*.gcno
modules.builtin
Module.symvers
+*.dwo
#
# Top-level generic files
<function>drm_dev_unregister()</function> followed by a call to
<function>drm_dev_unref()</function>.
</para>
-!Edrivers/gpu/drm/drm_stub.c
+!Edrivers/gpu/drm/drm_drv.c
</sect2>
<sect2>
<title>Driver Load</title>
The path number in the range 0 ... (<num_paths> - 1).
Expressed in hexadecimal (WITHOUT any prefix like 0x).
+R<n>,<m>
+ This parameter allows repetitive patterns to be loaded quickly. <n> and <m>
+ are hexadecimal numbers. The last <n> mappings are repeated in the next <m>
+ slots.
+
Status
======
Set mappings for the first 7 entries to point to devices switch0, switch1,
switch2, switch0, switch1, switch2, switch1:
dmsetup message switch 0 set_region_mappings 0:0 :1 :2 :0 :1 :2 :1
+
+Set repetitive mapping. This command:
+ dmsetup message switch 0 set_region_mappings 1000:1 :2 R2,10
+is equivalent to:
+ dmsetup message switch 0 set_region_mappings 1000:1 :2 :1 :2 :1 :2 :1 :2 \
+ :1 :2 :1 :2 :1 :2 :1 :2 :1 :2
+
1) Interrupt client nodes
-------------------------
-Nodes that describe devices which generate interrupts must contain an either an
-"interrupts" property or an "interrupts-extended" property. These properties
-contain a list of interrupt specifiers, one per output interrupt. The format of
-the interrupt specifier is determined by the interrupt controller to which the
-interrupts are routed; see section 2 below for details.
+Nodes that describe devices which generate interrupts must contain an
+"interrupts" property, an "interrupts-extended" property, or both. If both are
+present, the latter should take precedence; the former may be provided simply
+for compatibility with software that does not recognize the latter. These
+properties contain a list of interrupt specifiers, one per output interrupt. The
+format of the interrupt specifier is determined by the interrupt controller to
+which the interrupts are routed; see section 2 below for details.
Example:
interrupt-parent = <&intc1>;
- if CIU clock divider value is 0 (that is divide by 1), both tx and rx
phase shift clocks should be 0.
-Required properties for a slot:
+Required properties for a slot (Deprecated - Recommend to use one slot per host):
* gpios: specifies a list of gpios used for command, clock and data bus. The
first gpio is the command line and the second gpio is the clock line. The
rest of the gpios (depending on the bus-width property) are the data lines in
no particular order. The format of the gpio specifier depends on the gpio
controller.
+(Deprecated - Refer to Documentation/devicetree/binding/pinctrl/samsung-pinctrl.txt)
Example:
dwmmc0@12200000 {
num-slots = <1>;
- supports-highspeed;
+ cap-mmc-highspeed;
+ cap-sd-highspeed;
broken-cd;
fifo-depth = <0x80>;
card-detect-delay = <200>;
samsung,dw-mshc-ciu-div = <3>;
samsung,dw-mshc-sdr-timing = <2 3>;
samsung,dw-mshc-ddr-timing = <1 2>;
-
- slot@0 {
- reg = <0>;
- bus-width = <8>;
- gpios = <&gpc0 0 2 0 3>, <&gpc0 1 2 0 3>,
- <&gpc1 0 2 3 3>, <&gpc1 1 2 3 3>,
- <&gpc1 2 2 3 3>, <&gpc1 3 2 3 3>,
- <&gpc0 3 2 3 3>, <&gpc0 4 2 3 3>,
- <&gpc0 5 2 3 3>, <&gpc0 6 2 3 3>;
- };
+ bus-width = <8>;
};
num-slots = <1>;
vmmc-supply = <&ldo12>;
fifo-depth = <0x100>;
- supports-highspeed;
pinctrl-names = "default";
pinctrl-0 = <&sd_pmx_pins &sd_cfg_func1 &sd_cfg_func2>;
- slot@0 {
- reg = <0>;
- bus-width = <4>;
- disable-wp;
- cd-gpios = <&gpio10 3 0>;
- };
+ bus-width = <4>;
+ disable-wp;
+ cd-gpios = <&gpio10 3 0>;
+ cap-mmc-highspeed;
+ cap-sd-highspeed;
};
- cap-power-off-card: powering off the card is safe
- cap-sdio-irq: enable SDIO IRQ signalling on this interface
- full-pwr-cycle: full power cycle of the card is supported
-- mmc-highspeed-ddr-1_8v: eMMC high-speed DDR mode(1.8V I/O) is supported
-- mmc-highspeed-ddr-1_2v: eMMC high-speed DDR mode(1.2V I/O) is supported
+- mmc-ddr-1_8v: eMMC high-speed DDR mode(1.8V I/O) is supported
+- mmc-ddr-1_2v: eMMC high-speed DDR mode(1.2V I/O) is supported
- mmc-hs200-1_8v: eMMC HS200 mode(1.8V I/O) is supported
- mmc-hs200-1_2v: eMMC HS200 mode(1.2V I/O) is supported
- mmc-hs400-1_8v: eMMC HS400 mode(1.8V I/O) is supported
--- /dev/null
+* Renesas Multi Media Card Interface (MMCIF) Controller
+
+This file documents differences between the core properties in mmc.txt
+and the properties used by the MMCIF device.
+
+
+Required properties:
+
+- compatible: must contain one of the following
+ - "renesas,mmcif-r8a7740" for the MMCIF found in r8a7740 SoCs
+ - "renesas,mmcif-r8a7790" for the MMCIF found in r8a7790 SoCs
+ - "renesas,mmcif-r8a7791" for the MMCIF found in r8a7791 SoCs
+ - "renesas,sh-mmcif" for the generic MMCIF
+
+- clocks: reference to the functional clock
+
+- dmas: reference to the DMA channels, one per channel name listed in the
+ dma-names property.
+- dma-names: must contain "tx" for the transmit DMA channel and "rx" for the
+ receive DMA channel.
+
+
+Example: R8A7790 (R-Car H2) MMCIF0
+
+ mmcif0: mmc@ee200000 {
+ compatible = "renesas,mmcif-r8a7790", "renesas,sh-mmcif";
+ reg = <0 0xee200000 0 0x80>;
+ interrupts = <0 169 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&mstp3_clks R8A7790_CLK_MMCIF0>;
+ dmas = <&dmac0 0xd1>, <&dmac0 0xd2>;
+ dma-names = "tx", "rx";
+ };
bus-width = <8>;
non-removable;
- vmmc = <&pm8941_l20>;
- vqmmc = <&pm8941_s3>;
+ vmmc-supply = <&pm8941_l20>;
+ vqmmc-supply = <&pm8941_s3>;
pinctrl-names = "default";
pinctrl-0 = <&sdc1_clk &sdc1_cmd &sdc1_data>;
bus-width = <4>;
cd-gpios = <&msmgpio 62 0x1>;
- vmmc = <&pm8941_l21>;
- vqmmc = <&pm8941_l13>;
+ vmmc-supply = <&pm8941_l21>;
+ vqmmc-supply = <&pm8941_l13>;
pinctrl-names = "default";
pinctrl-0 = <&sdc2_clk &sdc2_cmd &sdc2_data>;
--- /dev/null
+* STMicroelectronics sdhci-st MMC/SD controller
+
+This file documents the differences between the core properties in
+Documentation/devicetree/bindings/mmc/mmc.txt and the properties
+used by the sdhci-st driver.
+
+Required properties:
+- compatible : Must be "st,sdhci"
+- clock-names : Should be "mmc"
+ See: Documentation/devicetree/bindings/resource-names.txt
+- clocks : Phandle of the clock used by the sdhci controler
+ See: Documentation/devicetree/bindings/clock/clock-bindings.txt
+
+Optional properties:
+- non-removable: non-removable slot
+ See: Documentation/devicetree/bindings/mmc/mmc.txt
+- bus-width: Number of data lines
+ See: Documentation/devicetree/bindings/mmc/mmc.txt
+
+Example:
+
+mmc0: sdhci@fe81e000 {
+ compatible = "st,sdhci";
+ status = "disabled";
+ reg = <0xfe81e000 0x1000>;
+ interrupts = <GIC_SPI 127 IRQ_TYPE_NONE>;
+ interrupt-names = "mmcirq";
+ pinctrl-names = "default";
+ pinctrl-0 = <&pinctrl_mmc0>;
+ clock-names = "mmc";
+ clocks = <&clk_s_a1_ls 1>;
+ bus-width = <8>
+};
* card-detect-delay: Delay in milli-seconds before detecting card after card
insert event. The default value is 0.
-* supports-highspeed: Enables support for high speed cards (up to 50MHz)
+* supports-highspeed (DEPRECATED): Enables support for high speed cards (up to 50MHz)
+ (use "cap-mmc-highspeed" or "cap-sd-highspeed" instead)
* broken-cd: as documented in mmc core bindings.
clock-frequency = <400000000>;
clock-freq-min-max = <400000 200000000>;
num-slots = <1>;
- supports-highspeed;
broken-cd;
fifo-depth = <0x80>;
card-detect-delay = <200>;
vmmc-supply = <&buck8>;
-
- slot@0 {
- reg = <0>;
- bus-width = <8>;
- };
+ bus-width = <8>;
+ cap-mmc-highspeed;
+ cap-sd-highspeed;
};
Should be "ti,omap3-hsmmc", for OMAP3 controllers
Should be "ti,omap3-pre-es3-hsmmc" for OMAP3 controllers pre ES3.0
Should be "ti,omap4-hsmmc", for OMAP4 controllers
+ Should be "ti,am33xx-hsmmc", for AM335x controllers
- ti,hwmods: Must be "mmc<n>", n is controller instance starting 1
Optional properties:
&edma 25>;
dma-names = "tx", "rx";
};
+
+[workaround for missing swakeup on am33xx]
+
+This SOC is missing the swakeup line, it will not detect SDIO irq
+while in suspend.
+
+ ------
+ | PRCM |
+ ------
+ ^ |
+ swakeup | | fclk
+ | v
+ ------ ------- -----
+ | card | -- CIRQ --> | hsmmc | -- IRQ --> | CPU |
+ ------ ------- -----
+
+In suspend the fclk is off and the module is disfunctional. Even register reads
+will fail. A small logic in the host will request fclk restore, when an
+external event is detected. Once the clock is restored, the host detects the
+event normally. Since am33xx doesn't have this line it never wakes from
+suspend.
+
+The workaround is to reconfigure the dat1 line as a GPIO upon suspend. To make
+this work, we need to set the named pinctrl states "default" and "idle".
+Prepare idle to remux dat1 as a gpio, and default to remux it back as sdio
+dat1. The MMC driver will then toggle between idle and default state during
+runtime.
+
+In summary:
+1. select matching 'compatible' section, see example below.
+2. specify pinctrl states "default" and "idle", "sleep" is optional.
+3. specify the gpio irq used for detecting sdio irq in suspend
+
+If configuration is incomplete, a warning message is emitted "falling back to
+polling". Also check the "sdio irq mode" in /sys/kernel/debug/mmc0/regs. Mind
+not every application needs SDIO irq, e.g. MMC cards.
+
+ mmc1: mmc@48060100 {
+ compatible = "ti,am33xx-hsmmc";
+ ...
+ pinctrl-names = "default", "idle", "sleep"
+ pinctrl-0 = <&mmc1_pins>;
+ pinctrl-1 = <&mmc1_idle>;
+ pinctrl-2 = <&mmc1_sleep>;
+ ...
+ interrupts-extended = <&intc 64 &gpio2 28 0>;
+ };
+
+ mmc1_idle : pinmux_cirq_pin {
+ pinctrl-single,pins = <
+ 0x0f8 0x3f /* GPIO2_28 */
+ >;
+ };
"renesas,sdhi-r8a7778" - SDHI IP on R8A7778 SoC
"renesas,sdhi-r8a7779" - SDHI IP on R8A7779 SoC
"renesas,sdhi-r8a7790" - SDHI IP on R8A7790 SoC
+ "renesas,sdhi-r8a7791" - SDHI IP on R8A7791 SoC
Optional properties:
- toshiba,mmc-wrprotect-disable: write-protect detection is unavailable
Required properties:
- compatible: should contain "snps,dw-pcie" to identify the core.
+- reg: Should contain the configuration address space.
+- reg-names: Must be "config" for the PCIe configuration space.
+ (The old way of getting the configuration address space from "ranges"
+ is deprecated and should be avoided.)
- #address-cells: set to <3>
- #size-cells: set to <2>
- device_type: set to "pci"
--- /dev/null
+TI PCI Controllers
+
+PCIe Designware Controller
+ - compatible: Should be "ti,dra7-pcie""
+ - reg : Two register ranges as listed in the reg-names property
+ - reg-names : The first entry must be "ti-conf" for the TI specific registers
+ The second entry must be "rc-dbics" for the designware pcie
+ registers
+ The third entry must be "config" for the PCIe configuration space
+ - phys : list of PHY specifiers (used by generic PHY framework)
+ - phy-names : must be "pcie-phy0", "pcie-phy1", "pcie-phyN".. based on the
+ number of PHYs as specified in *phys* property.
+ - ti,hwmods : Name of the hwmod associated to the pcie, "pcie<X>",
+ where <X> is the instance number of the pcie from the HW spec.
+ - interrupts : Two interrupt entries must be specified. The first one is for
+ main interrupt line and the second for MSI interrupt line.
+ - #address-cells,
+ #size-cells,
+ #interrupt-cells,
+ device_type,
+ ranges,
+ num-lanes,
+ interrupt-map-mask,
+ interrupt-map : as specified in ../designware-pcie.txt
+
+Example:
+axi {
+ compatible = "simple-bus";
+ #size-cells = <1>;
+ #address-cells = <1>;
+ ranges = <0x51000000 0x51000000 0x3000
+ 0x0 0x20000000 0x10000000>;
+ pcie@51000000 {
+ compatible = "ti,dra7-pcie";
+ reg = <0x51000000 0x2000>, <0x51002000 0x14c>, <0x1000 0x2000>;
+ reg-names = "rc_dbics", "ti_conf", "config";
+ interrupts = <0 232 0x4>, <0 233 0x4>;
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ ranges = <0x81000000 0 0 0x03000 0 0x00010000
+ 0x82000000 0 0x20013000 0x13000 0 0xffed000>;
+ #interrupt-cells = <1>;
+ num-lanes = <1>;
+ ti,hwmods = "pcie1";
+ phys = <&pcie1_phy>;
+ phy-names = "pcie-phy0";
+ interrupt-map-mask = <0 0 0 7>;
+ interrupt-map = <0 0 0 1 &pcie_intc 1>,
+ <0 0 0 2 &pcie_intc 2>,
+ <0 0 0 3 &pcie_intc 3>,
+ <0 0 0 4 &pcie_intc 4>;
+ pcie_intc: interrupt-controller {
+ interrupt-controller;
+ #address-cells = <0>;
+ #interrupt-cells = <1>;
+ };
+ };
+};
--- /dev/null
+A DT changeset is a method which allows one to apply changes
+in the live tree in such a way that either the full set of changes
+will be applied, or none of them will be. If an error occurs partway
+through applying the changeset, then the tree will be rolled back to the
+previous state. A changeset can also be removed after it has been
+applied.
+
+When a changeset is applied, all of the changes get applied to the tree
+at once before emitting OF_RECONFIG notifiers. This is so that the
+receiver sees a complete and consistent state of the tree when it
+receives the notifier.
+
+The sequence of a changeset is as follows.
+
+1. of_changeset_init() - initializes a changeset
+
+2. A number of DT tree change calls, of_changeset_attach_node(),
+of_changeset_detach_node(), of_changeset_add_property(),
+of_changeset_remove_property, of_changeset_update_property() to prepare
+a set of changes. No changes to the active tree are made at this point.
+All the change operations are recorded in the of_changeset 'entries'
+list.
+
+3. mutex_lock(of_mutex) - starts a changeset; The global of_mutex
+ensures there can only be one editor at a time.
+
+4. of_changeset_apply() - Apply the changes to the tree. Either the
+entire changeset will get applied, or if there is an error the tree will
+be restored to the previous state
+
+5. mutex_unlock(of_mutex) - All operations complete, release the mutex
+
+If a successfully applied changeset needs to be removed, it can be done
+with the following sequence.
+
+1. mutex_lock(of_mutex)
+
+2. of_changeset_revert()
+
+3. mutex_unlock(of_mutex)
--- /dev/null
+Todo list for devicetree:
+
+=== General structure ===
+- Switch from custom lists to (h)list_head for nodes and properties structure
+- Remove of_allnodes list and iterate using list of child nodes alone
+
+=== CONFIG_OF_DYNAMIC ===
+- Switch to RCU for tree updates and get rid of global spinlock
+- Document node lifecycle for CONFIG_OF_DYNAMIC
+- Always set ->full_name at of_attach_node() time
+- pseries: Get rid of open-coded tree modification from arch/powerpc/platforms/pseries/dlpar.c
locking rules:
inode->i_lock may block
fl_copy_lock: yes no
-fl_release_private: maybe no
+fl_release_private: maybe maybe[1]
+
+[1]: ->fl_release_private for flock or POSIX locks is currently allowed
+to block. Leases however can still be freed while the i_lock is held and
+so fl_release_private called on a lease should not block.
----------------------- lock_manager_operations ---------------------------
prototypes:
ioctl. Also, all agents registered through a file descriptor will
be unregistered when the descriptor is closed.
+ 2014 -- a new registration ioctl is now provided which allows additional
+ fields to be provided during registration.
+ Users of this registration call are implicitly setting the use of
+ pkey_index (see below).
+
Receiving MADs
MADs are received using read(). The receive side now supports
The old ib_umad interface did not allow setting the P_Key index for
MADs that are sent and did not provide a way for obtaining the P_Key
index of received MADs. A new layout for struct ib_user_mad_hdr
- with a pkey_index member has been defined; however, to preserve
- binary compatibility with older applications, this new layout will
- not be used unless the IB_USER_MAD_ENABLE_PKEY ioctl is called
- before a file descriptor is used for anything else.
+ with a pkey_index member has been defined; however, to preserve binary
+ compatibility with older applications, this new layout will not be used
+ unless one of IB_USER_MAD_ENABLE_PKEY or IB_USER_MAD_REGISTER_AGENT2 ioctl's
+ are called before a file descriptor is used for anything else.
In September 2008, the IB_USER_MAD_ABI_VERSION will be incremented
to 6, the new layout of struct ib_user_mad_hdr will be used by
00-INDEX
- this file: info on the kernel build process
+headers_install.txt
+ - how to export Linux headers for use by userspace
kbuild.txt
- developer information on kbuild
kconfig.txt
--- /dev/null
+Exporting kernel headers for use by userspace
+=============================================
+
+The "make headers_install" command exports the kernel's header files in a
+form suitable for use by userspace programs.
+
+The linux kernel's exported header files describe the API for user space
+programs attempting to use kernel services. These kernel header files are
+used by the system's C library (such as glibc or uClibc) to define available
+system calls, as well as constants and structures to be used with these
+system calls. The C library's header files include the kernel header files
+from the "linux" subdirectory. The system's libc headers are usually
+installed at the default location /usr/include and the kernel headers in
+subdirectories under that (most notably /usr/include/linux and
+/usr/include/asm).
+
+Kernel headers are backwards compatible, but not forwards compatible. This
+means that a program built against a C library using older kernel headers
+should run on a newer kernel (although it may not have access to new
+features), but a program built against newer kernel headers may not work on an
+older kernel.
+
+The "make headers_install" command can be run in the top level directory of the
+kernel source code (or using a standard out-of-tree build). It takes two
+optional arguments:
+
+ make headers_install ARCH=i386 INSTALL_HDR_PATH=/usr/include
+
+ARCH indicates which architecture to produce headers for, and defaults to the
+current architecture. The linux/asm directory of the exported kernel headers
+is platform-specific, to see a complete list of supported architectures use
+the command:
+
+ ls -d include/asm-* | sed 's/.*-//'
+
+INSTALL_HDR_PATH indicates where to install the headers. It defaults to
+"./usr/include".
+
+The command "make headers_install_all" exports headers for all architectures
+simultaneously. (This is mostly of interest to distribution maintainers,
+who create an architecture-independent tarball from the resulting include
+directory.) You also can use HDR_ARCH_LIST to specify list of architectures.
+Remember to provide the appropriate linux/asm directory via "mv" or "ln -s"
+before building a C library with headers exported this way.
+
+The kernel header export infrastructure is maintained by David Woodhouse
+<dwmw2@infradead.org>.
=== 4 Host Program support
--- 4.1 Simple Host Program
--- 4.2 Composite Host Programs
- --- 4.3 Defining shared libraries
- --- 4.4 Using C++ for host programs
- --- 4.5 Controlling compiler options for host programs
- --- 4.6 When host programs are actually built
- --- 4.7 Using hostprogs-$(CONFIG_FOO)
+ --- 4.3 Using C++ for host programs
+ --- 4.4 Controlling compiler options for host programs
+ --- 4.5 When host programs are actually built
+ --- 4.6 Using hostprogs-$(CONFIG_FOO)
=== 5 Kbuild clean infrastructure
Finally, the two .o files are linked to the executable, lxdialog.
Note: The syntax <executable>-y is not permitted for host-programs.
---- 4.3 Defining shared libraries
-
- Objects with extension .so are considered shared libraries, and
- will be compiled as position independent objects.
- Kbuild provides support for shared libraries, but the usage
- shall be restricted.
- In the following example the libkconfig.so shared library is used
- to link the executable conf.
-
- Example:
- #scripts/kconfig/Makefile
- hostprogs-y := conf
- conf-objs := conf.o libkconfig.so
- libkconfig-objs := expr.o type.o
-
- Shared libraries always require a corresponding -objs line, and
- in the example above the shared library libkconfig is composed by
- the two objects expr.o and type.o.
- expr.o and type.o will be built as position independent code and
- linked as a shared library libkconfig.so. C++ is not supported for
- shared libraries.
-
---- 4.4 Using C++ for host programs
+--- 4.3 Using C++ for host programs
kbuild offers support for host programs written in C++. This was
introduced solely to support kconfig, and is not recommended
qconf-cxxobjs := qconf.o
qconf-objs := check.o
---- 4.5 Controlling compiler options for host programs
+--- 4.4 Controlling compiler options for host programs
When compiling host programs, it is possible to set specific flags.
The programs will always be compiled utilising $(HOSTCC) passed
When linking qconf, it will be passed the extra option
"-L$(QTDIR)/lib".
---- 4.6 When host programs are actually built
+--- 4.5 When host programs are actually built
Kbuild will only build host-programs when they are referenced
as a prerequisite.
This will tell kbuild to build lxdialog even if not referenced in
any rule.
---- 4.7 Using hostprogs-$(CONFIG_FOO)
+--- 4.6 Using hostprogs-$(CONFIG_FOO)
A typical pattern in a Kbuild file looks like this:
- info on Linux Sony Programmable I/O Device support.
thinkpad-acpi.txt
- information on the (IBM and Lenovo) ThinkPad ACPI Extras driver.
+toshiba_haps.txt
+ - information on the Toshiba HDD Active Protection Sensor driver.
--- /dev/null
+Kernel driver toshiba_haps
+Toshiba HDD Active Protection Sensor
+====================================
+
+Author: Azael Avalos <coproscefalo@gmail.com>
+
+
+0. Contents
+-----------
+
+1. Description
+2. Interface
+3. Accelerometer axes
+4. Supported devices
+5. Usage
+
+
+1. Description
+--------------
+
+This driver provides support for the accelerometer found in various Toshiba
+laptops, being called "Toshiba HDD Protection - Shock Sensor" officialy,
+and detects laptops automatically with this device.
+On Windows, Toshiba provided software monitors this device and provides
+automatic HDD protection (head unload) on sudden moves or harsh vibrations,
+however, this driver only provides a notification via a sysfs file to let
+userspace tools or daemons act accordingly, as well as providing a sysfs
+file to set the desired protection level or sensor sensibility.
+
+
+2. Interface
+------------
+
+This device comes with 3 methods:
+_STA - Checks existence of the device, returning Zero if the device does not
+ exists or is not supported.
+PTLV - Sets the desired protection level.
+RSSS - Shuts down the HDD protection interface for a few seconds,
+ then restores normal operation.
+
+Note:
+The presence of Solid State Drives (SSD) can make this driver to fail loading,
+given the fact that such drives have no movable parts, and thus, not requiring
+any "protection" as well as failing during the evaluation of the _STA method
+found under this device.
+
+
+3. Accelerometer axes
+---------------------
+
+This device does not report any axes, however, to query the sensor position
+a couple HCI (Hardware Configuration Interface) calls (0x6D and 0xA6) are
+provided to query such information, handled by the kernel module toshiba_acpi
+since kernel version 3.15.
+
+
+4. Supported devices
+--------------------
+
+This driver binds itself to the ACPI device TOS620A, and any Toshiba laptop
+with this device is supported, given the fact that they have the presence of
+conventional HDD and not only SSD, or a combination of both HDD and SSD.
+
+
+5. Usage
+--------
+
+The sysfs files under /sys/devices/LNXSYSTM:00/LNXSYBUS:00/TOS620A:00/ are:
+protection_level - The protection_level is readable and writeable, and
+ provides a way to let userspace query the current protection
+ level, as well as set the desired protection level, the
+ available protection levels are:
+ 0 - Disabled | 1 - Low | 2 - Medium | 3 - High
+reset_protection - The reset_protection entry is writeable only, being "1"
+ the only parameter it accepts, it is used to trigger
+ a reset of the protection interface.
+++ /dev/null
-Exporting kernel headers for use by userspace
-=============================================
-
-The "make headers_install" command exports the kernel's header files in a
-form suitable for use by userspace programs.
-
-The linux kernel's exported header files describe the API for user space
-programs attempting to use kernel services. These kernel header files are
-used by the system's C library (such as glibc or uClibc) to define available
-system calls, as well as constants and structures to be used with these
-system calls. The C library's header files include the kernel header files
-from the "linux" subdirectory. The system's libc headers are usually
-installed at the default location /usr/include and the kernel headers in
-subdirectories under that (most notably /usr/include/linux and
-/usr/include/asm).
-
-Kernel headers are backwards compatible, but not forwards compatible. This
-means that a program built against a C library using older kernel headers
-should run on a newer kernel (although it may not have access to new
-features), but a program built against newer kernel headers may not work on an
-older kernel.
-
-The "make headers_install" command can be run in the top level directory of the
-kernel source code (or using a standard out-of-tree build). It takes two
-optional arguments:
-
- make headers_install ARCH=i386 INSTALL_HDR_PATH=/usr/include
-
-ARCH indicates which architecture to produce headers for, and defaults to the
-current architecture. The linux/asm directory of the exported kernel headers
-is platform-specific, to see a complete list of supported architectures use
-the command:
-
- ls -d include/asm-* | sed 's/.*-//'
-
-INSTALL_HDR_PATH indicates where to install the headers. It defaults to
-"./usr/include".
-
-The command "make headers_install_all" exports headers for all architectures
-simultaneously. (This is mostly of interest to distribution maintainers,
-who create an architecture-independent tarball from the resulting include
-directory.) You also can use HDR_ARCH_LIST to specify list of architectures.
-Remember to provide the appropriate linux/asm directory via "mv" or "ln -s"
-before building a C library with headers exported this way.
-
-The kernel header export infrastructure is maintained by David Woodhouse
-<dwmw2@infradead.org>.
F: Documentation/filesystems/befs.txt
F: fs/befs/
+BECKHOFF CX5020 ETHERCAT MASTER DRIVER
+M: Dariusz Marcinkiewicz <reksio@newterm.pl>
+L: netdev@vger.kernel.org
+S: Maintained
+F: drivers/net/ethernet/ec_bhf.c
+
BFS FILE SYSTEM
M: "Tigran A. Aivazian" <tigran@aivazian.fsnet.co.uk>
S: Maintained
FREESCALE SOC SOUND DRIVERS
M: Timur Tabi <timur@tabi.org>
+M: Nicolin Chen <nicoleotsuka@gmail.com>
+M: Xiubo Li <Li.Xiubo@freescale.com>
L: alsa-devel@alsa-project.org (moderated for non-subscribers)
L: linuxppc-dev@lists.ozlabs.org
S: Maintained
F: sound/soc/fsl/fsl*
+F: sound/soc/fsl/imx*
F: sound/soc/fsl/mpc8610_hpcd.c
FREEVXFS FILESYSTEM
F: include/uapi/linux/i2c.h
F: include/uapi/linux/i2c-*.h
+I2C ACPI SUPPORT
+M: Mika Westerberg <mika.westerberg@linux.intel.com>
+L: linux-i2c@vger.kernel.org
+L: linux-acpi@vger.kernel.org
+S: Maintained
+F: drivers/i2c/i2c-acpi.c
+
I2C-TAOS-EVM DRIVER
M: Jean Delvare <jdelvare@suse.de>
L: linux-i2c@vger.kernel.org
S: Maintained
F: drivers/media/radio/radio-mr800.c
+MRF24J40 IEEE 802.15.4 RADIO DRIVER
+M: Alan Ott <alan@signal11.us>
+L: linux-wpan@vger.kernel.org
+S: Maintained
+F: drivers/net/ieee802154/mrf24j40.c
+
MSI LAPTOP SUPPORT
M: "Lee, Chun-Yi" <jlee@suse.com>
L: platform-driver-x86@vger.kernel.org
F: Documentation/devicetree/bindings/pci/nvidia,tegra20-pcie.txt
F: drivers/pci/host/pci-tegra.c
+PCI DRIVER FOR TI DRA7XX
+M: Kishon Vijay Abraham I <kishon@ti.com>
+L: linux-omap@vger.kernel.org
+L: linux-pci@vger.kernel.org
+S: Supported
+F: Documentation/devicetree/bindings/pci/ti-pci.txt
+F: drivers/pci/host/pci-dra7xx.c
+
PCI DRIVER FOR RENESAS R-CAR
M: Simon Horman <horms@verge.net.au>
L: linux-pci@vger.kernel.org
PMC SIERRA PM8001 DRIVER
M: xjtuwjp@gmail.com
M: lindar_liu@usish.com
+L: pmchba@pmcs.com
L: linux-scsi@vger.kernel.org
S: Supported
F: drivers/scsi/pm8001/
X86 PLATFORM DRIVERS
M: Matthew Garrett <matthew.garrett@nebula.com>
L: platform-driver-x86@vger.kernel.org
-T: git git://git.kernel.org/pub/scm/linux/kernel/git/mjg59/platform-drivers-x86.git
+T: git git://cavan.codon.org.uk/platform-drivers-x86.git
S: Maintained
F: drivers/platform/x86/
VERSION = 3
-PATCHLEVEL = 16
+PATCHLEVEL = 17
SUBLEVEL = 0
-EXTRAVERSION =
+EXTRAVERSION = -rc1
NAME = Shuffling Zombie Juror
# *DOCUMENTATION*
INSTALLKERNEL := installkernel
DEPMOD = /sbin/depmod
PERL = perl
+PYTHON = python
CHECK = sparse
CHECKFLAGS := -D__linux__ -Dlinux -D__STDC__ -Dunix -D__unix__ \
export VERSION PATCHLEVEL SUBLEVEL KERNELRELEASE KERNELVERSION
export ARCH SRCARCH CONFIG_SHELL HOSTCC HOSTCFLAGS CROSS_COMPILE AS LD CC
export CPP AR NM STRIP OBJCOPY OBJDUMP
-export MAKE AWK GENKSYMS INSTALLKERNEL PERL UTS_MACHINE
+export MAKE AWK GENKSYMS INSTALLKERNEL PERL PYTHON UTS_MACHINE
export HOSTCXX HOSTCXXFLAGS LDFLAGS_MODULE CHECK CHECKFLAGS
export KBUILD_CPPFLAGS NOSTDINC_FLAGS LINUXINCLUDE OBJCOPYFLAGS LDFLAGS
# source of a reference will be _MergedGlobals and not on of the whitelisted names.
# See modpost pattern 2
KBUILD_CFLAGS += $(call cc-option, -mno-global-merge,)
+KBUILD_CFLAGS += $(call cc-option, -fcatch-undefined-behavior)
else
# This warning generated too much noise in a regular build.
KBUILD_CFLAGS += $(call cc-option, -fno-var-tracking-assignments)
ifdef CONFIG_DEBUG_INFO
+ifdef CONFIG_DEBUG_INFO_SPLIT
+KBUILD_CFLAGS += $(call cc-option, -gsplit-dwarf, -g)
+else
KBUILD_CFLAGS += -g
+endif
KBUILD_AFLAGS += -Wa,-gdwarf-2
endif
+ifdef CONFIG_DEBUG_INFO_DWARF4
+KBUILD_CFLAGS += $(call cc-option, -gdwarf-4,)
+endif
ifdef CONFIG_DEBUG_INFO_REDUCED
KBUILD_CFLAGS += $(call cc-option, -femit-struct-debug-baseonly) \
$(Q)$(MAKE) $(hdr-inst)=include/uapi HDRCHECK=1
$(Q)$(MAKE) $(hdr-inst)=arch/$(hdr-arch)/include/uapi/asm $(hdr-dst) HDRCHECK=1
+# ---------------------------------------------------------------------------
+# Kernel selftest
+
+PHONY += kselftest
+kselftest:
+ $(Q)$(MAKE) -C tools/testing/selftests run_tests
+
# ---------------------------------------------------------------------------
# Modules
@echo ' tags/TAGS - Generate tags file for editors'
@echo ' cscope - Generate cscope index'
@echo ' gtags - Generate GNU GLOBAL index'
- @echo ' kernelrelease - Output the release version string'
- @echo ' kernelversion - Output the version stored in Makefile'
- @echo ' image_name - Output the image name'
+ @echo ' kernelrelease - Output the release version string (use with make -s)'
+ @echo ' kernelversion - Output the version stored in Makefile (use with make -s)'
+ @echo ' image_name - Output the image name (use with make -s)'
@echo ' headers_install - Install sanitised kernel headers to INSTALL_HDR_PATH'; \
echo ' (default: $(INSTALL_HDR_PATH))'; \
echo ''
@echo ' headerdep - Detect inclusion cycles in headers'
@$(MAKE) -f $(srctree)/scripts/Makefile.help checker-help
@echo ''
+ @echo 'Kernel selftest'
+ @echo ' kselftest - Build and run kernel selftest (run as root)'
+ @echo ' Build, install, and boot kernel before'
+ @echo ' running kselftest on it'
+ @echo ''
@echo 'Kernel packaging:'
@$(MAKE) $(build)=$(package-dir) help
@echo ''
@find $(if $(KBUILD_EXTMOD), $(KBUILD_EXTMOD), .) $(RCS_FIND_IGNORE) \
\( -name '*.[oas]' -o -name '*.ko' -o -name '.*.cmd' \
-o -name '*.ko.*' \
+ -o -name '*.dwo' \
-o -name '.*.d' -o -name '.*.tmp' -o -name '*.mod.c' \
-o -name '*.symtypes' -o -name 'modules.order' \
-o -name modules.builtin -o -name '.tmp_*.o.*' \
i2c0 = &i2c0;
};
+ chosen {
+ stdout-path = &uart0;
+ };
+
memory {
reg = <0x0 0x08000000>;
};
};
};
};
-
-#include <testcases.dtsi>
{
return 0;
}
-
-int arch_gnttab_init(unsigned long nr_shared, unsigned long nr_status)
-{
- return 0;
-}
# The byte offset of the kernel image in RAM from the start of RAM.
ifeq ($(CONFIG_ARM64_RANDOMIZE_TEXT_OFFSET), y)
-TEXT_OFFSET := $(shell awk 'BEGIN {srand(); printf "0x%04x0\n", int(65535 * rand())}')
+TEXT_OFFSET := $(shell awk 'BEGIN {srand(); printf "0x%03x000\n", int(512 * rand())}')
else
TEXT_OFFSET := 0x00080000
endif
CONFIG_BLK_DEV_SD=y
# CONFIG_SCSI_LOWLEVEL is not set
CONFIG_ATA=y
+CONFIG_AHCI_XGENE=y
+CONFIG_PHY_XGENE=y
CONFIG_PATA_PLATFORM=y
CONFIG_PATA_OF_PLATFORM=y
CONFIG_NETDEVICES=y
CONFIG_VIRTIO_NET=y
CONFIG_SMC91X=y
CONFIG_SMSC911X=y
+CONFIG_NET_XGENE=y
# CONFIG_WLAN is not set
CONFIG_INPUT_EVDEV=y
# CONFIG_SERIO_SERPORT is not set
#define __ASM_SPARSEMEM_H
#ifdef CONFIG_SPARSEMEM
-#define MAX_PHYSMEM_BITS 40
+#define MAX_PHYSMEM_BITS 48
#define SECTION_SIZE_BITS 30
#endif
#define __ARM_NR_compat_cacheflush (__ARM_NR_COMPAT_BASE+2)
#define __ARM_NR_compat_set_tls (__ARM_NR_COMPAT_BASE+5)
-#define __NR_compat_syscalls 383
+#define __NR_compat_syscalls 386
#endif
#define __ARCH_WANT_SYS_CLONE
__SYSCALL(__NR_sched_getattr, sys_sched_getattr)
#define __NR_renameat2 382
__SYSCALL(__NR_renameat2, sys_renameat2)
+ /* 383 for seccomp */
+#define __NR_getrandom 384
+__SYSCALL(__NR_getrandom, sys_getrandom)
+#define __NR_memfd_create 385
+__SYSCALL(__NR_memfd_create, sys_memfd_create)
if (l1ip != ICACHE_POLICY_PIPT)
set_bit(ICACHEF_ALIASING, &__icache_flags);
- if (l1ip == ICACHE_POLICY_AIVIVT);
+ if (l1ip == ICACHE_POLICY_AIVIVT)
set_bit(ICACHEF_AIVIVT, &__icache_flags);
pr_info("Detected %s I-cache on CPU%d\n", icache_policy_str[l1ip], cpu);
if (uefi_debug)
pr_cont("\n");
}
+
+ set_bit(EFI_MEMMAP, &efi.flags);
}
#define KERNEL_RAM_VADDR (PAGE_OFFSET + TEXT_OFFSET)
-#if (TEXT_OFFSET & 0xf) != 0
-#error TEXT_OFFSET must be at least 16B aligned
-#elif (PAGE_OFFSET & 0xfffff) != 0
+#if (TEXT_OFFSET & 0xfff) != 0
+#error TEXT_OFFSET must be at least 4KB aligned
+#elif (PAGE_OFFSET & 0x1fffff) != 0
#error PAGE_OFFSET must be at least 2MB aligned
-#elif TEXT_OFFSET > 0xfffff
+#elif TEXT_OFFSET > 0x1fffff
#error TEXT_OFFSET must be less than 2MB
#endif
if (test_thread_flag(TIF_SYSCALL_TRACEPOINT))
trace_sys_enter(regs, regs->syscallno);
-#ifdef CONFIG_AUDITSYSCALL
audit_syscall_entry(syscall_get_arch(), regs->syscallno,
regs->orig_x0, regs->regs[1], regs->regs[2], regs->regs[3]);
-#endif
return regs->syscallno;
}
asmlinkage void syscall_trace_exit(struct pt_regs *regs)
{
-#ifdef CONFIG_AUDITSYSCALL
audit_syscall_exit(regs);
-#endif
if (test_thread_flag(TIF_SYSCALL_TRACEPOINT))
trace_sys_exit(regs, regs_return_value(regs));
#include <linux/of_fdt.h>
#include <linux/dma-mapping.h>
#include <linux/dma-contiguous.h>
+#include <linux/efi.h>
#include <asm/fixmap.h>
#include <asm/sections.h>
memblock_reserve(__virt_to_phys(initrd_start), initrd_end - initrd_start);
#endif
- early_init_fdt_scan_reserved_mem();
+ if (!efi_enabled(EFI_MEMMAP))
+ early_init_fdt_scan_reserved_mem();
/* 4GB maximum for 32-bit only capable devices */
if (IS_ENABLED(CONFIG_ZONE_DMA))
#define KSTK_EIP(tsk) ((tsk)->thread.frame0->pc)
#define KSTK_ESP(tsk) ((tsk)->thread.frame0->sp)
-#define cpu_relax() barrier()
+#define cpu_relax() barrier()
+#define cpu_relax_lowlatency() cpu_relax()
/* data cache prefetch */
#define ARCH_HAS_PREFETCH
$(Q)$(MAKE) $(build)=$(boot) $@
unwcheck: vmlinux
- -$(Q)READELF=$(READELF) python $(srctree)/arch/ia64/scripts/unwcheck.py $<
+ -$(Q)READELF=$(READELF) $(PYTHON) $(srctree)/arch/ia64/scripts/unwcheck.py $<
archclean:
$(Q)$(MAKE) $(clean)=$(boot)
-#define NR_syscalls 316 /* length of syscall table */
+#define NR_syscalls 317 /* length of syscall table */
/*
* The following defines stop scripts/checksyscalls.sh from complaining about
#define __NR_sched_getattr 1337
#define __NR_renameat2 1338
#define __NR_getrandom 1339
+#define __NR_memfd_create 1339
#endif /* _UAPI_ASM_IA64_UNISTD_H */
data8 sys_sched_getattr
data8 sys_renameat2
data8 sys_getrandom
+ data8 sys_memfd_create // 1340
.org sys_call_table + 8*NR_syscalls // guard against failures to increase NR_syscalls
#endif /* __IA64_ASM_PARAVIRTUALIZED_NATIVE */
#define __NR_sched_setattr 381
#define __NR_sched_getattr 382
#define __NR_renameat2 383
+#define __NR_seccomp 384
+#define __NR_getrandom 385
+#define __NR_memfd_create 386
#endif /* _UAPI_ASM_MICROBLAZE_UNISTD_H */
.long sys_sched_setattr
.long sys_sched_getattr
.long sys_renameat2
+ .long sys_seccomp
+ .long sys_getrandom /* 385 */
+ .long sys_memfd_create
r = zlib_inflate(&state->s, Z_FULL_FLUSH);
if (r != Z_OK && r != Z_STREAM_END)
fatal("inflate returned %d msg: %s\n\r", r, state->s.msg);
- len = state->s.next_out - (unsigned char *)dst;
+ len = state->s.next_out - (Byte *)dst;
} else {
/* uncompressed image */
- len = min(state->s.avail_in, (unsigned)dstlen);
+ len = min(state->s.avail_in, (uLong)dstlen);
memcpy(dst, state->s.next_in, len);
state->s.next_in += len;
state->s.avail_in -= len;
#define CPU_FTRS_POSSIBLE \
(CPU_FTRS_POWER4 | CPU_FTRS_PPC970 | CPU_FTRS_POWER5 | \
CPU_FTRS_POWER6 | CPU_FTRS_POWER7 | CPU_FTRS_POWER8E | \
- CPU_FTRS_POWER8 | CPU_FTRS_CELL | CPU_FTRS_PA6T | CPU_FTR_VSX)
+ CPU_FTRS_POWER8 | CPU_FTRS_POWER8_DD1 | CPU_FTRS_CELL | \
+ CPU_FTRS_PA6T | CPU_FTR_VSX)
#endif
#else
enum {
#define CPU_FTRS_ALWAYS \
(CPU_FTRS_POWER4 & CPU_FTRS_PPC970 & CPU_FTRS_POWER5 & \
CPU_FTRS_POWER6 & CPU_FTRS_POWER7 & CPU_FTRS_CELL & \
- CPU_FTRS_PA6T & CPU_FTRS_POSSIBLE)
+ CPU_FTRS_PA6T & CPU_FTRS_POWER8 & CPU_FTRS_POWER8E & \
+ CPU_FTRS_POWER8_DD1 & CPU_FTRS_POSSIBLE)
#endif
#else
enum {
void (*hpte_removebolted)(unsigned long ea,
int psize, int ssize);
void (*flush_hash_range)(unsigned long number, int local);
- void (*hugepage_invalidate)(struct mm_struct *mm,
+ void (*hugepage_invalidate)(unsigned long vsid,
+ unsigned long addr,
unsigned char *hpte_slot_array,
- unsigned long addr, int psize);
-
+ int psize, int ssize);
/* special for kexec, to be called in real mode, linear mapping is
* destroyed as well */
void (*hpte_clear_all)(void);
#define OPAL_DUMP_INFO2 94
#define OPAL_PCI_EEH_FREEZE_SET 97
#define OPAL_HANDLE_HMI 98
+#define OPAL_REGISTER_DUMP_REGION 101
+#define OPAL_UNREGISTER_DUMP_REGION 102
#ifndef __ASSEMBLY__
uint64_t length);
int64_t opal_sensor_read(uint32_t sensor_hndl, int token, __be32 *sensor_data);
int64_t opal_handle_hmi(void);
+int64_t opal_register_dump_region(uint32_t id, uint64_t start, uint64_t end);
+int64_t opal_unregister_dump_region(uint32_t id);
/* Internal functions */
extern int early_init_dt_scan_opal(unsigned long node, const char *uname,
unsigned long vmalloc_size);
void opal_free_sg_list(struct opal_sg_list *sg);
+/*
+ * Dump region ID range usable by the OS
+ */
+#define OPAL_DUMP_REGION_HOST_START 0x80
+#define OPAL_DUMP_REGION_LOG_BUF 0x80
+#define OPAL_DUMP_REGION_HOST_END 0xFF
+
#endif /* __ASSEMBLY__ */
#endif /* __OPAL_H */
}
extern void hpte_do_hugepage_flush(struct mm_struct *mm, unsigned long addr,
- pmd_t *pmdp);
+ pmd_t *pmdp, unsigned long old_pmd);
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
extern pmd_t pfn_pmd(unsigned long pfn, pgprot_t pgprot);
extern pmd_t mk_pmd(struct page *page, pgprot_t pgprot);
* in order to deal with 64K made of 4K HW pages. Thus we override the
* generic accessors and iterators here
*/
-#define __real_pte(e,p) ((real_pte_t) { \
- (e), (pte_val(e) & _PAGE_COMBO) ? \
- (pte_val(*((p) + PTRS_PER_PTE))) : 0 })
-#define __rpte_to_hidx(r,index) ((pte_val((r).pte) & _PAGE_COMBO) ? \
- (((r).hidx >> ((index)<<2)) & 0xf) : ((pte_val((r).pte) >> 12) & 0xf))
+#define __real_pte __real_pte
+static inline real_pte_t __real_pte(pte_t pte, pte_t *ptep)
+{
+ real_pte_t rpte;
+
+ rpte.pte = pte;
+ rpte.hidx = 0;
+ if (pte_val(pte) & _PAGE_COMBO) {
+ /*
+ * Make sure we order the hidx load against the _PAGE_COMBO
+ * check. The store side ordering is done in __hash_page_4K
+ */
+ smp_rmb();
+ rpte.hidx = pte_val(*((ptep) + PTRS_PER_PTE));
+ }
+ return rpte;
+}
+
+static inline unsigned long __rpte_to_hidx(real_pte_t rpte, unsigned long index)
+{
+ if ((pte_val(rpte.pte) & _PAGE_COMBO))
+ return (rpte.hidx >> (index<<2)) & 0xf;
+ return (pte_val(rpte.pte) >> 12) & 0xf;
+}
+
#define __rpte_to_pte(r) ((r).pte)
#define __rpte_sub_valid(rpte, index) \
(pte_val(rpte.pte) & (_PAGE_HPTE_SUB0 >> (index)))
#define SPRN_ACOP 0x1F /* Available Coprocessor Register */
#define SPRN_TFIAR 0x81 /* Transaction Failure Inst Addr */
#define SPRN_TEXASR 0x82 /* Transaction EXception & Summary */
-#define TEXASR_FS __MASK(63-36) /* Transaction Failure Summary */
#define SPRN_TEXASRU 0x83 /* '' '' '' Upper 32 */
-#define TEXASR_FS __MASK(63-36) /* TEXASR Failure Summary */
+#define TEXASR_FS __MASK(63-36) /* TEXASR Failure Summary */
#define SPRN_TFHAR 0x80 /* Transaction Failure Handler Addr */
#define SPRN_CTRLF 0x088
#define SPRN_CTRLT 0x098
static inline int arch_spin_is_locked(arch_spinlock_t *lock)
{
+ smp_mb();
return !arch_spin_value_unlocked(*lock);
}
MASKABLE_EXCEPTION_HV_OOL(0xe62, hmi_exception)
KVM_HANDLER(PACA_EXGEN, EXC_HV, 0xe62)
- .globl hmi_exception_early
-hmi_exception_early:
- EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0xe60)
- mr r10,r1 /* Save r1 */
- ld r1,PACAEMERGSP(r13) /* Use emergency stack */
- subi r1,r1,INT_FRAME_SIZE /* alloc stack frame */
- std r9,_CCR(r1) /* save CR in stackframe */
- mfspr r11,SPRN_HSRR0 /* Save HSRR0 */
- std r11,_NIP(r1) /* save HSRR0 in stackframe */
- mfspr r12,SPRN_HSRR1 /* Save SRR1 */
- std r12,_MSR(r1) /* save SRR1 in stackframe */
- std r10,0(r1) /* make stack chain pointer */
- std r0,GPR0(r1) /* save r0 in stackframe */
- std r10,GPR1(r1) /* save r1 in stackframe */
- EXCEPTION_PROLOG_COMMON_2(PACA_EXGEN)
- EXCEPTION_PROLOG_COMMON_3(0xe60)
- addi r3,r1,STACK_FRAME_OVERHEAD
- bl hmi_exception_realmode
- /* Windup the stack. */
- /* Clear MSR_RI before setting SRR0 and SRR1. */
- li r0,MSR_RI
- mfmsr r9 /* get MSR value */
- andc r9,r9,r0
- mtmsrd r9,1 /* Clear MSR_RI */
- /* Move original HSRR0 and HSRR1 into the respective regs */
- ld r9,_MSR(r1)
- mtspr SPRN_HSRR1,r9
- ld r3,_NIP(r1)
- mtspr SPRN_HSRR0,r3
- ld r9,_CTR(r1)
- mtctr r9
- ld r9,_XER(r1)
- mtxer r9
- ld r9,_LINK(r1)
- mtlr r9
- REST_GPR(0, r1)
- REST_8GPRS(2, r1)
- REST_GPR(10, r1)
- ld r11,_CCR(r1)
- mtcr r11
- REST_GPR(11, r1)
- REST_2GPRS(12, r1)
- /* restore original r1. */
- ld r1,GPR1(r1)
-
- /*
- * Go to virtual mode and pull the HMI event information from
- * firmware.
- */
- .globl hmi_exception_after_realmode
-hmi_exception_after_realmode:
- SET_SCRATCH0(r13)
- EXCEPTION_PROLOG_0(PACA_EXGEN)
- b hmi_exception_hv
-
MASKABLE_EXCEPTION_HV_OOL(0xe82, h_doorbell)
KVM_HANDLER(PACA_EXGEN, EXC_HV, 0xe82)
. = 0x8000
#endif /* defined(CONFIG_PPC_PSERIES) || defined(CONFIG_PPC_POWERNV) */
+ .globl hmi_exception_early
+hmi_exception_early:
+ EXCEPTION_PROLOG_1(PACA_EXGEN, NOTEST, 0xe60)
+ mr r10,r1 /* Save r1 */
+ ld r1,PACAEMERGSP(r13) /* Use emergency stack */
+ subi r1,r1,INT_FRAME_SIZE /* alloc stack frame */
+ std r9,_CCR(r1) /* save CR in stackframe */
+ mfspr r11,SPRN_HSRR0 /* Save HSRR0 */
+ std r11,_NIP(r1) /* save HSRR0 in stackframe */
+ mfspr r12,SPRN_HSRR1 /* Save SRR1 */
+ std r12,_MSR(r1) /* save SRR1 in stackframe */
+ std r10,0(r1) /* make stack chain pointer */
+ std r0,GPR0(r1) /* save r0 in stackframe */
+ std r10,GPR1(r1) /* save r1 in stackframe */
+ EXCEPTION_PROLOG_COMMON_2(PACA_EXGEN)
+ EXCEPTION_PROLOG_COMMON_3(0xe60)
+ addi r3,r1,STACK_FRAME_OVERHEAD
+ bl hmi_exception_realmode
+ /* Windup the stack. */
+ /* Clear MSR_RI before setting SRR0 and SRR1. */
+ li r0,MSR_RI
+ mfmsr r9 /* get MSR value */
+ andc r9,r9,r0
+ mtmsrd r9,1 /* Clear MSR_RI */
+ /* Move original HSRR0 and HSRR1 into the respective regs */
+ ld r9,_MSR(r1)
+ mtspr SPRN_HSRR1,r9
+ ld r3,_NIP(r1)
+ mtspr SPRN_HSRR0,r3
+ ld r9,_CTR(r1)
+ mtctr r9
+ ld r9,_XER(r1)
+ mtxer r9
+ ld r9,_LINK(r1)
+ mtlr r9
+ REST_GPR(0, r1)
+ REST_8GPRS(2, r1)
+ REST_GPR(10, r1)
+ ld r11,_CCR(r1)
+ mtcr r11
+ REST_GPR(11, r1)
+ REST_2GPRS(12, r1)
+ /* restore original r1. */
+ ld r1,GPR1(r1)
+
+ /*
+ * Go to virtual mode and pull the HMI event information from
+ * firmware.
+ */
+ .globl hmi_exception_after_realmode
+hmi_exception_after_realmode:
+ SET_SCRATCH0(r13)
+ EXCEPTION_PROLOG_0(PACA_EXGEN)
+ b hmi_exception_hv
+
#ifdef CONFIG_PPC_POWERNV
_GLOBAL(opal_mc_secondary_handler)
HMT_MEDIUM_PPR_DISCARD
/* We configure icbi to invalidate 128 bytes at a time since the
* current 32-bit kernel code isn't too happy with icache != dcache
- * block size
+ * block size. We also disable the BTAC as this can cause errors
+ * in some circumstances (see IBM Erratum 47).
*/
mfspr r3,SPRN_CCR0
oris r3,r3,0x0020
+ ori r3,r3,0x0040
mtspr SPRN_CCR0,r3
isync
int iommu_add_device(struct device *dev)
{
struct iommu_table *tbl;
- int ret = 0;
- if (WARN_ON(dev->iommu_group)) {
- pr_warn("iommu_tce: device %s is already in iommu group %d, skipping\n",
- dev_name(dev),
- iommu_group_id(dev->iommu_group));
+ /*
+ * The sysfs entries should be populated before
+ * binding IOMMU group. If sysfs entries isn't
+ * ready, we simply bail.
+ */
+ if (!device_is_registered(dev))
+ return -ENOENT;
+
+ if (dev->iommu_group) {
+ pr_debug("%s: Skipping device %s with iommu group %d\n",
+ __func__, dev_name(dev),
+ iommu_group_id(dev->iommu_group));
return -EBUSY;
}
tbl = get_iommu_table_base(dev);
if (!tbl || !tbl->it_group) {
- pr_debug("iommu_tce: skipping device %s with no tbl\n",
- dev_name(dev));
+ pr_debug("%s: Skipping device %s with no tbl\n",
+ __func__, dev_name(dev));
return 0;
}
- pr_debug("iommu_tce: adding %s to iommu group %d\n",
- dev_name(dev), iommu_group_id(tbl->it_group));
+ pr_debug("%s: Adding %s to iommu group %d\n",
+ __func__, dev_name(dev),
+ iommu_group_id(tbl->it_group));
if (PAGE_SIZE < IOMMU_PAGE_SIZE(tbl)) {
- pr_err("iommu_tce: unsupported iommu page size.");
- pr_err("%s has not been added\n", dev_name(dev));
+ pr_err("%s: Invalid IOMMU page size %lx (%lx) on %s\n",
+ __func__, IOMMU_PAGE_SIZE(tbl),
+ PAGE_SIZE, dev_name(dev));
return -EINVAL;
}
- ret = iommu_group_add_device(tbl->it_group, dev);
- if (ret < 0)
- pr_err("iommu_tce: %s has not been added, ret=%d\n",
- dev_name(dev), ret);
-
- return ret;
+ return iommu_group_add_device(tbl->it_group, dev);
}
EXPORT_SYMBOL_GPL(iommu_add_device);
}
EXPORT_SYMBOL(cpu_to_chip_id);
-#ifdef CONFIG_PPC_PSERIES
-/*
- * Fix up the uninitialized fields in a new device node:
- * name, type and pci-specific fields
- */
-
-static int of_finish_dynamic_node(struct device_node *node)
-{
- struct device_node *parent = of_get_parent(node);
- int err = 0;
- const phandle *ibm_phandle;
-
- node->name = of_get_property(node, "name", NULL);
- node->type = of_get_property(node, "device_type", NULL);
-
- if (!node->name)
- node->name = "<NULL>";
- if (!node->type)
- node->type = "<NULL>";
-
- if (!parent) {
- err = -ENODEV;
- goto out;
- }
-
- /* We don't support that function on PowerMac, at least
- * not yet
- */
- if (machine_is(powermac))
- return -ENODEV;
-
- /* fix up new node's phandle field */
- if ((ibm_phandle = of_get_property(node, "ibm,phandle", NULL)))
- node->phandle = *ibm_phandle;
-
-out:
- of_node_put(parent);
- return err;
-}
-
-static int prom_reconfig_notifier(struct notifier_block *nb,
- unsigned long action, void *node)
-{
- int err;
-
- switch (action) {
- case OF_RECONFIG_ATTACH_NODE:
- err = of_finish_dynamic_node(node);
- if (err < 0)
- printk(KERN_ERR "finish_node returned %d\n", err);
- break;
- default:
- err = 0;
- break;
- }
- return notifier_from_errno(err);
-}
-
-static struct notifier_block prom_reconfig_nb = {
- .notifier_call = prom_reconfig_notifier,
- .priority = 10, /* This one needs to run first */
-};
-
-static int __init prom_reconfig_setup(void)
-{
- return of_reconfig_notifier_register(&prom_reconfig_nb);
-}
-__initcall(prom_reconfig_setup);
-#endif
-
bool arch_match_cpu_phys_id(int cpu, u64 phys_id)
{
return (int)phys_id == get_hard_smp_processor_id(cpu);
GFP_KERNEL, cpu_to_node(cpu));
zalloc_cpumask_var_node(&per_cpu(cpu_core_map, cpu),
GFP_KERNEL, cpu_to_node(cpu));
+ /*
+ * numa_node_id() works after this.
+ */
+ set_cpu_numa_node(cpu, numa_cpu_lookup_table[cpu]);
+ set_cpu_numa_mem(cpu, local_memory_node(numa_cpu_lookup_table[cpu]));
}
cpumask_set_cpu(boot_cpuid, cpu_sibling_mask(boot_cpuid));
}
traverse_core_siblings(cpu, true);
- /*
- * numa_node_id() works after this.
- */
- set_numa_node(numa_cpu_lookup_table[cpu]);
- set_numa_mem(local_memory_node(numa_cpu_lookup_table[cpu]));
-
smp_wmb();
notify_cpu_starting(cpu);
set_cpu_online(cpu, true);
ri = kmalloc(sizeof(struct kvm_rma_info), GFP_KERNEL);
if (!ri)
return NULL;
- page = cma_alloc(kvm_cma, kvm_rma_pages, get_order(kvm_rma_pages));
+ page = cma_alloc(kvm_cma, kvm_rma_pages, order_base_2(kvm_rma_pages));
if (!page)
goto err_out;
atomic_set(&ri->use_count, 1);
{
unsigned long align_pages = HPT_ALIGN_PAGES;
- VM_BUG_ON(get_order(nr_pages) < KVM_CMA_CHUNK_ORDER - PAGE_SHIFT);
+ VM_BUG_ON(order_base_2(nr_pages) < KVM_CMA_CHUNK_ORDER - PAGE_SHIFT);
/* Old CPUs require HPT aligned on a multiple of its size */
if (!cpu_has_feature(CPU_FTR_ARCH_206))
align_pages = nr_pages;
- return cma_alloc(kvm_cma, nr_pages, get_order(align_pages));
+ return cma_alloc(kvm_cma, nr_pages, order_base_2(align_pages));
}
EXPORT_SYMBOL_GPL(kvm_alloc_hpt);
void arch_spin_unlock_wait(arch_spinlock_t *lock)
{
+ smp_mb();
+
while (lock->slock) {
HMT_low();
if (SHARED_PROCESSOR)
__spin_yield(lock);
}
HMT_medium();
+
+ smp_mb();
}
EXPORT_SYMBOL(arch_spin_unlock_wait);
local_irq_restore(flags);
}
-static void native_hugepage_invalidate(struct mm_struct *mm,
+static void native_hugepage_invalidate(unsigned long vsid,
+ unsigned long addr,
unsigned char *hpte_slot_array,
- unsigned long addr, int psize)
+ int psize, int ssize)
{
- int ssize = 0, i;
- int lock_tlbie;
+ int i;
struct hash_pte *hptep;
int actual_psize = MMU_PAGE_16M;
unsigned int max_hpte_count, valid;
unsigned long flags, s_addr = addr;
unsigned long hpte_v, want_v, shift;
- unsigned long hidx, vpn = 0, vsid, hash, slot;
+ unsigned long hidx, vpn = 0, hash, slot;
shift = mmu_psize_defs[psize].shift;
max_hpte_count = 1U << (PMD_SHIFT - shift);
/* get the vpn */
addr = s_addr + (i * (1ul << shift));
- if (!is_kernel_addr(addr)) {
- ssize = user_segment_size(addr);
- vsid = get_vsid(mm->context.id, addr, ssize);
- WARN_ON(vsid == 0);
- } else {
- vsid = get_kernel_vsid(addr, mmu_kernel_ssize);
- ssize = mmu_kernel_ssize;
- }
-
vpn = hpt_vpn(addr, vsid, ssize);
hash = hpt_hash(vpn, shift, ssize);
if (hidx & _PTEIDX_SECONDARY)
else
/* Invalidate the hpte. NOTE: this also unlocks it */
hptep->v = 0;
+ /*
+ * We need to do tlb invalidate for all the address, tlbie
+ * instruction compares entry_VA in tlb with the VA specified
+ * here
+ */
+ tlbie(vpn, psize, actual_psize, ssize, 0);
}
- /*
- * Since this is a hugepage, we just need a single tlbie.
- * use the last vpn.
- */
- lock_tlbie = !mmu_has_feature(MMU_FTR_LOCKLESS_TLBIE);
- if (lock_tlbie)
- raw_spin_lock(&native_tlbie_lock);
-
- asm volatile("ptesync":::"memory");
- __tlbie(vpn, psize, actual_psize, ssize);
- asm volatile("eieio; tlbsync; ptesync":::"memory");
-
- if (lock_tlbie)
- raw_spin_unlock(&native_tlbie_lock);
-
local_irq_restore(flags);
}
#include <linux/mm.h>
#include <asm/machdep.h>
+static void invalidate_old_hpte(unsigned long vsid, unsigned long addr,
+ pmd_t *pmdp, unsigned int psize, int ssize)
+{
+ int i, max_hpte_count, valid;
+ unsigned long s_addr;
+ unsigned char *hpte_slot_array;
+ unsigned long hidx, shift, vpn, hash, slot;
+
+ s_addr = addr & HPAGE_PMD_MASK;
+ hpte_slot_array = get_hpte_slot_array(pmdp);
+ /*
+ * IF we try to do a HUGE PTE update after a withdraw is done.
+ * we will find the below NULL. This happens when we do
+ * split_huge_page_pmd
+ */
+ if (!hpte_slot_array)
+ return;
+
+ if (ppc_md.hugepage_invalidate)
+ return ppc_md.hugepage_invalidate(vsid, s_addr, hpte_slot_array,
+ psize, ssize);
+ /*
+ * No bluk hpte removal support, invalidate each entry
+ */
+ shift = mmu_psize_defs[psize].shift;
+ max_hpte_count = HPAGE_PMD_SIZE >> shift;
+ for (i = 0; i < max_hpte_count; i++) {
+ /*
+ * 8 bits per each hpte entries
+ * 000| [ secondary group (one bit) | hidx (3 bits) | valid bit]
+ */
+ valid = hpte_valid(hpte_slot_array, i);
+ if (!valid)
+ continue;
+ hidx = hpte_hash_index(hpte_slot_array, i);
+
+ /* get the vpn */
+ addr = s_addr + (i * (1ul << shift));
+ vpn = hpt_vpn(addr, vsid, ssize);
+ hash = hpt_hash(vpn, shift, ssize);
+ if (hidx & _PTEIDX_SECONDARY)
+ hash = ~hash;
+
+ slot = (hash & htab_hash_mask) * HPTES_PER_GROUP;
+ slot += hidx & _PTEIDX_GROUP_IX;
+ ppc_md.hpte_invalidate(slot, vpn, psize,
+ MMU_PAGE_16M, ssize, 0);
+ }
+}
+
+
int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid,
pmd_t *pmdp, unsigned long trap, int local, int ssize,
unsigned int psize)
* atomically mark the linux large page PMD busy and dirty
*/
do {
- old_pmd = pmd_val(*pmdp);
+ pmd_t pmd = ACCESS_ONCE(*pmdp);
+
+ old_pmd = pmd_val(pmd);
/* If PMD busy, retry the access */
if (unlikely(old_pmd & _PAGE_BUSY))
return 0;
vpn = hpt_vpn(ea, vsid, ssize);
hash = hpt_hash(vpn, shift, ssize);
hpte_slot_array = get_hpte_slot_array(pmdp);
+ if (psize == MMU_PAGE_4K) {
+ /*
+ * invalidate the old hpte entry if we have that mapped via 64K
+ * base page size. This is because demote_segment won't flush
+ * hash page table entries.
+ */
+ if ((old_pmd & _PAGE_HASHPTE) && !(old_pmd & _PAGE_COMBO))
+ invalidate_old_hpte(vsid, ea, pmdp, MMU_PAGE_64K, ssize);
+ }
valid = hpte_valid(hpte_slot_array, index);
if (valid) {
* safely update this here.
*/
valid = 0;
- new_pmd &= ~_PAGE_HPTEFLAGS;
hpte_slot_array[index] = 0;
- } else
- /* clear the busy bits and set the hash pte bits */
- new_pmd = (new_pmd & ~_PAGE_HPTEFLAGS) | _PAGE_HASHPTE;
+ }
}
if (!valid) {
/* insert new entry */
pa = pmd_pfn(__pmd(old_pmd)) << PAGE_SHIFT;
-repeat:
- hpte_group = ((hash & htab_hash_mask) * HPTES_PER_GROUP) & ~0x7UL;
-
- /* clear the busy bits and set the hash pte bits */
- new_pmd = (new_pmd & ~_PAGE_HPTEFLAGS) | _PAGE_HASHPTE;
+ new_pmd |= _PAGE_HASHPTE;
/* Add in WIMG bits */
rflags |= (new_pmd & (_PAGE_WRITETHRU | _PAGE_NO_CACHE |
* enable the memory coherence always
*/
rflags |= HPTE_R_M;
+repeat:
+ hpte_group = ((hash & htab_hash_mask) * HPTES_PER_GROUP) & ~0x7UL;
/* Insert into the hash table, primary slot */
slot = ppc_md.hpte_insert(hpte_group, vpn, pa, rflags, 0,
mark_hpte_slot_valid(hpte_slot_array, index, slot);
}
/*
- * No need to use ldarx/stdcx here
+ * Mark the pte with _PAGE_COMBO, if we are trying to hash it with
+ * base page size 4k.
+ */
+ if (psize == MMU_PAGE_4K)
+ new_pmd |= _PAGE_COMBO;
+ /*
+ * The hpte valid is stored in the pgtable whose address is in the
+ * second half of the PMD. Order this against clearing of the busy bit in
+ * huge pmd.
*/
+ smp_wmb();
*pmdp = __pmd(new_pmd & ~_PAGE_BUSY);
return 0;
}
void __init do_init_bootmem(void)
{
- int nid;
+ int nid, cpu;
min_low_pfn = 0;
max_low_pfn = memblock_end_of_DRAM() >> PAGE_SHIFT;
reset_numa_cpu_lookup_table();
register_cpu_notifier(&ppc64_numa_nb);
- cpu_numa_callback(&ppc64_numa_nb, CPU_UP_PREPARE,
- (void *)(unsigned long)boot_cpuid);
+ /*
+ * We need the numa_cpu_lookup_table to be accurate for all CPUs,
+ * even before we online them, so that we can use cpu_to_{node,mem}
+ * early in boot, cf. smp_prepare_cpus().
+ */
+ for_each_possible_cpu(cpu) {
+ cpu_numa_callback(&ppc64_numa_nb, CPU_UP_PREPARE,
+ (void *)(unsigned long)cpu);
+ }
}
void __init paging_init(void)
#include "mmu_decl.h"
+#define CREATE_TRACE_POINTS
+#include <trace/events/thp.h>
+
/* Some sanity checking */
#if TASK_SIZE_USER64 > PGTABLE_RANGE
#error TASK_SIZE_USER64 exceeds pagetable range
old = pmd_val(*pmdp);
*pmdp = __pmd((old & ~clr) | set);
#endif
+ trace_hugepage_update(addr, old, clr, set);
if (old & _PAGE_HASHPTE)
- hpte_do_hugepage_flush(mm, addr, pmdp);
+ hpte_do_hugepage_flush(mm, addr, pmdp, old);
return old;
}
* If we didn't had the splitting flag set, go and flush the
* HPTE entries.
*/
+ trace_hugepage_splitting(address, old);
if (!(old & _PAGE_SPLITTING)) {
/* We need to flush the hpte */
if (old & _PAGE_HASHPTE)
- hpte_do_hugepage_flush(vma->vm_mm, address, pmdp);
+ hpte_do_hugepage_flush(vma->vm_mm, address, pmdp, old);
}
/*
* This ensures that generic code that rely on IRQ disabling
assert_spin_locked(&mm->page_table_lock);
WARN_ON(!pmd_trans_huge(pmd));
#endif
+ trace_hugepage_set_pmd(addr, pmd);
return set_pte_at(mm, addr, pmdp_ptep(pmdp), pmd_pte(pmd));
}
* neesd to be flushed.
*/
void hpte_do_hugepage_flush(struct mm_struct *mm, unsigned long addr,
- pmd_t *pmdp)
+ pmd_t *pmdp, unsigned long old_pmd)
{
int ssize, i;
unsigned long s_addr;
if (!hpte_slot_array)
return;
- /* get the base page size */
+ /* get the base page size,vsid and segment size */
+#ifdef CONFIG_DEBUG_VM
psize = get_slice_psize(mm, s_addr);
+ BUG_ON(psize == MMU_PAGE_16M);
+#endif
+ if (old_pmd & _PAGE_COMBO)
+ psize = MMU_PAGE_4K;
+ else
+ psize = MMU_PAGE_64K;
+
+ if (!is_kernel_addr(s_addr)) {
+ ssize = user_segment_size(s_addr);
+ vsid = get_vsid(mm->context.id, s_addr, ssize);
+ WARN_ON(vsid == 0);
+ } else {
+ vsid = get_kernel_vsid(s_addr, mmu_kernel_ssize);
+ ssize = mmu_kernel_ssize;
+ }
if (ppc_md.hugepage_invalidate)
- return ppc_md.hugepage_invalidate(mm, hpte_slot_array,
- s_addr, psize);
+ return ppc_md.hugepage_invalidate(vsid, s_addr,
+ hpte_slot_array,
+ psize, ssize);
/*
* No bluk hpte removal support, invalidate each entry
*/
/* get the vpn */
addr = s_addr + (i * (1ul << shift));
- if (!is_kernel_addr(addr)) {
- ssize = user_segment_size(addr);
- vsid = get_vsid(mm->context.id, addr, ssize);
- WARN_ON(vsid == 0);
- } else {
- vsid = get_kernel_vsid(addr, mmu_kernel_ssize);
- ssize = mmu_kernel_ssize;
- }
-
vpn = hpt_vpn(addr, vsid, ssize);
hash = hpt_hash(vpn, shift, ssize);
if (hidx & _PTEIDX_SECONDARY)
#include <asm/tlb.h>
#include <asm/bug.h>
+#include <trace/events/thp.h>
+
DEFINE_PER_CPU(struct ppc64_tlb_batch, ppc64_tlb_batch);
/*
if (ptep == NULL)
continue;
pte = pte_val(*ptep);
+ if (hugepage_shift)
+ trace_hugepage_invalidate(start, pte_val(pte));
if (!(pte & _PAGE_HASHPTE))
continue;
if (unlikely(hugepage_shift && pmd_trans_huge(*(pmd_t *)pte)))
- hpte_do_hugepage_flush(mm, start, (pmd_t *)pte);
+ hpte_do_hugepage_flush(mm, start, (pmd_t *)ptep, pte);
else
hpte_need_flush(mm, start, ptep, pte, 0);
}
/*
* Early initialization of the MMU TLB code
*/
-static void __early_init_mmu(int boot_cpu)
+static void early_init_this_mmu(void)
{
unsigned int mas4;
- /* XXX This will have to be decided at runtime, but right
- * now our boot and TLB miss code hard wires it. Ideally
- * we should find out a suitable page size and patch the
- * TLB miss code (either that or use the PACA to store
- * the value we want)
- */
- mmu_linear_psize = MMU_PAGE_1G;
-
- /* XXX This should be decided at runtime based on supported
- * page sizes in the TLB, but for now let's assume 16M is
- * always there and a good fit (which it probably is)
- *
- * Freescale booke only supports 4K pages in TLB0, so use that.
- */
- if (mmu_has_feature(MMU_FTR_TYPE_FSL_E))
- mmu_vmemmap_psize = MMU_PAGE_4K;
- else
- mmu_vmemmap_psize = MMU_PAGE_16M;
-
- /* XXX This code only checks for TLB 0 capabilities and doesn't
- * check what page size combos are supported by the HW. It
- * also doesn't handle the case where a separate array holds
- * the IND entries from the array loaded by the PT.
- */
- if (boot_cpu) {
- /* Look for supported page sizes */
- setup_page_sizes();
-
- /* Look for HW tablewalk support */
- setup_mmu_htw();
- }
-
/* Set MAS4 based on page table setting */
mas4 = 0x4 << MAS4_WIMGED_SHIFT;
}
mtspr(SPRN_MAS4, mas4);
- /* Set the global containing the top of the linear mapping
- * for use by the TLB miss code
- */
- linear_map_top = memblock_end_of_DRAM();
-
#ifdef CONFIG_PPC_FSL_BOOK3E
if (mmu_has_feature(MMU_FTR_TYPE_FSL_E)) {
unsigned int num_cams;
/* use a quarter of the TLBCAM for bolted linear map */
num_cams = (mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY) / 4;
linear_map_top = map_mem_in_cams(linear_map_top, num_cams);
+ }
+#endif
- /* limit memory so we dont have linear faults */
- memblock_enforce_memory_limit(linear_map_top);
+ /* A sync won't hurt us after mucking around with
+ * the MMU configuration
+ */
+ mb();
+}
+static void __init early_init_mmu_global(void)
+{
+ /* XXX This will have to be decided at runtime, but right
+ * now our boot and TLB miss code hard wires it. Ideally
+ * we should find out a suitable page size and patch the
+ * TLB miss code (either that or use the PACA to store
+ * the value we want)
+ */
+ mmu_linear_psize = MMU_PAGE_1G;
+
+ /* XXX This should be decided at runtime based on supported
+ * page sizes in the TLB, but for now let's assume 16M is
+ * always there and a good fit (which it probably is)
+ *
+ * Freescale booke only supports 4K pages in TLB0, so use that.
+ */
+ if (mmu_has_feature(MMU_FTR_TYPE_FSL_E))
+ mmu_vmemmap_psize = MMU_PAGE_4K;
+ else
+ mmu_vmemmap_psize = MMU_PAGE_16M;
+
+ /* XXX This code only checks for TLB 0 capabilities and doesn't
+ * check what page size combos are supported by the HW. It
+ * also doesn't handle the case where a separate array holds
+ * the IND entries from the array loaded by the PT.
+ */
+ /* Look for supported page sizes */
+ setup_page_sizes();
+
+ /* Look for HW tablewalk support */
+ setup_mmu_htw();
+
+#ifdef CONFIG_PPC_FSL_BOOK3E
+ if (mmu_has_feature(MMU_FTR_TYPE_FSL_E)) {
if (book3e_htw_mode == PPC_HTW_NONE) {
extlb_level_exc = EX_TLB_SIZE;
patch_exception(0x1c0, exc_data_tlb_miss_bolted_book3e);
}
#endif
- /* A sync won't hurt us after mucking around with
- * the MMU configuration
+ /* Set the global containing the top of the linear mapping
+ * for use by the TLB miss code
*/
- mb();
+ linear_map_top = memblock_end_of_DRAM();
+}
+
+static void __init early_mmu_set_memory_limit(void)
+{
+#ifdef CONFIG_PPC_FSL_BOOK3E
+ if (mmu_has_feature(MMU_FTR_TYPE_FSL_E)) {
+ /*
+ * Limit memory so we dont have linear faults.
+ * Unlike memblock_set_current_limit, which limits
+ * memory available during early boot, this permanently
+ * reduces the memory available to Linux. We need to
+ * do this because highmem is not supported on 64-bit.
+ */
+ memblock_enforce_memory_limit(linear_map_top);
+ }
+#endif
memblock_set_current_limit(linear_map_top);
}
+/* boot cpu only */
void __init early_init_mmu(void)
{
- __early_init_mmu(1);
+ early_init_mmu_global();
+ early_init_this_mmu();
+ early_mmu_set_memory_limit();
}
void early_init_mmu_secondary(void)
{
- __early_init_mmu(0);
+ early_init_this_mmu();
}
void setup_initial_memory_limit(phys_addr_t first_memblock_base,
pr_err("h_get_24x7_catalog_page(ver=%lld, page=%lld) failed:"
" rc=%ld\n",
catalog_version_num, page_offset, hret);
- kfree(page);
+ kmem_cache_free(hv_page_cache, page);
pr_devel("catalog_read: offset=%lld(%lld) count=%zu(%zu) catalog_len=%zu(%zu) => %zd\n",
offset, page_offset, count, page_count, catalog_len,
/* Enable GMAC for now for PCI probing. It will be disabled
* later on after PCI probe
*/
- np = of_find_node_by_name(NULL, "ethernet");
- while(np) {
+ for_each_node_by_name(np, "ethernet")
if (of_device_is_compatible(np, "K2-GMAC"))
g5_gmac_enable(np, 0, 1);
- np = of_find_node_by_name(np, "ethernet");
- }
/* Enable FW before PCI probe. Will be disabled later on
* Note: We should have a batter way to check that we are
* dealing with uninorth internal cell and not a PCI cell
* on the external PCI. The code below works though.
*/
- np = of_find_node_by_name(NULL, "firewire");
- while(np) {
+ for_each_node_by_name(np, "firewire") {
if (of_device_is_compatible(np, "pci106b,5811")) {
macio_chips[0].flags |= MACIO_FLAG_FW_SUPPORTED;
g5_fw_enable(np, 0, 1);
}
- np = of_find_node_by_name(np, "firewire");
}
}
#else /* CONFIG_PPC64 */
/* Enable GMAC for now for PCI probing. It will be disabled
* later on after PCI probe
*/
- np = of_find_node_by_name(NULL, "ethernet");
- while(np) {
+ for_each_node_by_name(np, "ethernet") {
if (np->parent
&& of_device_is_compatible(np->parent, "uni-north")
&& of_device_is_compatible(np, "gmac"))
core99_gmac_enable(np, 0, 1);
- np = of_find_node_by_name(np, "ethernet");
}
/* Enable FW before PCI probe. Will be disabled later on
* dealing with uninorth internal cell and not a PCI cell
* on the external PCI. The code below works though.
*/
- np = of_find_node_by_name(NULL, "firewire");
- while(np) {
+ for_each_node_by_name(np, "firewire") {
if (np->parent
&& of_device_is_compatible(np->parent, "uni-north")
&& (of_device_is_compatible(np, "pci106b,18") ||
macio_chips[0].flags |= MACIO_FLAG_FW_SUPPORTED;
core99_firewire_enable(np, 0, 1);
}
- np = of_find_node_by_name(np, "firewire");
}
/* Enable ATA-100 before PCI probe. */
np = of_find_node_by_name(NULL, "ata-6");
- while(np) {
+ for_each_node_by_name(np, "ata-6") {
if (np->parent
&& of_device_is_compatible(np->parent, "uni-north")
&& of_device_is_compatible(np, "kauai-ata")) {
core99_ata100_enable(np, 1);
}
- np = of_find_node_by_name(np, "ata-6");
}
/* Switch airport off */
{
struct device_node *nec;
- for (nec = NULL; (nec = of_find_node_by_name(nec, "usb")) != NULL;) {
+ for_each_node_by_name(nec, "usb") {
struct pci_controller *hose;
u32 data;
const u32 *prop;
int ok;
/* Look for the clock chip */
- while ((cc = of_find_node_by_name(cc, "i2c-hwclock")) != NULL) {
+ for_each_node_by_name(cc, "i2c-hwclock") {
p = of_get_parent(cc);
ok = p && of_device_is_compatible(p, "uni-n-i2c");
of_node_put(p);
* of type "adb". If not, we return a failure, but we keep the
* bext output set for now
*/
- for (np = NULL; (np = of_find_node_by_name(np, "keyboard")) != NULL;) {
+ for_each_node_by_name(np, "keyboard") {
struct device_node *parent = of_get_parent(np);
int found = (parent && strcmp(parent->type, "adb") == 0);
of_node_put(parent);
OPAL_CALL(opal_get_param, OPAL_GET_PARAM);
OPAL_CALL(opal_set_param, OPAL_SET_PARAM);
OPAL_CALL(opal_handle_hmi, OPAL_HANDLE_HMI);
+OPAL_CALL(opal_register_dump_region, OPAL_REGISTER_DUMP_REGION);
+OPAL_CALL(opal_unregister_dump_region, OPAL_UNREGISTER_DUMP_REGION);
return 0;
}
+static void __init opal_dump_region_init(void)
+{
+ void *addr;
+ uint64_t size;
+ int rc;
+
+ /* Register kernel log buffer */
+ addr = log_buf_addr_get();
+ size = log_buf_len_get();
+ rc = opal_register_dump_region(OPAL_DUMP_REGION_LOG_BUF,
+ __pa(addr), size);
+ /* Don't warn if this is just an older OPAL that doesn't
+ * know about that call
+ */
+ if (rc && rc != OPAL_UNSUPPORTED)
+ pr_warn("DUMP: Failed to register kernel log buffer. "
+ "rc = %d\n", rc);
+}
static int __init opal_init(void)
{
struct device_node *np, *consoles;
/* Create "opal" kobject under /sys/firmware */
rc = opal_sysfs_init();
if (rc == 0) {
+ /* Setup dump region interface */
+ opal_dump_region_init();
/* Setup error log interface */
rc = opal_elog_init();
/* Setup code update interface */
else
mdelay(10);
}
+
+ /* Unregister memory dump region */
+ opal_unregister_dump_region(OPAL_DUMP_REGION_LOG_BUF);
}
/* Export this so that test modules can use it */
pe = &phb->ioda.pe_array[pdn->pe_number];
WARN_ON(get_dma_ops(&pdev->dev) != &dma_iommu_ops);
- set_iommu_table_base(&pdev->dev, &pe->tce32_table);
+ set_iommu_table_base_and_group(&pdev->dev, &pe->tce32_table);
}
static int pnv_pci_ioda_dma_set_mask(struct pnv_phb *phb,
}
static inline int pseries_remove_mem_node(struct device_node *np)
{
- return -EOPNOTSUPP;
+ return 0;
}
#endif /* CONFIG_MEMORY_HOTREMOVE */
if (!memblock_size)
return -EINVAL;
- p = (u32 *)of_get_property(pr->dn, "ibm,dynamic-memory", NULL);
+ p = (u32 *) pr->old_prop->value;
if (!p)
return -EINVAL;
return retval;
}
- last_p_partition_ID = pi_buff[0];
- last_p_unit_address = pi_buff[1];
+ last_p_partition_ID = be64_to_cpu(pi_buff[0]);
+ last_p_unit_address = be64_to_cpu(pi_buff[1]);
/* This indicates that there are no further partners */
if (last_p_partition_ID == ~0UL
early_param("disable_ddw", disable_ddw_setup);
-static void remove_ddw(struct device_node *np)
+static void remove_ddw(struct device_node *np, bool remove_prop)
{
struct dynamic_dma_window_prop *dwp;
struct property *win64;
const u32 *ddw_avail;
u64 liobn;
- int len, ret;
+ int len, ret = 0;
ddw_avail = of_get_property(np, "ibm,ddw-applicable", &len);
win64 = of_find_property(np, DIRECT64_PROPNAME, NULL);
np->full_name, ret, ddw_avail[2], liobn);
delprop:
- ret = of_remove_property(np, win64);
+ if (remove_prop)
+ ret = of_remove_property(np, win64);
if (ret)
pr_warning("%s: failed to remove direct window property: %d\n",
np->full_name, ret);
window = kzalloc(sizeof(*window), GFP_KERNEL);
if (!window || len < sizeof(struct dynamic_dma_window_prop)) {
kfree(window);
- remove_ddw(pdn);
+ remove_ddw(pdn, true);
continue;
}
kfree(window);
out_clear_window:
- remove_ddw(pdn);
+ remove_ddw(pdn, true);
out_free_prop:
kfree(win64->name);
switch (action) {
case OF_RECONFIG_DETACH_NODE:
- remove_ddw(np);
+ /*
+ * Removing the property will invoke the reconfig
+ * notifier again, which causes dead-lock on the
+ * read-write semaphore of the notifier chain. So
+ * we have to remove the property when releasing
+ * the device node.
+ */
+ remove_ddw(np, false);
if (pci && pci->iommu_table)
iommu_free_table(pci->iommu_table, np->full_name);
spin_unlock_irqrestore(&pSeries_lpar_tlbie_lock, flags);
}
-static void pSeries_lpar_hugepage_invalidate(struct mm_struct *mm,
- unsigned char *hpte_slot_array,
- unsigned long addr, int psize)
+static void pSeries_lpar_hugepage_invalidate(unsigned long vsid,
+ unsigned long addr,
+ unsigned char *hpte_slot_array,
+ int psize, int ssize)
{
- int ssize = 0, i, index = 0;
+ int i, index = 0;
unsigned long s_addr = addr;
unsigned int max_hpte_count, valid;
unsigned long vpn_array[PPC64_HUGE_HPTE_BATCH];
unsigned long slot_array[PPC64_HUGE_HPTE_BATCH];
- unsigned long shift, hidx, vpn = 0, vsid, hash, slot;
+ unsigned long shift, hidx, vpn = 0, hash, slot;
shift = mmu_psize_defs[psize].shift;
max_hpte_count = 1U << (PMD_SHIFT - shift);
/* get the vpn */
addr = s_addr + (i * (1ul << shift));
- if (!is_kernel_addr(addr)) {
- ssize = user_segment_size(addr);
- vsid = get_vsid(mm->context.id, addr, ssize);
- WARN_ON(vsid == 0);
- } else {
- vsid = get_kernel_vsid(addr, mmu_kernel_ssize);
- ssize = mmu_kernel_ssize;
- }
-
vpn = hpt_vpn(addr, vsid, ssize);
hash = hpt_hash(vpn, shift, ssize);
if (hidx & _PTEIDX_SECONDARY)
struct device_node *np;
const char *typep;
- for (np = NULL; (np = of_find_node_by_name(np,
- "interrupt-controller"));) {
+ for_each_node_by_name(np, "interrupt-controller") {
typep = of_get_property(np, "compatible", NULL);
if (strstr(typep, "open-pic")) {
pSeries_mpic_node = of_node_get(np);
#include <linux/interrupt.h>
#include <linux/irq.h>
#include <linux/bug.h>
+#include <linux/nmi.h>
#include <asm/ptrace.h>
#include <asm/string.h>
#endif
local_irq_save(flags);
+ hard_irq_disable();
bp = in_breakpoint_table(regs->nip, &offset);
if (bp != NULL) {
#endif
insert_cpu_bpts();
+ touch_nmi_watchdog();
local_irq_restore(flags);
return cmd != 'X' && cmd != EOF;
#define __NR_renameat2 345
#define __NR_seccomp 346
#define __NR_getrandom 347
+#define __NR_memfd_create 348
-#define NR_syscalls 348
+#define NR_syscalls 349
/* Bitmask values returned from kern_features system call. */
#define KERN_FEATURE_MIXED_MODE_STACK 0x00000001
node->full_name);
return;
}
+
+ if (ofpci_verbose)
+ printk(" Bridge bus range [%u --> %u]\n",
+ busrange[0], busrange[1]);
+
ranges = of_get_property(node, "ranges", &len);
simba = 0;
if (ranges == NULL) {
pci_bus_insert_busn_res(bus, busrange[0], busrange[1]);
bus->bridge_ctl = 0;
+ if (ofpci_verbose)
+ printk(" Bridge ranges[%p] simba[%d]\n",
+ ranges, simba);
+
/* parse ranges property, or cook one up by hand for Simba */
/* PCI #address-cells == 3 and #size-cells == 2 always */
res = &dev->resource[PCI_BRIDGE_RESOURCES];
}
i = 1;
for (; len >= 32; len -= 32, ranges += 8) {
+ u64 start;
+
+ if (ofpci_verbose)
+ printk(" RAW Range[%08x:%08x:%08x:%08x:%08x:%08x:"
+ "%08x:%08x]\n",
+ ranges[0], ranges[1], ranges[2], ranges[3],
+ ranges[4], ranges[5], ranges[6], ranges[7]);
+
flags = pci_parse_of_flags(ranges[0]);
size = GET_64BIT(ranges, 6);
if (flags == 0 || size == 0)
continue;
+
+ /* On PCI-Express systems, PCI bridges that have no devices downstream
+ * have a bogus size value where the first 32-bit cell is 0xffffffff.
+ * This results in a bogus range where start + size overflows.
+ *
+ * Just skip these otherwise the kernel will complain when the resource
+ * tries to be claimed.
+ */
+ if (size >> 32 == 0xffffffff)
+ continue;
+
if (flags & IORESOURCE_IO) {
res = bus->resource[0];
if (res->flags) {
}
res->flags = flags;
- region.start = GET_64BIT(ranges, 1);
+ region.start = start = GET_64BIT(ranges, 1);
region.end = region.start + size - 1;
+
+ if (ofpci_verbose)
+ printk(" Using flags[%08x] start[%016llx] size[%016llx]\n",
+ flags, start, size);
+
pcibios_bus_to_resource(dev->bus, res, ®ion);
}
after_ranges:
pci_bus_register_of_sysfs(child_bus);
}
+static void pci_claim_bus_resources(struct pci_bus *bus)
+{
+ struct pci_bus *child_bus;
+ struct pci_dev *dev;
+
+ list_for_each_entry(dev, &bus->devices, bus_list) {
+ int i;
+
+ for (i = 0; i < PCI_NUM_RESOURCES; i++) {
+ struct resource *r = &dev->resource[i];
+
+ if (r->parent || !r->start || !r->flags)
+ continue;
+
+ if (ofpci_verbose)
+ printk("PCI: Claiming %s: "
+ "Resource %d: %016llx..%016llx [%x]\n",
+ pci_name(dev), i,
+ (unsigned long long)r->start,
+ (unsigned long long)r->end,
+ (unsigned int)r->flags);
+
+ pci_claim_resource(dev, i);
+ }
+ }
+
+ list_for_each_entry(child_bus, &bus->children, node)
+ pci_claim_bus_resources(child_bus);
+}
+
struct pci_bus *pci_scan_one_pbm(struct pci_pbm_info *pbm,
struct device *parent)
{
pci_bus_add_devices(bus);
pci_bus_register_of_sysfs(bus);
+ pci_claim_bus_resources(bus);
+
return bus;
}
/*330*/ .long sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime
/*335*/ .long sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev
/*340*/ .long sys_ni_syscall, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr
-/*345*/ .long sys_renameat2, sys_seccomp, sys_getrandom
+/*345*/ .long sys_renameat2, sys_seccomp, sys_getrandom, sys_memfd_create
/*330*/ .word compat_sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, compat_sys_open_by_handle_at, compat_sys_clock_adjtime
.word sys_syncfs, compat_sys_sendmmsg, sys_setns, compat_sys_process_vm_readv, compat_sys_process_vm_writev
/*340*/ .word sys_kern_features, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr
- .word sys32_renameat2, sys_seccomp, sys_getrandom
+ .word sys32_renameat2, sys_seccomp, sys_getrandom, sys_memfd_create
#endif /* CONFIG_COMPAT */
/*330*/ .word sys_fanotify_mark, sys_prlimit64, sys_name_to_handle_at, sys_open_by_handle_at, sys_clock_adjtime
.word sys_syncfs, sys_sendmmsg, sys_setns, sys_process_vm_readv, sys_process_vm_writev
/*340*/ .word sys_kern_features, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr
- .word sys_renameat2, sys_seccomp, sys_getrandom
+ .word sys_renameat2, sys_seccomp, sys_getrandom, sys_memfd_create
#define KVM_REFILL_PAGES 25
#define KVM_MAX_CPUID_ENTRIES 80
#define KVM_NR_FIXED_MTRR_REGION 88
-#define KVM_NR_VAR_MTRR 10
+#define KVM_NR_VAR_MTRR 8
#define ASYNC_PF_PER_VCPU 64
#define MSR_CORE_C1_RES 0x00000660
+#define MSR_CC6_DEMOTION_POLICY_CONFIG 0x00000668
+#define MSR_MC6_DEMOTION_POLICY_CONFIG 0x00000669
+
#define MSR_AMD64_MC0_MASK 0xc0010044
#define MSR_IA32_MCx_CTL(x) (MSR_IA32_MC0_CTL + 4*(x))
NULL,
};
-static DEFINE_PCI_DEVICE_TABLE(snbep_uncore_pci_ids) = {
+static const struct pci_device_id snbep_uncore_pci_ids[] = {
{ /* Home Agent */
PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_UNC_HA),
.driver_data = UNCORE_PCI_DEV_DATA(SNBEP_PCI_UNCORE_HA, 0),
NULL,
};
-static DEFINE_PCI_DEVICE_TABLE(ivt_uncore_pci_ids) = {
+static const struct pci_device_id ivt_uncore_pci_ids[] = {
{ /* Home Agent 0 */
PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0xe30),
.driver_data = UNCORE_PCI_DEV_DATA(IVT_PCI_UNCORE_HA, 0),
NULL,
};
-static DEFINE_PCI_DEVICE_TABLE(snb_uncore_pci_ids) = {
+static const struct pci_device_id snb_uncore_pci_ids[] = {
{ /* IMC */
PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SNB_IMC),
.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
{ /* end: all zeroes */ },
};
-static DEFINE_PCI_DEVICE_TABLE(ivb_uncore_pci_ids) = {
+static const struct pci_device_id ivb_uncore_pci_ids[] = {
{ /* IMC */
PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IVB_IMC),
.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
{ /* end: all zeroes */ },
};
-static DEFINE_PCI_DEVICE_TABLE(hsw_uncore_pci_ids) = {
+static const struct pci_device_id hsw_uncore_pci_ids[] = {
{ /* IMC */
PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_HSW_IMC),
.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
return 0;
}
-static DEFINE_PCI_DEVICE_TABLE(iosf_mbi_pci_ids) = {
+static const struct pci_device_id iosf_mbi_pci_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_BAYTRAIL) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_QUARK_X1000) },
{ 0, },
goto exception;
break;
case VCPU_SREG_CS:
- if (in_task_switch && rpl != dpl)
- goto exception;
-
if (!(seg_desc.type & 8))
goto exception;
ctxt->execute = opcode.u.execute;
+ if (unlikely(ctxt->ud) && likely(!(ctxt->d & EmulateOnUD)))
+ return EMULATION_FAILED;
+
if (unlikely(ctxt->d &
- (NotImpl|EmulateOnUD|Stack|Op3264|Sse|Mmx|Intercept|CheckPerm))) {
+ (NotImpl|Stack|Op3264|Sse|Mmx|Intercept|CheckPerm))) {
/*
* These are copied unconditionally here, and checked unconditionally
* in x86_emulate_insn.
if (ctxt->d & NotImpl)
return EMULATION_FAILED;
- if (!(ctxt->d & EmulateOnUD) && ctxt->ud)
- return EMULATION_FAILED;
-
if (mode == X86EMUL_MODE_PROT64 && (ctxt->d & Stack))
ctxt->op_bytes = 8;
{
struct page **pages;
xen_pfn_t *pfns;
+ void *vaddr;
int rc;
unsigned int i;
unsigned long nr_grant_frames = gnttab_max_grant_frames();
for (i = 0; i < nr_grant_frames; i++)
pfns[i] = page_to_pfn(pages[i]);
- rc = arch_gnttab_map_shared(pfns, nr_grant_frames, nr_grant_frames,
- &xen_auto_xlat_grant_frames.vaddr);
-
- if (rc) {
+ vaddr = vmap(pages, nr_grant_frames, 0, PAGE_KERNEL);
+ if (!vaddr) {
pr_warn("%s Couldn't map %ld pfns rc:%d\n", __func__,
nr_grant_frames, rc);
free_xenballooned_pages(nr_grant_frames, pages);
kfree(pages);
kfree(pfns);
- return rc;
+ return -ENOMEM;
}
kfree(pages);
xen_auto_xlat_grant_frames.pfn = pfns;
xen_auto_xlat_grant_frames.count = nr_grant_frames;
+ xen_auto_xlat_grant_frames.vaddr = vaddr;
return 0;
}
irq = bind_virq_to_irqhandler(VIRQ_TIMER, cpu, xen_timer_interrupt,
IRQF_PERCPU|IRQF_NOBALANCING|IRQF_TIMER|
- IRQF_FORCE_RESUME,
+ IRQF_FORCE_RESUME|IRQF_EARLY_RESUME,
name, NULL);
(void)xen_set_irq_priority(irq, XEN_IRQ_PRIORITY_MAX);
bs->bvec_integrity_pool);
if (!bip->bip_vec)
goto err;
+ bip->bip_max_vcnt = bvec_nr_vecs(idx);
} else {
bip->bip_vec = bip->bip_inline_vecs;
+ bip->bip_max_vcnt = inline_vecs;
}
bip->bip_slab = idx;
}
EXPORT_SYMBOL(bio_integrity_free);
-static inline unsigned int bip_integrity_vecs(struct bio_integrity_payload *bip)
-{
- if (bip->bip_slab == BIO_POOL_NONE)
- return BIP_INLINE_VECS;
-
- return bvec_nr_vecs(bip->bip_slab);
-}
-
/**
* bio_integrity_add_page - Attach integrity metadata
* @bio: bio to update
struct bio_integrity_payload *bip = bio->bi_integrity;
struct bio_vec *iv;
- if (bip->bip_vcnt >= bip_integrity_vecs(bip)) {
+ if (bip->bip_vcnt >= bip->bip_max_vcnt) {
printk(KERN_ERR "%s: bip_vec full\n", __func__);
return 0;
}
bslab = &bio_slabs[entry];
snprintf(bslab->name, sizeof(bslab->name), "bio-%d", entry);
- slab = kmem_cache_create(bslab->name, sz, 0, SLAB_HWCACHE_ALIGN, NULL);
+ slab = kmem_cache_create(bslab->name, sz, ARCH_KMALLOC_MINALIGN,
+ SLAB_HWCACHE_ALIGN, NULL);
if (!slab)
goto out_unlock;
*/
void blk_queue_bypass_start(struct request_queue *q)
{
- bool drain;
-
spin_lock_irq(q->queue_lock);
- drain = !q->bypass_depth++;
+ q->bypass_depth++;
queue_flag_set(QUEUE_FLAG_BYPASS, q);
spin_unlock_irq(q->queue_lock);
- if (drain) {
+ /*
+ * Queues start drained. Skip actual draining till init is
+ * complete. This avoids lenghty delays during queue init which
+ * can happen many times during boot.
+ */
+ if (blk_queue_init_done(q)) {
spin_lock_irq(q->queue_lock);
__blk_drain_queue(q, false);
spin_unlock_irq(q->queue_lock);
* prevent that q->request_fn() gets invoked after draining finished.
*/
if (q->mq_ops) {
- blk_mq_drain_queue(q);
+ blk_mq_freeze_queue(q);
spin_lock_irq(lock);
} else {
spin_lock_irq(lock);
static int blk_mq_queue_enter(struct request_queue *q)
{
- int ret;
-
- __percpu_counter_add(&q->mq_usage_counter, 1, 1000000);
- smp_wmb();
-
- /* we have problems freezing the queue if it's initializing */
- if (!blk_queue_dying(q) &&
- (!blk_queue_bypass(q) || !blk_queue_init_done(q)))
- return 0;
-
- __percpu_counter_add(&q->mq_usage_counter, -1, 1000000);
+ while (true) {
+ int ret;
- spin_lock_irq(q->queue_lock);
- ret = wait_event_interruptible_lock_irq(q->mq_freeze_wq,
- !blk_queue_bypass(q) || blk_queue_dying(q),
- *q->queue_lock);
- /* inc usage with lock hold to avoid freeze_queue runs here */
- if (!ret && !blk_queue_dying(q))
- __percpu_counter_add(&q->mq_usage_counter, 1, 1000000);
- else if (blk_queue_dying(q))
- ret = -ENODEV;
- spin_unlock_irq(q->queue_lock);
+ if (percpu_ref_tryget_live(&q->mq_usage_counter))
+ return 0;
- return ret;
+ ret = wait_event_interruptible(q->mq_freeze_wq,
+ !q->mq_freeze_depth || blk_queue_dying(q));
+ if (blk_queue_dying(q))
+ return -ENODEV;
+ if (ret)
+ return ret;
+ }
}
static void blk_mq_queue_exit(struct request_queue *q)
{
- __percpu_counter_add(&q->mq_usage_counter, -1, 1000000);
+ percpu_ref_put(&q->mq_usage_counter);
}
-void blk_mq_drain_queue(struct request_queue *q)
+static void blk_mq_usage_counter_release(struct percpu_ref *ref)
{
- while (true) {
- s64 count;
-
- spin_lock_irq(q->queue_lock);
- count = percpu_counter_sum(&q->mq_usage_counter);
- spin_unlock_irq(q->queue_lock);
+ struct request_queue *q =
+ container_of(ref, struct request_queue, mq_usage_counter);
- if (count == 0)
- break;
- blk_mq_start_hw_queues(q);
- msleep(10);
- }
+ wake_up_all(&q->mq_freeze_wq);
}
/*
* Guarantee no request is in use, so we can change any data structure of
* the queue afterward.
*/
-static void blk_mq_freeze_queue(struct request_queue *q)
+void blk_mq_freeze_queue(struct request_queue *q)
{
- bool drain;
-
spin_lock_irq(q->queue_lock);
- drain = !q->bypass_depth++;
- queue_flag_set(QUEUE_FLAG_BYPASS, q);
+ q->mq_freeze_depth++;
spin_unlock_irq(q->queue_lock);
- if (drain)
- blk_mq_drain_queue(q);
+ percpu_ref_kill(&q->mq_usage_counter);
+ blk_mq_run_queues(q, false);
+ wait_event(q->mq_freeze_wq, percpu_ref_is_zero(&q->mq_usage_counter));
}
static void blk_mq_unfreeze_queue(struct request_queue *q)
bool wake = false;
spin_lock_irq(q->queue_lock);
- if (!--q->bypass_depth) {
- queue_flag_clear(QUEUE_FLAG_BYPASS, q);
- wake = true;
- }
- WARN_ON_ONCE(q->bypass_depth < 0);
+ wake = !--q->mq_freeze_depth;
+ WARN_ON_ONCE(q->mq_freeze_depth < 0);
spin_unlock_irq(q->queue_lock);
- if (wake)
+ if (wake) {
+ percpu_ref_reinit(&q->mq_usage_counter);
wake_up_all(&q->mq_freeze_wq);
+ }
}
bool blk_mq_can_queue(struct blk_mq_hw_ctx *hctx)
if (!q)
goto err_hctxs;
- if (percpu_counter_init(&q->mq_usage_counter, 0))
+ if (percpu_ref_init(&q->mq_usage_counter, blk_mq_usage_counter_release))
goto err_map;
setup_timer(&q->timeout, blk_mq_rq_timer, (unsigned long) q);
blk_mq_exit_hw_queues(q, set, set->nr_hw_queues);
blk_mq_free_hw_queues(q, set);
- percpu_counter_destroy(&q->mq_usage_counter);
+ percpu_ref_exit(&q->mq_usage_counter);
free_percpu(q->queue_ctx);
kfree(q->queue_hw_ctx);
{
blk_mq_cpu_init();
- /* Must be called after percpu_counter_hotcpu_callback() */
- hotcpu_notifier(blk_mq_queue_reinit_notify, -10);
+ hotcpu_notifier(blk_mq_queue_reinit_notify, 0);
return 0;
}
void __blk_mq_complete_request(struct request *rq);
void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async);
void blk_mq_init_flush(struct request_queue *q);
-void blk_mq_drain_queue(struct request_queue *q);
+void blk_mq_freeze_queue(struct request_queue *q);
void blk_mq_free_queue(struct request_queue *q);
void blk_mq_clone_flush_request(struct request *flush_rq,
struct request *orig_rq);
* Initialization must be complete by now. Finish the initial
* bypass from queue allocation.
*/
- blk_queue_bypass_end(q);
queue_flag_set_unlocked(QUEUE_FLAG_INIT_DONE, q);
+ blk_queue_bypass_end(q);
ret = blk_trace_init_sysfs(dev);
if (ret)
fmode_t mode = file->f_mode;
struct backing_dev_info *bdi;
loff_t size;
+ unsigned int max_sectors;
/*
* O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have
case BLKSSZGET: /* get block device hardware sector size */
return compat_put_int(arg, bdev_logical_block_size(bdev));
case BLKSECTGET:
- return compat_put_ushort(arg,
- queue_max_sectors(bdev_get_queue(bdev)));
+ max_sectors = min_t(unsigned int, USHRT_MAX,
+ queue_max_sectors(bdev_get_queue(bdev)));
+ return compat_put_ushort(arg, max_sectors);
case BLKROTATIONAL:
return compat_put_ushort(arg,
!blk_queue_nonrot(bdev_get_queue(bdev)));
struct backing_dev_info *bdi;
loff_t size;
int ret, n;
+ unsigned int max_sectors;
switch(cmd) {
case BLKFLSBUF:
case BLKDISCARDZEROES:
return put_uint(arg, bdev_discard_zeroes_data(bdev));
case BLKSECTGET:
- return put_ushort(arg, queue_max_sectors(bdev_get_queue(bdev)));
+ max_sectors = min_t(unsigned int, USHRT_MAX,
+ queue_max_sectors(bdev_get_queue(bdev)));
+ return put_ushort(arg, max_sectors);
case BLKROTATIONAL:
return put_ushort(arg, !blk_queue_nonrot(bdev_get_queue(bdev)));
case BLKRASET:
numlvs = be16_to_cpu(p->numlvs);
put_dev_sector(sect);
}
- lvip = kzalloc(sizeof(struct lv_info) * state->limit, GFP_KERNEL);
+ lvip = kcalloc(state->limit, sizeof(struct lv_info), GFP_KERNEL);
if (!lvip)
return 0;
if (numlvs && (d = read_part_sector(state, vgda_sector + 1, §))) {
continue;
}
lv_ix = be16_to_cpu(p->lv_ix) - 1;
- if (lv_ix > state->limit) {
+ if (lv_ix >= state->limit) {
cur_lv_ix = -1;
continue;
}
* Re-organised Feb 1998 Russell King
*/
+#define pr_fmt(fmt) fmt
+
#include <linux/types.h>
#include <linux/affs_hardblocks.h>
data = read_part_sector(state, blk, §);
if (!data) {
if (warn_no_part)
- printk("Dev %s: unable to read RDB block %d\n",
+ pr_err("Dev %s: unable to read RDB block %d\n",
bdevname(state->bdev, b), blk);
res = -1;
goto rdb_done;
*(__be32 *)(data+0xdc) = 0;
if (checksum_block((__be32 *)data,
be32_to_cpu(rdb->rdb_SummedLongs) & 0x7F)==0) {
- printk("Warning: Trashed word at 0xd0 in block %d "
- "ignored in checksum calculation\n",blk);
+ pr_err("Trashed word at 0xd0 in block %d ignored in checksum calculation\n",
+ blk);
break;
}
- printk("Dev %s: RDB in block %d has bad checksum\n",
+ pr_err("Dev %s: RDB in block %d has bad checksum\n",
bdevname(state->bdev, b), blk);
}
data = read_part_sector(state, blk, §);
if (!data) {
if (warn_no_part)
- printk("Dev %s: unable to read partition block %d\n",
+ pr_err("Dev %s: unable to read partition block %d\n",
bdevname(state->bdev, b), blk);
res = -1;
goto rdb_done;
/**
* efi_crc32() - EFI version of crc32 function
* @buf: buffer to calculate crc32 of
- * @len - length of buf
+ * @len: length of buf
*
* Description: Returns EFI-style CRC32 value for @buf
*
/**
* read_lba(): Read bytes from disk, starting at given LBA
- * @state
- * @lba
- * @buffer
- * @size_t
+ * @state: disk parsed partitions
+ * @lba: the Logical Block Address of the partition table
+ * @buffer: destination buffer
+ * @count: bytes to read
*
* Description: Reads @count bytes from @state->bdev into @buffer.
* Returns number of bytes read on success, 0 on error.
/**
* alloc_read_gpt_entries(): reads partition entries from disk
- * @state
- * @gpt - GPT header
+ * @state: disk parsed partitions
+ * @gpt: GPT header
*
* Description: Returns ptes on success, NULL on error.
* Allocates space for PTEs based on information found in @gpt.
/**
* alloc_read_gpt_header(): Allocates GPT header, reads into it from disk
- * @state
- * @lba is the Logical Block Address of the partition table
+ * @state: disk parsed partitions
+ * @lba: the Logical Block Address of the partition table
*
* Description: returns GPT header on success, NULL on error. Allocates
* and fills a GPT header starting at @ from @state->bdev.
/**
* is_gpt_valid() - tests one GPT header and PTEs for validity
- * @state
- * @lba is the logical block address of the GPT header to test
- * @gpt is a GPT header ptr, filled on return.
- * @ptes is a PTEs ptr, filled on return.
+ * @state: disk parsed partitions
+ * @lba: logical block address of the GPT header to test
+ * @gpt: GPT header ptr, filled on return.
+ * @ptes: PTEs ptr, filled on return.
*
* Description: returns 1 if valid, 0 on error.
* If valid, returns pointers to newly allocated GPT header and PTEs.
/**
* is_pte_valid() - tests one PTE for validity
- * @pte is the pte to check
- * @lastlba is last lba of the disk
+ * @pte:pte to check
+ * @lastlba: last lba of the disk
*
* Description: returns 1 if valid, 0 on error.
*/
/**
* compare_gpts() - Search disk for valid GPT headers and PTEs
- * @pgpt is the primary GPT header
- * @agpt is the alternate GPT header
- * @lastlba is the last LBA number
+ * @pgpt: primary GPT header
+ * @agpt: alternate GPT header
+ * @lastlba: last LBA number
+ *
* Description: Returns nothing. Sanity checks pgpt and agpt fields
* and prints warnings on discrepancies.
*
/**
* find_valid_gpt() - Search disk for valid GPT headers and PTEs
- * @state
- * @gpt is a GPT header ptr, filled on return.
- * @ptes is a PTEs ptr, filled on return.
+ * @state: disk parsed partitions
+ * @gpt: GPT header ptr, filled on return.
+ * @ptes: PTEs ptr, filled on return.
+ *
* Description: Returns 1 if valid, 0 on error.
* If valid, returns pointers to newly allocated GPT header and PTEs.
* Validity depends on PMBR being valid (or being overridden by the
/**
* efi_partition(struct parsed_partitions *state)
- * @state
+ * @state: disk parsed partitions
*
* Description: called from check.c, if the disk contains GPT
* partitions, sets up partition entries in the kernel.
/*
* First process the data partition(s)
*/
- for (i=0; i<4; i++, p++) {
+ for (i = 0; i < 4; i++, p++) {
sector_t offs, size, next;
+
if (!nr_sects(p) || is_extended_partition(p))
continue;
* It should be a link to the next logical partition.
*/
p -= 4;
- for (i=0; i<4; i++, p++)
+ for (i = 0; i < 4; i++, p++)
if (nr_sects(p) && is_extended_partition(p))
break;
if (i == 4)
return;
}
/* Ensure we can handle previous case of VTOC with 8 entries gracefully */
- max_nparts = le16_to_cpu (v->v_nparts) > 8 ? SOLARIS_X86_NUMSLICE : 8;
- for (i=0; i<max_nparts && state->next<state->limit; i++) {
+ max_nparts = le16_to_cpu(v->v_nparts) > 8 ? SOLARIS_X86_NUMSLICE : 8;
+ for (i = 0; i < max_nparts && state->next < state->limit; i++) {
struct solaris_x86_slice *s = &v->v_slice[i];
char tmp[3 + 10 + 1 + 1];
/* The first sector of a Minix partition can have either
* a secondary MBR describing its subpartitions, or
* the normal boot sector. */
- if (msdos_magic_present (data + 510) &&
+ if (msdos_magic_present(data + 510) &&
SYS_IND(p) == MINIX_PARTITION) { /* subpartition table present */
char tmp[1 + BDEVNAME_SIZE + 10 + 9 + 1];
for (slot = 1 ; slot <= 4 ; slot++, p++) {
sector_t start = start_sect(p)*sector_size;
sector_t size = nr_sects(p)*sector_size;
+
if (!size)
continue;
if (is_extended_partition(p)) {
* sector, although it may not be enough/proper.
*/
sector_t n = 2;
+
n = min(size, max(sector_size, n));
put_partition(state, slot, start, n);
return err;
}
+static int max_sectors_bytes(struct request_queue *q)
+{
+ unsigned int max_sectors = queue_max_sectors(q);
+
+ max_sectors = min_t(unsigned int, max_sectors, INT_MAX >> 9);
+
+ return max_sectors << 9;
+}
+
static int sg_get_reserved_size(struct request_queue *q, int __user *p)
{
- unsigned val = min(q->sg_reserved_size, queue_max_sectors(q) << 9);
+ int val = min_t(int, q->sg_reserved_size, max_sectors_bytes(q));
return put_user(val, p);
}
if (size < 0)
return -EINVAL;
- if (size > (queue_max_sectors(q) << 9))
- size = queue_max_sectors(q) << 9;
- q->sg_reserved_size = size;
+ q->sg_reserved_size = min(size, max_sectors_bytes(q));
return 0;
}
unsigned long start_time;
ssize_t ret = 0;
int writing = 0;
+ int at_head = 0;
struct request *rq;
char sense[SCSI_SENSE_BUFFERSIZE];
struct bio *bio;
case SG_DXFER_FROM_DEV:
break;
}
+ if (hdr->flags & SG_FLAG_Q_AT_HEAD)
+ at_head = 1;
rq = blk_get_request(q, writing ? WRITE : READ, GFP_KERNEL);
if (!rq)
* (if he doesn't check that is his problem).
* N.B. a non-zero SCSI status is _not_ necessarily an error.
*/
- blk_execute_rq(q, bd_disk, rq, 0);
+ blk_execute_rq(q, bd_disk, rq, at_head);
hdr->duration = jiffies_to_msecs(jiffies - start_time);
*/
if (battery->capacity_now > battery->full_charge_capacity
&& battery->full_charge_capacity != ACPI_BATTERY_VALUE_UNKNOWN) {
- battery->capacity_now = battery->full_charge_capacity;
if (battery->capacity_now != battery->design_capacity)
printk_once(KERN_WARNING FW_BUG
"battery: reported current charge level (%d) "
"is higher than reported maximum charge level (%d).\n",
battery->capacity_now, battery->full_charge_capacity);
+ battery->capacity_now = battery->full_charge_capacity;
}
if (test_bit(ACPI_BATTERY_QUIRK_PERCENTAGE_CAPACITY, &battery->flags)
unsigned int cpu = (unsigned long)hcpu;
struct acpi_processor *pr = per_cpu(processors, cpu);
struct acpi_device *device;
+ action &= ~CPU_TASKS_FROZEN;
/*
* CPU_STARTING and CPU_DYING must not sleep. Return here since
unsigned long long sta;
acpi_status status;
- if (device->handler->hotplug.demand_offline && !acpi_force_hot_remove) {
+ if (device->handler && device->handler->hotplug.demand_offline
+ && !acpi_force_hot_remove) {
if (!acpi_scan_is_offline(device, true))
return -EBUSY;
} else {
#include <linux/module.h>
#include <linux/of_device.h>
#include <linux/platform_device.h>
-#include <linux/tegra-powergate.h>
#include <linux/regulator/consumer.h>
+#include <soc/tegra/pmc.h>
#include "ahci.h"
#define SATA_CONFIGURATION_0 0x180
};
static const struct ata_port_info xgene_ahci_port_info = {
- .flags = AHCI_FLAG_COMMON | ATA_FLAG_NCQ,
+ .flags = AHCI_FLAG_COMMON,
.pio_mask = ATA_PIO4,
.udma_mask = ATA_UDMA6,
.port_ops = &xgene_ahci_ops,
/* Configure the host controller */
xgene_ahci_hw_init(hpriv);
- hpriv->flags = AHCI_HFLAG_NO_PMP | AHCI_HFLAG_YES_NCQ;
+ hpriv->flags = AHCI_HFLAG_NO_PMP | AHCI_HFLAG_NO_NCQ;
rc = ahci_platform_init_host(pdev, hpriv, &xgene_ahci_port_info);
if (rc)
{ "Micron_M500*", NULL, ATA_HORKAGE_NO_NCQ_TRIM, },
{ "Crucial_CT???M500SSD*", NULL, ATA_HORKAGE_NO_NCQ_TRIM, },
{ "Micron_M550*", NULL, ATA_HORKAGE_NO_NCQ_TRIM, },
- { "Crucial_CT???M550SSD*", NULL, ATA_HORKAGE_NO_NCQ_TRIM, },
+ { "Crucial_CT*M550SSD*", NULL, ATA_HORKAGE_NO_NCQ_TRIM, },
/*
* Some WD SATA-I drives spin up and down erratically when the link
/*
* pata_s3c_bus_softreset - PATA device software reset
*/
-static unsigned int pata_s3c_bus_softreset(struct ata_port *ap,
+static int pata_s3c_bus_softreset(struct ata_port *ap,
unsigned long deadline)
{
struct ata_ioports *ioaddr = &ap->ioaddr;
* Note: Original code is ata_bus_softreset().
*/
-static unsigned int scc_bus_softreset(struct ata_port *ap, unsigned int devmask,
+static int scc_bus_softreset(struct ata_port *ap, unsigned int devmask,
unsigned long deadline)
{
struct ata_ioports *ioaddr = &ap->ioaddr;
udelay(20);
out_be32(ioaddr->ctl_addr, ap->ctl);
- scc_wait_after_reset(&ap->link, devmask, deadline);
-
- return 0;
+ return scc_wait_after_reset(&ap->link, devmask, deadline);
}
/**
{
struct ata_port *ap = link->ap;
unsigned int slave_possible = ap->flags & ATA_FLAG_SLAVE_POSS;
- unsigned int devmask = 0, err_mask;
+ unsigned int devmask = 0;
+ int rc;
u8 err;
DPRINTK("ENTER\n");
/* issue bus reset */
DPRINTK("about to softreset, devmask=%x\n", devmask);
- err_mask = scc_bus_softreset(ap, devmask, deadline);
- if (err_mask) {
- ata_port_err(ap, "SRST failed (err_mask=0x%x)\n", err_mask);
+ rc = scc_bus_softreset(ap, devmask, deadline);
+ if (rc) {
+ ata_port_err(ap, "SRST failed (err_mask=0x%x)\n", rc);
return -EIO;
}
drbd-y += drbd_main.o drbd_strings.o drbd_nl.o
drbd-y += drbd_interval.o drbd_state.o
drbd-y += drbd_nla.o
+drbd-$(CONFIG_DEBUG_FS) += drbd_debugfs.o
obj-$(CONFIG_BLK_DEV_DRBD) += drbd.o
__be32 context[AL_CONTEXT_PER_TRANSACTION];
};
-struct update_odbm_work {
- struct drbd_work w;
- struct drbd_device *device;
- unsigned int enr;
-};
-
-struct update_al_work {
- struct drbd_work w;
- struct drbd_device *device;
- struct completion event;
- int err;
-};
-
-
-void *drbd_md_get_buffer(struct drbd_device *device)
+void *drbd_md_get_buffer(struct drbd_device *device, const char *intent)
{
int r;
wait_event(device->misc_wait,
- (r = atomic_cmpxchg(&device->md_io_in_use, 0, 1)) == 0 ||
+ (r = atomic_cmpxchg(&device->md_io.in_use, 0, 1)) == 0 ||
device->state.disk <= D_FAILED);
- return r ? NULL : page_address(device->md_io_page);
+ if (r)
+ return NULL;
+
+ device->md_io.current_use = intent;
+ device->md_io.start_jif = jiffies;
+ device->md_io.submit_jif = device->md_io.start_jif - 1;
+ return page_address(device->md_io.page);
}
void drbd_md_put_buffer(struct drbd_device *device)
{
- if (atomic_dec_and_test(&device->md_io_in_use))
+ if (atomic_dec_and_test(&device->md_io.in_use))
wake_up(&device->misc_wait);
}
static int _drbd_md_sync_page_io(struct drbd_device *device,
struct drbd_backing_dev *bdev,
- struct page *page, sector_t sector,
- int rw, int size)
+ sector_t sector, int rw)
{
struct bio *bio;
+ /* we do all our meta data IO in aligned 4k blocks. */
+ const int size = 4096;
int err;
device->md_io.done = 0;
if ((rw & WRITE) && !test_bit(MD_NO_FUA, &device->flags))
rw |= REQ_FUA | REQ_FLUSH;
- rw |= REQ_SYNC;
+ rw |= REQ_SYNC | REQ_NOIDLE;
bio = bio_alloc_drbd(GFP_NOIO);
bio->bi_bdev = bdev->md_bdev;
bio->bi_iter.bi_sector = sector;
err = -EIO;
- if (bio_add_page(bio, page, size, 0) != size)
+ if (bio_add_page(bio, device->md_io.page, size, 0) != size)
goto out;
- bio->bi_private = &device->md_io;
+ bio->bi_private = device;
bio->bi_end_io = drbd_md_io_complete;
bio->bi_rw = rw;
}
bio_get(bio); /* one bio_put() is in the completion handler */
- atomic_inc(&device->md_io_in_use); /* drbd_md_put_buffer() is in the completion handler */
+ atomic_inc(&device->md_io.in_use); /* drbd_md_put_buffer() is in the completion handler */
+ device->md_io.submit_jif = jiffies;
if (drbd_insert_fault(device, (rw & WRITE) ? DRBD_FAULT_MD_WR : DRBD_FAULT_MD_RD))
bio_endio(bio, -EIO);
else
sector_t sector, int rw)
{
int err;
- struct page *iop = device->md_io_page;
-
- D_ASSERT(device, atomic_read(&device->md_io_in_use) == 1);
+ D_ASSERT(device, atomic_read(&device->md_io.in_use) == 1);
BUG_ON(!bdev->md_bdev);
current->comm, current->pid, __func__,
(unsigned long long)sector, (rw & WRITE) ? "WRITE" : "READ");
- /* we do all our meta data IO in aligned 4k blocks. */
- err = _drbd_md_sync_page_io(device, bdev, iop, sector, rw, 4096);
+ err = _drbd_md_sync_page_io(device, bdev, sector, rw);
if (err) {
drbd_err(device, "drbd_md_sync_page_io(,%llus,%s) failed with error %d\n",
(unsigned long long)sector, (rw & WRITE) ? "WRITE" : "READ", err);
return need_transaction;
}
-static int al_write_transaction(struct drbd_device *device, bool delegate);
-
-/* When called through generic_make_request(), we must delegate
- * activity log I/O to the worker thread: a further request
- * submitted via generic_make_request() within the same task
- * would be queued on current->bio_list, and would only start
- * after this function returns (see generic_make_request()).
- *
- * However, if we *are* the worker, we must not delegate to ourselves.
- */
+static int al_write_transaction(struct drbd_device *device);
-/*
- * @delegate: delegate activity log I/O to the worker thread
- */
-void drbd_al_begin_io_commit(struct drbd_device *device, bool delegate)
+void drbd_al_begin_io_commit(struct drbd_device *device)
{
bool locked = false;
- BUG_ON(delegate && current == first_peer_device(device)->connection->worker.task);
-
/* Serialize multiple transactions.
* This uses test_and_set_bit, memory barrier is implicit.
*/
rcu_read_unlock();
if (write_al_updates)
- al_write_transaction(device, delegate);
+ al_write_transaction(device);
spin_lock_irq(&device->al_lock);
/* FIXME
if (err)
/*
* @delegate: delegate activity log I/O to the worker thread
*/
-void drbd_al_begin_io(struct drbd_device *device, struct drbd_interval *i, bool delegate)
+void drbd_al_begin_io(struct drbd_device *device, struct drbd_interval *i)
{
- BUG_ON(delegate && current == first_peer_device(device)->connection->worker.task);
-
if (drbd_al_begin_io_prepare(device, i))
- drbd_al_begin_io_commit(device, delegate);
+ drbd_al_begin_io_commit(device);
}
int drbd_al_begin_io_nonblock(struct drbd_device *device, struct drbd_interval *i)
/* We want all necessary updates for a given request within the same transaction
* We could first check how many updates are *actually* needed,
* and use that instead of the worst-case nr_al_extents */
- if (available_update_slots < nr_al_extents)
- return -EWOULDBLOCK;
+ if (available_update_slots < nr_al_extents) {
+ /* Too many activity log extents are currently "hot".
+ *
+ * If we have accumulated pending changes already,
+ * we made progress.
+ *
+ * If we cannot get even a single pending change through,
+ * stop the fast path until we made some progress,
+ * or requests to "cold" extents could be starved. */
+ if (!al->pending_changes)
+ __set_bit(__LC_STARVING, &device->act_log->flags);
+ return -ENOBUFS;
+ }
/* Is resync active in this area? */
for (enr = first; enr <= last; enr++) {
(AL_EXTENT_SHIFT - BM_BLOCK_SHIFT));
}
-static unsigned int rs_extent_to_bm_page(unsigned int rs_enr)
-{
- return rs_enr >>
- /* bit to page */
- ((PAGE_SHIFT + 3) -
- /* resync extent number to bit */
- (BM_EXT_SHIFT - BM_BLOCK_SHIFT));
-}
-
static sector_t al_tr_number_to_on_disk_sector(struct drbd_device *device)
{
const unsigned int stripes = device->ldev->md.al_stripes;
return device->ldev->md.md_offset + device->ldev->md.al_offset + t;
}
-static int
-_al_write_transaction(struct drbd_device *device)
+int al_write_transaction(struct drbd_device *device)
{
struct al_transaction_on_disk *buffer;
struct lc_element *e;
return -EIO;
}
- buffer = drbd_md_get_buffer(device); /* protects md_io_buffer, al_tr_cycle, ... */
+ /* protects md_io_buffer, al_tr_cycle, ... */
+ buffer = drbd_md_get_buffer(device, __func__);
if (!buffer) {
drbd_err(device, "disk failed while waiting for md_io buffer\n");
put_ldev(device);
return err;
}
-
-static int w_al_write_transaction(struct drbd_work *w, int unused)
-{
- struct update_al_work *aw = container_of(w, struct update_al_work, w);
- struct drbd_device *device = aw->device;
- int err;
-
- err = _al_write_transaction(device);
- aw->err = err;
- complete(&aw->event);
-
- return err != -EIO ? err : 0;
-}
-
-/* Calls from worker context (see w_restart_disk_io()) need to write the
- transaction directly. Others came through generic_make_request(),
- those need to delegate it to the worker. */
-static int al_write_transaction(struct drbd_device *device, bool delegate)
-{
- if (delegate) {
- struct update_al_work al_work;
- init_completion(&al_work.event);
- al_work.w.cb = w_al_write_transaction;
- al_work.device = device;
- drbd_queue_work_front(&first_peer_device(device)->connection->sender_work,
- &al_work.w);
- wait_for_completion(&al_work.event);
- return al_work.err;
- } else
- return _al_write_transaction(device);
-}
-
static int _try_lc_del(struct drbd_device *device, struct lc_element *al_ext)
{
int rv;
return 0;
}
-static int w_update_odbm(struct drbd_work *w, int unused)
-{
- struct update_odbm_work *udw = container_of(w, struct update_odbm_work, w);
- struct drbd_device *device = udw->device;
- struct sib_info sib = { .sib_reason = SIB_SYNC_PROGRESS, };
-
- if (!get_ldev(device)) {
- if (__ratelimit(&drbd_ratelimit_state))
- drbd_warn(device, "Can not update on disk bitmap, local IO disabled.\n");
- kfree(udw);
- return 0;
- }
-
- drbd_bm_write_page(device, rs_extent_to_bm_page(udw->enr));
- put_ldev(device);
-
- kfree(udw);
-
- if (drbd_bm_total_weight(device) <= device->rs_failed) {
- switch (device->state.conn) {
- case C_SYNC_SOURCE: case C_SYNC_TARGET:
- case C_PAUSED_SYNC_S: case C_PAUSED_SYNC_T:
- drbd_resync_finished(device);
- default:
- /* nothing to do */
- break;
- }
- }
- drbd_bcast_event(device, &sib);
-
- return 0;
-}
-
+static const char *drbd_change_sync_fname[] = {
+ [RECORD_RS_FAILED] = "drbd_rs_failed_io",
+ [SET_IN_SYNC] = "drbd_set_in_sync",
+ [SET_OUT_OF_SYNC] = "drbd_set_out_of_sync"
+};
/* ATTENTION. The AL's extents are 4MB each, while the extents in the
* resync LRU-cache are 16MB each.
* The caller of this function has to hold an get_ldev() reference.
*
+ * Adjusts the caching members ->rs_left (success) or ->rs_failed (!success),
+ * potentially pulling in (and recounting the corresponding bits)
+ * this resync extent into the resync extent lru cache.
+ *
+ * Returns whether all bits have been cleared for this resync extent,
+ * precisely: (rs_left <= rs_failed)
+ *
* TODO will be obsoleted once we have a caching lru of the on disk bitmap
*/
-static void drbd_try_clear_on_disk_bm(struct drbd_device *device, sector_t sector,
- int count, int success)
+static bool update_rs_extent(struct drbd_device *device,
+ unsigned int enr, int count,
+ enum update_sync_bits_mode mode)
{
struct lc_element *e;
- struct update_odbm_work *udw;
-
- unsigned int enr;
D_ASSERT(device, atomic_read(&device->local_cnt));
- /* I simply assume that a sector/size pair never crosses
- * a 16 MB extent border. (Currently this is true...) */
- enr = BM_SECT_TO_EXT(sector);
-
- e = lc_get(device->resync, enr);
+ /* When setting out-of-sync bits,
+ * we don't need it cached (lc_find).
+ * But if it is present in the cache,
+ * we should update the cached bit count.
+ * Otherwise, that extent should be in the resync extent lru cache
+ * already -- or we want to pull it in if necessary -- (lc_get),
+ * then update and check rs_left and rs_failed. */
+ if (mode == SET_OUT_OF_SYNC)
+ e = lc_find(device->resync, enr);
+ else
+ e = lc_get(device->resync, enr);
if (e) {
struct bm_extent *ext = lc_entry(e, struct bm_extent, lce);
if (ext->lce.lc_number == enr) {
- if (success)
+ if (mode == SET_IN_SYNC)
ext->rs_left -= count;
+ else if (mode == SET_OUT_OF_SYNC)
+ ext->rs_left += count;
else
ext->rs_failed += count;
if (ext->rs_left < ext->rs_failed) {
- drbd_warn(device, "BAD! sector=%llus enr=%u rs_left=%d "
+ drbd_warn(device, "BAD! enr=%u rs_left=%d "
"rs_failed=%d count=%d cstate=%s\n",
- (unsigned long long)sector,
ext->lce.lc_number, ext->rs_left,
ext->rs_failed, count,
drbd_conn_str(device->state.conn));
ext->lce.lc_number, ext->rs_failed);
}
ext->rs_left = rs_left;
- ext->rs_failed = success ? 0 : count;
+ ext->rs_failed = (mode == RECORD_RS_FAILED) ? count : 0;
/* we don't keep a persistent log of the resync lru,
* we can commit any change right away. */
lc_committed(device->resync);
}
- lc_put(device->resync, &ext->lce);
+ if (mode != SET_OUT_OF_SYNC)
+ lc_put(device->resync, &ext->lce);
/* no race, we are within the al_lock! */
- if (ext->rs_left == ext->rs_failed) {
+ if (ext->rs_left <= ext->rs_failed) {
ext->rs_failed = 0;
-
- udw = kmalloc(sizeof(*udw), GFP_ATOMIC);
- if (udw) {
- udw->enr = ext->lce.lc_number;
- udw->w.cb = w_update_odbm;
- udw->device = device;
- drbd_queue_work_front(&first_peer_device(device)->connection->sender_work,
- &udw->w);
- } else {
- drbd_warn(device, "Could not kmalloc an udw\n");
- }
+ return true;
}
- } else {
+ } else if (mode != SET_OUT_OF_SYNC) {
+ /* be quiet if lc_find() did not find it. */
drbd_err(device, "lc_get() failed! locked=%d/%d flags=%lu\n",
device->resync_locked,
device->resync->nr_elements,
device->resync->flags);
}
+ return false;
}
void drbd_advance_rs_marks(struct drbd_device *device, unsigned long still_to_go)
}
}
-/* clear the bit corresponding to the piece of storage in question:
- * size byte of data starting from sector. Only clear a bits of the affected
- * one ore more _aligned_ BM_BLOCK_SIZE blocks.
- *
- * called by worker on C_SYNC_TARGET and receiver on SyncSource.
- *
- */
-void __drbd_set_in_sync(struct drbd_device *device, sector_t sector, int size,
- const char *file, const unsigned int line)
+/* It is called lazy update, so don't do write-out too often. */
+static bool lazy_bitmap_update_due(struct drbd_device *device)
{
- /* Is called from worker and receiver context _only_ */
- unsigned long sbnr, ebnr, lbnr;
- unsigned long count = 0;
- sector_t esector, nr_sectors;
- int wake_up = 0;
- unsigned long flags;
+ return time_after(jiffies, device->rs_last_bcast + 2*HZ);
+}
- if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_DISCARD_SIZE) {
- drbd_err(device, "drbd_set_in_sync: sector=%llus size=%d nonsense!\n",
- (unsigned long long)sector, size);
+static void maybe_schedule_on_disk_bitmap_update(struct drbd_device *device, bool rs_done)
+{
+ if (rs_done)
+ set_bit(RS_DONE, &device->flags);
+ /* and also set RS_PROGRESS below */
+ else if (!lazy_bitmap_update_due(device))
return;
- }
-
- if (!get_ldev(device))
- return; /* no disk, no metadata, no bitmap to clear bits in */
-
- nr_sectors = drbd_get_capacity(device->this_bdev);
- esector = sector + (size >> 9) - 1;
-
- if (!expect(sector < nr_sectors))
- goto out;
- if (!expect(esector < nr_sectors))
- esector = nr_sectors - 1;
-
- lbnr = BM_SECT_TO_BIT(nr_sectors-1);
-
- /* we clear it (in sync).
- * round up start sector, round down end sector. we make sure we only
- * clear full, aligned, BM_BLOCK_SIZE (4K) blocks */
- if (unlikely(esector < BM_SECT_PER_BIT-1))
- goto out;
- if (unlikely(esector == (nr_sectors-1)))
- ebnr = lbnr;
- else
- ebnr = BM_SECT_TO_BIT(esector - (BM_SECT_PER_BIT-1));
- sbnr = BM_SECT_TO_BIT(sector + BM_SECT_PER_BIT-1);
- if (sbnr > ebnr)
- goto out;
+ drbd_device_post_work(device, RS_PROGRESS);
+}
+static int update_sync_bits(struct drbd_device *device,
+ unsigned long sbnr, unsigned long ebnr,
+ enum update_sync_bits_mode mode)
+{
/*
- * ok, (capacity & 7) != 0 sometimes, but who cares...
- * we count rs_{total,left} in bits, not sectors.
+ * We keep a count of set bits per resync-extent in the ->rs_left
+ * caching member, so we need to loop and work within the resync extent
+ * alignment. Typically this loop will execute exactly once.
*/
- count = drbd_bm_clear_bits(device, sbnr, ebnr);
- if (count) {
- drbd_advance_rs_marks(device, drbd_bm_total_weight(device));
- spin_lock_irqsave(&device->al_lock, flags);
- drbd_try_clear_on_disk_bm(device, sector, count, true);
- spin_unlock_irqrestore(&device->al_lock, flags);
-
- /* just wake_up unconditional now, various lc_chaged(),
- * lc_put() in drbd_try_clear_on_disk_bm(). */
- wake_up = 1;
+ unsigned long flags;
+ unsigned long count = 0;
+ unsigned int cleared = 0;
+ while (sbnr <= ebnr) {
+ /* set temporary boundary bit number to last bit number within
+ * the resync extent of the current start bit number,
+ * but cap at provided end bit number */
+ unsigned long tbnr = min(ebnr, sbnr | BM_BLOCKS_PER_BM_EXT_MASK);
+ unsigned long c;
+
+ if (mode == RECORD_RS_FAILED)
+ /* Only called from drbd_rs_failed_io(), bits
+ * supposedly still set. Recount, maybe some
+ * of the bits have been successfully cleared
+ * by application IO meanwhile.
+ */
+ c = drbd_bm_count_bits(device, sbnr, tbnr);
+ else if (mode == SET_IN_SYNC)
+ c = drbd_bm_clear_bits(device, sbnr, tbnr);
+ else /* if (mode == SET_OUT_OF_SYNC) */
+ c = drbd_bm_set_bits(device, sbnr, tbnr);
+
+ if (c) {
+ spin_lock_irqsave(&device->al_lock, flags);
+ cleared += update_rs_extent(device, BM_BIT_TO_EXT(sbnr), c, mode);
+ spin_unlock_irqrestore(&device->al_lock, flags);
+ count += c;
+ }
+ sbnr = tbnr + 1;
}
-out:
- put_ldev(device);
- if (wake_up)
+ if (count) {
+ if (mode == SET_IN_SYNC) {
+ unsigned long still_to_go = drbd_bm_total_weight(device);
+ bool rs_is_done = (still_to_go <= device->rs_failed);
+ drbd_advance_rs_marks(device, still_to_go);
+ if (cleared || rs_is_done)
+ maybe_schedule_on_disk_bitmap_update(device, rs_is_done);
+ } else if (mode == RECORD_RS_FAILED)
+ device->rs_failed += count;
wake_up(&device->al_wait);
+ }
+ return count;
}
-/*
- * this is intended to set one request worth of data out of sync.
- * affects at least 1 bit,
- * and at most 1+DRBD_MAX_BIO_SIZE/BM_BLOCK_SIZE bits.
+/* clear the bit corresponding to the piece of storage in question:
+ * size byte of data starting from sector. Only clear a bits of the affected
+ * one ore more _aligned_ BM_BLOCK_SIZE blocks.
+ *
+ * called by worker on C_SYNC_TARGET and receiver on SyncSource.
*
- * called by tl_clear and drbd_send_dblock (==drbd_make_request).
- * so this can be _any_ process.
*/
-int __drbd_set_out_of_sync(struct drbd_device *device, sector_t sector, int size,
- const char *file, const unsigned int line)
+int __drbd_change_sync(struct drbd_device *device, sector_t sector, int size,
+ enum update_sync_bits_mode mode,
+ const char *file, const unsigned int line)
{
- unsigned long sbnr, ebnr, flags;
+ /* Is called from worker and receiver context _only_ */
+ unsigned long sbnr, ebnr, lbnr;
+ unsigned long count = 0;
sector_t esector, nr_sectors;
- unsigned int enr, count = 0;
- struct lc_element *e;
- /* this should be an empty REQ_FLUSH */
- if (size == 0)
+ /* This would be an empty REQ_FLUSH, be silent. */
+ if ((mode == SET_OUT_OF_SYNC) && size == 0)
return 0;
- if (size < 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_DISCARD_SIZE) {
- drbd_err(device, "sector: %llus, size: %d\n",
- (unsigned long long)sector, size);
+ if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_DISCARD_SIZE) {
+ drbd_err(device, "%s: sector=%llus size=%d nonsense!\n",
+ drbd_change_sync_fname[mode],
+ (unsigned long long)sector, size);
return 0;
}
if (!get_ldev(device))
- return 0; /* no disk, no metadata, no bitmap to set bits in */
+ return 0; /* no disk, no metadata, no bitmap to manipulate bits in */
nr_sectors = drbd_get_capacity(device->this_bdev);
esector = sector + (size >> 9) - 1;
if (!expect(esector < nr_sectors))
esector = nr_sectors - 1;
- /* we set it out of sync,
- * we do not need to round anything here */
- sbnr = BM_SECT_TO_BIT(sector);
- ebnr = BM_SECT_TO_BIT(esector);
-
- /* ok, (capacity & 7) != 0 sometimes, but who cares...
- * we count rs_{total,left} in bits, not sectors. */
- spin_lock_irqsave(&device->al_lock, flags);
- count = drbd_bm_set_bits(device, sbnr, ebnr);
+ lbnr = BM_SECT_TO_BIT(nr_sectors-1);
- enr = BM_SECT_TO_EXT(sector);
- e = lc_find(device->resync, enr);
- if (e)
- lc_entry(e, struct bm_extent, lce)->rs_left += count;
- spin_unlock_irqrestore(&device->al_lock, flags);
+ if (mode == SET_IN_SYNC) {
+ /* Round up start sector, round down end sector. We make sure
+ * we only clear full, aligned, BM_BLOCK_SIZE blocks. */
+ if (unlikely(esector < BM_SECT_PER_BIT-1))
+ goto out;
+ if (unlikely(esector == (nr_sectors-1)))
+ ebnr = lbnr;
+ else
+ ebnr = BM_SECT_TO_BIT(esector - (BM_SECT_PER_BIT-1));
+ sbnr = BM_SECT_TO_BIT(sector + BM_SECT_PER_BIT-1);
+ } else {
+ /* We set it out of sync, or record resync failure.
+ * Should not round anything here. */
+ sbnr = BM_SECT_TO_BIT(sector);
+ ebnr = BM_SECT_TO_BIT(esector);
+ }
+ count = update_sync_bits(device, sbnr, ebnr, mode);
out:
put_ldev(device);
-
return count;
}
struct lc_element *e;
struct bm_extent *bm_ext;
int i;
+ bool throttle = drbd_rs_should_slow_down(device, sector, true);
+
+ /* If we need to throttle, a half-locked (only marked BME_NO_WRITES,
+ * not yet BME_LOCKED) extent needs to be kicked out explicitly if we
+ * need to throttle. There is at most one such half-locked extent,
+ * which is remembered in resync_wenr. */
+
+ if (throttle && device->resync_wenr != enr)
+ return -EAGAIN;
spin_lock_irq(&device->al_lock);
if (device->resync_wenr != LC_FREE && device->resync_wenr != enr) {
D_ASSERT(device, test_bit(BME_NO_WRITES, &bm_ext->flags));
clear_bit(BME_NO_WRITES, &bm_ext->flags);
device->resync_wenr = LC_FREE;
- if (lc_put(device->resync, &bm_ext->lce) == 0)
+ if (lc_put(device->resync, &bm_ext->lce) == 0) {
+ bm_ext->flags = 0;
device->resync_locked--;
+ }
wake_up(&device->al_wait);
} else {
drbd_alert(device, "LOGIC BUG\n");
return 0;
try_again:
- if (bm_ext)
- device->resync_wenr = enr;
+ if (bm_ext) {
+ if (throttle) {
+ D_ASSERT(device, !test_bit(BME_LOCKED, &bm_ext->flags));
+ D_ASSERT(device, test_bit(BME_NO_WRITES, &bm_ext->flags));
+ clear_bit(BME_NO_WRITES, &bm_ext->flags);
+ device->resync_wenr = LC_FREE;
+ if (lc_put(device->resync, &bm_ext->lce) == 0) {
+ bm_ext->flags = 0;
+ device->resync_locked--;
+ }
+ wake_up(&device->al_wait);
+ } else
+ device->resync_wenr = enr;
+ }
spin_unlock_irq(&device->al_lock);
return -EAGAIN;
}
return 0;
}
-
-/**
- * drbd_rs_failed_io() - Record information on a failure to resync the specified blocks
- * @device: DRBD device.
- * @sector: The sector number.
- * @size: Size of failed IO operation, in byte.
- */
-void drbd_rs_failed_io(struct drbd_device *device, sector_t sector, int size)
-{
- /* Is called from worker and receiver context _only_ */
- unsigned long sbnr, ebnr, lbnr;
- unsigned long count;
- sector_t esector, nr_sectors;
- int wake_up = 0;
-
- if (size <= 0 || !IS_ALIGNED(size, 512) || size > DRBD_MAX_DISCARD_SIZE) {
- drbd_err(device, "drbd_rs_failed_io: sector=%llus size=%d nonsense!\n",
- (unsigned long long)sector, size);
- return;
- }
- nr_sectors = drbd_get_capacity(device->this_bdev);
- esector = sector + (size >> 9) - 1;
-
- if (!expect(sector < nr_sectors))
- return;
- if (!expect(esector < nr_sectors))
- esector = nr_sectors - 1;
-
- lbnr = BM_SECT_TO_BIT(nr_sectors-1);
-
- /*
- * round up start sector, round down end sector. we make sure we only
- * handle full, aligned, BM_BLOCK_SIZE (4K) blocks */
- if (unlikely(esector < BM_SECT_PER_BIT-1))
- return;
- if (unlikely(esector == (nr_sectors-1)))
- ebnr = lbnr;
- else
- ebnr = BM_SECT_TO_BIT(esector - (BM_SECT_PER_BIT-1));
- sbnr = BM_SECT_TO_BIT(sector + BM_SECT_PER_BIT-1);
-
- if (sbnr > ebnr)
- return;
-
- /*
- * ok, (capacity & 7) != 0 sometimes, but who cares...
- * we count rs_{total,left} in bits, not sectors.
- */
- spin_lock_irq(&device->al_lock);
- count = drbd_bm_count_bits(device, sbnr, ebnr);
- if (count) {
- device->rs_failed += count;
-
- if (get_ldev(device)) {
- drbd_try_clear_on_disk_bm(device, sector, count, false);
- put_ldev(device);
- }
-
- /* just wake_up unconditional now, various lc_chaged(),
- * lc_put() in drbd_try_clear_on_disk_bm(). */
- wake_up = 1;
- }
- spin_unlock_irq(&device->al_lock);
- if (wake_up)
- wake_up(&device->al_wait);
-}
the Free Software Foundation, 675 Mass Ave, Cambridge, MA 02139, USA.
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/bitops.h>
#include <linux/vmalloc.h>
#include <linux/string.h>
for (i = 0; i < number; i++) {
if (!pages[i]) {
- printk(KERN_ALERT "drbd: bm_free_pages tried to free "
- "a NULL pointer; i=%lu n=%lu\n",
- i, number);
+ pr_alert("bm_free_pages tried to free a NULL pointer; i=%lu n=%lu\n",
+ i, number);
continue;
}
__free_page(pages[i]);
end = offset + len;
if (end > b->bm_words) {
- printk(KERN_ALERT "drbd: bm_memset end > bm_words\n");
+ pr_alert("bm_memset end > bm_words\n");
return;
}
p_addr = bm_map_pidx(b, idx);
bm = p_addr + MLPP(offset);
if (bm+do_now > p_addr + LWPP) {
- printk(KERN_ALERT "drbd: BUG BUG BUG! p_addr:%p bm:%p do_now:%d\n",
+ pr_alert("BUG BUG BUG! p_addr:%p bm:%p do_now:%d\n",
p_addr, bm, (int)do_now);
} else
memset(bm, c, do_now * sizeof(long));
spin_unlock_irq(&b->bm_lock);
}
-struct bm_aio_ctx {
- struct drbd_device *device;
- atomic_t in_flight;
- unsigned int done;
- unsigned flags;
-#define BM_AIO_COPY_PAGES 1
-#define BM_AIO_WRITE_HINTED 2
-#define BM_WRITE_ALL_PAGES 4
- int error;
- struct kref kref;
-};
-
-static void bm_aio_ctx_destroy(struct kref *kref)
+static void drbd_bm_aio_ctx_destroy(struct kref *kref)
{
- struct bm_aio_ctx *ctx = container_of(kref, struct bm_aio_ctx, kref);
+ struct drbd_bm_aio_ctx *ctx = container_of(kref, struct drbd_bm_aio_ctx, kref);
+ unsigned long flags;
+ spin_lock_irqsave(&ctx->device->resource->req_lock, flags);
+ list_del(&ctx->list);
+ spin_unlock_irqrestore(&ctx->device->resource->req_lock, flags);
put_ldev(ctx->device);
kfree(ctx);
}
/* bv_page may be a copy, or may be the original */
static void bm_async_io_complete(struct bio *bio, int error)
{
- struct bm_aio_ctx *ctx = bio->bi_private;
+ struct drbd_bm_aio_ctx *ctx = bio->bi_private;
struct drbd_device *device = ctx->device;
struct drbd_bitmap *b = device->bitmap;
unsigned int idx = bm_page_to_idx(bio->bi_io_vec[0].bv_page);
if (atomic_dec_and_test(&ctx->in_flight)) {
ctx->done = 1;
wake_up(&device->misc_wait);
- kref_put(&ctx->kref, &bm_aio_ctx_destroy);
+ kref_put(&ctx->kref, &drbd_bm_aio_ctx_destroy);
}
}
-static void bm_page_io_async(struct bm_aio_ctx *ctx, int page_nr, int rw) __must_hold(local)
+static void bm_page_io_async(struct drbd_bm_aio_ctx *ctx, int page_nr) __must_hold(local)
{
struct bio *bio = bio_alloc_drbd(GFP_NOIO);
struct drbd_device *device = ctx->device;
struct drbd_bitmap *b = device->bitmap;
struct page *page;
unsigned int len;
+ unsigned int rw = (ctx->flags & BM_AIO_READ) ? READ : WRITE;
sector_t on_disk_sector =
device->ldev->md.md_offset + device->ldev->md.bm_offset;
/*
* bm_rw: read/write the whole bitmap from/to its on disk location.
*/
-static int bm_rw(struct drbd_device *device, int rw, unsigned flags, unsigned lazy_writeout_upper_idx) __must_hold(local)
+static int bm_rw(struct drbd_device *device, const unsigned int flags, unsigned lazy_writeout_upper_idx) __must_hold(local)
{
- struct bm_aio_ctx *ctx;
+ struct drbd_bm_aio_ctx *ctx;
struct drbd_bitmap *b = device->bitmap;
int num_pages, i, count = 0;
unsigned long now;
* as we submit copies of pages anyways.
*/
- ctx = kmalloc(sizeof(struct bm_aio_ctx), GFP_NOIO);
+ ctx = kmalloc(sizeof(struct drbd_bm_aio_ctx), GFP_NOIO);
if (!ctx)
return -ENOMEM;
- *ctx = (struct bm_aio_ctx) {
+ *ctx = (struct drbd_bm_aio_ctx) {
.device = device,
+ .start_jif = jiffies,
.in_flight = ATOMIC_INIT(1),
.done = 0,
.flags = flags,
.kref = { ATOMIC_INIT(2) },
};
- if (!get_ldev_if_state(device, D_ATTACHING)) { /* put is in bm_aio_ctx_destroy() */
+ if (!get_ldev_if_state(device, D_ATTACHING)) { /* put is in drbd_bm_aio_ctx_destroy() */
drbd_err(device, "ASSERT FAILED: get_ldev_if_state() == 1 in bm_rw()\n");
kfree(ctx);
return -ENODEV;
}
+ /* Here D_ATTACHING is sufficient since drbd_bm_read() is called only from
+ drbd_adm_attach(), after device->ldev was assigned. */
- if (!ctx->flags)
+ if (0 == (ctx->flags & ~BM_AIO_READ))
WARN_ON(!(BM_LOCKED_MASK & b->bm_flags));
+ spin_lock_irq(&device->resource->req_lock);
+ list_add_tail(&ctx->list, &device->pending_bitmap_io);
+ spin_unlock_irq(&device->resource->req_lock);
+
num_pages = b->bm_number_of_pages;
now = jiffies;
/* ignore completely unchanged pages */
if (lazy_writeout_upper_idx && i == lazy_writeout_upper_idx)
break;
- if (rw & WRITE) {
+ if (!(flags & BM_AIO_READ)) {
if ((flags & BM_AIO_WRITE_HINTED) &&
!test_and_clear_bit(BM_PAGE_HINT_WRITEOUT,
&page_private(b->bm_pages[i])))
continue;
- if (!(flags & BM_WRITE_ALL_PAGES) &&
+ if (!(flags & BM_AIO_WRITE_ALL_PAGES) &&
bm_test_page_unchanged(b->bm_pages[i])) {
dynamic_drbd_dbg(device, "skipped bm write for idx %u\n", i);
continue;
}
}
atomic_inc(&ctx->in_flight);
- bm_page_io_async(ctx, i, rw);
+ bm_page_io_async(ctx, i);
++count;
cond_resched();
}
if (!atomic_dec_and_test(&ctx->in_flight))
wait_until_done_or_force_detached(device, device->ldev, &ctx->done);
else
- kref_put(&ctx->kref, &bm_aio_ctx_destroy);
+ kref_put(&ctx->kref, &drbd_bm_aio_ctx_destroy);
/* summary for global bitmap IO */
if (flags == 0)
drbd_info(device, "bitmap %s of %u pages took %lu jiffies\n",
- rw == WRITE ? "WRITE" : "READ",
+ (flags & BM_AIO_READ) ? "READ" : "WRITE",
count, jiffies - now);
if (ctx->error) {
err = -EIO; /* Disk timeout/force-detach during IO... */
now = jiffies;
- if (rw == WRITE) {
- drbd_md_flush(device);
- } else /* rw == READ */ {
+ if (flags & BM_AIO_READ) {
b->bm_set = bm_count_bits(b);
drbd_info(device, "recounting of set bits took additional %lu jiffies\n",
jiffies - now);
}
now = b->bm_set;
- if (flags == 0)
+ if ((flags & ~BM_AIO_READ) == 0)
drbd_info(device, "%s (%lu bits) marked out-of-sync by on disk bit-map.\n",
ppsize(ppb, now << (BM_BLOCK_SHIFT-10)), now);
- kref_put(&ctx->kref, &bm_aio_ctx_destroy);
+ kref_put(&ctx->kref, &drbd_bm_aio_ctx_destroy);
return err;
}
*/
int drbd_bm_read(struct drbd_device *device) __must_hold(local)
{
- return bm_rw(device, READ, 0, 0);
+ return bm_rw(device, BM_AIO_READ, 0);
}
/**
*/
int drbd_bm_write(struct drbd_device *device) __must_hold(local)
{
- return bm_rw(device, WRITE, 0, 0);
+ return bm_rw(device, 0, 0);
}
/**
*/
int drbd_bm_write_all(struct drbd_device *device) __must_hold(local)
{
- return bm_rw(device, WRITE, BM_WRITE_ALL_PAGES, 0);
+ return bm_rw(device, BM_AIO_WRITE_ALL_PAGES, 0);
+}
+
+/**
+ * drbd_bm_write_lazy() - Write bitmap pages 0 to @upper_idx-1, if they have changed.
+ * @device: DRBD device.
+ * @upper_idx: 0: write all changed pages; +ve: page index to stop scanning for changed pages
+ */
+int drbd_bm_write_lazy(struct drbd_device *device, unsigned upper_idx) __must_hold(local)
+{
+ return bm_rw(device, BM_AIO_COPY_PAGES, upper_idx);
}
/**
*/
int drbd_bm_write_copy_pages(struct drbd_device *device) __must_hold(local)
{
- return bm_rw(device, WRITE, BM_AIO_COPY_PAGES, 0);
+ return bm_rw(device, BM_AIO_COPY_PAGES, 0);
}
/**
*/
int drbd_bm_write_hinted(struct drbd_device *device) __must_hold(local)
{
- return bm_rw(device, WRITE, BM_AIO_WRITE_HINTED | BM_AIO_COPY_PAGES, 0);
-}
-
-/**
- * drbd_bm_write_page() - Writes a PAGE_SIZE aligned piece of bitmap
- * @device: DRBD device.
- * @idx: bitmap page index
- *
- * We don't want to special case on logical_block_size of the backend device,
- * so we submit PAGE_SIZE aligned pieces.
- * Note that on "most" systems, PAGE_SIZE is 4k.
- *
- * In case this becomes an issue on systems with larger PAGE_SIZE,
- * we may want to change this again to write 4k aligned 4k pieces.
- */
-int drbd_bm_write_page(struct drbd_device *device, unsigned int idx) __must_hold(local)
-{
- struct bm_aio_ctx *ctx;
- int err;
-
- if (bm_test_page_unchanged(device->bitmap->bm_pages[idx])) {
- dynamic_drbd_dbg(device, "skipped bm page write for idx %u\n", idx);
- return 0;
- }
-
- ctx = kmalloc(sizeof(struct bm_aio_ctx), GFP_NOIO);
- if (!ctx)
- return -ENOMEM;
-
- *ctx = (struct bm_aio_ctx) {
- .device = device,
- .in_flight = ATOMIC_INIT(1),
- .done = 0,
- .flags = BM_AIO_COPY_PAGES,
- .error = 0,
- .kref = { ATOMIC_INIT(2) },
- };
-
- if (!get_ldev_if_state(device, D_ATTACHING)) { /* put is in bm_aio_ctx_destroy() */
- drbd_err(device, "ASSERT FAILED: get_ldev_if_state() == 1 in drbd_bm_write_page()\n");
- kfree(ctx);
- return -ENODEV;
- }
-
- bm_page_io_async(ctx, idx, WRITE_SYNC);
- wait_until_done_or_force_detached(device, device->ldev, &ctx->done);
-
- if (ctx->error)
- drbd_chk_io_error(device, 1, DRBD_META_IO_ERROR);
- /* that causes us to detach, so the in memory bitmap will be
- * gone in a moment as well. */
-
- device->bm_writ_cnt++;
- err = atomic_read(&ctx->in_flight) ? -EIO : ctx->error;
- kref_put(&ctx->kref, &bm_aio_ctx_destroy);
- return err;
+ return bm_rw(device, BM_AIO_WRITE_HINTED | BM_AIO_COPY_PAGES, 0);
}
/* NOTE
--- /dev/null
+#define pr_fmt(fmt) "drbd debugfs: " fmt
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/stat.h>
+#include <linux/jiffies.h>
+#include <linux/list.h>
+
+#include "drbd_int.h"
+#include "drbd_req.h"
+#include "drbd_debugfs.h"
+
+
+/**********************************************************************
+ * Whenever you change the file format, remember to bump the version. *
+ **********************************************************************/
+
+static struct dentry *drbd_debugfs_root;
+static struct dentry *drbd_debugfs_version;
+static struct dentry *drbd_debugfs_resources;
+static struct dentry *drbd_debugfs_minors;
+
+static void seq_print_age_or_dash(struct seq_file *m, bool valid, unsigned long dt)
+{
+ if (valid)
+ seq_printf(m, "\t%d", jiffies_to_msecs(dt));
+ else
+ seq_printf(m, "\t-");
+}
+
+static void __seq_print_rq_state_bit(struct seq_file *m,
+ bool is_set, char *sep, const char *set_name, const char *unset_name)
+{
+ if (is_set && set_name) {
+ seq_putc(m, *sep);
+ seq_puts(m, set_name);
+ *sep = '|';
+ } else if (!is_set && unset_name) {
+ seq_putc(m, *sep);
+ seq_puts(m, unset_name);
+ *sep = '|';
+ }
+}
+
+static void seq_print_rq_state_bit(struct seq_file *m,
+ bool is_set, char *sep, const char *set_name)
+{
+ __seq_print_rq_state_bit(m, is_set, sep, set_name, NULL);
+}
+
+/* pretty print enum drbd_req_state_bits req->rq_state */
+static void seq_print_request_state(struct seq_file *m, struct drbd_request *req)
+{
+ unsigned int s = req->rq_state;
+ char sep = ' ';
+ seq_printf(m, "\t0x%08x", s);
+ seq_printf(m, "\tmaster: %s", req->master_bio ? "pending" : "completed");
+
+ /* RQ_WRITE ignored, already reported */
+ seq_puts(m, "\tlocal:");
+ seq_print_rq_state_bit(m, s & RQ_IN_ACT_LOG, &sep, "in-AL");
+ seq_print_rq_state_bit(m, s & RQ_POSTPONED, &sep, "postponed");
+ seq_print_rq_state_bit(m, s & RQ_COMPLETION_SUSP, &sep, "suspended");
+ sep = ' ';
+ seq_print_rq_state_bit(m, s & RQ_LOCAL_PENDING, &sep, "pending");
+ seq_print_rq_state_bit(m, s & RQ_LOCAL_COMPLETED, &sep, "completed");
+ seq_print_rq_state_bit(m, s & RQ_LOCAL_ABORTED, &sep, "aborted");
+ seq_print_rq_state_bit(m, s & RQ_LOCAL_OK, &sep, "ok");
+ if (sep == ' ')
+ seq_puts(m, " -");
+
+ /* for_each_connection ... */
+ seq_printf(m, "\tnet:");
+ sep = ' ';
+ seq_print_rq_state_bit(m, s & RQ_NET_PENDING, &sep, "pending");
+ seq_print_rq_state_bit(m, s & RQ_NET_QUEUED, &sep, "queued");
+ seq_print_rq_state_bit(m, s & RQ_NET_SENT, &sep, "sent");
+ seq_print_rq_state_bit(m, s & RQ_NET_DONE, &sep, "done");
+ seq_print_rq_state_bit(m, s & RQ_NET_SIS, &sep, "sis");
+ seq_print_rq_state_bit(m, s & RQ_NET_OK, &sep, "ok");
+ if (sep == ' ')
+ seq_puts(m, " -");
+
+ seq_printf(m, " :");
+ sep = ' ';
+ seq_print_rq_state_bit(m, s & RQ_EXP_RECEIVE_ACK, &sep, "B");
+ seq_print_rq_state_bit(m, s & RQ_EXP_WRITE_ACK, &sep, "C");
+ seq_print_rq_state_bit(m, s & RQ_EXP_BARR_ACK, &sep, "barr");
+ if (sep == ' ')
+ seq_puts(m, " -");
+ seq_printf(m, "\n");
+}
+
+static void seq_print_one_request(struct seq_file *m, struct drbd_request *req, unsigned long now)
+{
+ /* change anything here, fixup header below! */
+ unsigned int s = req->rq_state;
+
+#define RQ_HDR_1 "epoch\tsector\tsize\trw"
+ seq_printf(m, "0x%x\t%llu\t%u\t%s",
+ req->epoch,
+ (unsigned long long)req->i.sector, req->i.size >> 9,
+ (s & RQ_WRITE) ? "W" : "R");
+
+#define RQ_HDR_2 "\tstart\tin AL\tsubmit"
+ seq_printf(m, "\t%d", jiffies_to_msecs(now - req->start_jif));
+ seq_print_age_or_dash(m, s & RQ_IN_ACT_LOG, now - req->in_actlog_jif);
+ seq_print_age_or_dash(m, s & RQ_LOCAL_PENDING, now - req->pre_submit_jif);
+
+#define RQ_HDR_3 "\tsent\tacked\tdone"
+ seq_print_age_or_dash(m, s & RQ_NET_SENT, now - req->pre_send_jif);
+ seq_print_age_or_dash(m, (s & RQ_NET_SENT) && !(s & RQ_NET_PENDING), now - req->acked_jif);
+ seq_print_age_or_dash(m, s & RQ_NET_DONE, now - req->net_done_jif);
+
+#define RQ_HDR_4 "\tstate\n"
+ seq_print_request_state(m, req);
+}
+#define RQ_HDR RQ_HDR_1 RQ_HDR_2 RQ_HDR_3 RQ_HDR_4
+
+static void seq_print_minor_vnr_req(struct seq_file *m, struct drbd_request *req, unsigned long now)
+{
+ seq_printf(m, "%u\t%u\t", req->device->minor, req->device->vnr);
+ seq_print_one_request(m, req, now);
+}
+
+static void seq_print_resource_pending_meta_io(struct seq_file *m, struct drbd_resource *resource, unsigned long now)
+{
+ struct drbd_device *device;
+ unsigned int i;
+
+ seq_puts(m, "minor\tvnr\tstart\tsubmit\tintent\n");
+ rcu_read_lock();
+ idr_for_each_entry(&resource->devices, device, i) {
+ struct drbd_md_io tmp;
+ /* In theory this is racy,
+ * in the sense that there could have been a
+ * drbd_md_put_buffer(); drbd_md_get_buffer();
+ * between accessing these members here. */
+ tmp = device->md_io;
+ if (atomic_read(&tmp.in_use)) {
+ seq_printf(m, "%u\t%u\t%d\t",
+ device->minor, device->vnr,
+ jiffies_to_msecs(now - tmp.start_jif));
+ if (time_before(tmp.submit_jif, tmp.start_jif))
+ seq_puts(m, "-\t");
+ else
+ seq_printf(m, "%d\t", jiffies_to_msecs(now - tmp.submit_jif));
+ seq_printf(m, "%s\n", tmp.current_use);
+ }
+ }
+ rcu_read_unlock();
+}
+
+static void seq_print_waiting_for_AL(struct seq_file *m, struct drbd_resource *resource, unsigned long now)
+{
+ struct drbd_device *device;
+ unsigned int i;
+
+ seq_puts(m, "minor\tvnr\tage\t#waiting\n");
+ rcu_read_lock();
+ idr_for_each_entry(&resource->devices, device, i) {
+ unsigned long jif;
+ struct drbd_request *req;
+ int n = atomic_read(&device->ap_actlog_cnt);
+ if (n) {
+ spin_lock_irq(&device->resource->req_lock);
+ req = list_first_entry_or_null(&device->pending_master_completion[1],
+ struct drbd_request, req_pending_master_completion);
+ /* if the oldest request does not wait for the activity log
+ * it is not interesting for us here */
+ if (req && !(req->rq_state & RQ_IN_ACT_LOG))
+ jif = req->start_jif;
+ else
+ req = NULL;
+ spin_unlock_irq(&device->resource->req_lock);
+ }
+ if (n) {
+ seq_printf(m, "%u\t%u\t", device->minor, device->vnr);
+ if (req)
+ seq_printf(m, "%u\t", jiffies_to_msecs(now - jif));
+ else
+ seq_puts(m, "-\t");
+ seq_printf(m, "%u\n", n);
+ }
+ }
+ rcu_read_unlock();
+}
+
+static void seq_print_device_bitmap_io(struct seq_file *m, struct drbd_device *device, unsigned long now)
+{
+ struct drbd_bm_aio_ctx *ctx;
+ unsigned long start_jif;
+ unsigned int in_flight;
+ unsigned int flags;
+ spin_lock_irq(&device->resource->req_lock);
+ ctx = list_first_entry_or_null(&device->pending_bitmap_io, struct drbd_bm_aio_ctx, list);
+ if (ctx && ctx->done)
+ ctx = NULL;
+ if (ctx) {
+ start_jif = ctx->start_jif;
+ in_flight = atomic_read(&ctx->in_flight);
+ flags = ctx->flags;
+ }
+ spin_unlock_irq(&device->resource->req_lock);
+ if (ctx) {
+ seq_printf(m, "%u\t%u\t%c\t%u\t%u\n",
+ device->minor, device->vnr,
+ (flags & BM_AIO_READ) ? 'R' : 'W',
+ jiffies_to_msecs(now - start_jif),
+ in_flight);
+ }
+}
+
+static void seq_print_resource_pending_bitmap_io(struct seq_file *m, struct drbd_resource *resource, unsigned long now)
+{
+ struct drbd_device *device;
+ unsigned int i;
+
+ seq_puts(m, "minor\tvnr\trw\tage\t#in-flight\n");
+ rcu_read_lock();
+ idr_for_each_entry(&resource->devices, device, i) {
+ seq_print_device_bitmap_io(m, device, now);
+ }
+ rcu_read_unlock();
+}
+
+/* pretty print enum peer_req->flags */
+static void seq_print_peer_request_flags(struct seq_file *m, struct drbd_peer_request *peer_req)
+{
+ unsigned long f = peer_req->flags;
+ char sep = ' ';
+
+ __seq_print_rq_state_bit(m, f & EE_SUBMITTED, &sep, "submitted", "preparing");
+ __seq_print_rq_state_bit(m, f & EE_APPLICATION, &sep, "application", "internal");
+ seq_print_rq_state_bit(m, f & EE_CALL_AL_COMPLETE_IO, &sep, "in-AL");
+ seq_print_rq_state_bit(m, f & EE_SEND_WRITE_ACK, &sep, "C");
+ seq_print_rq_state_bit(m, f & EE_MAY_SET_IN_SYNC, &sep, "set-in-sync");
+
+ if (f & EE_IS_TRIM) {
+ seq_putc(m, sep);
+ sep = '|';
+ if (f & EE_IS_TRIM_USE_ZEROOUT)
+ seq_puts(m, "zero-out");
+ else
+ seq_puts(m, "trim");
+ }
+ seq_putc(m, '\n');
+}
+
+static void seq_print_peer_request(struct seq_file *m,
+ struct drbd_device *device, struct list_head *lh,
+ unsigned long now)
+{
+ bool reported_preparing = false;
+ struct drbd_peer_request *peer_req;
+ list_for_each_entry(peer_req, lh, w.list) {
+ if (reported_preparing && !(peer_req->flags & EE_SUBMITTED))
+ continue;
+
+ if (device)
+ seq_printf(m, "%u\t%u\t", device->minor, device->vnr);
+
+ seq_printf(m, "%llu\t%u\t%c\t%u\t",
+ (unsigned long long)peer_req->i.sector, peer_req->i.size >> 9,
+ (peer_req->flags & EE_WRITE) ? 'W' : 'R',
+ jiffies_to_msecs(now - peer_req->submit_jif));
+ seq_print_peer_request_flags(m, peer_req);
+ if (peer_req->flags & EE_SUBMITTED)
+ break;
+ else
+ reported_preparing = true;
+ }
+}
+
+static void seq_print_device_peer_requests(struct seq_file *m,
+ struct drbd_device *device, unsigned long now)
+{
+ seq_puts(m, "minor\tvnr\tsector\tsize\trw\tage\tflags\n");
+ spin_lock_irq(&device->resource->req_lock);
+ seq_print_peer_request(m, device, &device->active_ee, now);
+ seq_print_peer_request(m, device, &device->read_ee, now);
+ seq_print_peer_request(m, device, &device->sync_ee, now);
+ spin_unlock_irq(&device->resource->req_lock);
+ if (test_bit(FLUSH_PENDING, &device->flags)) {
+ seq_printf(m, "%u\t%u\t-\t-\tF\t%u\tflush\n",
+ device->minor, device->vnr,
+ jiffies_to_msecs(now - device->flush_jif));
+ }
+}
+
+static void seq_print_resource_pending_peer_requests(struct seq_file *m,
+ struct drbd_resource *resource, unsigned long now)
+{
+ struct drbd_device *device;
+ unsigned int i;
+
+ rcu_read_lock();
+ idr_for_each_entry(&resource->devices, device, i) {
+ seq_print_device_peer_requests(m, device, now);
+ }
+ rcu_read_unlock();
+}
+
+static void seq_print_resource_transfer_log_summary(struct seq_file *m,
+ struct drbd_resource *resource,
+ struct drbd_connection *connection,
+ unsigned long now)
+{
+ struct drbd_request *req;
+ unsigned int count = 0;
+ unsigned int show_state = 0;
+
+ seq_puts(m, "n\tdevice\tvnr\t" RQ_HDR);
+ spin_lock_irq(&resource->req_lock);
+ list_for_each_entry(req, &connection->transfer_log, tl_requests) {
+ unsigned int tmp = 0;
+ unsigned int s;
+ ++count;
+
+ /* don't disable irq "forever" */
+ if (!(count & 0x1ff)) {
+ struct drbd_request *req_next;
+ kref_get(&req->kref);
+ spin_unlock_irq(&resource->req_lock);
+ cond_resched();
+ spin_lock_irq(&resource->req_lock);
+ req_next = list_next_entry(req, tl_requests);
+ if (kref_put(&req->kref, drbd_req_destroy))
+ req = req_next;
+ if (&req->tl_requests == &connection->transfer_log)
+ break;
+ }
+
+ s = req->rq_state;
+
+ /* This is meant to summarize timing issues, to be able to tell
+ * local disk problems from network problems.
+ * Skip requests, if we have shown an even older request with
+ * similar aspects already. */
+ if (req->master_bio == NULL)
+ tmp |= 1;
+ if ((s & RQ_LOCAL_MASK) && (s & RQ_LOCAL_PENDING))
+ tmp |= 2;
+ if (s & RQ_NET_MASK) {
+ if (!(s & RQ_NET_SENT))
+ tmp |= 4;
+ if (s & RQ_NET_PENDING)
+ tmp |= 8;
+ if (!(s & RQ_NET_DONE))
+ tmp |= 16;
+ }
+ if ((tmp & show_state) == tmp)
+ continue;
+ show_state |= tmp;
+ seq_printf(m, "%u\t", count);
+ seq_print_minor_vnr_req(m, req, now);
+ if (show_state == 0x1f)
+ break;
+ }
+ spin_unlock_irq(&resource->req_lock);
+}
+
+/* TODO: transfer_log and friends should be moved to resource */
+static int in_flight_summary_show(struct seq_file *m, void *pos)
+{
+ struct drbd_resource *resource = m->private;
+ struct drbd_connection *connection;
+ unsigned long jif = jiffies;
+
+ connection = first_connection(resource);
+ /* This does not happen, actually.
+ * But be robust and prepare for future code changes. */
+ if (!connection || !kref_get_unless_zero(&connection->kref))
+ return -ESTALE;
+
+ /* BUMP me if you change the file format/content/presentation */
+ seq_printf(m, "v: %u\n\n", 0);
+
+ seq_puts(m, "oldest bitmap IO\n");
+ seq_print_resource_pending_bitmap_io(m, resource, jif);
+ seq_putc(m, '\n');
+
+ seq_puts(m, "meta data IO\n");
+ seq_print_resource_pending_meta_io(m, resource, jif);
+ seq_putc(m, '\n');
+
+ seq_puts(m, "socket buffer stats\n");
+ /* for each connection ... once we have more than one */
+ rcu_read_lock();
+ if (connection->data.socket) {
+ /* open coded SIOCINQ, the "relevant" part */
+ struct tcp_sock *tp = tcp_sk(connection->data.socket->sk);
+ int answ = tp->rcv_nxt - tp->copied_seq;
+ seq_printf(m, "unread receive buffer: %u Byte\n", answ);
+ /* open coded SIOCOUTQ, the "relevant" part */
+ answ = tp->write_seq - tp->snd_una;
+ seq_printf(m, "unacked send buffer: %u Byte\n", answ);
+ }
+ rcu_read_unlock();
+ seq_putc(m, '\n');
+
+ seq_puts(m, "oldest peer requests\n");
+ seq_print_resource_pending_peer_requests(m, resource, jif);
+ seq_putc(m, '\n');
+
+ seq_puts(m, "application requests waiting for activity log\n");
+ seq_print_waiting_for_AL(m, resource, jif);
+ seq_putc(m, '\n');
+
+ seq_puts(m, "oldest application requests\n");
+ seq_print_resource_transfer_log_summary(m, resource, connection, jif);
+ seq_putc(m, '\n');
+
+ jif = jiffies - jif;
+ if (jif)
+ seq_printf(m, "generated in %d ms\n", jiffies_to_msecs(jif));
+ kref_put(&connection->kref, drbd_destroy_connection);
+ return 0;
+}
+
+/* simple_positive(file->f_dentry) respectively debugfs_positive(),
+ * but neither is "reachable" from here.
+ * So we have our own inline version of it above. :-( */
+static inline int debugfs_positive(struct dentry *dentry)
+{
+ return dentry->d_inode && !d_unhashed(dentry);
+}
+
+/* make sure at *open* time that the respective object won't go away. */
+static int drbd_single_open(struct file *file, int (*show)(struct seq_file *, void *),
+ void *data, struct kref *kref,
+ void (*release)(struct kref *))
+{
+ struct dentry *parent;
+ int ret = -ESTALE;
+
+ /* Are we still linked,
+ * or has debugfs_remove() already been called? */
+ parent = file->f_dentry->d_parent;
+ /* not sure if this can happen: */
+ if (!parent || !parent->d_inode)
+ goto out;
+ /* serialize with d_delete() */
+ mutex_lock(&parent->d_inode->i_mutex);
+ /* Make sure the object is still alive */
+ if (debugfs_positive(file->f_dentry)
+ && kref_get_unless_zero(kref))
+ ret = 0;
+ mutex_unlock(&parent->d_inode->i_mutex);
+ if (!ret) {
+ ret = single_open(file, show, data);
+ if (ret)
+ kref_put(kref, release);
+ }
+out:
+ return ret;
+}
+
+static int in_flight_summary_open(struct inode *inode, struct file *file)
+{
+ struct drbd_resource *resource = inode->i_private;
+ return drbd_single_open(file, in_flight_summary_show, resource,
+ &resource->kref, drbd_destroy_resource);
+}
+
+static int in_flight_summary_release(struct inode *inode, struct file *file)
+{
+ struct drbd_resource *resource = inode->i_private;
+ kref_put(&resource->kref, drbd_destroy_resource);
+ return single_release(inode, file);
+}
+
+static const struct file_operations in_flight_summary_fops = {
+ .owner = THIS_MODULE,
+ .open = in_flight_summary_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = in_flight_summary_release,
+};
+
+void drbd_debugfs_resource_add(struct drbd_resource *resource)
+{
+ struct dentry *dentry;
+ if (!drbd_debugfs_resources)
+ return;
+
+ dentry = debugfs_create_dir(resource->name, drbd_debugfs_resources);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ resource->debugfs_res = dentry;
+
+ dentry = debugfs_create_dir("volumes", resource->debugfs_res);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ resource->debugfs_res_volumes = dentry;
+
+ dentry = debugfs_create_dir("connections", resource->debugfs_res);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ resource->debugfs_res_connections = dentry;
+
+ dentry = debugfs_create_file("in_flight_summary", S_IRUSR|S_IRGRP,
+ resource->debugfs_res, resource,
+ &in_flight_summary_fops);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ resource->debugfs_res_in_flight_summary = dentry;
+ return;
+
+fail:
+ drbd_debugfs_resource_cleanup(resource);
+ drbd_err(resource, "failed to create debugfs dentry\n");
+}
+
+static void drbd_debugfs_remove(struct dentry **dp)
+{
+ debugfs_remove(*dp);
+ *dp = NULL;
+}
+
+void drbd_debugfs_resource_cleanup(struct drbd_resource *resource)
+{
+ /* it is ok to call debugfs_remove(NULL) */
+ drbd_debugfs_remove(&resource->debugfs_res_in_flight_summary);
+ drbd_debugfs_remove(&resource->debugfs_res_connections);
+ drbd_debugfs_remove(&resource->debugfs_res_volumes);
+ drbd_debugfs_remove(&resource->debugfs_res);
+}
+
+static void seq_print_one_timing_detail(struct seq_file *m,
+ const struct drbd_thread_timing_details *tdp,
+ unsigned long now)
+{
+ struct drbd_thread_timing_details td;
+ /* No locking...
+ * use temporary assignment to get at consistent data. */
+ do {
+ td = *tdp;
+ } while (td.cb_nr != tdp->cb_nr);
+ if (!td.cb_addr)
+ return;
+ seq_printf(m, "%u\t%d\t%s:%u\t%ps\n",
+ td.cb_nr,
+ jiffies_to_msecs(now - td.start_jif),
+ td.caller_fn, td.line,
+ td.cb_addr);
+}
+
+static void seq_print_timing_details(struct seq_file *m,
+ const char *title,
+ unsigned int cb_nr, struct drbd_thread_timing_details *tdp, unsigned long now)
+{
+ unsigned int start_idx;
+ unsigned int i;
+
+ seq_printf(m, "%s\n", title);
+ /* If not much is going on, this will result in natural ordering.
+ * If it is very busy, we will possibly skip events, or even see wrap
+ * arounds, which could only be avoided with locking.
+ */
+ start_idx = cb_nr % DRBD_THREAD_DETAILS_HIST;
+ for (i = start_idx; i < DRBD_THREAD_DETAILS_HIST; i++)
+ seq_print_one_timing_detail(m, tdp+i, now);
+ for (i = 0; i < start_idx; i++)
+ seq_print_one_timing_detail(m, tdp+i, now);
+}
+
+static int callback_history_show(struct seq_file *m, void *ignored)
+{
+ struct drbd_connection *connection = m->private;
+ unsigned long jif = jiffies;
+
+ /* BUMP me if you change the file format/content/presentation */
+ seq_printf(m, "v: %u\n\n", 0);
+
+ seq_puts(m, "n\tage\tcallsite\tfn\n");
+ seq_print_timing_details(m, "worker", connection->w_cb_nr, connection->w_timing_details, jif);
+ seq_print_timing_details(m, "receiver", connection->r_cb_nr, connection->r_timing_details, jif);
+ return 0;
+}
+
+static int callback_history_open(struct inode *inode, struct file *file)
+{
+ struct drbd_connection *connection = inode->i_private;
+ return drbd_single_open(file, callback_history_show, connection,
+ &connection->kref, drbd_destroy_connection);
+}
+
+static int callback_history_release(struct inode *inode, struct file *file)
+{
+ struct drbd_connection *connection = inode->i_private;
+ kref_put(&connection->kref, drbd_destroy_connection);
+ return single_release(inode, file);
+}
+
+static const struct file_operations connection_callback_history_fops = {
+ .owner = THIS_MODULE,
+ .open = callback_history_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = callback_history_release,
+};
+
+static int connection_oldest_requests_show(struct seq_file *m, void *ignored)
+{
+ struct drbd_connection *connection = m->private;
+ unsigned long now = jiffies;
+ struct drbd_request *r1, *r2;
+
+ /* BUMP me if you change the file format/content/presentation */
+ seq_printf(m, "v: %u\n\n", 0);
+
+ spin_lock_irq(&connection->resource->req_lock);
+ r1 = connection->req_next;
+ if (r1)
+ seq_print_minor_vnr_req(m, r1, now);
+ r2 = connection->req_ack_pending;
+ if (r2 && r2 != r1) {
+ r1 = r2;
+ seq_print_minor_vnr_req(m, r1, now);
+ }
+ r2 = connection->req_not_net_done;
+ if (r2 && r2 != r1)
+ seq_print_minor_vnr_req(m, r2, now);
+ spin_unlock_irq(&connection->resource->req_lock);
+ return 0;
+}
+
+static int connection_oldest_requests_open(struct inode *inode, struct file *file)
+{
+ struct drbd_connection *connection = inode->i_private;
+ return drbd_single_open(file, connection_oldest_requests_show, connection,
+ &connection->kref, drbd_destroy_connection);
+}
+
+static int connection_oldest_requests_release(struct inode *inode, struct file *file)
+{
+ struct drbd_connection *connection = inode->i_private;
+ kref_put(&connection->kref, drbd_destroy_connection);
+ return single_release(inode, file);
+}
+
+static const struct file_operations connection_oldest_requests_fops = {
+ .owner = THIS_MODULE,
+ .open = connection_oldest_requests_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = connection_oldest_requests_release,
+};
+
+void drbd_debugfs_connection_add(struct drbd_connection *connection)
+{
+ struct dentry *conns_dir = connection->resource->debugfs_res_connections;
+ struct dentry *dentry;
+ if (!conns_dir)
+ return;
+
+ /* Once we enable mutliple peers,
+ * these connections will have descriptive names.
+ * For now, it is just the one connection to the (only) "peer". */
+ dentry = debugfs_create_dir("peer", conns_dir);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ connection->debugfs_conn = dentry;
+
+ dentry = debugfs_create_file("callback_history", S_IRUSR|S_IRGRP,
+ connection->debugfs_conn, connection,
+ &connection_callback_history_fops);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ connection->debugfs_conn_callback_history = dentry;
+
+ dentry = debugfs_create_file("oldest_requests", S_IRUSR|S_IRGRP,
+ connection->debugfs_conn, connection,
+ &connection_oldest_requests_fops);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ connection->debugfs_conn_oldest_requests = dentry;
+ return;
+
+fail:
+ drbd_debugfs_connection_cleanup(connection);
+ drbd_err(connection, "failed to create debugfs dentry\n");
+}
+
+void drbd_debugfs_connection_cleanup(struct drbd_connection *connection)
+{
+ drbd_debugfs_remove(&connection->debugfs_conn_callback_history);
+ drbd_debugfs_remove(&connection->debugfs_conn_oldest_requests);
+ drbd_debugfs_remove(&connection->debugfs_conn);
+}
+
+static void resync_dump_detail(struct seq_file *m, struct lc_element *e)
+{
+ struct bm_extent *bme = lc_entry(e, struct bm_extent, lce);
+
+ seq_printf(m, "%5d %s %s %s\n", bme->rs_left,
+ test_bit(BME_NO_WRITES, &bme->flags) ? "NO_WRITES" : "---------",
+ test_bit(BME_LOCKED, &bme->flags) ? "LOCKED" : "------",
+ test_bit(BME_PRIORITY, &bme->flags) ? "PRIORITY" : "--------"
+ );
+}
+
+static int device_resync_extents_show(struct seq_file *m, void *ignored)
+{
+ struct drbd_device *device = m->private;
+
+ /* BUMP me if you change the file format/content/presentation */
+ seq_printf(m, "v: %u\n\n", 0);
+
+ if (get_ldev_if_state(device, D_FAILED)) {
+ lc_seq_printf_stats(m, device->resync);
+ lc_seq_dump_details(m, device->resync, "rs_left flags", resync_dump_detail);
+ put_ldev(device);
+ }
+ return 0;
+}
+
+static int device_act_log_extents_show(struct seq_file *m, void *ignored)
+{
+ struct drbd_device *device = m->private;
+
+ /* BUMP me if you change the file format/content/presentation */
+ seq_printf(m, "v: %u\n\n", 0);
+
+ if (get_ldev_if_state(device, D_FAILED)) {
+ lc_seq_printf_stats(m, device->act_log);
+ lc_seq_dump_details(m, device->act_log, "", NULL);
+ put_ldev(device);
+ }
+ return 0;
+}
+
+static int device_oldest_requests_show(struct seq_file *m, void *ignored)
+{
+ struct drbd_device *device = m->private;
+ struct drbd_resource *resource = device->resource;
+ unsigned long now = jiffies;
+ struct drbd_request *r1, *r2;
+ int i;
+
+ /* BUMP me if you change the file format/content/presentation */
+ seq_printf(m, "v: %u\n\n", 0);
+
+ seq_puts(m, RQ_HDR);
+ spin_lock_irq(&resource->req_lock);
+ /* WRITE, then READ */
+ for (i = 1; i >= 0; --i) {
+ r1 = list_first_entry_or_null(&device->pending_master_completion[i],
+ struct drbd_request, req_pending_master_completion);
+ r2 = list_first_entry_or_null(&device->pending_completion[i],
+ struct drbd_request, req_pending_local);
+ if (r1)
+ seq_print_one_request(m, r1, now);
+ if (r2 && r2 != r1)
+ seq_print_one_request(m, r2, now);
+ }
+ spin_unlock_irq(&resource->req_lock);
+ return 0;
+}
+
+static int device_data_gen_id_show(struct seq_file *m, void *ignored)
+{
+ struct drbd_device *device = m->private;
+ struct drbd_md *md;
+ enum drbd_uuid_index idx;
+
+ if (!get_ldev_if_state(device, D_FAILED))
+ return -ENODEV;
+
+ md = &device->ldev->md;
+ spin_lock_irq(&md->uuid_lock);
+ for (idx = UI_CURRENT; idx <= UI_HISTORY_END; idx++) {
+ seq_printf(m, "0x%016llX\n", md->uuid[idx]);
+ }
+ spin_unlock_irq(&md->uuid_lock);
+ put_ldev(device);
+ return 0;
+}
+
+#define drbd_debugfs_device_attr(name) \
+static int device_ ## name ## _open(struct inode *inode, struct file *file) \
+{ \
+ struct drbd_device *device = inode->i_private; \
+ return drbd_single_open(file, device_ ## name ## _show, device, \
+ &device->kref, drbd_destroy_device); \
+} \
+static int device_ ## name ## _release(struct inode *inode, struct file *file) \
+{ \
+ struct drbd_device *device = inode->i_private; \
+ kref_put(&device->kref, drbd_destroy_device); \
+ return single_release(inode, file); \
+} \
+static const struct file_operations device_ ## name ## _fops = { \
+ .owner = THIS_MODULE, \
+ .open = device_ ## name ## _open, \
+ .read = seq_read, \
+ .llseek = seq_lseek, \
+ .release = device_ ## name ## _release, \
+};
+
+drbd_debugfs_device_attr(oldest_requests)
+drbd_debugfs_device_attr(act_log_extents)
+drbd_debugfs_device_attr(resync_extents)
+drbd_debugfs_device_attr(data_gen_id)
+
+void drbd_debugfs_device_add(struct drbd_device *device)
+{
+ struct dentry *vols_dir = device->resource->debugfs_res_volumes;
+ char minor_buf[8]; /* MINORMASK, MINORBITS == 20; */
+ char vnr_buf[8]; /* volume number vnr is even 16 bit only; */
+ char *slink_name = NULL;
+
+ struct dentry *dentry;
+ if (!vols_dir || !drbd_debugfs_minors)
+ return;
+
+ snprintf(vnr_buf, sizeof(vnr_buf), "%u", device->vnr);
+ dentry = debugfs_create_dir(vnr_buf, vols_dir);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ device->debugfs_vol = dentry;
+
+ snprintf(minor_buf, sizeof(minor_buf), "%u", device->minor);
+ slink_name = kasprintf(GFP_KERNEL, "../resources/%s/volumes/%u",
+ device->resource->name, device->vnr);
+ if (!slink_name)
+ goto fail;
+ dentry = debugfs_create_symlink(minor_buf, drbd_debugfs_minors, slink_name);
+ kfree(slink_name);
+ slink_name = NULL;
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ device->debugfs_minor = dentry;
+
+#define DCF(name) do { \
+ dentry = debugfs_create_file(#name, S_IRUSR|S_IRGRP, \
+ device->debugfs_vol, device, \
+ &device_ ## name ## _fops); \
+ if (IS_ERR_OR_NULL(dentry)) \
+ goto fail; \
+ device->debugfs_vol_ ## name = dentry; \
+ } while (0)
+
+ DCF(oldest_requests);
+ DCF(act_log_extents);
+ DCF(resync_extents);
+ DCF(data_gen_id);
+#undef DCF
+ return;
+
+fail:
+ drbd_debugfs_device_cleanup(device);
+ drbd_err(device, "failed to create debugfs entries\n");
+}
+
+void drbd_debugfs_device_cleanup(struct drbd_device *device)
+{
+ drbd_debugfs_remove(&device->debugfs_minor);
+ drbd_debugfs_remove(&device->debugfs_vol_oldest_requests);
+ drbd_debugfs_remove(&device->debugfs_vol_act_log_extents);
+ drbd_debugfs_remove(&device->debugfs_vol_resync_extents);
+ drbd_debugfs_remove(&device->debugfs_vol_data_gen_id);
+ drbd_debugfs_remove(&device->debugfs_vol);
+}
+
+void drbd_debugfs_peer_device_add(struct drbd_peer_device *peer_device)
+{
+ struct dentry *conn_dir = peer_device->connection->debugfs_conn;
+ struct dentry *dentry;
+ char vnr_buf[8];
+
+ if (!conn_dir)
+ return;
+
+ snprintf(vnr_buf, sizeof(vnr_buf), "%u", peer_device->device->vnr);
+ dentry = debugfs_create_dir(vnr_buf, conn_dir);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ peer_device->debugfs_peer_dev = dentry;
+ return;
+
+fail:
+ drbd_debugfs_peer_device_cleanup(peer_device);
+ drbd_err(peer_device, "failed to create debugfs entries\n");
+}
+
+void drbd_debugfs_peer_device_cleanup(struct drbd_peer_device *peer_device)
+{
+ drbd_debugfs_remove(&peer_device->debugfs_peer_dev);
+}
+
+static int drbd_version_show(struct seq_file *m, void *ignored)
+{
+ seq_printf(m, "# %s\n", drbd_buildtag());
+ seq_printf(m, "VERSION=%s\n", REL_VERSION);
+ seq_printf(m, "API_VERSION=%u\n", API_VERSION);
+ seq_printf(m, "PRO_VERSION_MIN=%u\n", PRO_VERSION_MIN);
+ seq_printf(m, "PRO_VERSION_MAX=%u\n", PRO_VERSION_MAX);
+ return 0;
+}
+
+static int drbd_version_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, drbd_version_show, NULL);
+}
+
+static struct file_operations drbd_version_fops = {
+ .owner = THIS_MODULE,
+ .open = drbd_version_open,
+ .llseek = seq_lseek,
+ .read = seq_read,
+ .release = single_release,
+};
+
+/* not __exit, may be indirectly called
+ * from the module-load-failure path as well. */
+void drbd_debugfs_cleanup(void)
+{
+ drbd_debugfs_remove(&drbd_debugfs_resources);
+ drbd_debugfs_remove(&drbd_debugfs_minors);
+ drbd_debugfs_remove(&drbd_debugfs_version);
+ drbd_debugfs_remove(&drbd_debugfs_root);
+}
+
+int __init drbd_debugfs_init(void)
+{
+ struct dentry *dentry;
+
+ dentry = debugfs_create_dir("drbd", NULL);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ drbd_debugfs_root = dentry;
+
+ dentry = debugfs_create_file("version", 0444, drbd_debugfs_root, NULL, &drbd_version_fops);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ drbd_debugfs_version = dentry;
+
+ dentry = debugfs_create_dir("resources", drbd_debugfs_root);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ drbd_debugfs_resources = dentry;
+
+ dentry = debugfs_create_dir("minors", drbd_debugfs_root);
+ if (IS_ERR_OR_NULL(dentry))
+ goto fail;
+ drbd_debugfs_minors = dentry;
+ return 0;
+
+fail:
+ drbd_debugfs_cleanup();
+ if (dentry)
+ return PTR_ERR(dentry);
+ else
+ return -EINVAL;
+}
--- /dev/null
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/debugfs.h>
+
+#include "drbd_int.h"
+
+#ifdef CONFIG_DEBUG_FS
+int __init drbd_debugfs_init(void);
+void drbd_debugfs_cleanup(void);
+
+void drbd_debugfs_resource_add(struct drbd_resource *resource);
+void drbd_debugfs_resource_cleanup(struct drbd_resource *resource);
+
+void drbd_debugfs_connection_add(struct drbd_connection *connection);
+void drbd_debugfs_connection_cleanup(struct drbd_connection *connection);
+
+void drbd_debugfs_device_add(struct drbd_device *device);
+void drbd_debugfs_device_cleanup(struct drbd_device *device);
+
+void drbd_debugfs_peer_device_add(struct drbd_peer_device *peer_device);
+void drbd_debugfs_peer_device_cleanup(struct drbd_peer_device *peer_device);
+#else
+
+static inline int __init drbd_debugfs_init(void) { return -ENODEV; }
+static inline void drbd_debugfs_cleanup(void) { }
+
+static inline void drbd_debugfs_resource_add(struct drbd_resource *resource) { }
+static inline void drbd_debugfs_resource_cleanup(struct drbd_resource *resource) { }
+
+static inline void drbd_debugfs_connection_add(struct drbd_connection *connection) { }
+static inline void drbd_debugfs_connection_cleanup(struct drbd_connection *connection) { }
+
+static inline void drbd_debugfs_device_add(struct drbd_device *device) { }
+static inline void drbd_debugfs_device_cleanup(struct drbd_device *device) { }
+
+static inline void drbd_debugfs_peer_device_add(struct drbd_peer_device *peer_device) { }
+static inline void drbd_debugfs_peer_device_cleanup(struct drbd_peer_device *peer_device) { }
+
+#endif
struct list_head tl_requests; /* ring list in the transfer log */
struct bio *master_bio; /* master bio pointer */
- unsigned long start_time;
+
+ /* see struct drbd_device */
+ struct list_head req_pending_master_completion;
+ struct list_head req_pending_local;
+
+ /* for generic IO accounting */
+ unsigned long start_jif;
+
+ /* for DRBD internal statistics */
+
+ /* Minimal set of time stamps to determine if we wait for activity log
+ * transactions, local disk or peer. 32 bit "jiffies" are good enough,
+ * we don't expect a DRBD request to be stalled for several month.
+ */
+
+ /* before actual request processing */
+ unsigned long in_actlog_jif;
+
+ /* local disk */
+ unsigned long pre_submit_jif;
+
+ /* per connection */
+ unsigned long pre_send_jif;
+ unsigned long acked_jif;
+ unsigned long net_done_jif;
+
+ /* Possibly even more detail to track each phase:
+ * master_completion_jif
+ * how long did it take to complete the master bio
+ * (application visible latency)
+ * allocated_jif
+ * how long the master bio was blocked until we finally allocated
+ * a tracking struct
+ * in_actlog_jif
+ * how long did we wait for activity log transactions
+ *
+ * net_queued_jif
+ * when did we finally queue it for sending
+ * pre_send_jif
+ * when did we start sending it
+ * post_send_jif
+ * how long did we block in the network stack trying to send it
+ * acked_jif
+ * when did we receive (or fake, in protocol A) a remote ACK
+ * net_done_jif
+ * when did we receive final acknowledgement (P_BARRIER_ACK),
+ * or decide, e.g. on connection loss, that we do no longer expect
+ * anything from this peer for this request.
+ *
+ * pre_submit_jif
+ * post_sub_jif
+ * when did we start submiting to the lower level device,
+ * and how long did we block in that submit function
+ * local_completion_jif
+ * how long did it take the lower level device to complete this request
+ */
+
/* once it hits 0, we may complete the master_bio */
atomic_t completion_ref;
struct drbd_interval i;
/* see comments on ee flag bits below */
unsigned long flags;
+ unsigned long submit_jif;
union {
u64 block_id;
struct digest_info *digest;
/* Is set when net_conf had two_primaries set while creating this peer_req */
__EE_IN_INTERVAL_TREE,
+
+ /* for debugfs: */
+ /* has this been submitted, or does it still wait for something else? */
+ __EE_SUBMITTED,
+
+ /* this is/was a write request */
+ __EE_WRITE,
+
+ /* this originates from application on peer
+ * (not some resync or verify or other DRBD internal request) */
+ __EE_APPLICATION,
};
#define EE_CALL_AL_COMPLETE_IO (1<<__EE_CALL_AL_COMPLETE_IO)
#define EE_MAY_SET_IN_SYNC (1<<__EE_MAY_SET_IN_SYNC)
#define EE_RESTART_REQUESTS (1<<__EE_RESTART_REQUESTS)
#define EE_SEND_WRITE_ACK (1<<__EE_SEND_WRITE_ACK)
#define EE_IN_INTERVAL_TREE (1<<__EE_IN_INTERVAL_TREE)
+#define EE_SUBMITTED (1<<__EE_SUBMITTED)
+#define EE_WRITE (1<<__EE_WRITE)
+#define EE_APPLICATION (1<<__EE_APPLICATION)
/* flag bits per device */
enum {
CONSIDER_RESYNC,
MD_NO_FUA, /* Users wants us to not use FUA/FLUSH on meta data dev */
+
SUSPEND_IO, /* suspend application io */
BITMAP_IO, /* suspend application io;
once no more io in flight, start bitmap io */
BITMAP_IO_QUEUED, /* Started bitmap IO */
- GO_DISKLESS, /* Disk is being detached, on io-error or admin request. */
WAS_IO_ERROR, /* Local disk failed, returned IO error */
WAS_READ_ERROR, /* Local disk READ failed (set additionally to the above) */
FORCE_DETACH, /* Force-detach from local disk, aborting any pending local IO */
B_RS_H_DONE, /* Before resync handler done (already executed) */
DISCARD_MY_DATA, /* discard_my_data flag per volume */
READ_BALANCE_RR,
+
+ FLUSH_PENDING, /* if set, device->flush_jif is when we submitted that flush
+ * from drbd_flush_after_epoch() */
+
+ /* cleared only after backing device related structures have been destroyed. */
+ GOING_DISKLESS, /* Disk is being detached, because of io-error, or admin request. */
+
+ /* to be used in drbd_device_post_work() */
+ GO_DISKLESS, /* tell worker to schedule cleanup before detach */
+ DESTROY_DISK, /* tell worker to close backing devices and destroy related structures. */
+ MD_SYNC, /* tell worker to call drbd_md_sync() */
+ RS_START, /* tell worker to start resync/OV */
+ RS_PROGRESS, /* tell worker that resync made significant progress */
+ RS_DONE, /* tell worker that resync is done */
};
struct drbd_bitmap; /* opaque for drbd_device */
};
struct drbd_md_io {
+ struct page *page;
+ unsigned long start_jif; /* last call to drbd_md_get_buffer */
+ unsigned long submit_jif; /* last _drbd_md_sync_page_io() submit */
+ const char *current_use;
+ atomic_t in_use;
unsigned int done;
int error;
};
* and potentially deadlock on, this drbd worker.
*/
DISCONNECT_SENT,
+
+ DEVICE_WORK_PENDING, /* tell worker that some device has pending work */
};
struct drbd_resource {
char *name;
+#ifdef CONFIG_DEBUG_FS
+ struct dentry *debugfs_res;
+ struct dentry *debugfs_res_volumes;
+ struct dentry *debugfs_res_connections;
+ struct dentry *debugfs_res_in_flight_summary;
+#endif
struct kref kref;
struct idr devices; /* volume number to device mapping */
struct list_head connections;
unsigned susp_nod:1; /* IO suspended because no data */
unsigned susp_fen:1; /* IO suspended because fence peer handler runs */
+ enum write_ordering_e write_ordering;
+
cpumask_var_t cpu_mask;
};
+struct drbd_thread_timing_details
+{
+ unsigned long start_jif;
+ void *cb_addr;
+ const char *caller_fn;
+ unsigned int line;
+ unsigned int cb_nr;
+};
+
struct drbd_connection {
struct list_head connections;
struct drbd_resource *resource;
+#ifdef CONFIG_DEBUG_FS
+ struct dentry *debugfs_conn;
+ struct dentry *debugfs_conn_callback_history;
+ struct dentry *debugfs_conn_oldest_requests;
+#endif
struct kref kref;
struct idr peer_devices; /* volume number to peer device mapping */
enum drbd_conns cstate; /* Only C_STANDALONE to C_WF_REPORT_PARAMS */
struct drbd_epoch *current_epoch;
spinlock_t epoch_lock;
unsigned int epochs;
- enum write_ordering_e write_ordering;
atomic_t current_tle_nr; /* transfer log epoch number */
unsigned current_tle_writes; /* writes seen within this tl epoch */
struct drbd_thread worker;
struct drbd_thread asender;
+ /* cached pointers,
+ * so we can look up the oldest pending requests more quickly.
+ * protected by resource->req_lock */
+ struct drbd_request *req_next; /* DRBD 9: todo.req_next */
+ struct drbd_request *req_ack_pending;
+ struct drbd_request *req_not_net_done;
+
/* sender side */
struct drbd_work_queue sender_work;
+#define DRBD_THREAD_DETAILS_HIST 16
+ unsigned int w_cb_nr; /* keeps counting up */
+ unsigned int r_cb_nr; /* keeps counting up */
+ struct drbd_thread_timing_details w_timing_details[DRBD_THREAD_DETAILS_HIST];
+ struct drbd_thread_timing_details r_timing_details[DRBD_THREAD_DETAILS_HIST];
+
struct {
/* whether this sender thread
* has processed a single write yet. */
} send;
};
+void __update_timing_details(
+ struct drbd_thread_timing_details *tdp,
+ unsigned int *cb_nr,
+ void *cb,
+ const char *fn, const unsigned int line);
+
+#define update_worker_timing_details(c, cb) \
+ __update_timing_details(c->w_timing_details, &c->w_cb_nr, cb, __func__ , __LINE__ )
+#define update_receiver_timing_details(c, cb) \
+ __update_timing_details(c->r_timing_details, &c->r_cb_nr, cb, __func__ , __LINE__ )
+
struct submit_worker {
struct workqueue_struct *wq;
struct work_struct worker;
- spinlock_t lock;
+ /* protected by ..->resource->req_lock */
struct list_head writes;
};
struct list_head peer_devices;
struct drbd_device *device;
struct drbd_connection *connection;
+#ifdef CONFIG_DEBUG_FS
+ struct dentry *debugfs_peer_dev;
+#endif
};
struct drbd_device {
struct drbd_resource *resource;
struct list_head peer_devices;
- int vnr; /* volume number within the connection */
+ struct list_head pending_bitmap_io;
+
+ unsigned long flush_jif;
+#ifdef CONFIG_DEBUG_FS
+ struct dentry *debugfs_minor;
+ struct dentry *debugfs_vol;
+ struct dentry *debugfs_vol_oldest_requests;
+ struct dentry *debugfs_vol_act_log_extents;
+ struct dentry *debugfs_vol_resync_extents;
+ struct dentry *debugfs_vol_data_gen_id;
+#endif
+
+ unsigned int vnr; /* volume number within the connection */
+ unsigned int minor; /* device minor number */
+
struct kref kref;
/* things that are stored as / read from meta data on disk */
unsigned long last_reattach_jif;
struct drbd_work resync_work;
struct drbd_work unplug_work;
- struct drbd_work go_diskless;
- struct drbd_work md_sync_work;
- struct drbd_work start_resync_work;
struct timer_list resync_timer;
struct timer_list md_sync_timer;
struct timer_list start_resync_timer;
struct timer_list request_timer;
-#ifdef DRBD_DEBUG_MD_SYNC
- struct {
- unsigned int line;
- const char* func;
- } last_md_mark_dirty;
-#endif
/* Used after attach while negotiating new disk state. */
union drbd_state new_state_tmp;
unsigned int al_writ_cnt;
unsigned int bm_writ_cnt;
atomic_t ap_bio_cnt; /* Requests we need to complete */
+ atomic_t ap_actlog_cnt; /* Requests waiting for activity log */
atomic_t ap_pending_cnt; /* AP data packets on the wire, ack expected */
atomic_t rs_pending_cnt; /* RS request/data packets on the wire */
atomic_t unacked_cnt; /* Need to send replies for */
struct rb_root read_requests;
struct rb_root write_requests;
+ /* for statistics and timeouts */
+ /* [0] read, [1] write */
+ struct list_head pending_master_completion[2];
+ struct list_head pending_completion[2];
+
+ /* use checksums for *this* resync */
+ bool use_csums;
/* blocks to resync in this run [unit BM_BLOCK_SIZE] */
unsigned long rs_total;
/* number of resync blocks that failed in this run */
atomic_t pp_in_use; /* allocated from page pool */
atomic_t pp_in_use_by_net; /* sendpage()d, still referenced by tcp */
wait_queue_head_t ee_wait;
- struct page *md_io_page; /* one page buffer for md_io */
struct drbd_md_io md_io;
- atomic_t md_io_in_use; /* protects the md_io, md_io_page and md_io_tmpp */
spinlock_t al_lock;
wait_queue_head_t al_wait;
struct lru_cache *act_log; /* activity log */
atomic_t packet_seq;
unsigned int peer_seq;
spinlock_t peer_seq_lock;
- unsigned int minor;
unsigned long comm_bm_set; /* communicated number of set bits. */
struct bm_io_work bm_io_work;
u64 ed_uuid; /* UUID of the exposed data */
struct submit_worker submit;
};
+struct drbd_bm_aio_ctx {
+ struct drbd_device *device;
+ struct list_head list; /* on device->pending_bitmap_io */;
+ unsigned long start_jif;
+ atomic_t in_flight;
+ unsigned int done;
+ unsigned flags;
+#define BM_AIO_COPY_PAGES 1
+#define BM_AIO_WRITE_HINTED 2
+#define BM_AIO_WRITE_ALL_PAGES 4
+#define BM_AIO_READ 8
+ int error;
+ struct kref kref;
+};
+
struct drbd_config_context {
/* assigned from drbd_genlmsghdr */
unsigned int minor;
extern int drbd_send_bitmap(struct drbd_device *device);
extern void drbd_send_sr_reply(struct drbd_peer_device *, enum drbd_state_rv retcode);
extern void conn_send_sr_reply(struct drbd_connection *connection, enum drbd_state_rv retcode);
-extern void drbd_free_bc(struct drbd_backing_dev *ldev);
+extern void drbd_free_ldev(struct drbd_backing_dev *ldev);
extern void drbd_device_cleanup(struct drbd_device *device);
void drbd_print_uuids(struct drbd_device *device, const char *text);
extern void drbd_md_set_flag(struct drbd_device *device, int flags) __must_hold(local);
extern void drbd_md_clear_flag(struct drbd_device *device, int flags)__must_hold(local);
extern int drbd_md_test_flag(struct drbd_backing_dev *, int);
-#ifndef DRBD_DEBUG_MD_SYNC
extern void drbd_md_mark_dirty(struct drbd_device *device);
-#else
-#define drbd_md_mark_dirty(m) drbd_md_mark_dirty_(m, __LINE__ , __func__ )
-extern void drbd_md_mark_dirty_(struct drbd_device *device,
- unsigned int line, const char *func);
-#endif
extern void drbd_queue_bitmap_io(struct drbd_device *device,
int (*io_fn)(struct drbd_device *),
void (*done)(struct drbd_device *, int),
extern int drbd_bitmap_io_from_worker(struct drbd_device *device,
int (*io_fn)(struct drbd_device *),
char *why, enum bm_flag flags);
-extern int drbd_bmio_set_n_write(struct drbd_device *device);
-extern int drbd_bmio_clear_n_write(struct drbd_device *device);
-extern void drbd_ldev_destroy(struct drbd_device *device);
+extern int drbd_bmio_set_n_write(struct drbd_device *device) __must_hold(local);
+extern int drbd_bmio_clear_n_write(struct drbd_device *device) __must_hold(local);
/* Meta data layout
*
/* in which _bitmap_ extent (resp. sector) the bit for a certain
* _storage_ sector is located in */
#define BM_SECT_TO_EXT(x) ((x)>>(BM_EXT_SHIFT-9))
+#define BM_BIT_TO_EXT(x) ((x) >> (BM_EXT_SHIFT - BM_BLOCK_SHIFT))
-/* how much _storage_ sectors we have per bitmap sector */
+/* first storage sector a bitmap extent corresponds to */
#define BM_EXT_TO_SECT(x) ((sector_t)(x) << (BM_EXT_SHIFT-9))
+/* how much _storage_ sectors we have per bitmap extent */
#define BM_SECT_PER_EXT BM_EXT_TO_SECT(1)
+/* how many bits are covered by one bitmap extent (resync extent) */
+#define BM_BITS_PER_EXT (1UL << (BM_EXT_SHIFT - BM_BLOCK_SHIFT))
+
+#define BM_BLOCKS_PER_BM_EXT_MASK (BM_BITS_PER_EXT - 1)
+
/* in one sector of the bitmap, we have this many activity_log extents. */
#define AL_EXT_PER_BM_SECT (1 << (BM_EXT_SHIFT - AL_EXTENT_SHIFT))
-#define BM_BLOCKS_PER_BM_EXT_B (BM_EXT_SHIFT - BM_BLOCK_SHIFT)
-#define BM_BLOCKS_PER_BM_EXT_MASK ((1<<BM_BLOCKS_PER_BM_EXT_B) - 1)
-
/* the extent in "PER_EXTENT" below is an activity log extent
* we need that many (long words/bytes) to store the bitmap
* of one AL_EXTENT_SIZE chunk of storage.
const unsigned long s, const unsigned long e);
extern int drbd_bm_test_bit(struct drbd_device *device, unsigned long bitnr);
extern int drbd_bm_e_weight(struct drbd_device *device, unsigned long enr);
-extern int drbd_bm_write_page(struct drbd_device *device, unsigned int idx) __must_hold(local);
extern int drbd_bm_read(struct drbd_device *device) __must_hold(local);
extern void drbd_bm_mark_for_writeout(struct drbd_device *device, int page_nr);
extern int drbd_bm_write(struct drbd_device *device) __must_hold(local);
extern int drbd_bm_write_hinted(struct drbd_device *device) __must_hold(local);
+extern int drbd_bm_write_lazy(struct drbd_device *device, unsigned upper_idx) __must_hold(local);
extern int drbd_bm_write_all(struct drbd_device *device) __must_hold(local);
extern int drbd_bm_write_copy_pages(struct drbd_device *device) __must_hold(local);
extern size_t drbd_bm_words(struct drbd_device *device);
extern unsigned long _drbd_bm_find_next_zero(struct drbd_device *device, unsigned long bm_fo);
extern unsigned long _drbd_bm_total_weight(struct drbd_device *device);
extern unsigned long drbd_bm_total_weight(struct drbd_device *device);
-extern int drbd_bm_rs_done(struct drbd_device *device);
/* for receive_bitmap */
extern void drbd_bm_merge_lel(struct drbd_device *device, size_t offset,
size_t number, unsigned long *buffer);
extern enum determine_dev_size
drbd_determine_dev_size(struct drbd_device *, enum dds_flags, struct resize_parms *) __must_hold(local);
extern void resync_after_online_grow(struct drbd_device *);
-extern void drbd_reconsider_max_bio_size(struct drbd_device *device);
+extern void drbd_reconsider_max_bio_size(struct drbd_device *device, struct drbd_backing_dev *bdev);
extern enum drbd_state_rv drbd_set_role(struct drbd_device *device,
enum drbd_role new_role,
int force);
extern void suspend_other_sg(struct drbd_device *device);
extern int drbd_resync_finished(struct drbd_device *device);
/* maybe rather drbd_main.c ? */
-extern void *drbd_md_get_buffer(struct drbd_device *device);
+extern void *drbd_md_get_buffer(struct drbd_device *device, const char *intent);
extern void drbd_md_put_buffer(struct drbd_device *device);
extern int drbd_md_sync_page_io(struct drbd_device *device,
struct drbd_backing_dev *bdev, sector_t sector, int rw);
extern int drbd_receiver(struct drbd_thread *thi);
extern int drbd_asender(struct drbd_thread *thi);
extern bool drbd_rs_c_min_rate_throttle(struct drbd_device *device);
-extern bool drbd_rs_should_slow_down(struct drbd_device *device, sector_t sector);
+extern bool drbd_rs_should_slow_down(struct drbd_device *device, sector_t sector,
+ bool throttle_if_app_is_waiting);
extern int drbd_submit_peer_request(struct drbd_device *,
struct drbd_peer_request *, const unsigned,
const int);
{
__release(local);
if (!bio->bi_bdev) {
- printk(KERN_ERR "drbd%d: drbd_generic_make_request: "
- "bio->bi_bdev == NULL\n",
- device_to_minor(device));
- dump_stack();
+ drbd_err(device, "drbd_generic_make_request: bio->bi_bdev == NULL\n");
bio_endio(bio, -ENODEV);
return;
}
generic_make_request(bio);
}
-void drbd_bump_write_ordering(struct drbd_connection *connection, enum write_ordering_e wo);
+void drbd_bump_write_ordering(struct drbd_resource *resource, struct drbd_backing_dev *bdev,
+ enum write_ordering_e wo);
/* drbd_proc.c */
extern struct proc_dir_entry *drbd_proc;
/* drbd_actlog.c */
extern bool drbd_al_begin_io_prepare(struct drbd_device *device, struct drbd_interval *i);
extern int drbd_al_begin_io_nonblock(struct drbd_device *device, struct drbd_interval *i);
-extern void drbd_al_begin_io_commit(struct drbd_device *device, bool delegate);
+extern void drbd_al_begin_io_commit(struct drbd_device *device);
extern bool drbd_al_begin_io_fastpath(struct drbd_device *device, struct drbd_interval *i);
-extern void drbd_al_begin_io(struct drbd_device *device, struct drbd_interval *i, bool delegate);
+extern void drbd_al_begin_io(struct drbd_device *device, struct drbd_interval *i);
extern void drbd_al_complete_io(struct drbd_device *device, struct drbd_interval *i);
extern void drbd_rs_complete_io(struct drbd_device *device, sector_t sector);
extern int drbd_rs_begin_io(struct drbd_device *device, sector_t sector);
extern void drbd_rs_failed_io(struct drbd_device *device,
sector_t sector, int size);
extern void drbd_advance_rs_marks(struct drbd_device *device, unsigned long still_to_go);
-extern void __drbd_set_in_sync(struct drbd_device *device, sector_t sector,
- int size, const char *file, const unsigned int line);
+
+enum update_sync_bits_mode { RECORD_RS_FAILED, SET_OUT_OF_SYNC, SET_IN_SYNC };
+extern int __drbd_change_sync(struct drbd_device *device, sector_t sector, int size,
+ enum update_sync_bits_mode mode,
+ const char *file, const unsigned int line);
#define drbd_set_in_sync(device, sector, size) \
- __drbd_set_in_sync(device, sector, size, __FILE__, __LINE__)
-extern int __drbd_set_out_of_sync(struct drbd_device *device, sector_t sector,
- int size, const char *file, const unsigned int line);
+ __drbd_change_sync(device, sector, size, SET_IN_SYNC, __FILE__, __LINE__)
#define drbd_set_out_of_sync(device, sector, size) \
- __drbd_set_out_of_sync(device, sector, size, __FILE__, __LINE__)
+ __drbd_change_sync(device, sector, size, SET_OUT_OF_SYNC, __FILE__, __LINE__)
+#define drbd_rs_failed_io(device, sector, size) \
+ __drbd_change_sync(device, sector, size, RECORD_RS_FAILED, __FILE__, __LINE__)
extern void drbd_al_shrink(struct drbd_device *device);
extern int drbd_initialize_al(struct drbd_device *, void *);
}
static inline void
-drbd_queue_work_front(struct drbd_work_queue *q, struct drbd_work *w)
+drbd_queue_work(struct drbd_work_queue *q, struct drbd_work *w)
{
unsigned long flags;
spin_lock_irqsave(&q->q_lock, flags);
- list_add(&w->list, &q->q);
+ list_add_tail(&w->list, &q->q);
spin_unlock_irqrestore(&q->q_lock, flags);
wake_up(&q->q_wait);
}
static inline void
-drbd_queue_work(struct drbd_work_queue *q, struct drbd_work *w)
+drbd_queue_work_if_unqueued(struct drbd_work_queue *q, struct drbd_work *w)
{
unsigned long flags;
spin_lock_irqsave(&q->q_lock, flags);
- list_add_tail(&w->list, &q->q);
+ if (list_empty_careful(&w->list))
+ list_add_tail(&w->list, &q->q);
spin_unlock_irqrestore(&q->q_lock, flags);
wake_up(&q->q_wait);
}
+static inline void
+drbd_device_post_work(struct drbd_device *device, int work_bit)
+{
+ if (!test_and_set_bit(work_bit, &device->flags)) {
+ struct drbd_connection *connection =
+ first_peer_device(device)->connection;
+ struct drbd_work_queue *q = &connection->sender_work;
+ if (!test_and_set_bit(DEVICE_WORK_PENDING, &connection->flags))
+ wake_up(&q->q_wait);
+ }
+}
+
extern void drbd_flush_workqueue(struct drbd_work_queue *work_queue);
static inline void wake_asender(struct drbd_connection *connection)
func, line, \
atomic_read(&device->which))
-#define dec_ap_pending(device) _dec_ap_pending(device, __FUNCTION__, __LINE__)
+#define dec_ap_pending(device) _dec_ap_pending(device, __func__, __LINE__)
static inline void _dec_ap_pending(struct drbd_device *device, const char *func, int line)
{
if (atomic_dec_and_test(&device->ap_pending_cnt))
atomic_inc(&device->rs_pending_cnt);
}
-#define dec_rs_pending(device) _dec_rs_pending(device, __FUNCTION__, __LINE__)
+#define dec_rs_pending(device) _dec_rs_pending(device, __func__, __LINE__)
static inline void _dec_rs_pending(struct drbd_device *device, const char *func, int line)
{
atomic_dec(&device->rs_pending_cnt);
atomic_inc(&device->unacked_cnt);
}
-#define dec_unacked(device) _dec_unacked(device, __FUNCTION__, __LINE__)
+#define dec_unacked(device) _dec_unacked(device, __func__, __LINE__)
static inline void _dec_unacked(struct drbd_device *device, const char *func, int line)
{
atomic_dec(&device->unacked_cnt);
ERR_IF_CNT_IS_NEGATIVE(unacked_cnt, func, line);
}
-#define sub_unacked(device, n) _sub_unacked(device, n, __FUNCTION__, __LINE__)
+#define sub_unacked(device, n) _sub_unacked(device, n, __func__, __LINE__)
static inline void _sub_unacked(struct drbd_device *device, int n, const char *func, int line)
{
atomic_sub(n, &device->unacked_cnt);
ERR_IF_CNT_IS_NEGATIVE(unacked_cnt, func, line);
}
+static inline bool is_sync_state(enum drbd_conns connection_state)
+{
+ return
+ (connection_state == C_SYNC_SOURCE
+ || connection_state == C_SYNC_TARGET
+ || connection_state == C_PAUSED_SYNC_S
+ || connection_state == C_PAUSED_SYNC_T);
+}
+
/**
* get_ldev() - Increase the ref count on device->ldev. Returns 0 if there is no ldev
* @M: DRBD device.
static inline void put_ldev(struct drbd_device *device)
{
+ enum drbd_disk_state ds = device->state.disk;
+ /* We must check the state *before* the atomic_dec becomes visible,
+ * or we have a theoretical race where someone hitting zero,
+ * while state still D_FAILED, will then see D_DISKLESS in the
+ * condition below and calling into destroy, where he must not, yet. */
int i = atomic_dec_return(&device->local_cnt);
/* This may be called from some endio handler,
__release(local);
D_ASSERT(device, i >= 0);
if (i == 0) {
- if (device->state.disk == D_DISKLESS)
+ if (ds == D_DISKLESS)
/* even internal references gone, safe to destroy */
- drbd_ldev_destroy(device);
- if (device->state.disk == D_FAILED) {
+ drbd_device_post_work(device, DESTROY_DISK);
+ if (ds == D_FAILED)
/* all application IO references gone. */
- if (!test_and_set_bit(GO_DISKLESS, &device->flags))
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
- &device->go_diskless);
- }
+ if (!test_and_set_bit(GOING_DISKLESS, &device->flags))
+ drbd_device_post_work(device, GO_DISKLESS);
wake_up(&device->misc_wait);
}
}
extern int _get_ldev_if_state(struct drbd_device *device, enum drbd_disk_state mins);
#endif
-/* you must have an "get_ldev" reference */
-static inline void drbd_get_syncer_progress(struct drbd_device *device,
- unsigned long *bits_left, unsigned int *per_mil_done)
-{
- /* this is to break it at compile time when we change that, in case we
- * want to support more than (1<<32) bits on a 32bit arch. */
- typecheck(unsigned long, device->rs_total);
-
- /* note: both rs_total and rs_left are in bits, i.e. in
- * units of BM_BLOCK_SIZE.
- * for the percentage, we don't care. */
-
- if (device->state.conn == C_VERIFY_S || device->state.conn == C_VERIFY_T)
- *bits_left = device->ov_left;
- else
- *bits_left = drbd_bm_total_weight(device) - device->rs_failed;
- /* >> 10 to prevent overflow,
- * +1 to prevent division by zero */
- if (*bits_left > device->rs_total) {
- /* doh. maybe a logic bug somewhere.
- * may also be just a race condition
- * between this and a disconnect during sync.
- * for now, just prevent in-kernel buffer overflow.
- */
- smp_rmb();
- drbd_warn(device, "cs:%s rs_left=%lu > rs_total=%lu (rs_failed %lu)\n",
- drbd_conn_str(device->state.conn),
- *bits_left, device->rs_total, device->rs_failed);
- *per_mil_done = 0;
- } else {
- /* Make sure the division happens in long context.
- * We allow up to one petabyte storage right now,
- * at a granularity of 4k per bit that is 2**38 bits.
- * After shift right and multiplication by 1000,
- * this should still fit easily into a 32bit long,
- * so we don't need a 64bit division on 32bit arch.
- * Note: currently we don't support such large bitmaps on 32bit
- * arch anyways, but no harm done to be prepared for it here.
- */
- unsigned int shift = device->rs_total > UINT_MAX ? 16 : 10;
- unsigned long left = *bits_left >> shift;
- unsigned long total = 1UL + (device->rs_total >> shift);
- unsigned long tmp = 1000UL - left * 1000UL/total;
- *per_mil_done = tmp;
- }
-}
-
-
/* this throttles on-the-fly application requests
* according to max_buffers settings;
* maybe re-implement using semaphores? */
return QUEUE_ORDERED_NONE;
}
-static inline void drbd_md_flush(struct drbd_device *device)
-{
- int r;
-
- if (device->ldev == NULL) {
- drbd_warn(device, "device->ldev == NULL in drbd_md_flush\n");
- return;
- }
-
- if (test_bit(MD_NO_FUA, &device->flags))
- return;
-
- r = blkdev_issue_flush(device->ldev->md_bdev, GFP_NOIO, NULL);
- if (r) {
- set_bit(MD_NO_FUA, &device->flags);
- drbd_err(device, "meta data flush failed with status %d, disabling md-flushes\n", r);
- }
-}
-
static inline struct drbd_connection *first_connection(struct drbd_resource *resource)
{
return list_first_entry_or_null(&resource->connections,
unsigned int size; /* size in bytes */
sector_t end; /* highest interval end in subtree */
int local:1 /* local or remote request? */;
- int waiting:1;
+ int waiting:1; /* someone is waiting for this to complete */
+ int completed:1; /* this has been completed already;
+ * ignore for conflict detection */
};
static inline void drbd_clear_interval(struct drbd_interval *i)
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/module.h>
+#include <linux/jiffies.h>
#include <linux/drbd.h>
#include <asm/uaccess.h>
#include <asm/types.h>
#include "drbd_int.h"
#include "drbd_protocol.h"
#include "drbd_req.h" /* only for _req_mod in tl_release and tl_clear */
-
#include "drbd_vli.h"
+#include "drbd_debugfs.h"
static DEFINE_MUTEX(drbd_main_mutex);
static int drbd_open(struct block_device *bdev, fmode_t mode);
static void drbd_release(struct gendisk *gd, fmode_t mode);
-static int w_md_sync(struct drbd_work *w, int unused);
static void md_sync_timer_fn(unsigned long data);
static int w_bitmap_io(struct drbd_work *w, int unused);
-static int w_go_diskless(struct drbd_work *w, int unused);
MODULE_AUTHOR("Philipp Reisner <phil@linbit.com>, "
"Lars Ellenberg <lars@linbit.com>");
/**
* _tl_restart() - Walks the transfer log, and applies an action to all requests
- * @device: DRBD device.
+ * @connection: DRBD connection to operate on.
* @what: The action/event to perform with all request objects
*
* @what might be one of CONNECTION_LOST_WHILE_PENDING, RESEND, FAIL_FROZEN_DISK_IO,
msg_flags);
if (data && !err)
err = drbd_send_all(connection, sock->socket, data, size, 0);
+ /* DRBD protocol "pings" are latency critical.
+ * This is supposed to trigger tcp_push_pending_frames() */
+ if (!err && (cmd == P_PING || cmd == P_PING_ACK))
+ drbd_tcp_nodelay(sock->socket);
+
return err;
}
if (peer_device->connection->agreed_pro_version >= 100) {
if (req->rq_state & RQ_EXP_RECEIVE_ACK)
dp_flags |= DP_SEND_RECEIVE_ACK;
- if (req->rq_state & RQ_EXP_WRITE_ACK)
+ /* During resync, request an explicit write ack,
+ * even in protocol != C */
+ if (req->rq_state & RQ_EXP_WRITE_ACK
+ || (dp_flags & DP_MAY_SET_IN_SYNC))
dp_flags |= DP_SEND_WRITE_ACK;
}
p->dp_flags = cpu_to_be32(dp_flags);
drbd_set_defaults(device);
atomic_set(&device->ap_bio_cnt, 0);
+ atomic_set(&device->ap_actlog_cnt, 0);
atomic_set(&device->ap_pending_cnt, 0);
atomic_set(&device->rs_pending_cnt, 0);
atomic_set(&device->unacked_cnt, 0);
atomic_set(&device->rs_sect_in, 0);
atomic_set(&device->rs_sect_ev, 0);
atomic_set(&device->ap_in_flight, 0);
- atomic_set(&device->md_io_in_use, 0);
+ atomic_set(&device->md_io.in_use, 0);
mutex_init(&device->own_state_mutex);
device->state_mutex = &device->own_state_mutex;
INIT_LIST_HEAD(&device->resync_reads);
INIT_LIST_HEAD(&device->resync_work.list);
INIT_LIST_HEAD(&device->unplug_work.list);
- INIT_LIST_HEAD(&device->go_diskless.list);
- INIT_LIST_HEAD(&device->md_sync_work.list);
- INIT_LIST_HEAD(&device->start_resync_work.list);
INIT_LIST_HEAD(&device->bm_io_work.w.list);
+ INIT_LIST_HEAD(&device->pending_master_completion[0]);
+ INIT_LIST_HEAD(&device->pending_master_completion[1]);
+ INIT_LIST_HEAD(&device->pending_completion[0]);
+ INIT_LIST_HEAD(&device->pending_completion[1]);
device->resync_work.cb = w_resync_timer;
device->unplug_work.cb = w_send_write_hint;
- device->go_diskless.cb = w_go_diskless;
- device->md_sync_work.cb = w_md_sync;
device->bm_io_work.w.cb = w_bitmap_io;
- device->start_resync_work.cb = w_start_resync;
init_timer(&device->resync_timer);
init_timer(&device->md_sync_timer);
drbd_bm_cleanup(device);
}
- drbd_free_bc(device->ldev);
+ drbd_free_ldev(device->ldev);
device->ldev = NULL;
clear_bit(AL_SUSPENDED, &device->flags);
D_ASSERT(device, list_empty(&first_peer_device(device)->connection->sender_work.q));
D_ASSERT(device, list_empty(&device->resync_work.list));
D_ASSERT(device, list_empty(&device->unplug_work.list));
- D_ASSERT(device, list_empty(&device->go_diskless.list));
drbd_set_defaults(device);
}
return -ENOMEM;
}
-static int drbd_notify_sys(struct notifier_block *this, unsigned long code,
- void *unused)
-{
- /* just so we have it. you never know what interesting things we
- * might want to do here some day...
- */
-
- return NOTIFY_DONE;
-}
-
-static struct notifier_block drbd_notifier = {
- .notifier_call = drbd_notify_sys,
-};
-
static void drbd_release_all_peer_reqs(struct drbd_device *device)
{
int rr;
{
struct drbd_device *device = container_of(kref, struct drbd_device, kref);
struct drbd_resource *resource = device->resource;
- struct drbd_connection *connection;
+ struct drbd_peer_device *peer_device, *tmp_peer_device;
del_timer_sync(&device->request_timer);
if (device->this_bdev)
bdput(device->this_bdev);
- drbd_free_bc(device->ldev);
+ drbd_free_ldev(device->ldev);
device->ldev = NULL;
drbd_release_all_peer_reqs(device);
if (device->bitmap) /* should no longer be there. */
drbd_bm_cleanup(device);
- __free_page(device->md_io_page);
+ __free_page(device->md_io.page);
put_disk(device->vdisk);
blk_cleanup_queue(device->rq_queue);
kfree(device->rs_plan_s);
- kfree(first_peer_device(device));
- kfree(device);
- for_each_connection(connection, resource)
- kref_put(&connection->kref, drbd_destroy_connection);
+ /* not for_each_connection(connection, resource):
+ * those may have been cleaned up and disassociated already.
+ */
+ for_each_peer_device_safe(peer_device, tmp_peer_device, device) {
+ kref_put(&peer_device->connection->kref, drbd_destroy_connection);
+ kfree(peer_device);
+ }
+ memset(device, 0xfd, sizeof(*device));
+ kfree(device);
kref_put(&resource->kref, drbd_destroy_resource);
}
list_for_each_entry_safe(req, tmp, &writes, tl_requests) {
struct drbd_device *device = req->device;
struct bio *bio = req->master_bio;
- unsigned long start_time = req->start_time;
+ unsigned long start_jif = req->start_jif;
bool expected;
expected =
/* We are not just doing generic_make_request(),
* as we want to keep the start_time information. */
inc_ap_bio(device);
- __drbd_make_request(device, bio, start_time);
+ __drbd_make_request(device, bio, start_jif);
}
}
+/* called via drbd_req_put_completion_ref(),
+ * holds resource->req_lock */
void drbd_restart_request(struct drbd_request *req)
{
unsigned long flags;
idr_destroy(&resource->devices);
free_cpumask_var(resource->cpu_mask);
kfree(resource->name);
+ memset(resource, 0xf2, sizeof(*resource));
kfree(resource);
}
for_each_connection_safe(connection, tmp, resource) {
list_del(&connection->connections);
+ drbd_debugfs_connection_cleanup(connection);
kref_put(&connection->kref, drbd_destroy_connection);
}
+ drbd_debugfs_resource_cleanup(resource);
kref_put(&resource->kref, drbd_destroy_resource);
}
struct drbd_device *device;
struct drbd_resource *resource, *tmp;
- unregister_reboot_notifier(&drbd_notifier);
-
/* first remove proc,
* drbdsetup uses it's presence to detect
* whether DRBD is loaded.
destroy_workqueue(retry.wq);
drbd_genl_unregister();
+ drbd_debugfs_cleanup();
idr_for_each_entry(&drbd_devices, device, i)
drbd_delete_device(device);
idr_destroy(&drbd_devices);
- printk(KERN_INFO "drbd: module cleanup done.\n");
+ pr_info("module cleanup done.\n");
}
/**
if (nr_cpu_ids > 1 && res_opts->cpu_mask[0] != 0) {
err = bitmap_parse(res_opts->cpu_mask, DRBD_CPU_MASK_SIZE,
cpumask_bits(new_cpu_mask), nr_cpu_ids);
+ if (err == -EOVERFLOW) {
+ /* So what. mask it out. */
+ cpumask_var_t tmp_cpu_mask;
+ if (zalloc_cpumask_var(&tmp_cpu_mask, GFP_KERNEL)) {
+ cpumask_setall(tmp_cpu_mask);
+ cpumask_and(new_cpu_mask, new_cpu_mask, tmp_cpu_mask);
+ drbd_warn(resource, "Overflow in bitmap_parse(%.12s%s), truncating to %u bits\n",
+ res_opts->cpu_mask,
+ strlen(res_opts->cpu_mask) > 12 ? "..." : "",
+ nr_cpu_ids);
+ free_cpumask_var(tmp_cpu_mask);
+ err = 0;
+ }
+ }
if (err) {
drbd_warn(resource, "bitmap_parse() failed with %d\n", err);
/* retcode = ERR_CPU_MASK_PARSE; */
kref_init(&resource->kref);
idr_init(&resource->devices);
INIT_LIST_HEAD(&resource->connections);
+ resource->write_ordering = WO_bdev_flush;
list_add_tail_rcu(&resource->resources, &drbd_resources);
mutex_init(&resource->conf_update);
mutex_init(&resource->adm_mutex);
spin_lock_init(&resource->req_lock);
+ drbd_debugfs_resource_add(resource);
return resource;
fail_free_name:
return NULL;
}
-/* caller must be under genl_lock() */
+/* caller must be under adm_mutex */
struct drbd_connection *conn_create(const char *name, struct res_opts *res_opts)
{
struct drbd_resource *resource;
INIT_LIST_HEAD(&connection->current_epoch->list);
connection->epochs = 1;
spin_lock_init(&connection->epoch_lock);
- connection->write_ordering = WO_bdev_flush;
connection->send.seen_any_write_yet = false;
connection->send.current_epoch_nr = 0;
kref_get(&resource->kref);
list_add_tail_rcu(&connection->connections, &resource->connections);
+ drbd_debugfs_connection_add(connection);
return connection;
fail_resource:
drbd_free_socket(&connection->data);
kfree(connection->int_dig_in);
kfree(connection->int_dig_vv);
+ memset(connection, 0xfc, sizeof(*connection));
kfree(connection);
kref_put(&resource->kref, drbd_destroy_resource);
}
return -ENOMEM;
INIT_WORK(&device->submit.worker, do_submit);
- spin_lock_init(&device->submit.lock);
INIT_LIST_HEAD(&device->submit.writes);
return 0;
}
blk_queue_merge_bvec(q, drbd_merge_bvec);
q->queue_lock = &resource->req_lock;
- device->md_io_page = alloc_page(GFP_KERNEL);
- if (!device->md_io_page)
+ device->md_io.page = alloc_page(GFP_KERNEL);
+ if (!device->md_io.page)
goto out_no_io_page;
if (drbd_bm_init(device))
kref_get(&device->kref);
INIT_LIST_HEAD(&device->peer_devices);
+ INIT_LIST_HEAD(&device->pending_bitmap_io);
for_each_connection(connection, resource) {
peer_device = kzalloc(sizeof(struct drbd_peer_device), GFP_KERNEL);
if (!peer_device)
for_each_peer_device(peer_device, device)
drbd_connected(peer_device);
}
-
+ /* move to create_peer_device() */
+ for_each_peer_device(peer_device, device)
+ drbd_debugfs_peer_device_add(peer_device);
+ drbd_debugfs_device_add(device);
return NO_ERROR;
out_idr_remove_vol:
out_no_minor_idr:
drbd_bm_cleanup(device);
out_no_bitmap:
- __free_page(device->md_io_page);
+ __free_page(device->md_io.page);
out_no_io_page:
put_disk(disk);
out_no_disk:
{
struct drbd_resource *resource = device->resource;
struct drbd_connection *connection;
+ struct drbd_peer_device *peer_device;
int refs = 3;
+ /* move to free_peer_device() */
+ for_each_peer_device(peer_device, device)
+ drbd_debugfs_peer_device_cleanup(peer_device);
+ drbd_debugfs_device_cleanup(device);
for_each_connection(connection, resource) {
idr_remove(&connection->peer_devices, device->vnr);
refs++;
kref_sub(&device->kref, refs, drbd_destroy_device);
}
-int __init drbd_init(void)
+static int __init drbd_init(void)
{
int err;
if (minor_count < DRBD_MINOR_COUNT_MIN || minor_count > DRBD_MINOR_COUNT_MAX) {
- printk(KERN_ERR
- "drbd: invalid minor_count (%d)\n", minor_count);
+ pr_err("invalid minor_count (%d)\n", minor_count);
#ifdef MODULE
return -EINVAL;
#else
err = register_blkdev(DRBD_MAJOR, "drbd");
if (err) {
- printk(KERN_ERR
- "drbd: unable to register block device major %d\n",
+ pr_err("unable to register block device major %d\n",
DRBD_MAJOR);
return err;
}
- register_reboot_notifier(&drbd_notifier);
-
/*
* allocate all necessary structs
*/
err = drbd_genl_register();
if (err) {
- printk(KERN_ERR "drbd: unable to register generic netlink family\n");
+ pr_err("unable to register generic netlink family\n");
goto fail;
}
err = -ENOMEM;
drbd_proc = proc_create_data("drbd", S_IFREG | S_IRUGO , NULL, &drbd_proc_fops, NULL);
if (!drbd_proc) {
- printk(KERN_ERR "drbd: unable to register proc file\n");
+ pr_err("unable to register proc file\n");
goto fail;
}
retry.wq = create_singlethread_workqueue("drbd-reissue");
if (!retry.wq) {
- printk(KERN_ERR "drbd: unable to create retry workqueue\n");
+ pr_err("unable to create retry workqueue\n");
goto fail;
}
INIT_WORK(&retry.worker, do_retry);
spin_lock_init(&retry.lock);
INIT_LIST_HEAD(&retry.writes);
- printk(KERN_INFO "drbd: initialized. "
+ if (drbd_debugfs_init())
+ pr_notice("failed to initialize debugfs -- will not be available\n");
+
+ pr_info("initialized. "
"Version: " REL_VERSION " (api:%d/proto:%d-%d)\n",
API_VERSION, PRO_VERSION_MIN, PRO_VERSION_MAX);
- printk(KERN_INFO "drbd: %s\n", drbd_buildtag());
- printk(KERN_INFO "drbd: registered as block device major %d\n",
- DRBD_MAJOR);
-
+ pr_info("%s\n", drbd_buildtag());
+ pr_info("registered as block device major %d\n", DRBD_MAJOR);
return 0; /* Success! */
fail:
drbd_cleanup();
if (err == -ENOMEM)
- printk(KERN_ERR "drbd: ran out of memory\n");
+ pr_err("ran out of memory\n");
else
- printk(KERN_ERR "drbd: initialization failure\n");
+ pr_err("initialization failure\n");
return err;
}
-void drbd_free_bc(struct drbd_backing_dev *ldev)
+void drbd_free_ldev(struct drbd_backing_dev *ldev)
{
if (ldev == NULL)
return;
kfree(ldev);
}
-void drbd_free_sock(struct drbd_connection *connection)
+static void drbd_free_one_sock(struct drbd_socket *ds)
{
- if (connection->data.socket) {
- mutex_lock(&connection->data.mutex);
- kernel_sock_shutdown(connection->data.socket, SHUT_RDWR);
- sock_release(connection->data.socket);
- connection->data.socket = NULL;
- mutex_unlock(&connection->data.mutex);
- }
- if (connection->meta.socket) {
- mutex_lock(&connection->meta.mutex);
- kernel_sock_shutdown(connection->meta.socket, SHUT_RDWR);
- sock_release(connection->meta.socket);
- connection->meta.socket = NULL;
- mutex_unlock(&connection->meta.mutex);
+ struct socket *s;
+ mutex_lock(&ds->mutex);
+ s = ds->socket;
+ ds->socket = NULL;
+ mutex_unlock(&ds->mutex);
+ if (s) {
+ /* so debugfs does not need to mutex_lock() */
+ synchronize_rcu();
+ kernel_sock_shutdown(s, SHUT_RDWR);
+ sock_release(s);
}
}
+void drbd_free_sock(struct drbd_connection *connection)
+{
+ if (connection->data.socket)
+ drbd_free_one_sock(&connection->data);
+ if (connection->meta.socket)
+ drbd_free_one_sock(&connection->meta);
+}
+
/* meta data management */
void conn_md_sync(struct drbd_connection *connection)
if (!get_ldev_if_state(device, D_FAILED))
return;
- buffer = drbd_md_get_buffer(device);
+ buffer = drbd_md_get_buffer(device, __func__);
if (!buffer)
goto out;
if (device->state.disk != D_DISKLESS)
return ERR_DISK_CONFIGURED;
- buffer = drbd_md_get_buffer(device);
+ buffer = drbd_md_get_buffer(device, __func__);
if (!buffer)
return ERR_NOMEM;
*
* Sets all bits in the bitmap and writes the whole bitmap to stable storage.
*/
-int drbd_bmio_set_n_write(struct drbd_device *device)
+int drbd_bmio_set_n_write(struct drbd_device *device) __must_hold(local)
{
int rv = -EIO;
- if (get_ldev_if_state(device, D_ATTACHING)) {
- drbd_md_set_flag(device, MDF_FULL_SYNC);
- drbd_md_sync(device);
- drbd_bm_set_all(device);
-
- rv = drbd_bm_write(device);
+ drbd_md_set_flag(device, MDF_FULL_SYNC);
+ drbd_md_sync(device);
+ drbd_bm_set_all(device);
- if (!rv) {
- drbd_md_clear_flag(device, MDF_FULL_SYNC);
- drbd_md_sync(device);
- }
+ rv = drbd_bm_write(device);
- put_ldev(device);
+ if (!rv) {
+ drbd_md_clear_flag(device, MDF_FULL_SYNC);
+ drbd_md_sync(device);
}
return rv;
*
* Clears all bits in the bitmap and writes the whole bitmap to stable storage.
*/
-int drbd_bmio_clear_n_write(struct drbd_device *device)
+int drbd_bmio_clear_n_write(struct drbd_device *device) __must_hold(local)
{
- int rv = -EIO;
-
drbd_resume_al(device);
- if (get_ldev_if_state(device, D_ATTACHING)) {
- drbd_bm_clear_all(device);
- rv = drbd_bm_write(device);
- put_ldev(device);
- }
-
- return rv;
+ drbd_bm_clear_all(device);
+ return drbd_bm_write(device);
}
static int w_bitmap_io(struct drbd_work *w, int unused)
return 0;
}
-void drbd_ldev_destroy(struct drbd_device *device)
-{
- lc_destroy(device->resync);
- device->resync = NULL;
- lc_destroy(device->act_log);
- device->act_log = NULL;
- __no_warn(local,
- drbd_free_bc(device->ldev);
- device->ldev = NULL;);
-
- clear_bit(GO_DISKLESS, &device->flags);
-}
-
-static int w_go_diskless(struct drbd_work *w, int unused)
-{
- struct drbd_device *device =
- container_of(w, struct drbd_device, go_diskless);
-
- D_ASSERT(device, device->state.disk == D_FAILED);
- /* we cannot assert local_cnt == 0 here, as get_ldev_if_state will
- * inc/dec it frequently. Once we are D_DISKLESS, no one will touch
- * the protected members anymore, though, so once put_ldev reaches zero
- * again, it will be safe to free them. */
-
- /* Try to write changed bitmap pages, read errors may have just
- * set some bits outside the area covered by the activity log.
- *
- * If we have an IO error during the bitmap writeout,
- * we will want a full sync next time, just in case.
- * (Do we want a specific meta data flag for this?)
- *
- * If that does not make it to stable storage either,
- * we cannot do anything about that anymore.
- *
- * We still need to check if both bitmap and ldev are present, we may
- * end up here after a failed attach, before ldev was even assigned.
- */
- if (device->bitmap && device->ldev) {
- /* An interrupted resync or similar is allowed to recounts bits
- * while we detach.
- * Any modifications would not be expected anymore, though.
- */
- if (drbd_bitmap_io_from_worker(device, drbd_bm_write,
- "detach", BM_LOCKED_TEST_ALLOWED)) {
- if (test_bit(WAS_READ_ERROR, &device->flags)) {
- drbd_md_set_flag(device, MDF_FULL_SYNC);
- drbd_md_sync(device);
- }
- }
- }
-
- drbd_force_state(device, NS(disk, D_DISKLESS));
- return 0;
-}
-
/**
* drbd_queue_bitmap_io() - Queues an IO operation on the whole bitmap
* @device: DRBD device.
* that drbd_set_out_of_sync() can not be called. This function MAY ONLY be
* called from worker context. It MUST NOT be used while a previous such
* work is still pending!
+ *
+ * Its worker function encloses the call of io_fn() by get_ldev() and
+ * put_ldev().
*/
void drbd_queue_bitmap_io(struct drbd_device *device,
int (*io_fn)(struct drbd_device *),
static void md_sync_timer_fn(unsigned long data)
{
struct drbd_device *device = (struct drbd_device *) data;
-
- /* must not double-queue! */
- if (list_empty(&device->md_sync_work.list))
- drbd_queue_work_front(&first_peer_device(device)->connection->sender_work,
- &device->md_sync_work);
-}
-
-static int w_md_sync(struct drbd_work *w, int unused)
-{
- struct drbd_device *device =
- container_of(w, struct drbd_device, md_sync_work);
-
- drbd_warn(device, "md_sync_timer expired! Worker calls drbd_md_sync().\n");
-#ifdef DEBUG
- drbd_warn(device, "last md_mark_dirty: %s:%u\n",
- device->last_md_mark_dirty.func, device->last_md_mark_dirty.line);
-#endif
- drbd_md_sync(device);
- return 0;
+ drbd_device_post_work(device, MD_SYNC);
}
const char *cmdname(enum drbd_packet cmd)
*/
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/module.h>
#include <linux/drbd.h>
#include <linux/in.h>
{
genlmsg_end(skb, genlmsg_data(nlmsg_data(nlmsg_hdr(skb))));
if (genlmsg_reply(skb, info))
- printk(KERN_ERR "drbd: error sending genl reply\n");
+ pr_err("error sending genl reply\n");
}
/* Used on a fresh "drbd_adm_prepare"d reply_skb, this cannot fail: The only
}
enum drbd_state_rv
-drbd_set_role(struct drbd_device *device, enum drbd_role new_role, int force)
+drbd_set_role(struct drbd_device *const device, enum drbd_role new_role, int force)
{
+ struct drbd_peer_device *const peer_device = first_peer_device(device);
+ struct drbd_connection *const connection = peer_device ? peer_device->connection : NULL;
const int max_tries = 4;
enum drbd_state_rv rv = SS_UNKNOWN_ERROR;
struct net_conf *nc;
device->state.disk == D_CONSISTENT && mask.pdsk == 0) {
D_ASSERT(device, device->state.pdsk == D_UNKNOWN);
- if (conn_try_outdate_peer(first_peer_device(device)->connection)) {
+ if (conn_try_outdate_peer(connection)) {
val.disk = D_UP_TO_DATE;
mask.disk = D_MASK;
}
if (rv == SS_NOTHING_TO_DO)
goto out;
if (rv == SS_PRIMARY_NOP && mask.pdsk == 0) {
- if (!conn_try_outdate_peer(first_peer_device(device)->connection) && force) {
+ if (!conn_try_outdate_peer(connection) && force) {
drbd_warn(device, "Forced into split brain situation!\n");
mask.pdsk = D_MASK;
val.pdsk = D_OUTDATED;
retry at most once more in this case. */
int timeo;
rcu_read_lock();
- nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
+ nc = rcu_dereference(connection->net_conf);
timeo = nc ? (nc->ping_timeo + 1) * HZ / 10 : 1;
rcu_read_unlock();
schedule_timeout_interruptible(timeo);
/* FIXME also wait for all pending P_BARRIER_ACK? */
if (new_role == R_SECONDARY) {
- set_disk_ro(device->vdisk, true);
if (get_ldev(device)) {
device->ldev->md.uuid[UI_CURRENT] &= ~(u64)1;
put_ldev(device);
}
} else {
- /* Called from drbd_adm_set_role only.
- * We are still holding the conf_update mutex. */
- nc = first_peer_device(device)->connection->net_conf;
+ mutex_lock(&device->resource->conf_update);
+ nc = connection->net_conf;
if (nc)
nc->discard_my_data = 0; /* without copy; single bit op is atomic */
+ mutex_unlock(&device->resource->conf_update);
- set_disk_ro(device->vdisk, false);
if (get_ldev(device)) {
if (((device->state.conn < C_CONNECTED ||
device->state.pdsk <= D_FAILED)
if (device->state.conn >= C_WF_REPORT_PARAMS) {
/* if this was forced, we should consider sync */
if (forced)
- drbd_send_uuids(first_peer_device(device));
- drbd_send_current_state(first_peer_device(device));
+ drbd_send_uuids(peer_device);
+ drbd_send_current_state(peer_device);
}
drbd_md_sync(device);
-
+ set_disk_ro(device->vdisk, new_role == R_SECONDARY);
kobject_uevent(&disk_to_dev(device->vdisk)->kobj, KOBJ_CHANGE);
out:
mutex_unlock(device->state_mutex);
* still lock the act_log to not trigger ASSERTs there.
*/
drbd_suspend_io(device);
- buffer = drbd_md_get_buffer(device); /* Lock meta-data IO */
+ buffer = drbd_md_get_buffer(device, __func__); /* Lock meta-data IO */
if (!buffer) {
drbd_resume_io(device);
return DS_ERROR;
if (la_size_changed || md_moved || rs) {
u32 prev_flags;
+ /* We do some synchronous IO below, which may take some time.
+ * Clear the timer, to avoid scary "timer expired!" messages,
+ * "Superblock" is written out at least twice below, anyways. */
+ del_timer(&device->md_sync_timer);
drbd_al_shrink(device); /* All extents inactive. */
prev_flags = md->flags;
return 0;
}
-static void drbd_setup_queue_param(struct drbd_device *device, unsigned int max_bio_size)
+static void drbd_setup_queue_param(struct drbd_device *device, struct drbd_backing_dev *bdev,
+ unsigned int max_bio_size)
{
struct request_queue * const q = device->rq_queue;
unsigned int max_hw_sectors = max_bio_size >> 9;
unsigned int max_segments = 0;
struct request_queue *b = NULL;
- if (get_ldev_if_state(device, D_ATTACHING)) {
- b = device->ldev->backing_bdev->bd_disk->queue;
+ if (bdev) {
+ b = bdev->backing_bdev->bd_disk->queue;
max_hw_sectors = min(queue_max_hw_sectors(b), max_bio_size >> 9);
rcu_read_lock();
b->backing_dev_info.ra_pages);
q->backing_dev_info.ra_pages = b->backing_dev_info.ra_pages;
}
- put_ldev(device);
}
}
-void drbd_reconsider_max_bio_size(struct drbd_device *device)
+void drbd_reconsider_max_bio_size(struct drbd_device *device, struct drbd_backing_dev *bdev)
{
unsigned int now, new, local, peer;
local = device->local_max_bio_size; /* Eventually last known value, from volatile memory */
peer = device->peer_max_bio_size; /* Eventually last known value, from meta data */
- if (get_ldev_if_state(device, D_ATTACHING)) {
- local = queue_max_hw_sectors(device->ldev->backing_bdev->bd_disk->queue) << 9;
+ if (bdev) {
+ local = queue_max_hw_sectors(bdev->backing_bdev->bd_disk->queue) << 9;
device->local_max_bio_size = local;
- put_ldev(device);
}
local = min(local, DRBD_MAX_BIO_SIZE);
if (new != now)
drbd_info(device, "max BIO size = %u\n", new);
- drbd_setup_queue_param(device, new);
+ drbd_setup_queue_param(device, bdev, new);
}
/* Starts the worker thread */
return (al_size_4k - 1) * AL_CONTEXT_PER_TRANSACTION;
}
+static bool write_ordering_changed(struct disk_conf *a, struct disk_conf *b)
+{
+ return a->disk_barrier != b->disk_barrier ||
+ a->disk_flushes != b->disk_flushes ||
+ a->disk_drain != b->disk_drain;
+}
+
int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info)
{
struct drbd_config_context adm_ctx;
else
set_bit(MD_NO_FUA, &device->flags);
- drbd_bump_write_ordering(first_peer_device(device)->connection, WO_bdev_flush);
+ if (write_ordering_changed(old_disk_conf, new_disk_conf))
+ drbd_bump_write_ordering(device->resource, NULL, WO_bdev_flush);
drbd_md_sync(device);
{
struct drbd_config_context adm_ctx;
struct drbd_device *device;
+ struct drbd_peer_device *peer_device;
+ struct drbd_connection *connection;
int err;
enum drbd_ret_code retcode;
enum determine_dev_size dd;
device = adm_ctx.device;
mutex_lock(&adm_ctx.resource->adm_mutex);
- conn_reconfig_start(first_peer_device(device)->connection);
+ peer_device = first_peer_device(device);
+ connection = peer_device ? peer_device->connection : NULL;
+ conn_reconfig_start(connection);
/* if you want to reconfigure, please tear down first */
if (device->state.disk > D_DISKLESS) {
* drbd_ldev_destroy is done already, we may end up here very fast,
* e.g. if someone calls attach from the on-io-error handler,
* to realize a "hot spare" feature (not that I'd recommend that) */
- wait_event(device->misc_wait, !atomic_read(&device->local_cnt));
+ wait_event(device->misc_wait, !test_bit(GOING_DISKLESS, &device->flags));
/* make sure there is no leftover from previous force-detach attempts */
clear_bit(FORCE_DETACH, &device->flags);
goto fail;
rcu_read_lock();
- nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
+ nc = rcu_dereference(connection->net_conf);
if (nc) {
if (new_disk_conf->fencing == FP_STONITH && nc->wire_protocol == DRBD_PROT_A) {
rcu_read_unlock();
*/
wait_event(device->misc_wait, !atomic_read(&device->ap_pending_cnt) || drbd_suspended(device));
/* and for any other previously queued work */
- drbd_flush_workqueue(&first_peer_device(device)->connection->sender_work);
+ drbd_flush_workqueue(&connection->sender_work);
rv = _drbd_request_state(device, NS(disk, D_ATTACHING), CS_VERBOSE);
retcode = rv; /* FIXME: Type mismatch. */
new_disk_conf = NULL;
new_plan = NULL;
- drbd_bump_write_ordering(first_peer_device(device)->connection, WO_bdev_flush);
+ drbd_bump_write_ordering(device->resource, device->ldev, WO_bdev_flush);
if (drbd_md_test_flag(device->ldev, MDF_CRASHED_PRIMARY))
set_bit(CRASHED_PRIMARY, &device->flags);
device->read_cnt = 0;
device->writ_cnt = 0;
- drbd_reconsider_max_bio_size(device);
+ drbd_reconsider_max_bio_size(device, device->ldev);
/* If I am currently not R_PRIMARY,
* but meta data primary indicator is set,
kobject_uevent(&disk_to_dev(device->vdisk)->kobj, KOBJ_CHANGE);
put_ldev(device);
- conn_reconfig_done(first_peer_device(device)->connection);
+ conn_reconfig_done(connection);
mutex_unlock(&adm_ctx.resource->adm_mutex);
drbd_adm_finish(&adm_ctx, info, retcode);
return 0;
drbd_force_state(device, NS(disk, D_DISKLESS));
drbd_md_sync(device);
fail:
- conn_reconfig_done(first_peer_device(device)->connection);
+ conn_reconfig_done(connection);
if (nbc) {
if (nbc->backing_bdev)
blkdev_put(nbc->backing_bdev,
}
drbd_suspend_io(device); /* so no-one is stuck in drbd_al_begin_io */
- drbd_md_get_buffer(device); /* make sure there is no in-flight meta-data IO */
+ drbd_md_get_buffer(device, __func__); /* make sure there is no in-flight meta-data IO */
retcode = drbd_request_state(device, NS(disk, D_FAILED));
drbd_md_put_buffer(device);
/* D_FAILED will transition to DISKLESS. */
if (retcode != NO_ERROR)
goto out;
- mutex_lock(&adm_ctx.resource->adm_mutex);
device = adm_ctx.device;
+ if (!get_ldev(device)) {
+ retcode = ERR_NO_DISK;
+ goto out;
+ }
+
+ mutex_lock(&adm_ctx.resource->adm_mutex);
/* If there is still bitmap IO pending, probably because of a previous
* resync just being finished, wait for it before requesting a new resync.
retcode = drbd_request_state(device, NS(conn, C_STARTING_SYNC_T));
drbd_resume_io(device);
mutex_unlock(&adm_ctx.resource->adm_mutex);
+ put_ldev(device);
out:
drbd_adm_finish(&adm_ctx, info, retcode);
return 0;
return 0;
}
-static int drbd_bmio_set_susp_al(struct drbd_device *device)
+static int drbd_bmio_set_susp_al(struct drbd_device *device) __must_hold(local)
{
int rv;
if (retcode != NO_ERROR)
goto out;
- mutex_lock(&adm_ctx.resource->adm_mutex);
device = adm_ctx.device;
+ if (!get_ldev(device)) {
+ retcode = ERR_NO_DISK;
+ goto out;
+ }
+
+ mutex_lock(&adm_ctx.resource->adm_mutex);
/* If there is still bitmap IO pending, probably because of a previous
* resync just being finished, wait for it before requesting a new resync.
retcode = drbd_request_state(device, NS(conn, C_STARTING_SYNC_S));
drbd_resume_io(device);
mutex_unlock(&adm_ctx.resource->adm_mutex);
+ put_ldev(device);
out:
drbd_adm_finish(&adm_ctx, info, retcode);
return 0;
return list_first_entry(&resource->connections, struct drbd_connection, connections);
}
-int nla_put_status_info(struct sk_buff *skb, struct drbd_device *device,
+static int nla_put_status_info(struct sk_buff *skb, struct drbd_device *device,
const struct sib_info *sib)
{
struct drbd_resource *resource = device->resource;
unsigned seq;
int err = -ENOMEM;
- if (sib->sib_reason == SIB_SYNC_PROGRESS) {
- if (time_after(jiffies, device->rs_last_bcast + HZ))
- device->rs_last_bcast = jiffies;
- else
- return;
- }
-
seq = atomic_inc_return(&drbd_genl_seq);
msg = genlmsg_new(NLMSG_GOODSIZE, GFP_NOIO);
if (!msg)
seq_printf(seq, "%ld", v);
}
+static void drbd_get_syncer_progress(struct drbd_device *device,
+ union drbd_dev_state state, unsigned long *rs_total,
+ unsigned long *bits_left, unsigned int *per_mil_done)
+{
+ /* this is to break it at compile time when we change that, in case we
+ * want to support more than (1<<32) bits on a 32bit arch. */
+ typecheck(unsigned long, device->rs_total);
+ *rs_total = device->rs_total;
+
+ /* note: both rs_total and rs_left are in bits, i.e. in
+ * units of BM_BLOCK_SIZE.
+ * for the percentage, we don't care. */
+
+ if (state.conn == C_VERIFY_S || state.conn == C_VERIFY_T)
+ *bits_left = device->ov_left;
+ else
+ *bits_left = drbd_bm_total_weight(device) - device->rs_failed;
+ /* >> 10 to prevent overflow,
+ * +1 to prevent division by zero */
+ if (*bits_left > *rs_total) {
+ /* D'oh. Maybe a logic bug somewhere. More likely just a race
+ * between state change and reset of rs_total.
+ */
+ *bits_left = *rs_total;
+ *per_mil_done = *rs_total ? 0 : 1000;
+ } else {
+ /* Make sure the division happens in long context.
+ * We allow up to one petabyte storage right now,
+ * at a granularity of 4k per bit that is 2**38 bits.
+ * After shift right and multiplication by 1000,
+ * this should still fit easily into a 32bit long,
+ * so we don't need a 64bit division on 32bit arch.
+ * Note: currently we don't support such large bitmaps on 32bit
+ * arch anyways, but no harm done to be prepared for it here.
+ */
+ unsigned int shift = *rs_total > UINT_MAX ? 16 : 10;
+ unsigned long left = *bits_left >> shift;
+ unsigned long total = 1UL + (*rs_total >> shift);
+ unsigned long tmp = 1000UL - left * 1000UL/total;
+ *per_mil_done = tmp;
+ }
+}
+
+
/*lge
* progress bars shamelessly adapted from driver/md/md.c
* output looks like
* [=====>..............] 33.5% (23456/123456)
* finish: 2:20:20 speed: 6,345 (6,456) K/sec
*/
-static void drbd_syncer_progress(struct drbd_device *device, struct seq_file *seq)
+static void drbd_syncer_progress(struct drbd_device *device, struct seq_file *seq,
+ union drbd_dev_state state)
{
- unsigned long db, dt, dbdt, rt, rs_left;
+ unsigned long db, dt, dbdt, rt, rs_total, rs_left;
unsigned int res;
int i, x, y;
int stalled = 0;
- drbd_get_syncer_progress(device, &rs_left, &res);
+ drbd_get_syncer_progress(device, state, &rs_total, &rs_left, &res);
x = res/50;
y = 20-x;
seq_printf(seq, ".");
seq_printf(seq, "] ");
- if (device->state.conn == C_VERIFY_S || device->state.conn == C_VERIFY_T)
+ if (state.conn == C_VERIFY_S || state.conn == C_VERIFY_T)
seq_printf(seq, "verified:");
else
seq_printf(seq, "sync'ed:");
seq_printf(seq, "%3u.%u%% ", res / 10, res % 10);
/* if more than a few GB, display in MB */
- if (device->rs_total > (4UL << (30 - BM_BLOCK_SHIFT)))
+ if (rs_total > (4UL << (30 - BM_BLOCK_SHIFT)))
seq_printf(seq, "(%lu/%lu)M",
(unsigned long) Bit2KB(rs_left >> 10),
- (unsigned long) Bit2KB(device->rs_total >> 10));
+ (unsigned long) Bit2KB(rs_total >> 10));
else
seq_printf(seq, "(%lu/%lu)K\n\t",
(unsigned long) Bit2KB(rs_left),
- (unsigned long) Bit2KB(device->rs_total));
+ (unsigned long) Bit2KB(rs_total));
/* see drivers/md/md.c
* We do not want to overflow, so the order of operands and
dt = (jiffies - device->rs_start - device->rs_paused) / HZ;
if (dt == 0)
dt = 1;
- db = device->rs_total - rs_left;
+ db = rs_total - rs_left;
dbdt = Bit2KB(db/dt);
seq_printf_with_thousands_grouping(seq, dbdt);
seq_printf(seq, ")");
- if (device->state.conn == C_SYNC_TARGET ||
- device->state.conn == C_VERIFY_S) {
+ if (state.conn == C_SYNC_TARGET ||
+ state.conn == C_VERIFY_S) {
seq_printf(seq, " want: ");
seq_printf_with_thousands_grouping(seq, device->c_sync_rate);
}
unsigned long bm_bits = drbd_bm_bits(device);
unsigned long bit_pos;
unsigned long long stop_sector = 0;
- if (device->state.conn == C_VERIFY_S ||
- device->state.conn == C_VERIFY_T) {
+ if (state.conn == C_VERIFY_S ||
+ state.conn == C_VERIFY_T) {
bit_pos = bm_bits - device->ov_left;
if (verify_can_do_stop_sector(device))
stop_sector = device->ov_stop_sector;
}
}
-static void resync_dump_detail(struct seq_file *seq, struct lc_element *e)
-{
- struct bm_extent *bme = lc_entry(e, struct bm_extent, lce);
-
- seq_printf(seq, "%5d %s %s\n", bme->rs_left,
- bme->flags & BME_NO_WRITES ? "NO_WRITES" : "---------",
- bme->flags & BME_LOCKED ? "LOCKED" : "------"
- );
-}
-
static int drbd_seq_show(struct seq_file *seq, void *v)
{
int i, prev_i = -1;
const char *sn;
struct drbd_device *device;
struct net_conf *nc;
+ union drbd_dev_state state;
char wp;
static char write_ordering_chars[] = {
seq_printf(seq, "\n");
prev_i = i;
- sn = drbd_conn_str(device->state.conn);
+ state = device->state;
+ sn = drbd_conn_str(state.conn);
- if (device->state.conn == C_STANDALONE &&
- device->state.disk == D_DISKLESS &&
- device->state.role == R_SECONDARY) {
+ if (state.conn == C_STANDALONE &&
+ state.disk == D_DISKLESS &&
+ state.role == R_SECONDARY) {
seq_printf(seq, "%2d: cs:Unconfigured\n", i);
} else {
/* reset device->congestion_reason */
" ns:%u nr:%u dw:%u dr:%u al:%u bm:%u "
"lo:%d pe:%d ua:%d ap:%d ep:%d wo:%c",
i, sn,
- drbd_role_str(device->state.role),
- drbd_role_str(device->state.peer),
- drbd_disk_str(device->state.disk),
- drbd_disk_str(device->state.pdsk),
+ drbd_role_str(state.role),
+ drbd_role_str(state.peer),
+ drbd_disk_str(state.disk),
+ drbd_disk_str(state.pdsk),
wp,
drbd_suspended(device) ? 's' : 'r',
- device->state.aftr_isp ? 'a' : '-',
- device->state.peer_isp ? 'p' : '-',
- device->state.user_isp ? 'u' : '-',
+ state.aftr_isp ? 'a' : '-',
+ state.peer_isp ? 'p' : '-',
+ state.user_isp ? 'u' : '-',
device->congestion_reason ?: '-',
test_bit(AL_SUSPENDED, &device->flags) ? 's' : '-',
device->send_cnt/2,
atomic_read(&device->unacked_cnt),
atomic_read(&device->ap_bio_cnt),
first_peer_device(device)->connection->epochs,
- write_ordering_chars[first_peer_device(device)->connection->write_ordering]
+ write_ordering_chars[device->resource->write_ordering]
);
seq_printf(seq, " oos:%llu\n",
Bit2KB((unsigned long long)
drbd_bm_total_weight(device)));
}
- if (device->state.conn == C_SYNC_SOURCE ||
- device->state.conn == C_SYNC_TARGET ||
- device->state.conn == C_VERIFY_S ||
- device->state.conn == C_VERIFY_T)
- drbd_syncer_progress(device, seq);
+ if (state.conn == C_SYNC_SOURCE ||
+ state.conn == C_SYNC_TARGET ||
+ state.conn == C_VERIFY_S ||
+ state.conn == C_VERIFY_T)
+ drbd_syncer_progress(device, seq, state);
if (proc_details >= 1 && get_ldev_if_state(device, D_FAILED)) {
lc_seq_printf_stats(seq, device->resync);
put_ldev(device);
}
- if (proc_details >= 2) {
- if (device->resync) {
- lc_seq_dump_details(seq, device->resync, "rs_left",
- resync_dump_detail);
- }
- }
+ if (proc_details >= 2)
+ seq_printf(seq, "\tblocked on activity log: %d\n", atomic_read(&device->ap_actlog_cnt));
}
rcu_read_unlock();
int err;
if (try_module_get(THIS_MODULE)) {
- err = single_open(file, drbd_seq_show, PDE_DATA(inode));
+ err = single_open(file, drbd_seq_show, NULL);
if (err)
module_put(THIS_MODULE);
return err;
goto fail;
}
+ memset(peer_req, 0, sizeof(*peer_req));
+ INIT_LIST_HEAD(&peer_req->w.list);
drbd_clear_interval(&peer_req->i);
peer_req->i.size = data_size;
peer_req->i.sector = sector;
- peer_req->i.local = false;
- peer_req->i.waiting = false;
-
- peer_req->epoch = NULL;
+ peer_req->submit_jif = jiffies;
peer_req->peer_device = peer_device;
peer_req->pages = page;
- atomic_set(&peer_req->pending_bios, 0);
- peer_req->flags = 0;
/*
* The block_id is opaque to the receiver. It is not endianness
* converted, and sent back to the sender unchanged.
void __drbd_free_peer_req(struct drbd_device *device, struct drbd_peer_request *peer_req,
int is_net)
{
+ might_sleep();
if (peer_req->flags & EE_HAS_DIGEST)
kfree(peer_req->digest);
drbd_free_pages(device, peer_req->pages, is_net);
D_ASSERT(device, atomic_read(&peer_req->pending_bios) == 0);
D_ASSERT(device, drbd_interval_empty(&peer_req->i));
+ if (!expect(!(peer_req->flags & EE_CALL_AL_COMPLETE_IO))) {
+ peer_req->flags &= ~EE_CALL_AL_COMPLETE_IO;
+ drbd_al_complete_io(device, &peer_req->i);
+ }
mempool_free(peer_req, drbd_ee_mempool);
}
{
unsigned int header_size = drbd_header_size(connection);
struct packet_info pi;
+ struct net_conf *nc;
int err;
+ rcu_read_lock();
+ nc = rcu_dereference(connection->net_conf);
+ if (!nc) {
+ rcu_read_unlock();
+ return -EIO;
+ }
+ sock->sk->sk_rcvtimeo = nc->ping_timeo * 4 * HZ / 10;
+ rcu_read_unlock();
+
err = drbd_recv_short(sock, connection->data.rbuf, header_size, 0);
if (err != header_size) {
if (err >= 0)
* drbd_socket_okay() - Free the socket if its connection is not okay
* @sock: pointer to the pointer to the socket.
*/
-static int drbd_socket_okay(struct socket **sock)
+static bool drbd_socket_okay(struct socket **sock)
{
int rr;
char tb[4];
return false;
}
}
+
+static bool connection_established(struct drbd_connection *connection,
+ struct socket **sock1,
+ struct socket **sock2)
+{
+ struct net_conf *nc;
+ int timeout;
+ bool ok;
+
+ if (!*sock1 || !*sock2)
+ return false;
+
+ rcu_read_lock();
+ nc = rcu_dereference(connection->net_conf);
+ timeout = (nc->sock_check_timeo ?: nc->ping_timeo) * HZ / 10;
+ rcu_read_unlock();
+ schedule_timeout_interruptible(timeout);
+
+ ok = drbd_socket_okay(sock1);
+ ok = drbd_socket_okay(sock2) && ok;
+
+ return ok;
+}
+
/* Gets called if a connection is established, or if a new minor gets created
in a connection */
int drbd_connected(struct drbd_peer_device *peer_device)
struct drbd_socket sock, msock;
struct drbd_peer_device *peer_device;
struct net_conf *nc;
- int vnr, timeout, h, ok;
- bool discard_my_data;
+ int vnr, timeout, h;
+ bool discard_my_data, ok;
enum drbd_state_rv rv;
struct accept_wait_data ad = {
.connection = connection,
}
}
- if (sock.socket && msock.socket) {
- rcu_read_lock();
- nc = rcu_dereference(connection->net_conf);
- timeout = nc->ping_timeo * HZ / 10;
- rcu_read_unlock();
- schedule_timeout_interruptible(timeout);
- ok = drbd_socket_okay(&sock.socket);
- ok = drbd_socket_okay(&msock.socket) && ok;
- if (ok)
- break;
- }
+ if (connection_established(connection, &sock.socket, &msock.socket))
+ break;
retry:
s = drbd_wait_for_connect(connection, &ad);
goto out_release_sockets;
}
- ok = drbd_socket_okay(&sock.socket);
- ok = drbd_socket_okay(&msock.socket) && ok;
+ ok = connection_established(connection, &sock.socket, &msock.socket);
} while (!ok);
if (ad.s_listen)
struct drbd_peer_device *peer_device;
int vnr;
- if (connection->write_ordering >= WO_bdev_flush) {
+ if (connection->resource->write_ordering >= WO_bdev_flush) {
rcu_read_lock();
idr_for_each_entry(&connection->peer_devices, peer_device, vnr) {
struct drbd_device *device = peer_device->device;
kref_get(&device->kref);
rcu_read_unlock();
+ /* Right now, we have only this one synchronous code path
+ * for flushes between request epochs.
+ * We may want to make those asynchronous,
+ * or at least parallelize the flushes to the volume devices.
+ */
+ device->flush_jif = jiffies;
+ set_bit(FLUSH_PENDING, &device->flags);
rv = blkdev_issue_flush(device->ldev->backing_bdev,
GFP_NOIO, NULL);
+ clear_bit(FLUSH_PENDING, &device->flags);
if (rv) {
drbd_info(device, "local disk flush failed with status %d\n", rv);
/* would rather check on EOPNOTSUPP, but that is not reliable.
* don't try again for ANY return value != 0
* if (rv == -EOPNOTSUPP) */
- drbd_bump_write_ordering(connection, WO_drain_io);
+ drbd_bump_write_ordering(connection->resource, NULL, WO_drain_io);
}
put_ldev(device);
kref_put(&device->kref, drbd_destroy_device);
return rv;
}
+static enum write_ordering_e
+max_allowed_wo(struct drbd_backing_dev *bdev, enum write_ordering_e wo)
+{
+ struct disk_conf *dc;
+
+ dc = rcu_dereference(bdev->disk_conf);
+
+ if (wo == WO_bdev_flush && !dc->disk_flushes)
+ wo = WO_drain_io;
+ if (wo == WO_drain_io && !dc->disk_drain)
+ wo = WO_none;
+
+ return wo;
+}
+
/**
* drbd_bump_write_ordering() - Fall back to an other write ordering method
* @connection: DRBD connection.
* @wo: Write ordering method to try.
*/
-void drbd_bump_write_ordering(struct drbd_connection *connection, enum write_ordering_e wo)
+void drbd_bump_write_ordering(struct drbd_resource *resource, struct drbd_backing_dev *bdev,
+ enum write_ordering_e wo)
{
- struct disk_conf *dc;
- struct drbd_peer_device *peer_device;
+ struct drbd_device *device;
enum write_ordering_e pwo;
int vnr;
static char *write_ordering_str[] = {
[WO_bdev_flush] = "flush",
};
- pwo = connection->write_ordering;
- wo = min(pwo, wo);
+ pwo = resource->write_ordering;
+ if (wo != WO_bdev_flush)
+ wo = min(pwo, wo);
rcu_read_lock();
- idr_for_each_entry(&connection->peer_devices, peer_device, vnr) {
- struct drbd_device *device = peer_device->device;
+ idr_for_each_entry(&resource->devices, device, vnr) {
+ if (get_ldev(device)) {
+ wo = max_allowed_wo(device->ldev, wo);
+ if (device->ldev == bdev)
+ bdev = NULL;
+ put_ldev(device);
+ }
+ }
- if (!get_ldev_if_state(device, D_ATTACHING))
- continue;
- dc = rcu_dereference(device->ldev->disk_conf);
+ if (bdev)
+ wo = max_allowed_wo(bdev, wo);
- if (wo == WO_bdev_flush && !dc->disk_flushes)
- wo = WO_drain_io;
- if (wo == WO_drain_io && !dc->disk_drain)
- wo = WO_none;
- put_ldev(device);
- }
rcu_read_unlock();
- connection->write_ordering = wo;
- if (pwo != connection->write_ordering || wo == WO_bdev_flush)
- drbd_info(connection, "Method to ensure write ordering: %s\n", write_ordering_str[connection->write_ordering]);
+
+ resource->write_ordering = wo;
+ if (pwo != resource->write_ordering || wo == WO_bdev_flush)
+ drbd_info(resource, "Method to ensure write ordering: %s\n", write_ordering_str[resource->write_ordering]);
}
/**
/* wait for all pending IO completions, before we start
* zeroing things out. */
conn_wait_active_ee_empty(first_peer_device(device)->connection);
+ /* add it to the active list now,
+ * so we can find it to present it in debugfs */
+ peer_req->submit_jif = jiffies;
+ peer_req->flags |= EE_SUBMITTED;
+ spin_lock_irq(&device->resource->req_lock);
+ list_add_tail(&peer_req->w.list, &device->active_ee);
+ spin_unlock_irq(&device->resource->req_lock);
if (blkdev_issue_zeroout(device->ldev->backing_bdev,
sector, ds >> 9, GFP_NOIO))
peer_req->flags |= EE_WAS_ERROR;
D_ASSERT(device, page == NULL);
atomic_set(&peer_req->pending_bios, n_bios);
+ /* for debugfs: update timestamp, mark as submitted */
+ peer_req->submit_jif = jiffies;
+ peer_req->flags |= EE_SUBMITTED;
do {
bio = bios;
bios = bios->bi_next;
* R_PRIMARY crashes now.
* Therefore we must send the barrier_ack after the barrier request was
* completed. */
- switch (connection->write_ordering) {
+ switch (connection->resource->write_ordering) {
case WO_none:
if (rv == FE_RECYCLED)
return 0;
return 0;
default:
- drbd_err(connection, "Strangeness in connection->write_ordering %d\n", connection->write_ordering);
+ drbd_err(connection, "Strangeness in connection->write_ordering %d\n",
+ connection->resource->write_ordering);
return -EIO;
}
struct drbd_peer_request *peer_req;
struct page *page;
int dgs, ds, err;
- int data_size = pi->size;
+ unsigned int data_size = pi->size;
void *dig_in = peer_device->connection->int_dig_in;
void *dig_vv = peer_device->connection->int_dig_vv;
unsigned long *data;
if (!peer_req)
return NULL;
+ peer_req->flags |= EE_WRITE;
if (trim)
return peer_req;
* respective _drbd_clear_done_ee */
peer_req->w.cb = e_end_resync_block;
+ peer_req->submit_jif = jiffies;
spin_lock_irq(&device->resource->req_lock);
- list_add(&peer_req->w.list, &device->sync_ee);
+ list_add_tail(&peer_req->w.list, &device->sync_ee);
spin_unlock_irq(&device->resource->req_lock);
atomic_add(pi->size >> 9, &device->rs_sect_ev);
}
dec_unacked(device);
}
+
/* we delete from the conflict detection hash _after_ we sent out the
* P_WRITE_ACK / P_NEG_ACK, to get the sequence number right. */
if (peer_req->flags & EE_IN_INTERVAL_TREE) {
drbd_for_each_overlap(i, &device->write_requests, sector, size) {
if (i == &peer_req->i)
continue;
+ if (i->completed)
+ continue;
if (!i->local) {
/*
(unsigned long long)sector, size,
superseded ? "local" : "remote");
- inc_unacked(device);
peer_req->w.cb = superseded ? e_send_superseded :
e_send_retry_write;
list_add_tail(&peer_req->w.list, &device->done_ee);
{
struct drbd_peer_device *peer_device;
struct drbd_device *device;
+ struct net_conf *nc;
sector_t sector;
struct drbd_peer_request *peer_req;
struct p_data *p = pi->data;
}
peer_req->w.cb = e_end_block;
+ peer_req->submit_jif = jiffies;
+ peer_req->flags |= EE_APPLICATION;
dp_flags = be32_to_cpu(p->dp_flags);
rw |= wire_flags_to_bio(dp_flags);
spin_unlock(&connection->epoch_lock);
rcu_read_lock();
- tp = rcu_dereference(peer_device->connection->net_conf)->two_primaries;
+ nc = rcu_dereference(peer_device->connection->net_conf);
+ tp = nc->two_primaries;
+ if (peer_device->connection->agreed_pro_version < 100) {
+ switch (nc->wire_protocol) {
+ case DRBD_PROT_C:
+ dp_flags |= DP_SEND_WRITE_ACK;
+ break;
+ case DRBD_PROT_B:
+ dp_flags |= DP_SEND_RECEIVE_ACK;
+ break;
+ }
+ }
rcu_read_unlock();
+
+ if (dp_flags & DP_SEND_WRITE_ACK) {
+ peer_req->flags |= EE_SEND_WRITE_ACK;
+ inc_unacked(device);
+ /* corresponding dec_unacked() in e_end_block()
+ * respective _drbd_clear_done_ee */
+ }
+
+ if (dp_flags & DP_SEND_RECEIVE_ACK) {
+ /* I really don't like it that the receiver thread
+ * sends on the msock, but anyways */
+ drbd_send_ack(first_peer_device(device), P_RECV_ACK, peer_req);
+ }
+
if (tp) {
+ /* two primaries implies protocol C */
+ D_ASSERT(device, dp_flags & DP_SEND_WRITE_ACK);
peer_req->flags |= EE_IN_INTERVAL_TREE;
err = wait_for_and_update_peer_seq(peer_device, peer_seq);
if (err)
* active_ee to become empty in drbd_submit_peer_request();
* better not add ourselves here. */
if ((peer_req->flags & EE_IS_TRIM_USE_ZEROOUT) == 0)
- list_add(&peer_req->w.list, &device->active_ee);
+ list_add_tail(&peer_req->w.list, &device->active_ee);
spin_unlock_irq(&device->resource->req_lock);
if (device->state.conn == C_SYNC_TARGET)
wait_event(device->ee_wait, !overlapping_resync_write(device, peer_req));
- if (peer_device->connection->agreed_pro_version < 100) {
- rcu_read_lock();
- switch (rcu_dereference(peer_device->connection->net_conf)->wire_protocol) {
- case DRBD_PROT_C:
- dp_flags |= DP_SEND_WRITE_ACK;
- break;
- case DRBD_PROT_B:
- dp_flags |= DP_SEND_RECEIVE_ACK;
- break;
- }
- rcu_read_unlock();
- }
-
- if (dp_flags & DP_SEND_WRITE_ACK) {
- peer_req->flags |= EE_SEND_WRITE_ACK;
- inc_unacked(device);
- /* corresponding dec_unacked() in e_end_block()
- * respective _drbd_clear_done_ee */
- }
-
- if (dp_flags & DP_SEND_RECEIVE_ACK) {
- /* I really don't like it that the receiver thread
- * sends on the msock, but anyways */
- drbd_send_ack(first_peer_device(device), P_RECV_ACK, peer_req);
- }
-
if (device->state.pdsk < D_INCONSISTENT) {
/* In case we have the only disk of the cluster, */
drbd_set_out_of_sync(device, peer_req->i.sector, peer_req->i.size);
- peer_req->flags |= EE_CALL_AL_COMPLETE_IO;
peer_req->flags &= ~EE_MAY_SET_IN_SYNC;
- drbd_al_begin_io(device, &peer_req->i, true);
+ drbd_al_begin_io(device, &peer_req->i);
+ peer_req->flags |= EE_CALL_AL_COMPLETE_IO;
}
err = drbd_submit_peer_request(device, peer_req, rw, DRBD_FAULT_DT_WR);
list_del(&peer_req->w.list);
drbd_remove_epoch_entry_interval(device, peer_req);
spin_unlock_irq(&device->resource->req_lock);
- if (peer_req->flags & EE_CALL_AL_COMPLETE_IO)
+ if (peer_req->flags & EE_CALL_AL_COMPLETE_IO) {
+ peer_req->flags &= ~EE_CALL_AL_COMPLETE_IO;
drbd_al_complete_io(device, &peer_req->i);
+ }
out_interrupted:
drbd_may_finish_epoch(connection, peer_req->epoch, EV_PUT + EV_CLEANUP);
* The current sync rate used here uses only the most recent two step marks,
* to have a short time average so we can react faster.
*/
-bool drbd_rs_should_slow_down(struct drbd_device *device, sector_t sector)
+bool drbd_rs_should_slow_down(struct drbd_device *device, sector_t sector,
+ bool throttle_if_app_is_waiting)
{
struct lc_element *tmp;
- bool throttle = true;
+ bool throttle = drbd_rs_c_min_rate_throttle(device);
- if (!drbd_rs_c_min_rate_throttle(device))
- return false;
+ if (!throttle || throttle_if_app_is_waiting)
+ return throttle;
spin_lock_irq(&device->al_lock);
tmp = lc_find(device->resync, BM_SECT_TO_EXT(sector));
struct bm_extent *bm_ext = lc_entry(tmp, struct bm_extent, lce);
if (test_bit(BME_PRIORITY, &bm_ext->flags))
throttle = false;
- /* Do not slow down if app IO is already waiting for this extent */
+ /* Do not slow down if app IO is already waiting for this extent,
+ * and our progress is necessary for application IO to complete. */
}
spin_unlock_irq(&device->al_lock);
curr_events = (int)part_stat_read(&disk->part0, sectors[0]) +
(int)part_stat_read(&disk->part0, sectors[1]) -
atomic_read(&device->rs_sect_ev);
- if (!device->rs_last_events || curr_events - device->rs_last_events > 64) {
+
+ if (atomic_read(&device->ap_actlog_cnt)
+ || !device->rs_last_events || curr_events - device->rs_last_events > 64) {
unsigned long rs_left;
int i;
peer_req->w.cb = w_e_end_data_req;
fault_type = DRBD_FAULT_DT_RD;
/* application IO, don't drbd_rs_begin_io */
+ peer_req->flags |= EE_APPLICATION;
goto submit;
case P_RS_DATA_REQUEST:
peer_req->w.cb = w_e_end_csum_rs_req;
/* used in the sector offset progress display */
device->bm_resync_fo = BM_SECT_TO_BIT(sector);
+ /* remember to report stats in drbd_resync_finished */
+ device->use_csums = true;
} else if (pi->cmd == P_OV_REPLY) {
/* track progress, we may need to throttle */
atomic_add(size >> 9, &device->rs_sect_in);
* we would also throttle its application reads.
* In that case, throttling is done on the SyncTarget only.
*/
- if (device->state.peer != R_PRIMARY && drbd_rs_should_slow_down(device, sector))
+
+ /* Even though this may be a resync request, we do add to "read_ee";
+ * "sync_ee" is only used for resync WRITEs.
+ * Add to list early, so debugfs can find this request
+ * even if we have to sleep below. */
+ spin_lock_irq(&device->resource->req_lock);
+ list_add_tail(&peer_req->w.list, &device->read_ee);
+ spin_unlock_irq(&device->resource->req_lock);
+
+ update_receiver_timing_details(connection, drbd_rs_should_slow_down);
+ if (device->state.peer != R_PRIMARY
+ && drbd_rs_should_slow_down(device, sector, false))
schedule_timeout_uninterruptible(HZ/10);
+ update_receiver_timing_details(connection, drbd_rs_begin_io);
if (drbd_rs_begin_io(device, sector))
goto out_free_e;
atomic_add(size >> 9, &device->rs_sect_ev);
submit:
+ update_receiver_timing_details(connection, drbd_submit_peer_request);
inc_unacked(device);
- spin_lock_irq(&device->resource->req_lock);
- list_add_tail(&peer_req->w.list, &device->read_ee);
- spin_unlock_irq(&device->resource->req_lock);
-
if (drbd_submit_peer_request(device, peer_req, READ, fault_type) == 0)
return 0;
/* don't care for the reason here */
drbd_err(device, "submit failed, triggering re-connect\n");
+
+out_free_e:
spin_lock_irq(&device->resource->req_lock);
list_del(&peer_req->w.list);
spin_unlock_irq(&device->resource->req_lock);
/* no drbd_rs_complete_io(), we are dropping the connection anyways */
-out_free_e:
put_ldev(device);
drbd_free_peer_req(device, peer_req);
return -EIO;
-1091 requires proto 91
-1096 requires proto 96
*/
-static int drbd_uuid_compare(struct drbd_device *device, int *rule_nr) __must_hold(local)
+static int drbd_uuid_compare(struct drbd_device *const device, int *rule_nr) __must_hold(local)
{
+ struct drbd_peer_device *const peer_device = first_peer_device(device);
+ struct drbd_connection *const connection = peer_device ? peer_device->connection : NULL;
u64 self, peer;
int i, j;
if (device->p_uuid[UI_BITMAP] == (u64)0 && device->ldev->md.uuid[UI_BITMAP] != (u64)0) {
- if (first_peer_device(device)->connection->agreed_pro_version < 91)
+ if (connection->agreed_pro_version < 91)
return -1091;
if ((device->ldev->md.uuid[UI_BITMAP] & ~((u64)1)) == (device->p_uuid[UI_HISTORY_START] & ~((u64)1)) &&
if (device->ldev->md.uuid[UI_BITMAP] == (u64)0 && device->p_uuid[UI_BITMAP] != (u64)0) {
- if (first_peer_device(device)->connection->agreed_pro_version < 91)
+ if (connection->agreed_pro_version < 91)
return -1091;
if ((device->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) == (device->p_uuid[UI_BITMAP] & ~((u64)1)) &&
case 1: /* self_pri && !peer_pri */ return 1;
case 2: /* !self_pri && peer_pri */ return -1;
case 3: /* self_pri && peer_pri */
- dc = test_bit(RESOLVE_CONFLICTS, &first_peer_device(device)->connection->flags);
+ dc = test_bit(RESOLVE_CONFLICTS, &connection->flags);
return dc ? -1 : 1;
}
}
*rule_nr = 51;
peer = device->p_uuid[UI_HISTORY_START] & ~((u64)1);
if (self == peer) {
- if (first_peer_device(device)->connection->agreed_pro_version < 96 ?
+ if (connection->agreed_pro_version < 96 ?
(device->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1)) ==
(device->p_uuid[UI_HISTORY_START + 1] & ~((u64)1)) :
peer + UUID_NEW_BM_OFFSET == (device->p_uuid[UI_BITMAP] & ~((u64)1))) {
/* The last P_SYNC_UUID did not get though. Undo the last start of
resync as sync source modifications of the peer's UUIDs. */
- if (first_peer_device(device)->connection->agreed_pro_version < 91)
+ if (connection->agreed_pro_version < 91)
return -1091;
device->p_uuid[UI_BITMAP] = device->p_uuid[UI_HISTORY_START];
*rule_nr = 71;
self = device->ldev->md.uuid[UI_HISTORY_START] & ~((u64)1);
if (self == peer) {
- if (first_peer_device(device)->connection->agreed_pro_version < 96 ?
+ if (connection->agreed_pro_version < 96 ?
(device->ldev->md.uuid[UI_HISTORY_START + 1] & ~((u64)1)) ==
(device->p_uuid[UI_HISTORY_START] & ~((u64)1)) :
self + UUID_NEW_BM_OFFSET == (device->ldev->md.uuid[UI_BITMAP] & ~((u64)1))) {
/* The last P_SYNC_UUID did not get though. Undo the last start of
resync as sync source modifications of our UUIDs. */
- if (first_peer_device(device)->connection->agreed_pro_version < 91)
+ if (connection->agreed_pro_version < 91)
return -1091;
__drbd_uuid_set(device, UI_BITMAP, device->ldev->md.uuid[UI_HISTORY_START]);
* return: NULL (alg name was "")
* ERR_PTR(error) if something goes wrong
* or the crypto hash ptr, if it worked out ok. */
-static
-struct crypto_hash *drbd_crypto_alloc_digest_safe(const struct drbd_device *device,
+static struct crypto_hash *drbd_crypto_alloc_digest_safe(const struct drbd_device *device,
const char *alg, const char *name)
{
struct crypto_hash *tfm;
struct drbd_device *device;
struct p_sizes *p = pi->data;
enum determine_dev_size dd = DS_UNCHANGED;
- sector_t p_size, p_usize, my_usize;
+ sector_t p_size, p_usize, p_csize, my_usize;
int ldsc = 0; /* local disk size changed */
enum dds_flags ddsf;
p_size = be64_to_cpu(p->d_size);
p_usize = be64_to_cpu(p->u_size);
+ p_csize = be64_to_cpu(p->c_size);
/* just store the peer's disk size for now.
* we still need to figure out whether we accept that. */
}
device->peer_max_bio_size = be32_to_cpu(p->max_bio_size);
- drbd_reconsider_max_bio_size(device);
/* Leave drbd_reconsider_max_bio_size() before drbd_determine_dev_size().
In case we cleared the QUEUE_FLAG_DISCARD from our queue in
drbd_reconsider_max_bio_size(), we can be sure that after
ddsf = be16_to_cpu(p->dds_flags);
if (get_ldev(device)) {
+ drbd_reconsider_max_bio_size(device, device->ldev);
dd = drbd_determine_dev_size(device, ddsf, NULL);
put_ldev(device);
if (dd == DS_ERROR)
return -EIO;
drbd_md_sync(device);
} else {
- /* I am diskless, need to accept the peer's size. */
- drbd_set_my_capacity(device, p_size);
+ /*
+ * I am diskless, need to accept the peer's *current* size.
+ * I must NOT accept the peers backing disk size,
+ * it may have been larger than mine all along...
+ *
+ * At this point, the peer knows more about my disk, or at
+ * least about what we last agreed upon, than myself.
+ * So if his c_size is less than his d_size, the most likely
+ * reason is that *my* d_size was smaller last time we checked.
+ *
+ * However, if he sends a zero current size,
+ * take his (user-capped or) backing disk size anyways.
+ */
+ drbd_reconsider_max_bio_size(device, NULL);
+ drbd_set_my_capacity(device, p_csize ?: p_usize ?: p_size);
}
if (get_ldev(device)) {
struct data_cmd *cmd;
drbd_thread_current_set_cpu(&connection->receiver);
+ update_receiver_timing_details(connection, drbd_recv_header);
if (drbd_recv_header(connection, &pi))
goto err_out;
}
if (shs) {
+ update_receiver_timing_details(connection, drbd_recv_all_warn);
err = drbd_recv_all_warn(connection, pi.data, shs);
if (err)
goto err_out;
pi.size -= shs;
}
+ update_receiver_timing_details(connection, cmd->fn);
err = cmd->fn(connection, &pi);
if (err) {
drbd_err(connection, "error receiving %s, e: %d l: %d!\n",
static void _drbd_end_io_acct(struct drbd_device *device, struct drbd_request *req)
{
int rw = bio_data_dir(req->master_bio);
- unsigned long duration = jiffies - req->start_time;
+ unsigned long duration = jiffies - req->start_jif;
int cpu;
cpu = part_stat_lock();
part_stat_add(cpu, &device->vdisk->part0, ticks[rw], duration);
{
struct drbd_request *req;
- req = mempool_alloc(drbd_request_mempool, GFP_NOIO);
+ req = mempool_alloc(drbd_request_mempool, GFP_NOIO | __GFP_ZERO);
if (!req)
return NULL;
INIT_LIST_HEAD(&req->tl_requests);
INIT_LIST_HEAD(&req->w.list);
+ INIT_LIST_HEAD(&req->req_pending_master_completion);
+ INIT_LIST_HEAD(&req->req_pending_local);
/* one reference to be put by __drbd_make_request */
atomic_set(&req->completion_ref, 1);
return req;
}
+static void drbd_remove_request_interval(struct rb_root *root,
+ struct drbd_request *req)
+{
+ struct drbd_device *device = req->device;
+ struct drbd_interval *i = &req->i;
+
+ drbd_remove_interval(root, i);
+
+ /* Wake up any processes waiting for this request to complete. */
+ if (i->waiting)
+ wake_up(&device->misc_wait);
+}
+
void drbd_req_destroy(struct kref *kref)
{
struct drbd_request *req = container_of(kref, struct drbd_request, kref);
return;
}
- /* remove it from the transfer log.
- * well, only if it had been there in the first
- * place... if it had not (local only or conflicting
- * and never sent), it should still be "empty" as
- * initialized in drbd_req_new(), so we can list_del() it
- * here unconditionally */
+ /* If called from mod_rq_state (expected normal case) or
+ * drbd_send_and_submit (the less likely normal path), this holds the
+ * req_lock, and req->tl_requests will typicaly be on ->transfer_log,
+ * though it may be still empty (never added to the transfer log).
+ *
+ * If called from do_retry(), we do NOT hold the req_lock, but we are
+ * still allowed to unconditionally list_del(&req->tl_requests),
+ * because it will be on a local on-stack list only. */
list_del_init(&req->tl_requests);
+ /* finally remove the request from the conflict detection
+ * respective block_id verification interval tree. */
+ if (!drbd_interval_empty(&req->i)) {
+ struct rb_root *root;
+
+ if (s & RQ_WRITE)
+ root = &device->write_requests;
+ else
+ root = &device->read_requests;
+ drbd_remove_request_interval(root, req);
+ } else if (s & (RQ_NET_MASK & ~RQ_NET_DONE) && req->i.size != 0)
+ drbd_err(device, "drbd_req_destroy: Logic BUG: interval empty, but: rq_state=0x%x, sect=%llu, size=%u\n",
+ s, (unsigned long long)req->i.sector, req->i.size);
+
/* if it was a write, we may have to set the corresponding
* bit(s) out-of-sync first. If it had a local part, we need to
* release the reference to the activity log. */
}
-static void drbd_remove_request_interval(struct rb_root *root,
- struct drbd_request *req)
-{
- struct drbd_device *device = req->device;
- struct drbd_interval *i = &req->i;
-
- drbd_remove_interval(root, i);
-
- /* Wake up any processes waiting for this request to complete. */
- if (i->waiting)
- wake_up(&device->misc_wait);
-}
-
/* Helper for __req_mod().
* Set m->bio to the master bio, if it is fit to be completed,
* or leave it alone (it is initialized to NULL in __req_mod),
ok = (s & RQ_LOCAL_OK) || (s & RQ_NET_OK);
error = PTR_ERR(req->private_bio);
- /* remove the request from the conflict detection
- * respective block_id verification hash */
- if (!drbd_interval_empty(&req->i)) {
- struct rb_root *root;
-
- if (rw == WRITE)
- root = &device->write_requests;
- else
- root = &device->read_requests;
- drbd_remove_request_interval(root, req);
- }
-
/* Before we can signal completion to the upper layers,
* we may need to close the current transfer log epoch.
* We are within the request lock, so we can simply compare
m->error = ok ? 0 : (error ?: -EIO);
m->bio = req->master_bio;
req->master_bio = NULL;
+ /* We leave it in the tree, to be able to verify later
+ * write-acks in protocol != C during resync.
+ * But we mark it as "complete", so it won't be counted as
+ * conflict in a multi-primary setup. */
+ req->i.completed = true;
}
+
+ if (req->i.waiting)
+ wake_up(&device->misc_wait);
+
+ /* Either we are about to complete to upper layers,
+ * or we will restart this request.
+ * In either case, the request object will be destroyed soon,
+ * so better remove it from all lists. */
+ list_del_init(&req->req_pending_master_completion);
}
+/* still holds resource->req_lock */
static int drbd_req_put_completion_ref(struct drbd_request *req, struct bio_and_error *m, int put)
{
struct drbd_device *device = req->device;
return 1;
}
+static void set_if_null_req_next(struct drbd_peer_device *peer_device, struct drbd_request *req)
+{
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
+ if (!connection)
+ return;
+ if (connection->req_next == NULL)
+ connection->req_next = req;
+}
+
+static void advance_conn_req_next(struct drbd_peer_device *peer_device, struct drbd_request *req)
+{
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
+ if (!connection)
+ return;
+ if (connection->req_next != req)
+ return;
+ list_for_each_entry_continue(req, &connection->transfer_log, tl_requests) {
+ const unsigned s = req->rq_state;
+ if (s & RQ_NET_QUEUED)
+ break;
+ }
+ if (&req->tl_requests == &connection->transfer_log)
+ req = NULL;
+ connection->req_next = req;
+}
+
+static void set_if_null_req_ack_pending(struct drbd_peer_device *peer_device, struct drbd_request *req)
+{
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
+ if (!connection)
+ return;
+ if (connection->req_ack_pending == NULL)
+ connection->req_ack_pending = req;
+}
+
+static void advance_conn_req_ack_pending(struct drbd_peer_device *peer_device, struct drbd_request *req)
+{
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
+ if (!connection)
+ return;
+ if (connection->req_ack_pending != req)
+ return;
+ list_for_each_entry_continue(req, &connection->transfer_log, tl_requests) {
+ const unsigned s = req->rq_state;
+ if ((s & RQ_NET_SENT) && (s & RQ_NET_PENDING))
+ break;
+ }
+ if (&req->tl_requests == &connection->transfer_log)
+ req = NULL;
+ connection->req_ack_pending = req;
+}
+
+static void set_if_null_req_not_net_done(struct drbd_peer_device *peer_device, struct drbd_request *req)
+{
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
+ if (!connection)
+ return;
+ if (connection->req_not_net_done == NULL)
+ connection->req_not_net_done = req;
+}
+
+static void advance_conn_req_not_net_done(struct drbd_peer_device *peer_device, struct drbd_request *req)
+{
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
+ if (!connection)
+ return;
+ if (connection->req_not_net_done != req)
+ return;
+ list_for_each_entry_continue(req, &connection->transfer_log, tl_requests) {
+ const unsigned s = req->rq_state;
+ if ((s & RQ_NET_SENT) && !(s & RQ_NET_DONE))
+ break;
+ }
+ if (&req->tl_requests == &connection->transfer_log)
+ req = NULL;
+ connection->req_not_net_done = req;
+}
+
/* I'd like this to be the only place that manipulates
* req->completion_ref and req->kref. */
static void mod_rq_state(struct drbd_request *req, struct bio_and_error *m,
int clear, int set)
{
struct drbd_device *device = req->device;
+ struct drbd_peer_device *peer_device = first_peer_device(device);
unsigned s = req->rq_state;
int c_put = 0;
int k_put = 0;
atomic_inc(&req->completion_ref);
}
- if (!(s & RQ_NET_QUEUED) && (set & RQ_NET_QUEUED))
+ if (!(s & RQ_NET_QUEUED) && (set & RQ_NET_QUEUED)) {
atomic_inc(&req->completion_ref);
+ set_if_null_req_next(peer_device, req);
+ }
if (!(s & RQ_EXP_BARR_ACK) && (set & RQ_EXP_BARR_ACK))
kref_get(&req->kref); /* wait for the DONE */
- if (!(s & RQ_NET_SENT) && (set & RQ_NET_SENT))
- atomic_add(req->i.size >> 9, &device->ap_in_flight);
+ if (!(s & RQ_NET_SENT) && (set & RQ_NET_SENT)) {
+ /* potentially already completed in the asender thread */
+ if (!(s & RQ_NET_DONE)) {
+ atomic_add(req->i.size >> 9, &device->ap_in_flight);
+ set_if_null_req_not_net_done(peer_device, req);
+ }
+ if (s & RQ_NET_PENDING)
+ set_if_null_req_ack_pending(peer_device, req);
+ }
if (!(s & RQ_COMPLETION_SUSP) && (set & RQ_COMPLETION_SUSP))
atomic_inc(&req->completion_ref);
++k_put;
else
++c_put;
+ list_del_init(&req->req_pending_local);
}
if ((s & RQ_NET_PENDING) && (clear & RQ_NET_PENDING)) {
dec_ap_pending(device);
++c_put;
+ req->acked_jif = jiffies;
+ advance_conn_req_ack_pending(peer_device, req);
}
- if ((s & RQ_NET_QUEUED) && (clear & RQ_NET_QUEUED))
+ if ((s & RQ_NET_QUEUED) && (clear & RQ_NET_QUEUED)) {
++c_put;
+ advance_conn_req_next(peer_device, req);
+ }
- if ((s & RQ_EXP_BARR_ACK) && !(s & RQ_NET_DONE) && (set & RQ_NET_DONE)) {
- if (req->rq_state & RQ_NET_SENT)
+ if (!(s & RQ_NET_DONE) && (set & RQ_NET_DONE)) {
+ if (s & RQ_NET_SENT)
atomic_sub(req->i.size >> 9, &device->ap_in_flight);
- ++k_put;
+ if (s & RQ_EXP_BARR_ACK)
+ ++k_put;
+ req->net_done_jif = jiffies;
+
+ /* in ahead/behind mode, or just in case,
+ * before we finally destroy this request,
+ * the caching pointers must not reference it anymore */
+ advance_conn_req_next(peer_device, req);
+ advance_conn_req_ack_pending(peer_device, req);
+ advance_conn_req_not_net_done(peer_device, req);
}
/* potentially complete and destroy */
bdevname(device->ldev->backing_bdev, b));
}
+/* Helper for HANDED_OVER_TO_NETWORK.
+ * Is this a protocol A write (neither WRITE_ACK nor RECEIVE_ACK expected)?
+ * Is it also still "PENDING"?
+ * --> If so, clear PENDING and set NET_OK below.
+ * If it is a protocol A write, but not RQ_PENDING anymore, neg-ack was faster
+ * (and we must not set RQ_NET_OK) */
+static inline bool is_pending_write_protocol_A(struct drbd_request *req)
+{
+ return (req->rq_state &
+ (RQ_WRITE|RQ_NET_PENDING|RQ_EXP_WRITE_ACK|RQ_EXP_RECEIVE_ACK))
+ == (RQ_WRITE|RQ_NET_PENDING);
+}
+
/* obviously this could be coded as many single functions
* instead of one huge switch,
* or by putting the code directly in the respective locations
int __req_mod(struct drbd_request *req, enum drbd_req_event what,
struct bio_and_error *m)
{
- struct drbd_device *device = req->device;
+ struct drbd_device *const device = req->device;
+ struct drbd_peer_device *const peer_device = first_peer_device(device);
+ struct drbd_connection *const connection = peer_device ? peer_device->connection : NULL;
struct net_conf *nc;
int p, rv = 0;
* and from w_read_retry_remote */
D_ASSERT(device, !(req->rq_state & RQ_NET_MASK));
rcu_read_lock();
- nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
+ nc = rcu_dereference(connection->net_conf);
p = nc->wire_protocol;
rcu_read_unlock();
req->rq_state |=
D_ASSERT(device, (req->rq_state & RQ_LOCAL_MASK) == 0);
mod_rq_state(req, m, 0, RQ_NET_QUEUED);
req->w.cb = w_send_read_req;
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
+ drbd_queue_work(&connection->sender_work,
&req->w);
break;
D_ASSERT(device, req->rq_state & RQ_NET_PENDING);
mod_rq_state(req, m, 0, RQ_NET_QUEUED|RQ_EXP_BARR_ACK);
req->w.cb = w_send_dblock;
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
+ drbd_queue_work(&connection->sender_work,
&req->w);
/* close the epoch, in case it outgrew the limit */
rcu_read_lock();
- nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
+ nc = rcu_dereference(connection->net_conf);
p = nc->max_epoch_size;
rcu_read_unlock();
- if (first_peer_device(device)->connection->current_tle_writes >= p)
- start_new_tl_epoch(first_peer_device(device)->connection);
+ if (connection->current_tle_writes >= p)
+ start_new_tl_epoch(connection);
break;
case QUEUE_FOR_SEND_OOS:
mod_rq_state(req, m, 0, RQ_NET_QUEUED);
req->w.cb = w_send_out_of_sync;
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
+ drbd_queue_work(&connection->sender_work,
&req->w);
break;
case HANDED_OVER_TO_NETWORK:
/* assert something? */
- if (bio_data_dir(req->master_bio) == WRITE &&
- !(req->rq_state & (RQ_EXP_RECEIVE_ACK | RQ_EXP_WRITE_ACK))) {
+ if (is_pending_write_protocol_A(req))
/* this is what is dangerous about protocol A:
* pretend it was successfully written on the peer. */
- if (req->rq_state & RQ_NET_PENDING)
- mod_rq_state(req, m, RQ_NET_PENDING, RQ_NET_OK);
- /* else: neg-ack was faster... */
- /* it is still not yet RQ_NET_DONE until the
- * corresponding epoch barrier got acked as well,
- * so we know what to dirty on connection loss */
- }
- mod_rq_state(req, m, RQ_NET_QUEUED, RQ_NET_SENT);
+ mod_rq_state(req, m, RQ_NET_QUEUED|RQ_NET_PENDING,
+ RQ_NET_SENT|RQ_NET_OK);
+ else
+ mod_rq_state(req, m, RQ_NET_QUEUED, RQ_NET_SENT);
+ /* It is still not yet RQ_NET_DONE until the
+ * corresponding epoch barrier got acked as well,
+ * so we know what to dirty on connection loss. */
break;
case OOS_HANDED_TO_NETWORK:
case WRITE_ACKED_BY_PEER_AND_SIS:
req->rq_state |= RQ_NET_SIS;
case WRITE_ACKED_BY_PEER:
- D_ASSERT(device, req->rq_state & RQ_EXP_WRITE_ACK);
- /* protocol C; successfully written on peer.
+ /* Normal operation protocol C: successfully written on peer.
+ * During resync, even in protocol != C,
+ * we requested an explicit write ack anyways.
+ * Which means we cannot even assert anything here.
* Nothing more to do here.
* We want to keep the tl in place for all protocols, to cater
* for volatile write-back caches on lower level devices. */
-
goto ack_common;
case RECV_ACKED_BY_PEER:
D_ASSERT(device, req->rq_state & RQ_EXP_RECEIVE_ACK);
* see also notes above in HANDED_OVER_TO_NETWORK about
* protocol != C */
ack_common:
- D_ASSERT(device, req->rq_state & RQ_NET_PENDING);
mod_rq_state(req, m, RQ_NET_PENDING, RQ_NET_OK);
break;
get_ldev(device); /* always succeeds in this call path */
req->w.cb = w_restart_disk_io;
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
+ drbd_queue_work(&connection->sender_work,
&req->w);
break;
mod_rq_state(req, m, RQ_COMPLETION_SUSP, RQ_NET_QUEUED|RQ_NET_PENDING);
if (req->w.cb) {
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
+ /* w.cb expected to be w_send_dblock, or w_send_read_req */
+ drbd_queue_work(&connection->sender_work,
&req->w);
rv = req->rq_state & RQ_WRITE ? MR_WRITE : MR_READ;
} /* else: FIXME can this happen? */
break;
case QUEUE_AS_DRBD_BARRIER:
- start_new_tl_epoch(first_peer_device(device)->connection);
+ start_new_tl_epoch(connection);
mod_rq_state(req, m, 0, RQ_NET_OK|RQ_NET_DONE);
break;
};
connection->agreed_pro_version < 96)
return;
+ if (on_congestion == OC_PULL_AHEAD && device->state.conn == C_AHEAD)
+ return; /* nothing to do ... */
+
/* If I don't even have good local storage, we can not reasonably try
* to pull ahead of the peer. We also need the local reference to make
* sure device->act_log is there.
* stable storage, and this is a WRITE, we may not even submit
* this bio. */
if (get_ldev(device)) {
+ req->pre_submit_jif = jiffies;
if (drbd_insert_fault(device,
rw == WRITE ? DRBD_FAULT_DT_WR
: rw == READ ? DRBD_FAULT_DT_RD
static void drbd_queue_write(struct drbd_device *device, struct drbd_request *req)
{
- spin_lock(&device->submit.lock);
+ spin_lock_irq(&device->resource->req_lock);
list_add_tail(&req->tl_requests, &device->submit.writes);
- spin_unlock(&device->submit.lock);
+ list_add_tail(&req->req_pending_master_completion,
+ &device->pending_master_completion[1 /* WRITE */]);
+ spin_unlock_irq(&device->resource->req_lock);
queue_work(device->submit.wq, &device->submit.worker);
+ /* do_submit() may sleep internally on al_wait, too */
+ wake_up(&device->al_wait);
}
/* returns the new drbd_request pointer, if the caller is expected to
* Returns ERR_PTR(-ENOMEM) if we cannot allocate a drbd_request.
*/
static struct drbd_request *
-drbd_request_prepare(struct drbd_device *device, struct bio *bio, unsigned long start_time)
+drbd_request_prepare(struct drbd_device *device, struct bio *bio, unsigned long start_jif)
{
const int rw = bio_data_dir(bio);
struct drbd_request *req;
bio_endio(bio, -ENOMEM);
return ERR_PTR(-ENOMEM);
}
- req->start_time = start_time;
+ req->start_jif = start_jif;
if (!get_ldev(device)) {
bio_put(req->private_bio);
if (rw == WRITE && req->private_bio && req->i.size
&& !test_bit(AL_SUSPENDED, &device->flags)) {
if (!drbd_al_begin_io_fastpath(device, &req->i)) {
+ atomic_inc(&device->ap_actlog_cnt);
drbd_queue_write(device, req);
return NULL;
}
req->rq_state |= RQ_IN_ACT_LOG;
+ req->in_actlog_jif = jiffies;
}
return req;
static void drbd_send_and_submit(struct drbd_device *device, struct drbd_request *req)
{
+ struct drbd_resource *resource = device->resource;
const int rw = bio_rw(req->master_bio);
struct bio_and_error m = { NULL, };
bool no_remote = false;
+ bool submit_private_bio = false;
- spin_lock_irq(&device->resource->req_lock);
+ spin_lock_irq(&resource->req_lock);
if (rw == WRITE) {
/* This may temporarily give up the req_lock,
* but will re-aquire it before it returns here.
no_remote = true;
}
+ /* If it took the fast path in drbd_request_prepare, add it here.
+ * The slow path has added it already. */
+ if (list_empty(&req->req_pending_master_completion))
+ list_add_tail(&req->req_pending_master_completion,
+ &device->pending_master_completion[rw == WRITE]);
if (req->private_bio) {
/* needs to be marked within the same spinlock */
+ list_add_tail(&req->req_pending_local,
+ &device->pending_completion[rw == WRITE]);
_req_mod(req, TO_BE_SUBMITTED);
/* but we need to give up the spinlock to submit */
- spin_unlock_irq(&device->resource->req_lock);
- drbd_submit_req_private_bio(req);
- spin_lock_irq(&device->resource->req_lock);
+ submit_private_bio = true;
} else if (no_remote) {
nodata:
if (__ratelimit(&drbd_ratelimit_state))
out:
if (drbd_req_put_completion_ref(req, &m, 1))
kref_put(&req->kref, drbd_req_destroy);
- spin_unlock_irq(&device->resource->req_lock);
-
+ spin_unlock_irq(&resource->req_lock);
+
+ /* Even though above is a kref_put(), this is safe.
+ * As long as we still need to submit our private bio,
+ * we hold a completion ref, and the request cannot disappear.
+ * If however this request did not even have a private bio to submit
+ * (e.g. remote read), req may already be invalid now.
+ * That's why we cannot check on req->private_bio. */
+ if (submit_private_bio)
+ drbd_submit_req_private_bio(req);
if (m.bio)
complete_master_bio(device, &m);
}
-void __drbd_make_request(struct drbd_device *device, struct bio *bio, unsigned long start_time)
+void __drbd_make_request(struct drbd_device *device, struct bio *bio, unsigned long start_jif)
{
- struct drbd_request *req = drbd_request_prepare(device, bio, start_time);
+ struct drbd_request *req = drbd_request_prepare(device, bio, start_jif);
if (IS_ERR_OR_NULL(req))
return;
drbd_send_and_submit(device, req);
continue;
req->rq_state |= RQ_IN_ACT_LOG;
+ req->in_actlog_jif = jiffies;
+ atomic_dec(&device->ap_actlog_cnt);
}
list_del_init(&req->tl_requests);
static bool prepare_al_transaction_nonblock(struct drbd_device *device,
struct list_head *incoming,
- struct list_head *pending)
+ struct list_head *pending,
+ struct list_head *later)
{
struct drbd_request *req, *tmp;
int wake = 0;
spin_lock_irq(&device->al_lock);
list_for_each_entry_safe(req, tmp, incoming, tl_requests) {
err = drbd_al_begin_io_nonblock(device, &req->i);
+ if (err == -ENOBUFS)
+ break;
if (err == -EBUSY)
wake = 1;
if (err)
- continue;
- req->rq_state |= RQ_IN_ACT_LOG;
- list_move_tail(&req->tl_requests, pending);
+ list_move_tail(&req->tl_requests, later);
+ else
+ list_move_tail(&req->tl_requests, pending);
}
spin_unlock_irq(&device->al_lock);
if (wake)
wake_up(&device->al_wait);
-
return !list_empty(pending);
}
+void send_and_submit_pending(struct drbd_device *device, struct list_head *pending)
+{
+ struct drbd_request *req, *tmp;
+
+ list_for_each_entry_safe(req, tmp, pending, tl_requests) {
+ req->rq_state |= RQ_IN_ACT_LOG;
+ req->in_actlog_jif = jiffies;
+ atomic_dec(&device->ap_actlog_cnt);
+ list_del_init(&req->tl_requests);
+ drbd_send_and_submit(device, req);
+ }
+}
+
void do_submit(struct work_struct *ws)
{
struct drbd_device *device = container_of(ws, struct drbd_device, submit.worker);
- LIST_HEAD(incoming);
- LIST_HEAD(pending);
- struct drbd_request *req, *tmp;
+ LIST_HEAD(incoming); /* from drbd_make_request() */
+ LIST_HEAD(pending); /* to be submitted after next AL-transaction commit */
+ LIST_HEAD(busy); /* blocked by resync requests */
+
+ /* grab new incoming requests */
+ spin_lock_irq(&device->resource->req_lock);
+ list_splice_tail_init(&device->submit.writes, &incoming);
+ spin_unlock_irq(&device->resource->req_lock);
for (;;) {
- spin_lock(&device->submit.lock);
- list_splice_tail_init(&device->submit.writes, &incoming);
- spin_unlock(&device->submit.lock);
+ DEFINE_WAIT(wait);
+ /* move used-to-be-busy back to front of incoming */
+ list_splice_init(&busy, &incoming);
submit_fast_path(device, &incoming);
if (list_empty(&incoming))
break;
-skip_fast_path:
- wait_event(device->al_wait, prepare_al_transaction_nonblock(device, &incoming, &pending));
- /* Maybe more was queued, while we prepared the transaction?
- * Try to stuff them into this transaction as well.
- * Be strictly non-blocking here, no wait_event, we already
- * have something to commit.
- * Stop if we don't make any more progres.
- */
for (;;) {
+ prepare_to_wait(&device->al_wait, &wait, TASK_UNINTERRUPTIBLE);
+
+ list_splice_init(&busy, &incoming);
+ prepare_al_transaction_nonblock(device, &incoming, &pending, &busy);
+ if (!list_empty(&pending))
+ break;
+
+ schedule();
+
+ /* If all currently "hot" activity log extents are kept busy by
+ * incoming requests, we still must not totally starve new
+ * requests to "cold" extents.
+ * Something left on &incoming means there had not been
+ * enough update slots available, and the activity log
+ * has been marked as "starving".
+ *
+ * Try again now, without looking for new requests,
+ * effectively blocking all new requests until we made
+ * at least _some_ progress with what we currently have.
+ */
+ if (!list_empty(&incoming))
+ continue;
+
+ /* Nothing moved to pending, but nothing left
+ * on incoming: all moved to busy!
+ * Grab new and iterate. */
+ spin_lock_irq(&device->resource->req_lock);
+ list_splice_tail_init(&device->submit.writes, &incoming);
+ spin_unlock_irq(&device->resource->req_lock);
+ }
+ finish_wait(&device->al_wait, &wait);
+
+ /* If the transaction was full, before all incoming requests
+ * had been processed, skip ahead to commit, and iterate
+ * without splicing in more incoming requests from upper layers.
+ *
+ * Else, if all incoming have been processed,
+ * they have become either "pending" (to be submitted after
+ * next transaction commit) or "busy" (blocked by resync).
+ *
+ * Maybe more was queued, while we prepared the transaction?
+ * Try to stuff those into this transaction as well.
+ * Be strictly non-blocking here,
+ * we already have something to commit.
+ *
+ * Commit if we don't make any more progres.
+ */
+
+ while (list_empty(&incoming)) {
LIST_HEAD(more_pending);
LIST_HEAD(more_incoming);
bool made_progress;
if (list_empty(&device->submit.writes))
break;
- spin_lock(&device->submit.lock);
+ spin_lock_irq(&device->resource->req_lock);
list_splice_tail_init(&device->submit.writes, &more_incoming);
- spin_unlock(&device->submit.lock);
+ spin_unlock_irq(&device->resource->req_lock);
if (list_empty(&more_incoming))
break;
- made_progress = prepare_al_transaction_nonblock(device, &more_incoming, &more_pending);
+ made_progress = prepare_al_transaction_nonblock(device, &more_incoming, &more_pending, &busy);
list_splice_tail_init(&more_pending, &pending);
list_splice_tail_init(&more_incoming, &incoming);
-
if (!made_progress)
break;
}
- drbd_al_begin_io_commit(device, false);
-
- list_for_each_entry_safe(req, tmp, &pending, tl_requests) {
- list_del_init(&req->tl_requests);
- drbd_send_and_submit(device, req);
- }
- /* If all currently hot activity log extents are kept busy by
- * incoming requests, we still must not totally starve new
- * requests to cold extents. In that case, prepare one request
- * in blocking mode. */
- list_for_each_entry_safe(req, tmp, &incoming, tl_requests) {
- list_del_init(&req->tl_requests);
- req->rq_state |= RQ_IN_ACT_LOG;
- if (!drbd_al_begin_io_prepare(device, &req->i)) {
- /* Corresponding extent was hot after all? */
- drbd_send_and_submit(device, req);
- } else {
- /* Found a request to a cold extent.
- * Put on "pending" list,
- * and try to cumulate with more. */
- list_add(&req->tl_requests, &pending);
- goto skip_fast_path;
- }
- }
+ drbd_al_begin_io_commit(device);
+ send_and_submit_pending(device, &pending);
}
}
void drbd_make_request(struct request_queue *q, struct bio *bio)
{
struct drbd_device *device = (struct drbd_device *) q->queuedata;
- unsigned long start_time;
+ unsigned long start_jif;
- start_time = jiffies;
+ start_jif = jiffies;
/*
* what we "blindly" assume:
D_ASSERT(device, IS_ALIGNED(bio->bi_iter.bi_size, 512));
inc_ap_bio(device);
- __drbd_make_request(device, bio, start_time);
+ __drbd_make_request(device, bio, start_jif);
}
/* This is called by bio_add_page().
return limit;
}
-static void find_oldest_requests(
- struct drbd_connection *connection,
- struct drbd_device *device,
- struct drbd_request **oldest_req_waiting_for_peer,
- struct drbd_request **oldest_req_waiting_for_disk)
-{
- struct drbd_request *r;
- *oldest_req_waiting_for_peer = NULL;
- *oldest_req_waiting_for_disk = NULL;
- list_for_each_entry(r, &connection->transfer_log, tl_requests) {
- const unsigned s = r->rq_state;
- if (!*oldest_req_waiting_for_peer
- && ((s & RQ_NET_MASK) && !(s & RQ_NET_DONE)))
- *oldest_req_waiting_for_peer = r;
-
- if (!*oldest_req_waiting_for_disk
- && (s & RQ_LOCAL_PENDING) && r->device == device)
- *oldest_req_waiting_for_disk = r;
-
- if (*oldest_req_waiting_for_peer && *oldest_req_waiting_for_disk)
- break;
- }
-}
-
void request_timer_fn(unsigned long data)
{
struct drbd_device *device = (struct drbd_device *) data;
struct drbd_connection *connection = first_peer_device(device)->connection;
- struct drbd_request *req_disk, *req_peer; /* oldest request */
+ struct drbd_request *req_read, *req_write, *req_peer; /* oldest request */
struct net_conf *nc;
+ unsigned long oldest_submit_jif;
unsigned long ent = 0, dt = 0, et, nt; /* effective timeout = ko_count * timeout */
unsigned long now;
return; /* Recurring timer stopped */
now = jiffies;
+ nt = now + et;
spin_lock_irq(&device->resource->req_lock);
- find_oldest_requests(connection, device, &req_peer, &req_disk);
- if (req_peer == NULL && req_disk == NULL) {
- spin_unlock_irq(&device->resource->req_lock);
- mod_timer(&device->request_timer, now + et);
- return;
- }
+ req_read = list_first_entry_or_null(&device->pending_completion[0], struct drbd_request, req_pending_local);
+ req_write = list_first_entry_or_null(&device->pending_completion[1], struct drbd_request, req_pending_local);
+ req_peer = connection->req_not_net_done;
+ /* maybe the oldest request waiting for the peer is in fact still
+ * blocking in tcp sendmsg */
+ if (!req_peer && connection->req_next && connection->req_next->pre_send_jif)
+ req_peer = connection->req_next;
+
+ /* evaluate the oldest peer request only in one timer! */
+ if (req_peer && req_peer->device != device)
+ req_peer = NULL;
+
+ /* do we have something to evaluate? */
+ if (req_peer == NULL && req_write == NULL && req_read == NULL)
+ goto out;
+
+ oldest_submit_jif =
+ (req_write && req_read)
+ ? ( time_before(req_write->pre_submit_jif, req_read->pre_submit_jif)
+ ? req_write->pre_submit_jif : req_read->pre_submit_jif )
+ : req_write ? req_write->pre_submit_jif
+ : req_read ? req_read->pre_submit_jif : now;
/* The request is considered timed out, if
* - we have some effective timeout from the configuration,
* to expire twice (worst case) to become effective. Good enough.
*/
if (ent && req_peer &&
- time_after(now, req_peer->start_time + ent) &&
+ time_after(now, req_peer->pre_send_jif + ent) &&
!time_in_range(now, connection->last_reconnect_jif, connection->last_reconnect_jif + ent)) {
drbd_warn(device, "Remote failed to finish a request within ko-count * timeout\n");
_drbd_set_state(_NS(device, conn, C_TIMEOUT), CS_VERBOSE | CS_HARD, NULL);
}
- if (dt && req_disk &&
- time_after(now, req_disk->start_time + dt) &&
+ if (dt && oldest_submit_jif != now &&
+ time_after(now, oldest_submit_jif + dt) &&
!time_in_range(now, device->last_reattach_jif, device->last_reattach_jif + dt)) {
drbd_warn(device, "Local backing device failed to meet the disk-timeout\n");
__drbd_chk_io_error(device, DRBD_FORCE_DETACH);
/* Reschedule timer for the nearest not already expired timeout.
* Fallback to now + min(effective network timeout, disk timeout). */
- ent = (ent && req_peer && time_before(now, req_peer->start_time + ent))
- ? req_peer->start_time + ent : now + et;
- dt = (dt && req_disk && time_before(now, req_disk->start_time + dt))
- ? req_disk->start_time + dt : now + et;
+ ent = (ent && req_peer && time_before(now, req_peer->pre_send_jif + ent))
+ ? req_peer->pre_send_jif + ent : now + et;
+ dt = (dt && oldest_submit_jif != now && time_before(now, oldest_submit_jif + dt))
+ ? oldest_submit_jif + dt : now + et;
nt = time_before(ent, dt) ? ent : dt;
+out:
spin_unlock_irq(&connection->resource->req_lock);
mod_timer(&device->request_timer, nt);
}
extern void request_timer_fn(unsigned long data);
extern void tl_restart(struct drbd_connection *connection, enum drbd_req_event what);
extern void _tl_restart(struct drbd_connection *connection, enum drbd_req_event what);
+extern void tl_abort_disk_io(struct drbd_device *device);
/* this is in drbd_main.c */
extern void drbd_restart_request(struct drbd_request *req);
return rv;
}
-static void print_st(struct drbd_device *device, char *name, union drbd_state ns)
+static void print_st(struct drbd_device *device, const char *name, union drbd_state ns)
{
drbd_err(device, " %s = { cs:%s ro:%s/%s ds:%s/%s %c%c%c%c%c%c }\n",
name,
__drbd_set_state(struct drbd_device *device, union drbd_state ns,
enum chg_state_flags flags, struct completion *done)
{
+ struct drbd_peer_device *peer_device = first_peer_device(device);
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
union drbd_state os;
enum drbd_state_rv rv = SS_SUCCESS;
enum sanitize_state_warnings ssw;
struct after_state_chg_work *ascw;
- bool did_remote, should_do_remote;
os = drbd_read_state(device);
this happen...*/
if (is_valid_state(device, os) == rv)
- rv = is_valid_soft_transition(os, ns, first_peer_device(device)->connection);
+ rv = is_valid_soft_transition(os, ns, connection);
} else
- rv = is_valid_soft_transition(os, ns, first_peer_device(device)->connection);
+ rv = is_valid_soft_transition(os, ns, connection);
}
if (rv < SS_SUCCESS) {
sanitize_state(). Only display it here if we where not called from
_conn_request_state() */
if (!(flags & CS_DC_SUSP))
- conn_pr_state_change(first_peer_device(device)->connection, os, ns,
+ conn_pr_state_change(connection, os, ns,
(flags & ~CS_DC_MASK) | CS_DC_SUSP);
/* if we are going -> D_FAILED or D_DISKLESS, grab one extra reference
(os.disk != D_DISKLESS && ns.disk == D_DISKLESS))
atomic_inc(&device->local_cnt);
- did_remote = drbd_should_do_remote(device->state);
+ if (!is_sync_state(os.conn) && is_sync_state(ns.conn))
+ clear_bit(RS_DONE, &device->flags);
+
+ /* changes to local_cnt and device flags should be visible before
+ * changes to state, which again should be visible before anything else
+ * depending on that change happens. */
+ smp_wmb();
device->state.i = ns.i;
- should_do_remote = drbd_should_do_remote(device->state);
device->resource->susp = ns.susp;
device->resource->susp_nod = ns.susp_nod;
device->resource->susp_fen = ns.susp_fen;
+ smp_wmb();
/* put replicated vs not-replicated requests in seperate epochs */
- if (did_remote != should_do_remote)
- start_new_tl_epoch(first_peer_device(device)->connection);
+ if (drbd_should_do_remote((union drbd_dev_state)os.i) !=
+ drbd_should_do_remote((union drbd_dev_state)ns.i))
+ start_new_tl_epoch(connection);
if (os.disk == D_ATTACHING && ns.disk >= D_NEGOTIATING)
drbd_print_uuids(device, "attached to UUIDs");
/* Wake up role changes, that were delayed because of connection establishing */
if (os.conn == C_WF_REPORT_PARAMS && ns.conn != C_WF_REPORT_PARAMS &&
- no_peer_wf_report_params(first_peer_device(device)->connection))
- clear_bit(STATE_SENT, &first_peer_device(device)->connection->flags);
+ no_peer_wf_report_params(connection))
+ clear_bit(STATE_SENT, &connection->flags);
wake_up(&device->misc_wait);
wake_up(&device->state_wait);
- wake_up(&first_peer_device(device)->connection->ping_wait);
+ wake_up(&connection->ping_wait);
/* Aborted verify run, or we reached the stop sector.
* Log the last position, unless end-of-device. */
/* Receiver should clean up itself */
if (os.conn != C_DISCONNECTING && ns.conn == C_DISCONNECTING)
- drbd_thread_stop_nowait(&first_peer_device(device)->connection->receiver);
+ drbd_thread_stop_nowait(&connection->receiver);
/* Now the receiver finished cleaning up itself, it should die */
if (os.conn != C_STANDALONE && ns.conn == C_STANDALONE)
- drbd_thread_stop_nowait(&first_peer_device(device)->connection->receiver);
+ drbd_thread_stop_nowait(&connection->receiver);
/* Upon network failure, we need to restart the receiver. */
if (os.conn > C_WF_CONNECTION &&
ns.conn <= C_TEAR_DOWN && ns.conn >= C_TIMEOUT)
- drbd_thread_restart_nowait(&first_peer_device(device)->connection->receiver);
+ drbd_thread_restart_nowait(&connection->receiver);
/* Resume AL writing if we get a connection */
if (os.conn < C_CONNECTED && ns.conn >= C_CONNECTED) {
drbd_resume_al(device);
- first_peer_device(device)->connection->connect_cnt++;
+ connection->connect_cnt++;
}
/* remember last attach time so request_timer_fn() won't
ascw->w.cb = w_after_state_ch;
ascw->device = device;
ascw->done = done;
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
+ drbd_queue_work(&connection->sender_work,
&ascw->w);
} else {
drbd_err(device, "Could not kmalloc an ascw\n");
union drbd_state ns, enum chg_state_flags flags)
{
struct drbd_resource *resource = device->resource;
+ struct drbd_peer_device *peer_device = first_peer_device(device);
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
struct sib_info sib;
sib.sib_reason = SIB_STATE_CHANGE;
sib.os = os;
sib.ns = ns;
- if (os.conn != C_CONNECTED && ns.conn == C_CONNECTED) {
+ if ((os.disk != D_UP_TO_DATE || os.pdsk != D_UP_TO_DATE)
+ && (ns.disk == D_UP_TO_DATE && ns.pdsk == D_UP_TO_DATE)) {
clear_bit(CRASHED_PRIMARY, &device->flags);
if (device->p_uuid)
device->p_uuid[UI_FLAGS] &= ~((u64)2);
state change. This function might sleep */
if (ns.susp_nod) {
- struct drbd_connection *connection = first_peer_device(device)->connection;
enum drbd_req_event what = NOTHING;
spin_lock_irq(&device->resource->req_lock);
}
if (ns.susp_fen) {
- struct drbd_connection *connection = first_peer_device(device)->connection;
-
spin_lock_irq(&device->resource->req_lock);
if (resource->susp_fen && conn_lowest_conn(connection) >= C_CONNECTED) {
/* case2: The connection was established again: */
* which is unexpected. */
if ((os.conn != C_SYNC_SOURCE && os.conn != C_PAUSED_SYNC_S) &&
(ns.conn == C_SYNC_SOURCE || ns.conn == C_PAUSED_SYNC_S) &&
- first_peer_device(device)->connection->agreed_pro_version >= 96 && get_ldev(device)) {
- drbd_gen_and_send_sync_uuid(first_peer_device(device));
+ connection->agreed_pro_version >= 96 && get_ldev(device)) {
+ drbd_gen_and_send_sync_uuid(peer_device);
put_ldev(device);
}
atomic_set(&device->rs_pending_cnt, 0);
drbd_rs_cancel_all(device);
- drbd_send_uuids(first_peer_device(device));
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_uuids(peer_device);
+ drbd_send_state(peer_device, ns);
}
/* No point in queuing send_bitmap if we don't have a connection
* anymore, so check also the _current_ state, not only the new state
set_bit(NEW_CUR_UUID, &device->flags);
} else {
drbd_uuid_new_current(device);
- drbd_send_uuids(first_peer_device(device));
+ drbd_send_uuids(peer_device);
}
}
put_ldev(device);
if (os.peer == R_SECONDARY && ns.peer == R_PRIMARY &&
device->ldev->md.uuid[UI_BITMAP] == 0 && ns.disk >= D_UP_TO_DATE) {
drbd_uuid_new_current(device);
- drbd_send_uuids(first_peer_device(device));
+ drbd_send_uuids(peer_device);
}
/* D_DISKLESS Peer becomes secondary */
if (os.peer == R_PRIMARY && ns.peer == R_SECONDARY)
/* Last part of the attaching process ... */
if (ns.conn >= C_CONNECTED &&
os.disk == D_ATTACHING && ns.disk == D_NEGOTIATING) {
- drbd_send_sizes(first_peer_device(device), 0, 0); /* to start sync... */
- drbd_send_uuids(first_peer_device(device));
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_sizes(peer_device, 0, 0); /* to start sync... */
+ drbd_send_uuids(peer_device);
+ drbd_send_state(peer_device, ns);
}
/* We want to pause/continue resync, tell peer. */
if (ns.conn >= C_CONNECTED &&
((os.aftr_isp != ns.aftr_isp) ||
(os.user_isp != ns.user_isp)))
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
/* In case one of the isp bits got set, suspend other devices. */
if ((!os.aftr_isp && !os.peer_isp && !os.user_isp) &&
/* Make sure the peer gets informed about eventual state
changes (ISP bits) while we were in WFReportParams. */
if (os.conn == C_WF_REPORT_PARAMS && ns.conn >= C_CONNECTED)
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
if (os.conn != C_AHEAD && ns.conn == C_AHEAD)
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
/* We are in the progress to start a full sync... */
if ((os.conn != C_STARTING_SYNC_T && ns.conn == C_STARTING_SYNC_T) ||
drbd_disk_str(device->state.disk));
if (ns.conn >= C_CONNECTED)
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
drbd_rs_cancel_all(device);
drbd_disk_str(device->state.disk));
if (ns.conn >= C_CONNECTED)
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
/* corresponding get_ldev in __drbd_set_state
* this may finally trigger drbd_ldev_destroy. */
put_ldev(device);
/* Notify peer that I had a local IO error, and did not detached.. */
if (os.disk == D_UP_TO_DATE && ns.disk == D_INCONSISTENT && ns.conn >= C_CONNECTED)
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
/* Disks got bigger while they were detached */
if (ns.disk > D_NEGOTIATING && ns.pdsk > D_NEGOTIATING &&
/* sync target done with resync. Explicitly notify peer, even though
* it should (at least for non-empty resyncs) already know itself. */
if (os.disk < D_UP_TO_DATE && os.conn >= C_SYNC_SOURCE && ns.conn == C_CONNECTED)
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
/* Verify finished, or reached stop sector. Peer did not know about
* the stop sector, and we may even have changed the stop sector during
* verify to interrupt/stop early. Send the new state. */
if (os.conn == C_VERIFY_S && ns.conn == C_CONNECTED
&& verify_can_do_stop_sector(device))
- drbd_send_state(first_peer_device(device), ns);
+ drbd_send_state(peer_device, ns);
/* This triggers bitmap writeout of potentially still unwritten pages
* if the resync finished cleanly, or aborted because of peer disk
old_conf = connection->net_conf;
connection->my_addr_len = 0;
connection->peer_addr_len = 0;
- rcu_assign_pointer(connection->net_conf, NULL);
+ RCU_INIT_POINTER(connection->net_conf, NULL);
conn_free_crypto(connection);
mutex_unlock(&connection->resource->conf_update);
return 0;
}
-void conn_old_common_state(struct drbd_connection *connection, union drbd_state *pcs, enum chg_state_flags *pf)
+static void conn_old_common_state(struct drbd_connection *connection, union drbd_state *pcs, enum chg_state_flags *pf)
{
enum chg_state_flags flags = ~0;
struct drbd_peer_device *peer_device;
return rv;
}
-void
+static void
conn_set_state(struct drbd_connection *connection, union drbd_state mask, union drbd_state val,
union drbd_state *pns_min, union drbd_state *pns_max, enum chg_state_flags flags)
{
*/
void drbd_md_io_complete(struct bio *bio, int error)
{
- struct drbd_md_io *md_io;
struct drbd_device *device;
- md_io = (struct drbd_md_io *)bio->bi_private;
- device = container_of(md_io, struct drbd_device, md_io);
-
- md_io->error = error;
+ device = bio->bi_private;
+ device->md_io.error = error;
/* We grabbed an extra reference in _drbd_md_sync_page_io() to be able
* to timeout on the lower level device, and eventually detach from it.
* ASSERT(atomic_read(&device->md_io_in_use) == 1) there.
*/
drbd_md_put_buffer(device);
- md_io->done = 1;
+ device->md_io.done = 1;
wake_up(&device->misc_wait);
bio_put(bio);
if (device->ldev) /* special case: drbd_md_read() during drbd_adm_attach() */
i = peer_req->i;
do_al_complete_io = peer_req->flags & EE_CALL_AL_COMPLETE_IO;
block_id = peer_req->block_id;
+ peer_req->flags &= ~EE_CALL_AL_COMPLETE_IO;
spin_lock_irqsave(&device->resource->req_lock, flags);
device->writ_cnt += peer_req->i.size >> 9;
if (!get_ldev(device))
return -EIO;
- if (drbd_rs_should_slow_down(device, sector))
- goto defer;
-
/* GFP_TRY, because if there is no memory available right now, this may
* be rescheduled for later. It is "only" background resync, after all. */
peer_req = drbd_alloc_peer_req(peer_device, ID_SYNCER /* unused */, sector,
peer_req->w.cb = w_e_send_csum;
spin_lock_irq(&device->resource->req_lock);
- list_add(&peer_req->w.list, &device->read_ee);
+ list_add_tail(&peer_req->w.list, &device->read_ee);
spin_unlock_irq(&device->resource->req_lock);
atomic_add(size >> 9, &device->rs_sect_ev);
{
struct drbd_device *device = (struct drbd_device *) data;
- if (list_empty(&device->resync_work.list))
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
- &device->resync_work);
+ drbd_queue_work_if_unqueued(
+ &first_peer_device(device)->connection->sender_work,
+ &device->resync_work);
}
static void fifo_set(struct fifo_buffer *fb, int value)
static int drbd_rs_controller(struct drbd_device *device, unsigned int sect_in)
{
struct disk_conf *dc;
- unsigned int want; /* The number of sectors we want in the proxy */
+ unsigned int want; /* The number of sectors we want in-flight */
int req_sect; /* Number of sectors to request in this turn */
- int correction; /* Number of sectors more we need in the proxy*/
+ int correction; /* Number of sectors more we need in-flight */
int cps; /* correction per invocation of drbd_rs_controller() */
int steps; /* Number of time steps to plan ahead */
int curr_corr;
* potentially causing a distributed deadlock on congestion during
* online-verify or (checksum-based) resync, if max-buffers,
* socket buffer sizes and resync rate settings are mis-configured. */
- if (mxb - device->rs_in_flight < number)
- number = mxb - device->rs_in_flight;
+
+ /* note that "number" is in units of "BM_BLOCK_SIZE" (which is 4k),
+ * mxb (as used here, and in drbd_alloc_pages on the peer) is
+ * "number of pages" (typically also 4k),
+ * but "rs_in_flight" is in "sectors" (512 Byte). */
+ if (mxb - device->rs_in_flight/8 < number)
+ number = mxb - device->rs_in_flight/8;
return number;
}
-static int make_resync_request(struct drbd_device *device, int cancel)
+static int make_resync_request(struct drbd_device *const device, int cancel)
{
+ struct drbd_peer_device *const peer_device = first_peer_device(device);
+ struct drbd_connection *const connection = peer_device ? peer_device->connection : NULL;
unsigned long bit;
sector_t sector;
const sector_t capacity = drbd_get_capacity(device->this_bdev);
int max_bio_size;
int number, rollback_i, size;
- int align, queued, sndbuf;
+ int align, requeue = 0;
int i = 0;
if (unlikely(cancel))
goto requeue;
for (i = 0; i < number; i++) {
- /* Stop generating RS requests, when half of the send buffer is filled */
- mutex_lock(&first_peer_device(device)->connection->data.mutex);
- if (first_peer_device(device)->connection->data.socket) {
- queued = first_peer_device(device)->connection->data.socket->sk->sk_wmem_queued;
- sndbuf = first_peer_device(device)->connection->data.socket->sk->sk_sndbuf;
- } else {
- queued = 1;
- sndbuf = 0;
- }
- mutex_unlock(&first_peer_device(device)->connection->data.mutex);
- if (queued > sndbuf / 2)
+ /* Stop generating RS requests when half of the send buffer is filled,
+ * but notify TCP that we'd like to have more space. */
+ mutex_lock(&connection->data.mutex);
+ if (connection->data.socket) {
+ struct sock *sk = connection->data.socket->sk;
+ int queued = sk->sk_wmem_queued;
+ int sndbuf = sk->sk_sndbuf;
+ if (queued > sndbuf / 2) {
+ requeue = 1;
+ if (sk->sk_socket)
+ set_bit(SOCK_NOSPACE, &sk->sk_socket->flags);
+ }
+ } else
+ requeue = 1;
+ mutex_unlock(&connection->data.mutex);
+ if (requeue)
goto requeue;
next_sector:
sector = BM_BIT_TO_SECT(bit);
- if (drbd_rs_should_slow_down(device, sector) ||
- drbd_try_rs_begin_io(device, sector)) {
+ if (drbd_try_rs_begin_io(device, sector)) {
device->bm_resync_fo = bit;
goto requeue;
}
/* adjust very last sectors, in case we are oddly sized */
if (sector + (size>>9) > capacity)
size = (capacity-sector)<<9;
- if (first_peer_device(device)->connection->agreed_pro_version >= 89 &&
- first_peer_device(device)->connection->csums_tfm) {
- switch (read_for_csum(first_peer_device(device), sector, size)) {
+
+ if (device->use_csums) {
+ switch (read_for_csum(peer_device, sector, size)) {
case -EIO: /* Disk failure */
put_ldev(device);
return -EIO;
int err;
inc_rs_pending(device);
- err = drbd_send_drequest(first_peer_device(device), P_RS_DATA_REQUEST,
+ err = drbd_send_drequest(peer_device, P_RS_DATA_REQUEST,
sector, size, ID_SYNCER);
if (err) {
drbd_err(device, "drbd_send_drequest() failed, aborting...\n");
size = BM_BLOCK_SIZE;
- if (drbd_rs_should_slow_down(device, sector) ||
- drbd_try_rs_begin_io(device, sector)) {
+ if (drbd_try_rs_begin_io(device, sector)) {
device->ov_position = sector;
goto requeue;
}
if (os.conn == C_SYNC_TARGET || os.conn == C_PAUSED_SYNC_T)
khelper_cmd = "after-resync-target";
- if (first_peer_device(device)->connection->csums_tfm && device->rs_total) {
+ if (device->use_csums && device->rs_total) {
const unsigned long s = device->rs_same_csum;
const unsigned long t = device->rs_total;
const int ratio =
{
struct drbd_request *req = container_of(w, struct drbd_request, w);
struct drbd_device *device = req->device;
- struct drbd_connection *connection = first_peer_device(device)->connection;
+ struct drbd_peer_device *const peer_device = first_peer_device(device);
+ struct drbd_connection *const connection = peer_device->connection;
int err;
if (unlikely(cancel)) {
req_mod(req, SEND_CANCELED);
return 0;
}
+ req->pre_send_jif = jiffies;
/* this time, no connection->send.current_epoch_writes++;
* If it was sent, it was the closing barrier for the last
* No more barriers will be sent, until we leave AHEAD mode again. */
maybe_send_barrier(connection, req->epoch);
- err = drbd_send_out_of_sync(first_peer_device(device), req);
+ err = drbd_send_out_of_sync(peer_device, req);
req_mod(req, OOS_HANDED_TO_NETWORK);
return err;
{
struct drbd_request *req = container_of(w, struct drbd_request, w);
struct drbd_device *device = req->device;
- struct drbd_connection *connection = first_peer_device(device)->connection;
+ struct drbd_peer_device *const peer_device = first_peer_device(device);
+ struct drbd_connection *connection = peer_device->connection;
int err;
if (unlikely(cancel)) {
req_mod(req, SEND_CANCELED);
return 0;
}
+ req->pre_send_jif = jiffies;
re_init_if_first_write(connection, req->epoch);
maybe_send_barrier(connection, req->epoch);
connection->send.current_epoch_writes++;
- err = drbd_send_dblock(first_peer_device(device), req);
+ err = drbd_send_dblock(peer_device, req);
req_mod(req, err ? SEND_FAILED : HANDED_OVER_TO_NETWORK);
return err;
{
struct drbd_request *req = container_of(w, struct drbd_request, w);
struct drbd_device *device = req->device;
- struct drbd_connection *connection = first_peer_device(device)->connection;
+ struct drbd_peer_device *const peer_device = first_peer_device(device);
+ struct drbd_connection *connection = peer_device->connection;
int err;
if (unlikely(cancel)) {
req_mod(req, SEND_CANCELED);
return 0;
}
+ req->pre_send_jif = jiffies;
/* Even read requests may close a write epoch,
* if there was any yet. */
maybe_send_barrier(connection, req->epoch);
- err = drbd_send_drequest(first_peer_device(device), P_DATA_REQUEST, req->i.sector, req->i.size,
+ err = drbd_send_drequest(peer_device, P_DATA_REQUEST, req->i.sector, req->i.size,
(unsigned long)req);
req_mod(req, err ? SEND_FAILED : HANDED_OVER_TO_NETWORK);
struct drbd_device *device = req->device;
if (bio_data_dir(req->master_bio) == WRITE && req->rq_state & RQ_IN_ACT_LOG)
- drbd_al_begin_io(device, &req->i, false);
+ drbd_al_begin_io(device, &req->i);
drbd_req_make_private_bio(req, req->master_bio);
req->private_bio->bi_bdev = device->ldev->backing_bdev;
void start_resync_timer_fn(unsigned long data)
{
struct drbd_device *device = (struct drbd_device *) data;
-
- drbd_queue_work(&first_peer_device(device)->connection->sender_work,
- &device->start_resync_work);
+ drbd_device_post_work(device, RS_START);
}
-int w_start_resync(struct drbd_work *w, int cancel)
+static void do_start_resync(struct drbd_device *device)
{
- struct drbd_device *device =
- container_of(w, struct drbd_device, start_resync_work);
-
if (atomic_read(&device->unacked_cnt) || atomic_read(&device->rs_pending_cnt)) {
- drbd_warn(device, "w_start_resync later...\n");
+ drbd_warn(device, "postponing start_resync ...\n");
device->start_resync_timer.expires = jiffies + HZ/10;
add_timer(&device->start_resync_timer);
- return 0;
+ return;
}
drbd_start_resync(device, C_SYNC_SOURCE);
clear_bit(AHEAD_TO_SYNC_SOURCE, &device->flags);
- return 0;
+}
+
+static bool use_checksum_based_resync(struct drbd_connection *connection, struct drbd_device *device)
+{
+ bool csums_after_crash_only;
+ rcu_read_lock();
+ csums_after_crash_only = rcu_dereference(connection->net_conf)->csums_after_crash_only;
+ rcu_read_unlock();
+ return connection->agreed_pro_version >= 89 && /* supported? */
+ connection->csums_tfm && /* configured? */
+ (csums_after_crash_only == 0 /* use for each resync? */
+ || test_bit(CRASHED_PRIMARY, &device->flags)); /* or only after Primary crash? */
}
/**
*/
void drbd_start_resync(struct drbd_device *device, enum drbd_conns side)
{
+ struct drbd_peer_device *peer_device = first_peer_device(device);
+ struct drbd_connection *connection = peer_device ? peer_device->connection : NULL;
union drbd_state ns;
int r;
if (r > 0) {
drbd_info(device, "before-resync-target handler returned %d, "
"dropping connection.\n", r);
- conn_request_state(first_peer_device(device)->connection, NS(conn, C_DISCONNECTING), CS_HARD);
+ conn_request_state(connection, NS(conn, C_DISCONNECTING), CS_HARD);
return;
}
} else /* C_SYNC_SOURCE */ {
} else {
drbd_info(device, "before-resync-source handler returned %d, "
"dropping connection.\n", r);
- conn_request_state(first_peer_device(device)->connection,
+ conn_request_state(connection,
NS(conn, C_DISCONNECTING), CS_HARD);
return;
}
}
}
- if (current == first_peer_device(device)->connection->worker.task) {
+ if (current == connection->worker.task) {
/* The worker should not sleep waiting for state_mutex,
that can take long */
if (!mutex_trylock(device->state_mutex)) {
device->rs_mark_time[i] = now;
}
_drbd_pause_after(device);
+ /* Forget potentially stale cached per resync extent bit-counts.
+ * Open coded drbd_rs_cancel_all(device), we already have IRQs
+ * disabled, and know the disk state is ok. */
+ spin_lock(&device->al_lock);
+ lc_reset(device->resync);
+ device->resync_locked = 0;
+ device->resync_wenr = LC_FREE;
+ spin_unlock(&device->al_lock);
}
write_unlock(&global_state_lock);
spin_unlock_irq(&device->resource->req_lock);
if (r == SS_SUCCESS) {
+ wake_up(&device->al_wait); /* for lc_reset() above */
/* reset rs_last_bcast when a resync or verify is started,
* to deal with potential jiffies wrap. */
device->rs_last_bcast = jiffies - HZ;
drbd_conn_str(ns.conn),
(unsigned long) device->rs_total << (BM_BLOCK_SHIFT-10),
(unsigned long) device->rs_total);
- if (side == C_SYNC_TARGET)
+ if (side == C_SYNC_TARGET) {
device->bm_resync_fo = 0;
+ device->use_csums = use_checksum_based_resync(connection, device);
+ } else {
+ device->use_csums = 0;
+ }
/* Since protocol 96, we must serialize drbd_gen_and_send_sync_uuid
* with w_send_oos, or the sync target will get confused as to
* drbd_resync_finished from here in that case.
* We drbd_gen_and_send_sync_uuid here for protocol < 96,
* and from after_state_ch otherwise. */
- if (side == C_SYNC_SOURCE &&
- first_peer_device(device)->connection->agreed_pro_version < 96)
- drbd_gen_and_send_sync_uuid(first_peer_device(device));
+ if (side == C_SYNC_SOURCE && connection->agreed_pro_version < 96)
+ drbd_gen_and_send_sync_uuid(peer_device);
- if (first_peer_device(device)->connection->agreed_pro_version < 95 &&
- device->rs_total == 0) {
+ if (connection->agreed_pro_version < 95 && device->rs_total == 0) {
/* This still has a race (about when exactly the peers
* detect connection loss) that can lead to a full sync
* on next handshake. In 8.3.9 we fixed this with explicit
int timeo;
rcu_read_lock();
- nc = rcu_dereference(first_peer_device(device)->connection->net_conf);
+ nc = rcu_dereference(connection->net_conf);
timeo = nc->ping_int * HZ + nc->ping_timeo * HZ / 9;
rcu_read_unlock();
schedule_timeout_interruptible(timeo);
mutex_unlock(device->state_mutex);
}
+static void update_on_disk_bitmap(struct drbd_device *device, bool resync_done)
+{
+ struct sib_info sib = { .sib_reason = SIB_SYNC_PROGRESS, };
+ device->rs_last_bcast = jiffies;
+
+ if (!get_ldev(device))
+ return;
+
+ drbd_bm_write_lazy(device, 0);
+ if (resync_done && is_sync_state(device->state.conn))
+ drbd_resync_finished(device);
+
+ drbd_bcast_event(device, &sib);
+ /* update timestamp, in case it took a while to write out stuff */
+ device->rs_last_bcast = jiffies;
+ put_ldev(device);
+}
+
+static void drbd_ldev_destroy(struct drbd_device *device)
+{
+ lc_destroy(device->resync);
+ device->resync = NULL;
+ lc_destroy(device->act_log);
+ device->act_log = NULL;
+ __no_warn(local,
+ drbd_free_ldev(device->ldev);
+ device->ldev = NULL;);
+ clear_bit(GOING_DISKLESS, &device->flags);
+ wake_up(&device->misc_wait);
+}
+
+static void go_diskless(struct drbd_device *device)
+{
+ D_ASSERT(device, device->state.disk == D_FAILED);
+ /* we cannot assert local_cnt == 0 here, as get_ldev_if_state will
+ * inc/dec it frequently. Once we are D_DISKLESS, no one will touch
+ * the protected members anymore, though, so once put_ldev reaches zero
+ * again, it will be safe to free them. */
+
+ /* Try to write changed bitmap pages, read errors may have just
+ * set some bits outside the area covered by the activity log.
+ *
+ * If we have an IO error during the bitmap writeout,
+ * we will want a full sync next time, just in case.
+ * (Do we want a specific meta data flag for this?)
+ *
+ * If that does not make it to stable storage either,
+ * we cannot do anything about that anymore.
+ *
+ * We still need to check if both bitmap and ldev are present, we may
+ * end up here after a failed attach, before ldev was even assigned.
+ */
+ if (device->bitmap && device->ldev) {
+ /* An interrupted resync or similar is allowed to recounts bits
+ * while we detach.
+ * Any modifications would not be expected anymore, though.
+ */
+ if (drbd_bitmap_io_from_worker(device, drbd_bm_write,
+ "detach", BM_LOCKED_TEST_ALLOWED)) {
+ if (test_bit(WAS_READ_ERROR, &device->flags)) {
+ drbd_md_set_flag(device, MDF_FULL_SYNC);
+ drbd_md_sync(device);
+ }
+ }
+ }
+
+ drbd_force_state(device, NS(disk, D_DISKLESS));
+}
+
+static int do_md_sync(struct drbd_device *device)
+{
+ drbd_warn(device, "md_sync_timer expired! Worker calls drbd_md_sync().\n");
+ drbd_md_sync(device);
+ return 0;
+}
+
+/* only called from drbd_worker thread, no locking */
+void __update_timing_details(
+ struct drbd_thread_timing_details *tdp,
+ unsigned int *cb_nr,
+ void *cb,
+ const char *fn, const unsigned int line)
+{
+ unsigned int i = *cb_nr % DRBD_THREAD_DETAILS_HIST;
+ struct drbd_thread_timing_details *td = tdp + i;
+
+ td->start_jif = jiffies;
+ td->cb_addr = cb;
+ td->caller_fn = fn;
+ td->line = line;
+ td->cb_nr = *cb_nr;
+
+ i = (i+1) % DRBD_THREAD_DETAILS_HIST;
+ td = tdp + i;
+ memset(td, 0, sizeof(*td));
+
+ ++(*cb_nr);
+}
+
+#define WORK_PENDING(work_bit, todo) (todo & (1UL << work_bit))
+static void do_device_work(struct drbd_device *device, const unsigned long todo)
+{
+ if (WORK_PENDING(MD_SYNC, todo))
+ do_md_sync(device);
+ if (WORK_PENDING(RS_DONE, todo) ||
+ WORK_PENDING(RS_PROGRESS, todo))
+ update_on_disk_bitmap(device, WORK_PENDING(RS_DONE, todo));
+ if (WORK_PENDING(GO_DISKLESS, todo))
+ go_diskless(device);
+ if (WORK_PENDING(DESTROY_DISK, todo))
+ drbd_ldev_destroy(device);
+ if (WORK_PENDING(RS_START, todo))
+ do_start_resync(device);
+}
+
+#define DRBD_DEVICE_WORK_MASK \
+ ((1UL << GO_DISKLESS) \
+ |(1UL << DESTROY_DISK) \
+ |(1UL << MD_SYNC) \
+ |(1UL << RS_START) \
+ |(1UL << RS_PROGRESS) \
+ |(1UL << RS_DONE) \
+ )
+
+static unsigned long get_work_bits(unsigned long *flags)
+{
+ unsigned long old, new;
+ do {
+ old = *flags;
+ new = old & ~DRBD_DEVICE_WORK_MASK;
+ } while (cmpxchg(flags, old, new) != old);
+ return old & DRBD_DEVICE_WORK_MASK;
+}
+
+static void do_unqueued_work(struct drbd_connection *connection)
+{
+ struct drbd_peer_device *peer_device;
+ int vnr;
+
+ rcu_read_lock();
+ idr_for_each_entry(&connection->peer_devices, peer_device, vnr) {
+ struct drbd_device *device = peer_device->device;
+ unsigned long todo = get_work_bits(&device->flags);
+ if (!todo)
+ continue;
+
+ kref_get(&device->kref);
+ rcu_read_unlock();
+ do_device_work(device, todo);
+ kref_put(&device->kref, drbd_destroy_device);
+ rcu_read_lock();
+ }
+ rcu_read_unlock();
+}
+
static bool dequeue_work_batch(struct drbd_work_queue *queue, struct list_head *work_list)
{
spin_lock_irq(&queue->q_lock);
- list_splice_init(&queue->q, work_list);
+ list_splice_tail_init(&queue->q, work_list);
spin_unlock_irq(&queue->q_lock);
return !list_empty(work_list);
}
/* dequeue single item only,
* we still use drbd_queue_work_front() in some places */
if (!list_empty(&connection->sender_work.q))
- list_move(connection->sender_work.q.next, work_list);
+ list_splice_tail_init(&connection->sender_work.q, work_list);
spin_unlock(&connection->sender_work.q_lock); /* FIXME get rid of this one? */
if (!list_empty(work_list) || signal_pending(current)) {
spin_unlock_irq(&connection->resource->req_lock);
if (send_barrier)
maybe_send_barrier(connection,
connection->send.current_epoch_nr + 1);
+
+ if (test_bit(DEVICE_WORK_PENDING, &connection->flags))
+ break;
+
+ /* drbd_send() may have called flush_signals() */
+ if (get_t_state(&connection->worker) != RUNNING)
+ break;
+
schedule();
/* may be woken up for other things but new work, too,
* e.g. if the current epoch got closed.
while (get_t_state(thi) == RUNNING) {
drbd_thread_current_set_cpu(thi);
- /* as long as we use drbd_queue_work_front(),
- * we may only dequeue single work items here, not batches. */
- if (list_empty(&work_list))
+ if (list_empty(&work_list)) {
+ update_worker_timing_details(connection, wait_for_work);
wait_for_work(connection, &work_list);
+ }
+
+ if (test_and_clear_bit(DEVICE_WORK_PENDING, &connection->flags)) {
+ update_worker_timing_details(connection, do_unqueued_work);
+ do_unqueued_work(connection);
+ }
if (signal_pending(current)) {
flush_signals(current);
while (!list_empty(&work_list)) {
w = list_first_entry(&work_list, struct drbd_work, list);
list_del_init(&w->list);
+ update_worker_timing_details(connection, w->cb);
if (w->cb(w, connection->cstate < C_WF_REPORT_PARAMS) == 0)
continue;
if (connection->cstate >= C_WF_REPORT_PARAMS)
}
do {
+ if (test_and_clear_bit(DEVICE_WORK_PENDING, &connection->flags)) {
+ update_worker_timing_details(connection, do_unqueued_work);
+ do_unqueued_work(connection);
+ }
while (!list_empty(&work_list)) {
w = list_first_entry(&work_list, struct drbd_work, list);
list_del_init(&w->list);
+ update_worker_timing_details(connection, w->cb);
w->cb(w, 1);
}
dequeue_work_batch(&connection->sender_work, &work_list);
- } while (!list_empty(&work_list));
+ } while (!list_empty(&work_list) || test_bit(DEVICE_WORK_PENDING, &connection->flags));
rcu_read_lock();
idr_for_each_entry(&connection->peer_devices, peer_device, vnr) {
}
/* Table of device ids supported by this driver. */
-static DEFINE_PCI_DEVICE_TABLE(mtip_pci_tbl) = {
+static const struct pci_device_id mtip_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_MICRON, P320H_DEVICE_ID) },
{ PCI_DEVICE(PCI_VENDOR_ID_MICRON, P320M_DEVICE_ID) },
{ PCI_DEVICE(PCI_VENDOR_ID_MICRON, P320S_DEVICE_ID) },
.slot_reset = rsxx_slot_reset,
};
-static DEFINE_PCI_DEVICE_TABLE(rsxx_pci_ids) = {
+static const struct pci_device_id rsxx_pci_ids[] = {
{PCI_DEVICE(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_FS70_FLASH)},
{PCI_DEVICE(PCI_VENDOR_ID_IBM, PCI_DEVICE_ID_FS80_FLASH)},
{0,},
*****************************************************************************
*/
-static DEFINE_PCI_DEVICE_TABLE(skd_pci_tbl) = {
+static const struct pci_device_id skd_pci_tbl[] = {
{ PCI_VENDOR_ID_STEC, PCI_DEVICE_ID_S1120,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, },
{ 0 } /* terminate list */
#include <linux/numa.h>
#define PART_BITS 4
+#define VQ_NAME_LEN 16
static int major;
static DEFINE_IDA(vd_index_ida);
static struct workqueue_struct *virtblk_wq;
+struct virtio_blk_vq {
+ struct virtqueue *vq;
+ spinlock_t lock;
+ char name[VQ_NAME_LEN];
+} ____cacheline_aligned_in_smp;
+
struct virtio_blk
{
struct virtio_device *vdev;
- struct virtqueue *vq;
- spinlock_t vq_lock;
/* The disk structure for the kernel. */
struct gendisk *disk;
/* Ida index - used to track minor number allocations. */
int index;
+
+ /* num of vqs */
+ int num_vqs;
+ struct virtio_blk_vq *vqs;
};
struct virtblk_req
{
struct virtio_blk *vblk = vq->vdev->priv;
bool req_done = false;
+ int qid = vq->index;
struct virtblk_req *vbr;
unsigned long flags;
unsigned int len;
- spin_lock_irqsave(&vblk->vq_lock, flags);
+ spin_lock_irqsave(&vblk->vqs[qid].lock, flags);
do {
virtqueue_disable_cb(vq);
- while ((vbr = virtqueue_get_buf(vblk->vq, &len)) != NULL) {
+ while ((vbr = virtqueue_get_buf(vblk->vqs[qid].vq, &len)) != NULL) {
blk_mq_complete_request(vbr->req);
req_done = true;
}
/* In case queue is stopped waiting for more buffers. */
if (req_done)
blk_mq_start_stopped_hw_queues(vblk->disk->queue, true);
- spin_unlock_irqrestore(&vblk->vq_lock, flags);
+ spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
}
static int virtio_queue_rq(struct blk_mq_hw_ctx *hctx, struct request *req)
struct virtblk_req *vbr = blk_mq_rq_to_pdu(req);
unsigned long flags;
unsigned int num;
+ int qid = hctx->queue_num;
const bool last = (req->cmd_flags & REQ_END) != 0;
int err;
bool notify = false;
vbr->out_hdr.type |= VIRTIO_BLK_T_IN;
}
- spin_lock_irqsave(&vblk->vq_lock, flags);
- err = __virtblk_add_req(vblk->vq, vbr, vbr->sg, num);
+ spin_lock_irqsave(&vblk->vqs[qid].lock, flags);
+ err = __virtblk_add_req(vblk->vqs[qid].vq, vbr, vbr->sg, num);
if (err) {
- virtqueue_kick(vblk->vq);
+ virtqueue_kick(vblk->vqs[qid].vq);
blk_mq_stop_hw_queue(hctx);
- spin_unlock_irqrestore(&vblk->vq_lock, flags);
+ spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
/* Out of mem doesn't actually happen, since we fall back
* to direct descriptors */
if (err == -ENOMEM || err == -ENOSPC)
return BLK_MQ_RQ_QUEUE_ERROR;
}
- if (last && virtqueue_kick_prepare(vblk->vq))
+ if (last && virtqueue_kick_prepare(vblk->vqs[qid].vq))
notify = true;
- spin_unlock_irqrestore(&vblk->vq_lock, flags);
+ spin_unlock_irqrestore(&vblk->vqs[qid].lock, flags);
if (notify)
- virtqueue_notify(vblk->vq);
+ virtqueue_notify(vblk->vqs[qid].vq);
return BLK_MQ_RQ_QUEUE_OK;
}
static int init_vq(struct virtio_blk *vblk)
{
int err = 0;
+ int i;
+ vq_callback_t **callbacks;
+ const char **names;
+ struct virtqueue **vqs;
+ unsigned short num_vqs;
+ struct virtio_device *vdev = vblk->vdev;
+
+ err = virtio_cread_feature(vdev, VIRTIO_BLK_F_MQ,
+ struct virtio_blk_config, num_queues,
+ &num_vqs);
+ if (err)
+ num_vqs = 1;
+
+ vblk->vqs = kmalloc(sizeof(*vblk->vqs) * num_vqs, GFP_KERNEL);
+ if (!vblk->vqs) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ names = kmalloc(sizeof(*names) * num_vqs, GFP_KERNEL);
+ if (!names)
+ goto err_names;
+
+ callbacks = kmalloc(sizeof(*callbacks) * num_vqs, GFP_KERNEL);
+ if (!callbacks)
+ goto err_callbacks;
+
+ vqs = kmalloc(sizeof(*vqs) * num_vqs, GFP_KERNEL);
+ if (!vqs)
+ goto err_vqs;
- /* We expect one virtqueue, for output. */
- vblk->vq = virtio_find_single_vq(vblk->vdev, virtblk_done, "requests");
- if (IS_ERR(vblk->vq))
- err = PTR_ERR(vblk->vq);
+ for (i = 0; i < num_vqs; i++) {
+ callbacks[i] = virtblk_done;
+ snprintf(vblk->vqs[i].name, VQ_NAME_LEN, "req.%d", i);
+ names[i] = vblk->vqs[i].name;
+ }
+
+ /* Discover virtqueues and write information to configuration. */
+ err = vdev->config->find_vqs(vdev, num_vqs, vqs, callbacks, names);
+ if (err)
+ goto err_find_vqs;
+ for (i = 0; i < num_vqs; i++) {
+ spin_lock_init(&vblk->vqs[i].lock);
+ vblk->vqs[i].vq = vqs[i];
+ }
+ vblk->num_vqs = num_vqs;
+
+ err_find_vqs:
+ kfree(vqs);
+ err_vqs:
+ kfree(callbacks);
+ err_callbacks:
+ kfree(names);
+ err_names:
+ if (err)
+ kfree(vblk->vqs);
+ out:
return err;
}
err = init_vq(vblk);
if (err)
goto out_free_vblk;
- spin_lock_init(&vblk->vq_lock);
/* FIXME: How many partitions? How long is a piece of string? */
vblk->disk = alloc_disk(1 << PART_BITS);
/* Default queue sizing is to fill the ring. */
if (!virtblk_queue_depth) {
- virtblk_queue_depth = vblk->vq->num_free;
+ virtblk_queue_depth = vblk->vqs[0].vq->num_free;
/* ... but without indirect descs, we use 2 descs per req */
if (!virtio_has_feature(vdev, VIRTIO_RING_F_INDIRECT_DESC))
virtblk_queue_depth /= 2;
memset(&vblk->tag_set, 0, sizeof(vblk->tag_set));
vblk->tag_set.ops = &virtio_mq_ops;
- vblk->tag_set.nr_hw_queues = 1;
vblk->tag_set.queue_depth = virtblk_queue_depth;
vblk->tag_set.numa_node = NUMA_NO_NODE;
vblk->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
sizeof(struct virtblk_req) +
sizeof(struct scatterlist) * sg_elems;
vblk->tag_set.driver_data = vblk;
+ vblk->tag_set.nr_hw_queues = vblk->num_vqs;
err = blk_mq_alloc_tag_set(&vblk->tag_set);
if (err)
refc = atomic_read(&disk_to_dev(vblk->disk)->kobj.kref.refcount);
put_disk(vblk->disk);
vdev->config->del_vqs(vdev);
+ kfree(vblk->vqs);
kfree(vblk);
/* Only free device id if we don't have any users */
static unsigned int features[] = {
VIRTIO_BLK_F_SEG_MAX, VIRTIO_BLK_F_SIZE_MAX, VIRTIO_BLK_F_GEOMETRY,
VIRTIO_BLK_F_RO, VIRTIO_BLK_F_BLK_SIZE, VIRTIO_BLK_F_SCSI,
- VIRTIO_BLK_F_WCE, VIRTIO_BLK_F_TOPOLOGY, VIRTIO_BLK_F_CONFIG_WCE
+ VIRTIO_BLK_F_WCE, VIRTIO_BLK_F_TOPOLOGY, VIRTIO_BLK_F_CONFIG_WCE,
+ VIRTIO_BLK_F_MQ,
};
static struct virtio_driver virtio_blk = {
.cleanup = virtio_cleanup,
.priv = (unsigned long)vi,
.name = vi->name,
+ .quality = 1000,
};
vdev->priv = vi;
#include <linux/cpufreq.h>
#include <linux/cpumask.h>
#include <linux/export.h>
+#include <linux/module.h>
#include <linux/mutex.h>
#include <linux/of_platform.h>
#include <linux/pm_opp.h>
arm_bL_ops = NULL;
}
EXPORT_SYMBOL_GPL(bL_cpufreq_unregister);
+
+MODULE_AUTHOR("Viresh Kumar <viresh.kumar@linaro.org>");
+MODULE_DESCRIPTION("Generic ARM big LITTLE cpufreq driver");
+MODULE_LICENSE("GPL v2");
MODULE_AUTHOR("Viresh Kumar <viresh.kumar@linaro.org>");
MODULE_DESCRIPTION("Generic ARM big LITTLE cpufreq driver via DT");
-MODULE_LICENSE("GPL");
+MODULE_LICENSE("GPL v2");
* not yet registered, we should try defering probe.
*/
if (PTR_ERR(cpu_reg) == -EPROBE_DEFER) {
- dev_err(cpu_dev, "cpu0 regulator not ready, retry\n");
+ dev_dbg(cpu_dev, "cpu0 regulator not ready, retry\n");
ret = -EPROBE_DEFER;
goto out_put_node;
}
goto out;
}
- freq_table = kzalloc(sizeof(*freq_table) * (max_opps + 1), GFP_KERNEL);
+ freq_table = kcalloc(sizeof(*freq_table), (max_opps + 1), GFP_ATOMIC);
if (!freq_table) {
ret = -ENOMEM;
goto out;
* Bind to the specified CPU. When this call returns,
* we should be running on the right CPU.
*/
- set_cpus_allowed(current, cpumask_of_cpu(cpu));
+ set_cpus_allowed_ptr(current, cpumask_of(cpu));
BUG_ON(cpu != smp_processor_id());
/* get current setting */
freqs.new = icst_hz(&cclk_params, vco) / 1000;
if (freqs.old == freqs.new) {
- set_cpus_allowed(current, cpus_allowed);
+ set_cpus_allowed_ptr(current, &cpus_allowed);
return 0;
}
/*
* Restore the CPUs allowed mask.
*/
- set_cpus_allowed(current, cpus_allowed);
+ set_cpus_allowed_ptr(current, &cpus_allowed);
cpufreq_freq_transition_end(policy, &freqs, 0);
cpus_allowed = current->cpus_allowed;
- set_cpus_allowed(current, cpumask_of_cpu(cpu));
+ set_cpus_allowed_ptr(current, cpumask_of(cpu));
BUG_ON(cpu != smp_processor_id());
/* detect memory etc. */
current_freq = icst_hz(&cclk_params, vco) / 1000; /* current freq */
- set_cpus_allowed(current, cpus_allowed);
+ set_cpus_allowed_ptr(current, &cpus_allowed);
return current_freq;
}
}
/* Lookup the i2c hwclock */
- for (hwclock = NULL;
- (hwclock = of_find_node_by_name(hwclock, "i2c-hwclock")) != NULL;){
+ for_each_node_by_name(hwclock, "i2c-hwclock") {
const char *loc = of_get_property(hwclock,
"hwctrl-location", NULL);
if (loc == NULL)
return -ENODEV;
}
- pr_debug("signature:0x%.8ulx, command:0x%.8ulx, "
- "event:0x%.8ulx, perf_level:0x%.8ulx.\n",
+ pr_debug("signature:0x%.8x, command:0x%.8x, "
+ "event:0x%.8x, perf_level:0x%.8x.\n",
ist_info.signature, ist_info.command,
ist_info.event, ist_info.perf_level);
* The default values do not overflow.
*/
#define BUCKETS 12
-#define INTERVALS 8
+#define INTERVAL_SHIFT 3
+#define INTERVALS (1UL << INTERVAL_SHIFT)
#define RESOLUTION 1024
#define DECAY 8
#define MAX_INTERESTING 50000
#define LOAD_INT(x) ((x) >> FSHIFT)
#define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1-1)) * 100)
-static int get_loadavg(void)
+static inline int get_loadavg(unsigned long load)
{
- unsigned long this = this_cpu_load();
-
-
- return LOAD_INT(this) * 10 + LOAD_FRAC(this) / 10;
+ return LOAD_INT(load) * 10 + LOAD_FRAC(load) / 10;
}
-static inline int which_bucket(unsigned int duration)
+static inline int which_bucket(unsigned int duration, unsigned long nr_iowaiters)
{
int bucket = 0;
* This allows us to calculate
* E(duration)|iowait
*/
- if (nr_iowait_cpu(smp_processor_id()))
+ if (nr_iowaiters)
bucket = BUCKETS/2;
if (duration < 10)
* to be, the higher this multiplier, and thus the higher
* the barrier to go to an expensive C state.
*/
-static inline int performance_multiplier(void)
+static inline int performance_multiplier(unsigned long nr_iowaiters, unsigned long load)
{
int mult = 1;
/* for higher loadavg, we are more reluctant */
- mult += 2 * get_loadavg();
+ mult += 2 * get_loadavg(load);
/* for IO wait tasks (per cpu!) we add 5x each */
- mult += 10 * nr_iowait_cpu(smp_processor_id());
+ mult += 10 * nr_iowaiters;
return mult;
}
max = value;
}
}
- do_div(avg, divisor);
+ if (divisor == INTERVALS)
+ avg >>= INTERVAL_SHIFT;
+ else
+ do_div(avg, divisor);
/* Then try to determine standard deviation */
stddev = 0;
stddev += diff * diff;
}
}
- do_div(stddev, divisor);
+ if (divisor == INTERVALS)
+ stddev >>= INTERVAL_SHIFT;
+ else
+ do_div(stddev, divisor);
+
/*
* The typical interval is obtained when standard deviation is small
* or standard deviation is small compared to the average interval.
int latency_req = pm_qos_request(PM_QOS_CPU_DMA_LATENCY);
int i;
unsigned int interactivity_req;
- struct timespec t;
+ unsigned long nr_iowaiters, cpu_load;
if (data->needs_update) {
menu_update(drv, dev);
return 0;
/* determine the expected residency time, round up */
- t = ktime_to_timespec(tick_nohz_get_sleep_length());
- data->next_timer_us =
- t.tv_sec * USEC_PER_SEC + t.tv_nsec / NSEC_PER_USEC;
-
+ data->next_timer_us = ktime_to_us(tick_nohz_get_sleep_length());
- data->bucket = which_bucket(data->next_timer_us);
+ get_iowait_load(&nr_iowaiters, &cpu_load);
+ data->bucket = which_bucket(data->next_timer_us, nr_iowaiters);
/*
* Force the result of multiplication to be 64 bits even if both
* duration / latency ratio. Adjust the latency limit if
* necessary.
*/
- interactivity_req = data->predicted_us / performance_multiplier();
+ interactivity_req = data->predicted_us / performance_multiplier(nr_iowaiters, cpu_load);
if (latency_req > interactivity_req)
latency_req = interactivity_req;
}
#endif
-static DEFINE_PCI_DEVICE_TABLE(ccp_pci_table) = {
+static const struct pci_device_id ccp_pci_table[] = {
{ PCI_VDEVICE(AMD, 0x1537), },
/* Last entry must be zero */
{ 0, }
goto error_out;
}
- /* Set ptr to new property if provided */
- if (new_prop) {
- /* Single property */
- if (!strncmp(new_prop->name, "status", new_prop->length)) {
- status = new_prop;
-
- } else if (!strncmp(new_prop->name, "ibm,max-sg-len",
- new_prop->length)) {
- maxsglen = new_prop;
-
- } else if (!strncmp(new_prop->name, "ibm,max-sync-cop",
- new_prop->length)) {
- maxsyncop = new_prop;
-
- } else {
- /*
- * Skip the update, the property being updated
- * has no impact.
- */
- goto out;
- }
- }
+ /*
+ * If this is a property update, there are only certain properties that
+ * we care about. Bail if it isn't in the below list
+ */
+ if (new_prop && (strncmp(new_prop->name, "status", new_prop->length) ||
+ strncmp(new_prop->name, "ibm,max-sg-len", new_prop->length) ||
+ strncmp(new_prop->name, "ibm,max-sync-cop", new_prop->length)))
+ goto out;
/* Perform property updates */
ret = nx842_OF_upd_status(new_devdata, status);
Clarksboro MCH (Intel 7300 chipset).
config EDAC_SBRIDGE
- tristate "Intel Sandy-Bridge Integrated MC"
+ tristate "Intel Sandy-Bridge/Ivy-Bridge/Haswell Integrated MC"
depends on EDAC_MM_EDAC && PCI && X86_64 && X86_MCE_INTEL
depends on PCI_MMCONFIG
help
Support for error detection and correction the Intel
- Sandy Bridge Integrated Memory Controller.
+ Sandy Bridge, Ivy Bridge and Haswell Integrated Memory Controllers.
config EDAC_MPC85XX
tristate "Freescale MPC83xx / MPC85xx"
int j;
u32 nr_pages;
- for (np = NULL;
- (np = of_find_node_by_name(np, "memory")) != NULL;) {
+ for_each_node_by_name(np, "memory") {
struct resource r;
/* We "know" that the Cell firmware only creates one entry
[MEM_RDDR2] = "Registered-DDR2",
[MEM_XDR] = "XDR",
[MEM_DDR3] = "Unbuffered-DDR3",
- [MEM_RDDR3] = "Registered-DDR3"
+ [MEM_RDDR3] = "Registered-DDR3",
+ [MEM_DDR4] = "Unbuffered-DDR4",
+ [MEM_RDDR4] = "Registered-DDR4"
};
static const char * const dev_types[] = {
#define DRAM_ATTR(reg) GET_BITFIELD(reg, 2, 3)
#define INTERLEAVE_MODE(reg) GET_BITFIELD(reg, 1, 1)
#define DRAM_RULE_ENABLE(reg) GET_BITFIELD(reg, 0, 0)
+#define A7MODE(reg) GET_BITFIELD(reg, 26, 26)
static char *get_dram_attr(u32 reg)
{
#define TOLM 0x80
#define TOHM 0x84
+#define HASWELL_TOHM_0 0xd4
+#define HASWELL_TOHM_1 0xd8
#define GET_TOLM(reg) ((GET_BITFIELD(reg, 0, 3) << 28) | 0x3ffffff)
#define GET_TOHM(reg) ((GET_BITFIELD(reg, 0, 20) << 25) | 0x3ffffff)
#define SAD_CONTROL 0xf4
-#define NODE_ID(reg) GET_BITFIELD(reg, 0, 2)
-
/* Device 14 function 0 */
static const u32 tad_dram_rule[] = {
#define IS_RIR_VALID(reg) GET_BITFIELD(reg, 31, 31)
#define RIR_WAY(reg) GET_BITFIELD(reg, 28, 29)
-#define RIR_LIMIT(reg) ((GET_BITFIELD(reg, 1, 10) << 29)| 0x1fffffff)
#define MAX_RIR_WAY 8
#define IB_RANK_CFG_A 0x0320
-#define IS_RDIMM_ENABLED(reg) GET_BITFIELD(reg, 11, 11)
-
/*
* sbridge structs
*/
enum type {
SANDY_BRIDGE,
IVY_BRIDGE,
+ HASWELL,
};
struct sbridge_pvt;
u32 rankcfgr;
u64 (*get_tolm)(struct sbridge_pvt *pvt);
u64 (*get_tohm)(struct sbridge_pvt *pvt);
+ u64 (*rir_limit)(u32 reg);
const u32 *dram_rule;
const u32 *interleave_list;
const struct interleave_pkg *interleave_pkg;
u8 max_sad;
u8 max_interleave;
+ u8 (*get_node_id)(struct sbridge_pvt *pvt);
+ enum mem_type (*get_memory_type)(struct sbridge_pvt *pvt);
+ struct pci_dev *pci_vtd;
};
struct sbridge_channel {
};
struct pci_id_descr {
- int dev;
- int func;
- int dev_id;
+ int dev_id;
int optional;
};
struct pci_dev *pci_sad0, *pci_sad1;
struct pci_dev *pci_ha0, *pci_ha1;
struct pci_dev *pci_br0, *pci_br1;
+ struct pci_dev *pci_ha1_ta;
struct pci_dev *pci_tad[NUM_CHANNELS];
struct sbridge_dev *sbridge_dev;
u64 tolm, tohm;
};
-#define PCI_DESCR(device, function, device_id, opt) \
- .dev = (device), \
- .func = (function), \
- .dev_id = (device_id), \
+#define PCI_DESCR(device_id, opt) \
+ .dev_id = (device_id), \
.optional = opt
static const struct pci_id_descr pci_dev_descr_sbridge[] = {
/* Processor Home Agent */
- { PCI_DESCR(14, 0, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_HA0, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_HA0, 0) },
/* Memory controller */
- { PCI_DESCR(15, 0, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TA, 0) },
- { PCI_DESCR(15, 1, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_RAS, 0) },
- { PCI_DESCR(15, 2, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD0, 0) },
- { PCI_DESCR(15, 3, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD1, 0) },
- { PCI_DESCR(15, 4, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD2, 0) },
- { PCI_DESCR(15, 5, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD3, 0) },
- { PCI_DESCR(17, 0, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_DDRIO, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TA, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_RAS, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD0, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD1, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD2, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD3, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_DDRIO, 1) },
/* System Address Decoder */
- { PCI_DESCR(12, 6, PCI_DEVICE_ID_INTEL_SBRIDGE_SAD0, 0) },
- { PCI_DESCR(12, 7, PCI_DEVICE_ID_INTEL_SBRIDGE_SAD1, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_SAD0, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_SAD1, 0) },
/* Broadcast Registers */
- { PCI_DESCR(13, 6, PCI_DEVICE_ID_INTEL_SBRIDGE_BR, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_SBRIDGE_BR, 0) },
};
#define PCI_ID_TABLE_ENTRY(A) { .descr=A, .n_devs = ARRAY_SIZE(A) }
static const struct pci_id_descr pci_dev_descr_ibridge[] = {
/* Processor Home Agent */
- { PCI_DESCR(14, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0, 0) },
/* Memory controller */
- { PCI_DESCR(15, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA, 0) },
- { PCI_DESCR(15, 1, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_RAS, 0) },
- { PCI_DESCR(15, 2, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD0, 0) },
- { PCI_DESCR(15, 3, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD1, 0) },
- { PCI_DESCR(15, 4, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD2, 0) },
- { PCI_DESCR(15, 5, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD3, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_RAS, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD0, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD1, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD2, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD3, 0) },
/* System Address Decoder */
- { PCI_DESCR(22, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_SAD, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_SAD, 0) },
/* Broadcast Registers */
- { PCI_DESCR(22, 1, PCI_DEVICE_ID_INTEL_IBRIDGE_BR0, 1) },
- { PCI_DESCR(22, 2, PCI_DEVICE_ID_INTEL_IBRIDGE_BR1, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_BR0, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_BR1, 0) },
/* Optional, mode 2HA */
- { PCI_DESCR(28, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1, 1) },
#if 0
- { PCI_DESCR(29, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TA, 1) },
- { PCI_DESCR(29, 1, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_RAS, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TA, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_RAS, 1) },
#endif
- { PCI_DESCR(29, 2, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD0, 1) },
- { PCI_DESCR(29, 3, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD1, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD0, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD1, 1) },
- { PCI_DESCR(17, 0, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_1HA_DDRIO0, 1) },
- { PCI_DESCR(17, 4, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_2HA_DDRIO0, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_1HA_DDRIO0, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_2HA_DDRIO0, 1) },
};
static const struct pci_id_table pci_dev_descr_ibridge_table[] = {
{0,} /* 0 terminated list. */
};
+/* Haswell support */
+/* EN processor:
+ * - 1 IMC
+ * - 3 DDR3 channels, 2 DPC per channel
+ * EP processor:
+ * - 1 or 2 IMC
+ * - 4 DDR4 channels, 3 DPC per channel
+ * EP 4S processor:
+ * - 2 IMC
+ * - 4 DDR4 channels, 3 DPC per channel
+ * EX processor:
+ * - 2 IMC
+ * - each IMC interfaces with a SMI 2 channel
+ * - each SMI channel interfaces with a scalable memory buffer
+ * - each scalable memory buffer supports 4 DDR3/DDR4 channels, 3 DPC
+ */
+#define HASWELL_DDRCRCLKCONTROLS 0xa10
+#define HASWELL_HASYSDEFEATURE2 0x84
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_VTD_MISC 0x2f28
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0 0x2fa0
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1 0x2f60
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TA 0x2fa8
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_THERMAL 0x2f71
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TA 0x2f68
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_THERMAL 0x2f79
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_CBO_SAD0 0x2ffc
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_CBO_SAD1 0x2ffd
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD0 0x2faa
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD1 0x2fab
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD2 0x2fac
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD3 0x2fad
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD0 0x2f6a
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD1 0x2f6b
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD2 0x2f6c
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD3 0x2f6d
+#define PCI_DEVICE_ID_INTEL_HASWELL_IMC_DDRIO0 0x2fbd
+static const struct pci_id_descr pci_dev_descr_haswell[] = {
+ /* first item must be the HA */
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0, 0) },
+
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_CBO_SAD0, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_CBO_SAD1, 0) },
+
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1, 1) },
+
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TA, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_THERMAL, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD0, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD1, 0) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD2, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD3, 1) },
+
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_DDRIO0, 1) },
+
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TA, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_THERMAL, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD0, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD1, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD2, 1) },
+ { PCI_DESCR(PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD3, 1) },
+};
+
+static const struct pci_id_table pci_dev_descr_haswell_table[] = {
+ PCI_ID_TABLE_ENTRY(pci_dev_descr_haswell),
+ {0,} /* 0 terminated list. */
+};
+
/*
* pci_device_id table for which devices we are looking for
*/
static const struct pci_device_id sbridge_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TA)},
{PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA)},
+ {PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0)},
{0,} /* 0 terminated list. */
};
Ancillary status routines
****************************************************************************/
-static inline int numrank(u32 mtr)
+static inline int numrank(enum type type, u32 mtr)
{
int ranks = (1 << RANK_CNT_BITS(mtr));
+ int max = 4;
- if (ranks > 4) {
- edac_dbg(0, "Invalid number of ranks: %d (max = 4) raw value = %x (%04x)\n",
- ranks, (unsigned int)RANK_CNT_BITS(mtr), mtr);
+ if (type == HASWELL)
+ max = 8;
+
+ if (ranks > max) {
+ edac_dbg(0, "Invalid number of ranks: %d (max = %i) raw value = %x (%04x)\n",
+ ranks, max, (unsigned int)RANK_CNT_BITS(mtr), mtr);
return -EINVAL;
}
return GET_TOHM(reg);
}
+static u64 rir_limit(u32 reg)
+{
+ return ((u64)GET_BITFIELD(reg, 1, 10) << 29) | 0x1fffffff;
+}
+
+static enum mem_type get_memory_type(struct sbridge_pvt *pvt)
+{
+ u32 reg;
+ enum mem_type mtype;
+
+ if (pvt->pci_ddrio) {
+ pci_read_config_dword(pvt->pci_ddrio, pvt->info.rankcfgr,
+ ®);
+ if (GET_BITFIELD(reg, 11, 11))
+ /* FIXME: Can also be LRDIMM */
+ mtype = MEM_RDDR3;
+ else
+ mtype = MEM_DDR3;
+ } else
+ mtype = MEM_UNKNOWN;
+
+ return mtype;
+}
+
+static enum mem_type haswell_get_memory_type(struct sbridge_pvt *pvt)
+{
+ u32 reg;
+ bool registered = false;
+ enum mem_type mtype = MEM_UNKNOWN;
+
+ if (!pvt->pci_ddrio)
+ goto out;
+
+ pci_read_config_dword(pvt->pci_ddrio,
+ HASWELL_DDRCRCLKCONTROLS, ®);
+ /* Is_Rdimm */
+ if (GET_BITFIELD(reg, 16, 16))
+ registered = true;
+
+ pci_read_config_dword(pvt->pci_ta, MCMTR, ®);
+ if (GET_BITFIELD(reg, 14, 14)) {
+ if (registered)
+ mtype = MEM_RDDR4;
+ else
+ mtype = MEM_DDR4;
+ } else {
+ if (registered)
+ mtype = MEM_RDDR3;
+ else
+ mtype = MEM_DDR3;
+ }
+
+out:
+ return mtype;
+}
+
+static u8 get_node_id(struct sbridge_pvt *pvt)
+{
+ u32 reg;
+ pci_read_config_dword(pvt->pci_br0, SAD_CONTROL, ®);
+ return GET_BITFIELD(reg, 0, 2);
+}
+
+static u8 haswell_get_node_id(struct sbridge_pvt *pvt)
+{
+ u32 reg;
+
+ pci_read_config_dword(pvt->pci_sad1, SAD_CONTROL, ®);
+ return GET_BITFIELD(reg, 0, 3);
+}
+
+static u64 haswell_get_tolm(struct sbridge_pvt *pvt)
+{
+ u32 reg;
+
+ pci_read_config_dword(pvt->info.pci_vtd, TOLM, ®);
+ return (GET_BITFIELD(reg, 26, 31) << 26) | 0x1ffffff;
+}
+
+static u64 haswell_get_tohm(struct sbridge_pvt *pvt)
+{
+ u64 rc;
+ u32 reg;
+
+ pci_read_config_dword(pvt->info.pci_vtd, HASWELL_TOHM_0, ®);
+ rc = GET_BITFIELD(reg, 26, 31);
+ pci_read_config_dword(pvt->info.pci_vtd, HASWELL_TOHM_1, ®);
+ rc = ((reg << 6) | rc) << 26;
+
+ return rc | 0x1ffffff;
+}
+
+static u64 haswell_rir_limit(u32 reg)
+{
+ return (((u64)GET_BITFIELD(reg, 1, 11) + 1) << 29) - 1;
+}
+
static inline u8 sad_pkg_socket(u8 pkg)
{
/* on Ivy Bridge, nodeID is SASS, where A is HA and S is node id */
- return (pkg >> 3) | (pkg & 0x3);
+ return ((pkg >> 3) << 2) | (pkg & 0x3);
}
static inline u8 sad_pkg_ha(u8 pkg)
/****************************************************************************
Memory check routines
****************************************************************************/
-static struct pci_dev *get_pdev_slot_func(u8 bus, unsigned slot,
- unsigned func)
+static struct pci_dev *get_pdev_same_bus(u8 bus, u32 id)
{
- struct sbridge_dev *sbridge_dev = get_sbridge_dev(bus);
- int i;
-
- if (!sbridge_dev)
- return NULL;
-
- for (i = 0; i < sbridge_dev->n_devs; i++) {
- if (!sbridge_dev->pdev[i])
- continue;
+ struct pci_dev *pdev = NULL;
- if (PCI_SLOT(sbridge_dev->pdev[i]->devfn) == slot &&
- PCI_FUNC(sbridge_dev->pdev[i]->devfn) == func) {
- edac_dbg(1, "Associated %02x.%02x.%d with %p\n",
- bus, slot, func, sbridge_dev->pdev[i]);
- return sbridge_dev->pdev[i];
- }
- }
+ do {
+ pdev = pci_get_device(PCI_VENDOR_ID_INTEL, id, pdev);
+ if (pdev && pdev->bus->number == bus)
+ break;
+ } while (pdev);
- return NULL;
+ return pdev;
}
/**
* check_if_ecc_is_active() - Checks if ECC is active
- * bus: Device bus
+ * @bus: Device bus
+ * @type: Memory controller type
+ * returns: 0 in case ECC is active, -ENODEV if it can't be determined or
+ * disabled
*/
-static int check_if_ecc_is_active(const u8 bus)
+static int check_if_ecc_is_active(const u8 bus, enum type type)
{
struct pci_dev *pdev = NULL;
- u32 mcmtr;
+ u32 mcmtr, id;
+
+ if (type == IVY_BRIDGE)
+ id = PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA;
+ else if (type == HASWELL)
+ id = PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TA;
+ else
+ id = PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TA;
- pdev = get_pdev_slot_func(bus, 15, 0);
+ pdev = get_pdev_same_bus(bus, id);
if (!pdev) {
sbridge_printk(KERN_ERR, "Couldn't find PCI device "
- "%2x.%02d.%d!!!\n",
- bus, 15, 0);
+ "%04x:%04x! on bus %02d\n",
+ PCI_VENDOR_ID_INTEL, id, bus);
return -ENODEV;
}
enum edac_type mode;
enum mem_type mtype;
- pci_read_config_dword(pvt->pci_br0, SAD_TARGET, ®);
+ if (pvt->info.type == HASWELL)
+ pci_read_config_dword(pvt->pci_sad1, SAD_TARGET, ®);
+ else
+ pci_read_config_dword(pvt->pci_br0, SAD_TARGET, ®);
+
pvt->sbridge_dev->source_id = SOURCE_ID(reg);
- pci_read_config_dword(pvt->pci_br0, SAD_CONTROL, ®);
- pvt->sbridge_dev->node_id = NODE_ID(reg);
+ pvt->sbridge_dev->node_id = pvt->info.get_node_id(pvt);
edac_dbg(0, "mc#%d: Node ID: %d, source ID: %d\n",
pvt->sbridge_dev->mc,
pvt->sbridge_dev->node_id,
pvt->is_close_pg = false;
}
- if (pvt->pci_ddrio) {
- pci_read_config_dword(pvt->pci_ddrio, pvt->info.rankcfgr,
- ®);
- if (IS_RDIMM_ENABLED(reg)) {
- /* FIXME: Can also be LRDIMM */
- edac_dbg(0, "Memory is registered\n");
- mtype = MEM_RDDR3;
- } else {
- edac_dbg(0, "Memory is unregistered\n");
- mtype = MEM_DDR3;
- }
- } else {
+ mtype = pvt->info.get_memory_type(pvt);
+ if (mtype == MEM_RDDR3 || mtype == MEM_RDDR4)
+ edac_dbg(0, "Memory is registered\n");
+ else if (mtype == MEM_UNKNOWN)
edac_dbg(0, "Cannot determine memory type\n");
- mtype = MEM_UNKNOWN;
- }
+ else
+ edac_dbg(0, "Memory is unregistered\n");
- /* On all supported DDR3 DIMM types, there are 8 banks available */
- banks = 8;
+ if (mtype == MEM_DDR4 || MEM_RDDR4)
+ banks = 16;
+ else
+ banks = 8;
for (i = 0; i < NUM_CHANNELS; i++) {
u32 mtr;
if (IS_DIMM_PRESENT(mtr)) {
pvt->channel[i].dimms++;
- ranks = numrank(mtr);
+ ranks = numrank(pvt->info.type, mtr);
rows = numrow(mtr);
cols = numcol(mtr);
- /* DDR3 has 8 I/O banks */
size = ((u64)rows * cols * banks * ranks) >> (20 - 3);
npages = MiB_TO_PAGES(size);
dimm->nr_pages = npages;
dimm->grain = 32;
- dimm->dtype = (banks == 8) ? DEV_X8 : DEV_X4;
+ switch (banks) {
+ case 16:
+ dimm->dtype = DEV_X16;
+ break;
+ case 8:
+ dimm->dtype = DEV_X8;
+ break;
+ case 4:
+ dimm->dtype = DEV_X4;
+ break;
+ }
dimm->mtype = mtype;
dimm->edac_mode = mode;
snprintf(dimm->label, sizeof(dimm->label),
if (!IS_RIR_VALID(reg))
continue;
- tmp_mb = RIR_LIMIT(reg) >> 20;
+ tmp_mb = pvt->info.rir_limit(reg) >> 20;
rir_way = 1 << RIR_WAY(reg);
mb = div_u64_rem(tmp_mb, 1000, &kb);
edac_dbg(0, "CH#%d RIR#%d, limit: %u.%03u GB (0x%016Lx), way: %d, reg=0x%08x\n",
struct mem_ctl_info *new_mci;
struct sbridge_pvt *pvt = mci->pvt_info;
struct pci_dev *pci_ha;
- int n_rir, n_sads, n_tads, sad_way, sck_xch;
+ int n_rir, n_sads, n_tads, sad_way, sck_xch;
int sad_interl, idx, base_ch;
- int interleave_mode;
+ int interleave_mode, shiftup = 0;
unsigned sad_interleave[pvt->info.max_interleave];
- u32 reg;
+ u32 reg, dram_rule;
u8 ch_way, sck_way, pkg, sad_ha = 0;
u32 tad_offset;
u32 rir_way;
sprintf(msg, "Can't discover the memory socket");
return -EINVAL;
}
- *area_type = get_dram_attr(reg);
- interleave_mode = INTERLEAVE_MODE(reg);
+ dram_rule = reg;
+ *area_type = get_dram_attr(dram_rule);
+ interleave_mode = INTERLEAVE_MODE(dram_rule);
pci_read_config_dword(pvt->pci_sad0, pvt->info.interleave_list[n_sads],
®);
*socket = sad_interleave[idx];
edac_dbg(0, "SAD interleave index: %d (wayness %d) = CPU socket %d\n",
idx, sad_way, *socket);
+ } else if (pvt->info.type == HASWELL) {
+ int bits, a7mode = A7MODE(dram_rule);
+
+ if (a7mode) {
+ /* A7 mode swaps P9 with P6 */
+ bits = GET_BITFIELD(addr, 7, 8) << 1;
+ bits |= GET_BITFIELD(addr, 9, 9);
+ } else
+ bits = GET_BITFIELD(addr, 7, 9);
+
+ if (interleave_mode) {
+ /* interleave mode will XOR {8,7,6} with {18,17,16} */
+ idx = GET_BITFIELD(addr, 16, 18);
+ idx ^= bits;
+ } else
+ idx = bits;
+
+ pkg = sad_pkg(pvt->info.interleave_pkg, reg, idx);
+ *socket = sad_pkg_socket(pkg);
+ sad_ha = sad_pkg_ha(pkg);
+
+ if (a7mode) {
+ /* MCChanShiftUpEnable */
+ pci_read_config_dword(pvt->pci_ha0,
+ HASWELL_HASYSDEFEATURE2, ®);
+ shiftup = GET_BITFIELD(reg, 22, 22);
+ }
+
+ edac_dbg(0, "SAD interleave package: %d = CPU socket %d, HA %i, shiftup: %i\n",
+ idx, *socket, sad_ha, shiftup);
} else {
/* Ivy Bridge's SAD mode doesn't support XOR interleave mode */
idx = (addr >> 6) & 7;
if (ch_way == 3)
idx = addr >> 6;
else
- idx = addr >> (6 + sck_way);
+ idx = (addr >> (6 + sck_way + shiftup)) & 0x3;
idx = idx % ch_way;
/*
if (!IS_RIR_VALID(reg))
continue;
- limit = RIR_LIMIT(reg);
+ limit = pvt->info.rir_limit(reg);
mb = div_u64_rem(limit >> 20, 1000, &kb);
edac_dbg(0, "RIR#%d, limit: %u.%03u GB (0x%016Lx), way: %d\n",
n_rir,
return -EINVAL;
}
rir_way = RIR_WAY(reg);
+
if (pvt->is_close_pg)
idx = (ch_addr >> 6);
else
{
struct sbridge_dev *sbridge_dev;
const struct pci_id_descr *dev_descr = &table->descr[devno];
-
struct pci_dev *pdev = NULL;
u8 bus = 0;
sbridge_printk(KERN_DEBUG,
- "Seeking for: dev %02x.%d PCI ID %04x:%04x\n",
- dev_descr->dev, dev_descr->func,
+ "Seeking for: PCI ID %04x:%04x\n",
PCI_VENDOR_ID_INTEL, dev_descr->dev_id);
pdev = pci_get_device(PCI_VENDOR_ID_INTEL,
if (dev_descr->optional)
return 0;
+ /* if the HA wasn't found */
if (devno == 0)
return -ENODEV;
sbridge_printk(KERN_INFO,
- "Device not found: dev %02x.%d PCI ID %04x:%04x\n",
- dev_descr->dev, dev_descr->func,
+ "Device not found: %04x:%04x\n",
PCI_VENDOR_ID_INTEL, dev_descr->dev_id);
/* End of list, leave */
if (sbridge_dev->pdev[devno]) {
sbridge_printk(KERN_ERR,
- "Duplicated device for "
- "dev %02x:%d.%d PCI ID %04x:%04x\n",
- bus, dev_descr->dev, dev_descr->func,
+ "Duplicated device for %04x:%04x\n",
PCI_VENDOR_ID_INTEL, dev_descr->dev_id);
pci_dev_put(pdev);
return -ENODEV;
sbridge_dev->pdev[devno] = pdev;
- /* Sanity check */
- if (unlikely(PCI_SLOT(pdev->devfn) != dev_descr->dev ||
- PCI_FUNC(pdev->devfn) != dev_descr->func)) {
- sbridge_printk(KERN_ERR,
- "Device PCI ID %04x:%04x "
- "has dev %02x:%d.%d instead of dev %02x:%02x.%d\n",
- PCI_VENDOR_ID_INTEL, dev_descr->dev_id,
- bus, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn),
- bus, dev_descr->dev, dev_descr->func);
- return -ENODEV;
- }
-
/* Be sure that the device is enabled */
if (unlikely(pci_enable_device(pdev) < 0)) {
sbridge_printk(KERN_ERR,
- "Couldn't enable "
- "dev %02x:%d.%d PCI ID %04x:%04x\n",
- bus, dev_descr->dev, dev_descr->func,
+ "Couldn't enable %04x:%04x\n",
PCI_VENDOR_ID_INTEL, dev_descr->dev_id);
return -ENODEV;
}
- edac_dbg(0, "Detected dev %02x:%d.%d PCI ID %04x:%04x\n",
- bus, dev_descr->dev, dev_descr->func,
+ edac_dbg(0, "Detected %04x:%04x\n",
PCI_VENDOR_ID_INTEL, dev_descr->dev_id);
/*
/*
* sbridge_get_all_devices - Find and perform 'get' operation on the MCH's
- * device/functions we want to reference for this driver.
- * Need to 'get' device 16 func 1 and func 2.
+ * devices we want to reference for this driver.
* @num_mc: pointer to the memory controllers count, to be incremented in case
- * of success.
+ * of success.
* @table: model specific table
*
* returns 0 in case of success or error code
{
struct sbridge_pvt *pvt = mci->pvt_info;
struct pci_dev *pdev;
- int i, func, slot;
+ int i;
for (i = 0; i < sbridge_dev->n_devs; i++) {
pdev = sbridge_dev->pdev[i];
if (!pdev)
continue;
- slot = PCI_SLOT(pdev->devfn);
- func = PCI_FUNC(pdev->devfn);
- switch (slot) {
- case 12:
- switch (func) {
- case 6:
- pvt->pci_sad0 = pdev;
- break;
- case 7:
- pvt->pci_sad1 = pdev;
- break;
- default:
- goto error;
- }
+
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_SAD0:
+ pvt->pci_sad0 = pdev;
break;
- case 13:
- switch (func) {
- case 6:
- pvt->pci_br0 = pdev;
- break;
- default:
- goto error;
- }
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_SAD1:
+ pvt->pci_sad1 = pdev;
break;
- case 14:
- switch (func) {
- case 0:
- pvt->pci_ha0 = pdev;
- break;
- default:
- goto error;
- }
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_BR:
+ pvt->pci_br0 = pdev;
break;
- case 15:
- switch (func) {
- case 0:
- pvt->pci_ta = pdev;
- break;
- case 1:
- pvt->pci_ras = pdev;
- break;
- case 2:
- case 3:
- case 4:
- case 5:
- pvt->pci_tad[func - 2] = pdev;
- break;
- default:
- goto error;
- }
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_HA0:
+ pvt->pci_ha0 = pdev;
break;
- case 17:
- switch (func) {
- case 0:
- pvt->pci_ddrio = pdev;
- break;
- default:
- goto error;
- }
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TA:
+ pvt->pci_ta = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_RAS:
+ pvt->pci_ras = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD0:
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD1:
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD2:
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD3:
+ {
+ int id = pdev->device - PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TAD0;
+ pvt->pci_tad[id] = pdev;
+ }
+ break;
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_DDRIO:
+ pvt->pci_ddrio = pdev;
break;
default:
goto error;
}
- edac_dbg(0, "Associated PCI %02x.%02d.%d with dev = %p\n",
+ edac_dbg(0, "Associated PCI %02x:%02x, bus %d with dev = %p\n",
+ pdev->vendor, pdev->device,
sbridge_dev->bus,
- PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn),
pdev);
}
return -ENODEV;
error:
- sbridge_printk(KERN_ERR, "Device %d, function %d "
- "is out of the expected range\n",
- slot, func);
+ sbridge_printk(KERN_ERR, "Unexpected device %02x:%02x\n",
+ PCI_VENDOR_ID_INTEL, pdev->device);
return -EINVAL;
}
{
struct sbridge_pvt *pvt = mci->pvt_info;
struct pci_dev *pdev, *tmp;
- int i, func, slot;
+ int i;
bool mode_2ha = false;
tmp = pci_get_device(PCI_VENDOR_ID_INTEL,
pdev = sbridge_dev->pdev[i];
if (!pdev)
continue;
- slot = PCI_SLOT(pdev->devfn);
- func = PCI_FUNC(pdev->devfn);
- switch (slot) {
- case 14:
- if (func == 0) {
- pvt->pci_ha0 = pdev;
- break;
- }
- goto error;
- case 15:
- switch (func) {
- case 0:
- pvt->pci_ta = pdev;
- break;
- case 1:
- pvt->pci_ras = pdev;
- break;
- case 4:
- case 5:
- /* if we have 2 HAs active, channels 2 and 3
- * are in other device */
- if (mode_2ha)
- break;
- /* fall through */
- case 2:
- case 3:
- pvt->pci_tad[func - 2] = pdev;
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0:
+ pvt->pci_ha0 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA:
+ pvt->pci_ta = pdev;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_RAS:
+ pvt->pci_ras = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD2:
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD3:
+ /* if we have 2 HAs active, channels 2 and 3
+ * are in other device */
+ if (mode_2ha)
break;
- default:
- goto error;
- }
+ /* fall through */
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD0:
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD1:
+ {
+ int id = pdev->device - PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TAD0;
+ pvt->pci_tad[id] = pdev;
+ }
break;
- case 17:
- if (func == 4) {
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_2HA_DDRIO0:
+ pvt->pci_ddrio = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_1HA_DDRIO0:
+ if (!mode_2ha)
pvt->pci_ddrio = pdev;
- break;
- } else if (func == 0) {
- if (!mode_2ha)
- pvt->pci_ddrio = pdev;
- break;
- }
- goto error;
- case 22:
- switch (func) {
- case 0:
- pvt->pci_sad0 = pdev;
- break;
- case 1:
- pvt->pci_br0 = pdev;
- break;
- case 2:
- pvt->pci_br1 = pdev;
- break;
- default:
- goto error;
- }
break;
- case 28:
- if (func == 0) {
- pvt->pci_ha1 = pdev;
- break;
- }
- goto error;
- case 29:
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_SAD:
+ pvt->pci_sad0 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_BR0:
+ pvt->pci_br0 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_BR1:
+ pvt->pci_br1 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1:
+ pvt->pci_ha1 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD0:
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD1:
+ {
+ int id = pdev->device - PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA1_TAD0 + 2;
+
/* we shouldn't have this device if we have just one
* HA present */
WARN_ON(!mode_2ha);
- if (func == 2 || func == 3) {
- pvt->pci_tad[func] = pdev;
- break;
- }
- goto error;
+ pvt->pci_tad[id] = pdev;
+ }
+ break;
default:
goto error;
}
error:
sbridge_printk(KERN_ERR,
- "Device %d, function %d is out of the expected range\n",
- slot, func);
+ "Unexpected device %02x:%02x\n", PCI_VENDOR_ID_INTEL,
+ pdev->device);
return -EINVAL;
}
+static int haswell_mci_bind_devs(struct mem_ctl_info *mci,
+ struct sbridge_dev *sbridge_dev)
+{
+ struct sbridge_pvt *pvt = mci->pvt_info;
+ struct pci_dev *pdev, *tmp;
+ int i;
+ bool mode_2ha = false;
+
+ tmp = pci_get_device(PCI_VENDOR_ID_INTEL,
+ PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1, NULL);
+ if (tmp) {
+ mode_2ha = true;
+ pci_dev_put(tmp);
+ }
+
+ /* there's only one device per system; not tied to any bus */
+ if (pvt->info.pci_vtd == NULL)
+ /* result will be checked later */
+ pvt->info.pci_vtd = pci_get_device(PCI_VENDOR_ID_INTEL,
+ PCI_DEVICE_ID_INTEL_HASWELL_IMC_VTD_MISC,
+ NULL);
+
+ for (i = 0; i < sbridge_dev->n_devs; i++) {
+ pdev = sbridge_dev->pdev[i];
+ if (!pdev)
+ continue;
+
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_CBO_SAD0:
+ pvt->pci_sad0 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_CBO_SAD1:
+ pvt->pci_sad1 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0:
+ pvt->pci_ha0 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TA:
+ pvt->pci_ta = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_THERMAL:
+ pvt->pci_ras = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD0:
+ pvt->pci_tad[0] = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD1:
+ pvt->pci_tad[1] = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD2:
+ if (!mode_2ha)
+ pvt->pci_tad[2] = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0_TAD3:
+ if (!mode_2ha)
+ pvt->pci_tad[3] = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_DDRIO0:
+ pvt->pci_ddrio = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1:
+ pvt->pci_ha1 = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TA:
+ pvt->pci_ha1_ta = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD0:
+ if (mode_2ha)
+ pvt->pci_tad[2] = pdev;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA1_TAD1:
+ if (mode_2ha)
+ pvt->pci_tad[3] = pdev;
+ break;
+ default:
+ break;
+ }
+
+ edac_dbg(0, "Associated PCI %02x.%02d.%d with dev = %p\n",
+ sbridge_dev->bus,
+ PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn),
+ pdev);
+ }
+
+ /* Check if everything were registered */
+ if (!pvt->pci_sad0 || !pvt->pci_ha0 || !pvt->pci_sad1 ||
+ !pvt->pci_ras || !pvt->pci_ta || !pvt->info.pci_vtd)
+ goto enodev;
+
+ for (i = 0; i < NUM_CHANNELS; i++) {
+ if (!pvt->pci_tad[i])
+ goto enodev;
+ }
+ return 0;
+
+enodev:
+ sbridge_printk(KERN_ERR, "Some needed devices are missing\n");
+ return -ENODEV;
+}
+
/****************************************************************************
Error check routines
****************************************************************************/
* EDAC core should be handling the channel mask, in order to point
* to the group of dimm's where the error may be happening.
*/
+ if (!pvt->is_lockstep && !pvt->is_mirrored && !pvt->is_close_pg)
+ channel = first_channel;
+
snprintf(msg, sizeof(msg),
"%s%s area:%s err_code:%04x:%04x socket:%d channel_mask:%ld rank:%d",
overflow ? " OVERFLOW" : "",
"%u APIC %x\n", mce->cpuvendor, mce->cpuid,
mce->time, mce->socketid, mce->apicid);
- /* Only handle if it is the right mc controller */
- if (cpu_data(mce->cpu).phys_proc_id != pvt->sbridge_dev->mc)
- return NOTIFY_DONE;
-
smp_rmb();
if ((pvt->mce_out + 1) % MCE_LOG_LEN == pvt->mce_in) {
smp_wmb();
int rc;
/* Check the number of active and not disabled channels */
- rc = check_if_ecc_is_active(sbridge_dev->bus);
+ rc = check_if_ecc_is_active(sbridge_dev->bus, type);
if (unlikely(rc < 0))
return rc;
mci->edac_check = sbridge_check_error;
pvt->info.type = type;
- if (type == IVY_BRIDGE) {
+ switch (type) {
+ case IVY_BRIDGE:
pvt->info.rankcfgr = IB_RANK_CFG_A;
pvt->info.get_tolm = ibridge_get_tolm;
pvt->info.get_tohm = ibridge_get_tohm;
pvt->info.dram_rule = ibridge_dram_rule;
+ pvt->info.get_memory_type = get_memory_type;
+ pvt->info.get_node_id = get_node_id;
+ pvt->info.rir_limit = rir_limit;
pvt->info.max_sad = ARRAY_SIZE(ibridge_dram_rule);
pvt->info.interleave_list = ibridge_interleave_list;
pvt->info.max_interleave = ARRAY_SIZE(ibridge_interleave_list);
rc = ibridge_mci_bind_devs(mci, sbridge_dev);
if (unlikely(rc < 0))
goto fail0;
- } else {
+ break;
+ case SANDY_BRIDGE:
pvt->info.rankcfgr = SB_RANK_CFG_A;
pvt->info.get_tolm = sbridge_get_tolm;
pvt->info.get_tohm = sbridge_get_tohm;
pvt->info.dram_rule = sbridge_dram_rule;
+ pvt->info.get_memory_type = get_memory_type;
+ pvt->info.get_node_id = get_node_id;
+ pvt->info.rir_limit = rir_limit;
pvt->info.max_sad = ARRAY_SIZE(sbridge_dram_rule);
pvt->info.interleave_list = sbridge_interleave_list;
pvt->info.max_interleave = ARRAY_SIZE(sbridge_interleave_list);
rc = sbridge_mci_bind_devs(mci, sbridge_dev);
if (unlikely(rc < 0))
goto fail0;
- }
+ break;
+ case HASWELL:
+ /* rankcfgr isn't used */
+ pvt->info.get_tolm = haswell_get_tolm;
+ pvt->info.get_tohm = haswell_get_tohm;
+ pvt->info.dram_rule = ibridge_dram_rule;
+ pvt->info.get_memory_type = haswell_get_memory_type;
+ pvt->info.get_node_id = haswell_get_node_id;
+ pvt->info.rir_limit = haswell_rir_limit;
+ pvt->info.max_sad = ARRAY_SIZE(ibridge_dram_rule);
+ pvt->info.interleave_list = ibridge_interleave_list;
+ pvt->info.max_interleave = ARRAY_SIZE(ibridge_interleave_list);
+ pvt->info.interleave_pkg = ibridge_interleave_pkg;
+ mci->ctl_name = kasprintf(GFP_KERNEL, "Haswell Socket#%d", mci->mc_idx);
+ /* Store pci devices at mci for faster access */
+ rc = haswell_mci_bind_devs(mci, sbridge_dev);
+ if (unlikely(rc < 0))
+ goto fail0;
+ break;
+ }
/* Get dimm basic config and the memory layout */
get_dimm_config(mci);
static int sbridge_probe(struct pci_dev *pdev, const struct pci_device_id *id)
{
- int rc;
+ int rc = -ENODEV;
u8 mc, num_mc = 0;
struct sbridge_dev *sbridge_dev;
- enum type type;
+ enum type type = SANDY_BRIDGE;
/* get the pci devices we want to reserve for our use */
mutex_lock(&sbridge_edac_lock);
}
probed++;
- if (pdev->device == PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA) {
+ switch (pdev->device) {
+ case PCI_DEVICE_ID_INTEL_IBRIDGE_IMC_HA0_TA:
rc = sbridge_get_all_devices(&num_mc, pci_dev_descr_ibridge_table);
type = IVY_BRIDGE;
- } else {
+ break;
+ case PCI_DEVICE_ID_INTEL_SBRIDGE_IMC_TA:
rc = sbridge_get_all_devices(&num_mc, pci_dev_descr_sbridge_table);
type = SANDY_BRIDGE;
+ break;
+ case PCI_DEVICE_ID_INTEL_HASWELL_IMC_HA0:
+ rc = sbridge_get_all_devices(&num_mc, pci_dev_descr_haswell_table);
+ type = HASWELL;
+ break;
}
if (unlikely(rc < 0))
goto fail0;
list_for_each_entry(sbridge_dev, &sbridge_edac_list, list) {
edac_dbg(0, "Registering MC#%d (%d of %d)\n",
mc, mc + 1, num_mc);
+
sbridge_dev->mc = mc++;
rc = sbridge_register_mci(sbridge_dev, type);
if (unlikely(rc < 0))
.subdevice = PCI_ANY_ID, \
.driver_data = (unsigned long) info }
-static DEFINE_PCI_DEVICE_TABLE(pciidlist) = {
+static const struct pci_device_id pciidlist[] = {
AST_VGA_DEVICE(PCI_CHIP_AST2000, NULL),
AST_VGA_DEVICE(PCI_CHIP_AST2100, NULL),
/* AST_VGA_DEVICE(PCI_CHIP_AST1180, NULL), - don't bind to 1180 for now */
drm_put_dev(dev);
}
-static DEFINE_PCI_DEVICE_TABLE(bochs_pci_tbl) = {
+static const struct pci_device_id bochs_pci_tbl[] = {
{
.vendor = 0x1234,
.device = 0x1111,
static struct drm_driver driver;
/* only bind to the cirrus chip in qemu */
-static DEFINE_PCI_DEVICE_TABLE(pciidlist) = {
+static const struct pci_device_id pciidlist[] = {
{ PCI_VENDOR_ID_CIRRUS, PCI_DEVICE_ID_CIRRUS_5446, 0x1af4, 0x1100, 0,
0, 0 },
{0,}
kfree(gma_encoder);
}
-static DEFINE_PCI_DEVICE_TABLE(hdmi_ids) = {
+static const struct pci_device_id hdmi_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, 0x080d) },
{ 0 }
};
* PowerVR SGX545 - Cedartrail - Intel GMA 3650, Intel Atom D2550, D2700,
* N2800
*/
-static DEFINE_PCI_DEVICE_TABLE(pciidlist) = {
+static const struct pci_device_id pciidlist[] = {
{ 0x8086, 0x8108, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &psb_chip_ops },
{ 0x8086, 0x8109, PCI_ANY_ID, PCI_ANY_ID, 0, 0, (long) &psb_chip_ops },
#if defined(CONFIG_DRM_GMA600)
static struct drm_driver driver;
-static DEFINE_PCI_DEVICE_TABLE(pciidlist) = {
+static const struct pci_device_id pciidlist[] = {
{ PCI_VENDOR_ID_MATROX, 0x522, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_SE_A },
{ PCI_VENDOR_ID_MATROX, 0x524, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_SE_B },
{ PCI_VENDOR_ID_MATROX, 0x530, PCI_ANY_ID, PCI_ANY_ID, 0, 0, G200_EV },
if (ret == 0) {
client->notify[index] = notify;
notify->client = client;
- return 0;
+ return index;
}
}
kfree(notify);
- return 0;
+ return ret;
}
static int
device->oclass[NVDEV_SUBDEV_BUS ] = nvc0_bus_oclass;
device->oclass[NVDEV_SUBDEV_TIMER ] = &gk20a_timer_oclass;
device->oclass[NVDEV_SUBDEV_FB ] = gk20a_fb_oclass;
+ device->oclass[NVDEV_SUBDEV_LTC ] = gk104_ltc_oclass;
device->oclass[NVDEV_SUBDEV_IBUS ] = &gk20a_ibus_oclass;
device->oclass[NVDEV_SUBDEV_INSTMEM] = nv50_instmem_oclass;
device->oclass[NVDEV_SUBDEV_VM ] = &nvc0_vmmgr_oclass;
}
}
+ if (zbc < 0)
+ return zbc;
+
memcpy(priv->zbc_color[zbc].ds, ds, sizeof(priv->zbc_color[zbc].ds));
memcpy(priv->zbc_color[zbc].l2, l2, sizeof(priv->zbc_color[zbc].l2));
priv->zbc_color[zbc].format = format;
}
}
+ if (zbc < 0)
+ return zbc;
+
priv->zbc_depth[zbc].format = format;
priv->zbc_depth[zbc].ds = ds;
priv->zbc_depth[zbc].l2 = l2;
void *data;
int (*ntfy)(const void *, u32, const void *, u32);
- struct nvkm_client_notify *notify[8];
+ struct nvkm_client_notify *notify[16];
};
static inline struct nouveau_client *
struct nouveau_mem *mem, struct nouveau_object **pobject)
{
struct nouveau_object *engine = nv_object(bar);
- return nouveau_object_ctor(parent, engine, &nouveau_barobj_oclass,
- mem, 0, pobject);
+ int ret = -ENOMEM;
+ if (bar->iomem) {
+ ret = nouveau_object_ctor(parent, engine,
+ &nouveau_barobj_oclass,
+ mem, 0, pobject);
+ }
+ return ret;
}
int
if (ret)
return ret;
- if (nv_device_resource_len(device, 3) != 0)
+ if (nv_device_resource_len(device, 3) != 0) {
bar->iomem = ioremap(nv_device_resource_start(device, 3),
nv_device_resource_len(device, 3));
+ if (!bar->iomem)
+ nv_warn(bar, "PRAMIN ioremap failed\n");
+ }
return 0;
}
} else {
/* otherwise, address lowest common amount from 0GiB */
ret = nouveau_mm_init(&pfb->vram, rsvd_head,
- (bsize << 8) * parts, 1);
+ (bsize << 8) * parts - rsvd_head, 1);
if (ret)
return ret;
/* and the rest starting from (8GiB + common_size) */
offset = (0x0200000000ULL >> 12) + (bsize << 8);
- length = (ram->size >> 12) - (bsize << 8) - rsvd_tail;
+ length = (ram->size >> 12) - ((bsize * parts) << 8) - rsvd_tail;
ret = nouveau_mm_init(&pfb->vram, offset, length, 0);
if (ret)
if (ret) {
priv->num_tags = 0;
} else {
- u64 tag_base = (priv->tag_ram->offset << 12) + tag_margin;
+ u64 tag_base = ((u64)priv->tag_ram->offset << 12) + tag_margin;
tag_base += tag_align - 1;
ret = do_div(tag_base, tag_align);
struct ttm_mem_type_manager *man = &bdev->man[mem->mem_type];
struct nouveau_drm *drm = nouveau_bdev(bdev);
struct nouveau_mem *node = mem->mm_node;
- struct drm_device *dev = drm->dev;
int ret;
mem->bus.addr = NULL;
if (drm->agp.stat == ENABLED) {
mem->bus.offset = mem->start << PAGE_SHIFT;
mem->bus.base = drm->agp.base;
- mem->bus.is_iomem = !dev->agp->cant_use_aperture;
+ mem->bus.is_iomem = !drm->dev->agp->cant_use_aperture;
}
#endif
if (drm->device.info.family < NV_DEVICE_INFO_V0_TESLA || !node->memtype)
if (!nouveau_fb || !nouveau_fb->nvbo)
continue;
- nouveau_bo_pin(nouveau_fb->nvbo, TTM_PL_FLAG_VRAM);
+ ret = nouveau_bo_pin(nouveau_fb->nvbo, TTM_PL_FLAG_VRAM);
+ if (ret)
+ NV_ERROR(drm, "Could not pin framebuffer\n");
}
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
}
}
-void
+static void
nouveau_fbcon_accel_fini(struct drm_device *dev)
{
struct nouveau_drm *drm = nouveau_drm(dev);
}
}
-void
+static void
nouveau_fbcon_accel_init(struct drm_device *dev)
{
struct nouveau_drm *drm = nouveau_drm(dev);
static int nouveau_platform_remove(struct platform_device *pdev)
{
struct drm_device *drm_dev = platform_get_drvdata(pdev);
- struct nouveau_device *device = nouveau_dev(drm_dev);
+ struct nouveau_drm *drm = nouveau_drm(drm_dev);
+ struct nouveau_device *device = nvkm_device(&drm->device);
struct nouveau_platform_gpu *gpu = nv_device_to_platform(device)->gpu;
nouveau_drm_device_remove(drm_dev);
struct nv50_disp_dac_load_v0 {
__u8 version;
__u8 load;
- __u16 data;
- __u8 pad04[4];
+ __u8 pad02[2];
+ __u32 data;
};
struct nv50_disp_sor_pwr_v0 {
return 0;
}
+static inline int
+nvif_notify_func(struct nvif_notify *notify, bool keep)
+{
+ int ret = notify->func(notify);
+ if (ret == NVIF_NOTIFY_KEEP ||
+ !test_and_clear_bit(NVKM_NOTIFY_USER, ¬ify->flags)) {
+ if (!keep)
+ atomic_dec(¬ify->putcnt);
+ else
+ nvif_notify_get_(notify);
+ }
+ return ret;
+}
+
static void
nvif_notify_work(struct work_struct *work)
{
struct nvif_notify *notify = container_of(work, typeof(*notify), work);
- if (notify->func(notify) == NVIF_NOTIFY_KEEP)
- nvif_notify_get_(notify);
+ nvif_notify_func(notify, true);
}
int
if (!WARN_ON(notify == NULL)) {
struct nvif_client *client = nvif_client(notify->object);
if (!WARN_ON(notify->size != size)) {
+ atomic_inc(¬ify->putcnt);
if (test_bit(NVIF_NOTIFY_WORK, ¬ify->flags)) {
- atomic_inc(¬ify->putcnt);
memcpy((void *)notify->data, data, size);
schedule_work(¬ify->work);
return NVIF_NOTIFY_DROP;
}
notify->data = data;
- ret = notify->func(notify);
+ ret = nvif_notify_func(notify, client->driver->keep);
notify->data = NULL;
- if (ret != NVIF_NOTIFY_DROP && client->driver->keep) {
- atomic_inc(¬ify->putcnt);
- nvif_notify_get_(notify);
- }
}
}
if (notify) {
int ret = nvif_notify_init(object, nvif_notify_del, func, work,
type, data, size, reply, notify);
- if (ret)
+ if (ret) {
kfree(notify);
+ notify = NULL;
+ }
*pnotify = notify;
return ret;
}
if (object) {
int ret = nvif_object_init(parent, nvif_object_del, handle,
oclass, data, size, object);
- if (ret)
+ if (ret) {
kfree(object);
+ object = NULL;
+ }
*pobject = object;
return ret;
}
#include "qxl_object.h"
extern int qxl_max_ioctls;
-static DEFINE_PCI_DEVICE_TABLE(pciidlist) = {
+static const struct pci_device_id pciidlist[] = {
{ 0x1b36, 0x100, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_VGA << 8,
0xffff00, 0 },
{ 0x1b36, 0x100, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_DISPLAY_OTHER << 8,
static __u8 *ch_report_fixup(struct hid_device *hdev, __u8 *rdesc,
unsigned int *rsize)
{
- if (*rsize >= 17 && rdesc[11] == 0x3c && rdesc[12] == 0x02) {
+ if (*rsize >= 18 && rdesc[11] == 0x3c && rdesc[12] == 0x02) {
hid_info(hdev, "fixing up Cherry Cymotion report descriptor\n");
rdesc[11] = rdesc[16] = 0xff;
rdesc[12] = rdesc[17] = 0x03;
0xC0 /* End Collection */
};
+/* Parameter indices */
+enum huion_prm {
+ HUION_PRM_X_LM = 1,
+ HUION_PRM_Y_LM = 2,
+ HUION_PRM_PRESSURE_LM = 4,
+ HUION_PRM_RESOLUTION = 5,
+ HUION_PRM_NUM
+};
+
/* Driver data */
struct huion_drvdata {
__u8 *rdesc;
int rc;
struct usb_device *usb_dev = hid_to_usb_dev(hdev);
struct huion_drvdata *drvdata = hid_get_drvdata(hdev);
- __le16 buf[6];
+ __le16 *buf = NULL;
+ size_t len;
+ s32 params[HUION_PH_ID_NUM];
+ s32 resolution;
+ __u8 *p;
+ s32 v;
/*
* Read string descriptor containing tablet parameters. The specific
* driver traffic.
* NOTE: This enables fully-functional tablet mode.
*/
+ len = HUION_PRM_NUM * sizeof(*buf);
+ buf = kmalloc(len, GFP_KERNEL);
+ if (buf == NULL) {
+ hid_err(hdev, "failed to allocate parameter buffer\n");
+ rc = -ENOMEM;
+ goto cleanup;
+ }
rc = usb_control_msg(usb_dev, usb_rcvctrlpipe(usb_dev, 0),
USB_REQ_GET_DESCRIPTOR, USB_DIR_IN,
(USB_DT_STRING << 8) + 0x64,
- 0x0409, buf, sizeof(buf),
+ 0x0409, buf, len,
USB_CTRL_GET_TIMEOUT);
- if (rc == -EPIPE)
- hid_warn(hdev, "device parameters not found\n");
- else if (rc < 0)
- hid_warn(hdev, "failed to get device parameters: %d\n", rc);
- else if (rc != sizeof(buf))
- hid_warn(hdev, "invalid device parameters\n");
- else {
- s32 params[HUION_PH_ID_NUM];
- s32 resolution;
- __u8 *p;
- s32 v;
+ if (rc == -EPIPE) {
+ hid_err(hdev, "device parameters not found\n");
+ rc = -ENODEV;
+ goto cleanup;
+ } else if (rc < 0) {
+ hid_err(hdev, "failed to get device parameters: %d\n", rc);
+ rc = -ENODEV;
+ goto cleanup;
+ } else if (rc != len) {
+ hid_err(hdev, "invalid device parameters\n");
+ rc = -ENODEV;
+ goto cleanup;
+ }
- /* Extract device parameters */
- params[HUION_PH_ID_X_LM] = le16_to_cpu(buf[1]);
- params[HUION_PH_ID_Y_LM] = le16_to_cpu(buf[2]);
- params[HUION_PH_ID_PRESSURE_LM] = le16_to_cpu(buf[4]);
- resolution = le16_to_cpu(buf[5]);
- if (resolution == 0) {
- params[HUION_PH_ID_X_PM] = 0;
- params[HUION_PH_ID_Y_PM] = 0;
- } else {
- params[HUION_PH_ID_X_PM] = params[HUION_PH_ID_X_LM] *
- 1000 / resolution;
- params[HUION_PH_ID_Y_PM] = params[HUION_PH_ID_Y_LM] *
- 1000 / resolution;
- }
+ /* Extract device parameters */
+ params[HUION_PH_ID_X_LM] = le16_to_cpu(buf[HUION_PRM_X_LM]);
+ params[HUION_PH_ID_Y_LM] = le16_to_cpu(buf[HUION_PRM_Y_LM]);
+ params[HUION_PH_ID_PRESSURE_LM] =
+ le16_to_cpu(buf[HUION_PRM_PRESSURE_LM]);
+ resolution = le16_to_cpu(buf[HUION_PRM_RESOLUTION]);
+ if (resolution == 0) {
+ params[HUION_PH_ID_X_PM] = 0;
+ params[HUION_PH_ID_Y_PM] = 0;
+ } else {
+ params[HUION_PH_ID_X_PM] = params[HUION_PH_ID_X_LM] *
+ 1000 / resolution;
+ params[HUION_PH_ID_Y_PM] = params[HUION_PH_ID_Y_LM] *
+ 1000 / resolution;
+ }
- /* Allocate fixed report descriptor */
- drvdata->rdesc = devm_kmalloc(&hdev->dev,
- sizeof(huion_tablet_rdesc_template),
- GFP_KERNEL);
- if (drvdata->rdesc == NULL) {
- hid_err(hdev, "failed to allocate fixed rdesc\n");
- return -ENOMEM;
- }
- drvdata->rsize = sizeof(huion_tablet_rdesc_template);
+ /* Allocate fixed report descriptor */
+ drvdata->rdesc = devm_kmalloc(&hdev->dev,
+ sizeof(huion_tablet_rdesc_template),
+ GFP_KERNEL);
+ if (drvdata->rdesc == NULL) {
+ hid_err(hdev, "failed to allocate fixed rdesc\n");
+ rc = -ENOMEM;
+ goto cleanup;
+ }
+ drvdata->rsize = sizeof(huion_tablet_rdesc_template);
- /* Format fixed report descriptor */
- memcpy(drvdata->rdesc, huion_tablet_rdesc_template,
- drvdata->rsize);
- for (p = drvdata->rdesc;
- p <= drvdata->rdesc + drvdata->rsize - 4;) {
- if (p[0] == 0xFE && p[1] == 0xED && p[2] == 0x1D &&
- p[3] < sizeof(params)) {
- v = params[p[3]];
- put_unaligned(cpu_to_le32(v), (s32 *)p);
- p += 4;
- } else {
- p++;
- }
+ /* Format fixed report descriptor */
+ memcpy(drvdata->rdesc, huion_tablet_rdesc_template,
+ drvdata->rsize);
+ for (p = drvdata->rdesc;
+ p <= drvdata->rdesc + drvdata->rsize - 4;) {
+ if (p[0] == 0xFE && p[1] == 0xED && p[2] == 0x1D &&
+ p[3] < sizeof(params)) {
+ v = params[p[3]];
+ put_unaligned(cpu_to_le32(v), (s32 *)p);
+ p += 4;
+ } else {
+ p++;
}
}
- return 0;
+ rc = 0;
+
+cleanup:
+ kfree(buf);
+ return rc;
}
static int huion_probe(struct hid_device *hdev, const struct hid_device_id *id)
* - change the button usage range to 4-7 for the extra
* buttons
*/
- if (*rsize >= 74 &&
+ if (*rsize >= 75 &&
rdesc[61] == 0x05 && rdesc[62] == 0x08 &&
rdesc[63] == 0x19 && rdesc[64] == 0x08 &&
rdesc[65] == 0x29 && rdesc[66] == 0x0f &&
struct usb_device_descriptor *udesc;
__u16 bcdDevice, rev_maj, rev_min;
- if ((drv_data->quirks & LG_RDESC) && *rsize >= 90 && rdesc[83] == 0x26 &&
+ if ((drv_data->quirks & LG_RDESC) && *rsize >= 91 && rdesc[83] == 0x26 &&
rdesc[84] == 0x8c && rdesc[85] == 0x02) {
hid_info(hdev,
"fixing up Logitech keyboard report descriptor\n");
rdesc[84] = rdesc[89] = 0x4d;
rdesc[85] = rdesc[90] = 0x10;
}
- if ((drv_data->quirks & LG_RDESC_REL_ABS) && *rsize >= 50 &&
+ if ((drv_data->quirks & LG_RDESC_REL_ABS) && *rsize >= 51 &&
rdesc[32] == 0x81 && rdesc[33] == 0x06 &&
rdesc[49] == 0x81 && rdesc[50] == 0x06) {
hid_info(hdev,
drv_data = hid_get_drvdata(hid);
if (!drv_data) {
hid_err(hid, "Private driver data not found!\n");
- return 0;
+ return -EINVAL;
}
entry = drv_data->device_props;
if (!entry) {
hid_err(hid, "Device properties not found!\n");
- return 0;
+ return -EINVAL;
}
if (range == 0)
return;
}
- if ((dj_report->device_index < DJ_DEVICE_INDEX_MIN) ||
- (dj_report->device_index > DJ_DEVICE_INDEX_MAX)) {
- dev_err(&djrcv_hdev->dev, "%s: invalid device index:%d\n",
- __func__, dj_report->device_index);
- return;
- }
-
if (djrcv_dev->paired_dj_devices[dj_report->device_index]) {
/* The device is already known. No need to reallocate it. */
dbg_hid("%s: device is already known\n", __func__);
if (!out_buf)
return -ENOMEM;
- if (count < DJREPORT_SHORT_LENGTH - 2)
+ if (count > DJREPORT_SHORT_LENGTH - 2)
count = DJREPORT_SHORT_LENGTH - 2;
out_buf[0] = REPORT_ID_DJ_SHORT;
* device (via hid_input_report() ) and return 1 so hid-core does not do
* anything else with it.
*/
+ if ((dj_report->device_index < DJ_DEVICE_INDEX_MIN) ||
+ (dj_report->device_index > DJ_DEVICE_INDEX_MAX)) {
+ dev_err(&hdev->dev, "%s: invalid device index:%d\n",
+ __func__, dj_report->device_index);
+ return false;
+ }
spin_lock_irqsave(&djrcv_dev->lock, flags);
if (dj_report->report_id == REPORT_ID_DJ_SHORT) {
static __u8 *mr_report_fixup(struct hid_device *hdev, __u8 *rdesc,
unsigned int *rsize)
{
- if (*rsize >= 30 && rdesc[29] == 0x05 && rdesc[30] == 0x09) {
+ if (*rsize >= 31 && rdesc[29] == 0x05 && rdesc[30] == 0x09) {
hid_info(hdev, "fixing up button/consumer in HID report descriptor\n");
rdesc[30] = 0x0c;
}
static __u8 *pl_report_fixup(struct hid_device *hdev, __u8 *rdesc,
unsigned int *rsize)
{
- if (*rsize >= 60 && rdesc[39] == 0x2a && rdesc[40] == 0xf5 &&
+ if (*rsize >= 62 && rdesc[39] == 0x2a && rdesc[40] == 0xf5 &&
rdesc[41] == 0x00 && rdesc[59] == 0x26 &&
rdesc[60] == 0xf9 && rdesc[61] == 0x00) {
hid_info(hdev, "fixing up Petalynx Maxter Remote report descriptor\n");
return ret;
}
- if (!test_bit(RMI_STARTED, &data->flags)) {
- hid_hw_stop(hdev);
- return -EIO;
- }
+ if (!test_bit(RMI_STARTED, &data->flags))
+ /*
+ * The device maybe in the bootloader if rmi_input_configured
+ * failed to find F11 in the PDT. Print an error, but don't
+ * return an error from rmi_probe so that hidraw will be
+ * accessible from userspace. That way a userspace tool
+ * can be used to reload working firmware on the touchpad.
+ */
+ hid_err(hdev, "Device failed to be properly configured\n");
return 0;
}
ret = -EINVAL;
goto err_stop_hw;
}
- sd->hid_sensor_hub_client_devs = kzalloc(dev_cnt *
- sizeof(struct mfd_cell),
- GFP_KERNEL);
+ sd->hid_sensor_hub_client_devs = devm_kzalloc(&hdev->dev, dev_cnt *
+ sizeof(struct mfd_cell),
+ GFP_KERNEL);
if (sd->hid_sensor_hub_client_devs == NULL) {
hid_err(hdev, "Failed to allocate memory for mfd cells\n");
ret = -ENOMEM;
if (collection->type == HID_COLLECTION_PHYSICAL) {
- hsdev = kzalloc(sizeof(*hsdev), GFP_KERNEL);
+ hsdev = devm_kzalloc(&hdev->dev, sizeof(*hsdev),
+ GFP_KERNEL);
if (!hsdev) {
hid_err(hdev, "cannot allocate hid_sensor_hub_device\n");
ret = -ENOMEM;
- goto err_no_mem;
+ goto err_stop_hw;
}
hsdev->hdev = hdev;
hsdev->vendor_id = hdev->vendor;
if (last_hsdev)
last_hsdev->end_collection_index = i;
last_hsdev = hsdev;
- name = kasprintf(GFP_KERNEL, "HID-SENSOR-%x",
- collection->usage);
+ name = devm_kasprintf(&hdev->dev, GFP_KERNEL,
+ "HID-SENSOR-%x",
+ collection->usage);
if (name == NULL) {
hid_err(hdev, "Failed MFD device name\n");
ret = -ENOMEM;
- kfree(hsdev);
- goto err_no_mem;
+ goto err_stop_hw;
}
sd->hid_sensor_hub_client_devs[
sd->hid_sensor_client_cnt].id =
ret = mfd_add_devices(&hdev->dev, 0, sd->hid_sensor_hub_client_devs,
sd->hid_sensor_client_cnt, NULL, 0, NULL);
if (ret < 0)
- goto err_no_mem;
+ goto err_stop_hw;
return ret;
-err_no_mem:
- for (i = 0; i < sd->hid_sensor_client_cnt; ++i) {
- kfree(sd->hid_sensor_hub_client_devs[i].name);
- kfree(sd->hid_sensor_hub_client_devs[i].platform_data);
- }
- kfree(sd->hid_sensor_hub_client_devs);
err_stop_hw:
hid_hw_stop(hdev);
{
struct sensor_hub_data *data = hid_get_drvdata(hdev);
unsigned long flags;
- int i;
hid_dbg(hdev, " hardware removed\n");
hid_hw_close(hdev);
complete(&data->pending.ready);
spin_unlock_irqrestore(&data->lock, flags);
mfd_remove_devices(&hdev->dev);
- for (i = 0; i < data->hid_sensor_client_cnt; ++i) {
- kfree(data->hid_sensor_hub_client_devs[i].name);
- kfree(data->hid_sensor_hub_client_devs[i].platform_data);
- }
- kfree(data->hid_sensor_hub_client_devs);
hid_set_drvdata(hdev, NULL);
mutex_destroy(&data->mutex);
}
static __u8 *sp_report_fixup(struct hid_device *hdev, __u8 *rdesc,
unsigned int *rsize)
{
- if (*rsize >= 107 && rdesc[104] == 0x26 && rdesc[105] == 0x80 &&
+ if (*rsize >= 112 && rdesc[104] == 0x26 && rdesc[105] == 0x80 &&
rdesc[106] == 0x03) {
hid_info(hdev, "fixing up Sunplus Wireless Desktop report descriptor\n");
rdesc[105] = rdesc[110] = 0x03;
kfree(wacom);
}
+#ifdef CONFIG_PM
static int wacom_resume(struct hid_device *hdev)
{
struct wacom *wacom = hid_get_drvdata(hdev);
{
return wacom_resume(hdev);
}
+#endif /* CONFIG_PM */
static struct hid_driver wacom_driver = {
.name = "wacom",
if (err)
return err;
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
if (err)
return err;
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
dev_err(&client->dev,
"invalid gain on %s\n",
node->full_name);
+ return -EINVAL;
}
}
dev_err(&client->dev,
"invalid data_rate on %s\n",
node->full_name);
+ return -EINVAL;
}
}
err = kstrtoul(buf, 10, &val);
if (err)
return err;
+
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
u8 pwm_acz[3];
u8 pwm_freq[6];
u8 pwm_rr[2];
- u8 zone_low[3];
- u8 zone_abs[3];
+ s8 zone_low[3];
+ s8 zone_abs[3];
u8 zone_hyst[2];
u32 alarms;
};
return (reg * nominal + (3 << (res - 3))) / (3 << (res - 2));
}
-static inline int IN_TO_REG(int val, int nominal)
+static inline int IN_TO_REG(long val, int nominal)
{
return clamp_val((val * 192 + nominal / 2) / nominal, 0, 255);
}
return (reg * 1000) >> (res - 8);
}
-static inline int TEMP_TO_REG(int val)
+static inline int TEMP_TO_REG(long val)
{
return clamp_val((val < 0 ? val - 500 : val + 500) / 1000, -128, 127);
}
return TEMP_RANGE[(reg >> 4) & 0x0f];
}
-static int TEMP_RANGE_TO_REG(int val, int reg)
+static int TEMP_RANGE_TO_REG(long val, int reg)
{
int i;
return (((ix == 1) ? reg : reg >> 4) & 0x0f) * 1000;
}
-static inline int TEMP_HYST_TO_REG(int val, int ix, int reg)
+static inline int TEMP_HYST_TO_REG(long val, int ix, int reg)
{
int hyst = clamp_val((val + 500) / 1000, 0, 15);
return (reg == 0 || reg == 0xffff) ? 0 : 90000 * 60 / reg;
}
-static inline int FAN_TO_REG(int val, int tpc)
+static inline int FAN_TO_REG(long val, int tpc)
{
if (tpc) {
return clamp_val(val / tpc, 0, 0xffff);
return (edge > 0) ? 1 << (edge - 1) : 0;
}
-static inline int FAN_TYPE_TO_REG(int val, int reg)
+static inline int FAN_TYPE_TO_REG(long val, int reg)
{
int edge = (val == 4) ? 3 : val;
return 1000 + i * 500;
}
-static int FAN_MAX_TO_REG(int val)
+static int FAN_MAX_TO_REG(long val)
{
int i;
return acz[(reg >> 5) & 0x07];
}
-static inline int PWM_ACZ_TO_REG(int val, int reg)
+static inline int PWM_ACZ_TO_REG(long val, int reg)
{
int acz = (val == 4) ? 2 : val - 1;
return PWM_FREQ[reg & 0x0f];
}
-static int PWM_FREQ_TO_REG(int val, int reg)
+static int PWM_FREQ_TO_REG(long val, int reg)
{
int i;
return (rr & 0x08) ? PWM_RR[rr & 0x07] : 0;
}
-static int PWM_RR_TO_REG(int val, int ix, int reg)
+static int PWM_RR_TO_REG(long val, int ix, int reg)
{
int i;
return PWM_RR_FROM_REG(reg, ix) ? 1 : 0;
}
-static inline int PWM_RR_EN_TO_REG(int val, int ix, int reg)
+static inline int PWM_RR_EN_TO_REG(long val, int ix, int reg)
{
int en = (ix == 1) ? 0x80 : 0x08;
const char *buf, size_t count)
{
struct dme1737_data *data = dev_get_drvdata(dev);
- long val;
+ unsigned long val;
int err;
- err = kstrtol(buf, 10, &val);
+ err = kstrtoul(buf, 10, &val);
if (err)
return err;
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
if (err < 0)
return err;
- val /= 1000;
+ val = DIV_ROUND_CLOSEST(val, 1000);
reg = (sf == min) ? EMC6W201_REG_TEMP_LOW(nr)
: EMC6W201_REG_TEMP_HIGH(nr);
mutex_lock(&data->update_lock);
- data->temp[sf][nr] = clamp_val(val, -127, 128);
+ data->temp[sf][nr] = clamp_val(val, -127, 127);
err = emc6w201_write8(client, reg, data->temp[sf][nr]);
mutex_unlock(&data->update_lock);
hih6130->client = client;
mutex_init(&hih6130->lock);
+ if (!i2c_check_functionality(client->adapter, I2C_FUNC_SMBUS_QUICK))
+ hih6130->write_length = 1;
+
hwmon_dev = devm_hwmon_device_register_with_groups(dev, client->name,
hih6130,
hih6130_groups);
err = kstrtoul(buf, 10, &val);
if (err)
return err;
+
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
return reg / 8 * 625 / 10;
}
-static inline s16 TEMP_TO_REG(int val)
+static inline s16 TEMP_TO_REG(long val)
{
- if (val <= -60000)
- return -60000 * 10 / 625 * 8;
- if (val >= 160000)
- return 160000 * 10 / 625 * 8;
+ val = clamp_val(val, -60000, 160000);
return val * 10 / 625 * 8;
}
if (err)
return err;
+ val = clamp_val(val, -120000, 220000);
mutex_lock(&data->update_lock);
- data->temp[t_hyst] = TEMP_FROM_REG(data->temp[attr->index]) - val;
+ data->temp[t_hyst] =
+ TEMP_TO_REG(TEMP_FROM_REG(data->temp[attr->index]) - val);
i2c_smbus_write_word_swapped(client, LM92_REG_TEMP_HYST,
- TEMP_TO_REG(data->temp[t_hyst]));
+ data->temp[t_hyst]);
mutex_unlock(&data->update_lock);
return count;
}
if (err)
return err;
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
struct regmap *regmap;
int ret;
- if (!i2c_check_functionality(client->adapter,
- I2C_FUNC_SMBUS_BYTE_DATA)) {
- dev_err(&client->dev,
- "adapter doesn't support SMBus byte transactions\n");
- return -ENODEV;
- }
-
regmap = devm_regmap_init_i2c(client, &tmp103_regmap_config);
if (IS_ERR(regmap)) {
dev_err(dev, "failed to allocate register map\n");
if (err)
return err;
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
err = kstrtoul(buf, 10, &val);
if (err)
return err;
+
+ if (val > 255)
+ return -EINVAL;
data->vrm = val;
return count;
if (err)
return err;
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
if (err)
return err;
+ if (val > 255)
+ return -EINVAL;
+
data->vrm = val;
return count;
}
config HWSPINLOCK_OMAP
tristate "OMAP Hardware Spinlock device"
- depends on ARCH_OMAP4 || SOC_OMAP5
+ depends on ARCH_OMAP4 || SOC_OMAP5 || SOC_DRA7XX || SOC_AM33XX || SOC_AM43XX
select HWSPINLOCK
help
Say y here to support the OMAP Hardware Spinlock device (firstly
if (!io_base)
return -ENOMEM;
+ /*
+ * make sure the module is enabled and clocked before reading
+ * the module SYSSTATUS register
+ */
+ pm_runtime_enable(&pdev->dev);
+ ret = pm_runtime_get_sync(&pdev->dev);
+ if (ret < 0) {
+ pm_runtime_put_noidle(&pdev->dev);
+ goto iounmap_base;
+ }
+
/* Determine number of locks */
i = readl(io_base + SYSSTATUS_OFFSET);
i >>= SPINLOCK_NUMLOCKS_BIT_OFFSET;
+ /*
+ * runtime PM will make sure the clock of this module is
+ * enabled again iff at least one lock is requested
+ */
+ ret = pm_runtime_put(&pdev->dev);
+ if (ret < 0)
+ goto iounmap_base;
+
/* one of the four lsb's must be set, and nothing else */
if (hweight_long(i & 0xf) != 1 || i > 8) {
ret = -EINVAL;
for (i = 0, hwlock = &bank->lock[0]; i < num_locks; i++, hwlock++)
hwlock->priv = io_base + LOCK_BASE_OFFSET + sizeof(u32) * i;
- /*
- * runtime PM will make sure the clock of this module is
- * enabled iff at least one lock is requested
- */
- pm_runtime_enable(&pdev->dev);
-
ret = hwspin_lock_register(bank, &pdev->dev, &omap_hwspinlock_ops,
pdata->base_id, num_locks);
if (ret)
return 0;
reg_fail:
- pm_runtime_disable(&pdev->dev);
kfree(bank);
iounmap_base:
+ pm_runtime_disable(&pdev->dev);
iounmap(io_base);
return ret;
}
This I2C support can also be built as a module. If so, the module
will be called i2c-core.
-config I2C_ACPI
- bool "I2C ACPI support"
- select I2C
- depends on ACPI
+config ACPI_I2C_OPREGION
+ bool "ACPI I2C Operation region support"
+ depends on I2C=y && ACPI
default y
help
- Say Y here if you want to enable ACPI I2C support. This includes support
- for automatic enumeration of I2C slave devices and support for ACPI I2C
- Operation Regions. Operation Regions allow firmware (BIOS) code to
- access I2C slave devices, such as smart batteries through an I2C host
- controller driver.
+ Say Y here if you want to enable ACPI I2C operation region support.
+ Operation Regions allow firmware (BIOS) code to access I2C slave devices,
+ such as smart batteries through an I2C host controller driver.
if I2C
#
i2ccore-y := i2c-core.o
-i2ccore-$(CONFIG_I2C_ACPI) += i2c-acpi.o
+i2ccore-$(CONFIG_ACPI) += i2c-acpi.o
obj-$(CONFIG_I2C_BOARDINFO) += i2c-boardinfo.o
obj-$(CONFIG_I2C) += i2ccore.o
/* Older devices have their ID defined in <linux/pci_ids.h> */
#define PCI_DEVICE_ID_INTEL_BAYTRAIL_SMBUS 0x0f12
+#define PCI_DEVICE_ID_INTEL_BRASWELL_SMBUS 0x2292
#define PCI_DEVICE_ID_INTEL_COUGARPOINT_SMBUS 0x1c22
#define PCI_DEVICE_ID_INTEL_PATSBURG_SMBUS 0x1d22
/* Patsburg also has three 'Integrated Device Function' SMBus controllers */
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WILDCATPOINT_SMBUS) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_WILDCATPOINT_LP_SMBUS) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_BAYTRAIL_SMBUS) },
+ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_BRASWELL_SMBUS) },
{ 0, }
};
dev_warn(&adap->dev, "failed to enumerate I2C slaves\n");
}
+#ifdef CONFIG_ACPI_I2C_OPREGION
static int acpi_gsb_i2c_read_bytes(struct i2c_client *client,
u8 cmd, u8 *data, u8 data_len)
{
acpi_bus_detach_private_data(handle);
}
+#endif
* Indicate which enable bits to clear here.
*/
unsigned long auto_demotion_disable_flags;
+ bool byt_auto_demotion_disable_flag;
bool disable_promotion_to_c1e;
};
{
.enter = NULL }
};
+static struct cpuidle_state bdw_cstates[] = {
+ {
+ .name = "C1-BDW",
+ .desc = "MWAIT 0x00",
+ .flags = MWAIT2flg(0x00) | CPUIDLE_FLAG_TIME_VALID,
+ .exit_latency = 2,
+ .target_residency = 2,
+ .enter = &intel_idle },
+ {
+ .name = "C1E-BDW",
+ .desc = "MWAIT 0x01",
+ .flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_TIME_VALID,
+ .exit_latency = 10,
+ .target_residency = 20,
+ .enter = &intel_idle },
+ {
+ .name = "C3-BDW",
+ .desc = "MWAIT 0x10",
+ .flags = MWAIT2flg(0x10) | CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
+ .exit_latency = 40,
+ .target_residency = 100,
+ .enter = &intel_idle },
+ {
+ .name = "C6-BDW",
+ .desc = "MWAIT 0x20",
+ .flags = MWAIT2flg(0x20) | CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
+ .exit_latency = 133,
+ .target_residency = 400,
+ .enter = &intel_idle },
+ {
+ .name = "C7s-BDW",
+ .desc = "MWAIT 0x32",
+ .flags = MWAIT2flg(0x32) | CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
+ .exit_latency = 166,
+ .target_residency = 500,
+ .enter = &intel_idle },
+ {
+ .name = "C8-BDW",
+ .desc = "MWAIT 0x40",
+ .flags = MWAIT2flg(0x40) | CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
+ .exit_latency = 300,
+ .target_residency = 900,
+ .enter = &intel_idle },
+ {
+ .name = "C9-BDW",
+ .desc = "MWAIT 0x50",
+ .flags = MWAIT2flg(0x50) | CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
+ .exit_latency = 600,
+ .target_residency = 1800,
+ .enter = &intel_idle },
+ {
+ .name = "C10-BDW",
+ .desc = "MWAIT 0x60",
+ .flags = MWAIT2flg(0x60) | CPUIDLE_FLAG_TIME_VALID | CPUIDLE_FLAG_TLB_FLUSHED,
+ .exit_latency = 2600,
+ .target_residency = 7700,
+ .enter = &intel_idle },
+ {
+ .enter = NULL }
+};
static struct cpuidle_state atom_cstates[] = {
{
static const struct idle_cpu idle_cpu_byt = {
.state_table = byt_cstates,
.disable_promotion_to_c1e = true,
+ .byt_auto_demotion_disable_flag = true,
};
static const struct idle_cpu idle_cpu_ivb = {
.disable_promotion_to_c1e = true,
};
+static const struct idle_cpu idle_cpu_bdw = {
+ .state_table = bdw_cstates,
+ .disable_promotion_to_c1e = true,
+};
+
static const struct idle_cpu idle_cpu_avn = {
.state_table = avn_cstates,
.disable_promotion_to_c1e = true,
ICPU(0x3f, idle_cpu_hsw),
ICPU(0x45, idle_cpu_hsw),
ICPU(0x46, idle_cpu_hsw),
- ICPU(0x4D, idle_cpu_avn),
+ ICPU(0x4d, idle_cpu_avn),
+ ICPU(0x3d, idle_cpu_bdw),
+ ICPU(0x4f, idle_cpu_bdw),
+ ICPU(0x56, idle_cpu_bdw),
{}
};
MODULE_DEVICE_TABLE(x86cpu, intel_idle_ids);
if (icpu->auto_demotion_disable_flags)
on_each_cpu(auto_demotion_disable, NULL, 1);
+ if (icpu->byt_auto_demotion_disable_flag) {
+ wrmsrl(MSR_CC6_DEMOTION_POLICY_CONFIG, 0);
+ wrmsrl(MSR_MC6_DEMOTION_POLICY_CONFIG, 0);
+ }
+
if (icpu->disable_promotion_to_c1e) /* each-cpu is redundant */
on_each_cpu(c1e_promotion_disable, NULL, 1);
port_priv = ib_get_agent_port(device, port_num);
if (!port_priv) {
- printk(KERN_ERR SPFX "Unable to find port agent\n");
+ dev_err(&device->dev, "Unable to find port agent\n");
return;
}
agent = port_priv->agent[qpn];
ah = ib_create_ah_from_wc(agent->qp->pd, wc, grh, port_num);
if (IS_ERR(ah)) {
- printk(KERN_ERR SPFX "ib_create_ah_from_wc error %ld\n",
+ dev_err(&device->dev, "ib_create_ah_from_wc error %ld\n",
PTR_ERR(ah));
return;
}
IB_MGMT_MAD_HDR, IB_MGMT_MAD_DATA,
GFP_KERNEL);
if (IS_ERR(send_buf)) {
- printk(KERN_ERR SPFX "ib_create_send_mad error\n");
+ dev_err(&device->dev, "ib_create_send_mad error\n");
goto err1;
}
}
if (ib_post_send_mad(send_buf, NULL)) {
- printk(KERN_ERR SPFX "ib_post_send_mad error\n");
+ dev_err(&device->dev, "ib_post_send_mad error\n");
goto err2;
}
return;
/* Create new device info */
port_priv = kzalloc(sizeof *port_priv, GFP_KERNEL);
if (!port_priv) {
- printk(KERN_ERR SPFX "No memory for ib_agent_port_private\n");
+ dev_err(&device->dev, "No memory for ib_agent_port_private\n");
ret = -ENOMEM;
goto error1;
}
port_priv->agent[0] = ib_register_mad_agent(device, port_num,
IB_QPT_SMI, NULL, 0,
&agent_send_handler,
- NULL, NULL);
+ NULL, NULL, 0);
if (IS_ERR(port_priv->agent[0])) {
ret = PTR_ERR(port_priv->agent[0]);
goto error2;
port_priv->agent[1] = ib_register_mad_agent(device, port_num,
IB_QPT_GSI, NULL, 0,
&agent_send_handler,
- NULL, NULL);
+ NULL, NULL, 0);
if (IS_ERR(port_priv->agent[1])) {
ret = PTR_ERR(port_priv->agent[1]);
goto error3;
port_priv = __ib_get_agent_port(device, port_num);
if (port_priv == NULL) {
spin_unlock_irqrestore(&ib_agent_port_list_lock, flags);
- printk(KERN_ERR SPFX "Port %d not found\n", port_num);
+ dev_err(&device->dev, "Port %d not found\n", port_num);
return -ENODEV;
}
list_del(&port_priv->port_list);
struct cm_port *port;
struct ib_mad_reg_req reg_req = {
.mgmt_class = IB_MGMT_CLASS_CM,
- .mgmt_class_version = IB_CM_CLASS_VERSION
+ .mgmt_class_version = IB_CM_CLASS_VERSION,
};
struct ib_port_modify port_modify = {
.set_port_cap_mask = IB_PORT_CM_SUP
0,
cm_send_handler,
cm_recv_handler,
- port);
+ port,
+ 0);
if (IS_ERR(port->mad_agent))
goto error2;
#include <linux/completion.h>
#include <linux/slab.h>
#include <linux/module.h>
+#include <linux/sysctl.h>
#include <rdma/iw_cm.h>
#include <rdma/ib_addr.h>
struct list_head free_list;
};
+static unsigned int default_backlog = 256;
+
+static struct ctl_table_header *iwcm_ctl_table_hdr;
+static struct ctl_table iwcm_ctl_table[] = {
+ {
+ .procname = "default_backlog",
+ .data = &default_backlog,
+ .maxlen = sizeof(default_backlog),
+ .mode = 0644,
+ .proc_handler = proc_dointvec,
+ },
+ { }
+};
+
/*
* The following services provide a mechanism for pre-allocating iwcm_work
* elements. The design pre-allocates them based on the cm_id type:
cm_id_priv = container_of(cm_id, struct iwcm_id_private, id);
+ if (!backlog)
+ backlog = default_backlog;
+
ret = alloc_work_entries(cm_id_priv, backlog);
if (ret)
return ret;
if (!iwcm_wq)
return -ENOMEM;
+ iwcm_ctl_table_hdr = register_net_sysctl(&init_net, "net/iw_cm",
+ iwcm_ctl_table);
+ if (!iwcm_ctl_table_hdr) {
+ pr_err("iw_cm: couldn't register sysctl paths\n");
+ destroy_workqueue(iwcm_wq);
+ return -ENOMEM;
+ }
+
return 0;
}
static void __exit iw_cm_cleanup(void)
{
+ unregister_net_sysctl_table(iwcm_ctl_table_hdr);
destroy_workqueue(iwcm_wq);
}
* SOFTWARE.
*
*/
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
#include <linux/dma-mapping.h>
#include <linux/slab.h>
#include <linux/module.h>
u8 rmpp_version,
ib_mad_send_handler send_handler,
ib_mad_recv_handler recv_handler,
- void *context)
+ void *context,
+ u32 registration_flags)
{
struct ib_mad_port_private *port_priv;
struct ib_mad_agent *ret = ERR_PTR(-EINVAL);
/* Validate parameters */
qpn = get_spl_qp_index(qp_type);
- if (qpn == -1)
+ if (qpn == -1) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: invalid QP Type %d\n",
+ qp_type);
goto error1;
+ }
- if (rmpp_version && rmpp_version != IB_MGMT_RMPP_VERSION)
+ if (rmpp_version && rmpp_version != IB_MGMT_RMPP_VERSION) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: invalid RMPP Version %u\n",
+ rmpp_version);
goto error1;
+ }
/* Validate MAD registration request if supplied */
if (mad_reg_req) {
- if (mad_reg_req->mgmt_class_version >= MAX_MGMT_VERSION)
+ if (mad_reg_req->mgmt_class_version >= MAX_MGMT_VERSION) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: invalid Class Version %u\n",
+ mad_reg_req->mgmt_class_version);
goto error1;
- if (!recv_handler)
+ }
+ if (!recv_handler) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: no recv_handler\n");
goto error1;
+ }
if (mad_reg_req->mgmt_class >= MAX_MGMT_CLASS) {
/*
* IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE is the only
* one in this range currently allowed
*/
if (mad_reg_req->mgmt_class !=
- IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)
+ IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: Invalid Mgmt Class 0x%x\n",
+ mad_reg_req->mgmt_class);
goto error1;
+ }
} else if (mad_reg_req->mgmt_class == 0) {
/*
* Class 0 is reserved in IBA and is used for
* aliasing of IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE
*/
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: Invalid Mgmt Class 0\n");
goto error1;
} else if (is_vendor_class(mad_reg_req->mgmt_class)) {
/*
* If class is in "new" vendor range,
* ensure supplied OUI is not zero
*/
- if (!is_vendor_oui(mad_reg_req->oui))
+ if (!is_vendor_oui(mad_reg_req->oui)) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: No OUI specified for class 0x%x\n",
+ mad_reg_req->mgmt_class);
goto error1;
+ }
}
/* Make sure class supplied is consistent with RMPP */
if (!ib_is_mad_class_rmpp(mad_reg_req->mgmt_class)) {
- if (rmpp_version)
+ if (rmpp_version) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: RMPP version for non-RMPP class 0x%x\n",
+ mad_reg_req->mgmt_class);
goto error1;
+ }
}
+
/* Make sure class supplied is consistent with QP type */
if (qp_type == IB_QPT_SMI) {
if ((mad_reg_req->mgmt_class !=
IB_MGMT_CLASS_SUBN_LID_ROUTED) &&
(mad_reg_req->mgmt_class !=
- IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE))
+ IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: Invalid SM QP type: class 0x%x\n",
+ mad_reg_req->mgmt_class);
goto error1;
+ }
} else {
if ((mad_reg_req->mgmt_class ==
IB_MGMT_CLASS_SUBN_LID_ROUTED) ||
(mad_reg_req->mgmt_class ==
- IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE))
+ IB_MGMT_CLASS_SUBN_DIRECTED_ROUTE)) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: Invalid GS QP type: class 0x%x\n",
+ mad_reg_req->mgmt_class);
goto error1;
+ }
}
} else {
/* No registration request supplied */
if (!send_handler)
goto error1;
+ if (registration_flags & IB_MAD_USER_RMPP)
+ goto error1;
}
/* Validate device and port */
port_priv = ib_get_mad_port(device, port_num);
if (!port_priv) {
+ dev_notice(&device->dev, "ib_register_mad_agent: Invalid port\n");
ret = ERR_PTR(-ENODEV);
goto error1;
}
/* Verify the QP requested is supported. For example, Ethernet devices
* will not have QP0 */
if (!port_priv->qp_info[qpn].qp) {
+ dev_notice(&device->dev,
+ "ib_register_mad_agent: QP %d not supported\n", qpn);
ret = ERR_PTR(-EPROTONOSUPPORT);
goto error1;
}
mad_agent_priv->agent.context = context;
mad_agent_priv->agent.qp = port_priv->qp_info[qpn].qp;
mad_agent_priv->agent.port_num = port_num;
+ mad_agent_priv->agent.flags = registration_flags;
spin_lock_init(&mad_agent_priv->lock);
INIT_LIST_HEAD(&mad_agent_priv->send_list);
INIT_LIST_HEAD(&mad_agent_priv->wait_list);
smi_handle_dr_smp_send(smp, device->node_type, port_num) ==
IB_SMI_DISCARD) {
ret = -EINVAL;
- printk(KERN_ERR PFX "Invalid directed route\n");
+ dev_err(&device->dev, "Invalid directed route\n");
goto out;
}
local = kmalloc(sizeof *local, GFP_ATOMIC);
if (!local) {
ret = -ENOMEM;
- printk(KERN_ERR PFX "No memory for ib_mad_local_private\n");
+ dev_err(&device->dev, "No memory for ib_mad_local_private\n");
goto out;
}
local->mad_priv = NULL;
mad_priv = kmem_cache_alloc(ib_mad_cache, GFP_ATOMIC);
if (!mad_priv) {
ret = -ENOMEM;
- printk(KERN_ERR PFX "No memory for local response MAD\n");
+ dev_err(&device->dev, "No memory for local response MAD\n");
kfree(local);
goto out;
}
for (left = send_buf->data_len + pad; left > 0; left -= seg_size) {
seg = kmalloc(sizeof (*seg) + seg_size, gfp_mask);
if (!seg) {
- printk(KERN_ERR "alloc_send_rmpp_segs: RMPP mem "
- "alloc failed for len %zd, gfp %#x\n",
- sizeof (*seg) + seg_size, gfp_mask);
+ dev_err(&send_buf->mad_agent->device->dev,
+ "alloc_send_rmpp_segs: RMPP mem alloc failed for len %zd, gfp %#x\n",
+ sizeof (*seg) + seg_size, gfp_mask);
free_send_rmpp_list(send_wr);
return -ENOMEM;
}
return 0;
}
+int ib_mad_kernel_rmpp_agent(struct ib_mad_agent *agent)
+{
+ return agent->rmpp_version && !(agent->flags & IB_MAD_USER_RMPP);
+}
+EXPORT_SYMBOL(ib_mad_kernel_rmpp_agent);
+
struct ib_mad_send_buf * ib_create_send_mad(struct ib_mad_agent *mad_agent,
u32 remote_qpn, u16 pkey_index,
int rmpp_active,
pad = get_pad_size(hdr_len, data_len);
message_size = hdr_len + data_len + pad;
- if ((!mad_agent->rmpp_version &&
- (rmpp_active || message_size > sizeof(struct ib_mad))) ||
- (!rmpp_active && message_size > sizeof(struct ib_mad)))
- return ERR_PTR(-EINVAL);
+ if (ib_mad_kernel_rmpp_agent(mad_agent)) {
+ if (!rmpp_active && message_size > sizeof(struct ib_mad))
+ return ERR_PTR(-EINVAL);
+ } else
+ if (rmpp_active || message_size > sizeof(struct ib_mad))
+ return ERR_PTR(-EINVAL);
size = rmpp_active ? hdr_len : sizeof(struct ib_mad);
buf = kzalloc(sizeof *mad_send_wr + size, gfp_mask);
&mad_agent_priv->send_list);
spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
- if (mad_agent_priv->agent.rmpp_version) {
+ if (ib_mad_kernel_rmpp_agent(&mad_agent_priv->agent)) {
ret = ib_send_rmpp_mad(mad_send_wr);
if (ret >= 0 && ret != IB_RMPP_RESULT_CONSUMED)
ret = ib_send_mad(mad_send_wr);
int ib_process_mad_wc(struct ib_mad_agent *mad_agent,
struct ib_wc *wc)
{
- printk(KERN_ERR PFX "ib_process_mad_wc() not implemented yet\n");
+ dev_err(&mad_agent->device->dev,
+ "ib_process_mad_wc() not implemented yet\n");
return 0;
}
EXPORT_SYMBOL(ib_process_mad_wc);
for_each_set_bit(i, mad_reg_req->method_mask, IB_MGMT_MAX_METHODS) {
if ((*method)->agent[i]) {
- printk(KERN_ERR PFX "Method %d already in use\n", i);
+ pr_err("Method %d already in use\n", i);
return -EINVAL;
}
}
/* Allocate management method table */
*method = kzalloc(sizeof **method, GFP_ATOMIC);
if (!*method) {
- printk(KERN_ERR PFX "No memory for "
- "ib_mad_mgmt_method_table\n");
+ pr_err("No memory for ib_mad_mgmt_method_table\n");
return -ENOMEM;
}
/* Allocate management class table for "new" class version */
*class = kzalloc(sizeof **class, GFP_ATOMIC);
if (!*class) {
- printk(KERN_ERR PFX "No memory for "
- "ib_mad_mgmt_class_table\n");
+ dev_err(&agent_priv->agent.device->dev,
+ "No memory for ib_mad_mgmt_class_table\n");
ret = -ENOMEM;
goto error1;
}
/* Allocate mgmt vendor class table for "new" class version */
vendor = kzalloc(sizeof *vendor, GFP_ATOMIC);
if (!vendor) {
- printk(KERN_ERR PFX "No memory for "
- "ib_mad_mgmt_vendor_class_table\n");
+ dev_err(&agent_priv->agent.device->dev,
+ "No memory for ib_mad_mgmt_vendor_class_table\n");
goto error1;
}
/* Allocate table for this management vendor class */
vendor_class = kzalloc(sizeof *vendor_class, GFP_ATOMIC);
if (!vendor_class) {
- printk(KERN_ERR PFX "No memory for "
- "ib_mad_mgmt_vendor_class\n");
+ dev_err(&agent_priv->agent.device->dev,
+ "No memory for ib_mad_mgmt_vendor_class\n");
goto error2;
}
goto check_in_use;
}
}
- printk(KERN_ERR PFX "All OUI slots in use\n");
+ dev_err(&agent_priv->agent.device->dev, "All OUI slots in use\n");
goto error3;
check_in_use:
if (mad_agent->agent.recv_handler)
atomic_inc(&mad_agent->refcount);
else {
- printk(KERN_NOTICE PFX "No receive handler for client "
- "%p on port %d\n",
- &mad_agent->agent, port_priv->port_num);
+ dev_notice(&port_priv->device->dev,
+ "No receive handler for client %p on port %d\n",
+ &mad_agent->agent, port_priv->port_num);
mad_agent = NULL;
}
}
/* Make sure MAD base version is understood */
if (mad->mad_hdr.base_version != IB_MGMT_BASE_VERSION) {
- printk(KERN_ERR PFX "MAD received with unsupported base "
- "version %d\n", mad->mad_hdr.base_version);
+ pr_err("MAD received with unsupported base version %d\n",
+ mad->mad_hdr.base_version);
goto out;
}
rmpp_mad = (struct ib_rmpp_mad *)mad_hdr;
return !mad_agent_priv->agent.rmpp_version ||
+ !ib_mad_kernel_rmpp_agent(&mad_agent_priv->agent) ||
!(ib_get_rmpp_flags(&rmpp_mad->rmpp_hdr) &
IB_MGMT_RMPP_FLAG_ACTIVE) ||
(rmpp_mad->rmpp_hdr.rmpp_type == IB_MGMT_RMPP_TYPE_DATA);
INIT_LIST_HEAD(&mad_recv_wc->rmpp_list);
list_add(&mad_recv_wc->recv_buf.list, &mad_recv_wc->rmpp_list);
- if (mad_agent_priv->agent.rmpp_version) {
+ if (ib_mad_kernel_rmpp_agent(&mad_agent_priv->agent)) {
mad_recv_wc = ib_process_rmpp_recv_wc(mad_agent_priv,
mad_recv_wc);
if (!mad_recv_wc) {
mad_send_wr = ib_find_send_mad(mad_agent_priv, mad_recv_wc);
if (!mad_send_wr) {
spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
- ib_free_recv_mad(mad_recv_wc);
- deref_mad_agent(mad_agent_priv);
- return;
- }
- ib_mark_mad_done(mad_send_wr);
- spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
+ if (!ib_mad_kernel_rmpp_agent(&mad_agent_priv->agent)
+ && ib_is_mad_class_rmpp(mad_recv_wc->recv_buf.mad->mad_hdr.mgmt_class)
+ && (ib_get_rmpp_flags(&((struct ib_rmpp_mad *)mad_recv_wc->recv_buf.mad)->rmpp_hdr)
+ & IB_MGMT_RMPP_FLAG_ACTIVE)) {
+ /* user rmpp is in effect
+ * and this is an active RMPP MAD
+ */
+ mad_recv_wc->wc->wr_id = 0;
+ mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent,
+ mad_recv_wc);
+ atomic_dec(&mad_agent_priv->refcount);
+ } else {
+ /* not user rmpp, revert to normal behavior and
+ * drop the mad */
+ ib_free_recv_mad(mad_recv_wc);
+ deref_mad_agent(mad_agent_priv);
+ return;
+ }
+ } else {
+ ib_mark_mad_done(mad_send_wr);
+ spin_unlock_irqrestore(&mad_agent_priv->lock, flags);
- /* Defined behavior is to complete response before request */
- mad_recv_wc->wc->wr_id = (unsigned long) &mad_send_wr->send_buf;
- mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent,
- mad_recv_wc);
- atomic_dec(&mad_agent_priv->refcount);
+ /* Defined behavior is to complete response before request */
+ mad_recv_wc->wc->wr_id = (unsigned long) &mad_send_wr->send_buf;
+ mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent,
+ mad_recv_wc);
+ atomic_dec(&mad_agent_priv->refcount);
- mad_send_wc.status = IB_WC_SUCCESS;
- mad_send_wc.vendor_err = 0;
- mad_send_wc.send_buf = &mad_send_wr->send_buf;
- ib_mad_complete_send_wr(mad_send_wr, &mad_send_wc);
+ mad_send_wc.status = IB_WC_SUCCESS;
+ mad_send_wc.vendor_err = 0;
+ mad_send_wc.send_buf = &mad_send_wr->send_buf;
+ ib_mad_complete_send_wr(mad_send_wr, &mad_send_wc);
+ }
} else {
mad_agent_priv->agent.recv_handler(&mad_agent_priv->agent,
mad_recv_wc);
response = kmem_cache_alloc(ib_mad_cache, GFP_KERNEL);
if (!response) {
- printk(KERN_ERR PFX "ib_mad_recv_done_handler no memory "
- "for response buffer\n");
+ dev_err(&port_priv->device->dev,
+ "ib_mad_recv_done_handler no memory for response buffer\n");
goto out;
}
mad_agent_priv = mad_send_wr->mad_agent_priv;
spin_lock_irqsave(&mad_agent_priv->lock, flags);
- if (mad_agent_priv->agent.rmpp_version) {
+ if (ib_mad_kernel_rmpp_agent(&mad_agent_priv->agent)) {
ret = ib_process_rmpp_send_wc(mad_send_wr, mad_send_wc);
if (ret == IB_RMPP_RESULT_CONSUMED)
goto done;
ret = ib_post_send(qp_info->qp, &queued_send_wr->send_wr,
&bad_send_wr);
if (ret) {
- printk(KERN_ERR PFX "ib_post_send failed: %d\n", ret);
+ dev_err(&port_priv->device->dev,
+ "ib_post_send failed: %d\n", ret);
mad_send_wr = queued_send_wr;
wc->status = IB_WC_LOC_QP_OP_ERR;
goto retry;
IB_QP_STATE | IB_QP_CUR_STATE);
kfree(attr);
if (ret)
- printk(KERN_ERR PFX "mad_error_handler - "
- "ib_modify_qp to RTS : %d\n", ret);
+ dev_err(&port_priv->device->dev,
+ "mad_error_handler - ib_modify_qp to RTS : %d\n",
+ ret);
else
mark_sends_for_retry(qp_info);
}
if (local->mad_priv) {
recv_mad_agent = local->recv_mad_agent;
if (!recv_mad_agent) {
- printk(KERN_ERR PFX "No receive MAD agent for local completion\n");
+ dev_err(&mad_agent_priv->agent.device->dev,
+ "No receive MAD agent for local completion\n");
free_mad = 1;
goto local_send_completion;
}
mad_send_wr->timeout = msecs_to_jiffies(mad_send_wr->send_buf.timeout_ms);
- if (mad_send_wr->mad_agent_priv->agent.rmpp_version) {
+ if (ib_mad_kernel_rmpp_agent(&mad_send_wr->mad_agent_priv->agent)) {
ret = ib_retry_rmpp(mad_send_wr);
switch (ret) {
case IB_RMPP_RESULT_UNHANDLED:
} else {
mad_priv = kmem_cache_alloc(ib_mad_cache, GFP_KERNEL);
if (!mad_priv) {
- printk(KERN_ERR PFX "No memory for receive buffer\n");
+ dev_err(&qp_info->port_priv->device->dev,
+ "No memory for receive buffer\n");
ret = -ENOMEM;
break;
}
sizeof mad_priv->header,
DMA_FROM_DEVICE);
kmem_cache_free(ib_mad_cache, mad_priv);
- printk(KERN_ERR PFX "ib_post_recv failed: %d\n", ret);
+ dev_err(&qp_info->port_priv->device->dev,
+ "ib_post_recv failed: %d\n", ret);
break;
}
} while (post);
attr = kmalloc(sizeof *attr, GFP_KERNEL);
if (!attr) {
- printk(KERN_ERR PFX "Couldn't kmalloc ib_qp_attr\n");
+ dev_err(&port_priv->device->dev,
+ "Couldn't kmalloc ib_qp_attr\n");
return -ENOMEM;
}
ret = ib_modify_qp(qp, attr, IB_QP_STATE |
IB_QP_PKEY_INDEX | IB_QP_QKEY);
if (ret) {
- printk(KERN_ERR PFX "Couldn't change QP%d state to "
- "INIT: %d\n", i, ret);
+ dev_err(&port_priv->device->dev,
+ "Couldn't change QP%d state to INIT: %d\n",
+ i, ret);
goto out;
}
attr->qp_state = IB_QPS_RTR;
ret = ib_modify_qp(qp, attr, IB_QP_STATE);
if (ret) {
- printk(KERN_ERR PFX "Couldn't change QP%d state to "
- "RTR: %d\n", i, ret);
+ dev_err(&port_priv->device->dev,
+ "Couldn't change QP%d state to RTR: %d\n",
+ i, ret);
goto out;
}
attr->sq_psn = IB_MAD_SEND_Q_PSN;
ret = ib_modify_qp(qp, attr, IB_QP_STATE | IB_QP_SQ_PSN);
if (ret) {
- printk(KERN_ERR PFX "Couldn't change QP%d state to "
- "RTS: %d\n", i, ret);
+ dev_err(&port_priv->device->dev,
+ "Couldn't change QP%d state to RTS: %d\n",
+ i, ret);
goto out;
}
}
ret = ib_req_notify_cq(port_priv->cq, IB_CQ_NEXT_COMP);
if (ret) {
- printk(KERN_ERR PFX "Failed to request completion "
- "notification: %d\n", ret);
+ dev_err(&port_priv->device->dev,
+ "Failed to request completion notification: %d\n",
+ ret);
goto out;
}
ret = ib_mad_post_receive_mads(&port_priv->qp_info[i], NULL);
if (ret) {
- printk(KERN_ERR PFX "Couldn't post receive WRs\n");
+ dev_err(&port_priv->device->dev,
+ "Couldn't post receive WRs\n");
goto out;
}
}
struct ib_mad_qp_info *qp_info = qp_context;
/* It's worse than that! He's dead, Jim! */
- printk(KERN_ERR PFX "Fatal error (%d) on MAD QP (%d)\n",
+ dev_err(&qp_info->port_priv->device->dev,
+ "Fatal error (%d) on MAD QP (%d)\n",
event->event, qp_info->qp->qp_num);
}
qp_init_attr.event_handler = qp_event_handler;
qp_info->qp = ib_create_qp(qp_info->port_priv->pd, &qp_init_attr);
if (IS_ERR(qp_info->qp)) {
- printk(KERN_ERR PFX "Couldn't create ib_mad QP%d\n",
- get_spl_qp_index(qp_type));
+ dev_err(&qp_info->port_priv->device->dev,
+ "Couldn't create ib_mad QP%d\n",
+ get_spl_qp_index(qp_type));
ret = PTR_ERR(qp_info->qp);
goto error;
}
/* Create new device info */
port_priv = kzalloc(sizeof *port_priv, GFP_KERNEL);
if (!port_priv) {
- printk(KERN_ERR PFX "No memory for ib_mad_port_private\n");
+ dev_err(&device->dev, "No memory for ib_mad_port_private\n");
return -ENOMEM;
}
ib_mad_thread_completion_handler,
NULL, port_priv, cq_size, 0);
if (IS_ERR(port_priv->cq)) {
- printk(KERN_ERR PFX "Couldn't create ib_mad CQ\n");
+ dev_err(&device->dev, "Couldn't create ib_mad CQ\n");
ret = PTR_ERR(port_priv->cq);
goto error3;
}
port_priv->pd = ib_alloc_pd(device);
if (IS_ERR(port_priv->pd)) {
- printk(KERN_ERR PFX "Couldn't create ib_mad PD\n");
+ dev_err(&device->dev, "Couldn't create ib_mad PD\n");
ret = PTR_ERR(port_priv->pd);
goto error4;
}
port_priv->mr = ib_get_dma_mr(port_priv->pd, IB_ACCESS_LOCAL_WRITE);
if (IS_ERR(port_priv->mr)) {
- printk(KERN_ERR PFX "Couldn't get ib_mad DMA MR\n");
+ dev_err(&device->dev, "Couldn't get ib_mad DMA MR\n");
ret = PTR_ERR(port_priv->mr);
goto error5;
}
ret = ib_mad_port_start(port_priv);
if (ret) {
- printk(KERN_ERR PFX "Couldn't start port\n");
+ dev_err(&device->dev, "Couldn't start port\n");
goto error9;
}
port_priv = __ib_get_mad_port(device, port_num);
if (port_priv == NULL) {
spin_unlock_irqrestore(&ib_mad_port_list_lock, flags);
- printk(KERN_ERR PFX "Port %d not found\n", port_num);
+ dev_err(&device->dev, "Port %d not found\n", port_num);
return -ENODEV;
}
list_del_init(&port_priv->port_list);
for (i = start; i <= end; i++) {
if (ib_mad_port_open(device, i)) {
- printk(KERN_ERR PFX "Couldn't open %s port %d\n",
- device->name, i);
+ dev_err(&device->dev, "Couldn't open port %d\n", i);
goto error;
}
if (ib_agent_port_open(device, i)) {
- printk(KERN_ERR PFX "Couldn't open %s port %d "
- "for agents\n",
- device->name, i);
+ dev_err(&device->dev,
+ "Couldn't open port %d for agents\n", i);
goto error_agent;
}
}
error_agent:
if (ib_mad_port_close(device, i))
- printk(KERN_ERR PFX "Couldn't close %s port %d\n",
- device->name, i);
+ dev_err(&device->dev, "Couldn't close port %d\n", i);
error:
i--;
while (i >= start) {
if (ib_agent_port_close(device, i))
- printk(KERN_ERR PFX "Couldn't close %s port %d "
- "for agents\n",
- device->name, i);
+ dev_err(&device->dev,
+ "Couldn't close port %d for agents\n", i);
if (ib_mad_port_close(device, i))
- printk(KERN_ERR PFX "Couldn't close %s port %d\n",
- device->name, i);
+ dev_err(&device->dev, "Couldn't close port %d\n", i);
i--;
}
}
}
for (i = 0; i < num_ports; i++, cur_port++) {
if (ib_agent_port_close(device, cur_port))
- printk(KERN_ERR PFX "Couldn't close %s port %d "
- "for agents\n",
- device->name, cur_port);
+ dev_err(&device->dev,
+ "Couldn't close port %d for agents\n",
+ cur_port);
if (ib_mad_port_close(device, cur_port))
- printk(KERN_ERR PFX "Couldn't close %s port %d\n",
- device->name, cur_port);
+ dev_err(&device->dev, "Couldn't close port %d\n",
+ cur_port);
}
}
SLAB_HWCACHE_ALIGN,
NULL);
if (!ib_mad_cache) {
- printk(KERN_ERR PFX "Couldn't create ib_mad cache\n");
+ pr_err("Couldn't create ib_mad cache\n");
ret = -ENOMEM;
goto error1;
}
INIT_LIST_HEAD(&ib_mad_port_list);
if (ib_register_client(&mad_client)) {
- printk(KERN_ERR PFX "Couldn't register ib_mad client\n");
+ pr_err("Couldn't register ib_mad client\n");
ret = -EINVAL;
goto error2;
}
#include <rdma/ib_mad.h>
#include <rdma/ib_smi.h>
-
-#define PFX "ib_mad: "
-
#define IB_MAD_QPS_CORE 2 /* Always QP0 and QP1 as a minimum */
/* QP and CQ parameters */
sa_dev->port[i].agent =
ib_register_mad_agent(device, i + s, IB_QPT_GSI,
NULL, 0, send_handler,
- recv_handler, sa_dev);
+ recv_handler, sa_dev, 0);
if (IS_ERR(sa_dev->port[i].agent))
goto err;
* SOFTWARE.
*/
+#define pr_fmt(fmt) "user_mad: " fmt
+
#include <linux/module.h>
#include <linux/init.h>
#include <linux/device.h>
rmpp_mad = (struct ib_rmpp_mad *) packet->mad.data;
hdr_len = ib_get_mad_data_offset(rmpp_mad->mad_hdr.mgmt_class);
- if (!ib_is_mad_class_rmpp(rmpp_mad->mad_hdr.mgmt_class)) {
- copy_offset = IB_MGMT_MAD_HDR;
- rmpp_active = 0;
- } else {
+
+ if (ib_is_mad_class_rmpp(rmpp_mad->mad_hdr.mgmt_class)
+ && ib_mad_kernel_rmpp_agent(agent)) {
copy_offset = IB_MGMT_RMPP_HDR;
rmpp_active = ib_get_rmpp_flags(&rmpp_mad->rmpp_hdr) &
- IB_MGMT_RMPP_FLAG_ACTIVE;
+ IB_MGMT_RMPP_FLAG_ACTIVE;
+ } else {
+ copy_offset = IB_MGMT_MAD_HDR;
+ rmpp_active = 0;
}
data_len = count - hdr_size(file) - hdr_len;
rmpp_mad->mad_hdr.tid = *tid;
}
- spin_lock_irq(&file->send_lock);
- ret = is_duplicate(file, packet);
- if (!ret)
+ if (!ib_mad_kernel_rmpp_agent(agent)
+ && ib_is_mad_class_rmpp(rmpp_mad->mad_hdr.mgmt_class)
+ && (ib_get_rmpp_flags(&rmpp_mad->rmpp_hdr) & IB_MGMT_RMPP_FLAG_ACTIVE)) {
+ spin_lock_irq(&file->send_lock);
list_add_tail(&packet->list, &file->send_list);
- spin_unlock_irq(&file->send_lock);
- if (ret) {
- ret = -EINVAL;
- goto err_msg;
+ spin_unlock_irq(&file->send_lock);
+ } else {
+ spin_lock_irq(&file->send_lock);
+ ret = is_duplicate(file, packet);
+ if (!ret)
+ list_add_tail(&packet->list, &file->send_list);
+ spin_unlock_irq(&file->send_lock);
+ if (ret) {
+ ret = -EINVAL;
+ goto err_msg;
+ }
}
ret = ib_post_send_mad(packet->msg, NULL);
mutex_lock(&file->mutex);
if (!file->port->ib_dev) {
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent: invalid device\n");
ret = -EPIPE;
goto out;
}
}
if (ureq.qpn != 0 && ureq.qpn != 1) {
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent: invalid QPN %d specified\n",
+ ureq.qpn);
ret = -EINVAL;
goto out;
}
if (!__get_agent(file, agent_id))
goto found;
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent: Max Agents (%u) reached\n",
+ IB_UMAD_MAX_AGENTS);
ret = -ENOMEM;
goto out;
found:
if (ureq.mgmt_class) {
+ memset(&req, 0, sizeof(req));
req.mgmt_class = ureq.mgmt_class;
req.mgmt_class_version = ureq.mgmt_class_version;
memcpy(req.oui, ureq.oui, sizeof req.oui);
ureq.qpn ? IB_QPT_GSI : IB_QPT_SMI,
ureq.mgmt_class ? &req : NULL,
ureq.rmpp_version,
- send_handler, recv_handler, file);
+ send_handler, recv_handler, file, 0);
if (IS_ERR(agent)) {
ret = PTR_ERR(agent);
agent = NULL;
if (!file->already_used) {
file->already_used = 1;
if (!file->use_pkey_index) {
- printk(KERN_WARNING "user_mad: process %s did not enable "
- "P_Key index support.\n", current->comm);
- printk(KERN_WARNING "user_mad: Documentation/infiniband/user_mad.txt "
- "has info on the new ABI.\n");
+ dev_warn(file->port->dev,
+ "process %s did not enable P_Key index support.\n",
+ current->comm);
+ dev_warn(file->port->dev,
+ " Documentation/infiniband/user_mad.txt has info on the new ABI.\n");
}
}
return ret;
}
+static int ib_umad_reg_agent2(struct ib_umad_file *file, void __user *arg)
+{
+ struct ib_user_mad_reg_req2 ureq;
+ struct ib_mad_reg_req req;
+ struct ib_mad_agent *agent = NULL;
+ int agent_id;
+ int ret;
+
+ mutex_lock(&file->port->file_mutex);
+ mutex_lock(&file->mutex);
+
+ if (!file->port->ib_dev) {
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent2: invalid device\n");
+ ret = -EPIPE;
+ goto out;
+ }
+
+ if (copy_from_user(&ureq, arg, sizeof(ureq))) {
+ ret = -EFAULT;
+ goto out;
+ }
+
+ if (ureq.qpn != 0 && ureq.qpn != 1) {
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent2: invalid QPN %d specified\n",
+ ureq.qpn);
+ ret = -EINVAL;
+ goto out;
+ }
+
+ if (ureq.flags & ~IB_USER_MAD_REG_FLAGS_CAP) {
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent2 failed: invalid registration flags specified 0x%x; supported 0x%x\n",
+ ureq.flags, IB_USER_MAD_REG_FLAGS_CAP);
+ ret = -EINVAL;
+
+ if (put_user((u32)IB_USER_MAD_REG_FLAGS_CAP,
+ (u32 __user *) (arg + offsetof(struct
+ ib_user_mad_reg_req2, flags))))
+ ret = -EFAULT;
+
+ goto out;
+ }
+
+ for (agent_id = 0; agent_id < IB_UMAD_MAX_AGENTS; ++agent_id)
+ if (!__get_agent(file, agent_id))
+ goto found;
+
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent2: Max Agents (%u) reached\n",
+ IB_UMAD_MAX_AGENTS);
+ ret = -ENOMEM;
+ goto out;
+
+found:
+ if (ureq.mgmt_class) {
+ memset(&req, 0, sizeof(req));
+ req.mgmt_class = ureq.mgmt_class;
+ req.mgmt_class_version = ureq.mgmt_class_version;
+ if (ureq.oui & 0xff000000) {
+ dev_notice(file->port->dev,
+ "ib_umad_reg_agent2 failed: oui invalid 0x%08x\n",
+ ureq.oui);
+ ret = -EINVAL;
+ goto out;
+ }
+ req.oui[2] = ureq.oui & 0x0000ff;
+ req.oui[1] = (ureq.oui & 0x00ff00) >> 8;
+ req.oui[0] = (ureq.oui & 0xff0000) >> 16;
+ memcpy(req.method_mask, ureq.method_mask,
+ sizeof(req.method_mask));
+ }
+
+ agent = ib_register_mad_agent(file->port->ib_dev, file->port->port_num,
+ ureq.qpn ? IB_QPT_GSI : IB_QPT_SMI,
+ ureq.mgmt_class ? &req : NULL,
+ ureq.rmpp_version,
+ send_handler, recv_handler, file,
+ ureq.flags);
+ if (IS_ERR(agent)) {
+ ret = PTR_ERR(agent);
+ agent = NULL;
+ goto out;
+ }
+
+ if (put_user(agent_id,
+ (u32 __user *)(arg +
+ offsetof(struct ib_user_mad_reg_req2, id)))) {
+ ret = -EFAULT;
+ goto out;
+ }
+
+ if (!file->already_used) {
+ file->already_used = 1;
+ file->use_pkey_index = 1;
+ }
+
+ file->agent[agent_id] = agent;
+ ret = 0;
+
+out:
+ mutex_unlock(&file->mutex);
+
+ if (ret && agent)
+ ib_unregister_mad_agent(agent);
+
+ mutex_unlock(&file->port->file_mutex);
+
+ return ret;
+}
+
+
static int ib_umad_unreg_agent(struct ib_umad_file *file, u32 __user *arg)
{
struct ib_mad_agent *agent = NULL;
return ib_umad_unreg_agent(filp->private_data, (__u32 __user *) arg);
case IB_USER_MAD_ENABLE_PKEY:
return ib_umad_enable_pkey(filp->private_data);
+ case IB_USER_MAD_REGISTER_AGENT2:
+ return ib_umad_reg_agent2(filp->private_data, (void __user *) arg);
default:
return -ENOIOCTLCMD;
}
return ib_umad_unreg_agent(filp->private_data, compat_ptr(arg));
case IB_USER_MAD_ENABLE_PKEY:
return ib_umad_enable_pkey(filp->private_data);
+ case IB_USER_MAD_REGISTER_AGENT2:
+ return ib_umad_reg_agent2(filp->private_data, compat_ptr(arg));
default:
return -ENOIOCTLCMD;
}
static dev_t overflow_maj;
static DECLARE_BITMAP(overflow_map, IB_UMAD_MAX_PORTS);
-static int find_overflow_devnum(void)
+static int find_overflow_devnum(struct ib_device *device)
{
int ret;
ret = alloc_chrdev_region(&overflow_maj, 0, IB_UMAD_MAX_PORTS * 2,
"infiniband_mad");
if (ret) {
- printk(KERN_ERR "user_mad: couldn't register dynamic device number\n");
+ dev_err(&device->dev,
+ "couldn't register dynamic device number\n");
return ret;
}
}
devnum = find_first_zero_bit(dev_map, IB_UMAD_MAX_PORTS);
if (devnum >= IB_UMAD_MAX_PORTS) {
spin_unlock(&port_lock);
- devnum = find_overflow_devnum();
+ devnum = find_overflow_devnum(device);
if (devnum < 0)
return -1;
ret = register_chrdev_region(base_dev, IB_UMAD_MAX_PORTS * 2,
"infiniband_mad");
if (ret) {
- printk(KERN_ERR "user_mad: couldn't register device number\n");
+ pr_err("couldn't register device number\n");
goto out;
}
umad_class = class_create(THIS_MODULE, "infiniband_mad");
if (IS_ERR(umad_class)) {
ret = PTR_ERR(umad_class);
- printk(KERN_ERR "user_mad: couldn't create class infiniband_mad\n");
+ pr_err("couldn't create class infiniband_mad\n");
goto out_chrdev;
}
ret = class_create_file(umad_class, &class_attr_abi_version.attr);
if (ret) {
- printk(KERN_ERR "user_mad: couldn't create abi_version attribute\n");
+ pr_err("couldn't create abi_version attribute\n");
goto out_class;
}
ret = ib_register_client(&umad_client);
if (ret) {
- printk(KERN_ERR "user_mad: couldn't register ib_umad client\n");
+ pr_err("couldn't register ib_umad client\n");
goto out_class;
}
IB_UVERBS_DECLARE_CMD(alloc_pd);
IB_UVERBS_DECLARE_CMD(dealloc_pd);
IB_UVERBS_DECLARE_CMD(reg_mr);
+IB_UVERBS_DECLARE_CMD(rereg_mr);
IB_UVERBS_DECLARE_CMD(dereg_mr);
IB_UVERBS_DECLARE_CMD(alloc_mw);
IB_UVERBS_DECLARE_CMD(dealloc_mw);
return ret;
}
+ssize_t ib_uverbs_rereg_mr(struct ib_uverbs_file *file,
+ const char __user *buf, int in_len,
+ int out_len)
+{
+ struct ib_uverbs_rereg_mr cmd;
+ struct ib_uverbs_rereg_mr_resp resp;
+ struct ib_udata udata;
+ struct ib_pd *pd = NULL;
+ struct ib_mr *mr;
+ struct ib_pd *old_pd;
+ int ret;
+ struct ib_uobject *uobj;
+
+ if (out_len < sizeof(resp))
+ return -ENOSPC;
+
+ if (copy_from_user(&cmd, buf, sizeof(cmd)))
+ return -EFAULT;
+
+ INIT_UDATA(&udata, buf + sizeof(cmd),
+ (unsigned long) cmd.response + sizeof(resp),
+ in_len - sizeof(cmd), out_len - sizeof(resp));
+
+ if (cmd.flags & ~IB_MR_REREG_SUPPORTED || !cmd.flags)
+ return -EINVAL;
+
+ if ((cmd.flags & IB_MR_REREG_TRANS) &&
+ (!cmd.start || !cmd.hca_va || 0 >= cmd.length ||
+ (cmd.start & ~PAGE_MASK) != (cmd.hca_va & ~PAGE_MASK)))
+ return -EINVAL;
+
+ uobj = idr_write_uobj(&ib_uverbs_mr_idr, cmd.mr_handle,
+ file->ucontext);
+
+ if (!uobj)
+ return -EINVAL;
+
+ mr = uobj->object;
+
+ if (cmd.flags & IB_MR_REREG_ACCESS) {
+ ret = ib_check_mr_access(cmd.access_flags);
+ if (ret)
+ goto put_uobjs;
+ }
+
+ if (cmd.flags & IB_MR_REREG_PD) {
+ pd = idr_read_pd(cmd.pd_handle, file->ucontext);
+ if (!pd) {
+ ret = -EINVAL;
+ goto put_uobjs;
+ }
+ }
+
+ if (atomic_read(&mr->usecnt)) {
+ ret = -EBUSY;
+ goto put_uobj_pd;
+ }
+
+ old_pd = mr->pd;
+ ret = mr->device->rereg_user_mr(mr, cmd.flags, cmd.start,
+ cmd.length, cmd.hca_va,
+ cmd.access_flags, pd, &udata);
+ if (!ret) {
+ if (cmd.flags & IB_MR_REREG_PD) {
+ atomic_inc(&pd->usecnt);
+ mr->pd = pd;
+ atomic_dec(&old_pd->usecnt);
+ }
+ } else {
+ goto put_uobj_pd;
+ }
+
+ memset(&resp, 0, sizeof(resp));
+ resp.lkey = mr->lkey;
+ resp.rkey = mr->rkey;
+
+ if (copy_to_user((void __user *)(unsigned long)cmd.response,
+ &resp, sizeof(resp)))
+ ret = -EFAULT;
+ else
+ ret = in_len;
+
+put_uobj_pd:
+ if (cmd.flags & IB_MR_REREG_PD)
+ put_pd_read(pd);
+
+put_uobjs:
+
+ put_uobj_write(mr->uobject);
+
+ return ret;
+}
+
ssize_t ib_uverbs_dereg_mr(struct ib_uverbs_file *file,
const char __user *buf, int in_len,
int out_len)
[IB_USER_VERBS_CMD_ALLOC_PD] = ib_uverbs_alloc_pd,
[IB_USER_VERBS_CMD_DEALLOC_PD] = ib_uverbs_dealloc_pd,
[IB_USER_VERBS_CMD_REG_MR] = ib_uverbs_reg_mr,
+ [IB_USER_VERBS_CMD_REREG_MR] = ib_uverbs_rereg_mr,
[IB_USER_VERBS_CMD_DEREG_MR] = ib_uverbs_dereg_mr,
[IB_USER_VERBS_CMD_ALLOC_MW] = ib_uverbs_alloc_mw,
[IB_USER_VERBS_CMD_DEALLOC_MW] = ib_uverbs_dealloc_mw,
mq->msg_pool.host, dma_unmap_addr(mq, mapping));
}
-static int c2_alloc_cq_buf(struct c2_dev *c2dev, struct c2_mq *mq, int q_size,
- int msg_size)
+static int c2_alloc_cq_buf(struct c2_dev *c2dev, struct c2_mq *mq,
+ size_t q_size, size_t msg_size)
{
u8 *pool_start;
+ if (q_size > SIZE_MAX / msg_size)
+ return -EINVAL;
+
pool_start = dma_alloc_coherent(&c2dev->pcidev->dev, q_size * msg_size,
&mq->host_dma, GFP_KERNEL);
if (!pool_start)
chp = get_chp(dev, qid);
if (chp) {
+ t4_clear_cq_armed(&chp->cq);
spin_lock_irqsave(&chp->comp_handler_lock, flag);
(*chp->ibcq.comp_handler)(&chp->ibcq, chp->ibcq.cq_context);
spin_unlock_irqrestore(&chp->comp_handler_lock, flag);
struct c4iw_cq *schp)
{
int count;
- int flushed;
+ int rq_flushed, sq_flushed;
unsigned long flag;
PDBG("%s qhp %p rchp %p schp %p\n", __func__, qhp, rchp, schp);
c4iw_flush_hw_cq(rchp);
c4iw_count_rcqes(&rchp->cq, &qhp->wq, &count);
- flushed = c4iw_flush_rq(&qhp->wq, &rchp->cq, count);
+ rq_flushed = c4iw_flush_rq(&qhp->wq, &rchp->cq, count);
spin_unlock(&qhp->lock);
spin_unlock_irqrestore(&rchp->lock, flag);
- if (flushed) {
- spin_lock_irqsave(&rchp->comp_handler_lock, flag);
- (*rchp->ibcq.comp_handler)(&rchp->ibcq, rchp->ibcq.cq_context);
- spin_unlock_irqrestore(&rchp->comp_handler_lock, flag);
- }
/* locking hierarchy: cq lock first, then qp lock. */
spin_lock_irqsave(&schp->lock, flag);
spin_lock(&qhp->lock);
if (schp != rchp)
c4iw_flush_hw_cq(schp);
- flushed = c4iw_flush_sq(qhp);
+ sq_flushed = c4iw_flush_sq(qhp);
spin_unlock(&qhp->lock);
spin_unlock_irqrestore(&schp->lock, flag);
- if (flushed) {
- spin_lock_irqsave(&schp->comp_handler_lock, flag);
- (*schp->ibcq.comp_handler)(&schp->ibcq, schp->ibcq.cq_context);
- spin_unlock_irqrestore(&schp->comp_handler_lock, flag);
+
+ if (schp == rchp) {
+ if (t4_clear_cq_armed(&rchp->cq) &&
+ (rq_flushed || sq_flushed)) {
+ spin_lock_irqsave(&rchp->comp_handler_lock, flag);
+ (*rchp->ibcq.comp_handler)(&rchp->ibcq,
+ rchp->ibcq.cq_context);
+ spin_unlock_irqrestore(&rchp->comp_handler_lock, flag);
+ }
+ } else {
+ if (t4_clear_cq_armed(&rchp->cq) && rq_flushed) {
+ spin_lock_irqsave(&rchp->comp_handler_lock, flag);
+ (*rchp->ibcq.comp_handler)(&rchp->ibcq,
+ rchp->ibcq.cq_context);
+ spin_unlock_irqrestore(&rchp->comp_handler_lock, flag);
+ }
+ if (t4_clear_cq_armed(&schp->cq) && sq_flushed) {
+ spin_lock_irqsave(&schp->comp_handler_lock, flag);
+ (*schp->ibcq.comp_handler)(&schp->ibcq,
+ schp->ibcq.cq_context);
+ spin_unlock_irqrestore(&schp->comp_handler_lock, flag);
+ }
}
}
return !wq->rq.queue[wq->rq.size].status.db_off;
}
+enum t4_cq_flags {
+ CQ_ARMED = 1,
+};
+
struct t4_cq {
struct t4_cqe *queue;
dma_addr_t dma_addr;
u16 cidx_inc;
u8 gen;
u8 error;
+ unsigned long flags;
};
+static inline int t4_clear_cq_armed(struct t4_cq *cq)
+{
+ return test_and_clear_bit(CQ_ARMED, &cq->flags);
+}
+
static inline int t4_arm_cq(struct t4_cq *cq, int se)
{
u32 val;
+ set_bit(CQ_ARMED, &cq->flags);
while (cq->cidx_inc > CIDXINC_MASK) {
val = SEINTARM(0) | CIDXINC(CIDXINC_MASK) | TIMERREG(7) |
INGRESSQID(cq->cqid);
* @dd: the infinipath device
* @pkeys: the PKEY table
*/
-static int set_pkeys(struct ipath_devdata *dd, u16 *pkeys)
+static int set_pkeys(struct ipath_devdata *dd, u16 *pkeys, u8 port)
{
struct ipath_portdata *pd;
int i;
}
if (changed) {
u64 pkey;
+ struct ib_event event;
pkey = (u64) dd->ipath_pkeys[0] |
((u64) dd->ipath_pkeys[1] << 16) |
(unsigned long long) pkey);
ipath_write_kreg(dd, dd->ipath_kregs->kr_partitionkey,
pkey);
+
+ event.event = IB_EVENT_PKEY_CHANGE;
+ event.device = &dd->verbs_dev->ibdev;
+ event.element.port_num = port;
+ ib_dispatch_event(&event);
}
return 0;
}
static int recv_subn_set_pkeytable(struct ib_smp *smp,
- struct ib_device *ibdev)
+ struct ib_device *ibdev, u8 port)
{
u32 startpx = 32 * (be32_to_cpu(smp->attr_mod) & 0xffff);
__be16 *p = (__be16 *) smp->data;
for (i = 0; i < n; i++)
q[i] = be16_to_cpu(p[i]);
- if (startpx != 0 || set_pkeys(dev->dd, q) != 0)
+ if (startpx != 0 || set_pkeys(dev->dd, q, port) != 0)
smp->status |= IB_SMP_INVALID_FIELD;
return recv_subn_get_pkeytable(smp, ibdev);
ret = recv_subn_set_portinfo(smp, ibdev, port_num);
goto bail;
case IB_SMP_ATTR_PKEY_TABLE:
- ret = recv_subn_set_pkeytable(smp, ibdev);
+ ret = recv_subn_set_pkeytable(smp, ibdev, port_num);
goto bail;
case IB_SMP_ATTR_SM_INFO:
if (dev->port_cap_flags & IB_PORT_SM_DISABLED) {
agent = ib_register_mad_agent(&dev->ib_dev, p + 1,
q ? IB_QPT_GSI : IB_QPT_SMI,
NULL, 0, send_handler,
- NULL, NULL);
+ NULL, NULL, 0);
if (IS_ERR(agent)) {
ret = PTR_ERR(agent);
goto err;
const struct default_rules *pdefault_rules = default_table;
u8 link_layer = rdma_port_get_link_layer(qp->device, flow_attr->port);
- for (i = 0; i < sizeof(default_table)/sizeof(default_table[0]); i++,
- pdefault_rules++) {
+ for (i = 0; i < ARRAY_SIZE(default_table); i++, pdefault_rules++) {
__u32 field_types[IB_FLOW_SPEC_SUPPORT_LAYERS];
memset(&field_types, 0, sizeof(field_types));
int size = 0;
int i;
- for (i = 0; i < sizeof(pdefault_rules->rules_create_list)/
- sizeof(pdefault_rules->rules_create_list[0]); i++) {
+ for (i = 0; i < ARRAY_SIZE(pdefault_rules->rules_create_list); i++) {
int ret;
union ib_flow_spec ib_spec;
switch (pdefault_rules->rules_create_list[i]) {
(1ull << IB_USER_VERBS_CMD_ALLOC_PD) |
(1ull << IB_USER_VERBS_CMD_DEALLOC_PD) |
(1ull << IB_USER_VERBS_CMD_REG_MR) |
+ (1ull << IB_USER_VERBS_CMD_REREG_MR) |
(1ull << IB_USER_VERBS_CMD_DEREG_MR) |
(1ull << IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL) |
(1ull << IB_USER_VERBS_CMD_CREATE_CQ) |
ibdev->ib_dev.req_notify_cq = mlx4_ib_arm_cq;
ibdev->ib_dev.get_dma_mr = mlx4_ib_get_dma_mr;
ibdev->ib_dev.reg_user_mr = mlx4_ib_reg_user_mr;
+ ibdev->ib_dev.rereg_user_mr = mlx4_ib_rereg_user_mr;
ibdev->ib_dev.dereg_mr = mlx4_ib_dereg_mr;
ibdev->ib_dev.alloc_fast_reg_mr = mlx4_ib_alloc_fast_reg_mr;
ibdev->ib_dev.alloc_fast_reg_page_list = mlx4_ib_alloc_fast_reg_page_list;
void mlx4_ib_steer_qp_free(struct mlx4_ib_dev *dev, u32 qpn, int count);
int mlx4_ib_steer_qp_reg(struct mlx4_ib_dev *mdev, struct mlx4_ib_qp *mqp,
int is_attach);
+int mlx4_ib_rereg_user_mr(struct ib_mr *mr, int flags,
+ u64 start, u64 length, u64 virt_addr,
+ int mr_access_flags, struct ib_pd *pd,
+ struct ib_udata *udata);
#endif /* MLX4_IB_H */
if (!mr)
return ERR_PTR(-ENOMEM);
+ /* Force registering the memory as writable. */
+ /* Used for memory re-registeration. HCA protects the access */
mr->umem = ib_umem_get(pd->uobject->context, start, length,
- access_flags, 0);
+ access_flags | IB_ACCESS_LOCAL_WRITE, 0);
if (IS_ERR(mr->umem)) {
err = PTR_ERR(mr->umem);
goto err_free;
return ERR_PTR(err);
}
+int mlx4_ib_rereg_user_mr(struct ib_mr *mr, int flags,
+ u64 start, u64 length, u64 virt_addr,
+ int mr_access_flags, struct ib_pd *pd,
+ struct ib_udata *udata)
+{
+ struct mlx4_ib_dev *dev = to_mdev(mr->device);
+ struct mlx4_ib_mr *mmr = to_mmr(mr);
+ struct mlx4_mpt_entry *mpt_entry;
+ struct mlx4_mpt_entry **pmpt_entry = &mpt_entry;
+ int err;
+
+ /* Since we synchronize this call and mlx4_ib_dereg_mr via uverbs,
+ * we assume that the calls can't run concurrently. Otherwise, a
+ * race exists.
+ */
+ err = mlx4_mr_hw_get_mpt(dev->dev, &mmr->mmr, &pmpt_entry);
+
+ if (err)
+ return err;
+
+ if (flags & IB_MR_REREG_PD) {
+ err = mlx4_mr_hw_change_pd(dev->dev, *pmpt_entry,
+ to_mpd(pd)->pdn);
+
+ if (err)
+ goto release_mpt_entry;
+ }
+
+ if (flags & IB_MR_REREG_ACCESS) {
+ err = mlx4_mr_hw_change_access(dev->dev, *pmpt_entry,
+ convert_access(mr_access_flags));
+
+ if (err)
+ goto release_mpt_entry;
+ }
+
+ if (flags & IB_MR_REREG_TRANS) {
+ int shift;
+ int err;
+ int n;
+
+ mlx4_mr_rereg_mem_cleanup(dev->dev, &mmr->mmr);
+ ib_umem_release(mmr->umem);
+ mmr->umem = ib_umem_get(mr->uobject->context, start, length,
+ mr_access_flags |
+ IB_ACCESS_LOCAL_WRITE,
+ 0);
+ if (IS_ERR(mmr->umem)) {
+ err = PTR_ERR(mmr->umem);
+ mmr->umem = NULL;
+ goto release_mpt_entry;
+ }
+ n = ib_umem_page_count(mmr->umem);
+ shift = ilog2(mmr->umem->page_size);
+
+ mmr->mmr.iova = virt_addr;
+ mmr->mmr.size = length;
+ err = mlx4_mr_rereg_mem_write(dev->dev, &mmr->mmr,
+ virt_addr, length, n, shift,
+ *pmpt_entry);
+ if (err) {
+ ib_umem_release(mmr->umem);
+ goto release_mpt_entry;
+ }
+
+ err = mlx4_ib_umem_write_mtt(dev, &mmr->mmr.mtt, mmr->umem);
+ if (err) {
+ mlx4_mr_rereg_mem_cleanup(dev->dev, &mmr->mmr);
+ ib_umem_release(mmr->umem);
+ goto release_mpt_entry;
+ }
+ }
+
+ /* If we couldn't transfer the MR to the HCA, just remember to
+ * return a failure. But dereg_mr will free the resources.
+ */
+ err = mlx4_mr_hw_write_mpt(dev->dev, &mmr->mmr, pmpt_entry);
+
+release_mpt_entry:
+ mlx4_mr_hw_put_mpt(dev->dev, pmpt_entry);
+
+ return err;
+}
+
int mlx4_ib_dereg_mr(struct ib_mr *ibmr)
{
struct mlx4_ib_mr *mr = to_mmr(ibmr);
spin_lock_irqsave(&qp->sq.lock, flags);
for (nreq = 0; wr; nreq++, wr = wr->next) {
- if (unlikely(wr->opcode >= sizeof(mlx5_ib_opcode) / sizeof(mlx5_ib_opcode[0]))) {
+ if (unlikely(wr->opcode >= ARRAY_SIZE(mlx5_ib_opcode))) {
mlx5_ib_warn(dev, "\n");
err = -EINVAL;
*bad_wr = wr;
agent = ib_register_mad_agent(&dev->ib_dev, p + 1,
q ? IB_QPT_GSI : IB_QPT_SMI,
NULL, 0, send_handler,
- NULL, NULL);
+ NULL, NULL, 0);
if (IS_ERR(agent)) {
ret = PTR_ERR(agent);
goto err;
#include <be_roce.h>
#include "ocrdma_sli.h"
-#define OCRDMA_ROCE_DRV_VERSION "10.2.145.0u"
+#define OCRDMA_ROCE_DRV_VERSION "10.2.287.0u"
#define OCRDMA_ROCE_DRV_DESC "Emulex OneConnect RoCE Driver"
#define OCRDMA_NODE_DESC "Emulex OneConnect RoCE HCA"
u16 cqe_status;
u16 ext_status;
bool cmd_done;
+ bool fw_error_state;
};
struct ocrdma_hw_mr {
struct list_head entry;
struct rcu_head rcu;
int id;
- u64 stag_arr[OCRDMA_MAX_STAG];
+ u64 *stag_arr;
+ u8 sl; /* service level */
+ bool pfc_state;
+ atomic_t update_sl;
u16 pvid;
u32 asic_id;
OCRDMA_SLI_ASIC_GEN_NUM_SHIFT;
}
+static inline u8 ocrdma_get_pfc_prio(u8 *pfc, u8 prio)
+{
+ return *(pfc + prio);
+}
+
+static inline u8 ocrdma_get_app_prio(u8 *app_prio, u8 prio)
+{
+ return *(app_prio + prio);
+}
+
+static inline u8 ocrdma_is_enabled_and_synced(u32 state)
+{ /* May also be used to interpret TC-state, QCN-state
+ * Appl-state and Logical-link-state in future.
+ */
+ return (state & OCRDMA_STATE_FLAG_ENABLED) &&
+ (state & OCRDMA_STATE_FLAG_SYNC);
+}
+
#endif
#include "ocrdma_ah.h"
#include "ocrdma_hw.h"
+#define OCRDMA_VID_PCP_SHIFT 0xD
+
static inline int set_av_attr(struct ocrdma_dev *dev, struct ocrdma_ah *ah,
struct ib_ah_attr *attr, int pdid)
{
if (vlan_tag && (vlan_tag < 0x1000)) {
eth.eth_type = cpu_to_be16(0x8100);
eth.roce_eth_type = cpu_to_be16(OCRDMA_ROCE_ETH_TYPE);
- vlan_tag |= (attr->sl & 7) << 13;
+ vlan_tag |= (dev->sl & 0x07) << OCRDMA_VID_PCP_SHIFT;
eth.vlan_tag = cpu_to_be16(vlan_tag);
eth_sz = sizeof(struct ocrdma_eth_vlan);
vlan_enabled = true;
if (!(attr->ah_flags & IB_AH_GRH))
return ERR_PTR(-EINVAL);
+ if (atomic_cmpxchg(&dev->update_sl, 1, 0))
+ ocrdma_init_service_level(dev);
ah = kzalloc(sizeof(*ah), GFP_ATOMIC);
if (!ah)
return ERR_PTR(-ENOMEM);
cmd->ev_cnt_flags = OCRDMA_CREATE_CQ_DEF_FLAGS;
cmd->eqn = eq->id;
- cmd->cqe_count = cq->size / sizeof(struct ocrdma_mcqe);
+ cmd->pdid_cqecnt = cq->size / sizeof(struct ocrdma_mcqe);
ocrdma_build_q_pages(&cmd->pa[0], cq->size / OCRDMA_MIN_Q_PAGE_SIZE,
cq->dma, PAGE_SIZE_4K);
{
struct ocrdma_qp *qp = NULL;
struct ocrdma_cq *cq = NULL;
- struct ib_event ib_evt = { 0 };
+ struct ib_event ib_evt;
int cq_event = 0;
int qp_event = 1;
int srq_event = 0;
if (cqe->cqvalid_cqid & OCRDMA_AE_MCQE_CQVALID)
cq = dev->cq_tbl[cqe->cqvalid_cqid & OCRDMA_AE_MCQE_CQID_MASK];
+ memset(&ib_evt, 0, sizeof(ib_evt));
+
ib_evt.device = &dev->ibdev;
switch (type) {
OCRDMA_AE_PVID_MCQE_TAG_MASK) >>
OCRDMA_AE_PVID_MCQE_TAG_SHIFT);
break;
+
+ case OCRDMA_ASYNC_EVENT_COS_VALUE:
+ atomic_set(&dev->update_sl, 1);
+ break;
default:
/* Not interested evts. */
break;
msecs_to_jiffies(30000));
if (status)
return 0;
- else
+ else {
+ dev->mqe_ctx.fw_error_state = true;
+ pr_err("%s(%d) mailbox timeout: fw not responding\n",
+ __func__, dev->id);
return -1;
+ }
}
/* issue a mailbox command on the MQ */
struct ocrdma_mbx_rsp *rsp = NULL;
mutex_lock(&dev->mqe_ctx.lock);
+ if (dev->mqe_ctx.fw_error_state)
+ goto mbx_err;
ocrdma_post_mqe(dev, mqe);
status = ocrdma_wait_mqe_cmpl(dev);
if (status)
OCRDMA_MBX_QUERY_CFG_CA_ACK_DELAY_SHIFT;
attr->max_mw = rsp->max_mw;
attr->max_mr = rsp->max_mr;
- attr->max_mr_size = ~0ull;
+ attr->max_mr_size = ((u64)rsp->max_mr_size_hi << 32) |
+ rsp->max_mr_size_lo;
attr->max_fmr = 0;
attr->max_pages_per_frmr = rsp->max_pages_per_frmr;
attr->max_num_mr_pbl = rsp->max_num_mr_pbl;
ctrl_attr_rsp = (struct ocrdma_get_ctrl_attribs_rsp *)dma.va;
hba_attribs = &ctrl_attr_rsp->ctrl_attribs.hba_attribs;
- dev->hba_port_num = hba_attribs->phy_port;
+ dev->hba_port_num = (hba_attribs->ptpnum_maxdoms_hbast_cv &
+ OCRDMA_HBA_ATTRB_PTNUM_MASK)
+ >> OCRDMA_HBA_ATTRB_PTNUM_SHIFT;
strncpy(dev->model_number,
hba_attribs->controller_model_number, 31);
}
goto mbx_err;
rsp = (struct ocrdma_get_link_speed_rsp *)cmd;
- *lnk_speed = rsp->phys_port_speed;
+ *lnk_speed = (rsp->pflt_pps_ld_pnum & OCRDMA_PHY_PS_MASK)
+ >> OCRDMA_PHY_PS_SHIFT;
mbx_err:
kfree(cmd);
goto mbx_err;
rsp = (struct ocrdma_get_phy_info_rsp *)cmd;
- dev->phy.phy_type = le16_to_cpu(rsp->phy_type);
+ dev->phy.phy_type =
+ (rsp->ityp_ptyp & OCRDMA_PHY_TYPE_MASK);
+ dev->phy.interface_type =
+ (rsp->ityp_ptyp & OCRDMA_IF_TYPE_MASK)
+ >> OCRDMA_IF_TYPE_SHIFT;
dev->phy.auto_speeds_supported =
- le16_to_cpu(rsp->auto_speeds_supported);
+ (rsp->fspeed_aspeed & OCRDMA_ASPEED_SUPP_MASK);
dev->phy.fixed_speeds_supported =
- le16_to_cpu(rsp->fixed_speeds_supported);
+ (rsp->fspeed_aspeed & OCRDMA_FSPEED_SUPP_MASK)
+ >> OCRDMA_FSPEED_SUPP_SHIFT;
mbx_err:
kfree(cmd);
return status;
pbes = (struct ocrdma_pbe *)dev->av_tbl.pbl.va;
for (i = 0; i < dev->av_tbl.size / OCRDMA_MIN_Q_PAGE_SIZE; i++) {
- pbes[i].pa_lo = (u32) (pa & 0xffffffff);
- pbes[i].pa_hi = (u32) upper_32_bits(pa);
+ pbes[i].pa_lo = (u32)cpu_to_le32(pa & 0xffffffff);
+ pbes[i].pa_hi = (u32)cpu_to_le32(upper_32_bits(pa));
pa += PAGE_SIZE;
}
cmd->tbl_addr[0].lo = (u32)(dev->av_tbl.pbl.pa & 0xFFFFFFFF);
ocrdma_mbx_cmd(dev, (struct ocrdma_mqe *)cmd);
dma_free_coherent(&pdev->dev, dev->av_tbl.size, dev->av_tbl.va,
dev->av_tbl.pa);
+ dev->av_tbl.va = NULL;
dma_free_coherent(&pdev->dev, PAGE_SIZE, dev->av_tbl.pbl.va,
dev->av_tbl.pbl.pa);
kfree(cmd);
cmd->cmd.pgsz_pgcnt |= OCRDMA_CREATE_CQ_DPP <<
OCRDMA_CREATE_CQ_TYPE_SHIFT;
cq->phase_change = false;
- cmd->cmd.cqe_count = (cq->len / cqe_size);
+ cmd->cmd.pdid_cqecnt = (cq->len / cqe_size);
} else {
- cmd->cmd.cqe_count = (cq->len / cqe_size) - 1;
+ cmd->cmd.pdid_cqecnt = (cq->len / cqe_size) - 1;
cmd->cmd.ev_cnt_flags |= OCRDMA_CREATE_CQ_FLAGS_AUTO_VALID;
cq->phase_change = true;
}
- cmd->cmd.pd_id = pd_id; /* valid only for v3 */
+ /* pd_id valid only for v3 */
+ cmd->cmd.pdid_cqecnt |= (pd_id <<
+ OCRDMA_CREATE_CQ_CMD_PDID_SHIFT);
ocrdma_build_q_pages(&cmd->cmd.pa[0], hw_pages, cq->pa, page_size);
status = ocrdma_mbx_cmd(dev, (struct ocrdma_mqe *)cmd);
if (status)
OCRDMA_CREATE_QP_REQ_RQ_CQID_MASK;
qp->rq_cq = cq;
- if (pd->dpp_enabled && pd->num_dpp_qp) {
+ if (pd->dpp_enabled && attrs->cap.max_inline_data && pd->num_dpp_qp &&
+ (attrs->cap.max_inline_data <= dev->attr.max_inline_data)) {
ocrdma_set_create_qp_dpp_cmd(cmd, pd, qp, enable_dpp_cq,
dpp_cq_id);
}
if ((ah_attr->ah_flags & IB_AH_GRH) == 0)
return -EINVAL;
+ if (atomic_cmpxchg(&qp->dev->update_sl, 1, 0))
+ ocrdma_init_service_level(qp->dev);
cmd->params.tclass_sq_psn |=
(ah_attr->grh.traffic_class << OCRDMA_QP_PARAMS_TCLASS_SHIFT);
cmd->params.rnt_rc_sl_fl |=
cmd->params.vlan_dmac_b4_to_b5 |=
vlan_id << OCRDMA_QP_PARAMS_VLAN_SHIFT;
cmd->flags |= OCRDMA_QP_PARA_VLAN_EN_VALID;
+ cmd->params.rnt_rc_sl_fl |=
+ (qp->dev->sl & 0x07) << OCRDMA_QP_PARAMS_SL_SHIFT;
}
return 0;
}
return status;
}
+static int ocrdma_mbx_get_dcbx_config(struct ocrdma_dev *dev, u32 ptype,
+ struct ocrdma_dcbx_cfg *dcbxcfg)
+{
+ int status = 0;
+ dma_addr_t pa;
+ struct ocrdma_mqe cmd;
+
+ struct ocrdma_get_dcbx_cfg_req *req = NULL;
+ struct ocrdma_get_dcbx_cfg_rsp *rsp = NULL;
+ struct pci_dev *pdev = dev->nic_info.pdev;
+ struct ocrdma_mqe_sge *mqe_sge = cmd.u.nonemb_req.sge;
+
+ memset(&cmd, 0, sizeof(struct ocrdma_mqe));
+ cmd.hdr.pyld_len = max_t (u32, sizeof(struct ocrdma_get_dcbx_cfg_rsp),
+ sizeof(struct ocrdma_get_dcbx_cfg_req));
+ req = dma_alloc_coherent(&pdev->dev, cmd.hdr.pyld_len, &pa, GFP_KERNEL);
+ if (!req) {
+ status = -ENOMEM;
+ goto mem_err;
+ }
+
+ cmd.hdr.spcl_sge_cnt_emb |= (1 << OCRDMA_MQE_HDR_SGE_CNT_SHIFT) &
+ OCRDMA_MQE_HDR_SGE_CNT_MASK;
+ mqe_sge->pa_lo = (u32) (pa & 0xFFFFFFFFUL);
+ mqe_sge->pa_hi = (u32) upper_32_bits(pa);
+ mqe_sge->len = cmd.hdr.pyld_len;
+
+ memset(req, 0, sizeof(struct ocrdma_get_dcbx_cfg_req));
+ ocrdma_init_mch(&req->hdr, OCRDMA_CMD_GET_DCBX_CONFIG,
+ OCRDMA_SUBSYS_DCBX, cmd.hdr.pyld_len);
+ req->param_type = ptype;
+
+ status = ocrdma_mbx_cmd(dev, &cmd);
+ if (status)
+ goto mbx_err;
+
+ rsp = (struct ocrdma_get_dcbx_cfg_rsp *)req;
+ ocrdma_le32_to_cpu(rsp, sizeof(struct ocrdma_get_dcbx_cfg_rsp));
+ memcpy(dcbxcfg, &rsp->cfg, sizeof(struct ocrdma_dcbx_cfg));
+
+mbx_err:
+ dma_free_coherent(&pdev->dev, cmd.hdr.pyld_len, req, pa);
+mem_err:
+ return status;
+}
+
+#define OCRDMA_MAX_SERVICE_LEVEL_INDEX 0x08
+#define OCRDMA_DEFAULT_SERVICE_LEVEL 0x05
+
+static int ocrdma_parse_dcbxcfg_rsp(struct ocrdma_dev *dev, int ptype,
+ struct ocrdma_dcbx_cfg *dcbxcfg,
+ u8 *srvc_lvl)
+{
+ int status = -EINVAL, indx, slindx;
+ int ventry_cnt;
+ struct ocrdma_app_parameter *app_param;
+ u8 valid, proto_sel;
+ u8 app_prio, pfc_prio;
+ u16 proto;
+
+ if (!(dcbxcfg->tcv_aev_opv_st & OCRDMA_DCBX_STATE_MASK)) {
+ pr_info("%s ocrdma%d DCBX is disabled\n",
+ dev_name(&dev->nic_info.pdev->dev), dev->id);
+ goto out;
+ }
+
+ if (!ocrdma_is_enabled_and_synced(dcbxcfg->pfc_state)) {
+ pr_info("%s ocrdma%d priority flow control(%s) is %s%s\n",
+ dev_name(&dev->nic_info.pdev->dev), dev->id,
+ (ptype > 0 ? "operational" : "admin"),
+ (dcbxcfg->pfc_state & OCRDMA_STATE_FLAG_ENABLED) ?
+ "enabled" : "disabled",
+ (dcbxcfg->pfc_state & OCRDMA_STATE_FLAG_SYNC) ?
+ "" : ", not sync'ed");
+ goto out;
+ } else {
+ pr_info("%s ocrdma%d priority flow control is enabled and sync'ed\n",
+ dev_name(&dev->nic_info.pdev->dev), dev->id);
+ }
+
+ ventry_cnt = (dcbxcfg->tcv_aev_opv_st >>
+ OCRDMA_DCBX_APP_ENTRY_SHIFT)
+ & OCRDMA_DCBX_STATE_MASK;
+
+ for (indx = 0; indx < ventry_cnt; indx++) {
+ app_param = &dcbxcfg->app_param[indx];
+ valid = (app_param->valid_proto_app >>
+ OCRDMA_APP_PARAM_VALID_SHIFT)
+ & OCRDMA_APP_PARAM_VALID_MASK;
+ proto_sel = (app_param->valid_proto_app
+ >> OCRDMA_APP_PARAM_PROTO_SEL_SHIFT)
+ & OCRDMA_APP_PARAM_PROTO_SEL_MASK;
+ proto = app_param->valid_proto_app &
+ OCRDMA_APP_PARAM_APP_PROTO_MASK;
+
+ if (
+ valid && proto == OCRDMA_APP_PROTO_ROCE &&
+ proto_sel == OCRDMA_PROTO_SELECT_L2) {
+ for (slindx = 0; slindx <
+ OCRDMA_MAX_SERVICE_LEVEL_INDEX; slindx++) {
+ app_prio = ocrdma_get_app_prio(
+ (u8 *)app_param->app_prio,
+ slindx);
+ pfc_prio = ocrdma_get_pfc_prio(
+ (u8 *)dcbxcfg->pfc_prio,
+ slindx);
+
+ if (app_prio && pfc_prio) {
+ *srvc_lvl = slindx;
+ status = 0;
+ goto out;
+ }
+ }
+ if (slindx == OCRDMA_MAX_SERVICE_LEVEL_INDEX) {
+ pr_info("%s ocrdma%d application priority not set for 0x%x protocol\n",
+ dev_name(&dev->nic_info.pdev->dev),
+ dev->id, proto);
+ }
+ }
+ }
+
+out:
+ return status;
+}
+
+void ocrdma_init_service_level(struct ocrdma_dev *dev)
+{
+ int status = 0, indx;
+ struct ocrdma_dcbx_cfg dcbxcfg;
+ u8 srvc_lvl = OCRDMA_DEFAULT_SERVICE_LEVEL;
+ int ptype = OCRDMA_PARAMETER_TYPE_OPER;
+
+ for (indx = 0; indx < 2; indx++) {
+ status = ocrdma_mbx_get_dcbx_config(dev, ptype, &dcbxcfg);
+ if (status) {
+ pr_err("%s(): status=%d\n", __func__, status);
+ ptype = OCRDMA_PARAMETER_TYPE_ADMIN;
+ continue;
+ }
+
+ status = ocrdma_parse_dcbxcfg_rsp(dev, ptype,
+ &dcbxcfg, &srvc_lvl);
+ if (status) {
+ ptype = OCRDMA_PARAMETER_TYPE_ADMIN;
+ continue;
+ }
+
+ break;
+ }
+
+ if (status)
+ pr_info("%s ocrdma%d service level default\n",
+ dev_name(&dev->nic_info.pdev->dev), dev->id);
+ else
+ pr_info("%s ocrdma%d service level %d\n",
+ dev_name(&dev->nic_info.pdev->dev), dev->id,
+ srvc_lvl);
+
+ dev->pfc_state = ocrdma_is_enabled_and_synced(dcbxcfg.pfc_state);
+ dev->sl = srvc_lvl;
+}
+
int ocrdma_alloc_av(struct ocrdma_dev *dev, struct ocrdma_ah *ah)
{
int i;
goto conf_err;
status = ocrdma_mbx_get_phy_info(dev);
if (status)
- goto conf_err;
+ goto info_attrb_err;
status = ocrdma_mbx_get_ctrl_attribs(dev);
if (status)
- goto conf_err;
+ goto info_attrb_err;
return 0;
+info_attrb_err:
+ ocrdma_mbx_delete_ah_tbl(dev);
conf_err:
ocrdma_destroy_mq(dev);
mq_err:
int ocrdma_mbx_rdma_stats(struct ocrdma_dev *, bool reset);
char *port_speed_string(struct ocrdma_dev *dev);
+void ocrdma_init_service_level(struct ocrdma_dev *);
+
#endif /* __OCRDMA_HW_H__ */
if (!dev->qp_tbl)
goto alloc_err;
}
+
+ dev->stag_arr = kzalloc(sizeof(u64) * OCRDMA_MAX_STAG, GFP_KERNEL);
+ if (dev->stag_arr == NULL)
+ goto alloc_err;
+
spin_lock_init(&dev->av_tbl.lock);
spin_lock_init(&dev->flush_q_lock);
return 0;
static void ocrdma_free_resources(struct ocrdma_dev *dev)
{
+ kfree(dev->stag_arr);
kfree(dev->qp_tbl);
kfree(dev->cq_tbl);
kfree(dev->sgid_tbl);
{
struct ocrdma_dev *dev = dev_get_drvdata(device);
- return scnprintf(buf, PAGE_SIZE, "%s", &dev->attr.fw_ver[0]);
+ return scnprintf(buf, PAGE_SIZE, "%s\n", &dev->attr.fw_ver[0]);
+}
+
+static ssize_t show_hca_type(struct device *device,
+ struct device_attribute *attr, char *buf)
+{
+ struct ocrdma_dev *dev = dev_get_drvdata(device);
+
+ return scnprintf(buf, PAGE_SIZE, "%s\n", &dev->model_number[0]);
}
static DEVICE_ATTR(hw_rev, S_IRUGO, show_rev, NULL);
static DEVICE_ATTR(fw_ver, S_IRUGO, show_fw_ver, NULL);
+static DEVICE_ATTR(hca_type, S_IRUGO, show_hca_type, NULL);
static struct device_attribute *ocrdma_attributes[] = {
&dev_attr_hw_rev,
- &dev_attr_fw_ver
+ &dev_attr_fw_ver,
+ &dev_attr_hca_type
};
static void ocrdma_remove_sysfiles(struct ocrdma_dev *dev)
device_remove_file(&dev->ibdev.dev, ocrdma_attributes[i]);
}
+static void ocrdma_init_ipv4_gids(struct ocrdma_dev *dev,
+ struct net_device *net)
+{
+ struct in_device *in_dev;
+ union ib_gid gid;
+ in_dev = in_dev_get(net);
+ if (in_dev) {
+ for_ifa(in_dev) {
+ ipv6_addr_set_v4mapped(ifa->ifa_address,
+ (struct in6_addr *)&gid);
+ ocrdma_add_sgid(dev, &gid);
+ }
+ endfor_ifa(in_dev);
+ in_dev_put(in_dev);
+ }
+}
+
+static void ocrdma_init_ipv6_gids(struct ocrdma_dev *dev,
+ struct net_device *net)
+{
+#if IS_ENABLED(CONFIG_IPV6)
+ struct inet6_dev *in6_dev;
+ union ib_gid *pgid;
+ struct inet6_ifaddr *ifp;
+ in6_dev = in6_dev_get(net);
+ if (in6_dev) {
+ read_lock_bh(&in6_dev->lock);
+ list_for_each_entry(ifp, &in6_dev->addr_list, if_list) {
+ pgid = (union ib_gid *)&ifp->addr;
+ ocrdma_add_sgid(dev, pgid);
+ }
+ read_unlock_bh(&in6_dev->lock);
+ in6_dev_put(in6_dev);
+ }
+#endif
+}
+
+static void ocrdma_init_gid_table(struct ocrdma_dev *dev)
+{
+ struct net_device *net_dev;
+
+ for_each_netdev(&init_net, net_dev) {
+ struct net_device *real_dev = rdma_vlan_dev_real_dev(net_dev) ?
+ rdma_vlan_dev_real_dev(net_dev) : net_dev;
+
+ if (real_dev == dev->nic_info.netdev) {
+ ocrdma_init_ipv4_gids(dev, net_dev);
+ ocrdma_init_ipv6_gids(dev, net_dev);
+ }
+ }
+}
+
static struct ocrdma_dev *ocrdma_add(struct be_dev_info *dev_info)
{
int status = 0, i;
if (status)
goto alloc_err;
+ ocrdma_init_service_level(dev);
+ ocrdma_init_gid_table(dev);
status = ocrdma_register_device(dev);
if (status)
goto alloc_err;
return 0;
}
+static void ocrdma_shutdown(struct ocrdma_dev *dev)
+{
+ ocrdma_close(dev);
+ ocrdma_remove(dev);
+}
+
/* event handling via NIC driver ensures that all the NIC specific
* initialization done before RoCE driver notifies
* event to stack.
case BE_DEV_DOWN:
ocrdma_close(dev);
break;
+ case BE_DEV_SHUTDOWN:
+ ocrdma_shutdown(dev);
+ break;
}
}
#define OCRDMA_SUBSYS_ROCE 10
enum {
OCRDMA_CMD_QUERY_CONFIG = 1,
- OCRDMA_CMD_ALLOC_PD,
- OCRDMA_CMD_DEALLOC_PD,
-
- OCRDMA_CMD_CREATE_AH_TBL,
- OCRDMA_CMD_DELETE_AH_TBL,
-
- OCRDMA_CMD_CREATE_QP,
- OCRDMA_CMD_QUERY_QP,
- OCRDMA_CMD_MODIFY_QP,
- OCRDMA_CMD_DELETE_QP,
-
- OCRDMA_CMD_RSVD1,
- OCRDMA_CMD_ALLOC_LKEY,
- OCRDMA_CMD_DEALLOC_LKEY,
- OCRDMA_CMD_REGISTER_NSMR,
- OCRDMA_CMD_REREGISTER_NSMR,
- OCRDMA_CMD_REGISTER_NSMR_CONT,
- OCRDMA_CMD_QUERY_NSMR,
- OCRDMA_CMD_ALLOC_MW,
- OCRDMA_CMD_QUERY_MW,
-
- OCRDMA_CMD_CREATE_SRQ,
- OCRDMA_CMD_QUERY_SRQ,
- OCRDMA_CMD_MODIFY_SRQ,
- OCRDMA_CMD_DELETE_SRQ,
-
- OCRDMA_CMD_ATTACH_MCAST,
- OCRDMA_CMD_DETACH_MCAST,
- OCRDMA_CMD_GET_RDMA_STATS,
+ OCRDMA_CMD_ALLOC_PD = 2,
+ OCRDMA_CMD_DEALLOC_PD = 3,
+
+ OCRDMA_CMD_CREATE_AH_TBL = 4,
+ OCRDMA_CMD_DELETE_AH_TBL = 5,
+
+ OCRDMA_CMD_CREATE_QP = 6,
+ OCRDMA_CMD_QUERY_QP = 7,
+ OCRDMA_CMD_MODIFY_QP = 8 ,
+ OCRDMA_CMD_DELETE_QP = 9,
+
+ OCRDMA_CMD_RSVD1 = 10,
+ OCRDMA_CMD_ALLOC_LKEY = 11,
+ OCRDMA_CMD_DEALLOC_LKEY = 12,
+ OCRDMA_CMD_REGISTER_NSMR = 13,
+ OCRDMA_CMD_REREGISTER_NSMR = 14,
+ OCRDMA_CMD_REGISTER_NSMR_CONT = 15,
+ OCRDMA_CMD_QUERY_NSMR = 16,
+ OCRDMA_CMD_ALLOC_MW = 17,
+ OCRDMA_CMD_QUERY_MW = 18,
+
+ OCRDMA_CMD_CREATE_SRQ = 19,
+ OCRDMA_CMD_QUERY_SRQ = 20,
+ OCRDMA_CMD_MODIFY_SRQ = 21,
+ OCRDMA_CMD_DELETE_SRQ = 22,
+
+ OCRDMA_CMD_ATTACH_MCAST = 23,
+ OCRDMA_CMD_DETACH_MCAST = 24,
+
+ OCRDMA_CMD_CREATE_RBQ = 25,
+ OCRDMA_CMD_DESTROY_RBQ = 26,
+
+ OCRDMA_CMD_GET_RDMA_STATS = 27,
OCRDMA_CMD_MAX
};
#define OCRDMA_MAX_QP 2048
#define OCRDMA_MAX_CQ 2048
-#define OCRDMA_MAX_STAG 8192
+#define OCRDMA_MAX_STAG 16384
enum {
OCRDMA_DB_RQ_OFFSET = 0xE0,
#define OCRDMA_ASYNC_RDMA_EVE_CODE 0x14
#define OCRDMA_ASYNC_GRP5_EVE_CODE 0x5
-#define OCRDMA_ASYNC_EVENT_PVID_STATE 0x3
+
+enum ocrdma_async_grp5_events {
+ OCRDMA_ASYNC_EVENT_QOS_VALUE = 0x01,
+ OCRDMA_ASYNC_EVENT_COS_VALUE = 0x02,
+ OCRDMA_ASYNC_EVENT_PVID_STATE = 0x03
+};
enum OCRDMA_ASYNC_EVENT_TYPE {
OCRDMA_CQ_ERROR = 0x00,
u32 max_ird_ord_per_qp;
u32 max_shared_ird_ord;
u32 max_mr;
- u32 max_mr_size_lo;
u32 max_mr_size_hi;
+ u32 max_mr_size_lo;
u32 max_num_mr_pbl;
u32 max_mw;
u32 max_fmr;
OCRDMA_FN_MODE_RDMA = 0x4
};
+enum {
+ OCRDMA_IF_TYPE_MASK = 0xFFFF0000,
+ OCRDMA_IF_TYPE_SHIFT = 0x10,
+ OCRDMA_PHY_TYPE_MASK = 0x0000FFFF,
+ OCRDMA_FUTURE_DETAILS_MASK = 0xFFFF0000,
+ OCRDMA_FUTURE_DETAILS_SHIFT = 0x10,
+ OCRDMA_EX_PHY_DETAILS_MASK = 0x0000FFFF,
+ OCRDMA_FSPEED_SUPP_MASK = 0xFFFF0000,
+ OCRDMA_FSPEED_SUPP_SHIFT = 0x10,
+ OCRDMA_ASPEED_SUPP_MASK = 0x0000FFFF
+};
+
struct ocrdma_get_phy_info_rsp {
struct ocrdma_mqe_hdr hdr;
struct ocrdma_mbx_rsp rsp;
- u16 phy_type;
- u16 interface_type;
+ u32 ityp_ptyp;
u32 misc_params;
- u16 ext_phy_details;
- u16 rsvd;
- u16 auto_speeds_supported;
- u16 fixed_speeds_supported;
+ u32 ftrdtl_exphydtl;
+ u32 fspeed_aspeed;
u32 future_use[2];
};
OCRDMA_PHY_SPEED_40GBPS = 0x20
};
+enum {
+ OCRDMA_PORT_NUM_MASK = 0x3F,
+ OCRDMA_PT_MASK = 0xC0,
+ OCRDMA_PT_SHIFT = 0x6,
+ OCRDMA_LINK_DUP_MASK = 0x0000FF00,
+ OCRDMA_LINK_DUP_SHIFT = 0x8,
+ OCRDMA_PHY_PS_MASK = 0x00FF0000,
+ OCRDMA_PHY_PS_SHIFT = 0x10,
+ OCRDMA_PHY_PFLT_MASK = 0xFF000000,
+ OCRDMA_PHY_PFLT_SHIFT = 0x18,
+ OCRDMA_QOS_LNKSP_MASK = 0xFFFF0000,
+ OCRDMA_QOS_LNKSP_SHIFT = 0x10,
+ OCRDMA_LLST_MASK = 0xFF,
+ OCRDMA_PLFC_MASK = 0x00000400,
+ OCRDMA_PLFC_SHIFT = 0x8,
+ OCRDMA_PLRFC_MASK = 0x00000200,
+ OCRDMA_PLRFC_SHIFT = 0x8,
+ OCRDMA_PLTFC_MASK = 0x00000100,
+ OCRDMA_PLTFC_SHIFT = 0x8
+};
struct ocrdma_get_link_speed_rsp {
struct ocrdma_mqe_hdr hdr;
struct ocrdma_mbx_rsp rsp;
- u8 pt_port_num;
- u8 link_duplex;
- u8 phys_port_speed;
- u8 phys_port_fault;
- u16 rsvd1;
- u16 qos_lnk_speed;
- u8 logical_lnk_status;
- u8 rsvd2[3];
+ u32 pflt_pps_ld_pnum;
+ u32 qos_lsp;
+ u32 res_lls;
};
enum {
u32 pgsz_pgcnt;
u32 ev_cnt_flags;
u32 eqn;
- u16 cqe_count;
- u16 pd_id;
+ u32 pdid_cqecnt;
u32 rsvd6;
struct ocrdma_pa pa[OCRDMA_CREATE_CQ_MAX_PAGES];
};
struct ocrdma_create_cq_cmd cmd;
};
+enum {
+ OCRDMA_CREATE_CQ_CMD_PDID_SHIFT = 0x10
+};
+
enum {
OCRDMA_CREATE_CQ_RSP_CQ_ID_MASK = 0xFFFF
};
enum {
OCRDMA_ALLOC_PD_ENABLE_DPP = BIT(16),
- OCRDMA_PD_MAX_DPP_ENABLED_QP = 8,
OCRDMA_DPP_PAGE_SIZE = 4096
};
struct ocrdma_rx_dbg_stats rx_dbg_stats;
} __packed;
+enum {
+ OCRDMA_HBA_ATTRB_EPROM_VER_LO_MASK = 0xFF,
+ OCRDMA_HBA_ATTRB_EPROM_VER_HI_MASK = 0xFF00,
+ OCRDMA_HBA_ATTRB_EPROM_VER_HI_SHIFT = 0x08,
+ OCRDMA_HBA_ATTRB_CDBLEN_MASK = 0xFFFF,
+ OCRDMA_HBA_ATTRB_ASIC_REV_MASK = 0xFF0000,
+ OCRDMA_HBA_ATTRB_ASIC_REV_SHIFT = 0x10,
+ OCRDMA_HBA_ATTRB_GUID0_MASK = 0xFF000000,
+ OCRDMA_HBA_ATTRB_GUID0_SHIFT = 0x18,
+ OCRDMA_HBA_ATTRB_GUID13_MASK = 0xFF,
+ OCRDMA_HBA_ATTRB_GUID14_MASK = 0xFF00,
+ OCRDMA_HBA_ATTRB_GUID14_SHIFT = 0x08,
+ OCRDMA_HBA_ATTRB_GUID15_MASK = 0xFF0000,
+ OCRDMA_HBA_ATTRB_GUID15_SHIFT = 0x10,
+ OCRDMA_HBA_ATTRB_PCNT_MASK = 0xFF000000,
+ OCRDMA_HBA_ATTRB_PCNT_SHIFT = 0x18,
+ OCRDMA_HBA_ATTRB_LDTOUT_MASK = 0xFFFF,
+ OCRDMA_HBA_ATTRB_ISCSI_VER_MASK = 0xFF0000,
+ OCRDMA_HBA_ATTRB_ISCSI_VER_SHIFT = 0x10,
+ OCRDMA_HBA_ATTRB_MFUNC_DEV_MASK = 0xFF000000,
+ OCRDMA_HBA_ATTRB_MFUNC_DEV_SHIFT = 0x18,
+ OCRDMA_HBA_ATTRB_CV_MASK = 0xFF,
+ OCRDMA_HBA_ATTRB_HBA_ST_MASK = 0xFF00,
+ OCRDMA_HBA_ATTRB_HBA_ST_SHIFT = 0x08,
+ OCRDMA_HBA_ATTRB_MAX_DOMS_MASK = 0xFF0000,
+ OCRDMA_HBA_ATTRB_MAX_DOMS_SHIFT = 0x10,
+ OCRDMA_HBA_ATTRB_PTNUM_MASK = 0x3F000000,
+ OCRDMA_HBA_ATTRB_PTNUM_SHIFT = 0x18,
+ OCRDMA_HBA_ATTRB_PT_MASK = 0xC0000000,
+ OCRDMA_HBA_ATTRB_PT_SHIFT = 0x1E,
+ OCRDMA_HBA_ATTRB_ISCSI_FET_MASK = 0xFF,
+ OCRDMA_HBA_ATTRB_ASIC_GEN_MASK = 0xFF00,
+ OCRDMA_HBA_ATTRB_ASIC_GEN_SHIFT = 0x08,
+ OCRDMA_HBA_ATTRB_PCI_VID_MASK = 0xFFFF,
+ OCRDMA_HBA_ATTRB_PCI_DID_MASK = 0xFFFF0000,
+ OCRDMA_HBA_ATTRB_PCI_DID_SHIFT = 0x10,
+ OCRDMA_HBA_ATTRB_PCI_SVID_MASK = 0xFFFF,
+ OCRDMA_HBA_ATTRB_PCI_SSID_MASK = 0xFFFF0000,
+ OCRDMA_HBA_ATTRB_PCI_SSID_SHIFT = 0x10,
+ OCRDMA_HBA_ATTRB_PCI_BUSNUM_MASK = 0xFF,
+ OCRDMA_HBA_ATTRB_PCI_DEVNUM_MASK = 0xFF00,
+ OCRDMA_HBA_ATTRB_PCI_DEVNUM_SHIFT = 0x08,
+ OCRDMA_HBA_ATTRB_PCI_FUNCNUM_MASK = 0xFF0000,
+ OCRDMA_HBA_ATTRB_PCI_FUNCNUM_SHIFT = 0x10,
+ OCRDMA_HBA_ATTRB_IF_TYPE_MASK = 0xFF000000,
+ OCRDMA_HBA_ATTRB_IF_TYPE_SHIFT = 0x18,
+ OCRDMA_HBA_ATTRB_NETFIL_MASK =0xFF
+};
struct mgmt_hba_attribs {
u8 flashrom_version_string[32];
u8 manufacturer_name[32];
u32 supported_modes;
- u32 rsvd0[3];
+ u32 rsvd_eprom_verhi_verlo;
+ u32 mbx_ds_ver;
+ u32 epfw_ds_ver;
u8 ncsi_ver_string[12];
u32 default_extended_timeout;
u8 controller_model_number[32];
u8 driver_version_string[32];
u8 fw_on_flash_version_string[32];
u32 functionalities_supported;
- u16 max_cdblength;
- u8 asic_revision;
- u8 generational_guid[16];
- u8 hba_port_count;
- u16 default_link_down_timeout;
- u8 iscsi_ver_min_max;
- u8 multifunction_device;
- u8 cache_valid;
- u8 hba_status;
- u8 max_domains_supported;
- u8 phy_port;
+ u32 guid0_asicrev_cdblen;
+ u8 generational_guid[12];
+ u32 portcnt_guid15;
+ u32 mfuncdev_iscsi_ldtout;
+ u32 ptpnum_maxdoms_hbast_cv;
u32 firmware_post_status;
u32 hba_mtu[8];
- u32 rsvd1[4];
+ u32 res_asicgen_iscsi_feaures;
+ u32 rsvd1[3];
};
struct mgmt_controller_attrib {
struct mgmt_hba_attribs hba_attribs;
- u16 pci_vendor_id;
- u16 pci_device_id;
- u16 pci_sub_vendor_id;
- u16 pci_sub_system_id;
- u8 pci_bus_number;
- u8 pci_device_number;
- u8 pci_function_number;
- u8 interface_type;
- u64 unique_identifier;
- u32 rsvd0[5];
+ u32 pci_did_vid;
+ u32 pci_ssid_svid;
+ u32 ityp_fnum_devnum_bnum;
+ u32 uid_hi;
+ u32 uid_lo;
+ u32 res_nnetfil;
+ u32 rsvd0[4];
};
struct ocrdma_get_ctrl_attribs_rsp {
struct mgmt_controller_attrib ctrl_attribs;
};
+#define OCRDMA_SUBSYS_DCBX 0x10
+
+enum OCRDMA_DCBX_OPCODE {
+ OCRDMA_CMD_GET_DCBX_CONFIG = 0x01
+};
+
+enum OCRDMA_DCBX_PARAM_TYPE {
+ OCRDMA_PARAMETER_TYPE_ADMIN = 0x00,
+ OCRDMA_PARAMETER_TYPE_OPER = 0x01,
+ OCRDMA_PARAMETER_TYPE_PEER = 0x02
+};
+
+enum OCRDMA_DCBX_APP_PROTO {
+ OCRDMA_APP_PROTO_ROCE = 0x8915
+};
+
+enum OCRDMA_DCBX_PROTO {
+ OCRDMA_PROTO_SELECT_L2 = 0x00,
+ OCRDMA_PROTO_SELECT_L4 = 0x01
+};
+
+enum OCRDMA_DCBX_APP_PARAM {
+ OCRDMA_APP_PARAM_APP_PROTO_MASK = 0xFFFF,
+ OCRDMA_APP_PARAM_PROTO_SEL_MASK = 0xFF,
+ OCRDMA_APP_PARAM_PROTO_SEL_SHIFT = 0x10,
+ OCRDMA_APP_PARAM_VALID_MASK = 0xFF,
+ OCRDMA_APP_PARAM_VALID_SHIFT = 0x18
+};
+
+enum OCRDMA_DCBX_STATE_FLAGS {
+ OCRDMA_STATE_FLAG_ENABLED = 0x01,
+ OCRDMA_STATE_FLAG_ADDVERTISED = 0x02,
+ OCRDMA_STATE_FLAG_WILLING = 0x04,
+ OCRDMA_STATE_FLAG_SYNC = 0x08,
+ OCRDMA_STATE_FLAG_UNSUPPORTED = 0x40000000,
+ OCRDMA_STATE_FLAG_NEG_FAILD = 0x80000000
+};
+
+enum OCRDMA_TCV_AEV_OPV_ST {
+ OCRDMA_DCBX_TC_SUPPORT_MASK = 0xFF,
+ OCRDMA_DCBX_TC_SUPPORT_SHIFT = 0x18,
+ OCRDMA_DCBX_APP_ENTRY_SHIFT = 0x10,
+ OCRDMA_DCBX_OP_PARAM_SHIFT = 0x08,
+ OCRDMA_DCBX_STATE_MASK = 0xFF
+};
+
+struct ocrdma_app_parameter {
+ u32 valid_proto_app;
+ u32 oui;
+ u32 app_prio[2];
+};
+
+struct ocrdma_dcbx_cfg {
+ u32 tcv_aev_opv_st;
+ u32 tc_state;
+ u32 pfc_state;
+ u32 qcn_state;
+ u32 appl_state;
+ u32 ll_state;
+ u32 tc_bw[2];
+ u32 tc_prio[8];
+ u32 pfc_prio[2];
+ struct ocrdma_app_parameter app_param[15];
+};
+
+struct ocrdma_get_dcbx_cfg_req {
+ struct ocrdma_mbx_hdr hdr;
+ u32 param_type;
+} __packed;
+
+struct ocrdma_get_dcbx_cfg_rsp {
+ struct ocrdma_mbx_rsp hdr;
+ struct ocrdma_dcbx_cfg cfg;
+} __packed;
#endif /* __OCRDMA_SLI_H__ */
memcpy(&attr->fw_ver, &dev->attr.fw_ver[0],
min(sizeof(dev->attr.fw_ver), sizeof(attr->fw_ver)));
ocrdma_get_guid(dev, (u8 *)&attr->sys_image_guid);
- attr->max_mr_size = ~0ull;
+ attr->max_mr_size = dev->attr.max_mr_size;
attr->page_size_cap = 0xffff000;
attr->vendor_id = dev->nic_info.pdev->vendor;
attr->vendor_part_id = dev->nic_info.pdev->device;
- attr->hw_ver = 0;
+ attr->hw_ver = dev->asic_id;
attr->max_qp = dev->attr.max_qp;
attr->max_ah = OCRDMA_MAX_AH;
attr->max_qp_wr = dev->attr.max_wqe;
pd->dpp_enabled =
ocrdma_get_asic_type(dev) == OCRDMA_ASIC_GEN_SKH_R;
pd->num_dpp_qp =
- pd->dpp_enabled ? OCRDMA_PD_MAX_DPP_ENABLED_QP : 0;
+ pd->dpp_enabled ? (dev->nic_info.db_page_size /
+ dev->attr.wqe_size) : 0;
}
retry:
struct ocrdma_pd *pd = uctx->cntxt_pd;
struct ocrdma_dev *dev = get_ocrdma_dev(pd->ibpd.device);
- BUG_ON(uctx->pd_in_use);
+ if (uctx->pd_in_use) {
+ pr_err("%s(%d) Freeing in use pdid=0x%x.\n",
+ __func__, dev->id, pd->id);
+ }
uctx->cntxt_pd = NULL;
status = _ocrdma_dealloc_pd(dev, pd);
return status;
if (mr->umem)
ib_umem_release(mr->umem);
kfree(mr);
+
+ /* Don't stop cleanup, in case FW is unresponsive */
+ if (dev->mqe_ctx.fw_error_state) {
+ status = 0;
+ pr_err("%s(%d) fw not responding.\n",
+ __func__, dev->id);
+ }
return status;
}
}
while (wr) {
+ if (qp->qp_type == IB_QPT_UD &&
+ (wr->opcode != IB_WR_SEND &&
+ wr->opcode != IB_WR_SEND_WITH_IMM)) {
+ *bad_wr = wr;
+ status = -EINVAL;
+ break;
+ }
if (ocrdma_hwq_free_cnt(&qp->sq) == 0 ||
wr->num_sge > qp->sq.max_sges) {
*bad_wr = wr;
*stop = true;
expand = false;
}
+ } else if (is_hw_sq_empty(qp)) {
+ /* Do nothing */
+ expand = false;
+ *polled = false;
+ *stop = false;
} else {
*polled = true;
expand = ocrdma_update_err_scqe(ibwc, cqe, qp, status);
*stop = true;
expand = false;
}
+ } else if (is_hw_rq_empty(qp)) {
+ /* Do nothing */
+ expand = false;
+ *polled = false;
+ *stop = false;
} else {
*polled = true;
expand = ocrdma_update_err_rcqe(ibwc, cqe, qp, status);
#define DRIVER_LOAD_MSG "Intel " QIB_DRV_NAME " loaded: "
#define PFX QIB_DRV_NAME ": "
-static DEFINE_PCI_DEVICE_TABLE(qib_pci_tbl) = {
+static const struct pci_device_id qib_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_PATHSCALE, PCI_DEVICE_ID_QLOGIC_IB_6120) },
{ PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_IB_7220) },
{ PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_IB_7322) },
ibp = &dd->pport[p].ibport_data;
agent = ib_register_mad_agent(&dev->ibdev, p + 1, IB_QPT_SMI,
NULL, 0, send_handler,
- NULL, NULL);
+ NULL, NULL, 0);
if (IS_ERR(agent)) {
ret = PTR_ERR(agent);
goto err;
/* Start of PCI section */
-static DEFINE_PCI_DEVICE_TABLE(usnic_ib_pci_ids) = {
+static const struct pci_device_id usnic_ib_pci_ids[] = {
{PCI_DEVICE(PCI_VENDOR_ID_CISCO, PCI_DEVICE_ID_CISCO_VIC_USPACE_NIC)},
{0,}
};
IPOIB_FLAG_INITIALIZED = 1,
IPOIB_FLAG_ADMIN_UP = 2,
IPOIB_PKEY_ASSIGNED = 3,
- IPOIB_PKEY_STOP = 4,
IPOIB_FLAG_SUBINTERFACE = 5,
IPOIB_MCAST_RUN = 6,
IPOIB_STOP_REAPER = 7,
struct list_head multicast_list;
struct rb_root multicast_tree;
- struct delayed_work pkey_poll_task;
struct delayed_work mcast_task;
struct work_struct carrier_on_task;
struct work_struct flush_light;
void ipoib_pkey_event(struct work_struct *work);
void ipoib_ib_dev_cleanup(struct net_device *dev);
-int ipoib_ib_dev_open(struct net_device *dev);
+int ipoib_ib_dev_open(struct net_device *dev, int flush);
int ipoib_ib_dev_up(struct net_device *dev);
int ipoib_ib_dev_down(struct net_device *dev, int flush);
int ipoib_ib_dev_stop(struct net_device *dev, int flush);
+void ipoib_pkey_dev_check_presence(struct net_device *dev);
int ipoib_dev_init(struct net_device *dev, struct ib_device *ca, int port);
void ipoib_dev_cleanup(struct net_device *dev);
void ipoib_setup(struct net_device *dev);
-void ipoib_pkey_poll(struct work_struct *work);
-int ipoib_pkey_dev_delay_open(struct net_device *dev);
+void ipoib_pkey_open(struct ipoib_dev_priv *priv);
void ipoib_drain_cq(struct net_device *dev);
void ipoib_set_ethtool_ops(struct net_device *dev);
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
- if (priv->mcg_dentry)
- debugfs_remove(priv->mcg_dentry);
- if (priv->path_dentry)
- debugfs_remove(priv->path_dentry);
+ debugfs_remove(priv->mcg_dentry);
+ debugfs_remove(priv->path_dentry);
}
int ipoib_register_debugfs(void)
drain_tx_cq((struct net_device *)ctx);
}
-int ipoib_ib_dev_open(struct net_device *dev)
+int ipoib_ib_dev_open(struct net_device *dev, int flush)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
int ret;
- if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &priv->pkey_index)) {
- ipoib_warn(priv, "P_Key 0x%04x not found\n", priv->pkey);
- clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
+ ipoib_pkey_dev_check_presence(dev);
+
+ if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
+ ipoib_warn(priv, "P_Key 0x%04x is %s\n", priv->pkey,
+ (!(priv->pkey & 0x7fff) ? "Invalid" : "not found"));
return -1;
}
- set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
ret = ipoib_init_qp(dev);
if (ret) {
dev_stop:
if (!test_and_set_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
napi_enable(&priv->napi);
- ipoib_ib_dev_stop(dev, 1);
+ ipoib_ib_dev_stop(dev, flush);
return -1;
}
-static void ipoib_pkey_dev_check_presence(struct net_device *dev)
+void ipoib_pkey_dev_check_presence(struct net_device *dev)
{
struct ipoib_dev_priv *priv = netdev_priv(dev);
- u16 pkey_index = 0;
- if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &pkey_index))
+ if (!(priv->pkey & 0x7fff) ||
+ ib_find_pkey(priv->ca, priv->port, priv->pkey,
+ &priv->pkey_index))
clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
else
set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
clear_bit(IPOIB_FLAG_OPER_UP, &priv->flags);
netif_carrier_off(dev);
- /* Shutdown the P_Key thread if still active */
- if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
- mutex_lock(&pkey_mutex);
- set_bit(IPOIB_PKEY_STOP, &priv->flags);
- cancel_delayed_work_sync(&priv->pkey_poll_task);
- mutex_unlock(&pkey_mutex);
- }
-
ipoib_mcast_stop_thread(dev, flush);
ipoib_mcast_dev_flush(dev);
(unsigned long) dev);
if (dev->flags & IFF_UP) {
- if (ipoib_ib_dev_open(dev)) {
+ if (ipoib_ib_dev_open(dev, 1)) {
ipoib_transport_dev_cleanup(dev);
return -ENODEV;
}
return 1;
}
+/*
+ * returns 0 if pkey value was found in a different slot.
+ */
+static inline int update_child_pkey(struct ipoib_dev_priv *priv)
+{
+ u16 old_index = priv->pkey_index;
+
+ priv->pkey_index = 0;
+ ipoib_pkey_dev_check_presence(priv->dev);
+
+ if (test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags) &&
+ (old_index == priv->pkey_index))
+ return 1;
+ return 0;
+}
static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv,
enum ipoib_flush_level level)
{
struct ipoib_dev_priv *cpriv;
struct net_device *dev = priv->dev;
- u16 new_index;
int result;
down_read(&priv->vlan_rwsem);
up_read(&priv->vlan_rwsem);
- if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) {
- /* for non-child devices must check/update the pkey value here */
- if (level == IPOIB_FLUSH_HEAVY &&
- !test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags))
- update_parent_pkey(priv);
+ if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags) &&
+ level != IPOIB_FLUSH_HEAVY) {
ipoib_dbg(priv, "Not flushing - IPOIB_FLAG_INITIALIZED not set.\n");
return;
}
if (!test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) {
+ /* interface is down. update pkey and leave. */
+ if (level == IPOIB_FLUSH_HEAVY) {
+ if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags))
+ update_parent_pkey(priv);
+ else
+ update_child_pkey(priv);
+ }
ipoib_dbg(priv, "Not flushing - IPOIB_FLAG_ADMIN_UP not set.\n");
return;
}
* (parent) devices should always takes what present in pkey index 0
*/
if (test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) {
- if (ib_find_pkey(priv->ca, priv->port, priv->pkey, &new_index)) {
- clear_bit(IPOIB_PKEY_ASSIGNED, &priv->flags);
- ipoib_ib_dev_down(dev, 0);
- ipoib_ib_dev_stop(dev, 0);
- if (ipoib_pkey_dev_delay_open(dev))
- return;
- }
- /* restart QP only if P_Key index is changed */
- if (test_and_set_bit(IPOIB_PKEY_ASSIGNED, &priv->flags) &&
- new_index == priv->pkey_index) {
+ result = update_child_pkey(priv);
+ if (result) {
+ /* restart QP only if P_Key index is changed */
ipoib_dbg(priv, "Not flushing - P_Key index not changed.\n");
return;
}
- priv->pkey_index = new_index;
+
} else {
result = update_parent_pkey(priv);
/* restart QP only if P_Key value changed */
ipoib_ib_dev_down(dev, 0);
if (level == IPOIB_FLUSH_HEAVY) {
- ipoib_ib_dev_stop(dev, 0);
- ipoib_ib_dev_open(dev);
+ if (test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags))
+ ipoib_ib_dev_stop(dev, 0);
+ if (ipoib_ib_dev_open(dev, 0) != 0)
+ return;
+ if (netif_queue_stopped(dev))
+ netif_start_queue(dev);
}
/*
ipoib_transport_dev_cleanup(dev);
}
-/*
- * Delayed P_Key Assigment Interim Support
- *
- * The following is initial implementation of delayed P_Key assigment
- * mechanism. It is using the same approach implemented for the multicast
- * group join. The single goal of this implementation is to quickly address
- * Bug #2507. This implementation will probably be removed when the P_Key
- * change async notification is available.
- */
-
-void ipoib_pkey_poll(struct work_struct *work)
-{
- struct ipoib_dev_priv *priv =
- container_of(work, struct ipoib_dev_priv, pkey_poll_task.work);
- struct net_device *dev = priv->dev;
-
- ipoib_pkey_dev_check_presence(dev);
-
- if (test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags))
- ipoib_open(dev);
- else {
- mutex_lock(&pkey_mutex);
- if (!test_bit(IPOIB_PKEY_STOP, &priv->flags))
- queue_delayed_work(ipoib_workqueue,
- &priv->pkey_poll_task,
- HZ);
- mutex_unlock(&pkey_mutex);
- }
-}
-
-int ipoib_pkey_dev_delay_open(struct net_device *dev)
-{
- struct ipoib_dev_priv *priv = netdev_priv(dev);
-
- /* Look for the interface pkey value in the IB Port P_Key table and */
- /* set the interface pkey assigment flag */
- ipoib_pkey_dev_check_presence(dev);
- /* P_Key value not assigned yet - start polling */
- if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags)) {
- mutex_lock(&pkey_mutex);
- clear_bit(IPOIB_PKEY_STOP, &priv->flags);
- queue_delayed_work(ipoib_workqueue,
- &priv->pkey_poll_task,
- HZ);
- mutex_unlock(&pkey_mutex);
- return 1;
- }
-
- return 0;
-}
set_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags);
- if (ipoib_pkey_dev_delay_open(dev))
- return 0;
-
- if (ipoib_ib_dev_open(dev))
+ if (ipoib_ib_dev_open(dev, 1)) {
+ if (!test_bit(IPOIB_PKEY_ASSIGNED, &priv->flags))
+ return 0;
goto err_disable;
+ }
if (ipoib_ib_dev_up(dev))
goto err_stop;
INIT_LIST_HEAD(&priv->dead_ahs);
INIT_LIST_HEAD(&priv->multicast_list);
- INIT_DELAYED_WORK(&priv->pkey_poll_task, ipoib_pkey_poll);
INIT_DELAYED_WORK(&priv->mcast_task, ipoib_mcast_join_task);
INIT_WORK(&priv->carrier_on_task, ipoib_mcast_carrier_on_task);
INIT_WORK(&priv->flush_light, ipoib_ib_dev_flush_light);
struct iser_conn *ib_conn;
struct iscsi_endpoint *ep;
- ep = iscsi_create_endpoint(sizeof(*ib_conn));
+ ep = iscsi_create_endpoint(0);
if (!ep)
return ERR_PTR(-ENOMEM);
- ib_conn = ep->dd_data;
+ ib_conn = kzalloc(sizeof(*ib_conn), GFP_KERNEL);
+ if (!ib_conn) {
+ err = -ENOMEM;
+ goto failure;
+ }
+
+ ep->dd_data = ib_conn;
ib_conn->ep = ep;
iser_conn_init(ib_conn);
- err = iser_connect(ib_conn, NULL, (struct sockaddr_in *)dst_addr,
- non_blocking);
+ err = iser_connect(ib_conn, NULL, dst_addr, non_blocking);
if (err)
- return ERR_PTR(err);
+ goto failure;
return ep;
+failure:
+ iscsi_destroy_endpoint(ep);
+ return ERR_PTR(err);
}
static int
int rc;
ib_conn = ep->dd_data;
- rc = wait_event_interruptible_timeout(ib_conn->wait,
- ib_conn->state == ISER_CONN_UP,
- msecs_to_jiffies(timeout_ms));
-
+ rc = wait_for_completion_interruptible_timeout(&ib_conn->up_completion,
+ msecs_to_jiffies(timeout_ms));
/* if conn establishment failed, return error code to iscsi */
- if (!rc &&
- (ib_conn->state == ISER_CONN_TERMINATING ||
- ib_conn->state == ISER_CONN_DOWN))
- rc = -1;
+ if (rc == 0) {
+ mutex_lock(&ib_conn->state_mutex);
+ if (ib_conn->state == ISER_CONN_TERMINATING ||
+ ib_conn->state == ISER_CONN_DOWN)
+ rc = -1;
+ mutex_unlock(&ib_conn->state_mutex);
+ }
iser_info("ib conn %p rc = %d\n", ib_conn, rc);
ib_conn = ep->dd_data;
iser_info("ep %p ib conn %p state %d\n", ep, ib_conn, ib_conn->state);
+ mutex_lock(&ib_conn->state_mutex);
iser_conn_terminate(ib_conn);
/*
- * if iser_conn and iscsi_conn are bound, we must wait iscsi_conn_stop
- * call and ISER_CONN_DOWN state before freeing the iser resources.
- * otherwise we are safe to free resources immediately.
+ * if iser_conn and iscsi_conn are bound, we must wait for
+ * iscsi_conn_stop and flush errors completion before freeing
+ * the iser resources. Otherwise we are safe to free resources
+ * immediately.
*/
if (ib_conn->iscsi_conn) {
INIT_WORK(&ib_conn->release_work, iser_release_work);
queue_work(release_wq, &ib_conn->release_work);
+ mutex_unlock(&ib_conn->state_mutex);
} else {
+ ib_conn->state = ISER_CONN_DOWN;
+ mutex_unlock(&ib_conn->state_mutex);
iser_conn_release(ib_conn);
}
+ iscsi_destroy_endpoint(ep);
}
static umode_t iser_attr_is_visible(int param_type, int param)
struct iser_device *device; /* device context */
struct rdma_cm_id *cma_id; /* CMA ID */
struct ib_qp *qp; /* QP */
- wait_queue_head_t wait; /* waitq for conn/disconn */
unsigned qp_max_recv_dtos; /* num of rx buffers */
unsigned qp_max_recv_dtos_mask; /* above minus 1 */
unsigned min_posted_rx; /* qp_max_recv_dtos >> 2 */
char name[ISER_OBJECT_NAME_SIZE];
struct work_struct release_work;
struct completion stop_completion;
+ struct mutex state_mutex;
+ struct completion flush_completion;
+ struct completion up_completion;
struct list_head conn_list; /* entry in ig conn list */
char *login_buf;
enum iser_data_dir cmd_dir);
int iser_connect(struct iser_conn *ib_conn,
- struct sockaddr_in *src_addr,
- struct sockaddr_in *dst_addr,
+ struct sockaddr *src_addr,
+ struct sockaddr *dst_addr,
int non_blocking);
int iser_reg_page_vec(struct iser_conn *ib_conn,
}
/**
- * releases the QP objects, returns 0 on success,
- * -1 on failure
+ * releases the QP object
*/
-static int iser_free_ib_conn_res(struct iser_conn *ib_conn)
+static void iser_free_ib_conn_res(struct iser_conn *ib_conn)
{
int cq_index;
BUG_ON(ib_conn == NULL);
}
ib_conn->qp = NULL;
-
- return 0;
}
/**
mutex_unlock(&ig.device_list_mutex);
}
+/**
+ * Called with state mutex held
+ **/
static int iser_conn_state_comp_exch(struct iser_conn *ib_conn,
enum iser_ib_conn_state comp,
enum iser_ib_conn_state exch)
{
int ret;
- spin_lock_bh(&ib_conn->lock);
if ((ret = (ib_conn->state == comp)))
ib_conn->state = exch;
- spin_unlock_bh(&ib_conn->lock);
return ret;
}
void iser_release_work(struct work_struct *work)
{
struct iser_conn *ib_conn;
+ int rc;
ib_conn = container_of(work, struct iser_conn, release_work);
/* wait for .conn_stop callback */
- wait_for_completion(&ib_conn->stop_completion);
+ rc = wait_for_completion_timeout(&ib_conn->stop_completion, 30 * HZ);
+ WARN_ON(rc == 0);
/* wait for the qp`s post send and post receive buffers to empty */
- wait_event_interruptible(ib_conn->wait,
- ib_conn->state == ISER_CONN_DOWN);
+ rc = wait_for_completion_timeout(&ib_conn->flush_completion, 30 * HZ);
+ WARN_ON(rc == 0);
+
+ ib_conn->state = ISER_CONN_DOWN;
+
+ mutex_lock(&ib_conn->state_mutex);
+ ib_conn->state = ISER_CONN_DOWN;
+ mutex_unlock(&ib_conn->state_mutex);
iser_conn_release(ib_conn);
}
{
struct iser_device *device = ib_conn->device;
- BUG_ON(ib_conn->state == ISER_CONN_UP);
-
mutex_lock(&ig.connlist_mutex);
list_del(&ib_conn->conn_list);
mutex_unlock(&ig.connlist_mutex);
+
+ mutex_lock(&ib_conn->state_mutex);
+ BUG_ON(ib_conn->state != ISER_CONN_DOWN);
+
iser_free_rx_descriptors(ib_conn);
iser_free_ib_conn_res(ib_conn);
ib_conn->device = NULL;
/* on EVENT_ADDR_ERROR there's no device yet for this conn */
if (device != NULL)
iser_device_try_release(device);
+ mutex_unlock(&ib_conn->state_mutex);
+
/* if cma handler context, the caller actually destroy the id */
if (ib_conn->cma_id != NULL) {
rdma_destroy_id(ib_conn->cma_id);
ib_conn->cma_id = NULL;
}
- iscsi_destroy_endpoint(ib_conn->ep);
+ kfree(ib_conn);
}
/**
ib_conn,err);
}
+/**
+ * Called with state mutex held
+ **/
static void iser_connect_error(struct rdma_cm_id *cma_id)
{
struct iser_conn *ib_conn;
ib_conn = (struct iser_conn *)cma_id->context;
-
ib_conn->state = ISER_CONN_DOWN;
- wake_up_interruptible(&ib_conn->wait);
}
+/**
+ * Called with state mutex held
+ **/
static void iser_addr_handler(struct rdma_cm_id *cma_id)
{
struct iser_device *device;
struct iser_conn *ib_conn;
int ret;
+ ib_conn = (struct iser_conn *)cma_id->context;
+ if (ib_conn->state != ISER_CONN_PENDING)
+ /* bailout */
+ return;
+
device = iser_device_find_by_ib_device(cma_id);
if (!device) {
iser_err("device lookup/creation failed\n");
return;
}
- ib_conn = (struct iser_conn *)cma_id->context;
ib_conn->device = device;
/* connection T10-PI support */
}
}
+/**
+ * Called with state mutex held
+ **/
static void iser_route_handler(struct rdma_cm_id *cma_id)
{
struct rdma_conn_param conn_param;
int ret;
struct iser_cm_hdr req_hdr;
+ struct iser_conn *ib_conn = (struct iser_conn *)cma_id->context;
+ struct iser_device *device = ib_conn->device;
+
+ if (ib_conn->state != ISER_CONN_PENDING)
+ /* bailout */
+ return;
ret = iser_create_ib_conn_res((struct iser_conn *)cma_id->context);
if (ret)
goto failure;
memset(&conn_param, 0, sizeof conn_param);
- conn_param.responder_resources = 4;
+ conn_param.responder_resources = device->dev_attr.max_qp_rd_atom;
conn_param.initiator_depth = 1;
conn_param.retry_count = 7;
conn_param.rnr_retry_count = 6;
struct ib_qp_attr attr;
struct ib_qp_init_attr init_attr;
+ ib_conn = (struct iser_conn *)cma_id->context;
+ if (ib_conn->state != ISER_CONN_PENDING)
+ /* bailout */
+ return;
+
(void)ib_query_qp(cma_id->qp, &attr, ~0, &init_attr);
iser_info("remote qpn:%x my qpn:%x\n", attr.dest_qp_num, cma_id->qp->qp_num);
- ib_conn = (struct iser_conn *)cma_id->context;
- if (iser_conn_state_comp_exch(ib_conn, ISER_CONN_PENDING, ISER_CONN_UP))
- wake_up_interruptible(&ib_conn->wait);
+ ib_conn->state = ISER_CONN_UP;
+ complete(&ib_conn->up_completion);
}
static void iser_disconnected_handler(struct rdma_cm_id *cma_id)
iser_err("iscsi_iser connection isn't bound\n");
}
- /* Complete the termination process if no posts are pending */
+ /* Complete the termination process if no posts are pending. This code
+ * block also exists in iser_handle_comp_error(), but it is needed here
+ * for cases of no flushes at all, e.g. discovery over rdma.
+ */
if (ib_conn->post_recv_buf_count == 0 &&
(atomic_read(&ib_conn->post_send_buf_count) == 0)) {
- ib_conn->state = ISER_CONN_DOWN;
- wake_up_interruptible(&ib_conn->wait);
+ complete(&ib_conn->flush_completion);
}
}
static int iser_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *event)
{
+ struct iser_conn *ib_conn;
+
+ ib_conn = (struct iser_conn *)cma_id->context;
iser_info("event %d status %d conn %p id %p\n",
event->event, event->status, cma_id->context, cma_id);
+ mutex_lock(&ib_conn->state_mutex);
switch (event->event) {
case RDMA_CM_EVENT_ADDR_RESOLVED:
iser_addr_handler(cma_id);
case RDMA_CM_EVENT_DISCONNECTED:
case RDMA_CM_EVENT_DEVICE_REMOVAL:
case RDMA_CM_EVENT_ADDR_CHANGE:
+ case RDMA_CM_EVENT_TIMEWAIT_EXIT:
iser_disconnected_handler(cma_id);
break;
default:
iser_err("Unexpected RDMA CM event (%d)\n", event->event);
break;
}
+ mutex_unlock(&ib_conn->state_mutex);
return 0;
}
void iser_conn_init(struct iser_conn *ib_conn)
{
ib_conn->state = ISER_CONN_INIT;
- init_waitqueue_head(&ib_conn->wait);
ib_conn->post_recv_buf_count = 0;
atomic_set(&ib_conn->post_send_buf_count, 0);
init_completion(&ib_conn->stop_completion);
+ init_completion(&ib_conn->flush_completion);
+ init_completion(&ib_conn->up_completion);
INIT_LIST_HEAD(&ib_conn->conn_list);
spin_lock_init(&ib_conn->lock);
+ mutex_init(&ib_conn->state_mutex);
}
/**
* sleeps until the connection is established or rejected
*/
int iser_connect(struct iser_conn *ib_conn,
- struct sockaddr_in *src_addr,
- struct sockaddr_in *dst_addr,
+ struct sockaddr *src_addr,
+ struct sockaddr *dst_addr,
int non_blocking)
{
- struct sockaddr *src, *dst;
int err = 0;
- sprintf(ib_conn->name, "%pI4:%d",
- &dst_addr->sin_addr.s_addr, dst_addr->sin_port);
+ mutex_lock(&ib_conn->state_mutex);
+
+ sprintf(ib_conn->name, "%pISp", dst_addr);
+
+ iser_info("connecting to: %s\n", ib_conn->name);
/* the device is known only --after-- address resolution */
ib_conn->device = NULL;
- iser_info("connecting to: %pI4, port 0x%x\n",
- &dst_addr->sin_addr, dst_addr->sin_port);
-
ib_conn->state = ISER_CONN_PENDING;
ib_conn->cma_id = rdma_create_id(iser_cma_handler,
goto id_failure;
}
- src = (struct sockaddr *)src_addr;
- dst = (struct sockaddr *)dst_addr;
- err = rdma_resolve_addr(ib_conn->cma_id, src, dst, 1000);
+ err = rdma_resolve_addr(ib_conn->cma_id, src_addr, dst_addr, 1000);
if (err) {
iser_err("rdma_resolve_addr failed: %d\n", err);
goto addr_failure;
}
if (!non_blocking) {
- wait_event_interruptible(ib_conn->wait,
- (ib_conn->state != ISER_CONN_PENDING));
+ wait_for_completion_interruptible(&ib_conn->up_completion);
if (ib_conn->state != ISER_CONN_UP) {
err = -EIO;
goto connect_failure;
}
}
+ mutex_unlock(&ib_conn->state_mutex);
mutex_lock(&ig.connlist_mutex);
list_add(&ib_conn->conn_list, &ig.connlist);
addr_failure:
ib_conn->state = ISER_CONN_DOWN;
connect_failure:
+ mutex_unlock(&ib_conn->state_mutex);
iser_conn_release(ib_conn);
return err;
}
if (ib_conn->post_recv_buf_count == 0 &&
atomic_read(&ib_conn->post_send_buf_count) == 0) {
- /* getting here when the state is UP means that the conn is *
- * being terminated asynchronously from the iSCSI layer's *
- * perspective. */
- if (iser_conn_state_comp_exch(ib_conn, ISER_CONN_UP,
- ISER_CONN_TERMINATING))
+ /**
+ * getting here when the state is UP means that the conn is
+ * being terminated asynchronously from the iSCSI layer's
+ * perspective. It is safe to peek at the connection state
+ * since iscsi_conn_failure is allowed to be called twice.
+ **/
+ if (ib_conn->state == ISER_CONN_UP)
iscsi_conn_failure(ib_conn->iscsi_conn,
ISCSI_ERR_CONN_FAILED);
/* no more non completed posts to the QP, complete the
* termination process w.o worrying on disconnect event */
- ib_conn->state = ISER_CONN_DOWN;
- wake_up_interruptible(&ib_conn->wait);
+ complete(&ib_conn->flush_completion);
}
}
static int srp_cm_handler(struct ib_cm_id *cm_id, struct ib_cm_event *event);
static struct scsi_transport_template *ib_srp_transport_template;
+static struct workqueue_struct *srp_remove_wq;
static struct ib_client srp_client = {
.name = "srp",
spin_unlock_irq(&target->lock);
if (changed)
- queue_work(system_long_wq, &target->remove_work);
+ queue_work(srp_remove_wq, &target->remove_work);
return changed;
}
SCSI_SENSE_BUFFERSIZE));
}
- if (rsp->flags & (SRP_RSP_FLAG_DOOVER | SRP_RSP_FLAG_DOUNDER))
- scsi_set_resid(scmnd, be32_to_cpu(rsp->data_out_res_cnt));
- else if (rsp->flags & (SRP_RSP_FLAG_DIOVER | SRP_RSP_FLAG_DIUNDER))
+ if (unlikely(rsp->flags & SRP_RSP_FLAG_DIUNDER))
scsi_set_resid(scmnd, be32_to_cpu(rsp->data_in_res_cnt));
+ else if (unlikely(rsp->flags & SRP_RSP_FLAG_DIOVER))
+ scsi_set_resid(scmnd, -be32_to_cpu(rsp->data_in_res_cnt));
+ else if (unlikely(rsp->flags & SRP_RSP_FLAG_DOUNDER))
+ scsi_set_resid(scmnd, be32_to_cpu(rsp->data_out_res_cnt));
+ else if (unlikely(rsp->flags & SRP_RSP_FLAG_DOOVER))
+ scsi_set_resid(scmnd, -be32_to_cpu(rsp->data_out_res_cnt));
srp_free_req(target, req, scmnd,
be32_to_cpu(rsp->req_lim_delta));
spin_unlock(&host->target_lock);
/*
- * Wait for target port removal tasks.
+ * Wait for tl_err and target port removal tasks.
*/
flush_workqueue(system_long_wq);
+ flush_workqueue(srp_remove_wq);
kfree(host);
}
indirect_sg_entries = cmd_sg_entries;
}
+ srp_remove_wq = create_workqueue("srp_remove");
+ if (!srp_remove_wq) {
+ ret = -ENOMEM;
+ goto out;
+ }
+
+ ret = -ENOMEM;
ib_srp_transport_template =
srp_attach_transport(&ib_srp_transport_functions);
if (!ib_srp_transport_template)
- return -ENOMEM;
+ goto destroy_wq;
ret = class_register(&srp_class);
if (ret) {
pr_err("couldn't register class infiniband_srp\n");
- srp_release_transport(ib_srp_transport_template);
- return ret;
+ goto release_tr;
}
ib_sa_register_client(&srp_sa_client);
ret = ib_register_client(&srp_client);
if (ret) {
pr_err("couldn't register IB client\n");
- srp_release_transport(ib_srp_transport_template);
- ib_sa_unregister_client(&srp_sa_client);
- class_unregister(&srp_class);
- return ret;
+ goto unreg_sa;
}
- return 0;
+out:
+ return ret;
+
+unreg_sa:
+ ib_sa_unregister_client(&srp_sa_client);
+ class_unregister(&srp_class);
+
+release_tr:
+ srp_release_transport(ib_srp_transport_template);
+
+destroy_wq:
+ destroy_workqueue(srp_remove_wq);
+ goto out;
}
static void __exit srp_cleanup_module(void)
ib_sa_unregister_client(&srp_sa_client);
class_unregister(&srp_class);
srp_release_transport(ib_srp_transport_template);
+ destroy_workqueue(srp_remove_wq);
}
module_init(srp_init_module);
case IB_EVENT_PKEY_CHANGE:
case IB_EVENT_SM_CHANGE:
case IB_EVENT_CLIENT_REREGISTER:
+ case IB_EVENT_GID_CHANGE:
/* Refresh port data asynchronously. */
if (event->element.port_num <= sdev->device->phys_port_cnt) {
sport = &sdev->port[event->element.port_num - 1];
®_req, 0,
srpt_mad_send_handler,
srpt_mad_recv_handler,
- sport);
+ sport, 0);
if (IS_ERR(sport->mad_agent)) {
ret = PTR_ERR(sport->mad_agent);
sport->mad_agent = NULL;
}
EXPORT_SYMBOL(input_mt_report_pointer_emulation);
+/**
+ * input_mt_drop_unused() - Inactivate slots not seen in this frame
+ * @dev: input device with allocated MT slots
+ *
+ * Lift all slots not seen since the last call to this function.
+ */
+void input_mt_drop_unused(struct input_dev *dev)
+{
+ struct input_mt *mt = dev->mt;
+ int i;
+
+ if (!mt)
+ return;
+
+ for (i = 0; i < mt->num_slots; i++) {
+ if (!input_mt_is_used(mt, &mt->slots[i])) {
+ input_mt_slot(dev, i);
+ input_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
+ }
+ }
+
+ mt->frame++;
+}
+EXPORT_SYMBOL(input_mt_drop_unused);
+
/**
* input_mt_sync_frame() - synchronize mt frame
* @dev: input device with allocated MT slots
void input_mt_sync_frame(struct input_dev *dev)
{
struct input_mt *mt = dev->mt;
- struct input_mt_slot *s;
bool use_count = false;
if (!mt)
return;
- if (mt->flags & INPUT_MT_DROP_UNUSED) {
- for (s = mt->slots; s != mt->slots + mt->num_slots; s++) {
- if (input_mt_is_used(mt, s))
- continue;
- input_mt_slot(dev, s - mt->slots);
- input_event(dev, EV_ABS, ABS_MT_TRACKING_ID, -1);
- }
- }
+ if (mt->flags & INPUT_MT_DROP_UNUSED)
+ input_mt_drop_unused(dev);
if ((mt->flags & INPUT_MT_POINTER) && !(mt->flags & INPUT_MT_SEMI_MT))
use_count = true;
input_mt_report_pointer_emulation(dev, use_count);
-
- mt->frame++;
}
EXPORT_SYMBOL(input_mt_sync_frame);
#define GET_TIME(x) rdtscl(x)
#define DELTA(x,y) ((y)-(x))
#define TIME_NAME "TSC"
-#elif defined(__alpha__) || defined(CONFIG_MN10300) || defined(CONFIG_ARM) || defined(CONFIG_TILE)
+#elif defined(__alpha__) || defined(CONFIG_MN10300) || defined(CONFIG_ARM) || defined(CONFIG_ARM64) || defined(CONFIG_TILE)
#define GET_TIME(x) do { x = get_cycles(); } while (0)
#define DELTA(x,y) ((y)-(x))
#define TIME_NAME "get_cycles"
#define XTYPE_XBOX 0
#define XTYPE_XBOX360 1
#define XTYPE_XBOX360W 2
-#define XTYPE_UNKNOWN 3
+#define XTYPE_XBOXONE 3
+#define XTYPE_UNKNOWN 4
static bool dpad_to_buttons;
module_param(dpad_to_buttons, bool, S_IRUGO);
{ 0x045e, 0x0287, "Microsoft Xbox Controller S", 0, XTYPE_XBOX },
{ 0x045e, 0x0289, "Microsoft X-Box pad v2 (US)", 0, XTYPE_XBOX },
{ 0x045e, 0x028e, "Microsoft X-Box 360 pad", 0, XTYPE_XBOX360 },
+ { 0x045e, 0x02d1, "Microsoft X-Box One pad", 0, XTYPE_XBOXONE },
{ 0x045e, 0x0291, "Xbox 360 Wireless Receiver (XBOX)", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360W },
{ 0x045e, 0x0719, "Xbox 360 Wireless Receiver", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360W },
{ 0x044f, 0x0f07, "Thrustmaster, Inc. Controller", 0, XTYPE_XBOX },
-1
};
-/* Xbox 360 has a vendor-specific class, so we cannot match it with only
+/*
+ * Xbox 360 has a vendor-specific class, so we cannot match it with only
* USB_INTERFACE_INFO (also specifically refused by USB subsystem), so we
* match against vendor id as well. Wired Xbox 360 devices have protocol 1,
- * wireless controllers have protocol 129. */
+ * wireless controllers have protocol 129.
+ */
#define XPAD_XBOX360_VENDOR_PROTOCOL(vend,pr) \
.match_flags = USB_DEVICE_ID_MATCH_VENDOR | USB_DEVICE_ID_MATCH_INT_INFO, \
.idVendor = (vend), \
{ XPAD_XBOX360_VENDOR_PROTOCOL(vend,1) }, \
{ XPAD_XBOX360_VENDOR_PROTOCOL(vend,129) }
+/* The Xbox One controller uses subclass 71 and protocol 208. */
+#define XPAD_XBOXONE_VENDOR_PROTOCOL(vend, pr) \
+ .match_flags = USB_DEVICE_ID_MATCH_VENDOR | USB_DEVICE_ID_MATCH_INT_INFO, \
+ .idVendor = (vend), \
+ .bInterfaceClass = USB_CLASS_VENDOR_SPEC, \
+ .bInterfaceSubClass = 71, \
+ .bInterfaceProtocol = (pr)
+#define XPAD_XBOXONE_VENDOR(vend) \
+ { XPAD_XBOXONE_VENDOR_PROTOCOL(vend, 208) }
+
static struct usb_device_id xpad_table[] = {
{ USB_INTERFACE_INFO('X', 'B', 0) }, /* X-Box USB-IF not approved class */
XPAD_XBOX360_VENDOR(0x045e), /* Microsoft X-Box 360 controllers */
+ XPAD_XBOXONE_VENDOR(0x045e), /* Microsoft X-Box One controllers */
XPAD_XBOX360_VENDOR(0x046d), /* Logitech X-Box 360 style controllers */
XPAD_XBOX360_VENDOR(0x0738), /* Mad Catz X-Box 360 controllers */
{ USB_DEVICE(0x0738, 0x4540) }, /* Mad Catz Beat Pad */
struct urb *bulk_out;
unsigned char *bdata;
-#if defined(CONFIG_JOYSTICK_XPAD_FF) || defined(CONFIG_JOYSTICK_XPAD_LEDS)
struct urb *irq_out; /* urb for interrupt out report */
unsigned char *odata; /* output data */
dma_addr_t odata_dma;
struct mutex odata_mutex;
-#endif
#if defined(CONFIG_JOYSTICK_XPAD_LEDS)
struct xpad_led *led;
xpad360_process_packet(xpad, cmd, &data[4]);
}
+/*
+ * xpadone_process_buttons
+ *
+ * Process a button update packet from an Xbox one controller.
+ */
+static void xpadone_process_buttons(struct usb_xpad *xpad,
+ struct input_dev *dev,
+ unsigned char *data)
+{
+ /* menu/view buttons */
+ input_report_key(dev, BTN_START, data[4] & 0x04);
+ input_report_key(dev, BTN_SELECT, data[4] & 0x08);
+
+ /* buttons A,B,X,Y */
+ input_report_key(dev, BTN_A, data[4] & 0x10);
+ input_report_key(dev, BTN_B, data[4] & 0x20);
+ input_report_key(dev, BTN_X, data[4] & 0x40);
+ input_report_key(dev, BTN_Y, data[4] & 0x80);
+
+ /* digital pad */
+ if (xpad->mapping & MAP_DPAD_TO_BUTTONS) {
+ /* dpad as buttons (left, right, up, down) */
+ input_report_key(dev, BTN_TRIGGER_HAPPY1, data[5] & 0x04);
+ input_report_key(dev, BTN_TRIGGER_HAPPY2, data[5] & 0x08);
+ input_report_key(dev, BTN_TRIGGER_HAPPY3, data[5] & 0x01);
+ input_report_key(dev, BTN_TRIGGER_HAPPY4, data[5] & 0x02);
+ } else {
+ input_report_abs(dev, ABS_HAT0X,
+ !!(data[5] & 0x08) - !!(data[5] & 0x04));
+ input_report_abs(dev, ABS_HAT0Y,
+ !!(data[5] & 0x02) - !!(data[5] & 0x01));
+ }
+
+ /* TL/TR */
+ input_report_key(dev, BTN_TL, data[5] & 0x10);
+ input_report_key(dev, BTN_TR, data[5] & 0x20);
+
+ /* stick press left/right */
+ input_report_key(dev, BTN_THUMBL, data[5] & 0x40);
+ input_report_key(dev, BTN_THUMBR, data[5] & 0x80);
+
+ if (!(xpad->mapping & MAP_STICKS_TO_NULL)) {
+ /* left stick */
+ input_report_abs(dev, ABS_X,
+ (__s16) le16_to_cpup((__le16 *)(data + 10)));
+ input_report_abs(dev, ABS_Y,
+ ~(__s16) le16_to_cpup((__le16 *)(data + 12)));
+
+ /* right stick */
+ input_report_abs(dev, ABS_RX,
+ (__s16) le16_to_cpup((__le16 *)(data + 14)));
+ input_report_abs(dev, ABS_RY,
+ ~(__s16) le16_to_cpup((__le16 *)(data + 16)));
+ }
+
+ /* triggers left/right */
+ if (xpad->mapping & MAP_TRIGGERS_TO_BUTTONS) {
+ input_report_key(dev, BTN_TL2,
+ (__u16) le16_to_cpup((__le16 *)(data + 6)));
+ input_report_key(dev, BTN_TR2,
+ (__u16) le16_to_cpup((__le16 *)(data + 8)));
+ } else {
+ input_report_abs(dev, ABS_Z,
+ (__u16) le16_to_cpup((__le16 *)(data + 6)));
+ input_report_abs(dev, ABS_RZ,
+ (__u16) le16_to_cpup((__le16 *)(data + 8)));
+ }
+
+ input_sync(dev);
+}
+
+/*
+ * xpadone_process_packet
+ *
+ * Completes a request by converting the data into events for the
+ * input subsystem. This version is for the Xbox One controller.
+ *
+ * The report format was gleaned from
+ * https://github.com/kylelemons/xbox/blob/master/xbox.go
+ */
+
+static void xpadone_process_packet(struct usb_xpad *xpad,
+ u16 cmd, unsigned char *data)
+{
+ struct input_dev *dev = xpad->dev;
+
+ switch (data[0]) {
+ case 0x20:
+ xpadone_process_buttons(xpad, dev, data);
+ break;
+
+ case 0x07:
+ /* the xbox button has its own special report */
+ input_report_key(dev, BTN_MODE, data[4] & 0x01);
+ input_sync(dev);
+ break;
+ }
+}
+
static void xpad_irq_in(struct urb *urb)
{
struct usb_xpad *xpad = urb->context;
case XTYPE_XBOX360W:
xpad360w_process_packet(xpad, 0, xpad->idata);
break;
+ case XTYPE_XBOXONE:
+ xpadone_process_packet(xpad, 0, xpad->idata);
+ break;
default:
xpad_process_packet(xpad, 0, xpad->idata);
}
}
}
-#if defined(CONFIG_JOYSTICK_XPAD_FF) || defined(CONFIG_JOYSTICK_XPAD_LEDS)
static void xpad_irq_out(struct urb *urb)
{
struct usb_xpad *xpad = urb->context;
static int xpad_init_output(struct usb_interface *intf, struct usb_xpad *xpad)
{
struct usb_endpoint_descriptor *ep_irq_out;
+ int ep_irq_out_idx;
int error;
if (xpad->xtype == XTYPE_UNKNOWN)
goto fail2;
}
- ep_irq_out = &intf->cur_altsetting->endpoint[1].desc;
+ /* Xbox One controller has in/out endpoints swapped. */
+ ep_irq_out_idx = xpad->xtype == XTYPE_XBOXONE ? 0 : 1;
+ ep_irq_out = &intf->cur_altsetting->endpoint[ep_irq_out_idx].desc;
+
usb_fill_int_urb(xpad->irq_out, xpad->udev,
usb_sndintpipe(xpad->udev, ep_irq_out->bEndpointAddress),
xpad->odata, XPAD_PKT_LEN,
xpad->odata, xpad->odata_dma);
}
}
-#else
-static int xpad_init_output(struct usb_interface *intf, struct usb_xpad *xpad) { return 0; }
-static void xpad_deinit_output(struct usb_xpad *xpad) {}
-static void xpad_stop_output(struct usb_xpad *xpad) {}
-#endif
#ifdef CONFIG_JOYSTICK_XPAD_FF
static int xpad_play_effect(struct input_dev *dev, void *data, struct ff_effect *effect)
static int xpad_init_ff(struct usb_xpad *xpad)
{
- if (xpad->xtype == XTYPE_UNKNOWN)
+ if (xpad->xtype == XTYPE_UNKNOWN || xpad->xtype == XTYPE_XBOXONE)
return 0;
input_set_capability(xpad->dev, EV_FF, FF_RUMBLE);
if (usb_submit_urb(xpad->irq_in, GFP_KERNEL))
return -EIO;
+ if (xpad->xtype == XTYPE_XBOXONE) {
+ /* Xbox one controller needs to be initialized. */
+ xpad->odata[0] = 0x05;
+ xpad->odata[1] = 0x20;
+ xpad->irq_out->transfer_buffer_length = 2;
+ return usb_submit_urb(xpad->irq_out, GFP_KERNEL);
+ }
+
return 0;
}
static void xpad_set_up_abs(struct input_dev *input_dev, signed short abs)
{
+ struct usb_xpad *xpad = input_get_drvdata(input_dev);
set_bit(abs, input_dev->absbit);
switch (abs) {
break;
case ABS_Z:
case ABS_RZ: /* the triggers (if mapped to axes) */
- input_set_abs_params(input_dev, abs, 0, 255, 0, 0);
+ if (xpad->xtype == XTYPE_XBOXONE)
+ input_set_abs_params(input_dev, abs, 0, 1023, 0, 0);
+ else
+ input_set_abs_params(input_dev, abs, 0, 255, 0, 0);
break;
case ABS_HAT0X:
case ABS_HAT0Y: /* the d-pad (only if dpad is mapped to axes */
struct usb_xpad *xpad;
struct input_dev *input_dev;
struct usb_endpoint_descriptor *ep_irq_in;
+ int ep_irq_in_idx;
int i, error;
for (i = 0; xpad_device[i].idVendor; i++) {
break;
}
+ if (xpad_device[i].xtype == XTYPE_XBOXONE &&
+ intf->cur_altsetting->desc.bInterfaceNumber != 0) {
+ /*
+ * The Xbox One controller lists three interfaces all with the
+ * same interface class, subclass and protocol. Differentiate by
+ * interface number.
+ */
+ return -ENODEV;
+ }
+
xpad = kzalloc(sizeof(struct usb_xpad), GFP_KERNEL);
input_dev = input_allocate_device();
if (!xpad || !input_dev) {
__set_bit(xpad_common_btn[i], input_dev->keybit);
/* set up model-specific ones */
- if (xpad->xtype == XTYPE_XBOX360 || xpad->xtype == XTYPE_XBOX360W) {
+ if (xpad->xtype == XTYPE_XBOX360 || xpad->xtype == XTYPE_XBOX360W ||
+ xpad->xtype == XTYPE_XBOXONE) {
for (i = 0; xpad360_btn[i] >= 0; i++)
__set_bit(xpad360_btn[i], input_dev->keybit);
} else {
__set_bit(xpad_btn_pad[i], input_dev->keybit);
} else {
for (i = 0; xpad_abs_pad[i] >= 0; i++)
- xpad_set_up_abs(input_dev, xpad_abs_pad[i]);
+ xpad_set_up_abs(input_dev, xpad_abs_pad[i]);
}
if (xpad->mapping & MAP_TRIGGERS_TO_BUTTONS) {
if (error)
goto fail5;
- ep_irq_in = &intf->cur_altsetting->endpoint[0].desc;
+ /* Xbox One controller has in/out endpoints swapped. */
+ ep_irq_in_idx = xpad->xtype == XTYPE_XBOXONE ? 1 : 0;
+ ep_irq_in = &intf->cur_altsetting->endpoint[ep_irq_in_idx].desc;
+
usb_fill_int_urb(xpad->irq_in, udev,
usb_rcvintpipe(udev, ep_irq_in->bEndpointAddress),
xpad->idata, XPAD_PKT_LEN, xpad_irq_in,
struct input_dev *idev;
/* config */
- unsigned int keycodes[CAP1106_NUM_CHN];
+ unsigned short keycodes[CAP1106_NUM_CHN];
};
static const struct reg_default cap1106_reg_defaults[] = {
for (i = 0; i < CAP1106_NUM_CHN; i++)
__set_bit(priv->keycodes[i], priv->idev->keybit);
+ __clear_bit(KEY_RESERVED, priv->idev->keybit);
+
+ priv->idev->keycode = priv->keycodes;
+ priv->idev->keycodesize = sizeof(priv->keycodes[0]);
+ priv->idev->keycodemax = ARRAY_SIZE(priv->keycodes);
+
priv->idev->id.vendor = CAP1106_MANUFACTURER_ID;
priv->idev->id.product = CAP1106_PRODUCT_ID;
priv->idev->id.version = rev;
}
#ifdef CONFIG_MOUSE_PS2_SYNAPTICS
+
+static bool cr48_profile_sensor;
+
struct min_max_quirk {
const char * const *pnp_ids;
int x_min, x_max, y_min, y_max;
priv->agm_pending = false;
}
+static void synaptics_profile_sensor_process(struct psmouse *psmouse,
+ struct synaptics_hw_state *sgm,
+ int num_fingers)
+{
+ struct input_dev *dev = psmouse->dev;
+ struct synaptics_data *priv = psmouse->private;
+ struct synaptics_hw_state *hw[2] = { sgm, &priv->agm };
+ struct input_mt_pos pos[2];
+ int slot[2], nsemi, i;
+
+ nsemi = clamp_val(num_fingers, 0, 2);
+
+ for (i = 0; i < nsemi; i++) {
+ pos[i].x = hw[i]->x;
+ pos[i].y = synaptics_invert_y(hw[i]->y);
+ }
+
+ input_mt_assign_slots(dev, slot, pos, nsemi);
+
+ for (i = 0; i < nsemi; i++) {
+ input_mt_slot(dev, slot[i]);
+ input_mt_report_slot_state(dev, MT_TOOL_FINGER, true);
+ input_report_abs(dev, ABS_MT_POSITION_X, pos[i].x);
+ input_report_abs(dev, ABS_MT_POSITION_Y, pos[i].y);
+ input_report_abs(dev, ABS_MT_PRESSURE, hw[i]->z);
+ }
+
+ input_mt_drop_unused(dev);
+ input_mt_report_pointer_emulation(dev, false);
+ input_mt_report_finger_count(dev, num_fingers);
+
+ synaptics_report_buttons(psmouse, sgm);
+
+ input_sync(dev);
+}
+
/*
* called for each full received packet from the touchpad
*/
finger_width = 0;
}
+ if (cr48_profile_sensor) {
+ synaptics_profile_sensor_process(psmouse, &hw, num_fingers);
+ return;
+ }
+
if (SYN_CAP_ADV_GESTURE(priv->ext_cap_0c))
synaptics_report_semi_mt_data(dev, &hw, &priv->agm,
num_fingers);
set_abs_position_params(dev, priv, ABS_X, ABS_Y);
input_set_abs_params(dev, ABS_PRESSURE, 0, 255, 0, 0);
+ if (cr48_profile_sensor)
+ input_set_abs_params(dev, ABS_MT_PRESSURE, 0, 255, 0, 0);
+
if (SYN_CAP_IMAGE_SENSOR(priv->ext_cap_0c)) {
set_abs_position_params(dev, priv, ABS_MT_POSITION_X,
ABS_MT_POSITION_Y);
__set_bit(BTN_TOOL_QUADTAP, dev->keybit);
__set_bit(BTN_TOOL_QUINTTAP, dev->keybit);
} else if (SYN_CAP_ADV_GESTURE(priv->ext_cap_0c)) {
- /* Non-image sensors with AGM use semi-mt */
- __set_bit(INPUT_PROP_SEMI_MT, dev->propbit);
- input_mt_init_slots(dev, 2, 0);
set_abs_position_params(dev, priv, ABS_MT_POSITION_X,
ABS_MT_POSITION_Y);
+ /*
+ * Profile sensor in CR-48 tracks contacts reasonably well,
+ * other non-image sensors with AGM use semi-mt.
+ */
+ input_mt_init_slots(dev, 2,
+ INPUT_MT_POINTER |
+ (cr48_profile_sensor ?
+ INPUT_MT_TRACK : INPUT_MT_SEMI_MT));
}
if (SYN_CAP_PALMDETECT(priv->capabilities))
{ }
};
+static const struct dmi_system_id __initconst cr48_dmi_table[] = {
+#if defined(CONFIG_DMI) && defined(CONFIG_X86)
+ {
+ /* Cr-48 Chromebook (Codename Mario) */
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "IEC"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "Mario"),
+ },
+ },
+#endif
+ { }
+};
+
void __init synaptics_module_init(void)
{
impaired_toshiba_kbc = dmi_check_system(toshiba_dmi_table);
broken_olpc_ec = dmi_check_system(olpc_dmi_table);
+ cr48_profile_sensor = dmi_check_system(cr48_dmi_table);
}
static int __synaptics_init(struct psmouse *psmouse, bool absolute_mode)
msg.buf = val;
ret = i2c_transfer(data->client->adapter, &msg, 1);
-
if (ret == 1) {
ret = 0;
} else {
case 0x5b:
bootloader = appmode - 0x26;
break;
+
default:
dev_err(&data->client->dev,
"Appmode i2c address 0x%02x not found\n",
return 0;
}
-static int mxt_probe_bootloader(struct mxt_data *data, bool retry)
+static int mxt_probe_bootloader(struct mxt_data *data, bool alt_address)
{
struct device *dev = &data->client->dev;
- int ret;
+ int error;
u8 val;
bool crc_failure;
- ret = mxt_lookup_bootloader_address(data, retry);
- if (ret)
- return ret;
+ error = mxt_lookup_bootloader_address(data, alt_address);
+ if (error)
+ return error;
- ret = mxt_bootloader_read(data, &val, 1);
- if (ret)
- return ret;
+ error = mxt_bootloader_read(data, &val, 1);
+ if (error)
+ return error;
/* Check app crc fail mode */
crc_failure = (val & ~MXT_BOOT_STATUS_MASK) == MXT_APP_CRC_FAIL;
return crc;
}
+static int mxt_prepare_cfg_mem(struct mxt_data *data,
+ const struct firmware *cfg,
+ unsigned int data_pos,
+ unsigned int cfg_start_ofs,
+ u8 *config_mem,
+ size_t config_mem_size)
+{
+ struct device *dev = &data->client->dev;
+ struct mxt_object *object;
+ unsigned int type, instance, size, byte_offset;
+ int offset;
+ int ret;
+ int i;
+ u16 reg;
+ u8 val;
+
+ while (data_pos < cfg->size) {
+ /* Read type, instance, length */
+ ret = sscanf(cfg->data + data_pos, "%x %x %x%n",
+ &type, &instance, &size, &offset);
+ if (ret == 0) {
+ /* EOF */
+ break;
+ } else if (ret != 3) {
+ dev_err(dev, "Bad format: failed to parse object\n");
+ return -EINVAL;
+ }
+ data_pos += offset;
+
+ object = mxt_get_object(data, type);
+ if (!object) {
+ /* Skip object */
+ for (i = 0; i < size; i++) {
+ ret = sscanf(cfg->data + data_pos, "%hhx%n",
+ &val, &offset);
+ if (ret != 1) {
+ dev_err(dev, "Bad format in T%d at %d\n",
+ type, i);
+ return -EINVAL;
+ }
+ data_pos += offset;
+ }
+ continue;
+ }
+
+ if (size > mxt_obj_size(object)) {
+ /*
+ * Either we are in fallback mode due to wrong
+ * config or config from a later fw version,
+ * or the file is corrupt or hand-edited.
+ */
+ dev_warn(dev, "Discarding %zu byte(s) in T%u\n",
+ size - mxt_obj_size(object), type);
+ } else if (mxt_obj_size(object) > size) {
+ /*
+ * If firmware is upgraded, new bytes may be added to
+ * end of objects. It is generally forward compatible
+ * to zero these bytes - previous behaviour will be
+ * retained. However this does invalidate the CRC and
+ * will force fallback mode until the configuration is
+ * updated. We warn here but do nothing else - the
+ * malloc has zeroed the entire configuration.
+ */
+ dev_warn(dev, "Zeroing %zu byte(s) in T%d\n",
+ mxt_obj_size(object) - size, type);
+ }
+
+ if (instance >= mxt_obj_instances(object)) {
+ dev_err(dev, "Object instances exceeded!\n");
+ return -EINVAL;
+ }
+
+ reg = object->start_address + mxt_obj_size(object) * instance;
+
+ for (i = 0; i < size; i++) {
+ ret = sscanf(cfg->data + data_pos, "%hhx%n",
+ &val,
+ &offset);
+ if (ret != 1) {
+ dev_err(dev, "Bad format in T%d at %d\n",
+ type, i);
+ return -EINVAL;
+ }
+ data_pos += offset;
+
+ if (i > mxt_obj_size(object))
+ continue;
+
+ byte_offset = reg + i - cfg_start_ofs;
+
+ if (byte_offset >= 0 && byte_offset < config_mem_size) {
+ *(config_mem + byte_offset) = val;
+ } else {
+ dev_err(dev, "Bad object: reg:%d, T%d, ofs=%d\n",
+ reg, object->type, byte_offset);
+ return -EINVAL;
+ }
+ }
+ }
+
+ return 0;
+}
+
+static int mxt_upload_cfg_mem(struct mxt_data *data, unsigned int cfg_start,
+ u8 *config_mem, size_t config_mem_size)
+{
+ unsigned int byte_offset = 0;
+ int error;
+
+ /* Write configuration as blocks */
+ while (byte_offset < config_mem_size) {
+ unsigned int size = config_mem_size - byte_offset;
+
+ if (size > MXT_MAX_BLOCK_WRITE)
+ size = MXT_MAX_BLOCK_WRITE;
+
+ error = __mxt_write_reg(data->client,
+ cfg_start + byte_offset,
+ size, config_mem + byte_offset);
+ if (error) {
+ dev_err(&data->client->dev,
+ "Config write error, ret=%d\n", error);
+ return error;
+ }
+
+ byte_offset += size;
+ }
+
+ return 0;
+}
+
/*
* mxt_update_cfg - download configuration to chip
*
{
struct device *dev = &data->client->dev;
struct mxt_info cfg_info;
- struct mxt_object *object;
int ret;
int offset;
int data_pos;
- int byte_offset;
int i;
int cfg_start_ofs;
u32 info_crc, config_crc, calculated_crc;
u8 *config_mem;
size_t config_mem_size;
- unsigned int type, instance, size;
- u8 val;
- u16 reg;
mxt_update_crc(data, MXT_COMMAND_REPORTALL, 1);
if (strncmp(cfg->data, MXT_CFG_MAGIC, strlen(MXT_CFG_MAGIC))) {
dev_err(dev, "Unrecognised config file\n");
- ret = -EINVAL;
- goto release;
+ return -EINVAL;
}
data_pos = strlen(MXT_CFG_MAGIC);
&offset);
if (ret != 1) {
dev_err(dev, "Bad format\n");
- ret = -EINVAL;
- goto release;
+ return -EINVAL;
}
data_pos += offset;
if (cfg_info.family_id != data->info.family_id) {
dev_err(dev, "Family ID mismatch!\n");
- ret = -EINVAL;
- goto release;
+ return -EINVAL;
}
if (cfg_info.variant_id != data->info.variant_id) {
dev_err(dev, "Variant ID mismatch!\n");
- ret = -EINVAL;
- goto release;
+ return -EINVAL;
}
/* Read CRCs */
ret = sscanf(cfg->data + data_pos, "%x%n", &info_crc, &offset);
if (ret != 1) {
dev_err(dev, "Bad format: failed to parse Info CRC\n");
- ret = -EINVAL;
- goto release;
+ return -EINVAL;
}
data_pos += offset;
ret = sscanf(cfg->data + data_pos, "%x%n", &config_crc, &offset);
if (ret != 1) {
dev_err(dev, "Bad format: failed to parse Config CRC\n");
- ret = -EINVAL;
- goto release;
+ return -EINVAL;
}
data_pos += offset;
} else if (config_crc == data->config_crc) {
dev_dbg(dev, "Config CRC 0x%06X: OK\n",
data->config_crc);
- ret = 0;
- goto release;
+ return 0;
} else {
dev_info(dev, "Config CRC 0x%06X: does not match file 0x%06X\n",
data->config_crc, config_crc);
config_mem = kzalloc(config_mem_size, GFP_KERNEL);
if (!config_mem) {
dev_err(dev, "Failed to allocate memory\n");
- ret = -ENOMEM;
- goto release;
+ return -ENOMEM;
}
- while (data_pos < cfg->size) {
- /* Read type, instance, length */
- ret = sscanf(cfg->data + data_pos, "%x %x %x%n",
- &type, &instance, &size, &offset);
- if (ret == 0) {
- /* EOF */
- break;
- } else if (ret != 3) {
- dev_err(dev, "Bad format: failed to parse object\n");
- ret = -EINVAL;
- goto release_mem;
- }
- data_pos += offset;
-
- object = mxt_get_object(data, type);
- if (!object) {
- /* Skip object */
- for (i = 0; i < size; i++) {
- ret = sscanf(cfg->data + data_pos, "%hhx%n",
- &val,
- &offset);
- data_pos += offset;
- }
- continue;
- }
-
- if (size > mxt_obj_size(object)) {
- /*
- * Either we are in fallback mode due to wrong
- * config or config from a later fw version,
- * or the file is corrupt or hand-edited.
- */
- dev_warn(dev, "Discarding %zu byte(s) in T%u\n",
- size - mxt_obj_size(object), type);
- } else if (mxt_obj_size(object) > size) {
- /*
- * If firmware is upgraded, new bytes may be added to
- * end of objects. It is generally forward compatible
- * to zero these bytes - previous behaviour will be
- * retained. However this does invalidate the CRC and
- * will force fallback mode until the configuration is
- * updated. We warn here but do nothing else - the
- * malloc has zeroed the entire configuration.
- */
- dev_warn(dev, "Zeroing %zu byte(s) in T%d\n",
- mxt_obj_size(object) - size, type);
- }
-
- if (instance >= mxt_obj_instances(object)) {
- dev_err(dev, "Object instances exceeded!\n");
- ret = -EINVAL;
- goto release_mem;
- }
-
- reg = object->start_address + mxt_obj_size(object) * instance;
-
- for (i = 0; i < size; i++) {
- ret = sscanf(cfg->data + data_pos, "%hhx%n",
- &val,
- &offset);
- if (ret != 1) {
- dev_err(dev, "Bad format in T%d\n", type);
- ret = -EINVAL;
- goto release_mem;
- }
- data_pos += offset;
-
- if (i > mxt_obj_size(object))
- continue;
-
- byte_offset = reg + i - cfg_start_ofs;
-
- if ((byte_offset >= 0)
- && (byte_offset <= config_mem_size)) {
- *(config_mem + byte_offset) = val;
- } else {
- dev_err(dev, "Bad object: reg:%d, T%d, ofs=%d\n",
- reg, object->type, byte_offset);
- ret = -EINVAL;
- goto release_mem;
- }
- }
- }
+ ret = mxt_prepare_cfg_mem(data, cfg, data_pos, cfg_start_ofs,
+ config_mem, config_mem_size);
+ if (ret)
+ goto release_mem;
/* Calculate crc of the received configs (not the raw config file) */
if (data->T7_address < cfg_start_ofs) {
data->T7_address - cfg_start_ofs,
config_mem_size);
- if (config_crc > 0 && (config_crc != calculated_crc))
+ if (config_crc > 0 && config_crc != calculated_crc)
dev_warn(dev, "Config CRC error, calculated=%06X, file=%06X\n",
calculated_crc, config_crc);
- /* Write configuration as blocks */
- byte_offset = 0;
- while (byte_offset < config_mem_size) {
- size = config_mem_size - byte_offset;
-
- if (size > MXT_MAX_BLOCK_WRITE)
- size = MXT_MAX_BLOCK_WRITE;
-
- ret = __mxt_write_reg(data->client,
- cfg_start_ofs + byte_offset,
- size, config_mem + byte_offset);
- if (ret != 0) {
- dev_err(dev, "Config write error, ret=%d\n", ret);
- goto release_mem;
- }
-
- byte_offset += size;
- }
+ ret = mxt_upload_cfg_mem(data, cfg_start_ofs,
+ config_mem, config_mem_size);
+ if (ret)
+ goto release_mem;
mxt_update_crc(data, MXT_COMMAND_BACKUPNV, MXT_BACKUP_VALUE);
release_mem:
kfree(config_mem);
-release:
- release_firmware(cfg);
return ret;
}
switch (object->type) {
case MXT_GEN_MESSAGE_T5:
- if (data->info.family_id == 0x80) {
+ if (data->info.family_id == 0x80 &&
+ data->info.version < 0x20) {
/*
- * On mXT224 read and discard unused CRC byte
- * otherwise DMA reads are misaligned
+ * On mXT224 firmware versions prior to V2.0
+ * read and discard unused CRC byte otherwise
+ * DMA reads are misaligned.
*/
data->T5_msg_size = mxt_obj_size(object);
} else {
data->T5_msg_size = mxt_obj_size(object) - 1;
}
data->T5_address = object->start_address;
+ break;
case MXT_GEN_COMMAND_T6:
data->T6_reportid = min_id;
data->T6_address = object->start_address;
static void mxt_config_cb(const struct firmware *cfg, void *ctx)
{
mxt_configure_objects(ctx, cfg);
+ release_firmware(cfg);
}
static int mxt_initialize(struct mxt_data *data)
{
struct i2c_client *client = data->client;
+ int recovery_attempts = 0;
int error;
- bool alt_bootloader_addr = false;
- bool retry = false;
-retry_info:
- error = mxt_get_info(data);
- if (error) {
-retry_bootloader:
- error = mxt_probe_bootloader(data, alt_bootloader_addr);
+ while (1) {
+ error = mxt_get_info(data);
+ if (!error)
+ break;
+
+ /* Check bootloader state */
+ error = mxt_probe_bootloader(data, false);
if (error) {
- if (alt_bootloader_addr) {
+ dev_info(&client->dev, "Trying alternate bootloader address\n");
+ error = mxt_probe_bootloader(data, true);
+ if (error) {
/* Chip is not in appmode or bootloader mode */
return error;
}
+ }
- dev_info(&client->dev, "Trying alternate bootloader address\n");
- alt_bootloader_addr = true;
- goto retry_bootloader;
- } else {
- if (retry) {
- dev_err(&client->dev, "Could not recover from bootloader mode\n");
- /*
- * We can reflash from this state, so do not
- * abort init
- */
- data->in_bootloader = true;
- return 0;
- }
-
- /* Attempt to exit bootloader into app mode */
- mxt_send_bootloader_cmd(data, false);
- msleep(MXT_FW_RESET_TIME);
- retry = true;
- goto retry_info;
+ /* OK, we are in bootloader, see if we can recover */
+ if (++recovery_attempts > 1) {
+ dev_err(&client->dev, "Could not recover from bootloader mode\n");
+ /*
+ * We can reflash from this state, so do not
+ * abort initialization.
+ */
+ data->in_bootloader = true;
+ return 0;
}
+
+ /* Attempt to exit bootloader into app mode */
+ mxt_send_bootloader_cmd(data, false);
+ msleep(MXT_FW_RESET_TIME);
}
/* Get object table information */
return error;
}
- mxt_acquire_irq(data);
+ error = mxt_acquire_irq(data);
if (error)
goto err_free_object_table;
- request_firmware_nowait(THIS_MODULE, true, MXT_CFG_NAME,
- &data->client->dev, GFP_KERNEL, data,
- mxt_config_cb);
+ error = request_firmware_nowait(THIS_MODULE, true, MXT_CFG_NAME,
+ &client->dev, GFP_KERNEL, data,
+ mxt_config_cb);
+ if (error) {
+ dev_err(&client->dev, "Failed to invoke firmware loader: %d\n",
+ error);
+ goto err_free_object_table;
+ }
return 0;
case M06:
wrbuf[0] = tsdata->factory_mode ? 0xf3 : 0xfc;
wrbuf[1] = tsdata->factory_mode ? addr & 0x7f : addr & 0x3f;
- wrbuf[1] = tsdata->factory_mode ? addr & 0x7f : addr & 0x3f;
wrbuf[2] = value;
wrbuf[3] = wrbuf[0] ^ wrbuf[1] ^ wrbuf[2];
return edt_ft5x06_ts_readwrite(tsdata->client, 4,
static void cleanup_domain(struct protection_domain *domain)
{
- struct iommu_dev_data *dev_data, *next;
+ struct iommu_dev_data *entry;
unsigned long flags;
write_lock_irqsave(&amd_iommu_devtable_lock, flags);
- list_for_each_entry_safe(dev_data, next, &domain->dev_list, list) {
- __detach_device(dev_data);
- atomic_set(&dev_data->bind, 0);
+ while (!list_empty(&domain->dev_list)) {
+ entry = list_first_entry(&domain->dev_list,
+ struct iommu_dev_data, list);
+ __detach_device(entry);
+ atomic_set(&entry->bind, 0);
}
write_unlock_irqrestore(&amd_iommu_devtable_lock, flags);
action != BUS_NOTIFY_DEL_DEVICE)
return 0;
+ /*
+ * If the device is still attached to a device driver we can't
+ * tear down the domain yet as DMA mappings may still be in use.
+ * Wait for the BUS_NOTIFY_UNBOUND_DRIVER event to do that.
+ */
+ if (action == BUS_NOTIFY_DEL_DEVICE && dev->driver != NULL)
+ return 0;
+
domain = find_domain(dev);
if (!domain)
return 0;
size_t orig_size = size;
int ret = 0;
- if (unlikely(domain->ops->unmap == NULL ||
+ if (unlikely(domain->ops->map == NULL ||
domain->ops->pgsize_bitmap == 0UL))
return -ENODEV;
/* $Id: xdi_msg.h,v 1.1.2.2 2001/02/16 08:40:36 armin Exp $ */
-#ifndef __DIVA_XDI_UM_CFG_MESSSGE_H__
+#ifndef __DIVA_XDI_UM_CFG_MESSAGE_H__
#define __DIVA_XDI_UM_CFG_MESSAGE_H__
/*
mutex_unlock(&ca->set->bucket_lock);
blkdev_issue_discard(ca->bdev,
bucket_to_sector(ca->set, bucket),
- ca->sb.block_size, GFP_KERNEL, 0);
+ ca->sb.bucket_size, GFP_KERNEL, 0);
mutex_lock(&ca->set->bucket_lock);
}
* CACHE_SET_STOPPING always gets set first when we're closing down a cache set;
* we'll continue to run normally for awhile with CACHE_SET_STOPPING set (i.e.
* flushing dirty data).
+ *
+ * CACHE_SET_RUNNING means all cache devices have been registered and journal
+ * replay is complete.
*/
#define CACHE_SET_UNREGISTERING 0
#define CACHE_SET_STOPPING 1
+#define CACHE_SET_RUNNING 2
struct cache_set {
struct closure cl;
{
uint64_t start_time;
bool used_mempool = false;
- struct bset *out = (void *) __get_free_pages(__GFP_NOWARN|GFP_NOIO,
+ struct bset *out = (void *) __get_free_pages(__GFP_NOWARN|GFP_NOWAIT,
order);
if (!out) {
struct page *outp;
{
return (KEY_DIRTY(l) == KEY_DIRTY(r) &&
KEY_PTRS(l) == KEY_PTRS(r) &&
- KEY_CSUM(l) == KEY_CSUM(l));
+ KEY_CSUM(l) == KEY_CSUM(r));
}
/* Keylists */
({ \
int _r, l = (b)->level - 1; \
bool _w = l <= (op)->lock; \
- struct btree *_child = bch_btree_node_get((b)->c, op, key, l, _w);\
+ struct btree *_child = bch_btree_node_get((b)->c, op, key, l, \
+ _w, b); \
if (!IS_ERR(_child)) { \
- _child->parent = (b); \
_r = bch_btree_ ## fn(_child, op, ##__VA_ARGS__); \
rw_unlock(_w, _child); \
} else \
rw_lock(_w, _b, _b->level); \
if (_b == (c)->root && \
_w == insert_lock(op, _b)) { \
- _b->parent = NULL; \
_r = bch_btree_ ## fn(_b, op, ##__VA_ARGS__); \
} \
rw_unlock(_w, _b); \
struct bset *i = btree_bset_first(b);
struct btree_iter *iter;
- iter = mempool_alloc(b->c->fill_iter, GFP_NOWAIT);
+ iter = mempool_alloc(b->c->fill_iter, GFP_NOIO);
iter->size = b->c->sb.bucket_size / b->c->sb.block_size;
iter->used = 0;
SET_PTR_OFFSET(&k.key, 0, PTR_OFFSET(&k.key, 0) +
bset_sector_offset(&b->keys, i));
- if (!bio_alloc_pages(b->bio, GFP_NOIO)) {
+ if (!bio_alloc_pages(b->bio, __GFP_NOWARN|GFP_NOWAIT)) {
int j;
struct bio_vec *bv;
void *base = (void *) ((unsigned long) i & ~(PAGE_SIZE - 1));
* level and op->lock.
*/
struct btree *bch_btree_node_get(struct cache_set *c, struct btree_op *op,
- struct bkey *k, int level, bool write)
+ struct bkey *k, int level, bool write,
+ struct btree *parent)
{
int i = 0;
struct btree *b;
BUG_ON(b->level != level);
}
+ b->parent = parent;
b->accessed = 1;
for (; i <= b->keys.nsets && b->keys.set[i].size; i++) {
return b;
}
-static void btree_node_prefetch(struct cache_set *c, struct bkey *k, int level)
+static void btree_node_prefetch(struct btree *parent, struct bkey *k)
{
struct btree *b;
- mutex_lock(&c->bucket_lock);
- b = mca_alloc(c, NULL, k, level);
- mutex_unlock(&c->bucket_lock);
+ mutex_lock(&parent->c->bucket_lock);
+ b = mca_alloc(parent->c, NULL, k, parent->level - 1);
+ mutex_unlock(&parent->c->bucket_lock);
if (!IS_ERR_OR_NULL(b)) {
+ b->parent = parent;
bch_btree_node_read(b);
rw_unlock(true, b);
}
mutex_unlock(&b->c->bucket_lock);
}
-struct btree *bch_btree_node_alloc(struct cache_set *c, struct btree_op *op,
- int level)
+struct btree *__bch_btree_node_alloc(struct cache_set *c, struct btree_op *op,
+ int level, bool wait,
+ struct btree *parent)
{
BKEY_PADDED(key) k;
struct btree *b = ERR_PTR(-EAGAIN);
mutex_lock(&c->bucket_lock);
retry:
- if (__bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, 1, op != NULL))
+ if (__bch_bucket_alloc_set(c, RESERVE_BTREE, &k.key, 1, wait))
goto err;
bkey_put(c, &k.key);
}
b->accessed = 1;
+ b->parent = parent;
bch_bset_init_next(&b->keys, b->keys.set->data, bset_magic(&b->c->sb));
mutex_unlock(&c->bucket_lock);
err:
mutex_unlock(&c->bucket_lock);
- trace_bcache_btree_node_alloc_fail(b);
+ trace_bcache_btree_node_alloc_fail(c);
return b;
}
+static struct btree *bch_btree_node_alloc(struct cache_set *c,
+ struct btree_op *op, int level,
+ struct btree *parent)
+{
+ return __bch_btree_node_alloc(c, op, level, op != NULL, parent);
+}
+
static struct btree *btree_node_alloc_replacement(struct btree *b,
struct btree_op *op)
{
- struct btree *n = bch_btree_node_alloc(b->c, op, b->level);
+ struct btree *n = bch_btree_node_alloc(b->c, op, b->level, b->parent);
if (!IS_ERR_OR_NULL(n)) {
mutex_lock(&n->write_lock);
bch_btree_sort_into(&b->keys, &n->keys, &b->c->sort);
BUG_ON(btree_bset_first(new_nodes[0])->keys);
btree_node_free(new_nodes[0]);
rw_unlock(true, new_nodes[0]);
+ new_nodes[0] = NULL;
for (i = 0; i < nodes; i++) {
if (__bch_keylist_realloc(&keylist, bkey_u64s(&r[i].b->key)))
k = bch_btree_iter_next_filter(&iter, &b->keys, bch_ptr_bad);
if (k) {
r->b = bch_btree_node_get(b->c, op, k, b->level - 1,
- true);
+ true, b);
if (IS_ERR(r->b)) {
ret = PTR_ERR(r->b);
break;
k = bch_btree_iter_next_filter(&iter, &b->keys,
bch_ptr_bad);
if (k)
- btree_node_prefetch(b->c, k, b->level - 1);
+ btree_node_prefetch(b, k);
if (p)
ret = btree(check_recurse, p, b, op);
trace_bcache_btree_node_split(b, btree_bset_first(n1)->keys);
- n2 = bch_btree_node_alloc(b->c, op, b->level);
+ n2 = bch_btree_node_alloc(b->c, op, b->level, b->parent);
if (IS_ERR(n2))
goto err_free1;
if (!b->parent) {
- n3 = bch_btree_node_alloc(b->c, op, b->level + 1);
+ n3 = bch_btree_node_alloc(b->c, op, b->level + 1, NULL);
if (IS_ERR(n3))
goto err_free2;
}
void bch_btree_node_write(struct btree *, struct closure *);
void bch_btree_set_root(struct btree *);
-struct btree *bch_btree_node_alloc(struct cache_set *, struct btree_op *, int);
+struct btree *__bch_btree_node_alloc(struct cache_set *, struct btree_op *,
+ int, bool, struct btree *);
struct btree *bch_btree_node_get(struct cache_set *, struct btree_op *,
- struct bkey *, int, bool);
+ struct bkey *, int, bool, struct btree *);
int bch_btree_insert_check_key(struct btree *, struct btree_op *,
struct bkey *);
return false;
}
-static bool bch_extent_invalid(struct btree_keys *bk, const struct bkey *k)
+bool __bch_extent_invalid(struct cache_set *c, const struct bkey *k)
{
- struct btree *b = container_of(bk, struct btree, keys);
char buf[80];
if (!KEY_SIZE(k))
if (KEY_SIZE(k) > KEY_OFFSET(k))
goto bad;
- if (__ptr_invalid(b->c, k))
+ if (__ptr_invalid(c, k))
goto bad;
return false;
bad:
bch_extent_to_text(buf, sizeof(buf), k);
- cache_bug(b->c, "spotted extent %s: %s", buf, bch_ptr_status(b->c, k));
+ cache_bug(c, "spotted extent %s: %s", buf, bch_ptr_status(c, k));
return true;
}
+static bool bch_extent_invalid(struct btree_keys *bk, const struct bkey *k)
+{
+ struct btree *b = container_of(bk, struct btree, keys);
+ return __bch_extent_invalid(b->c, k);
+}
+
static bool bch_extent_bad_expensive(struct btree *b, const struct bkey *k,
unsigned ptr)
{
void bch_extent_to_text(char *, size_t, const struct bkey *);
bool __bch_btree_ptr_invalid(struct cache_set *, const struct bkey *);
+bool __bch_extent_invalid(struct cache_set *, const struct bkey *);
#endif /* _BCACHE_EXTENTS_H */
#include "bcache.h"
#include "btree.h"
#include "debug.h"
+#include "extents.h"
#include <trace/events/bcache.h>
if (read_bucket(l))
goto bsearch;
- if (list_empty(list))
+ /* no journal entries on this device? */
+ if (l == ca->sb.njournal_buckets)
continue;
bsearch:
+ BUG_ON(list_empty(list));
+
/* Binary search */
- m = r = find_next_bit(bitmap, ca->sb.njournal_buckets, l + 1);
+ m = l;
+ r = find_next_bit(bitmap, ca->sb.njournal_buckets, l + 1);
pr_debug("starting binary search, l %u r %u", l, r);
while (l + 1 < r) {
for (k = i->j.start;
k < bset_bkey_last(&i->j);
- k = bkey_next(k)) {
- unsigned j;
+ k = bkey_next(k))
+ if (!__bch_extent_invalid(c, k)) {
+ unsigned j;
- for (j = 0; j < KEY_PTRS(k); j++)
- if (ptr_available(c, k, j))
- atomic_inc(&PTR_BUCKET(c, k, j)->pin);
+ for (j = 0; j < KEY_PTRS(k); j++)
+ if (ptr_available(c, k, j))
+ atomic_inc(&PTR_BUCKET(c, k, j)->pin);
- bch_initial_mark_key(c, 0, k);
- }
+ bch_initial_mark_key(c, 0, k);
+ }
}
}
{
struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
- trace_bcache_write(op->bio, op->writeback, op->bypass);
+ trace_bcache_write(op->c, op->inode, op->bio,
+ op->writeback, op->bypass);
bch_keylist_init(&op->insert_keys);
bio_get(op->bio);
static void bcache_device_attach(struct bcache_device *d, struct cache_set *c,
unsigned id)
{
- BUG_ON(test_bit(CACHE_SET_STOPPING, &c->flags));
-
d->id = id;
d->c = c;
c->devices[id] = d;
list_move(&dc->list, &uncached_devices);
clear_bit(BCACHE_DEV_DETACHING, &dc->disk.flags);
+ clear_bit(BCACHE_DEV_UNLINK_DONE, &dc->disk.flags);
mutex_unlock(&bch_register_lock);
*/
atomic_set(&dc->count, 1);
+ if (bch_cached_dev_writeback_start(dc))
+ return -ENOMEM;
+
if (BDEV_STATE(&dc->sb) == BDEV_STATE_DIRTY) {
bch_sectors_dirty_init(dc);
atomic_set(&dc->has_dirty, 1);
struct cached_dev *dc = container_of(cl, struct cached_dev, disk.cl);
cancel_delayed_work_sync(&dc->writeback_rate_update);
- kthread_stop(dc->writeback_thread);
+ if (!IS_ERR_OR_NULL(dc->writeback_thread))
+ kthread_stop(dc->writeback_thread);
mutex_lock(&bch_register_lock);
mutex_unlock(&bch_register_lock);
- if (!IS_ERR_OR_NULL(dc->bdev)) {
- if (dc->bdev->bd_disk)
- blk_sync_queue(bdev_get_queue(dc->bdev));
-
+ if (!IS_ERR_OR_NULL(dc->bdev))
blkdev_put(dc->bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
- }
wake_up(&unregister_wait);
static void flash_dev_free(struct closure *cl)
{
struct bcache_device *d = container_of(cl, struct bcache_device, cl);
+ mutex_lock(&bch_register_lock);
bcache_device_free(d);
+ mutex_unlock(&bch_register_lock);
kobject_put(&d->kobj);
}
{
struct bcache_device *d = container_of(cl, struct bcache_device, cl);
+ mutex_lock(&bch_register_lock);
bcache_device_unlink(d);
+ mutex_unlock(&bch_register_lock);
kobject_del(&d->kobj);
continue_at(cl, flash_dev_free, system_wq);
}
if (test_bit(CACHE_SET_STOPPING, &c->flags))
return -EINTR;
+ if (!test_bit(CACHE_SET_RUNNING, &c->flags))
+ return -EPERM;
+
u = uuid_find_empty(c);
if (!u) {
pr_err("Can't create volume, no room for UUID");
bch_journal_free(c);
for_each_cache(ca, c, i)
- if (ca)
+ if (ca) {
+ ca->set = NULL;
+ c->cache[ca->sb.nr_this_dev] = NULL;
kobject_put(&ca->kobj);
+ }
bch_bset_sort_state_free(&c->sort);
free_pages((unsigned long) c->uuids, ilog2(bucket_pages(c)));
if (ca->alloc_thread)
kthread_stop(ca->alloc_thread);
- cancel_delayed_work_sync(&c->journal.work);
- /* flush last journal entry if needed */
- c->journal.work.work.func(&c->journal.work.work);
+ if (c->journal.cur) {
+ cancel_delayed_work_sync(&c->journal.work);
+ /* flush last journal entry if needed */
+ c->journal.work.work.func(&c->journal.work.work);
+ }
closure_return(cl);
}
goto err;
err = "error reading btree root";
- c->root = bch_btree_node_get(c, NULL, k, j->btree_level, true);
+ c->root = bch_btree_node_get(c, NULL, k, j->btree_level, true, NULL);
if (IS_ERR_OR_NULL(c->root))
goto err;
goto err;
err = "cannot allocate new btree root";
- c->root = bch_btree_node_alloc(c, NULL, 0);
+ c->root = __bch_btree_node_alloc(c, NULL, 0, true, NULL);
if (IS_ERR_OR_NULL(c->root))
goto err;
flash_devs_run(c);
+ set_bit(CACHE_SET_RUNNING, &c->flags);
return;
err:
closure_sync(&cl);
pr_debug("set version = %llu", c->sb.version);
}
+ kobject_get(&ca->kobj);
ca->set = c;
ca->set->cache[ca->sb.nr_this_dev] = ca;
c->cache_by_alloc[c->caches_loaded++] = ca;
struct cache *ca = container_of(kobj, struct cache, kobj);
unsigned i;
- if (ca->set)
+ if (ca->set) {
+ BUG_ON(ca->set->cache[ca->sb.nr_this_dev] != ca);
ca->set->cache[ca->sb.nr_this_dev] = NULL;
+ }
bio_split_pool_free(&ca->bio_split_hook);
if (ca->sb_bio.bi_inline_vecs[0].bv_page)
put_page(ca->sb_bio.bi_io_vec[0].bv_page);
- if (!IS_ERR_OR_NULL(ca->bdev)) {
- blk_sync_queue(bdev_get_queue(ca->bdev));
+ if (!IS_ERR_OR_NULL(ca->bdev))
blkdev_put(ca->bdev, FMODE_READ|FMODE_WRITE|FMODE_EXCL);
- }
kfree(ca);
module_put(THIS_MODULE);
}
static void register_cache(struct cache_sb *sb, struct page *sb_page,
- struct block_device *bdev, struct cache *ca)
+ struct block_device *bdev, struct cache *ca)
{
char name[BDEVNAME_SIZE];
const char *err = "cannot allocate memory";
goto err;
pr_info("registered cache device %s", bdevname(bdev, name));
+out:
+ kobject_put(&ca->kobj);
return;
err:
pr_notice("error opening %s: %s", bdevname(bdev, name), err);
- kobject_put(&ca->kobj);
+ goto out;
}
/* Global interfaces/init */
if (IS_ERR(bdev)) {
if (bdev == ERR_PTR(-EBUSY)) {
bdev = lookup_bdev(strim(path));
+ mutex_lock(&bch_register_lock);
if (!IS_ERR(bdev) && bch_is_open(bdev))
err = "device already registered";
else
err = "device busy";
+ mutex_unlock(&bch_register_lock);
}
goto err;
}
average_frequency, frequency_units); \
__print_time_stat(stats, name, \
average_duration, duration_units); \
- __print_time_stat(stats, name, \
- max_duration, duration_units); \
+ sysfs_print(name ## _ ##max_duration ## _ ## duration_units, \
+ div_u64((stats)->max_duration, NSEC_PER_ ## duration_units));\
\
sysfs_print(name ## _last_ ## frequency_units, (stats)->last \
? div_s64(local_clock() - (stats)->last, \
if (KEY_START(&w->key) != dc->last_read ||
jiffies_to_msecs(delay) > 50)
while (!kthread_should_stop() && delay)
- delay = schedule_timeout_uninterruptible(delay);
+ delay = schedule_timeout_interruptible(delay);
dc->last_read = KEY_OFFSET(&w->key);
while (delay &&
!kthread_should_stop() &&
!test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags))
- delay = schedule_timeout_uninterruptible(delay);
+ delay = schedule_timeout_interruptible(delay);
}
}
dc->disk.sectors_dirty_last = bcache_dev_sectors_dirty(&dc->disk);
}
-int bch_cached_dev_writeback_init(struct cached_dev *dc)
+void bch_cached_dev_writeback_init(struct cached_dev *dc)
{
sema_init(&dc->in_flight, 64);
init_rwsem(&dc->writeback_lock);
dc->writeback_rate_d_term = 30;
dc->writeback_rate_p_term_inverse = 6000;
+ INIT_DELAYED_WORK(&dc->writeback_rate_update, update_writeback_rate);
+}
+
+int bch_cached_dev_writeback_start(struct cached_dev *dc)
+{
dc->writeback_thread = kthread_create(bch_writeback_thread, dc,
"bcache_writeback");
if (IS_ERR(dc->writeback_thread))
return PTR_ERR(dc->writeback_thread);
- INIT_DELAYED_WORK(&dc->writeback_rate_update, update_writeback_rate);
schedule_delayed_work(&dc->writeback_rate_update,
dc->writeback_rate_update_seconds * HZ);
+ bch_writeback_queue(dc);
+
return 0;
}
void bcache_dev_sectors_dirty_add(struct cache_set *, unsigned, uint64_t, int);
void bch_sectors_dirty_init(struct cached_dev *dc);
-int bch_cached_dev_writeback_init(struct cached_dev *);
+void bch_cached_dev_writeback_init(struct cached_dev *);
+int bch_cached_dev_writeback_start(struct cached_dev *);
#endif
disk_super->discard_root = cpu_to_le64(cmd->discard_root);
disk_super->discard_block_size = cpu_to_le64(cmd->discard_block_size);
disk_super->discard_nr_blocks = cpu_to_le64(from_oblock(cmd->discard_nr_blocks));
- disk_super->metadata_block_size = cpu_to_le32(DM_CACHE_METADATA_BLOCK_SIZE >> SECTOR_SHIFT);
+ disk_super->metadata_block_size = cpu_to_le32(DM_CACHE_METADATA_BLOCK_SIZE);
disk_super->data_block_size = cpu_to_le32(cmd->data_block_size);
disk_super->cache_blocks = cpu_to_le32(0);
bool may_format_device)
{
int r;
- cmd->bm = dm_block_manager_create(cmd->bdev, DM_CACHE_METADATA_BLOCK_SIZE,
+ cmd->bm = dm_block_manager_create(cmd->bdev, DM_CACHE_METADATA_BLOCK_SIZE << SECTOR_SHIFT,
CACHE_METADATA_CACHE_SIZE,
CACHE_MAX_CONCURRENT_LOCKS);
if (IS_ERR(cmd->bm)) {
#include "dm-cache-block-types.h"
#include "dm-cache-policy-internal.h"
+#include "persistent-data/dm-space-map-metadata.h"
/*----------------------------------------------------------------*/
-#define DM_CACHE_METADATA_BLOCK_SIZE 4096
+#define DM_CACHE_METADATA_BLOCK_SIZE DM_SM_METADATA_BLOCK_SIZE
/* FIXME: remove this restriction */
/*
* The metadata device is currently limited in size.
- *
- * We have one block of index, which can hold 255 index entries. Each
- * index entry contains allocation info about 16k metadata blocks.
*/
-#define DM_CACHE_METADATA_MAX_SECTORS (255 * (1 << 14) * (DM_CACHE_METADATA_BLOCK_SIZE / (1 << SECTOR_SHIFT)))
+#define DM_CACHE_METADATA_MAX_SECTORS DM_SM_METADATA_MAX_SECTORS
/*
* A metadata device larger than 16GB triggers a warning.
return bio->bi_rw & (REQ_FLUSH | REQ_FUA);
}
+/*
+ * You must increment the deferred set whilst the prison cell is held. To
+ * encourage this, we ask for 'cell' to be passed in.
+ */
+static void inc_ds(struct cache *cache, struct bio *bio,
+ struct dm_bio_prison_cell *cell)
+{
+ size_t pb_data_size = get_per_bio_data_size(cache);
+ struct per_bio_data *pb = get_per_bio_data(bio, pb_data_size);
+
+ BUG_ON(!cell);
+ BUG_ON(pb->all_io_entry);
+
+ pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
+}
+
static void issue(struct cache *cache, struct bio *bio)
{
unsigned long flags;
spin_unlock_irqrestore(&cache->lock, flags);
}
+static void inc_and_issue(struct cache *cache, struct bio *bio, struct dm_bio_prison_cell *cell)
+{
+ inc_ds(cache, bio, cell);
+ issue(cache, bio);
+}
+
static void defer_writethrough_bio(struct cache *cache, struct bio *bio)
{
unsigned long flags;
dm_hook_bio(&pb->hook_info, bio, overwrite_endio, mg);
remap_to_cache_dirty(mg->cache, bio, mg->new_oblock, mg->cblock);
+
+ /*
+ * No need to inc_ds() here, since the cell will be held for the
+ * duration of the io.
+ */
generic_make_request(bio);
}
return;
INIT_LIST_HEAD(&work);
- if (pb->all_io_entry)
- dm_deferred_entry_dec(pb->all_io_entry, &work);
+ dm_deferred_entry_dec(pb->all_io_entry, &work);
if (!list_empty(&work))
queue_quiesced_migrations(cache, &work);
else
remap_to_cache(cache, bio, 0);
+ /*
+ * REQ_FLUSH is not directed at any particular block so we don't
+ * need to inc_ds(). REQ_FUA's are split into a write + REQ_FLUSH
+ * by dm-core.
+ */
issue(cache, bio);
}
&cache->stats.read_miss : &cache->stats.write_miss);
}
-static void issue_cache_bio(struct cache *cache, struct bio *bio,
- struct per_bio_data *pb,
- dm_oblock_t oblock, dm_cblock_t cblock)
-{
- pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
- remap_to_cache_dirty(cache, bio, oblock, cblock);
- issue(cache, bio);
-}
-
static void process_bio(struct cache *cache, struct prealloc *structs,
struct bio *bio)
{
dm_oblock_t block = get_bio_block(cache, bio);
struct dm_bio_prison_cell *cell_prealloc, *old_ocell, *new_ocell;
struct policy_result lookup_result;
- size_t pb_data_size = get_per_bio_data_size(cache);
- struct per_bio_data *pb = get_per_bio_data(bio, pb_data_size);
bool discarded_block = is_discarded_oblock(cache, block);
bool passthrough = passthrough_mode(&cache->features);
bool can_migrate = !passthrough && (discarded_block || spare_migration_bandwidth(cache));
} else {
/* FIXME: factor out issue_origin() */
- pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
remap_to_origin_clear_discard(cache, bio, block);
- issue(cache, bio);
+ inc_and_issue(cache, bio, new_ocell);
}
} else {
inc_hit_counter(cache, bio);
if (bio_data_dir(bio) == WRITE &&
writethrough_mode(&cache->features) &&
!is_dirty(cache, lookup_result.cblock)) {
- pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
remap_to_origin_then_cache(cache, bio, block, lookup_result.cblock);
- issue(cache, bio);
- } else
- issue_cache_bio(cache, bio, pb, block, lookup_result.cblock);
+ inc_and_issue(cache, bio, new_ocell);
+
+ } else {
+ remap_to_cache_dirty(cache, bio, block, lookup_result.cblock);
+ inc_and_issue(cache, bio, new_ocell);
+ }
}
break;
case POLICY_MISS:
inc_miss_counter(cache, bio);
- pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
remap_to_origin_clear_discard(cache, bio, block);
- issue(cache, bio);
+ inc_and_issue(cache, bio, new_ocell);
break;
case POLICY_NEW:
bio_list_init(&cache->deferred_flush_bios);
spin_unlock_irqrestore(&cache->lock, flags);
+ /*
+ * These bios have already been through inc_ds()
+ */
while ((bio = bio_list_pop(&bios)))
submit_bios ? generic_make_request(bio) : bio_io_error(bio);
}
bio_list_init(&cache->deferred_writethrough_bios);
spin_unlock_irqrestore(&cache->lock, flags);
+ /*
+ * These bios have already been through inc_ds()
+ */
while ((bio = bio_list_pop(&bios)))
generic_make_request(bio);
}
if (commit_if_needed(cache)) {
process_deferred_flush_bios(cache, false);
+ process_migrations(cache, &cache->need_commit_migrations, migration_failure);
/*
* FIXME: rollback metadata or just go into a
return r;
}
-static int cache_map(struct dm_target *ti, struct bio *bio)
+static int __cache_map(struct cache *cache, struct bio *bio, struct dm_bio_prison_cell **cell)
{
- struct cache *cache = ti->private;
-
int r;
dm_oblock_t block = get_bio_block(cache, bio);
size_t pb_data_size = get_per_bio_data_size(cache);
bool can_migrate = false;
bool discarded_block;
- struct dm_bio_prison_cell *cell;
struct policy_result lookup_result;
struct per_bio_data *pb = init_per_bio_data(bio, pb_data_size);
/*
* Check to see if that block is currently migrating.
*/
- cell = alloc_prison_cell(cache);
- if (!cell) {
+ *cell = alloc_prison_cell(cache);
+ if (!*cell) {
defer_bio(cache, bio);
return DM_MAPIO_SUBMITTED;
}
- r = bio_detain(cache, block, bio, cell,
+ r = bio_detain(cache, block, bio, *cell,
(cell_free_fn) free_prison_cell,
- cache, &cell);
+ cache, cell);
if (r) {
if (r < 0)
defer_bio(cache, bio);
r = policy_map(cache->policy, block, false, can_migrate, discarded_block,
bio, &lookup_result);
if (r == -EWOULDBLOCK) {
- cell_defer(cache, cell, true);
+ cell_defer(cache, *cell, true);
return DM_MAPIO_SUBMITTED;
} else if (r) {
DMERR_LIMIT("Unexpected return from cache replacement policy: %d", r);
+ cell_defer(cache, *cell, false);
bio_io_error(bio);
return DM_MAPIO_SUBMITTED;
}
* We need to invalidate this block, so
* defer for the worker thread.
*/
- cell_defer(cache, cell, true);
+ cell_defer(cache, *cell, true);
r = DM_MAPIO_SUBMITTED;
} else {
- pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
inc_miss_counter(cache, bio);
remap_to_origin_clear_discard(cache, bio, block);
-
- cell_defer(cache, cell, false);
}
} else {
inc_hit_counter(cache, bio);
- pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
-
if (bio_data_dir(bio) == WRITE && writethrough_mode(&cache->features) &&
!is_dirty(cache, lookup_result.cblock))
remap_to_origin_then_cache(cache, bio, block, lookup_result.cblock);
else
remap_to_cache_dirty(cache, bio, block, lookup_result.cblock);
-
- cell_defer(cache, cell, false);
}
break;
case POLICY_MISS:
inc_miss_counter(cache, bio);
- pb->all_io_entry = dm_deferred_entry_inc(cache->all_io_ds);
-
if (pb->req_nr != 0) {
/*
* This is a duplicate writethrough io that is no
* longer needed because the block has been demoted.
*/
bio_endio(bio, 0);
- cell_defer(cache, cell, false);
- return DM_MAPIO_SUBMITTED;
- } else {
+ cell_defer(cache, *cell, false);
+ r = DM_MAPIO_SUBMITTED;
+
+ } else
remap_to_origin_clear_discard(cache, bio, block);
- cell_defer(cache, cell, false);
- }
+
break;
default:
DMERR_LIMIT("%s: erroring bio: unknown policy op: %u", __func__,
(unsigned) lookup_result.op);
+ cell_defer(cache, *cell, false);
bio_io_error(bio);
r = DM_MAPIO_SUBMITTED;
}
return r;
}
+static int cache_map(struct dm_target *ti, struct bio *bio)
+{
+ int r;
+ struct dm_bio_prison_cell *cell;
+ struct cache *cache = ti->private;
+
+ r = __cache_map(cache, bio, &cell);
+ if (r == DM_MAPIO_REMAPPED) {
+ inc_ds(cache, bio, cell);
+ cell_defer(cache, cell, false);
+ }
+
+ return r;
+}
+
static int cache_end_io(struct dm_target *ti, struct bio *bio, int error)
{
struct cache *cache = ti->private;
residency = policy_residency(cache->policy);
DMEMIT("%u %llu/%llu %u %llu/%llu %u %u %u %u %u %u %lu ",
- (unsigned)(DM_CACHE_METADATA_BLOCK_SIZE >> SECTOR_SHIFT),
+ (unsigned)DM_CACHE_METADATA_BLOCK_SIZE,
(unsigned long long)(nr_blocks_metadata - nr_free_blocks_metadata),
(unsigned long long)nr_blocks_metadata,
cache->sectors_per_block,
*/
if (io_opt_sectors < cache->sectors_per_block ||
do_div(io_opt_sectors, cache->sectors_per_block)) {
- blk_limits_io_min(limits, 0);
+ blk_limits_io_min(limits, cache->sectors_per_block << SECTOR_SHIFT);
blk_limits_io_opt(limits, cache->sectors_per_block << SECTOR_SHIFT);
}
set_discard_limits(cache, limits);
static struct target_type cache_target = {
.name = "cache",
- .version = {1, 4, 0},
+ .version = {1, 5, 0},
.module = THIS_MODULE,
.ctr = cache_ctr,
.dtr = cache_dtr,
int error;
sector_t sector;
struct dm_crypt_io *base_io;
-};
+} CRYPTO_MINALIGN_ATTR;
struct dm_crypt_request {
struct convert_context *ctx;
*/
unsigned int dmreq_start;
+ unsigned int per_bio_data_size;
+
unsigned long flags;
unsigned int key_size;
unsigned int key_parts; /* independent parts in key buffer */
kcryptd_async_done, dmreq_of_req(cc, ctx->req));
}
+static void crypt_free_req(struct crypt_config *cc,
+ struct ablkcipher_request *req, struct bio *base_bio)
+{
+ struct dm_crypt_io *io = dm_per_bio_data(base_bio, cc->per_bio_data_size);
+
+ if ((struct ablkcipher_request *)(io + 1) != req)
+ mempool_free(req, cc->req_pool);
+}
+
/*
* Encrypt / decrypt data from one bio to another one (can be the same one)
*/
}
}
-static struct dm_crypt_io *crypt_io_alloc(struct crypt_config *cc,
- struct bio *bio, sector_t sector)
+static void crypt_io_init(struct dm_crypt_io *io, struct crypt_config *cc,
+ struct bio *bio, sector_t sector)
{
- struct dm_crypt_io *io;
-
- io = mempool_alloc(cc->io_pool, GFP_NOIO);
io->cc = cc;
io->base_bio = bio;
io->sector = sector;
io->base_io = NULL;
io->ctx.req = NULL;
atomic_set(&io->io_pending, 0);
-
- return io;
}
static void crypt_inc_pending(struct dm_crypt_io *io)
return;
if (io->ctx.req)
- mempool_free(io->ctx.req, cc->req_pool);
- mempool_free(io, cc->io_pool);
+ crypt_free_req(cc, io->ctx.req, base_bio);
+ if (io != dm_per_bio_data(base_bio, cc->per_bio_data_size))
+ mempool_free(io, cc->io_pool);
if (likely(!base_io))
bio_endio(base_bio, error);
* between fragments, so switch to a new dm_crypt_io structure.
*/
if (unlikely(!crypt_finished && remaining)) {
- new_io = crypt_io_alloc(io->cc, io->base_bio,
- sector);
+ new_io = mempool_alloc(cc->io_pool, GFP_NOIO);
+ crypt_io_init(new_io, io->cc, io->base_bio, sector);
crypt_inc_pending(new_io);
crypt_convert_init(cc, &new_io->ctx, NULL,
io->base_bio, sector);
if (error < 0)
io->error = -EIO;
- mempool_free(req_of_dmreq(cc, dmreq), cc->req_pool);
+ crypt_free_req(cc, req_of_dmreq(cc, dmreq), io->base_bio);
if (!atomic_dec_and_test(&ctx->cc_pending))
return;
goto bad;
}
+ cc->per_bio_data_size = ti->per_bio_data_size =
+ sizeof(struct dm_crypt_io) + cc->dmreq_start +
+ sizeof(struct dm_crypt_request) + cc->iv_size;
+
cc->page_pool = mempool_create_page_pool(MIN_POOL_PAGES, 0);
if (!cc->page_pool) {
ti->error = "Cannot allocate page mempool";
return DM_MAPIO_REMAPPED;
}
- io = crypt_io_alloc(cc, bio, dm_target_offset(ti, bio->bi_iter.bi_sector));
+ io = dm_per_bio_data(bio, cc->per_bio_data_size);
+ crypt_io_init(io, cc, bio, dm_target_offset(ti, bio->bi_iter.bi_sector));
+ io->ctx.req = (struct ablkcipher_request *)(io + 1);
if (bio_data_dir(io->base_bio) == READ) {
if (kcryptd_io_read(io, GFP_NOWAIT))
struct io {
unsigned long error_bits;
atomic_t count;
- struct completion *wait;
struct dm_io_client *client;
io_notify_fn callback;
void *context;
* We need an io object to keep track of the number of bios that
* have been dispatched for a particular io.
*---------------------------------------------------------------*/
-static void dec_count(struct io *io, unsigned int region, int error)
+static void complete_io(struct io *io)
{
- if (error)
- set_bit(region, &io->error_bits);
+ unsigned long error_bits = io->error_bits;
+ io_notify_fn fn = io->callback;
+ void *context = io->context;
- if (atomic_dec_and_test(&io->count)) {
- if (io->vma_invalidate_size)
- invalidate_kernel_vmap_range(io->vma_invalidate_address,
- io->vma_invalidate_size);
+ if (io->vma_invalidate_size)
+ invalidate_kernel_vmap_range(io->vma_invalidate_address,
+ io->vma_invalidate_size);
- if (io->wait)
- complete(io->wait);
+ mempool_free(io, io->client->pool);
+ fn(error_bits, context);
+}
- else {
- unsigned long r = io->error_bits;
- io_notify_fn fn = io->callback;
- void *context = io->context;
+static void dec_count(struct io *io, unsigned int region, int error)
+{
+ if (error)
+ set_bit(region, &io->error_bits);
- mempool_free(io, io->client->pool);
- fn(r, context);
- }
- }
+ if (atomic_dec_and_test(&io->count))
+ complete_io(io);
}
static void endio(struct bio *bio, int error)
dec_count(io, 0, 0);
}
+struct sync_io {
+ unsigned long error_bits;
+ struct completion wait;
+};
+
+static void sync_io_complete(unsigned long error, void *context)
+{
+ struct sync_io *sio = context;
+
+ sio->error_bits = error;
+ complete(&sio->wait);
+}
+
static int sync_io(struct dm_io_client *client, unsigned int num_regions,
struct dm_io_region *where, int rw, struct dpages *dp,
unsigned long *error_bits)
{
- /*
- * gcc <= 4.3 can't do the alignment for stack variables, so we must
- * align it on our own.
- * volatile prevents the optimizer from removing or reusing
- * "io_" field from the stack frame (allowed in ANSI C).
- */
- volatile char io_[sizeof(struct io) + __alignof__(struct io) - 1];
- struct io *io = (struct io *)PTR_ALIGN(&io_, __alignof__(struct io));
- DECLARE_COMPLETION_ONSTACK(wait);
+ struct io *io;
+ struct sync_io sio;
if (num_regions > 1 && (rw & RW_MASK) != WRITE) {
WARN_ON(1);
return -EIO;
}
+ init_completion(&sio.wait);
+
+ io = mempool_alloc(client->pool, GFP_NOIO);
io->error_bits = 0;
atomic_set(&io->count, 1); /* see dispatch_io() */
- io->wait = &wait;
io->client = client;
+ io->callback = sync_io_complete;
+ io->context = &sio;
io->vma_invalidate_address = dp->vma_invalidate_address;
io->vma_invalidate_size = dp->vma_invalidate_size;
dispatch_io(rw, num_regions, where, dp, io, 1);
- wait_for_completion_io(&wait);
+ wait_for_completion_io(&sio.wait);
if (error_bits)
- *error_bits = io->error_bits;
+ *error_bits = sio.error_bits;
- return io->error_bits ? -EIO : 0;
+ return sio.error_bits ? -EIO : 0;
}
static int async_io(struct dm_io_client *client, unsigned int num_regions,
io = mempool_alloc(client->pool, GFP_NOIO);
io->error_bits = 0;
atomic_set(&io->count, 1); /* see dispatch_io() */
- io->wait = NULL;
io->client = client;
io->callback = fn;
io->context = context;
* New collapsed (a)synchronous interface.
*
* If the IO is asynchronous (i.e. it has notify.fn), you must either unplug
- * the queue with blk_unplug() some time later or set REQ_SYNC in
-io_req->bi_rw. If you fail to do one of these, the IO will be submitted to
- * the disk after q->unplug_delay, which defaults to 3ms in blk-settings.c.
+ * the queue with blk_unplug() some time later or set REQ_SYNC in io_req->bi_rw.
+ * If you fail to do one of these, the IO will be submitted to the disk after
+ * q->unplug_delay, which defaults to 3ms in blk-settings.c.
*/
int dm_io(struct dm_io_request *io_req, unsigned num_regions,
struct dm_io_region *where, unsigned long *sync_error_bits)
dm_noflush_suspending(m->ti)));
}
-#define pg_ready(m) (!(m)->queue_io && !(m)->pg_init_required)
-
/*
* Map cloned requests
*/
if (!__must_push_back(m))
r = -EIO; /* Failed */
goto out_unlock;
- }
- if (!pg_ready(m)) {
+ } else if (m->queue_io || m->pg_init_required) {
__pg_init_all_paths(m);
goto out_unlock;
}
+
if (set_mapinfo(m, map_context) < 0)
/* ENOMEM, requeue */
goto out_unlock;
*bit *= sctx->region_table_entry_bits;
}
+static unsigned switch_region_table_read(struct switch_ctx *sctx, unsigned long region_nr)
+{
+ unsigned long region_index;
+ unsigned bit;
+
+ switch_get_position(sctx, region_nr, ®ion_index, &bit);
+
+ return (ACCESS_ONCE(sctx->region_table[region_index]) >> bit) &
+ ((1 << sctx->region_table_entry_bits) - 1);
+}
+
/*
* Find which path to use at given offset.
*/
static unsigned switch_get_path_nr(struct switch_ctx *sctx, sector_t offset)
{
- unsigned long region_index;
- unsigned bit, path_nr;
+ unsigned path_nr;
sector_t p;
p = offset;
else
sector_div(p, sctx->region_size);
- switch_get_position(sctx, p, ®ion_index, &bit);
- path_nr = (ACCESS_ONCE(sctx->region_table[region_index]) >> bit) &
- ((1 << sctx->region_table_entry_bits) - 1);
+ path_nr = switch_region_table_read(sctx, p);
/* This can only happen if the processor uses non-atomic stores. */
if (unlikely(path_nr >= sctx->nr_paths))
}
static int process_set_region_mappings(struct switch_ctx *sctx,
- unsigned argc, char **argv)
+ unsigned argc, char **argv)
{
unsigned i;
unsigned long region_index = 0;
unsigned long path_nr;
const char *string = argv[i];
+ if ((*string & 0xdf) == 'R') {
+ unsigned long cycle_length, num_write;
+
+ string++;
+ if (unlikely(*string == ',')) {
+ DMWARN("invalid set_region_mappings argument: '%s'", argv[i]);
+ return -EINVAL;
+ }
+ cycle_length = parse_hex(&string);
+ if (unlikely(*string != ',')) {
+ DMWARN("invalid set_region_mappings argument: '%s'", argv[i]);
+ return -EINVAL;
+ }
+ string++;
+ if (unlikely(!*string)) {
+ DMWARN("invalid set_region_mappings argument: '%s'", argv[i]);
+ return -EINVAL;
+ }
+ num_write = parse_hex(&string);
+ if (unlikely(*string)) {
+ DMWARN("invalid set_region_mappings argument: '%s'", argv[i]);
+ return -EINVAL;
+ }
+
+ if (unlikely(!cycle_length) || unlikely(cycle_length - 1 > region_index)) {
+ DMWARN("invalid set_region_mappings cycle length: %lu > %lu",
+ cycle_length - 1, region_index);
+ return -EINVAL;
+ }
+ if (unlikely(region_index + num_write < region_index) ||
+ unlikely(region_index + num_write >= sctx->nr_regions)) {
+ DMWARN("invalid set_region_mappings region number: %lu + %lu >= %lu",
+ region_index, num_write, sctx->nr_regions);
+ return -EINVAL;
+ }
+
+ while (num_write--) {
+ region_index++;
+ path_nr = switch_region_table_read(sctx, region_index - cycle_length);
+ switch_region_table_write(sctx, region_index, path_nr);
+ }
+
+ continue;
+ }
+
if (*string == ':')
region_index++;
else {
static struct target_type switch_target = {
.name = "switch",
- .version = {1, 0, 0},
+ .version = {1, 1, 0},
.module = THIS_MODULE,
.ctr = switch_ctr,
.dtr = switch_dtr,
return q && !blk_queue_add_random(q);
}
+static int queue_supports_sg_merge(struct dm_target *ti, struct dm_dev *dev,
+ sector_t start, sector_t len, void *data)
+{
+ struct request_queue *q = bdev_get_queue(dev->bdev);
+
+ return q && !test_bit(QUEUE_FLAG_NO_SG_MERGE, &q->queue_flags);
+}
+
static bool dm_table_all_devices_attribute(struct dm_table *t,
iterate_devices_callout_fn func)
{
return true;
}
+static int device_discard_capable(struct dm_target *ti, struct dm_dev *dev,
+ sector_t start, sector_t len, void *data)
+{
+ struct request_queue *q = bdev_get_queue(dev->bdev);
+
+ return q && blk_queue_discard(q);
+}
+
+static bool dm_table_supports_discards(struct dm_table *t)
+{
+ struct dm_target *ti;
+ unsigned i = 0;
+
+ /*
+ * Unless any target used by the table set discards_supported,
+ * require at least one underlying device to support discards.
+ * t->devices includes internal dm devices such as mirror logs
+ * so we need to use iterate_devices here, which targets
+ * supporting discard selectively must provide.
+ */
+ while (i < dm_table_get_num_targets(t)) {
+ ti = dm_table_get_target(t, i++);
+
+ if (!ti->num_discard_bios)
+ continue;
+
+ if (ti->discards_supported)
+ return 1;
+
+ if (ti->type->iterate_devices &&
+ ti->type->iterate_devices(ti, device_discard_capable, NULL))
+ return 1;
+ }
+
+ return 0;
+}
+
void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
struct queue_limits *limits)
{
if (!dm_table_supports_write_same(t))
q->limits.max_write_same_sectors = 0;
+ if (dm_table_all_devices_attribute(t, queue_supports_sg_merge))
+ queue_flag_clear_unlocked(QUEUE_FLAG_NO_SG_MERGE, q);
+ else
+ queue_flag_set_unlocked(QUEUE_FLAG_NO_SG_MERGE, q);
+
dm_table_set_integrity(t);
/*
}
EXPORT_SYMBOL(dm_table_run_md_queue_async);
-static int device_discard_capable(struct dm_target *ti, struct dm_dev *dev,
- sector_t start, sector_t len, void *data)
-{
- struct request_queue *q = bdev_get_queue(dev->bdev);
-
- return q && blk_queue_discard(q);
-}
-
-bool dm_table_supports_discards(struct dm_table *t)
-{
- struct dm_target *ti;
- unsigned i = 0;
-
- /*
- * Unless any target used by the table set discards_supported,
- * require at least one underlying device to support discards.
- * t->devices includes internal dm devices such as mirror logs
- * so we need to use iterate_devices here, which targets
- * supporting discard selectively must provide.
- */
- while (i < dm_table_get_num_targets(t)) {
- ti = dm_table_get_target(t, i++);
-
- if (!ti->num_discard_bios)
- continue;
-
- if (ti->discards_supported)
- return 1;
-
- if (ti->type->iterate_devices &&
- ti->type->iterate_devices(ti, device_discard_capable, NULL))
- return 1;
- }
-
- return 0;
-}
struct list_head list;
struct dm_dev *pool_dev;
struct dm_dev *origin_dev;
+ sector_t origin_size;
dm_thin_id dev_id;
struct pool *pool;
struct dm_thin_new_mapping {
struct list_head list;
- bool quiesced:1;
- bool prepared:1;
bool pass_discard:1;
bool definitely_not_shared:1;
+ /*
+ * Track quiescing, copying and zeroing preparation actions. When this
+ * counter hits zero the block is prepared and can be inserted into the
+ * btree.
+ */
+ atomic_t prepare_actions;
+
int err;
struct thin_c *tc;
dm_block_t virt_block;
bio_end_io_t *saved_bi_end_io;
};
-static void __maybe_add_mapping(struct dm_thin_new_mapping *m)
+static void __complete_mapping_preparation(struct dm_thin_new_mapping *m)
{
struct pool *pool = m->tc->pool;
- if (m->quiesced && m->prepared) {
+ if (atomic_dec_and_test(&m->prepare_actions)) {
list_add_tail(&m->list, &pool->prepared_mappings);
wake_worker(pool);
}
}
-static void copy_complete(int read_err, unsigned long write_err, void *context)
+static void complete_mapping_preparation(struct dm_thin_new_mapping *m)
{
unsigned long flags;
- struct dm_thin_new_mapping *m = context;
struct pool *pool = m->tc->pool;
- m->err = read_err || write_err ? -EIO : 0;
-
spin_lock_irqsave(&pool->lock, flags);
- m->prepared = true;
- __maybe_add_mapping(m);
+ __complete_mapping_preparation(m);
spin_unlock_irqrestore(&pool->lock, flags);
}
+static void copy_complete(int read_err, unsigned long write_err, void *context)
+{
+ struct dm_thin_new_mapping *m = context;
+
+ m->err = read_err || write_err ? -EIO : 0;
+ complete_mapping_preparation(m);
+}
+
static void overwrite_endio(struct bio *bio, int err)
{
- unsigned long flags;
struct dm_thin_endio_hook *h = dm_per_bio_data(bio, sizeof(struct dm_thin_endio_hook));
struct dm_thin_new_mapping *m = h->overwrite_mapping;
- struct pool *pool = m->tc->pool;
m->err = err;
-
- spin_lock_irqsave(&pool->lock, flags);
- m->prepared = true;
- __maybe_add_mapping(m);
- spin_unlock_irqrestore(&pool->lock, flags);
+ complete_mapping_preparation(m);
}
/*----------------------------------------------------------------*/
return m;
}
+static void ll_zero(struct thin_c *tc, struct dm_thin_new_mapping *m,
+ sector_t begin, sector_t end)
+{
+ int r;
+ struct dm_io_region to;
+
+ to.bdev = tc->pool_dev->bdev;
+ to.sector = begin;
+ to.count = end - begin;
+
+ r = dm_kcopyd_zero(tc->pool->copier, 1, &to, 0, copy_complete, m);
+ if (r < 0) {
+ DMERR_LIMIT("dm_kcopyd_zero() failed");
+ copy_complete(1, 1, m);
+ }
+}
+
+/*
+ * A partial copy also needs to zero the uncopied region.
+ */
static void schedule_copy(struct thin_c *tc, dm_block_t virt_block,
struct dm_dev *origin, dm_block_t data_origin,
dm_block_t data_dest,
- struct dm_bio_prison_cell *cell, struct bio *bio)
+ struct dm_bio_prison_cell *cell, struct bio *bio,
+ sector_t len)
{
int r;
struct pool *pool = tc->pool;
m->data_block = data_dest;
m->cell = cell;
+ /*
+ * quiesce action + copy action + an extra reference held for the
+ * duration of this function (we may need to inc later for a
+ * partial zero).
+ */
+ atomic_set(&m->prepare_actions, 3);
+
if (!dm_deferred_set_add_work(pool->shared_read_ds, &m->list))
- m->quiesced = true;
+ complete_mapping_preparation(m); /* already quiesced */
/*
* IO to pool_dev remaps to the pool target's data_dev.
from.bdev = origin->bdev;
from.sector = data_origin * pool->sectors_per_block;
- from.count = pool->sectors_per_block;
+ from.count = len;
to.bdev = tc->pool_dev->bdev;
to.sector = data_dest * pool->sectors_per_block;
- to.count = pool->sectors_per_block;
+ to.count = len;
r = dm_kcopyd_copy(pool->copier, &from, 1, &to,
0, copy_complete, m);
if (r < 0) {
- mempool_free(m, pool->mapping_pool);
DMERR_LIMIT("dm_kcopyd_copy() failed");
- cell_error(pool, cell);
+ copy_complete(1, 1, m);
+
+ /*
+ * We allow the zero to be issued, to simplify the
+ * error path. Otherwise we'd need to start
+ * worrying about decrementing the prepare_actions
+ * counter.
+ */
+ }
+
+ /*
+ * Do we need to zero a tail region?
+ */
+ if (len < pool->sectors_per_block && pool->pf.zero_new_blocks) {
+ atomic_inc(&m->prepare_actions);
+ ll_zero(tc, m,
+ data_dest * pool->sectors_per_block + len,
+ (data_dest + 1) * pool->sectors_per_block);
}
}
+
+ complete_mapping_preparation(m); /* drop our ref */
}
static void schedule_internal_copy(struct thin_c *tc, dm_block_t virt_block,
struct dm_bio_prison_cell *cell, struct bio *bio)
{
schedule_copy(tc, virt_block, tc->pool_dev,
- data_origin, data_dest, cell, bio);
-}
-
-static void schedule_external_copy(struct thin_c *tc, dm_block_t virt_block,
- dm_block_t data_dest,
- struct dm_bio_prison_cell *cell, struct bio *bio)
-{
- schedule_copy(tc, virt_block, tc->origin_dev,
- virt_block, data_dest, cell, bio);
+ data_origin, data_dest, cell, bio,
+ tc->pool->sectors_per_block);
}
static void schedule_zero(struct thin_c *tc, dm_block_t virt_block,
struct pool *pool = tc->pool;
struct dm_thin_new_mapping *m = get_next_mapping(pool);
- m->quiesced = true;
- m->prepared = false;
+ atomic_set(&m->prepare_actions, 1); /* no need to quiesce */
m->tc = tc;
m->virt_block = virt_block;
m->data_block = data_block;
save_and_set_endio(bio, &m->saved_bi_end_io, overwrite_endio);
inc_all_io_entry(pool, bio);
remap_and_issue(tc, bio, data_block);
- } else {
- int r;
- struct dm_io_region to;
- to.bdev = tc->pool_dev->bdev;
- to.sector = data_block * pool->sectors_per_block;
- to.count = pool->sectors_per_block;
+ } else
+ ll_zero(tc, m,
+ data_block * pool->sectors_per_block,
+ (data_block + 1) * pool->sectors_per_block);
+}
- r = dm_kcopyd_zero(pool->copier, 1, &to, 0, copy_complete, m);
- if (r < 0) {
- mempool_free(m, pool->mapping_pool);
- DMERR_LIMIT("dm_kcopyd_zero() failed");
- cell_error(pool, cell);
- }
- }
+static void schedule_external_copy(struct thin_c *tc, dm_block_t virt_block,
+ dm_block_t data_dest,
+ struct dm_bio_prison_cell *cell, struct bio *bio)
+{
+ struct pool *pool = tc->pool;
+ sector_t virt_block_begin = virt_block * pool->sectors_per_block;
+ sector_t virt_block_end = (virt_block + 1) * pool->sectors_per_block;
+
+ if (virt_block_end <= tc->origin_size)
+ schedule_copy(tc, virt_block, tc->origin_dev,
+ virt_block, data_dest, cell, bio,
+ pool->sectors_per_block);
+
+ else if (virt_block_begin < tc->origin_size)
+ schedule_copy(tc, virt_block, tc->origin_dev,
+ virt_block, data_dest, cell, bio,
+ tc->origin_size - virt_block_begin);
+
+ else
+ schedule_zero(tc, virt_block, data_dest, cell, bio);
}
/*
inc_all_io_entry(pool, bio);
cell_defer_no_holder(tc, cell);
- remap_to_origin_and_issue(tc, bio);
+ if (bio_end_sector(bio) <= tc->origin_size)
+ remap_to_origin_and_issue(tc, bio);
+
+ else if (bio->bi_iter.bi_sector < tc->origin_size) {
+ zero_fill_bio(bio);
+ bio->bi_iter.bi_size = (tc->origin_size - bio->bi_iter.bi_sector) << SECTOR_SHIFT;
+ remap_to_origin_and_issue(tc, bio);
+
+ } else {
+ zero_fill_bio(bio);
+ bio_endio(bio, 0);
+ }
} else
provision_block(tc, bio, block, cell);
break;
*/
if (io_opt_sectors < pool->sectors_per_block ||
do_div(io_opt_sectors, pool->sectors_per_block)) {
- blk_limits_io_min(limits, 0);
+ blk_limits_io_min(limits, pool->sectors_per_block << SECTOR_SHIFT);
blk_limits_io_opt(limits, pool->sectors_per_block << SECTOR_SHIFT);
}
.name = "thin-pool",
.features = DM_TARGET_SINGLETON | DM_TARGET_ALWAYS_WRITEABLE |
DM_TARGET_IMMUTABLE,
- .version = {1, 12, 0},
+ .version = {1, 13, 0},
.module = THIS_MODULE,
.ctr = pool_ctr,
.dtr = pool_dtr,
spin_lock_irqsave(&pool->lock, flags);
list_for_each_entry_safe(m, tmp, &work, list) {
list_del(&m->list);
- m->quiesced = true;
- __maybe_add_mapping(m);
+ __complete_mapping_preparation(m);
}
spin_unlock_irqrestore(&pool->lock, flags);
}
noflush_work(tc, do_noflush_stop);
}
+static int thin_preresume(struct dm_target *ti)
+{
+ struct thin_c *tc = ti->private;
+
+ if (tc->origin_dev)
+ tc->origin_size = get_dev_size(tc->origin_dev->bdev);
+
+ return 0;
+}
+
/*
* <nr mapped sectors> <highest mapped sector>
*/
static struct target_type thin_target = {
.name = "thin",
- .version = {1, 12, 0},
+ .version = {1, 13, 0},
.module = THIS_MODULE,
.ctr = thin_ctr,
.dtr = thin_dtr,
.map = thin_map,
.end_io = thin_endio,
+ .preresume = thin_preresume,
.presuspend = thin_presuspend,
.postsuspend = thin_postsuspend,
.status = thin_status,
unsigned dm_table_get_type(struct dm_table *t);
struct target_type *dm_table_get_immutable_target_type(struct dm_table *t);
bool dm_table_request_based(struct dm_table *t);
-bool dm_table_supports_discards(struct dm_table *t);
void dm_table_free_md_mempools(struct dm_table *t);
struct dm_md_mempools *dm_table_get_md_mempools(struct dm_table *t);
*/
if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery)) {
end_reshape(conf);
+ close_sync(conf);
return 0;
}
}
r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ r10_bio->state = 0;
raise_barrier(conf, rb2 != NULL);
atomic_set(&r10_bio->remaining, 0);
if (sync_blocks < max_sync)
max_sync = sync_blocks;
r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ r10_bio->state = 0;
r10_bio->mddev = mddev;
atomic_set(&r10_bio->remaining, 0);
read_more:
/* Now schedule reads for blocks from sector_nr to last */
r10_bio = mempool_alloc(conf->r10buf_pool, GFP_NOIO);
+ r10_bio->state = 0;
raise_barrier(conf, sectors_done != 0);
atomic_set(&r10_bio->remaining, 0);
r10_bio->mddev = mddev;
* on all the target devices.
*/
// FIXME
+ mempool_free(r10_bio, conf->r10buf_pool);
set_bit(MD_RECOVERY_INTR, &mddev->recovery);
return sectors_done;
}
read_bio->bi_private = r10_bio;
read_bio->bi_end_io = end_sync_read;
read_bio->bi_rw = READ;
- read_bio->bi_flags &= ~(BIO_POOL_MASK - 1);
+ read_bio->bi_flags &= (~0UL << BIO_RESET_BITS);
read_bio->bi_flags |= 1 << BIO_UPTODATE;
read_bio->bi_vcnt = 0;
read_bio->bi_iter.bi_size = 0;
(!test_bit(R5_Insync, &dev->flags) || test_bit(STRIPE_PREREAD_ACTIVE, &sh->state)) &&
!test_bit(R5_OVERWRITE, &fdev[0]->flags)) ||
(sh->raid_conf->level == 6 && s->failed && s->to_write &&
- s->to_write < sh->raid_conf->raid_disks - 2 &&
+ s->to_write - s->non_overwrite < sh->raid_conf->raid_disks - 2 &&
(!test_bit(R5_Insync, &dev->flags) || test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))))) {
/* we would like to get this block, possibly by computing it,
* otherwise read it if the backing disk is insync
set_bit(R5_Wantwrite, &dev->flags);
if (prexor)
continue;
+ if (s.failed > 1)
+ continue;
if (!test_bit(R5_Insync, &dev->flags) ||
((i == sh->pd_idx || i == sh->qd_idx) &&
s.failed == 0))
{ USB_DEVICE(0x0BDA, 0x0140) },
{ }
};
+MODULE_DEVICE_TABLE(usb, rtsx_usb_usb_ids);
static struct usb_driver rtsx_usb_driver = {
.name = "rtsx_usb",
/* complete ongoing async transfer before issuing discard */
if (card->host->areq)
mmc_blk_issue_rw_rq(mq, NULL);
- if (req->cmd_flags & REQ_SECURE &&
- !(card->quirks & MMC_QUIRK_SEC_ERASE_TRIM_BROKEN))
+ if (req->cmd_flags & REQ_SECURE)
ret = mmc_blk_issue_secdiscard_rq(mq, req);
else
ret = mmc_blk_issue_discard_rq(mq, req);
if (!(card->csd.cmdclass & CCC_BLOCK_READ))
return -ENODEV;
+ mmc_fixup_device(card, blk_fixups);
+
md = mmc_blk_alloc(card);
if (IS_ERR(md))
return PTR_ERR(md);
goto out;
mmc_set_drvdata(card, md);
- mmc_fixup_device(card, blk_fixups);
if (mmc_add_disk(md))
goto out;
#endif
#ifdef CONFIG_PM_RUNTIME
-
static int mmc_runtime_suspend(struct device *dev)
{
struct mmc_card *card = mmc_dev_to_card(dev);
return host->bus_ops->runtime_resume(host);
}
-
-static int mmc_runtime_idle(struct device *dev)
-{
- return 0;
-}
-
#endif /* !CONFIG_PM_RUNTIME */
static const struct dev_pm_ops mmc_bus_pm_ops = {
- SET_RUNTIME_PM_OPS(mmc_runtime_suspend, mmc_runtime_resume,
- mmc_runtime_idle)
+ SET_RUNTIME_PM_OPS(mmc_runtime_suspend, mmc_runtime_resume, NULL)
SET_SYSTEM_SLEEP_PM_OPS(mmc_bus_suspend, mmc_bus_resume)
};
int mmc_can_secure_erase_trim(struct mmc_card *card)
{
- if (card->ext_csd.sec_feature_support & EXT_CSD_SEC_ER_EN)
+ if ((card->ext_csd.sec_feature_support & EXT_CSD_SEC_ER_EN) &&
+ !(card->quirks & MMC_QUIRK_SEC_ERASE_TRIM_BROKEN))
return 1;
return 0;
}
}
}
+ /*
+ * The EXT_CSD format is meant to be forward compatible. As long
+ * as CSD_STRUCTURE does not change, all values for EXT_CSD_REV
+ * are authorized, see JEDEC JESD84-B50 section B.8.
+ */
card->ext_csd.rev = ext_csd[EXT_CSD_REV];
- if (card->ext_csd.rev > 7) {
- pr_err("%s: unrecognised EXT_CSD revision %d\n",
- mmc_hostname(card->host), card->ext_csd.rev);
- err = -EINVAL;
- goto out;
- }
card->ext_csd.raw_sectors[0] = ext_csd[EXT_CSD_SEC_CNT + 0];
card->ext_csd.raw_sectors[1] = ext_csd[EXT_CSD_SEC_CNT + 1];
(f->cis_device == card->cis.device ||
f->cis_device == (u16) SDIO_ANY_ID) &&
rev >= f->rev_start && rev <= f->rev_end) {
- dev_dbg(&card->dev, "calling %pF\n", f->vendor_fixup);
+ dev_dbg(&card->dev, "calling %pf\n", f->vendor_fixup);
f->vendor_fixup(card, f->data);
}
}
mmc_delay(10);
}
+ if (!i)
+ pr_err("%s: card never left busy state\n", mmc_hostname(host));
+
if (rocr && !mmc_host_is_spi(host))
*rocr = cmd.resp[0];
be found on some embedded hardware such as UC-7112-LX.
If you have a controller with this interface, say Y here.
+config MMC_SDHCI_ST
+ tristate "SDHCI support on STMicroelectronics SoC"
+ depends on ARCH_STI
+ depends on MMC_SDHCI_PLTFM
+ select MMC_SDHCI_IO_ACCESSORS
+ help
+ This selects the Secure Digital Host Controller Interface in
+ STMicroelectronics SoCs.
+
+ If you have a controller with this interface, say Y or M here.
+ If unsure, say N.
+
config MMC_OMAP
tristate "TI OMAP Multimedia Card Interface support"
depends on ARCH_OMAP
config MMC_OMAP_HS
tristate "TI OMAP High Speed Multimedia Card Interface support"
+ depends on HAS_DMA
depends on ARCH_OMAP2PLUS || COMPILE_TEST
help
This selects the TI OMAP High Speed Multimedia card Interface.
config MMC_SDHCI_MSM
tristate "Qualcomm SDHCI Controller Support"
- depends on ARCH_QCOM
+ depends on ARCH_QCOM || (ARM && COMPILE_TEST)
depends on MMC_SDHCI_PLTFM
help
This selects the Secure Digital Host Controller Interface (SDHCI)
config MMC_S3C
tristate "Samsung S3C SD/MMC Card Interface support"
depends on ARCH_S3C24XX
+ depends on S3C24XX_DMAC
help
This selects a driver for the MCI interface found in
Samsung's S3C2410, S3C2412, S3C2440, S3C2442 CPUs.
working properly and needs to be debugged before this
option is useful.
-config MMC_S3C_PIODMA
- bool "Support for both PIO and DMA"
- help
- Compile both the PIO and DMA transfer routines into the
- driver and let the platform select at run-time which one
- is best.
-
- See notes for the DMA option.
-
endchoice
config MMC_SDRICOH_CS
config MMC_SH_MMCIF
tristate "SuperH Internal MMCIF support"
- depends on MMC_BLOCK
+ depends on MMC_BLOCK && HAS_DMA
depends on SUPERH || ARCH_SHMOBILE || COMPILE_TEST
help
This selects the MMC Host Interface controller (MMCIF).
config MMC_USDHI6ROL0
tristate "Renesas USDHI6ROL0 SD/SDIO Host Controller support"
+ depends on HAS_DMA
help
This selects support for the Renesas USDHI6ROL0 SD/SDIO
Host Controller
obj-$(CONFIG_MMC_SDHCI_BCM_KONA) += sdhci-bcm-kona.o
obj-$(CONFIG_MMC_SDHCI_BCM2835) += sdhci-bcm2835.o
obj-$(CONFIG_MMC_SDHCI_MSM) += sdhci-msm.o
+obj-$(CONFIG_MMC_SDHCI_ST) += sdhci-st.o
ifeq ($(CONFIG_CB710_DEBUG),y)
CFLAGS-cb710-mmc += -DDEBUG
static SIMPLE_DEV_PM_OPS(dw_mci_pci_pmops, dw_mci_pci_suspend, dw_mci_pci_resume);
-static DEFINE_PCI_DEVICE_TABLE(dw_mci_pci_id) = {
+static const struct pci_device_id dw_mci_pci_id[] = {
{ PCI_DEVICE(SYNOPSYS_DW_MCI_VENDOR_ID, SYNOPSYS_DW_MCI_DEVICE_ID) },
{}
};
0xff, 0x77, 0x77, 0xff, 0x77, 0xbb, 0xdd, 0xee,
};
-static inline bool dw_mci_fifo_reset(struct dw_mci *host);
-static inline bool dw_mci_ctrl_all_reset(struct dw_mci *host);
+static bool dw_mci_reset(struct dw_mci *host);
#if defined(CONFIG_DEBUG_FS)
static int dw_mci_req_show(struct seq_file *s, void *v)
int gpio_ro = mmc_gpio_get_ro(mmc);
/* Use platform get_ro function, else try on board write protect */
- if (slot->quirks & DW_MCI_SLOT_QUIRK_NO_WRITE_PROTECT)
+ if ((slot->quirks & DW_MCI_SLOT_QUIRK_NO_WRITE_PROTECT) ||
+ (slot->host->quirks & DW_MCI_QUIRK_NO_WRITE_PROTECT))
read_only = 0;
else if (!IS_ERR_VALUE(gpio_ro))
read_only = gpio_ro;
* After an error, there may be data lingering
* in the FIFO
*/
- dw_mci_fifo_reset(host);
+ dw_mci_reset(host);
} else {
data->bytes_xfered = data->blocks * data->blksz;
data->error = 0;
/* CMD error in data command */
if (mrq->cmd->error && mrq->data)
- dw_mci_fifo_reset(host);
+ dw_mci_reset(host);
host->cmd = NULL;
host->data = NULL;
}
/* Power down slot */
- if (present == 0) {
- /* Clear down the FIFO */
- dw_mci_fifo_reset(host);
-#ifdef CONFIG_MMC_DW_IDMAC
- dw_mci_idmac_reset(host);
-#endif
-
- }
+ if (present == 0)
+ dw_mci_reset(host);
spin_unlock_bh(&host->lock);
/* get quirks */
for (idx = 0; idx < ARRAY_SIZE(of_slot_quirks); idx++)
- if (of_get_property(np, of_slot_quirks[idx].quirk, NULL))
+ if (of_get_property(np, of_slot_quirks[idx].quirk, NULL)) {
+ dev_warn(dev, "Slot quirk %s is deprecated\n",
+ of_slot_quirks[idx].quirk);
quirks |= of_slot_quirks[idx].id;
+ }
return quirks;
}
return false;
}
-static inline bool dw_mci_fifo_reset(struct dw_mci *host)
+static bool dw_mci_reset(struct dw_mci *host)
{
+ u32 flags = SDMMC_CTRL_RESET | SDMMC_CTRL_FIFO_RESET;
+ bool ret = false;
+
/*
* Reseting generates a block interrupt, hence setting
* the scatter-gather pointer to NULL.
host->sg = NULL;
}
- return dw_mci_ctrl_reset(host, SDMMC_CTRL_FIFO_RESET);
-}
+ if (host->use_dma)
+ flags |= SDMMC_CTRL_DMA_RESET;
-static inline bool dw_mci_ctrl_all_reset(struct dw_mci *host)
-{
- return dw_mci_ctrl_reset(host,
- SDMMC_CTRL_FIFO_RESET |
- SDMMC_CTRL_RESET |
- SDMMC_CTRL_DMA_RESET);
+ if (dw_mci_ctrl_reset(host, flags)) {
+ /*
+ * In all cases we clear the RAWINTS register to clear any
+ * interrupts.
+ */
+ mci_writel(host, RINTSTS, 0xFFFFFFFF);
+
+ /* if using dma we wait for dma_req to clear */
+ if (host->use_dma) {
+ unsigned long timeout = jiffies + msecs_to_jiffies(500);
+ u32 status;
+ do {
+ status = mci_readl(host, STATUS);
+ if (!(status & SDMMC_STATUS_DMA_REQ))
+ break;
+ cpu_relax();
+ } while (time_before(jiffies, timeout));
+
+ if (status & SDMMC_STATUS_DMA_REQ) {
+ dev_err(host->dev,
+ "%s: Timeout waiting for dma_req to "
+ "clear during reset\n", __func__);
+ goto ciu_out;
+ }
+
+ /* when using DMA next we reset the fifo again */
+ if (!dw_mci_ctrl_reset(host, SDMMC_CTRL_FIFO_RESET))
+ goto ciu_out;
+ }
+ } else {
+ /* if the controller reset bit did clear, then set clock regs */
+ if (!(mci_readl(host, CTRL) & SDMMC_CTRL_RESET)) {
+ dev_err(host->dev, "%s: fifo/dma reset bits didn't "
+ "clear but ciu was reset, doing clock update\n",
+ __func__);
+ goto ciu_out;
+ }
+ }
+
+#if IS_ENABLED(CONFIG_MMC_DW_IDMAC)
+ /* It is also recommended that we reset and reprogram idmac */
+ dw_mci_idmac_reset(host);
+#endif
+
+ ret = true;
+
+ciu_out:
+ /* After a CTRL reset we need to have CIU set clock registers */
+ mci_send_cmd(host->cur_slot, SDMMC_CMD_UPD_CLK, 0);
+
+ return ret;
}
#ifdef CONFIG_OF
{
.quirk = "broken-cd",
.id = DW_MCI_QUIRK_BROKEN_CARD_DETECTION,
+ }, {
+ .quirk = "disable-wp",
+ .id = DW_MCI_QUIRK_NO_WRITE_PROTECT,
},
};
}
/* Reset all blocks */
- if (!dw_mci_ctrl_all_reset(host))
+ if (!dw_mci_ctrl_reset(host, SDMMC_CTRL_ALL_RESET_FLAGS))
return -ENODEV;
host->dma_ops = host->pdata->dma_ops;
}
}
- if (!dw_mci_ctrl_all_reset(host)) {
+ if (!dw_mci_ctrl_reset(host, SDMMC_CTRL_ALL_RESET_FLAGS)) {
ret = -ENODEV;
return ret;
}
#define SDMMC_CMD_INDX(n) ((n) & 0x1F)
/* Status register defines */
#define SDMMC_GET_FCNT(x) (((x)>>17) & 0x1FFF)
+#define SDMMC_STATUS_DMA_REQ BIT(31)
/* FIFOTH register defines */
#define SDMMC_SET_FIFOTH(m, r, t) (((m) & 0x7) << 28 | \
((r) & 0xFFF) << 16 | \
/* Card read threshold */
#define SDMMC_SET_RD_THLD(v, x) (((v) & 0x1FFF) << 16 | (x))
+/* All ctrl reset bits */
+#define SDMMC_CTRL_ALL_RESET_FLAGS \
+ (SDMMC_CTRL_RESET | SDMMC_CTRL_FIFO_RESET | SDMMC_CTRL_DMA_RESET)
+
/* Register access macros */
#define mci_readl(dev, reg) \
__raw_readl((dev)->regs + SDMMC_##reg)
* struct variant_data - MMCI variant-specific quirks
* @clkreg: default value for MCICLOCK register
* @clkreg_enable: enable value for MMCICLOCK register
+ * @clkreg_8bit_bus_enable: enable value for 8 bit bus
+ * @clkreg_neg_edge_enable: enable value for inverted data/cmd output
* @datalength_bits: number of bits in the MMCIDATALENGTH register
* @fifosize: number of bytes that can be written when MMCI_TXFIFOEMPTY
* is asserted (likewise for RX)
* @fifohalfsize: number of bytes that can be written when MCI_TXFIFOHALFEMPTY
* is asserted (likewise for RX)
+ * @data_cmd_enable: enable value for data commands.
* @sdio: variant supports SDIO
* @st_clkdiv: true if using a ST-specific clock divider algorithm
+ * @datactrl_mask_ddrmode: ddr mode mask in datactrl register.
* @blksz_datactrl16: true if Block size is at b16..b30 position in datactrl register
+ * @blksz_datactrl4: true if Block size is at b4..b16 position in datactrl
+ * register
* @pwrreg_powerup: power up value for MMCIPOWER register
+ * @f_max: maximum clk frequency supported by the controller.
* @signal_direction: input/out direction of bus signals can be indicated
* @pwrreg_clkgate: MMCIPOWER register must be used to gate the clock
* @busy_detect: true if busy detection on dat0 is supported
* @pwrreg_nopower: bits in MMCIPOWER don't controls ext. power supply
+ * @explicit_mclk_control: enable explicit mclk control in driver.
+ * @qcom_fifo: enables qcom specific fifo pio read logic.
+ * @reversed_irq_handling: handle data irq before cmd irq.
*/
struct variant_data {
unsigned int clkreg;
unsigned int clkreg_enable;
+ unsigned int clkreg_8bit_bus_enable;
+ unsigned int clkreg_neg_edge_enable;
unsigned int datalength_bits;
unsigned int fifosize;
unsigned int fifohalfsize;
+ unsigned int data_cmd_enable;
+ unsigned int datactrl_mask_ddrmode;
bool sdio;
bool st_clkdiv;
bool blksz_datactrl16;
+ bool blksz_datactrl4;
u32 pwrreg_powerup;
+ u32 f_max;
bool signal_direction;
bool pwrreg_clkgate;
bool busy_detect;
bool pwrreg_nopower;
+ bool explicit_mclk_control;
+ bool qcom_fifo;
+ bool reversed_irq_handling;
};
static struct variant_data variant_arm = {
.fifohalfsize = 8 * 4,
.datalength_bits = 16,
.pwrreg_powerup = MCI_PWR_UP,
+ .f_max = 100000000,
+ .reversed_irq_handling = true,
};
static struct variant_data variant_arm_extended_fifo = {
.fifohalfsize = 64 * 4,
.datalength_bits = 16,
.pwrreg_powerup = MCI_PWR_UP,
+ .f_max = 100000000,
};
static struct variant_data variant_arm_extended_fifo_hwfc = {
.clkreg_enable = MCI_ARM_HWFCEN,
.datalength_bits = 16,
.pwrreg_powerup = MCI_PWR_UP,
+ .f_max = 100000000,
};
static struct variant_data variant_u300 = {
.fifosize = 16 * 4,
.fifohalfsize = 8 * 4,
.clkreg_enable = MCI_ST_U300_HWFCEN,
+ .clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
.datalength_bits = 16,
.sdio = true,
.pwrreg_powerup = MCI_PWR_ON,
+ .f_max = 100000000,
.signal_direction = true,
.pwrreg_clkgate = true,
.pwrreg_nopower = true,
.sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
+ .f_max = 100000000,
.signal_direction = true,
.pwrreg_clkgate = true,
.pwrreg_nopower = true,
.fifohalfsize = 8 * 4,
.clkreg = MCI_CLK_ENABLE,
.clkreg_enable = MCI_ST_UX500_HWFCEN,
+ .clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
+ .clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
.datalength_bits = 24,
.sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
+ .f_max = 100000000,
.signal_direction = true,
.pwrreg_clkgate = true,
.busy_detect = true,
.fifohalfsize = 8 * 4,
.clkreg = MCI_CLK_ENABLE,
.clkreg_enable = MCI_ST_UX500_HWFCEN,
+ .clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
+ .clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
+ .datactrl_mask_ddrmode = MCI_ST_DPSM_DDRMODE,
.datalength_bits = 24,
.sdio = true,
.st_clkdiv = true,
.blksz_datactrl16 = true,
.pwrreg_powerup = MCI_PWR_ON,
+ .f_max = 100000000,
.signal_direction = true,
.pwrreg_clkgate = true,
.busy_detect = true,
.pwrreg_nopower = true,
};
+static struct variant_data variant_qcom = {
+ .fifosize = 16 * 4,
+ .fifohalfsize = 8 * 4,
+ .clkreg = MCI_CLK_ENABLE,
+ .clkreg_enable = MCI_QCOM_CLK_FLOWENA |
+ MCI_QCOM_CLK_SELECT_IN_FBCLK,
+ .clkreg_8bit_bus_enable = MCI_QCOM_CLK_WIDEBUS_8,
+ .datactrl_mask_ddrmode = MCI_QCOM_CLK_SELECT_IN_DDR_MODE,
+ .data_cmd_enable = MCI_QCOM_CSPM_DATCMD,
+ .blksz_datactrl4 = true,
+ .datalength_bits = 24,
+ .pwrreg_powerup = MCI_PWR_UP,
+ .f_max = 208000000,
+ .explicit_mclk_control = true,
+ .qcom_fifo = true,
+};
+
static int mmci_card_busy(struct mmc_host *mmc)
{
struct mmci_host *host = mmc_priv(mmc);
host->cclk = 0;
if (desired) {
- if (desired >= host->mclk) {
+ if (variant->explicit_mclk_control) {
+ host->cclk = host->mclk;
+ } else if (desired >= host->mclk) {
clk = MCI_CLK_BYPASS;
if (variant->st_clkdiv)
clk |= MCI_ST_UX500_NEG_EDGE;
if (host->mmc->ios.bus_width == MMC_BUS_WIDTH_4)
clk |= MCI_4BIT_BUS;
if (host->mmc->ios.bus_width == MMC_BUS_WIDTH_8)
- clk |= MCI_ST_8BIT_BUS;
+ clk |= variant->clkreg_8bit_bus_enable;
if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50 ||
host->mmc->ios.timing == MMC_TIMING_MMC_DDR52)
- clk |= MCI_ST_UX500_NEG_EDGE;
+ clk |= variant->clkreg_neg_edge_enable;
mmci_write_clkreg(host, clk);
}
data->bytes_xfered = 0;
clks = (unsigned long long)data->timeout_ns * host->cclk;
- do_div(clks, 1000000000UL);
+ do_div(clks, NSEC_PER_SEC);
timeout = data->timeout_clks + (unsigned int)clks;
if (variant->blksz_datactrl16)
datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
+ else if (variant->blksz_datactrl4)
+ datactrl = MCI_DPSM_ENABLE | (data->blksz << 4);
else
datactrl = MCI_DPSM_ENABLE | blksz_bits << 4;
if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50 ||
host->mmc->ios.timing == MMC_TIMING_MMC_DDR52)
- datactrl |= MCI_ST_DPSM_DDRMODE;
+ datactrl |= variant->datactrl_mask_ddrmode;
/*
* Attempt to use DMA operation mode, if this
if (readl(base + MMCICOMMAND) & MCI_CPSM_ENABLE) {
writel(0, base + MMCICOMMAND);
- udelay(1);
+ mmci_reg_delay(host);
}
c |= cmd->opcode | MCI_CPSM_ENABLE;
if (/*interrupt*/0)
c |= MCI_CPSM_INTERRUPT;
+ if (mmc_cmd_type(cmd) == MMC_CMD_ADTC)
+ c |= host->variant->data_cmd_enable;
+
host->cmd = cmd;
writel(cmd->arg, base + MMCIARGUMENT);
mmci_data_irq(struct mmci_host *host, struct mmc_data *data,
unsigned int status)
{
+ /* Make sure we have data to handle */
+ if (!data)
+ return;
+
/* First check for errors */
if (status & (MCI_DATACRCFAIL|MCI_DATATIMEOUT|MCI_STARTBITERR|
MCI_TXUNDERRUN|MCI_RXOVERRUN)) {
unsigned int status)
{
void __iomem *base = host->base;
- bool sbc = (cmd == host->mrq->sbc);
- bool busy_resp = host->variant->busy_detect &&
- (cmd->flags & MMC_RSP_BUSY);
+ bool sbc, busy_resp;
+
+ if (!cmd)
+ return;
+
+ sbc = (cmd == host->mrq->sbc);
+ busy_resp = host->variant->busy_detect && (cmd->flags & MMC_RSP_BUSY);
+
+ if (!((status|host->busy_status) & (MCI_CMDCRCFAIL|MCI_CMDTIMEOUT|
+ MCI_CMDSENT|MCI_CMDRESPEND)))
+ return;
/* Check if we need to wait for busy completion. */
if (host->busy_status && (status & MCI_ST_CARDBUSY))
}
}
+static int mmci_get_rx_fifocnt(struct mmci_host *host, u32 status, int remain)
+{
+ return remain - (readl(host->base + MMCIFIFOCNT) << 2);
+}
+
+static int mmci_qcom_get_rx_fifocnt(struct mmci_host *host, u32 status, int r)
+{
+ /*
+ * on qcom SDCC4 only 8 words are used in each burst so only 8 addresses
+ * from the fifo range should be used
+ */
+ if (status & MCI_RXFIFOHALFFULL)
+ return host->variant->fifohalfsize;
+ else if (status & MCI_RXDATAAVLBL)
+ return 4;
+
+ return 0;
+}
+
static int mmci_pio_read(struct mmci_host *host, char *buffer, unsigned int remain)
{
void __iomem *base = host->base;
char *ptr = buffer;
- u32 status;
+ u32 status = readl(host->base + MMCISTATUS);
int host_remain = host->size;
do {
- int count = host_remain - (readl(base + MMCIFIFOCNT) << 2);
+ int count = host->get_rx_fifocnt(host, status, host_remain);
if (count > remain)
count = remain;
spin_lock(&host->lock);
do {
- struct mmc_command *cmd;
- struct mmc_data *data;
-
status = readl(host->base + MMCISTATUS);
if (host->singleirq) {
dev_dbg(mmc_dev(host->mmc), "irq0 (data+cmd) %08x\n", status);
- cmd = host->cmd;
- if ((status|host->busy_status) & (MCI_CMDCRCFAIL|MCI_CMDTIMEOUT|
- MCI_CMDSENT|MCI_CMDRESPEND) && cmd)
- mmci_cmd_irq(host, cmd, status);
-
- data = host->data;
- if (status & (MCI_DATACRCFAIL|MCI_DATATIMEOUT|MCI_STARTBITERR|
- MCI_TXUNDERRUN|MCI_RXOVERRUN|MCI_DATAEND|
- MCI_DATABLOCKEND) && data)
- mmci_data_irq(host, data, status);
+ if (host->variant->reversed_irq_handling) {
+ mmci_data_irq(host, host->data, status);
+ mmci_cmd_irq(host, host->cmd, status);
+ } else {
+ mmci_cmd_irq(host, host->cmd, status);
+ mmci_data_irq(host, host->data, status);
+ }
/* Don't poll for busy completion in irq context. */
if (host->busy_status)
if (!ios->clock && variant->pwrreg_clkgate)
pwr &= ~MCI_PWR_ON;
+ if (host->variant->explicit_mclk_control &&
+ ios->clock != host->clock_cache) {
+ ret = clk_set_rate(host->clk, ios->clock);
+ if (ret < 0)
+ dev_err(mmc_dev(host->mmc),
+ "Error setting clock rate (%d)\n", ret);
+ else
+ host->mclk = clk_get_rate(host->clk);
+ }
+ host->clock_cache = ios->clock;
+
spin_lock_irqsave(&host->lock, flags);
mmci_set_clkreg(host, ios->clock);
if (ret)
goto host_free;
+ if (variant->qcom_fifo)
+ host->get_rx_fifocnt = mmci_qcom_get_rx_fifocnt;
+ else
+ host->get_rx_fifocnt = mmci_get_rx_fifocnt;
+
host->plat = plat;
host->variant = variant;
host->mclk = clk_get_rate(host->clk);
* so we try to adjust the clock down to this,
* (if possible).
*/
- if (host->mclk > 100000000) {
- ret = clk_set_rate(host->clk, 100000000);
+ if (host->mclk > variant->f_max) {
+ ret = clk_set_rate(host->clk, variant->f_max);
if (ret < 0)
goto clk_disable;
host->mclk = clk_get_rate(host->clk);
* The ARM and ST versions of the block have slightly different
* clock divider equations which means that the minimum divider
* differs too.
+ * on Qualcomm like controllers get the nearest minimum clock to 100Khz
*/
if (variant->st_clkdiv)
mmc->f_min = DIV_ROUND_UP(host->mclk, 257);
+ else if (variant->explicit_mclk_control)
+ mmc->f_min = clk_round_rate(host->clk, 100000);
else
mmc->f_min = DIV_ROUND_UP(host->mclk, 512);
/*
* the block, of course.
*/
if (mmc->f_max)
- mmc->f_max = min(host->mclk, mmc->f_max);
+ mmc->f_max = variant->explicit_mclk_control ?
+ min(variant->f_max, mmc->f_max) :
+ min(host->mclk, mmc->f_max);
else
- mmc->f_max = min(host->mclk, fmax);
+ mmc->f_max = variant->explicit_mclk_control ?
+ fmax : min(host->mclk, fmax);
+
+
dev_dbg(mmc_dev(mmc), "clocking block at %u Hz\n", mmc->f_max);
/* Get regulators and the supported OCR mask */
.mask = 0xf0ffffff,
.data = &variant_ux500v2,
},
+ /* Qualcomm variants */
+ {
+ .id = 0x00051180,
+ .mask = 0x000fffff,
+ .data = &variant_qcom,
+ },
{ 0, 0 },
};
/* Modified PL180 on Versatile Express platform */
#define MCI_ARM_HWFCEN (1 << 12)
+/* Modified on Qualcomm Integrations */
+#define MCI_QCOM_CLK_WIDEBUS_8 (BIT(10) | BIT(11))
+#define MCI_QCOM_CLK_FLOWENA BIT(12)
+#define MCI_QCOM_CLK_INVERTOUT BIT(13)
+
+/* select in latch data and command in */
+#define MCI_QCOM_CLK_SELECT_IN_FBCLK BIT(15)
+#define MCI_QCOM_CLK_SELECT_IN_DDR_MODE (BIT(14) | BIT(15))
+
#define MMCIARGUMENT 0x008
#define MMCICOMMAND 0x00c
#define MCI_CPSM_RESPONSE (1 << 6)
#define MCI_ST_NIEN (1 << 13)
#define MCI_ST_CE_ATACMD (1 << 14)
+/* Modified on Qualcomm Integrations */
+#define MCI_QCOM_CSPM_DATCMD BIT(12)
+#define MCI_QCOM_CSPM_MCIABORT BIT(13)
+#define MCI_QCOM_CSPM_CCSENABLE BIT(14)
+#define MCI_QCOM_CSPM_CCSDISABLE BIT(15)
+#define MCI_QCOM_CSPM_AUTO_CMD19 BIT(16)
+#define MCI_QCOM_CSPM_AUTO_CMD21 BIT(21)
+
#define MMCIRESPCMD 0x010
#define MMCIRESPONSE0 0x014
#define MMCIRESPONSE1 0x018
spinlock_t lock;
unsigned int mclk;
+ /* cached value of requested clk in set_ios */
+ unsigned int clock_cache;
unsigned int cclk;
u32 pwr_reg;
u32 pwr_reg_add;
/* pio stuff */
struct sg_mapping_iter sg_miter;
unsigned int size;
+ int (*get_rx_fifocnt)(struct mmci_host *h, u32 status, int remain);
#ifdef CONFIG_DMA_ENGINE
/* DMA stuff */
* warranty of any kind, whether express or implied.
*/
-#include <linux/version.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/platform_device.h>
if (ret >= 0)
return ret;
- present = !(readl(ssp->base + HW_SSP_STATUS(ssp)) &
+ present = mmc->caps & MMC_CAP_NEEDS_POLL ||
+ !(readl(ssp->base + HW_SSP_STATUS(ssp)) &
BM_SSP_STATUS_CARD_DETECT);
if (mmc->caps2 & MMC_CAP2_CD_ACTIVE_HIGH)
#include <linux/timer.h>
#include <linux/clk.h>
#include <linux/of.h>
+#include <linux/of_irq.h>
#include <linux/of_gpio.h>
#include <linux/of_device.h>
#include <linux/omap-dmaengine.h>
#include <linux/mmc/core.h>
#include <linux/mmc/mmc.h>
#include <linux/io.h>
+#include <linux/irq.h>
#include <linux/gpio.h>
#include <linux/regulator/consumer.h>
#include <linux/pinctrl/consumer.h>
#define OMAP_HSMMC_RSP54 0x0118
#define OMAP_HSMMC_RSP76 0x011C
#define OMAP_HSMMC_DATA 0x0120
+#define OMAP_HSMMC_PSTATE 0x0124
#define OMAP_HSMMC_HCTL 0x0128
#define OMAP_HSMMC_SYSCTL 0x012C
#define OMAP_HSMMC_STAT 0x0130
#define BCE (1 << 1)
#define FOUR_BIT (1 << 1)
#define HSPE (1 << 2)
+#define IWE (1 << 24)
#define DDR (1 << 19)
+#define CLKEXTFREE (1 << 16)
+#define CTPL (1 << 11)
#define DW8 (1 << 5)
#define OD 0x1
#define STAT_CLEAR 0xFFFFFFFF
#define SRD (1 << 26)
#define SOFTRESET (1 << 1)
+/* PSTATE */
+#define DLEV_DAT(x) (1 << (20 + (x)))
+
/* Interrupt masks for IE and ISE register */
#define CC_EN (1 << 0)
#define TC_EN (1 << 1)
#define BWR_EN (1 << 4)
#define BRR_EN (1 << 5)
+#define CIRQ_EN (1 << 8)
#define ERR_EN (1 << 15)
#define CTO_EN (1 << 16)
#define CCRC_EN (1 << 17)
#define VDD_3V0 3000000 /* 300000 uV */
#define VDD_165_195 (ffs(MMC_VDD_165_195) - 1)
-#define AUTO_CMD23 (1 << 1) /* Auto CMD23 support */
/*
* One controller can have multiple slots, like on some omap boards using
* omap.c controller driver. Luckily this is not currently done on any known
u32 sysctl;
u32 capa;
int irq;
+ int wake_irq;
int use_dma, dma_ch;
struct dma_chan *tx_chan;
struct dma_chan *rx_chan;
int req_in_progress;
unsigned long clk_rate;
unsigned int flags;
+#define AUTO_CMD23 (1 << 0) /* Auto CMD23 support */
+#define HSMMC_SDIO_IRQ_ENABLED (1 << 1) /* SDIO irq enabled */
+#define HSMMC_WAKE_IRQ_ENABLED (1 << 2)
struct omap_hsmmc_next next_data;
struct omap_mmc_platform_data *pdata;
};
static void omap_hsmmc_enable_irq(struct omap_hsmmc_host *host,
struct mmc_command *cmd)
{
- unsigned int irq_mask;
+ u32 irq_mask = INT_EN_MASK;
+ unsigned long flags;
if (host->use_dma)
- irq_mask = INT_EN_MASK & ~(BRR_EN | BWR_EN);
- else
- irq_mask = INT_EN_MASK;
+ irq_mask &= ~(BRR_EN | BWR_EN);
/* Disable timeout for erases */
if (cmd->opcode == MMC_ERASE)
irq_mask &= ~DTO_EN;
+ spin_lock_irqsave(&host->irq_lock, flags);
OMAP_HSMMC_WRITE(host->base, STAT, STAT_CLEAR);
OMAP_HSMMC_WRITE(host->base, ISE, irq_mask);
+
+ /* latch pending CIRQ, but don't signal MMC core */
+ if (host->flags & HSMMC_SDIO_IRQ_ENABLED)
+ irq_mask |= CIRQ_EN;
OMAP_HSMMC_WRITE(host->base, IE, irq_mask);
+ spin_unlock_irqrestore(&host->irq_lock, flags);
}
static void omap_hsmmc_disable_irq(struct omap_hsmmc_host *host)
{
- OMAP_HSMMC_WRITE(host->base, ISE, 0);
- OMAP_HSMMC_WRITE(host->base, IE, 0);
+ u32 irq_mask = 0;
+ unsigned long flags;
+
+ spin_lock_irqsave(&host->irq_lock, flags);
+ /* no transfer running but need to keep cirq if enabled */
+ if (host->flags & HSMMC_SDIO_IRQ_ENABLED)
+ irq_mask |= CIRQ_EN;
+ OMAP_HSMMC_WRITE(host->base, ISE, irq_mask);
+ OMAP_HSMMC_WRITE(host->base, IE, irq_mask);
OMAP_HSMMC_WRITE(host->base, STAT, STAT_CLEAR);
+ spin_unlock_irqrestore(&host->irq_lock, flags);
}
/* Calculate divisor for the given clock frequency */
capa = VS18;
}
+ if (host->mmc->caps & MMC_CAP_SDIO_IRQ)
+ hctl |= IWE;
+
OMAP_HSMMC_WRITE(host->base, HCTL,
OMAP_HSMMC_READ(host->base, HCTL) | hctl);
&& time_before(jiffies, timeout))
;
- omap_hsmmc_disable_irq(host);
+ OMAP_HSMMC_WRITE(host->base, ISE, 0);
+ OMAP_HSMMC_WRITE(host->base, IE, 0);
+ OMAP_HSMMC_WRITE(host->base, STAT, STAT_CLEAR);
/* Do not initialize card-specific things if the power is off */
if (host->power_mode == MMC_POWER_OFF)
int status;
status = OMAP_HSMMC_READ(host->base, STAT);
- while (status & INT_EN_MASK && host->req_in_progress) {
- omap_hsmmc_do_irq(host, status);
+ while (status & (INT_EN_MASK | CIRQ_EN)) {
+ if (host->req_in_progress)
+ omap_hsmmc_do_irq(host, status);
+
+ if (status & CIRQ_EN)
+ mmc_signal_sdio_irq(host->mmc);
/* Flush posted write */
status = OMAP_HSMMC_READ(host->base, STAT);
return IRQ_HANDLED;
}
+static irqreturn_t omap_hsmmc_wake_irq(int irq, void *dev_id)
+{
+ struct omap_hsmmc_host *host = dev_id;
+
+ /* cirq is level triggered, disable to avoid infinite loop */
+ spin_lock(&host->irq_lock);
+ if (host->flags & HSMMC_WAKE_IRQ_ENABLED) {
+ disable_irq_nosync(host->wake_irq);
+ host->flags &= ~HSMMC_WAKE_IRQ_ENABLED;
+ }
+ spin_unlock(&host->irq_lock);
+ pm_request_resume(host->dev); /* no use counter */
+
+ return IRQ_HANDLED;
+}
+
static void set_sd_bus_power(struct omap_hsmmc_host *host)
{
unsigned long i;
mmc_slot(host).init_card(card);
}
+static void omap_hsmmc_enable_sdio_irq(struct mmc_host *mmc, int enable)
+{
+ struct omap_hsmmc_host *host = mmc_priv(mmc);
+ u32 irq_mask, con;
+ unsigned long flags;
+
+ spin_lock_irqsave(&host->irq_lock, flags);
+
+ con = OMAP_HSMMC_READ(host->base, CON);
+ irq_mask = OMAP_HSMMC_READ(host->base, ISE);
+ if (enable) {
+ host->flags |= HSMMC_SDIO_IRQ_ENABLED;
+ irq_mask |= CIRQ_EN;
+ con |= CTPL | CLKEXTFREE;
+ } else {
+ host->flags &= ~HSMMC_SDIO_IRQ_ENABLED;
+ irq_mask &= ~CIRQ_EN;
+ con &= ~(CTPL | CLKEXTFREE);
+ }
+ OMAP_HSMMC_WRITE(host->base, CON, con);
+ OMAP_HSMMC_WRITE(host->base, IE, irq_mask);
+
+ /*
+ * if enable, piggy back detection on current request
+ * but always disable immediately
+ */
+ if (!host->req_in_progress || !enable)
+ OMAP_HSMMC_WRITE(host->base, ISE, irq_mask);
+
+ /* flush posted write */
+ OMAP_HSMMC_READ(host->base, IE);
+
+ spin_unlock_irqrestore(&host->irq_lock, flags);
+}
+
+static int omap_hsmmc_configure_wake_irq(struct omap_hsmmc_host *host)
+{
+ struct mmc_host *mmc = host->mmc;
+ int ret;
+
+ /*
+ * For omaps with wake-up path, wakeirq will be irq from pinctrl and
+ * for other omaps, wakeirq will be from GPIO (dat line remuxed to
+ * gpio). wakeirq is needed to detect sdio irq in runtime suspend state
+ * with functional clock disabled.
+ */
+ if (!host->dev->of_node || !host->wake_irq)
+ return -ENODEV;
+
+ /* Prevent auto-enabling of IRQ */
+ irq_set_status_flags(host->wake_irq, IRQ_NOAUTOEN);
+ ret = devm_request_irq(host->dev, host->wake_irq, omap_hsmmc_wake_irq,
+ IRQF_TRIGGER_LOW | IRQF_ONESHOT,
+ mmc_hostname(mmc), host);
+ if (ret) {
+ dev_err(mmc_dev(host->mmc), "Unable to request wake IRQ\n");
+ goto err;
+ }
+
+ /*
+ * Some omaps don't have wake-up path from deeper idle states
+ * and need to remux SDIO DAT1 to GPIO for wake-up from idle.
+ */
+ if (host->pdata->controller_flags & OMAP_HSMMC_SWAKEUP_MISSING) {
+ struct pinctrl *p = devm_pinctrl_get(host->dev);
+ if (!p) {
+ ret = -ENODEV;
+ goto err_free_irq;
+ }
+ if (IS_ERR(pinctrl_lookup_state(p, PINCTRL_STATE_DEFAULT))) {
+ dev_info(host->dev, "missing default pinctrl state\n");
+ devm_pinctrl_put(p);
+ ret = -EINVAL;
+ goto err_free_irq;
+ }
+
+ if (IS_ERR(pinctrl_lookup_state(p, PINCTRL_STATE_IDLE))) {
+ dev_info(host->dev, "missing idle pinctrl state\n");
+ devm_pinctrl_put(p);
+ ret = -EINVAL;
+ goto err_free_irq;
+ }
+ devm_pinctrl_put(p);
+ }
+
+ OMAP_HSMMC_WRITE(host->base, HCTL,
+ OMAP_HSMMC_READ(host->base, HCTL) | IWE);
+ return 0;
+
+err_free_irq:
+ devm_free_irq(host->dev, host->wake_irq, host);
+err:
+ dev_warn(host->dev, "no SDIO IRQ support, falling back to polling\n");
+ host->wake_irq = 0;
+ return ret;
+}
+
static void omap_hsmmc_conf_bus_power(struct omap_hsmmc_host *host)
{
u32 hctl, capa, value;
.get_cd = omap_hsmmc_get_cd,
.get_ro = omap_hsmmc_get_ro,
.init_card = omap_hsmmc_init_card,
- /* NYET -- enable_sdio_irq */
+ .enable_sdio_irq = omap_hsmmc_enable_sdio_irq,
};
#ifdef CONFIG_DEBUG_FS
struct mmc_host *mmc = s->private;
struct omap_hsmmc_host *host = mmc_priv(mmc);
- seq_printf(s, "mmc%d:\n ctx_loss:\t%d\n\nregs:\n",
- mmc->index, host->context_loss);
+ seq_printf(s, "mmc%d:\n", mmc->index);
+ seq_printf(s, "sdio irq mode\t%s\n",
+ (mmc->caps & MMC_CAP_SDIO_IRQ) ? "interrupt" : "polling");
- pm_runtime_get_sync(host->dev);
+ if (mmc->caps & MMC_CAP_SDIO_IRQ) {
+ seq_printf(s, "sdio irq \t%s\n",
+ (host->flags & HSMMC_SDIO_IRQ_ENABLED) ? "enabled"
+ : "disabled");
+ }
+ seq_printf(s, "ctx_loss:\t%d\n", host->context_loss);
+ pm_runtime_get_sync(host->dev);
+ seq_puts(s, "\nregs:\n");
seq_printf(s, "CON:\t\t0x%08x\n",
OMAP_HSMMC_READ(host->base, CON));
+ seq_printf(s, "PSTATE:\t\t0x%08x\n",
+ OMAP_HSMMC_READ(host->base, PSTATE));
seq_printf(s, "HCTL:\t\t0x%08x\n",
OMAP_HSMMC_READ(host->base, HCTL));
seq_printf(s, "SYSCTL:\t\t0x%08x\n",
static const struct omap_mmc_of_data omap4_mmc_of_data = {
.reg_offset = 0x100,
};
+static const struct omap_mmc_of_data am33xx_mmc_of_data = {
+ .reg_offset = 0x100,
+ .controller_flags = OMAP_HSMMC_SWAKEUP_MISSING,
+};
static const struct of_device_id omap_mmc_of_match[] = {
{
.compatible = "ti,omap4-hsmmc",
.data = &omap4_mmc_of_data,
},
+ {
+ .compatible = "ti,am33xx-hsmmc",
+ .data = &am33xx_mmc_of_data,
+ },
{},
};
MODULE_DEVICE_TABLE(of, omap_mmc_of_match);
const struct of_device_id *match;
dma_cap_mask_t mask;
unsigned tx_req, rx_req;
- struct pinctrl *pinctrl;
const struct omap_mmc_of_data *data;
void __iomem *base;
platform_set_drvdata(pdev, host);
+ if (pdev->dev.of_node)
+ host->wake_irq = irq_of_parse_and_map(pdev->dev.of_node, 1);
+
mmc->ops = &omap_hsmmc_ops;
mmc->f_min = OMAP_MMC_MIN_CLOCK;
omap_hsmmc_disable_irq(host);
- pinctrl = devm_pinctrl_get_select_default(&pdev->dev);
- if (IS_ERR(pinctrl))
- dev_warn(&pdev->dev,
- "pins are not configured from the driver\n");
+ /*
+ * For now, only support SDIO interrupt if we have a separate
+ * wake-up interrupt configured from device tree. This is because
+ * the wake-up interrupt is needed for idle state and some
+ * platforms need special quirks. And we don't want to add new
+ * legacy mux platform init code callbacks any longer as we
+ * are moving to DT based booting anyways.
+ */
+ ret = omap_hsmmc_configure_wake_irq(host);
+ if (!ret)
+ mmc->caps |= MMC_CAP_SDIO_IRQ;
omap_hsmmc_protect_card(host);
pm_runtime_get_sync(host->dev);
if (!(host->mmc->pm_flags & MMC_PM_KEEP_POWER)) {
- omap_hsmmc_disable_irq(host);
+ OMAP_HSMMC_WRITE(host->base, ISE, 0);
+ OMAP_HSMMC_WRITE(host->base, IE, 0);
+ OMAP_HSMMC_WRITE(host->base, STAT, STAT_CLEAR);
OMAP_HSMMC_WRITE(host->base, HCTL,
OMAP_HSMMC_READ(host->base, HCTL) & ~SDBP);
}
+ /* do not wake up due to sdio irq */
+ if ((host->mmc->caps & MMC_CAP_SDIO_IRQ) &&
+ !(host->mmc->pm_flags & MMC_PM_WAKE_SDIO_IRQ))
+ disable_irq(host->wake_irq);
+
if (host->dbclk)
clk_disable_unprepare(host->dbclk);
omap_hsmmc_protect_card(host);
+ if ((host->mmc->caps & MMC_CAP_SDIO_IRQ) &&
+ !(host->mmc->pm_flags & MMC_PM_WAKE_SDIO_IRQ))
+ enable_irq(host->wake_irq);
+
pm_runtime_mark_last_busy(host->dev);
pm_runtime_put_autosuspend(host->dev);
return 0;
static int omap_hsmmc_runtime_suspend(struct device *dev)
{
struct omap_hsmmc_host *host;
+ unsigned long flags;
+ int ret = 0;
host = platform_get_drvdata(to_platform_device(dev));
omap_hsmmc_context_save(host);
dev_dbg(dev, "disabled\n");
- return 0;
+ spin_lock_irqsave(&host->irq_lock, flags);
+ if ((host->mmc->caps & MMC_CAP_SDIO_IRQ) &&
+ (host->flags & HSMMC_SDIO_IRQ_ENABLED)) {
+ /* disable sdio irq handling to prevent race */
+ OMAP_HSMMC_WRITE(host->base, ISE, 0);
+ OMAP_HSMMC_WRITE(host->base, IE, 0);
+
+ if (!(OMAP_HSMMC_READ(host->base, PSTATE) & DLEV_DAT(1))) {
+ /*
+ * dat1 line low, pending sdio irq
+ * race condition: possible irq handler running on
+ * multi-core, abort
+ */
+ dev_dbg(dev, "pending sdio irq, abort suspend\n");
+ OMAP_HSMMC_WRITE(host->base, STAT, STAT_CLEAR);
+ OMAP_HSMMC_WRITE(host->base, ISE, CIRQ_EN);
+ OMAP_HSMMC_WRITE(host->base, IE, CIRQ_EN);
+ pm_runtime_mark_last_busy(dev);
+ ret = -EBUSY;
+ goto abort;
+ }
+
+ pinctrl_pm_select_idle_state(dev);
+
+ WARN_ON(host->flags & HSMMC_WAKE_IRQ_ENABLED);
+ enable_irq(host->wake_irq);
+ host->flags |= HSMMC_WAKE_IRQ_ENABLED;
+ } else {
+ pinctrl_pm_select_idle_state(dev);
+ }
+
+abort:
+ spin_unlock_irqrestore(&host->irq_lock, flags);
+ return ret;
}
static int omap_hsmmc_runtime_resume(struct device *dev)
{
struct omap_hsmmc_host *host;
+ unsigned long flags;
host = platform_get_drvdata(to_platform_device(dev));
omap_hsmmc_context_restore(host);
dev_dbg(dev, "enabled\n");
+ spin_lock_irqsave(&host->irq_lock, flags);
+ if ((host->mmc->caps & MMC_CAP_SDIO_IRQ) &&
+ (host->flags & HSMMC_SDIO_IRQ_ENABLED)) {
+ /* sdio irq flag can't change while in runtime suspend */
+ if (host->flags & HSMMC_WAKE_IRQ_ENABLED) {
+ disable_irq_nosync(host->wake_irq);
+ host->flags &= ~HSMMC_WAKE_IRQ_ENABLED;
+ }
+
+ pinctrl_pm_select_default_state(host->dev);
+
+ /* irq lost, if pinmux incorrect */
+ OMAP_HSMMC_WRITE(host->base, STAT, STAT_CLEAR);
+ OMAP_HSMMC_WRITE(host->base, ISE, CIRQ_EN);
+ OMAP_HSMMC_WRITE(host->base, IE, CIRQ_EN);
+ } else {
+ pinctrl_pm_select_default_state(host->dev);
+ }
+ spin_unlock_irqrestore(&host->irq_lock, flags);
return 0;
}
*/
#include <linux/module.h>
+#include <linux/dmaengine.h>
#include <linux/dma-mapping.h>
#include <linux/clk.h>
#include <linux/mmc/host.h>
#include <mach/dma.h>
#include <mach/gpio-samsung.h>
+#include <linux/platform_data/dma-s3c24xx.h>
#include <linux/platform_data/mmc-s3cmci.h>
#include "s3cmci.h"
dev_dbg(&host->pdev->dev, args); \
} while (0)
-static struct s3c2410_dma_client s3cmci_dma_client = {
- .name = "s3c-mci",
-};
-
static void finalize_request(struct s3cmci_host *host);
static void s3cmci_send_request(struct mmc_host *mmc);
static void s3cmci_reset(struct s3cmci_host *host);
{
#ifdef CONFIG_MMC_S3C_PIO
return false;
-#elif defined(CONFIG_MMC_S3C_DMA)
+#else /* CONFIG_MMC_S3C_DMA */
return true;
-#else
- return host->dodma;
-#endif
-}
-
-/**
- * s3cmci_host_canpio - return true if host has pio code available
- *
- * Return true if the driver has been compiled with the PIO support code
- * available.
- */
-static inline bool s3cmci_host_canpio(void)
-{
-#ifdef CONFIG_MMC_S3C_PIO
- return true;
-#else
- return false;
#endif
}
return IRQ_HANDLED;
}
-static void s3cmci_dma_done_callback(struct s3c2410_dma_chan *dma_ch,
- void *buf_id, int size,
- enum s3c2410_dma_buffresult result)
+static void s3cmci_dma_done_callback(void *arg)
{
- struct s3cmci_host *host = buf_id;
+ struct s3cmci_host *host = arg;
unsigned long iflags;
- u32 mci_csta, mci_dsta, mci_fsta, mci_dcnt;
-
- mci_csta = readl(host->base + S3C2410_SDICMDSTAT);
- mci_dsta = readl(host->base + S3C2410_SDIDSTA);
- mci_fsta = readl(host->base + S3C2410_SDIFSTA);
- mci_dcnt = readl(host->base + S3C2410_SDIDCNT);
BUG_ON(!host->mrq);
BUG_ON(!host->mrq->data);
- BUG_ON(!host->dmatogo);
spin_lock_irqsave(&host->complete_lock, iflags);
- if (result != S3C2410_RES_OK) {
- dbg(host, dbg_fail, "DMA FAILED: csta=0x%08x dsta=0x%08x "
- "fsta=0x%08x dcnt:0x%08x result:0x%08x toGo:%u\n",
- mci_csta, mci_dsta, mci_fsta,
- mci_dcnt, result, host->dmatogo);
-
- goto fail_request;
- }
-
- host->dmatogo--;
- if (host->dmatogo) {
- dbg(host, dbg_dma, "DMA DONE Size:%i DSTA:[%08x] "
- "DCNT:[%08x] toGo:%u\n",
- size, mci_dsta, mci_dcnt, host->dmatogo);
-
- goto out;
- }
-
- dbg(host, dbg_dma, "DMA FINISHED Size:%i DSTA:%08x DCNT:%08x\n",
- size, mci_dsta, mci_dcnt);
+ dbg(host, dbg_dma, "DMA FINISHED\n");
host->dma_complete = 1;
host->complete_what = COMPLETION_FINALIZE;
-out:
tasklet_schedule(&host->pio_tasklet);
spin_unlock_irqrestore(&host->complete_lock, iflags);
- return;
-fail_request:
- host->mrq->data->error = -EINVAL;
- host->complete_what = COMPLETION_FINALIZE;
- clear_imask(host);
-
- goto out;
}
static void finalize_request(struct s3cmci_host *host)
* DMA channel and the fifo to clear out any garbage. */
if (mrq->data->error != 0) {
if (s3cmci_host_usedma(host))
- s3c2410_dma_ctrl(host->dma, S3C2410_DMAOP_FLUSH);
+ dmaengine_terminate_all(host->dma);
if (host->is2440) {
/* Clear failure register and reset fifo. */
mmc_request_done(host->mmc, mrq);
}
-static void s3cmci_dma_setup(struct s3cmci_host *host,
- enum dma_data_direction source)
-{
- static enum dma_data_direction last_source = -1;
- static int setup_ok;
-
- if (last_source == source)
- return;
-
- last_source = source;
-
- s3c2410_dma_devconfig(host->dma, source,
- host->mem->start + host->sdidata);
-
- if (!setup_ok) {
- s3c2410_dma_config(host->dma, 4);
- s3c2410_dma_set_buffdone_fn(host->dma,
- s3cmci_dma_done_callback);
- s3c2410_dma_setflags(host->dma, S3C2410_DMAF_AUTOSTART);
- setup_ok = 1;
- }
-}
-
static void s3cmci_send_command(struct s3cmci_host *host,
struct mmc_command *cmd)
{
static int s3cmci_prepare_dma(struct s3cmci_host *host, struct mmc_data *data)
{
- int dma_len, i;
int rw = data->flags & MMC_DATA_WRITE;
+ struct dma_async_tx_descriptor *desc;
+ struct dma_slave_config conf = {
+ .src_addr = host->mem->start + host->sdidata,
+ .dst_addr = host->mem->start + host->sdidata,
+ .src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES,
+ .dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES,
+ };
BUG_ON((data->flags & BOTH_DIR) == BOTH_DIR);
- s3cmci_dma_setup(host, rw ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
- s3c2410_dma_ctrl(host->dma, S3C2410_DMAOP_FLUSH);
-
- dma_len = dma_map_sg(mmc_dev(host->mmc), data->sg, data->sg_len,
- rw ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
-
- if (dma_len == 0)
- return -ENOMEM;
-
- host->dma_complete = 0;
- host->dmatogo = dma_len;
-
- for (i = 0; i < dma_len; i++) {
- int res;
-
- dbg(host, dbg_dma, "enqueue %i: %08x@%u\n", i,
- sg_dma_address(&data->sg[i]),
- sg_dma_len(&data->sg[i]));
+ /* Restore prescaler value */
+ writel(host->prescaler, host->base + S3C2410_SDIPRE);
- res = s3c2410_dma_enqueue(host->dma, host,
- sg_dma_address(&data->sg[i]),
- sg_dma_len(&data->sg[i]));
+ if (!rw)
+ conf.direction = DMA_DEV_TO_MEM;
+ else
+ conf.direction = DMA_MEM_TO_DEV;
- if (res) {
- s3c2410_dma_ctrl(host->dma, S3C2410_DMAOP_FLUSH);
- return -EBUSY;
- }
- }
+ dma_map_sg(mmc_dev(host->mmc), data->sg, data->sg_len,
+ rw ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
- s3c2410_dma_ctrl(host->dma, S3C2410_DMAOP_START);
+ dmaengine_slave_config(host->dma, &conf);
+ desc = dmaengine_prep_slave_sg(host->dma, data->sg, data->sg_len,
+ conf.direction,
+ DMA_CTRL_ACK | DMA_PREP_INTERRUPT);
+ if (!desc)
+ goto unmap_exit;
+ desc->callback = s3cmci_dma_done_callback;
+ desc->callback_param = host;
+ dmaengine_submit(desc);
+ dma_async_issue_pending(host->dma);
return 0;
+
+unmap_exit:
+ dma_unmap_sg(mmc_dev(host->mmc), data->sg, data->sg_len,
+ rw ? DMA_TO_DEVICE : DMA_FROM_DEVICE);
+ return -ENOMEM;
}
static void s3cmci_send_request(struct mmc_host *mmc)
host->complete_what = COMPLETION_NONE;
host->pio_active = XFER_NONE;
-#ifdef CONFIG_MMC_S3C_PIODMA
- host->dodma = host->pdata->use_dma;
-#endif
-
host->mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
if (!host->mem) {
dev_err(&pdev->dev,
/* depending on the dma state, get a dma channel to use. */
if (s3cmci_host_usedma(host)) {
- host->dma = s3c2410_dma_request(DMACH_SDI, &s3cmci_dma_client,
- host);
- if (host->dma < 0) {
+ dma_cap_mask_t mask;
+
+ dma_cap_zero(mask);
+ dma_cap_set(DMA_SLAVE, mask);
+
+ host->dma = dma_request_slave_channel_compat(mask,
+ s3c24xx_dma_filter, (void *)DMACH_SDI, &pdev->dev, "rx-tx");
+ if (!host->dma) {
dev_err(&pdev->dev, "cannot get DMA channel.\n");
- if (!s3cmci_host_canpio()) {
- ret = -EBUSY;
- goto probe_free_gpio_wp;
- } else {
- dev_warn(&pdev->dev, "falling back to PIO.\n");
- host->dodma = 0;
- }
+ ret = -EBUSY;
+ goto probe_free_gpio_wp;
}
}
goto probe_free_dma;
}
- ret = clk_enable(host->clk);
+ ret = clk_prepare_enable(host->clk);
if (ret) {
dev_err(&pdev->dev, "failed to enable clock source.\n");
goto clk_free;
mmc->max_segs = 128;
dbg(host, dbg_debug,
- "probe: mode:%s mapped mci_base:%p irq:%u irq_cd:%u dma:%u.\n",
+ "probe: mode:%s mapped mci_base:%p irq:%u irq_cd:%u dma:%p.\n",
(host->is2440?"2440":""),
host->base, host->irq, host->irq_cd, host->dma);
s3cmci_cpufreq_deregister(host);
free_dmabuf:
- clk_disable(host->clk);
+ clk_disable_unprepare(host->clk);
clk_free:
clk_put(host->clk);
probe_free_dma:
if (s3cmci_host_usedma(host))
- s3c2410_dma_free(host->dma, &s3cmci_dma_client);
+ dma_release_channel(host->dma);
probe_free_gpio_wp:
if (!host->pdata->no_wprotect)
s3cmci_debugfs_remove(host);
s3cmci_cpufreq_deregister(host);
mmc_remove_host(mmc);
- clk_disable(host->clk);
+ clk_disable_unprepare(host->clk);
}
static int s3cmci_remove(struct platform_device *pdev)
tasklet_disable(&host->pio_tasklet);
if (s3cmci_host_usedma(host))
- s3c2410_dma_free(host->dma, &s3cmci_dma_client);
+ dma_release_channel(host->dma);
free_irq(host->irq, host);
void __iomem *base;
int irq;
int irq_cd;
- int dma;
+ struct dma_chan *dma;
unsigned long clk_rate;
unsigned long clk_div;
int is2440;
unsigned sdiimsk;
unsigned sdidata;
- int dodma;
- int dmatogo;
bool irq_disabled;
bool irq_enabled;
static const struct sdhci_acpi_slot sdhci_acpi_slot_int_emmc = {
.chip = &sdhci_acpi_chip_int,
- .caps = MMC_CAP_8_BIT_DATA | MMC_CAP_NONREMOVABLE | MMC_CAP_HW_RESET,
+ .caps = MMC_CAP_8_BIT_DATA | MMC_CAP_NONREMOVABLE |
+ MMC_CAP_HW_RESET | MMC_CAP_1_8V_DDR,
.caps2 = MMC_CAP2_HC_ERASE_SZ,
.flags = SDHCI_ACPI_RUNTIME_PM,
+ .quirks2 = SDHCI_QUIRK2_PRESET_VALUE_BROKEN,
};
static const struct sdhci_acpi_slot sdhci_acpi_slot_int_sdio = {
#include <linux/module.h>
#include <linux/of_device.h>
-#include <linux/regulator/consumer.h>
#include <linux/delay.h>
#include <linux/mmc/mmc.h>
#include <linux/slab.h>
SDHCI_QUIRK_BROKEN_TIMEOUT_VAL,
};
+static const struct sdhci_pci_fixes sdhci_intel_qrk = {
+ .quirks = SDHCI_QUIRK_NO_HISPD_BIT,
+};
+
static int mrst_hc_probe_slot(struct sdhci_pci_slot *slot)
{
slot->host->mmc->caps |= MMC_CAP_8_BIT_DATA;
static int byt_emmc_probe_slot(struct sdhci_pci_slot *slot)
{
slot->host->mmc->caps |= MMC_CAP_8_BIT_DATA | MMC_CAP_NONREMOVABLE |
- MMC_CAP_HW_RESET;
+ MMC_CAP_HW_RESET | MMC_CAP_1_8V_DDR;
slot->host->mmc->caps2 |= MMC_CAP2_HC_ERASE_SZ;
slot->hw_reset = sdhci_pci_int_hw_reset;
return 0;
static const struct sdhci_pci_fixes sdhci_intel_byt_emmc = {
.allow_runtime_pm = true,
.probe_slot = byt_emmc_probe_slot,
+ .quirks2 = SDHCI_QUIRK2_PRESET_VALUE_BROKEN,
};
static const struct sdhci_pci_fixes sdhci_intel_byt_sdio = {
.driver_data = (kernel_ulong_t)&sdhci_rtsx,
},
+ {
+ .vendor = PCI_VENDOR_ID_INTEL,
+ .device = PCI_DEVICE_ID_INTEL_QRK_SD,
+ .subvendor = PCI_ANY_ID,
+ .subdevice = PCI_ANY_ID,
+ .driver_data = (kernel_ulong_t)&sdhci_intel_qrk,
+ },
+
{
.vendor = PCI_VENDOR_ID_INTEL,
.device = PCI_DEVICE_ID_INTEL_MRST_SD0,
goto err_pci_suspend;
}
- pci_save_state(pdev);
if (pm_flags & MMC_PM_KEEP_POWER) {
- if (pm_flags & MMC_PM_WAKE_SDIO_IRQ) {
- pci_pme_active(pdev, true);
- pci_enable_wake(pdev, PCI_D3hot, 1);
- }
- pci_set_power_state(pdev, PCI_D3hot);
- } else {
- pci_enable_wake(pdev, PCI_D3hot, 0);
- pci_disable_device(pdev);
- pci_set_power_state(pdev, PCI_D3hot);
- }
+ if (pm_flags & MMC_PM_WAKE_SDIO_IRQ)
+ device_init_wakeup(dev, true);
+ else
+ device_init_wakeup(dev, false);
+ } else
+ device_init_wakeup(dev, false);
return 0;
if (!chip)
return 0;
- pci_set_power_state(pdev, PCI_D0);
- pci_restore_state(pdev);
- ret = pci_enable_device(pdev);
- if (ret)
- return ret;
-
if (chip->fixes && chip->fixes->resume) {
ret = chip->fixes->resume(chip);
if (ret)
#define PCI_DEVICE_ID_INTEL_CLV_SDIO2 0x08fb
#define PCI_DEVICE_ID_INTEL_CLV_EMMC0 0x08e5
#define PCI_DEVICE_ID_INTEL_CLV_EMMC1 0x08e6
+#define PCI_DEVICE_ID_INTEL_QRK_SD 0x08A7
/*
* PCI registers
int ret;
struct clk *clk;
- pxa = kzalloc(sizeof(struct sdhci_pxa), GFP_KERNEL);
+ pxa = devm_kzalloc(&pdev->dev, sizeof(struct sdhci_pxa), GFP_KERNEL);
if (!pxa)
return -ENOMEM;
host = sdhci_pltfm_init(pdev, &sdhci_pxav3_pdata, 0);
- if (IS_ERR(host)) {
- kfree(pxa);
+ if (IS_ERR(host))
return PTR_ERR(host);
- }
if (of_device_is_compatible(np, "marvell,armada-380-sdhci")) {
ret = mv_conf_mbus_windows(pdev, mv_mbus_dram_info());
pltfm_host = sdhci_priv(host);
pltfm_host->priv = pxa;
- clk = clk_get(dev, NULL);
+ clk = devm_clk_get(dev, NULL);
if (IS_ERR(clk)) {
dev_err(dev, "failed to get io clock\n");
ret = PTR_ERR(clk);
pm_runtime_put_sync(&pdev->dev);
pm_runtime_disable(&pdev->dev);
clk_disable_unprepare(clk);
- clk_put(clk);
err_clk_get:
err_mbus_win:
sdhci_pltfm_free(pdev);
- kfree(pxa);
return ret;
}
{
struct sdhci_host *host = platform_get_drvdata(pdev);
struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
- struct sdhci_pxa *pxa = pltfm_host->priv;
pm_runtime_get_sync(&pdev->dev);
sdhci_remove_host(host, 1);
pm_runtime_disable(&pdev->dev);
clk_disable_unprepare(pltfm_host->clk);
- clk_put(pltfm_host->clk);
sdhci_pltfm_free(pdev);
- kfree(pxa);
return 0;
}
--- /dev/null
+/*
+ * Support for SDHCI on STMicroelectronics SoCs
+ *
+ * Copyright (C) 2014 STMicroelectronics Ltd
+ * Author: Giuseppe Cavallaro <peppe.cavallaro@st.com>
+ * Contributors: Peter Griffin <peter.griffin@linaro.org>
+ *
+ * Based on sdhci-cns3xxx.c
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/io.h>
+#include <linux/of.h>
+#include <linux/module.h>
+#include <linux/err.h>
+#include <linux/mmc/host.h>
+
+#include "sdhci-pltfm.h"
+
+static u32 sdhci_st_readl(struct sdhci_host *host, int reg)
+{
+ u32 ret;
+
+ switch (reg) {
+ case SDHCI_CAPABILITIES:
+ ret = readl_relaxed(host->ioaddr + reg);
+ /* Support 3.3V and 1.8V */
+ ret &= ~SDHCI_CAN_VDD_300;
+ break;
+ default:
+ ret = readl_relaxed(host->ioaddr + reg);
+ }
+ return ret;
+}
+
+static const struct sdhci_ops sdhci_st_ops = {
+ .get_max_clock = sdhci_pltfm_clk_get_max_clock,
+ .set_clock = sdhci_set_clock,
+ .set_bus_width = sdhci_set_bus_width,
+ .read_l = sdhci_st_readl,
+ .reset = sdhci_reset,
+};
+
+static const struct sdhci_pltfm_data sdhci_st_pdata = {
+ .ops = &sdhci_st_ops,
+ .quirks = SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC |
+ SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN,
+};
+
+
+static int sdhci_st_probe(struct platform_device *pdev)
+{
+ struct sdhci_host *host;
+ struct sdhci_pltfm_host *pltfm_host;
+ struct clk *clk;
+ int ret = 0;
+ u16 host_version;
+
+ clk = devm_clk_get(&pdev->dev, "mmc");
+ if (IS_ERR(clk)) {
+ dev_err(&pdev->dev, "Peripheral clk not found\n");
+ return PTR_ERR(clk);
+ }
+
+ host = sdhci_pltfm_init(pdev, &sdhci_st_pdata, 0);
+ if (IS_ERR(host)) {
+ dev_err(&pdev->dev, "Failed sdhci_pltfm_init\n");
+ return PTR_ERR(host);
+ }
+
+ ret = mmc_of_parse(host->mmc);
+
+ if (ret) {
+ dev_err(&pdev->dev, "Failed mmc_of_parse\n");
+ return ret;
+ }
+
+ clk_prepare_enable(clk);
+
+ pltfm_host = sdhci_priv(host);
+ pltfm_host->clk = clk;
+
+ ret = sdhci_add_host(host);
+ if (ret) {
+ dev_err(&pdev->dev, "Failed sdhci_add_host\n");
+ goto err_out;
+ }
+
+ platform_set_drvdata(pdev, host);
+
+ host_version = readw_relaxed((host->ioaddr + SDHCI_HOST_VERSION));
+
+ dev_info(&pdev->dev, "SDHCI ST Initialised: Host Version: 0x%x Vendor Version 0x%x\n",
+ ((host_version & SDHCI_SPEC_VER_MASK) >> SDHCI_SPEC_VER_SHIFT),
+ ((host_version & SDHCI_VENDOR_VER_MASK) >>
+ SDHCI_VENDOR_VER_SHIFT));
+
+ return 0;
+
+err_out:
+ clk_disable_unprepare(clk);
+ sdhci_pltfm_free(pdev);
+
+ return ret;
+}
+
+static int sdhci_st_remove(struct platform_device *pdev)
+{
+ struct sdhci_host *host = platform_get_drvdata(pdev);
+ struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+
+ clk_disable_unprepare(pltfm_host->clk);
+
+ return sdhci_pltfm_unregister(pdev);
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int sdhci_st_suspend(struct device *dev)
+{
+ struct sdhci_host *host = dev_get_drvdata(dev);
+ struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+ int ret = sdhci_suspend_host(host);
+
+ if (ret)
+ goto out;
+
+ clk_disable_unprepare(pltfm_host->clk);
+out:
+ return ret;
+}
+
+static int sdhci_st_resume(struct device *dev)
+{
+ struct sdhci_host *host = dev_get_drvdata(dev);
+ struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
+
+ clk_prepare_enable(pltfm_host->clk);
+
+ return sdhci_resume_host(host);
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(sdhci_st_pmops, sdhci_st_suspend, sdhci_st_resume);
+
+static const struct of_device_id st_sdhci_match[] = {
+ { .compatible = "st,sdhci" },
+ {},
+};
+
+MODULE_DEVICE_TABLE(of, st_sdhci_match);
+
+static struct platform_driver sdhci_st_driver = {
+ .probe = sdhci_st_probe,
+ .remove = sdhci_st_remove,
+ .driver = {
+ .name = "sdhci-st",
+ .pm = &sdhci_st_pmops,
+ .of_match_table = of_match_ptr(st_sdhci_match),
+ },
+};
+
+module_platform_driver(sdhci_st_driver);
+
+MODULE_DESCRIPTION("SDHCI driver for STMicroelectronics SoCs");
+MODULE_AUTHOR("Giuseppe Cavallaro <peppe.cavallaro@st.com>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:st-sdhci");
#include <linux/mmc/host.h>
#include <linux/mmc/slot-gpio.h>
-#include <asm/gpio.h>
-
#include "sdhci-pltfm.h"
/* Tegra SDHOST controller vendor register definitions */
static void sdhci_set_power(struct sdhci_host *host, unsigned char mode,
unsigned short vdd)
{
+ struct mmc_host *mmc = host->mmc;
u8 pwr = 0;
+ if (!IS_ERR(mmc->supply.vmmc)) {
+ spin_unlock_irq(&host->lock);
+ mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, vdd);
+ spin_lock_irq(&host->lock);
+ return;
+ }
+
if (mode != MMC_POWER_OFF) {
switch (1 << vdd) {
case MMC_VDD_165_195:
if (host->quirks & SDHCI_QUIRK_DELAY_AFTER_POWER)
mdelay(10);
}
-
- if (host->vmmc) {
- spin_unlock_irq(&host->lock);
- mmc_regulator_set_ocr(host->mmc, host->vmmc, vdd);
- spin_lock_irq(&host->lock);
- }
}
/*****************************************************************************\
{
unsigned long flags;
u8 ctrl;
+ struct mmc_host *mmc = host->mmc;
spin_lock_irqsave(&host->lock, flags);
if (host->flags & SDHCI_DEVICE_DEAD) {
spin_unlock_irqrestore(&host->lock, flags);
- if (host->vmmc && ios->power_mode == MMC_POWER_OFF)
- mmc_regulator_set_ocr(host->mmc, host->vmmc, 0);
+ if (!IS_ERR(mmc->supply.vmmc) &&
+ ios->power_mode == MMC_POWER_OFF)
+ mmc_regulator_set_ocr(mmc, mmc->supply.vmmc, 0);
return;
}
host->ops->set_clock(host, host->clock);
}
-
/* Reset SD Clock Enable */
clk = sdhci_readw(host, SDHCI_CLOCK_CONTROL);
clk &= ~SDHCI_CLOCK_CARD_EN;
static int sdhci_do_start_signal_voltage_switch(struct sdhci_host *host,
struct mmc_ios *ios)
{
+ struct mmc_host *mmc = host->mmc;
u16 ctrl;
int ret;
ctrl &= ~SDHCI_CTRL_VDD_180;
sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
- if (host->vqmmc) {
- ret = regulator_set_voltage(host->vqmmc, 2700000, 3600000);
+ if (!IS_ERR(mmc->supply.vqmmc)) {
+ ret = regulator_set_voltage(mmc->supply.vqmmc, 2700000,
+ 3600000);
if (ret) {
pr_warning("%s: Switching to 3.3V signalling voltage "
- " failed\n", mmc_hostname(host->mmc));
+ " failed\n", mmc_hostname(mmc));
return -EIO;
}
}
return 0;
pr_warning("%s: 3.3V regulator output did not became stable\n",
- mmc_hostname(host->mmc));
+ mmc_hostname(mmc));
return -EAGAIN;
case MMC_SIGNAL_VOLTAGE_180:
- if (host->vqmmc) {
- ret = regulator_set_voltage(host->vqmmc,
+ if (!IS_ERR(mmc->supply.vqmmc)) {
+ ret = regulator_set_voltage(mmc->supply.vqmmc,
1700000, 1950000);
if (ret) {
pr_warning("%s: Switching to 1.8V signalling voltage "
- " failed\n", mmc_hostname(host->mmc));
+ " failed\n", mmc_hostname(mmc));
return -EIO;
}
}
ctrl |= SDHCI_CTRL_VDD_180;
sdhci_writew(host, ctrl, SDHCI_HOST_CONTROL2);
- /* Wait for 5ms */
- usleep_range(5000, 5500);
-
/* 1.8V regulator output should be stable within 5 ms */
ctrl = sdhci_readw(host, SDHCI_HOST_CONTROL2);
if (ctrl & SDHCI_CTRL_VDD_180)
return 0;
pr_warning("%s: 1.8V regulator output did not became stable\n",
- mmc_hostname(host->mmc));
+ mmc_hostname(mmc));
return -EAGAIN;
case MMC_SIGNAL_VOLTAGE_120:
- if (host->vqmmc) {
- ret = regulator_set_voltage(host->vqmmc, 1100000, 1300000);
+ if (!IS_ERR(mmc->supply.vqmmc)) {
+ ret = regulator_set_voltage(mmc->supply.vqmmc, 1100000,
+ 1300000);
if (ret) {
pr_warning("%s: Switching to 1.2V signalling voltage "
- " failed\n", mmc_hostname(host->mmc));
+ " failed\n", mmc_hostname(mmc));
return -EIO;
}
}
int sdhci_runtime_suspend_host(struct sdhci_host *host)
{
unsigned long flags;
- int ret = 0;
/* Disable tuning since we are suspending */
if (host->flags & SDHCI_USING_RETUNING_TIMER) {
host->runtime_suspended = true;
spin_unlock_irqrestore(&host->lock, flags);
- return ret;
+ return 0;
}
EXPORT_SYMBOL_GPL(sdhci_runtime_suspend_host);
int sdhci_runtime_resume_host(struct sdhci_host *host)
{
unsigned long flags;
- int ret = 0, host_flags = host->flags;
+ int host_flags = host->flags;
if (host_flags & (SDHCI_USE_SDMA | SDHCI_USE_ADMA)) {
if (host->ops->enable_dma)
spin_unlock_irqrestore(&host->lock, flags);
- return ret;
+ return 0;
}
EXPORT_SYMBOL_GPL(sdhci_runtime_resume_host);
* (128) and potentially one alignment transfer for
* each of those entries.
*/
- host->adma_desc = dma_alloc_coherent(mmc_dev(host->mmc),
+ host->adma_desc = dma_alloc_coherent(mmc_dev(mmc),
ADMA_SIZE, &host->adma_addr,
GFP_KERNEL);
host->align_buffer = kmalloc(128 * 4, GFP_KERNEL);
if (!host->adma_desc || !host->align_buffer) {
- dma_free_coherent(mmc_dev(host->mmc), ADMA_SIZE,
+ dma_free_coherent(mmc_dev(mmc), ADMA_SIZE,
host->adma_desc, host->adma_addr);
kfree(host->align_buffer);
pr_warning("%s: Unable to allocate ADMA "
pr_warning("%s: unable to allocate aligned ADMA descriptor\n",
mmc_hostname(mmc));
host->flags &= ~SDHCI_USE_ADMA;
- dma_free_coherent(mmc_dev(host->mmc), ADMA_SIZE,
+ dma_free_coherent(mmc_dev(mmc), ADMA_SIZE,
host->adma_desc, host->adma_addr);
kfree(host->align_buffer);
host->adma_desc = NULL;
*/
if (!(host->flags & (SDHCI_USE_SDMA | SDHCI_USE_ADMA))) {
host->dma_mask = DMA_BIT_MASK(64);
- mmc_dev(host->mmc)->dma_mask = &host->dma_mask;
+ mmc_dev(mmc)->dma_mask = &host->dma_mask;
}
if (host->version >= SDHCI_SPEC_300)
mmc->caps |= MMC_CAP_SD_HIGHSPEED | MMC_CAP_MMC_HIGHSPEED;
if ((host->quirks & SDHCI_QUIRK_BROKEN_CARD_DETECTION) &&
- !(host->mmc->caps & MMC_CAP_NONREMOVABLE))
+ !(mmc->caps & MMC_CAP_NONREMOVABLE))
mmc->caps |= MMC_CAP_NEEDS_POLL;
+ /* If there are external regulators, get them */
+ if (mmc_regulator_get_supply(mmc) == -EPROBE_DEFER)
+ return -EPROBE_DEFER;
+
/* If vqmmc regulator and no 1.8V signalling, then there's no UHS */
- host->vqmmc = regulator_get_optional(mmc_dev(mmc), "vqmmc");
- if (IS_ERR_OR_NULL(host->vqmmc)) {
- if (PTR_ERR(host->vqmmc) < 0) {
- pr_info("%s: no vqmmc regulator found\n",
- mmc_hostname(mmc));
- host->vqmmc = NULL;
- }
- } else {
- ret = regulator_enable(host->vqmmc);
- if (!regulator_is_supported_voltage(host->vqmmc, 1700000,
- 1950000))
+ if (!IS_ERR(mmc->supply.vqmmc)) {
+ ret = regulator_enable(mmc->supply.vqmmc);
+ if (!regulator_is_supported_voltage(mmc->supply.vqmmc, 1700000,
+ 1950000))
caps[1] &= ~(SDHCI_SUPPORT_SDR104 |
SDHCI_SUPPORT_SDR50 |
SDHCI_SUPPORT_DDR50);
if (ret) {
pr_warn("%s: Failed to enable vqmmc regulator: %d\n",
mmc_hostname(mmc), ret);
- host->vqmmc = NULL;
+ mmc->supply.vqmmc = NULL;
}
}
ocr_avail = 0;
- host->vmmc = regulator_get_optional(mmc_dev(mmc), "vmmc");
- if (IS_ERR_OR_NULL(host->vmmc)) {
- if (PTR_ERR(host->vmmc) < 0) {
- pr_info("%s: no vmmc regulator found\n",
- mmc_hostname(mmc));
- host->vmmc = NULL;
- }
- }
-
-#ifdef CONFIG_REGULATOR
- /*
- * Voltage range check makes sense only if regulator reports
- * any voltage value.
- */
- if (host->vmmc && regulator_get_voltage(host->vmmc) > 0) {
- ret = regulator_is_supported_voltage(host->vmmc, 2700000,
- 3600000);
- if ((ret <= 0) || (!(caps[0] & SDHCI_CAN_VDD_330)))
- caps[0] &= ~SDHCI_CAN_VDD_330;
- if ((ret <= 0) || (!(caps[0] & SDHCI_CAN_VDD_300)))
- caps[0] &= ~SDHCI_CAN_VDD_300;
- ret = regulator_is_supported_voltage(host->vmmc, 1700000,
- 1950000);
- if ((ret <= 0) || (!(caps[0] & SDHCI_CAN_VDD_180)))
- caps[0] &= ~SDHCI_CAN_VDD_180;
- }
-#endif /* CONFIG_REGULATOR */
-
/*
* According to SD Host Controller spec v3.00, if the Host System
* can afford more than 150mA, Host Driver should set XPC to 1. Also
* value.
*/
max_current_caps = sdhci_readl(host, SDHCI_MAX_CURRENT);
- if (!max_current_caps && host->vmmc) {
- u32 curr = regulator_get_current_limit(host->vmmc);
+ if (!max_current_caps && !IS_ERR(mmc->supply.vmmc)) {
+ u32 curr = regulator_get_current_limit(mmc->supply.vmmc);
if (curr > 0) {
/* convert to SDHCI_MAX_CURRENT format */
SDHCI_MAX_CURRENT_MULTIPLIER;
}
+ /* If OCR set by external regulators, use it instead */
+ if (mmc->ocr_avail)
+ ocr_avail = mmc->ocr_avail;
+
if (host->ocr_mask)
- ocr_avail = host->ocr_mask;
+ ocr_avail &= host->ocr_mask;
mmc->ocr_avail = ocr_avail;
mmc->ocr_avail_sdio = ocr_avail;
void sdhci_remove_host(struct sdhci_host *host, int dead)
{
+ struct mmc_host *mmc = host->mmc;
unsigned long flags;
if (dead) {
if (host->mrq) {
pr_err("%s: Controller removed during "
- " transfer!\n", mmc_hostname(host->mmc));
+ " transfer!\n", mmc_hostname(mmc));
host->mrq->cmd->error = -ENOMEDIUM;
tasklet_schedule(&host->finish_tasklet);
sdhci_disable_card_detection(host);
- mmc_remove_host(host->mmc);
+ mmc_remove_host(mmc);
#ifdef SDHCI_USE_LEDS_CLASS
led_classdev_unregister(&host->led);
tasklet_kill(&host->finish_tasklet);
- if (host->vmmc) {
- regulator_disable(host->vmmc);
- regulator_put(host->vmmc);
- }
+ if (!IS_ERR(mmc->supply.vmmc))
+ regulator_disable(mmc->supply.vmmc);
- if (host->vqmmc) {
- regulator_disable(host->vqmmc);
- regulator_put(host->vqmmc);
- }
+ if (!IS_ERR(mmc->supply.vqmmc))
+ regulator_disable(mmc->supply.vqmmc);
if (host->adma_desc)
- dma_free_coherent(mmc_dev(host->mmc), ADMA_SIZE,
+ dma_free_coherent(mmc_dev(mmc), ADMA_SIZE,
host->adma_desc, host->adma_addr);
kfree(host->align_buffer);
struct sh_mmcif_plat_data *pdata,
enum dma_transfer_direction direction)
{
- struct dma_slave_config cfg;
+ struct dma_slave_config cfg = { 0, };
struct dma_chan *chan;
unsigned int slave_id;
struct resource *res;
/* In the OF case the driver will get the slave ID from the DT */
cfg.slave_id = slave_id;
cfg.direction = direction;
- cfg.dst_addr = res->start + MMCIF_CE_DATA;
- cfg.src_addr = 0;
+
+ if (direction == DMA_DEV_TO_MEM) {
+ cfg.src_addr = res->start + MMCIF_CE_DATA;
+ cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+ } else {
+ cfg.dst_addr = res->start + MMCIF_CE_DATA;
+ cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+ }
+
ret = dmaengine_slave_config(chan, &cfg);
if (ret < 0) {
dma_release_channel(chan);
dev_err(&pdev->dev, "Get irq error\n");
return -ENXIO;
}
+
res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
- if (!res) {
- dev_err(&pdev->dev, "platform_get_resource error.\n");
- return -ENXIO;
- }
- reg = ioremap(res->start, resource_size(res));
- if (!reg) {
- dev_err(&pdev->dev, "ioremap error.\n");
- return -ENOMEM;
- }
+ reg = devm_ioremap_resource(&pdev->dev, res);
+ if (IS_ERR(reg))
+ return PTR_ERR(reg);
mmc = mmc_alloc_host(sizeof(struct sh_mmcif_host), &pdev->dev);
- if (!mmc) {
- ret = -ENOMEM;
- goto ealloch;
- }
+ if (!mmc)
+ return -ENOMEM;
ret = mmc_of_parse(mmc);
if (ret < 0)
- goto eofparse;
+ goto err_host;
host = mmc_priv(mmc);
host->mmc = mmc;
pm_runtime_enable(&pdev->dev);
host->power = false;
- host->hclk = clk_get(&pdev->dev, NULL);
+ host->hclk = devm_clk_get(&pdev->dev, NULL);
if (IS_ERR(host->hclk)) {
ret = PTR_ERR(host->hclk);
dev_err(&pdev->dev, "cannot get clock: %d\n", ret);
- goto eclkget;
+ goto err_pm;
}
ret = sh_mmcif_clk_update(host);
if (ret < 0)
- goto eclkupdate;
+ goto err_pm;
ret = pm_runtime_resume(&pdev->dev);
if (ret < 0)
- goto eresume;
+ goto err_clk;
INIT_DELAYED_WORK(&host->timeout_work, mmcif_timeout_work);
sh_mmcif_writel(host->addr, MMCIF_CE_INT_MASK, MASK_ALL);
name = irq[1] < 0 ? dev_name(&pdev->dev) : "sh_mmc:error";
- ret = request_threaded_irq(irq[0], sh_mmcif_intr, sh_mmcif_irqt, 0, name, host);
+ ret = devm_request_threaded_irq(&pdev->dev, irq[0], sh_mmcif_intr,
+ sh_mmcif_irqt, 0, name, host);
if (ret) {
dev_err(&pdev->dev, "request_irq error (%s)\n", name);
- goto ereqirq0;
+ goto err_clk;
}
if (irq[1] >= 0) {
- ret = request_threaded_irq(irq[1], sh_mmcif_intr, sh_mmcif_irqt,
- 0, "sh_mmc:int", host);
+ ret = devm_request_threaded_irq(&pdev->dev, irq[1],
+ sh_mmcif_intr, sh_mmcif_irqt,
+ 0, "sh_mmc:int", host);
if (ret) {
dev_err(&pdev->dev, "request_irq error (sh_mmc:int)\n");
- goto ereqirq1;
+ goto err_clk;
}
}
if (pd && pd->use_cd_gpio) {
ret = mmc_gpio_request_cd(mmc, pd->cd_gpio, 0);
if (ret < 0)
- goto erqcd;
+ goto err_clk;
}
mutex_init(&host->thread_lock);
- clk_disable_unprepare(host->hclk);
ret = mmc_add_host(mmc);
if (ret < 0)
- goto emmcaddh;
+ goto err_clk;
dev_pm_qos_expose_latency_limit(&pdev->dev, 100);
- dev_info(&pdev->dev, "driver version %s\n", DRIVER_VERSION);
- dev_dbg(&pdev->dev, "chip ver H'%04x\n",
- sh_mmcif_readl(host->addr, MMCIF_CE_VERSION) & 0x0000ffff);
+ dev_info(&pdev->dev, "Chip version 0x%04x, clock rate %luMHz\n",
+ sh_mmcif_readl(host->addr, MMCIF_CE_VERSION) & 0xffff,
+ clk_get_rate(host->hclk) / 1000000UL);
+
+ clk_disable_unprepare(host->hclk);
return ret;
-emmcaddh:
-erqcd:
- if (irq[1] >= 0)
- free_irq(irq[1], host);
-ereqirq1:
- free_irq(irq[0], host);
-ereqirq0:
- pm_runtime_suspend(&pdev->dev);
-eresume:
+err_clk:
clk_disable_unprepare(host->hclk);
-eclkupdate:
- clk_put(host->hclk);
-eclkget:
+err_pm:
pm_runtime_disable(&pdev->dev);
-eofparse:
+err_host:
mmc_free_host(mmc);
-ealloch:
- iounmap(reg);
return ret;
}
static int sh_mmcif_remove(struct platform_device *pdev)
{
struct sh_mmcif_host *host = platform_get_drvdata(pdev);
- int irq[2];
host->dying = true;
clk_prepare_enable(host->hclk);
*/
cancel_delayed_work_sync(&host->timeout_work);
- if (host->addr)
- iounmap(host->addr);
-
- irq[0] = platform_get_irq(pdev, 0);
- irq[1] = platform_get_irq(pdev, 1);
-
- free_irq(irq[0], host);
- if (irq[1] >= 0)
- free_irq(irq[1], host);
-
clk_disable_unprepare(host->hclk);
mmc_free_host(host->mmc);
pm_runtime_put_sync(&pdev->dev);
cfg.slave_id = pdata->dma->slave_id_tx;
cfg.direction = DMA_MEM_TO_DEV;
cfg.dst_addr = res->start + (CTL_SD_DATA_PORT << host->pdata->bus_shift);
+ cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
cfg.src_addr = 0;
ret = dmaengine_slave_config(host->chan_tx, &cfg);
if (ret < 0)
cfg.slave_id = pdata->dma->slave_id_rx;
cfg.direction = DMA_DEV_TO_MEM;
cfg.src_addr = cfg.dst_addr;
+ cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_2_BYTES;
cfg.dst_addr = 0;
ret = dmaengine_slave_config(host->chan_rx, &cfg);
if (ret < 0)
#define BM_SPI_CS 0x20
#define BM_SD_POWER 0x40
#define BM_SOFT_RESET 0x80
-#define BM_ONEBIT_MASK 0xFD
/* SDMMC_BLKLEN bit fields */
#define BLKL_CRCERR_ABORT 0x0800
#define STS2_DATARSP_BUSY 0x20
#define STS2_DIS_FORCECLK 0x80
+/* SDMMC_EXTCTRL bit fields */
+#define EXT_EIGHTBIT 0x04
/* MMC/SD DMA Controller Registers */
#define SDDMA_GCR 0x100
static void wmt_mci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
{
struct wmt_mci_priv *priv;
- u32 reg_tmp;
+ u32 busmode, extctrl;
priv = mmc_priv(mmc);
if (ios->clock != 0)
clk_set_rate(priv->clk_sdmmc, ios->clock);
+ busmode = readb(priv->sdmmc_base + SDMMC_BUSMODE);
+ extctrl = readb(priv->sdmmc_base + SDMMC_EXTCTRL);
+
+ busmode &= ~(BM_EIGHTBIT_MODE | BM_FOURBIT_MODE);
+ extctrl &= ~EXT_EIGHTBIT;
+
switch (ios->bus_width) {
case MMC_BUS_WIDTH_8:
- reg_tmp = readb(priv->sdmmc_base + SDMMC_EXTCTRL);
- writeb(reg_tmp | 0x04, priv->sdmmc_base + SDMMC_EXTCTRL);
+ busmode |= BM_EIGHTBIT_MODE;
+ extctrl |= EXT_EIGHTBIT;
break;
case MMC_BUS_WIDTH_4:
- reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
- writeb(reg_tmp | BM_FOURBIT_MODE, priv->sdmmc_base +
- SDMMC_BUSMODE);
-
- reg_tmp = readb(priv->sdmmc_base + SDMMC_EXTCTRL);
- writeb(reg_tmp & 0xFB, priv->sdmmc_base + SDMMC_EXTCTRL);
+ busmode |= BM_FOURBIT_MODE;
break;
case MMC_BUS_WIDTH_1:
- reg_tmp = readb(priv->sdmmc_base + SDMMC_BUSMODE);
- writeb(reg_tmp & BM_ONEBIT_MASK, priv->sdmmc_base +
- SDMMC_BUSMODE);
-
- reg_tmp = readb(priv->sdmmc_base + SDMMC_EXTCTRL);
- writeb(reg_tmp & 0xFB, priv->sdmmc_base + SDMMC_EXTCTRL);
break;
}
+
+ writeb(busmode, priv->sdmmc_base + SDMMC_BUSMODE);
+ writeb(extctrl, priv->sdmmc_base + SDMMC_EXTCTRL);
}
static int wmt_mci_get_ro(struct mmc_host *mmc)
goto fail3;
}
- ret = request_irq(dma_irq, wmt_mci_dma_isr, 32, "sdmmc", priv);
+ ret = request_irq(dma_irq, wmt_mci_dma_isr, 0, "sdmmc", priv);
if (ret) {
dev_err(&pdev->dev, "Register DMA IRQ fail\n");
goto fail4;
free_netdev(dev);
}
-static DEFINE_PCI_DEVICE_TABLE(com20020pci_id_table) = {
+static const struct pci_device_id com20020pci_id_table[] = {
{ 0x1571, 0xa001, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ 0x1571, 0xa002, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ 0x1571, 0xa003, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
PCI_DEVICE(_vend, _dev), \
.driver_data = (unsigned long)&_driverdata, \
}
-static DEFINE_PCI_DEVICE_TABLE(c_can_pci_tbl) = {
+
+static const struct pci_device_id c_can_pci_tbl[] = {
C_CAN_ID(PCI_VENDOR_ID_STMICRO, PCI_DEVICE_ID_STMICRO_CAN,
c_can_sta2x11),
C_CAN_ID(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_PCH_CAN,
priv->raminit_ctrlreg = devm_ioremap(&pdev->dev, res->start,
resource_size(res));
- if (IS_ERR(priv->raminit_ctrlreg) || priv->instance < 0)
+ if (!priv->raminit_ctrlreg || priv->instance < 0)
dev_info(&pdev->dev, "control memory is not used for raminit\n");
else
priv->raminit = c_can_hw_raminit_ti;
/* process state changes depending on the new state */
switch (new_state) {
+ case CAN_STATE_ERROR_WARNING:
+ netdev_dbg(dev, "Error Warning\n");
+ cf->can_id |= CAN_ERR_CRTL;
+ cf->data[1] = (bec.txerr > bec.rxerr) ?
+ CAN_ERR_CRTL_TX_WARNING :
+ CAN_ERR_CRTL_RX_WARNING;
+ break;
case CAN_STATE_ERROR_ACTIVE:
netdev_dbg(dev, "Error Active\n");
cf->can_id |= CAN_ERR_PROT;
if (priv->devtype_data->features & FLEXCAN_HAS_BROKEN_ERR_STATE ||
priv->can.ctrlmode & CAN_CTRLMODE_BERR_REPORTING)
reg_ctrl |= FLEXCAN_CTRL_ERR_MSK;
+ else
+ reg_ctrl &= ~FLEXCAN_CTRL_ERR_MSK;
/* save for later use */
priv->reg_ctrl_default = reg_ctrl;
.brp_inc = 1,
};
-static DEFINE_PCI_DEVICE_TABLE(pch_pci_tbl) = {
+static const struct pci_device_id pch_pci_tbl[] = {
{PCI_VENDOR_ID_INTEL, 0x8818, PCI_ANY_ID, PCI_ANY_ID,},
{0,}
};
#define EMS_PCI_BASE_SIZE 4096 /* size of controller area */
-static DEFINE_PCI_DEVICE_TABLE(ems_pci_tbl) = {
+static const struct pci_device_id ems_pci_tbl[] = {
/* CPC-PCI v1 */
{PCI_VENDOR_ID_SIEMENS, 0x2104, PCI_ANY_ID, PCI_ANY_ID,},
/* CPC-PCI v2 */
#define KVASER_PCI_VENDOR_ID2 0x1a07 /* the PCI device and vendor IDs */
#define KVASER_PCI_DEVICE_ID2 0x0008
-static DEFINE_PCI_DEVICE_TABLE(kvaser_pci_tbl) = {
+static const struct pci_device_id kvaser_pci_tbl[] = {
{KVASER_PCI_VENDOR_ID1, KVASER_PCI_DEVICE_ID1, PCI_ANY_ID, PCI_ANY_ID,},
{KVASER_PCI_VENDOR_ID2, KVASER_PCI_DEVICE_ID2, PCI_ANY_ID, PCI_ANY_ID,},
{ 0,}
0x02, 0x01, 0x40, 0x80
};
-static DEFINE_PCI_DEVICE_TABLE(peak_pci_tbl) = {
+static const struct pci_device_id peak_pci_tbl[] = {
{PEAK_PCI_VENDOR_ID, PEAK_PCI_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID,},
{PEAK_PCI_VENDOR_ID, PEAK_PCIE_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID,},
{PEAK_PCI_VENDOR_ID, PEAK_MPCI_DEVICE_ID, PCI_ANY_ID, PCI_ANY_ID,},
/* based on PLX9030 */
};
-static DEFINE_PCI_DEVICE_TABLE(plx_pci_tbl) = {
+static const struct pci_device_id plx_pci_tbl[] = {
{
/* Adlink PCI-7841/cPCI-7841 */
ADLINK_PCI_VENDOR_ID, ADLINK_PCI_DEVICE_ID,
netdev_err(dev, "setting SJA1000 into normal mode failed!\n");
}
+/*
+ * initialize SJA1000 chip:
+ * - reset chip
+ * - set output mode
+ * - set baudrate
+ * - enable interrupts
+ * - start operating mode
+ */
+static void chipset_init(struct net_device *dev)
+{
+ struct sja1000_priv *priv = netdev_priv(dev);
+
+ /* set clock divider and output control register */
+ priv->write_reg(priv, SJA1000_CDR, priv->cdr | CDR_PELICAN);
+
+ /* set acceptance filter (accept all) */
+ priv->write_reg(priv, SJA1000_ACCC0, 0x00);
+ priv->write_reg(priv, SJA1000_ACCC1, 0x00);
+ priv->write_reg(priv, SJA1000_ACCC2, 0x00);
+ priv->write_reg(priv, SJA1000_ACCC3, 0x00);
+
+ priv->write_reg(priv, SJA1000_ACCM0, 0xFF);
+ priv->write_reg(priv, SJA1000_ACCM1, 0xFF);
+ priv->write_reg(priv, SJA1000_ACCM2, 0xFF);
+ priv->write_reg(priv, SJA1000_ACCM3, 0xFF);
+
+ priv->write_reg(priv, SJA1000_OCR, priv->ocr | OCR_MODE_NORMAL);
+}
+
static void sja1000_start(struct net_device *dev)
{
struct sja1000_priv *priv = netdev_priv(dev);
if (priv->can.state != CAN_STATE_STOPPED)
set_reset_mode(dev);
+ /* Initialize chip if uninitialized at this stage */
+ if (!(priv->read_reg(priv, SJA1000_CDR) & CDR_PELICAN))
+ chipset_init(dev);
+
/* Clear error counters and error code capture */
priv->write_reg(priv, SJA1000_TXERR, 0x0);
priv->write_reg(priv, SJA1000_RXERR, 0x0);
return 0;
}
-/*
- * initialize SJA1000 chip:
- * - reset chip
- * - set output mode
- * - set baudrate
- * - enable interrupts
- * - start operating mode
- */
-static void chipset_init(struct net_device *dev)
-{
- struct sja1000_priv *priv = netdev_priv(dev);
-
- /* set clock divider and output control register */
- priv->write_reg(priv, SJA1000_CDR, priv->cdr | CDR_PELICAN);
-
- /* set acceptance filter (accept all) */
- priv->write_reg(priv, SJA1000_ACCC0, 0x00);
- priv->write_reg(priv, SJA1000_ACCC1, 0x00);
- priv->write_reg(priv, SJA1000_ACCC2, 0x00);
- priv->write_reg(priv, SJA1000_ACCC3, 0x00);
-
- priv->write_reg(priv, SJA1000_ACCM0, 0xFF);
- priv->write_reg(priv, SJA1000_ACCM1, 0xFF);
- priv->write_reg(priv, SJA1000_ACCM2, 0xFF);
- priv->write_reg(priv, SJA1000_ACCM3, 0xFF);
-
- priv->write_reg(priv, SJA1000_OCR, priv->ocr | OCR_MODE_NORMAL);
-}
-
/*
* transmit a CAN message
* message layout in the sk_buff should be like this:
};
-static DEFINE_PCI_DEVICE_TABLE(vortex_pci_tbl) = {
+static const struct pci_device_id vortex_pci_tbl[] = {
{ 0x10B7, 0x5900, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_3C590 },
{ 0x10B7, 0x5920, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_3C592 },
{ 0x10B7, 0x5970, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_3C597 },
* bit 8 indicates if this is a (0) copper or (1) fiber card
* bits 12-16 indicate card type: (0) client and (1) server
*/
-static DEFINE_PCI_DEVICE_TABLE(typhoon_pci_tbl) = {
+static const struct pci_device_id typhoon_pci_tbl[] = {
{ PCI_VENDOR_ID_3COM, PCI_DEVICE_ID_3COM_3CR990,
PCI_ANY_ID, PCI_ANY_ID, 0, 0,TYPHOON_TX },
{ PCI_VENDOR_ID_3COM, PCI_DEVICE_ID_3COM_3CR990_TX_95,
};
-static DEFINE_PCI_DEVICE_TABLE(ne2k_pci_tbl) = {
+static const struct pci_device_id ne2k_pci_tbl[] = {
{ 0x10ec, 0x8029, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_RealTek_RTL_8029 },
{ 0x1050, 0x0940, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_Winbond_89C940 },
{ 0x11f6, 0x1401, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CH_Compex_RL2000 },
CH_6915 = 0,
};
-static DEFINE_PCI_DEVICE_TABLE(starfire_pci_tbl) = {
+static const struct pci_device_id starfire_pci_tbl[] = {
{ PCI_VDEVICE(ADAPTEC, 0x6915), CH_6915 },
{ 0, }
};
#define PCI_DEVICE_ID_SGI_ACENIC 0x0009
#endif
-static DEFINE_PCI_DEVICE_TABLE(acenic_pci_tbl) = {
+static const struct pci_device_id acenic_pci_tbl[] = {
{ PCI_VENDOR_ID_ALTEON, PCI_DEVICE_ID_ALTEON_ACENIC_FIBRE,
PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_NETWORK_ETHERNET << 8, 0xffff00, },
{ PCI_VENDOR_ID_ALTEON, PCI_DEVICE_ID_ALTEON_ACENIC_COPPER,
/*
* PCI device identifiers for "new style" Linux PCI Device Drivers
*/
-static DEFINE_PCI_DEVICE_TABLE(pcnet32_pci_tbl) = {
+static const struct pci_device_id pcnet32_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_LANCE_HOME), },
{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_LANCE), },
struct xgene_enet_desc_ring *ring;
ring = pdata->tx_ring;
- if (ring && ring->cp_ring && ring->cp_ring->cp_skb)
- devm_kfree(dev, ring->cp_ring->cp_skb);
- xgene_enet_free_desc_ring(ring);
+ if (ring) {
+ if (ring->cp_ring && ring->cp_ring->cp_skb)
+ devm_kfree(dev, ring->cp_ring->cp_skb);
+ xgene_enet_free_desc_ring(ring);
+ }
ring = pdata->rx_ring;
- if (ring && ring->buf_pool && ring->buf_pool->rx_skb)
- devm_kfree(dev, ring->buf_pool->rx_skb);
- xgene_enet_free_desc_ring(ring->buf_pool);
- xgene_enet_free_desc_ring(ring);
+ if (ring) {
+ if (ring->buf_pool) {
+ if (ring->buf_pool->rx_skb)
+ devm_kfree(dev, ring->buf_pool->rx_skb);
+ xgene_enet_free_desc_ring(ring->buf_pool);
+ }
+ xgene_enet_free_desc_ring(ring);
+ }
}
static struct xgene_enet_desc_ring *xgene_enet_create_desc_ring(
struct xgene_enet_desc_ring *ring;
struct xgene_enet_pdata *pdata = netdev_priv(ndev);
struct device *dev = ndev_to_dev(ndev);
- u32 size;
+ int size;
+
+ size = xgene_enet_get_ring_size(dev, cfgsize);
+ if (size < 0)
+ return NULL;
ring = devm_kzalloc(dev, sizeof(struct xgene_enet_desc_ring),
GFP_KERNEL);
ring->cfgsize = cfgsize;
ring->id = ring_id;
- size = xgene_enet_get_ring_size(dev, cfgsize);
ring->desc_addr = dma_zalloc_coherent(dev, size, &ring->dma,
GFP_KERNEL);
if (!ring->desc_addr) {
.resume = alx_pci_error_resume,
};
-static DEFINE_PCI_DEVICE_TABLE(alx_pci_tbl) = {
+static const struct pci_device_id alx_pci_tbl[] = {
{ PCI_VDEVICE(ATTANSIC, ALX_DEV_ID_AR8161),
.driver_data = ALX_DEV_QUIRK_MSI_INTX_DISABLE_BUG },
{ PCI_VDEVICE(ATTANSIC, ALX_DEV_ID_E2200),
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static DEFINE_PCI_DEVICE_TABLE(atl1c_pci_tbl) = {
+static const struct pci_device_id atl1c_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_ATTANSIC, PCI_DEVICE_ID_ATTANSIC_L1C)},
{PCI_DEVICE(PCI_VENDOR_ID_ATTANSIC, PCI_DEVICE_ID_ATTANSIC_L2C)},
{PCI_DEVICE(PCI_VENDOR_ID_ATTANSIC, PCI_DEVICE_ID_ATHEROS_L2C_B)},
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static DEFINE_PCI_DEVICE_TABLE(atl1e_pci_tbl) = {
+static const struct pci_device_id atl1e_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_ATTANSIC, PCI_DEVICE_ID_ATTANSIC_L1E)},
{PCI_DEVICE(PCI_VENDOR_ID_ATTANSIC, 0x1066)},
/* required last entry */
/*
* atl1_pci_tbl - PCI Device ID Table
*/
-static DEFINE_PCI_DEVICE_TABLE(atl1_pci_tbl) = {
+static const struct pci_device_id atl1_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_ATTANSIC, PCI_DEVICE_ID_ATTANSIC_L1)},
/* required last entry */
{0,}
/*
* atl2_pci_tbl - PCI Device ID Table
*/
-static DEFINE_PCI_DEVICE_TABLE(atl2_pci_tbl) = {
+static const struct pci_device_id atl2_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_ATTANSIC, PCI_DEVICE_ID_ATTANSIC_L2)},
/* required last entry */
{0,}
#ifdef CONFIG_B44_PCI
-static DEFINE_PCI_DEVICE_TABLE(b44_pci_tbl) = {
+static const struct pci_device_id b44_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_BCM4401) },
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_BCM4401B0) },
{ PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_BCM4401B1) },
{ "Broadcom NetXtreme II BCM5716 1000Base-SX" },
};
-static DEFINE_PCI_DEVICE_TABLE(bnx2_pci_tbl) = {
+static const struct pci_device_id bnx2_pci_tbl[] = {
{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706,
PCI_VENDOR_ID_HP, 0x3101, 0, 0, NC370T },
{ PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706,
#ifdef BNX2X_STOP_ON_ERROR
fp->tpa_queue_used |= (1 << queue);
-#ifdef _ASM_GENERIC_INT_L64_H
- DP(NETIF_MSG_RX_STATUS, "fp->tpa_queue_used = 0x%lx\n",
-#else
DP(NETIF_MSG_RX_STATUS, "fp->tpa_queue_used = 0x%llx\n",
-#endif
fp->tpa_queue_used);
#endif
}
#define PCI_DEVICE_ID_NX2_57811_VF CHIP_NUM_57811_VF
#endif
-static DEFINE_PCI_DEVICE_TABLE(bnx2x_pci_tbl) = {
+static const struct pci_device_id bnx2x_pci_tbl[] = {
{ PCI_VDEVICE(BROADCOM, PCI_DEVICE_ID_NX2_57710), BCM57710 },
{ PCI_VDEVICE(BROADCOM, PCI_DEVICE_ID_NX2_57711), BCM57711 },
{ PCI_VDEVICE(BROADCOM, PCI_DEVICE_ID_NX2_57711E), BCM57711E },
}
#define BNX2X_PREV_UNDI_PROD_ADDR(p) (BAR_TSTRORM_INTMEM + 0x1508 + ((p) << 4))
+#define BNX2X_PREV_UNDI_PROD_ADDR_H(f) (BAR_TSTRORM_INTMEM + \
+ 0x1848 + ((f) << 4))
#define BNX2X_PREV_UNDI_RCQ(val) ((val) & 0xffff)
#define BNX2X_PREV_UNDI_BD(val) ((val) >> 16 & 0xffff)
#define BNX2X_PREV_UNDI_PROD(rcq, bd) ((bd) << 16 | (rcq))
#define BCM_5710_UNDI_FW_MF_MAJOR (0x07)
#define BCM_5710_UNDI_FW_MF_MINOR (0x08)
#define BCM_5710_UNDI_FW_MF_VERS (0x05)
-#define BNX2X_PREV_UNDI_MF_PORT(p) (BAR_TSTRORM_INTMEM + 0x150c + ((p) << 4))
-#define BNX2X_PREV_UNDI_MF_FUNC(f) (BAR_TSTRORM_INTMEM + 0x184c + ((f) << 4))
static bool bnx2x_prev_is_after_undi(struct bnx2x *bp)
{
return false;
}
-static bool bnx2x_prev_unload_undi_fw_supports_mf(struct bnx2x *bp)
-{
- u8 major, minor, version;
- u32 fw;
-
- /* Must check that FW is loaded */
- if (!(REG_RD(bp, MISC_REG_RESET_REG_1) &
- MISC_REGISTERS_RESET_REG_1_RST_XSEM)) {
- BNX2X_DEV_INFO("XSEM is reset - UNDI MF FW is not loaded\n");
- return false;
- }
-
- /* Read Currently loaded FW version */
- fw = REG_RD(bp, XSEM_REG_PRAM);
- major = fw & 0xff;
- minor = (fw >> 0x8) & 0xff;
- version = (fw >> 0x10) & 0xff;
- BNX2X_DEV_INFO("Loaded FW: 0x%08x: Major 0x%02x Minor 0x%02x Version 0x%02x\n",
- fw, major, minor, version);
-
- if (major > BCM_5710_UNDI_FW_MF_MAJOR)
- return true;
-
- if ((major == BCM_5710_UNDI_FW_MF_MAJOR) &&
- (minor > BCM_5710_UNDI_FW_MF_MINOR))
- return true;
-
- if ((major == BCM_5710_UNDI_FW_MF_MAJOR) &&
- (minor == BCM_5710_UNDI_FW_MF_MINOR) &&
- (version >= BCM_5710_UNDI_FW_MF_VERS))
- return true;
-
- return false;
-}
-
-static void bnx2x_prev_unload_undi_mf(struct bnx2x *bp)
-{
- int i;
-
- /* Due to legacy (FW) code, the first function on each engine has a
- * different offset macro from the rest of the functions.
- * Setting this for all 8 functions is harmless regardless of whether
- * this is actually a multi-function device.
- */
- for (i = 0; i < 2; i++)
- REG_WR(bp, BNX2X_PREV_UNDI_MF_PORT(i), 1);
-
- for (i = 2; i < 8; i++)
- REG_WR(bp, BNX2X_PREV_UNDI_MF_FUNC(i - 2), 1);
-
- BNX2X_DEV_INFO("UNDI FW (MF) set to discard\n");
-}
-
-static void bnx2x_prev_unload_undi_inc(struct bnx2x *bp, u8 port, u8 inc)
+static void bnx2x_prev_unload_undi_inc(struct bnx2x *bp, u8 inc)
{
u16 rcq, bd;
- u32 tmp_reg = REG_RD(bp, BNX2X_PREV_UNDI_PROD_ADDR(port));
+ u32 addr, tmp_reg;
+ if (BP_FUNC(bp) < 2)
+ addr = BNX2X_PREV_UNDI_PROD_ADDR(BP_PORT(bp));
+ else
+ addr = BNX2X_PREV_UNDI_PROD_ADDR_H(BP_FUNC(bp) - 2);
+
+ tmp_reg = REG_RD(bp, addr);
rcq = BNX2X_PREV_UNDI_RCQ(tmp_reg) + inc;
bd = BNX2X_PREV_UNDI_BD(tmp_reg) + inc;
tmp_reg = BNX2X_PREV_UNDI_PROD(rcq, bd);
- REG_WR(bp, BNX2X_PREV_UNDI_PROD_ADDR(port), tmp_reg);
+ REG_WR(bp, addr, tmp_reg);
- BNX2X_DEV_INFO("UNDI producer [%d] rings bd -> 0x%04x, rcq -> 0x%04x\n",
- port, bd, rcq);
+ BNX2X_DEV_INFO("UNDI producer [%d/%d][%08x] rings bd -> 0x%04x, rcq -> 0x%04x\n",
+ BP_PORT(bp), BP_FUNC(bp), addr, bd, rcq);
}
static int bnx2x_prev_mcp_done(struct bnx2x *bp)
/* Reset should be performed after BRB is emptied */
if (reset_reg & MISC_REGISTERS_RESET_REG_1_RST_BRB1) {
u32 timer_count = 1000;
- bool need_write = true;
/* Close the MAC Rx to prevent BRB from filling up */
bnx2x_prev_unload_close_mac(bp, &mac_vals);
else
timer_count--;
- /* New UNDI FW supports MF and contains better
- * cleaning methods - might be redundant but harmless.
- */
- if (bnx2x_prev_unload_undi_fw_supports_mf(bp)) {
- if (need_write) {
- bnx2x_prev_unload_undi_mf(bp);
- need_write = false;
- }
- } else if (prev_undi) {
- /* If UNDI resides in memory,
- * manually increment it
- */
- bnx2x_prev_unload_undi_inc(bp, BP_PORT(bp), 1);
- }
+ /* If UNDI resides in memory, manually increment it */
+ if (prev_undi)
+ bnx2x_prev_unload_undi_inc(bp, 1);
+
udelay(10);
}
#define TG3_DRV_DATA_FLAG_10_100_ONLY 0x0001
#define TG3_DRV_DATA_FLAG_5705_10_100 0x0002
-static DEFINE_PCI_DEVICE_TABLE(tg3_pci_tbl) = {
+static const struct pci_device_id tg3_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5700)},
{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5701)},
{PCI_DEVICE(PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_TIGON3_5702)},
spin_lock_bh(&tp->lock);
if (!tp->hw_stats) {
+ *stats = tp->net_stats_prev;
spin_unlock_bh(&tp->lock);
- return &tp->net_stats_prev;
+ return stats;
}
tg3_get_nstats(tp, stats);
return TG3_RX_RET_MAX_SIZE_5705;
}
-static DEFINE_PCI_DEVICE_TABLE(tg3_write_reorder_chipsets) = {
+static const struct pci_device_id tg3_write_reorder_chipsets[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_FE_GATE_700C) },
{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_8131_BRIDGE) },
{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8385_0) },
#define TEST_BUFFER_SIZE 0x2000
-static DEFINE_PCI_DEVICE_TABLE(tg3_dma_wait_state_chipsets) = {
+static const struct pci_device_id tg3_dma_wait_state_chipsets[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_APPLE, PCI_DEVICE_ID_APPLE_UNI_N_PCI15) },
{ },
};
free_netdev(netdev);
}
-static DEFINE_PCI_DEVICE_TABLE(bnad_pci_id_table) = {
+static const struct pci_device_id bnad_pci_id_table[] = {
{
PCI_DEVICE(PCI_VENDOR_ID_BROCADE,
PCI_DEVICE_ID_BROCADE_CT),
};
-DEFINE_PCI_DEVICE_TABLE(t1_pci_tbl) = {
+const struct pci_device_id t1_pci_tbl[] = {
CH_DEVICE(8, 0, CH_BRD_T110_1CU),
CH_DEVICE(8, 1, CH_BRD_T110_1CU),
CH_DEVICE(7, 0, CH_BRD_N110_1F),
#define CH_DEVICE(devid, idx) \
{ PCI_VENDOR_ID_CHELSIO, devid, PCI_ANY_ID, PCI_ANY_ID, 0, 0, idx }
-static DEFINE_PCI_DEVICE_TABLE(cxgb3_pci_tbl) = {
+static const struct pci_device_id cxgb3_pci_tbl[] = {
CH_DEVICE(0x20, 0), /* PE9000 */
CH_DEVICE(0x21, 1), /* T302E */
CH_DEVICE(0x22, 2), /* T310E */
struct tid_info tids;
void **tid_release_head;
spinlock_t tid_release_lock;
+ struct workqueue_struct *workq;
struct work_struct tid_release_task;
struct work_struct db_full_task;
struct work_struct db_drop_task;
#define CH_DEVICE(devid, data) { PCI_VDEVICE(CHELSIO, devid), (data) }
-static DEFINE_PCI_DEVICE_TABLE(cxgb4_pci_tbl) = {
+static const struct pci_device_id cxgb4_pci_tbl[] = {
CH_DEVICE(0xa000, 0), /* PE10K */
CH_DEVICE(0x4001, -1),
CH_DEVICE(0x4002, -1),
return ret;
}
-static struct workqueue_struct *workq;
-
/**
* link_start - enable a port
* @dev: the port to enable
adap->tid_release_head = (void **)((uintptr_t)p | chan);
if (!adap->tid_release_task_busy) {
adap->tid_release_task_busy = true;
- queue_work(workq, &adap->tid_release_task);
+ queue_work(adap->workq, &adap->tid_release_task);
}
spin_unlock_bh(&adap->tid_release_lock);
}
notify_rdma_uld(adap, CXGB4_CONTROL_DB_FULL);
t4_set_reg_field(adap, SGE_INT_ENABLE3,
DBFIFO_HP_INT | DBFIFO_LP_INT, 0);
- queue_work(workq, &adap->db_full_task);
+ queue_work(adap->workq, &adap->db_full_task);
}
}
disable_dbs(adap);
notify_rdma_uld(adap, CXGB4_CONTROL_DB_FULL);
}
- queue_work(workq, &adap->db_drop_task);
+ queue_work(adap->workq, &adap->db_drop_task);
}
static void uld_attach(struct adapter *adap, unsigned int uld)
goto out_disable_device;
}
+ adapter->workq = create_singlethread_workqueue("cxgb4");
+ if (!adapter->workq) {
+ err = -ENOMEM;
+ goto out_free_adapter;
+ }
+
/* PCI device has been enabled */
adapter->flags |= DEV_ENABLED;
out_unmap_bar0:
iounmap(adapter->regs);
out_free_adapter:
+ if (adapter->workq)
+ destroy_workqueue(adapter->workq);
+
kfree(adapter);
out_disable_device:
pci_disable_pcie_error_reporting(pdev);
if (adapter) {
int i;
+ /* Tear down per-adapter Work Queue first since it can contain
+ * references to our adapter data structure.
+ */
+ destroy_workqueue(adapter->workq);
+
if (is_offload(adapter))
detach_ulds(adapter);
{
int ret;
- workq = create_singlethread_workqueue("cxgb4");
- if (!workq)
- return -ENOMEM;
-
/* Debugfs support is optional, just warn if this fails */
cxgb4_debugfs_root = debugfs_create_dir(KBUILD_MODNAME, NULL);
if (!cxgb4_debugfs_root)
pr_warn("could not create debugfs entry, continuing\n");
ret = pci_register_driver(&cxgb4_driver);
- if (ret < 0) {
+ if (ret < 0)
debugfs_remove(cxgb4_debugfs_root);
- destroy_workqueue(workq);
- }
register_inet6addr_notifier(&cxgb4_inet6addr_notifier);
unregister_inet6addr_notifier(&cxgb4_inet6addr_notifier);
pci_unregister_driver(&cxgb4_driver);
debugfs_remove(cxgb4_debugfs_root); /* NULL ok */
- flush_workqueue(workq);
- destroy_workqueue(workq);
}
module_init(cxgb4_init_module);
FW_EQ_ETH_CMD_PFN(adap->fn) | FW_EQ_ETH_CMD_VFN(0));
c.alloc_to_len16 = htonl(FW_EQ_ETH_CMD_ALLOC |
FW_EQ_ETH_CMD_EQSTART | FW_LEN16(c));
- c.viid_pkd = htonl(FW_EQ_ETH_CMD_VIID(pi->viid));
+ c.viid_pkd = htonl(FW_EQ_ETH_CMD_AUTOEQUEQE |
+ FW_EQ_ETH_CMD_VIID(pi->viid));
c.fetchszm_to_iqid = htonl(FW_EQ_ETH_CMD_HOSTFCMODE(2) |
FW_EQ_ETH_CMD_PCIECHN(pi->tx_chan) |
FW_EQ_ETH_CMD_FETCHRO(1) |
#define FW_EQ_ETH_CMD_CIDXFTHRESH(x) ((x) << 16)
#define FW_EQ_ETH_CMD_EQSIZE(x) ((x) << 0)
+#define FW_EQ_ETH_CMD_AUTOEQUEQE (1U << 30)
#define FW_EQ_ETH_CMD_VIID(x) ((x) << 16)
struct fw_eq_ctrl_cmd {
#define CH_DEVICE(devid, idx) \
{ PCI_VENDOR_ID_CHELSIO, devid, PCI_ANY_ID, PCI_ANY_ID, 0, 0, idx }
-static DEFINE_PCI_DEVICE_TABLE(cxgb4vf_pci_tbl) = {
+static const struct pci_device_id cxgb4vf_pci_tbl[] = {
CH_DEVICE(0xb000, 0), /* PE10K FPGA */
CH_DEVICE(0x4800, 0), /* T440-dbg */
CH_DEVICE(0x4801, 0), /* T420-cr */
cmd.alloc_to_len16 = cpu_to_be32(FW_EQ_ETH_CMD_ALLOC |
FW_EQ_ETH_CMD_EQSTART |
FW_LEN16(cmd));
- cmd.viid_pkd = cpu_to_be32(FW_EQ_ETH_CMD_VIID(pi->viid));
+ cmd.viid_pkd = cpu_to_be32(FW_EQ_ETH_CMD_AUTOEQUEQE |
+ FW_EQ_ETH_CMD_VIID(pi->viid));
cmd.fetchszm_to_iqid =
cpu_to_be32(FW_EQ_ETH_CMD_HOSTFCMODE(SGE_HOSTFCMODE_STPG) |
FW_EQ_ETH_CMD_PCIECHN(pi->port_id) |
#define PCI_DEVICE_ID_CISCO_VIC_ENET_VF 0x0071 /* enet SRIOV VF */
/* Supported devices */
-static DEFINE_PCI_DEVICE_TABLE(enic_id_table) = {
+static const struct pci_device_id enic_id_table[] = {
{ PCI_VDEVICE(CISCO, PCI_DEVICE_ID_CISCO_VIC_ENET) },
{ PCI_VDEVICE(CISCO, PCI_DEVICE_ID_CISCO_VIC_ENET_DYN) },
{ PCI_VDEVICE(CISCO, PCI_DEVICE_ID_CISCO_VIC_ENET_VF) },
static unsigned int de_ok_to_advertise (struct de_private *de, u32 new_media);
-static DEFINE_PCI_DEVICE_TABLE(de_pci_tbl) = {
+static const struct pci_device_id de_pci_tbl[] = {
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP_PLUS,
pci_disable_device (pdev);
}
-static DEFINE_PCI_DEVICE_TABLE(de4x5_pci_tbl) = {
+static const struct pci_device_id de4x5_pci_tbl[] = {
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP_PLUS,
-static DEFINE_PCI_DEVICE_TABLE(dmfe_pci_tbl) = {
+static const struct pci_device_id dmfe_pci_tbl[] = {
{ 0x1282, 0x9132, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PCI_DM9132_ID },
{ 0x1282, 0x9102, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PCI_DM9102_ID },
{ 0x1282, 0x9100, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PCI_DM9100_ID },
};
-static DEFINE_PCI_DEVICE_TABLE(tulip_pci_tbl) = {
+static const struct pci_device_id tulip_pci_tbl[] = {
{ 0x1011, 0x0009, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DC21140 },
{ 0x1011, 0x0019, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DC21143 },
{ 0x11AD, 0x0002, PCI_ANY_ID, PCI_ANY_ID, 0, 0, LC82C168 },
#endif
};
-DEFINE_PCI_DEVICE_TABLE(early_486_chipsets) = {
+const struct pci_device_id early_486_chipsets[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82424) },
{ PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_496) },
{ },
}
-static DEFINE_PCI_DEVICE_TABLE(uli526x_pci_tbl) = {
+static const struct pci_device_id uli526x_pci_tbl[] = {
{ 0x10B9, 0x5261, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PCI_ULI5261_ID },
{ 0x10B9, 0x5263, PCI_ANY_ID, PCI_ANY_ID, 0, 0, PCI_ULI5263_ID },
{ 0, }
CanHaveMII=1, HasBrokenTx=2, AlwaysFDX=4, FDXOnNoMII=8,
};
-static DEFINE_PCI_DEVICE_TABLE(w840_pci_tbl) = {
+static const struct pci_device_id w840_pci_tbl[] = {
{ 0x1050, 0x0840, PCI_ANY_ID, 0x8153, 0, 0, 0 },
{ 0x1050, 0x0840, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1 },
{ 0x11f6, 0x2011, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2 },
-static DEFINE_PCI_DEVICE_TABLE(xircom_pci_table) = {
+static const struct pci_device_id xircom_pci_table[] = {
{ PCI_VDEVICE(XIRCOM, 0x0003), },
{0,},
};
driver_data Data private to the driver.
*/
-static DEFINE_PCI_DEVICE_TABLE(rio_pci_tbl) = {
+static const struct pci_device_id rio_pci_tbl[] = {
{0x1186, 0x4000, PCI_ANY_ID, PCI_ANY_ID, },
{0x13f0, 0x1021, PCI_ANY_ID, PCI_ANY_ID, },
{ }
#define USE_IO_OPS 1
#endif
-static DEFINE_PCI_DEVICE_TABLE(sundance_pci_tbl) = {
+static const struct pci_device_id sundance_pci_tbl[] = {
{ 0x1186, 0x1002, 0x1186, 0x1002, 0, 0, 0 },
{ 0x1186, 0x1002, 0x1186, 0x1003, 0, 0, 1 },
{ 0x1186, 0x1002, 0x1186, 0x1012, 0, 0, 2 },
*/
void be_roce_dev_open(struct be_adapter *);
void be_roce_dev_close(struct be_adapter *);
+void be_roce_dev_shutdown(struct be_adapter *);
#endif /* BE_H */
module_param(rx_frag_size, ushort, S_IRUGO);
MODULE_PARM_DESC(rx_frag_size, "Size of a fragment that holds rcvd data.");
-static DEFINE_PCI_DEVICE_TABLE(be_dev_ids) = {
+static const struct pci_device_id be_dev_ids[] = {
{ PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID1) },
{ PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID2) },
{ PCI_DEVICE(BE_VENDOR_ID, OC_DEVICE_ID1) },
if (!adapter)
return;
+ be_roce_dev_shutdown(adapter);
cancel_delayed_work_sync(&adapter->work);
cancel_delayed_work_sync(&adapter->func_recovery_work);
{
if (ocrdma_drv && adapter->ocrdma_dev &&
ocrdma_drv->state_change_handler)
- ocrdma_drv->state_change_handler(adapter->ocrdma_dev, 0);
+ ocrdma_drv->state_change_handler(adapter->ocrdma_dev,
+ BE_DEV_UP);
}
void be_roce_dev_open(struct be_adapter *adapter)
{
if (ocrdma_drv && adapter->ocrdma_dev &&
ocrdma_drv->state_change_handler)
- ocrdma_drv->state_change_handler(adapter->ocrdma_dev, 1);
+ ocrdma_drv->state_change_handler(adapter->ocrdma_dev,
+ BE_DEV_DOWN);
}
void be_roce_dev_close(struct be_adapter *adapter)
}
}
+void be_roce_dev_shutdown(struct be_adapter *adapter)
+{
+ if (be_roce_supported(adapter)) {
+ mutex_lock(&be_adapter_list_lock);
+ if (ocrdma_drv && adapter->ocrdma_dev &&
+ ocrdma_drv->state_change_handler)
+ ocrdma_drv->state_change_handler(adapter->ocrdma_dev,
+ BE_DEV_SHUTDOWN);
+ mutex_unlock(&be_adapter_list_lock);
+ }
+}
+
int be_roce_register_driver(struct ocrdma_driver *drv)
{
struct be_adapter *dev;
enum {
BE_DEV_UP = 0,
- BE_DEV_DOWN = 1
+ BE_DEV_DOWN = 1,
+ BE_DEV_SHUTDOWN = 2
};
/* APIs for RoCE driver to register callback handlers,
return 0;
}
-static DEFINE_PCI_DEVICE_TABLE(fealnx_pci_tbl) = {
+static const struct pci_device_id fealnx_pci_tbl[] = {
{0x1516, 0x0800, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
{0x1516, 0x0803, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1},
{0x1516, 0x0891, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 2},
struct clk *clk_enet_out;
struct clk *clk_ptp;
+ bool ptp_clk_on;
+ struct mutex ptp_clk_mutex;
+
/* The saved address of a sent-in-place packet/buffer, for skfree(). */
unsigned char *tx_bounce[TX_RING_SIZE];
struct sk_buff *tx_skbuff[TX_RING_SIZE];
u32 cycle_speed;
int hwts_rx_en;
int hwts_tx_en;
- struct timer_list time_keep;
+ struct delayed_work time_keep;
struct regulator *reg_phy;
};
goto failed_clk_enet_out;
}
if (fep->clk_ptp) {
+ mutex_lock(&fep->ptp_clk_mutex);
ret = clk_prepare_enable(fep->clk_ptp);
- if (ret)
+ if (ret) {
+ mutex_unlock(&fep->ptp_clk_mutex);
goto failed_clk_ptp;
+ } else {
+ fep->ptp_clk_on = true;
+ }
+ mutex_unlock(&fep->ptp_clk_mutex);
}
} else {
clk_disable_unprepare(fep->clk_ahb);
clk_disable_unprepare(fep->clk_ipg);
if (fep->clk_enet_out)
clk_disable_unprepare(fep->clk_enet_out);
- if (fep->clk_ptp)
+ if (fep->clk_ptp) {
+ mutex_lock(&fep->ptp_clk_mutex);
clk_disable_unprepare(fep->clk_ptp);
+ fep->ptp_clk_on = false;
+ mutex_unlock(&fep->ptp_clk_mutex);
+ }
}
return 0;
if (IS_ERR(fep->clk_enet_out))
fep->clk_enet_out = NULL;
+ fep->ptp_clk_on = false;
+ mutex_init(&fep->ptp_clk_mutex);
fep->clk_ptp = devm_clk_get(&pdev->dev, "ptp");
fep->bufdesc_ex =
pdev->id_entry->driver_data & FEC_QUIRK_HAS_BUFDESC_EX;
struct net_device *ndev = platform_get_drvdata(pdev);
struct fec_enet_private *fep = netdev_priv(ndev);
+ cancel_delayed_work_sync(&fep->time_keep);
cancel_work_sync(&fep->tx_timeout_work);
unregister_netdev(ndev);
fec_enet_mii_remove(fep);
- del_timer_sync(&fep->time_keep);
if (fep->reg_phy)
regulator_disable(fep->reg_phy);
if (fep->ptp_clock)
u64 ns;
unsigned long flags;
+ mutex_lock(&fep->ptp_clk_mutex);
+ /* Check the ptp clock */
+ if (!fep->ptp_clk_on) {
+ mutex_unlock(&fep->ptp_clk_mutex);
+ return -EINVAL;
+ }
+
ns = ts->tv_sec * 1000000000ULL;
ns += ts->tv_nsec;
spin_lock_irqsave(&fep->tmreg_lock, flags);
timecounter_init(&fep->tc, &fep->cc, ns);
spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+ mutex_unlock(&fep->ptp_clk_mutex);
return 0;
}
* fec_time_keep - call timecounter_read every second to avoid timer overrun
* because ENET just support 32bit counter, will timeout in 4s
*/
-static void fec_time_keep(unsigned long _data)
+static void fec_time_keep(struct work_struct *work)
{
- struct fec_enet_private *fep = (struct fec_enet_private *)_data;
+ struct delayed_work *dwork = to_delayed_work(work);
+ struct fec_enet_private *fep = container_of(dwork, struct fec_enet_private, time_keep);
u64 ns;
unsigned long flags;
- spin_lock_irqsave(&fep->tmreg_lock, flags);
- ns = timecounter_read(&fep->tc);
- spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+ mutex_lock(&fep->ptp_clk_mutex);
+ if (fep->ptp_clk_on) {
+ spin_lock_irqsave(&fep->tmreg_lock, flags);
+ ns = timecounter_read(&fep->tc);
+ spin_unlock_irqrestore(&fep->tmreg_lock, flags);
+ }
+ mutex_unlock(&fep->ptp_clk_mutex);
- mod_timer(&fep->time_keep, jiffies + HZ);
+ schedule_delayed_work(&fep->time_keep, HZ);
}
/**
fec_ptp_start_cyclecounter(ndev);
- init_timer(&fep->time_keep);
- fep->time_keep.data = (unsigned long)fep;
- fep->time_keep.function = fec_time_keep;
- fep->time_keep.expires = jiffies + HZ;
- add_timer(&fep->time_keep);
+ INIT_DELAYED_WORK(&fep->time_keep, fec_time_keep);
fep->ptp_clock = ptp_clock_register(&fep->ptp_caps, &pdev->dev);
if (IS_ERR(fep->ptp_clock)) {
fep->ptp_clock = NULL;
pr_err("ptp_clock_register failed\n");
}
+
+ schedule_delayed_work(&fep->time_keep, HZ);
}
#endif
#ifdef CONFIG_PCI
-static DEFINE_PCI_DEVICE_TABLE(hp100_pci_tbl) = {
+static const struct pci_device_id hp100_pci_tbl[] = {
{PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_J2585A, PCI_ANY_ID, PCI_ANY_ID,},
{PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_J2585B, PCI_ANY_ID, PCI_ANY_ID,},
{PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_J2970A, PCI_ANY_ID, PCI_ANY_ID,},
#
# Makefile for the eHEA ethernet device driver for IBM eServer System p
#
-ehea-y = ehea_main.o ehea_phyp.o ehea_qmr.o ehea_ethtool.o ehea_phyp.o
+ehea-y = ehea_main.o ehea_phyp.o ehea_qmr.o ehea_ethtool.o
obj-$(CONFIG_EHEA) += ehea.o
atomic_add(buffers_added, &(pool->available));
}
+/*
+ * The final 8 bytes of the buffer list is a counter of frames dropped
+ * because there was not a buffer in the buffer list capable of holding
+ * the frame.
+ */
+static void ibmveth_update_rx_no_buffer(struct ibmveth_adapter *adapter)
+{
+ __be64 *p = adapter->buffer_list_addr + 4096 - 8;
+
+ adapter->rx_no_buffer = be64_to_cpup(p);
+}
+
/* replenish routine */
static void ibmveth_replenish_task(struct ibmveth_adapter *adapter)
{
ibmveth_replenish_buffer_pool(adapter, pool);
}
- adapter->rx_no_buffer = *(u64 *)(((char*)adapter->buffer_list_addr) +
- 4096 - 8);
+ ibmveth_update_rx_no_buffer(adapter);
}
/* empty and free ana buffer pool - also used to do cleanup in error paths */
free_irq(netdev->irq, netdev);
- adapter->rx_no_buffer = *(u64 *)(((char *)adapter->buffer_list_addr) +
- 4096 - 8);
+ ibmveth_update_rx_no_buffer(adapter);
ibmveth_cleanup(adapter);
"D-Link NIC IP1000A"
};
-static DEFINE_PCI_DEVICE_TABLE(ipg_pci_tbl) = {
+static const struct pci_device_id ipg_pci_tbl[] = {
{ PCI_VDEVICE(SUNDANCE, 0x1023), 0 },
{ PCI_VDEVICE(SUNDANCE, 0x2021), 1 },
{ PCI_VDEVICE(DLINK, 0x9021), 2 },
#define INTEL_8255X_ETHERNET_DEVICE(device_id, ich) {\
PCI_VENDOR_ID_INTEL, device_id, PCI_ANY_ID, PCI_ANY_ID, \
PCI_CLASS_NETWORK_ETHERNET << 8, 0xFFFF00, ich }
-static DEFINE_PCI_DEVICE_TABLE(e100_id_table) = {
+static const struct pci_device_id e100_id_table[] = {
INTEL_8255X_ETHERNET_DEVICE(0x1029, 0),
INTEL_8255X_ETHERNET_DEVICE(0x1030, 0),
INTEL_8255X_ETHERNET_DEVICE(0x1031, 3),
* Macro expands to...
* {PCI_DEVICE(PCI_VENDOR_ID_INTEL, device_id)}
*/
-static DEFINE_PCI_DEVICE_TABLE(e1000_pci_tbl) = {
+static const struct pci_device_id e1000_pci_tbl[] = {
INTEL_E1000_ETHERNET_DEVICE(0x1000),
INTEL_E1000_ETHERNET_DEVICE(0x1001),
INTEL_E1000_ETHERNET_DEVICE(0x1004),
* e1000_mng_enable_host_if - Checks host interface is enabled
* @hw: pointer to the HW structure
*
- * Returns E1000_success upon success, else E1000_ERR_HOST_INTERFACE_COMMAND
+ * Returns 0 upon success, else -E1000_ERR_HOST_INTERFACE_COMMAND
*
* This function checks whether the HOST IF is enabled for command operation
* and also checks whether the previous command is completed. It busy waits
}
if (i == E1000_MNG_DHCP_COMMAND_TIMEOUT) {
- e_dbg("Previous command timeout failed .\n");
+ e_dbg("Previous command timeout failed.\n");
return -E1000_ERR_HOST_INTERFACE_COMMAND;
}
#include <scsi/fc/fc_fcoe.h>
#include <scsi/libfc.h>
#include <scsi/libfcoe.h>
+#include <uapi/linux/dcbnl.h>
#include "i40e.h"
#include "i40e_fcoe.h"
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static DEFINE_PCI_DEVICE_TABLE(i40e_pci_tbl) = {
+static const struct pci_device_id i40e_pci_tbl[] = {
{PCI_VDEVICE(INTEL, I40E_DEV_ID_SFP_XL710), 0},
{PCI_VDEVICE(INTEL, I40E_DEV_ID_QEMU), 0},
{PCI_VDEVICE(INTEL, I40E_DEV_ID_KX_A), 0},
switch (vsi->back->hw.phy.link_info.link_speed) {
case I40E_LINK_SPEED_40GB:
- strncpy(speed, "40 Gbps", SPEED_SIZE);
+ strlcpy(speed, "40 Gbps", SPEED_SIZE);
break;
case I40E_LINK_SPEED_10GB:
- strncpy(speed, "10 Gbps", SPEED_SIZE);
+ strlcpy(speed, "10 Gbps", SPEED_SIZE);
break;
case I40E_LINK_SPEED_1GB:
- strncpy(speed, "1000 Mbps", SPEED_SIZE);
+ strlcpy(speed, "1000 Mbps", SPEED_SIZE);
break;
default:
break;
switch (vsi->back->hw.fc.current_mode) {
case I40E_FC_FULL:
- strncpy(fc, "RX/TX", FC_SIZE);
+ strlcpy(fc, "RX/TX", FC_SIZE);
break;
case I40E_FC_TX_PAUSE:
- strncpy(fc, "TX", FC_SIZE);
+ strlcpy(fc, "TX", FC_SIZE);
break;
case I40E_FC_RX_PAUSE:
- strncpy(fc, "RX", FC_SIZE);
+ strlcpy(fc, "RX", FC_SIZE);
break;
default:
- strncpy(fc, "None", FC_SIZE);
+ strlcpy(fc, "None", FC_SIZE);
break;
}
dv.minor_version = DRV_VERSION_MINOR;
dv.build_version = DRV_VERSION_BUILD;
dv.subbuild_version = 0;
- strncpy(dv.driver_string, DRV_VERSION, sizeof(dv.driver_string));
+ strlcpy(dv.driver_string, DRV_VERSION, sizeof(dv.driver_string));
i40e_aq_send_driver_version(&pf->hw, &dv, NULL);
}
if (alloc_qvectors) {
/* allocate memory for q_vector pointers */
- size = sizeof(struct i40e_q_vectors *) * vsi->num_q_vectors;
+ size = sizeof(struct i40e_q_vector *) * vsi->num_q_vectors;
vsi->q_vectors = kzalloc(size, GFP_KERNEL);
if (!vsi->q_vectors) {
ret = -ENOMEM;
*
* Writes a 16 bit words buffer to the Shadow RAM using the admin command.
**/
-i40e_status i40e_write_nvm_aq(struct i40e_hw *hw, u8 module_pointer,
- u32 offset, u16 words, void *data,
- bool last_command)
+static i40e_status i40e_write_nvm_aq(struct i40e_hw *hw, u8 module_pointer,
+ u32 offset, u16 words, void *data,
+ bool last_command)
{
i40e_status ret_code = I40E_ERR_NVM;
u32 prttsyn_stat;
int n;
- if (pf->flags & I40E_FLAG_PTP)
+ if (!(pf->flags & I40E_FLAG_PTP))
return;
prttsyn_stat = rd32(hw, I40E_PRTTSYN_STAT_1);
static int i40e_vc_send_msg_to_vf(struct i40e_vf *vf, u32 v_opcode,
u32 v_retval, u8 *msg, u16 msglen)
{
- struct i40e_pf *pf = vf->pf;
- struct i40e_hw *hw = &pf->hw;
- int abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
+ struct i40e_pf *pf;
+ struct i40e_hw *hw;
+ int abs_vf_id;
i40e_status aq_ret;
+ /* validate the request */
+ if (!vf || vf->vf_id >= vf->pf->num_alloc_vfs)
+ return -EINVAL;
+
+ pf = vf->pf;
+ hw = &pf->hw;
+ abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
+
/* single place to detect unsuccessful return values */
if (v_retval) {
vf->num_invalid_msgs++;
{
struct i40e_hw *hw = &pf->hw;
struct i40e_vf *vf = pf->vf;
- int abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
int i;
- for (i = 0; i < pf->num_alloc_vfs; i++) {
+ for (i = 0; i < pf->num_alloc_vfs; i++, vf++) {
+ int abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
+ /* Not all vfs are enabled so skip the ones that are not */
+ if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states) &&
+ !test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states))
+ continue;
+
/* Ignore return value on purpose - a given VF may fail, but
* we need to keep going and send to all of them
*/
i40e_aq_send_msg_to_vf(hw, abs_vf_id, v_opcode, v_retval,
msg, msglen, NULL);
- vf++;
- abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
}
}
struct i40e_hw *hw = &pf->hw;
struct i40e_vf *vf = pf->vf;
struct i40e_link_status *ls = &pf->hw.phy.link_info;
- int abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
int i;
pfe.event = I40E_VIRTCHNL_EVENT_LINK_CHANGE;
pfe.severity = I40E_PF_EVENT_SEVERITY_INFO;
- for (i = 0; i < pf->num_alloc_vfs; i++) {
+ for (i = 0; i < pf->num_alloc_vfs; i++, vf++) {
+ int abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
if (vf->link_forced) {
pfe.event_data.link_event.link_status = vf->link_up;
pfe.event_data.link_event.link_speed =
i40e_aq_send_msg_to_vf(hw, abs_vf_id, I40E_VIRTCHNL_OP_EVENT,
0, (u8 *)&pfe, sizeof(pfe),
NULL);
- vf++;
- abs_vf_id = vf->vf_id + hw->func_caps.vf_base_id;
}
}
void i40e_vc_notify_vf_reset(struct i40e_vf *vf)
{
struct i40e_virtchnl_pf_event pfe;
- int abs_vf_id = vf->vf_id + vf->pf->hw.func_caps.vf_base_id;
+ int abs_vf_id;
+
+ /* validate the request */
+ if (!vf || vf->vf_id >= vf->pf->num_alloc_vfs)
+ return;
+
+ /* verify if the VF is in either init or active before proceeding */
+ if (!test_bit(I40E_VF_STAT_INIT, &vf->vf_states) &&
+ !test_bit(I40E_VF_STAT_ACTIVE, &vf->vf_states))
+ return;
+
+ abs_vf_id = vf->vf_id + vf->pf->hw.func_caps.vf_base_id;
pfe.event = I40E_VIRTCHNL_EVENT_RESET_IMPENDING;
pfe.severity = I40E_PF_EVENT_SEVERITY_CERTAIN_DOOM;
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static DEFINE_PCI_DEVICE_TABLE(i40evf_pci_tbl) = {
+static const struct pci_device_id i40evf_pci_tbl[] = {
{PCI_VDEVICE(INTEL, I40E_DEV_ID_VF), 0},
/* required last entry */
{0, }
.resume = igbvf_io_resume,
};
-static DEFINE_PCI_DEVICE_TABLE(igbvf_pci_tbl) = {
+static const struct pci_device_id igbvf_pci_tbl[] = {
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82576_VF), board_vf },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_I350_VF), board_i350_vf },
{ } /* terminate list */
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static DEFINE_PCI_DEVICE_TABLE(ixgb_pci_tbl) = {
+static const struct pci_device_id ixgb_pci_tbl[] = {
{PCI_VENDOR_ID_INTEL, IXGB_DEVICE_ID_82597EX,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, IXGB_DEVICE_ID_82597EX_CX4,
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static DEFINE_PCI_DEVICE_TABLE(ixgbe_pci_tbl) = {
+static const struct pci_device_id ixgbe_pci_tbl[] = {
{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82598), board_82598 },
{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82598AF_DUAL_PORT), board_82598 },
{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82598AF_SINGLE_PORT), board_82598 },
* { Vendor ID, Device ID, SubVendor ID, SubDevice ID,
* Class, Class Mask, private data (not used) }
*/
-static DEFINE_PCI_DEVICE_TABLE(ixgbevf_pci_tbl) = {
+static const struct pci_device_id ixgbevf_pci_tbl[] = {
{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_82599_VF), board_82599_vf },
{PCI_VDEVICE(INTEL, IXGBE_DEV_ID_X540_VF), board_X540_vf },
/* required last entry */
#define JME_PM_OPS NULL
#endif
-static DEFINE_PCI_DEVICE_TABLE(jme_pci_tbl) = {
+static const struct pci_device_id jme_pci_tbl[] = {
{ PCI_VDEVICE(JMICRON, PCI_DEVICE_ID_JMICRON_JMC250) },
{ PCI_VDEVICE(JMICRON, PCI_DEVICE_ID_JMICRON_JMC260) },
{ }
module_param(debug, int, 0);
MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
-static DEFINE_PCI_DEVICE_TABLE(skge_id_table) = {
+static const struct pci_device_id skge_id_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_3COM, 0x1700) }, /* 3Com 3C940 */
{ PCI_DEVICE(PCI_VENDOR_ID_3COM, 0x80EB) }, /* 3Com 3C940B */
#ifdef CONFIG_SKGE_GENESIS
module_param(legacy_pme, int, 0);
MODULE_PARM_DESC(legacy_pme, "Legacy power management");
-static DEFINE_PCI_DEVICE_TABLE(sky2_id_table) = {
+static const struct pci_device_id sky2_id_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, 0x9000) }, /* SK-9Sxx */
{ PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, 0x9E00) }, /* SK-9Exx */
{ PCI_DEVICE(PCI_VENDOR_ID_SYSKONNECT, 0x9E01) }, /* SK-9E21M */
.verify = NULL,
.wrapper = mlx4_MAD_IFC_wrapper
},
+ {
+ .opcode = MLX4_CMD_MAD_DEMUX,
+ .has_inbox = false,
+ .has_outbox = false,
+ .out_is_imm = false,
+ .encode_slave_id = false,
+ .verify = NULL,
+ .wrapper = mlx4_CMD_EPERM_wrapper
+ },
{
.opcode = MLX4_CMD_QUERY_IF_STAT,
.has_inbox = false,
[7] = "FSM (MAC anti-spoofing) support",
[8] = "Dynamic QP updates support",
[9] = "Device managed flow steering IPoIB support",
- [10] = "TCP/IP offloads/flow-steering for VXLAN support"
+ [10] = "TCP/IP offloads/flow-steering for VXLAN support",
+ [11] = "MAD DEMUX (Secure-Host) support"
};
int i;
#define QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET 0xa0
#define QUERY_DEV_CAP_FW_REASSIGN_MAC 0x9d
#define QUERY_DEV_CAP_VXLAN 0x9e
+#define QUERY_DEV_CAP_MAD_DEMUX_OFFSET 0xb0
dev_cap->flags2 = 0;
mailbox = mlx4_alloc_cmd_mailbox(dev);
MLX4_GET(dev_cap->max_counters, outbox,
QUERY_DEV_CAP_MAX_COUNTERS_OFFSET);
+ MLX4_GET(field32, outbox,
+ QUERY_DEV_CAP_MAD_DEMUX_OFFSET);
+ if (field32 & (1 << 0))
+ dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_MAD_DEMUX;
+
MLX4_GET(field32, outbox, QUERY_DEV_CAP_EXT_2_FLAGS_OFFSET);
if (field32 & (1 << 16))
dev_cap->flags2 |= MLX4_DEV_CAP_FLAG2_UPDATE_QP;
out:
mlx4_free_cmd_mailbox(dev, mailbox);
}
+
+static int mlx4_check_smp_firewall_active(struct mlx4_dev *dev,
+ struct mlx4_cmd_mailbox *mailbox)
+{
+#define MLX4_CMD_MAD_DEMUX_SET_ATTR_OFFSET 0x10
+#define MLX4_CMD_MAD_DEMUX_GETRESP_ATTR_OFFSET 0x20
+#define MLX4_CMD_MAD_DEMUX_TRAP_ATTR_OFFSET 0x40
+#define MLX4_CMD_MAD_DEMUX_TRAP_REPRESS_ATTR_OFFSET 0x70
+
+ u32 set_attr_mask, getresp_attr_mask;
+ u32 trap_attr_mask, traprepress_attr_mask;
+
+ MLX4_GET(set_attr_mask, mailbox->buf,
+ MLX4_CMD_MAD_DEMUX_SET_ATTR_OFFSET);
+ mlx4_dbg(dev, "SMP firewall set_attribute_mask = 0x%x\n",
+ set_attr_mask);
+
+ MLX4_GET(getresp_attr_mask, mailbox->buf,
+ MLX4_CMD_MAD_DEMUX_GETRESP_ATTR_OFFSET);
+ mlx4_dbg(dev, "SMP firewall getresp_attribute_mask = 0x%x\n",
+ getresp_attr_mask);
+
+ MLX4_GET(trap_attr_mask, mailbox->buf,
+ MLX4_CMD_MAD_DEMUX_TRAP_ATTR_OFFSET);
+ mlx4_dbg(dev, "SMP firewall trap_attribute_mask = 0x%x\n",
+ trap_attr_mask);
+
+ MLX4_GET(traprepress_attr_mask, mailbox->buf,
+ MLX4_CMD_MAD_DEMUX_TRAP_REPRESS_ATTR_OFFSET);
+ mlx4_dbg(dev, "SMP firewall traprepress_attribute_mask = 0x%x\n",
+ traprepress_attr_mask);
+
+ if (set_attr_mask && getresp_attr_mask && trap_attr_mask &&
+ traprepress_attr_mask)
+ return 1;
+
+ return 0;
+}
+
+int mlx4_config_mad_demux(struct mlx4_dev *dev)
+{
+ struct mlx4_cmd_mailbox *mailbox;
+ int secure_host_active;
+ int err;
+
+ /* Check if mad_demux is supported */
+ if (!(dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_MAD_DEMUX))
+ return 0;
+
+ mailbox = mlx4_alloc_cmd_mailbox(dev);
+ if (IS_ERR(mailbox)) {
+ mlx4_warn(dev, "Failed to allocate mailbox for cmd MAD_DEMUX");
+ return -ENOMEM;
+ }
+
+ /* Query mad_demux to find out which MADs are handled by internal sma */
+ err = mlx4_cmd_box(dev, 0, mailbox->dma, 0x01 /* subn mgmt class */,
+ MLX4_CMD_MAD_DEMUX_QUERY_RESTR, MLX4_CMD_MAD_DEMUX,
+ MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE);
+ if (err) {
+ mlx4_warn(dev, "MLX4_CMD_MAD_DEMUX: query restrictions failed (%d)\n",
+ err);
+ goto out;
+ }
+
+ secure_host_active = mlx4_check_smp_firewall_active(dev, mailbox);
+
+ /* Config mad_demux to handle all MADs returned by the query above */
+ err = mlx4_cmd(dev, mailbox->dma, 0x01 /* subn mgmt class */,
+ MLX4_CMD_MAD_DEMUX_CONFIG, MLX4_CMD_MAD_DEMUX,
+ MLX4_CMD_TIME_CLASS_B, MLX4_CMD_NATIVE);
+ if (err) {
+ mlx4_warn(dev, "MLX4_CMD_MAD_DEMUX: configure failed (%d)\n", err);
+ goto out;
+ }
+
+ if (secure_host_active)
+ mlx4_warn(dev, "HCA operating in secure-host mode. SMP firewall activated.\n");
+out:
+ mlx4_free_cmd_mailbox(dev, mailbox);
+ return err;
+}
mlx4_err(dev, "Failed to initialize multicast group table, aborting\n");
goto err_mr_table_free;
}
+ err = mlx4_config_mad_demux(dev);
+ if (err) {
+ mlx4_err(dev, "Failed in config_mad_demux, aborting\n");
+ goto err_mcg_table_free;
+ }
}
err = mlx4_init_eq_table(dev);
return __mlx4_init_one(pdev, pci_dev_data);
}
-static DEFINE_PCI_DEVICE_TABLE(mlx4_pci_table) = {
+static const struct pci_device_id mlx4_pci_table[] = {
/* MT25408 "Hermon" SDR */
{ PCI_VDEVICE(MELLANOX, 0x6340), MLX4_PCI_DEV_FORCE_SENSE_PORT },
/* MT25408 "Hermon" DDR */
#define MLX4_MPT_FLAG_PHYSICAL (1 << 9)
#define MLX4_MPT_FLAG_REGION (1 << 8)
+#define MLX4_MPT_PD_MASK (0x1FFFFUL)
+#define MLX4_MPT_PD_VF_MASK (0xFE0000UL)
#define MLX4_MPT_PD_FLAG_FAST_REG (1 << 27)
#define MLX4_MPT_PD_FLAG_RAE (1 << 28)
#define MLX4_MPT_PD_FLAG_EN_INV (3 << 24)
int mlx4_get_slave_num_gids(struct mlx4_dev *dev, int slave, int port);
/* Returns the VF index of slave */
int mlx4_get_vf_indx(struct mlx4_dev *dev, int slave);
+int mlx4_config_mad_demux(struct mlx4_dev *dev);
#endif /* MLX4_H */
MLX4_CMD_TIME_CLASS_B, MLX4_CMD_WRAPPED);
}
+int mlx4_mr_hw_get_mpt(struct mlx4_dev *dev, struct mlx4_mr *mmr,
+ struct mlx4_mpt_entry ***mpt_entry)
+{
+ int err;
+ int key = key_to_hw_index(mmr->key) & (dev->caps.num_mpts - 1);
+ struct mlx4_cmd_mailbox *mailbox = NULL;
+
+ /* Make sure that at this point we have single-threaded access only */
+
+ if (mmr->enabled != MLX4_MPT_EN_HW)
+ return -EINVAL;
+
+ err = mlx4_HW2SW_MPT(dev, NULL, key);
+
+ if (err) {
+ mlx4_warn(dev, "HW2SW_MPT failed (%d).", err);
+ mlx4_warn(dev, "Most likely the MR has MWs bound to it.\n");
+ return err;
+ }
+
+ mmr->enabled = MLX4_MPT_EN_SW;
+
+ if (!mlx4_is_mfunc(dev)) {
+ **mpt_entry = mlx4_table_find(
+ &mlx4_priv(dev)->mr_table.dmpt_table,
+ key, NULL);
+ } else {
+ mailbox = mlx4_alloc_cmd_mailbox(dev);
+ if (IS_ERR_OR_NULL(mailbox))
+ return PTR_ERR(mailbox);
+
+ err = mlx4_cmd_box(dev, 0, mailbox->dma, key,
+ 0, MLX4_CMD_QUERY_MPT,
+ MLX4_CMD_TIME_CLASS_B,
+ MLX4_CMD_WRAPPED);
+
+ if (err)
+ goto free_mailbox;
+
+ *mpt_entry = (struct mlx4_mpt_entry **)&mailbox->buf;
+ }
+
+ if (!(*mpt_entry) || !(**mpt_entry)) {
+ err = -ENOMEM;
+ goto free_mailbox;
+ }
+
+ return 0;
+
+free_mailbox:
+ mlx4_free_cmd_mailbox(dev, mailbox);
+ return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_mr_hw_get_mpt);
+
+int mlx4_mr_hw_write_mpt(struct mlx4_dev *dev, struct mlx4_mr *mmr,
+ struct mlx4_mpt_entry **mpt_entry)
+{
+ int err;
+
+ if (!mlx4_is_mfunc(dev)) {
+ /* Make sure any changes to this entry are flushed */
+ wmb();
+
+ *(u8 *)(*mpt_entry) = MLX4_MPT_STATUS_HW;
+
+ /* Make sure the new status is written */
+ wmb();
+
+ err = mlx4_SYNC_TPT(dev);
+ } else {
+ int key = key_to_hw_index(mmr->key) & (dev->caps.num_mpts - 1);
+
+ struct mlx4_cmd_mailbox *mailbox =
+ container_of((void *)mpt_entry, struct mlx4_cmd_mailbox,
+ buf);
+
+ err = mlx4_SW2HW_MPT(dev, mailbox, key);
+ }
+
+ mmr->pd = be32_to_cpu((*mpt_entry)->pd_flags) & MLX4_MPT_PD_MASK;
+ if (!err)
+ mmr->enabled = MLX4_MPT_EN_HW;
+ return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_mr_hw_write_mpt);
+
+void mlx4_mr_hw_put_mpt(struct mlx4_dev *dev,
+ struct mlx4_mpt_entry **mpt_entry)
+{
+ if (mlx4_is_mfunc(dev)) {
+ struct mlx4_cmd_mailbox *mailbox =
+ container_of((void *)mpt_entry, struct mlx4_cmd_mailbox,
+ buf);
+ mlx4_free_cmd_mailbox(dev, mailbox);
+ }
+}
+EXPORT_SYMBOL_GPL(mlx4_mr_hw_put_mpt);
+
+int mlx4_mr_hw_change_pd(struct mlx4_dev *dev, struct mlx4_mpt_entry *mpt_entry,
+ u32 pdn)
+{
+ u32 pd_flags = be32_to_cpu(mpt_entry->pd_flags);
+ /* The wrapper function will put the slave's id here */
+ if (mlx4_is_mfunc(dev))
+ pd_flags &= ~MLX4_MPT_PD_VF_MASK;
+ mpt_entry->pd_flags = cpu_to_be32((pd_flags & ~MLX4_MPT_PD_MASK) |
+ (pdn & MLX4_MPT_PD_MASK)
+ | MLX4_MPT_PD_FLAG_EN_INV);
+ return 0;
+}
+EXPORT_SYMBOL_GPL(mlx4_mr_hw_change_pd);
+
+int mlx4_mr_hw_change_access(struct mlx4_dev *dev,
+ struct mlx4_mpt_entry *mpt_entry,
+ u32 access)
+{
+ u32 flags = (be32_to_cpu(mpt_entry->flags) & ~MLX4_PERM_MASK) |
+ (access & MLX4_PERM_MASK);
+
+ mpt_entry->flags = cpu_to_be32(flags);
+ return 0;
+}
+EXPORT_SYMBOL_GPL(mlx4_mr_hw_change_access);
+
static int mlx4_mr_alloc_reserved(struct mlx4_dev *dev, u32 mridx, u32 pd,
u64 iova, u64 size, u32 access, int npages,
int page_shift, struct mlx4_mr *mr)
}
EXPORT_SYMBOL_GPL(mlx4_mr_free);
+void mlx4_mr_rereg_mem_cleanup(struct mlx4_dev *dev, struct mlx4_mr *mr)
+{
+ mlx4_mtt_cleanup(dev, &mr->mtt);
+}
+EXPORT_SYMBOL_GPL(mlx4_mr_rereg_mem_cleanup);
+
+int mlx4_mr_rereg_mem_write(struct mlx4_dev *dev, struct mlx4_mr *mr,
+ u64 iova, u64 size, int npages,
+ int page_shift, struct mlx4_mpt_entry *mpt_entry)
+{
+ int err;
+
+ mpt_entry->start = cpu_to_be64(mr->iova);
+ mpt_entry->length = cpu_to_be64(mr->size);
+ mpt_entry->entity_size = cpu_to_be32(mr->mtt.page_shift);
+
+ err = mlx4_mtt_init(dev, npages, page_shift, &mr->mtt);
+ if (err)
+ return err;
+
+ if (mr->mtt.order < 0) {
+ mpt_entry->flags |= cpu_to_be32(MLX4_MPT_FLAG_PHYSICAL);
+ mpt_entry->mtt_addr = 0;
+ } else {
+ mpt_entry->mtt_addr = cpu_to_be64(mlx4_mtt_addr(dev,
+ &mr->mtt));
+ if (mr->mtt.page_shift == 0)
+ mpt_entry->mtt_sz = cpu_to_be32(1 << mr->mtt.order);
+ }
+ mr->enabled = MLX4_MPT_EN_SW;
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(mlx4_mr_rereg_mem_write);
+
int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr)
{
struct mlx4_cmd_mailbox *mailbox;
if (err)
return err;
- if (mpt->com.from_state != RES_MPT_HW) {
+ if (mpt->com.from_state == RES_MPT_MAPPED) {
+ /* In order to allow rereg in SRIOV, we need to alter the MPT entry. To do
+ * that, the VF must read the MPT. But since the MPT entry memory is not
+ * in the VF's virtual memory space, it must use QUERY_MPT to obtain the
+ * entry contents. To guarantee that the MPT cannot be changed, the driver
+ * must perform HW2SW_MPT before this query and return the MPT entry to HW
+ * ownership fofollowing the change. The change here allows the VF to
+ * perform QUERY_MPT also when the entry is in SW ownership.
+ */
+ struct mlx4_mpt_entry *mpt_entry = mlx4_table_find(
+ &mlx4_priv(dev)->mr_table.dmpt_table,
+ mpt->key, NULL);
+
+ if (NULL == mpt_entry || NULL == outbox->buf) {
+ err = -EINVAL;
+ goto out;
+ }
+
+ memcpy(outbox->buf, mpt_entry, sizeof(*mpt_entry));
+
+ err = 0;
+ } else if (mpt->com.from_state == RES_MPT_HW) {
+ err = mlx4_DMA_wrapper(dev, slave, vhcr, inbox, outbox, cmd);
+ } else {
err = -EBUSY;
goto out;
}
- err = mlx4_DMA_wrapper(dev, slave, vhcr, inbox, outbox, cmd);
out:
put_res(dev, slave, id, RES_MPT);
static char pcidev_name[] = "ksz884xp";
-static DEFINE_PCI_DEVICE_TABLE(pcidev_table) = {
+static const struct pci_device_id pcidev_table[] = {
{ PCI_VENDOR_ID_MICREL_KS, 0x8841,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ PCI_VENDOR_ID_MICREL_KS, 0x8842,
return -ENOMEM;
dmatest_bus = pci_map_page(mgp->pdev, dmatest_page, 0, PAGE_SIZE,
DMA_BIDIRECTIONAL);
+ if (unlikely(pci_dma_mapping_error(mgp->pdev, dmatest_bus))) {
+ __free_page(dmatest_page);
+ return -ENOMEM;
+ }
/* Run a small DMA test.
* The magic multipliers to the length tell the firmware
int bytes, int watchdog)
{
struct page *page;
+ dma_addr_t bus;
int idx;
#if MYRI10GE_ALLOC_SIZE > 4096
int end_offset;
rx->watchdog_needed = 1;
return;
}
+
+ bus = pci_map_page(mgp->pdev, page, 0,
+ MYRI10GE_ALLOC_SIZE,
+ PCI_DMA_FROMDEVICE);
+ if (unlikely(pci_dma_mapping_error(mgp->pdev, bus))) {
+ __free_pages(page, MYRI10GE_ALLOC_ORDER);
+ if (rx->fill_cnt - rx->cnt < 16)
+ rx->watchdog_needed = 1;
+ return;
+ }
+
rx->page = page;
rx->page_offset = 0;
- rx->bus = pci_map_page(mgp->pdev, page, 0,
- MYRI10GE_ALLOC_SIZE,
- PCI_DMA_FROMDEVICE);
+ rx->bus = bus;
+
}
rx->info[idx].page = rx->page;
rx->info[idx].page_offset = rx->page_offset;
mb();
}
+static void myri10ge_unmap_tx_dma(struct myri10ge_priv *mgp,
+ struct myri10ge_tx_buf *tx, int idx)
+{
+ unsigned int len;
+ int last_idx;
+
+ /* Free any DMA resources we've alloced and clear out the skb slot */
+ last_idx = (idx + 1) & tx->mask;
+ idx = tx->req & tx->mask;
+ do {
+ len = dma_unmap_len(&tx->info[idx], len);
+ if (len) {
+ if (tx->info[idx].skb != NULL)
+ pci_unmap_single(mgp->pdev,
+ dma_unmap_addr(&tx->info[idx],
+ bus), len,
+ PCI_DMA_TODEVICE);
+ else
+ pci_unmap_page(mgp->pdev,
+ dma_unmap_addr(&tx->info[idx],
+ bus), len,
+ PCI_DMA_TODEVICE);
+ dma_unmap_len_set(&tx->info[idx], len, 0);
+ tx->info[idx].skb = NULL;
+ }
+ idx = (idx + 1) & tx->mask;
+ } while (idx != last_idx);
+}
+
/*
* Transmit a packet. We need to split the packet so that a single
* segment does not cross myri10ge->tx_boundary, so this makes segment
u32 low;
__be32 high_swapped;
unsigned int len;
- int idx, last_idx, avail, frag_cnt, frag_idx, count, mss, max_segments;
+ int idx, avail, frag_cnt, frag_idx, count, mss, max_segments;
u16 pseudo_hdr_offset, cksum_offset, queue;
int cum_len, seglen, boundary, rdma_count;
u8 flags, odd_flag;
/* map the skb for DMA */
len = skb_headlen(skb);
+ bus = pci_map_single(mgp->pdev, skb->data, len, PCI_DMA_TODEVICE);
+ if (unlikely(pci_dma_mapping_error(mgp->pdev, bus)))
+ goto drop;
+
idx = tx->req & tx->mask;
tx->info[idx].skb = skb;
- bus = pci_map_single(mgp->pdev, skb->data, len, PCI_DMA_TODEVICE);
dma_unmap_addr_set(&tx->info[idx], bus, bus);
dma_unmap_len_set(&tx->info[idx], len, len);
break;
/* map next fragment for DMA */
- idx = (count + tx->req) & tx->mask;
frag = &skb_shinfo(skb)->frags[frag_idx];
frag_idx++;
len = skb_frag_size(frag);
bus = skb_frag_dma_map(&mgp->pdev->dev, frag, 0, len,
DMA_TO_DEVICE);
+ if (unlikely(pci_dma_mapping_error(mgp->pdev, bus))) {
+ myri10ge_unmap_tx_dma(mgp, tx, idx);
+ goto drop;
+ }
+ idx = (count + tx->req) & tx->mask;
dma_unmap_addr_set(&tx->info[idx], bus, bus);
dma_unmap_len_set(&tx->info[idx], len, len);
}
return NETDEV_TX_OK;
abort_linearize:
- /* Free any DMA resources we've alloced and clear out the skb
- * slot so as to not trip up assertions, and to avoid a
- * double-free if linearizing fails */
+ myri10ge_unmap_tx_dma(mgp, tx, idx);
- last_idx = (idx + 1) & tx->mask;
- idx = tx->req & tx->mask;
- tx->info[idx].skb = NULL;
- do {
- len = dma_unmap_len(&tx->info[idx], len);
- if (len) {
- if (tx->info[idx].skb != NULL)
- pci_unmap_single(mgp->pdev,
- dma_unmap_addr(&tx->info[idx],
- bus), len,
- PCI_DMA_TODEVICE);
- else
- pci_unmap_page(mgp->pdev,
- dma_unmap_addr(&tx->info[idx],
- bus), len,
- PCI_DMA_TODEVICE);
- dma_unmap_len_set(&tx->info[idx], len, 0);
- tx->info[idx].skb = NULL;
- }
- idx = (idx + 1) & tx->mask;
- } while (idx != last_idx);
if (skb_is_gso(skb)) {
netdev_err(mgp->dev, "TSO but wanted to linearize?!?!?\n");
goto drop;
#define PCI_DEVICE_ID_MYRICOM_MYRI10GE_Z8E 0x0008
#define PCI_DEVICE_ID_MYRICOM_MYRI10GE_Z8E_9 0x0009
-static DEFINE_PCI_DEVICE_TABLE(myri10ge_pci_tbl) = {
+static const struct pci_device_id myri10ge_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_MYRICOM, PCI_DEVICE_ID_MYRICOM_MYRI10GE_Z8E)},
{PCI_DEVICE
(PCI_VENDOR_ID_MYRICOM, PCI_DEVICE_ID_MYRICOM_MYRI10GE_Z8E_9)},
{ "NatSemi DP8381[56]", 0, 24 },
};
-static DEFINE_PCI_DEVICE_TABLE(natsemi_pci_tbl) = {
+static const struct pci_device_id natsemi_pci_tbl[] = {
{ PCI_VENDOR_ID_NS, 0x0020, 0x12d9, 0x000c, 0, 0, 0 },
{ PCI_VENDOR_ID_NS, 0x0020, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1 },
{ } /* terminate list */
free_netdev(ndev);
}
-static DEFINE_PCI_DEVICE_TABLE(ns83820_pci_tbl) = {
+static const struct pci_device_id ns83820_pci_tbl[] = {
{ 0x100b, 0x0022, PCI_ANY_ID, PCI_ANY_ID, 0, .driver_data = 0, },
{ 0, },
};
* S2IO device table.
* This table lists all the devices that this driver supports.
*/
-static DEFINE_PCI_DEVICE_TABLE(s2io_tbl) = {
+static const struct pci_device_id s2io_tbl[] = {
{PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_S2IO_WIN,
PCI_ANY_ID, PCI_ANY_ID},
{PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_S2IO_UNI,
MODULE_DESCRIPTION("Neterion's X3100 Series 10GbE PCIe I/O"
"Virtualized Server Adapter");
-static DEFINE_PCI_DEVICE_TABLE(vxge_id_table) = {
+static const struct pci_device_id vxge_id_table[] = {
{PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_TITAN_WIN, PCI_ANY_ID,
PCI_ANY_ID},
{PCI_VENDOR_ID_S2IO, PCI_DEVICE_ID_TITAN_UNI, PCI_ANY_ID,
#define nv_shutdown NULL
#endif /* CONFIG_PM */
-static DEFINE_PCI_DEVICE_TABLE(pci_tbl) = {
+static const struct pci_device_id pci_tbl[] = {
{ /* nForce Ethernet Controller */
PCI_DEVICE(0x10DE, 0x01C3),
.driver_data = DEV_NEED_TIMERIRQ|DEV_NEED_LINKTIMER,
.platform_init = pch_gbe_minnow_platform_init,
};
-static DEFINE_PCI_DEVICE_TABLE(pch_gbe_pcidev_id) = {
+static const struct pci_device_id pch_gbe_pcidev_id[] = {
{.vendor = PCI_VENDOR_ID_INTEL,
.device = PCI_DEVICE_ID_INTEL_IOH1_GBE,
.subvendor = PCI_VENDOR_ID_CIRCUITCO,
}
}
-static DEFINE_PCI_DEVICE_TABLE(hamachi_pci_tbl) = {
+static const struct pci_device_id hamachi_pci_tbl[] = {
{ 0x1318, 0x0911, PCI_ANY_ID, PCI_ANY_ID, },
{ 0, }
};
{ }
};
-static DEFINE_PCI_DEVICE_TABLE(yellowfin_pci_tbl) = {
+static const struct pci_device_id yellowfin_pci_tbl[] = {
{ 0x1000, 0x0702, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ 0x1000, 0x0701, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 1 },
{ }
free_netdev(netdev);
}
-static DEFINE_PCI_DEVICE_TABLE(pasemi_mac_pci_tbl) = {
+static const struct pci_device_id pasemi_mac_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_PASEMI, 0xa005) },
{ PCI_DEVICE(PCI_VENDOR_ID_PASEMI, 0xa006) },
{ },
{PCI_DEVICE(PCI_VENDOR_ID_NETXEN, (device)), \
.class = PCI_CLASS_NETWORK_ETHERNET << 8, .class_mask = ~0}
-static DEFINE_PCI_DEVICE_TABLE(netxen_pci_tbl) = {
+static const struct pci_device_id netxen_pci_tbl[] = {
ENTRY(PCI_DEVICE_ID_NX2031_10GXSR),
ENTRY(PCI_DEVICE_ID_NX2031_10GCX4),
ENTRY(PCI_DEVICE_ID_NX2031_4GCU),
module_param(msi, int, 0);
MODULE_PARM_DESC(msi, "Turn on Message Signaled Interrupts.");
-static DEFINE_PCI_DEVICE_TABLE(ql3xxx_pci_tbl) = {
+static const struct pci_device_id ql3xxx_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, QL3022_DEVICE_ID)},
{PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, QL3032_DEVICE_ID)},
/* required last entry */
u16 cksum;
u16 unused;
u8 model[16];
- u16 mfg_id;
+ u8 mfg_id;
u16 id;
u8 flag;
u8 erase_cmd;
return QLC_DEFAULT_VNIC_COUNT;
}
+static inline void qlcnic_swap32_buffer(u32 *buffer, int count)
+{
+#if defined(__BIG_ENDIAN)
+ u32 *tmp = buffer;
+ int i;
+
+ for (i = 0; i < count; i++) {
+ *tmp = swab32(*tmp);
+ tmp++;
+ }
+#endif
+}
+
#ifdef CONFIG_QLCNIC_HWMON
void qlcnic_register_hwmon_dev(struct qlcnic_adapter *);
void qlcnic_unregister_hwmon_dev(struct qlcnic_adapter *);
}
qlcnic_83xx_wrt_reg_indirect(adapter, QLC_83XX_FLASH_DIRECT_WINDOW,
- (addr));
+ (addr & 0xFFFF0000));
range = flash_offset + (count * sizeof(u32));
/* Check if data is spread across multiple sectors */
ret = qlcnic_83xx_lockless_flash_read32(adapter, QLCNIC_FDT_LOCATION,
(u8 *)&adapter->ahw->fdt,
count);
-
+ qlcnic_swap32_buffer((u32 *)&adapter->ahw->fdt, count);
qlcnic_83xx_unlock_flash(adapter);
return ret;
}
addr1 = (sector_start_addr & 0xFF) << 16;
addr2 = (sector_start_addr & 0xFF0000) >> 16;
- reversed_addr = addr1 | addr2;
+ reversed_addr = addr1 | addr2 | (sector_start_addr & 0xFF00);
qlcnic_83xx_wrt_reg_indirect(adapter, QLC_83XX_FLASH_WRDATA,
reversed_addr);
{
struct qlc_83xx_fw_info *fw_info = adapter->ahw->fw_info;
const struct firmware *fw = fw_info->fw;
- u32 dest, *p_cache;
+ u32 dest, *p_cache, *temp;
int i, ret = -EIO;
+ __le32 *temp_le;
u8 data[16];
size_t size;
u64 addr;
+ temp = kzalloc(fw->size, GFP_KERNEL);
+ if (!temp) {
+ release_firmware(fw);
+ fw_info->fw = NULL;
+ return -ENOMEM;
+ }
+
+ temp_le = (__le32 *)fw->data;
+
+ /* FW image in file is in little endian, swap the data to nullify
+ * the effect of writel() operation on big endian platform.
+ */
+ for (i = 0; i < fw->size / sizeof(u32); i++)
+ temp[i] = __le32_to_cpu(temp_le[i]);
+
dest = QLCRDX(adapter->ahw, QLCNIC_FW_IMAGE_ADDR);
size = (fw->size & ~0xF);
- p_cache = (u32 *)fw->data;
+ p_cache = temp;
addr = (u64)dest;
ret = qlcnic_ms_mem_write128(adapter, addr,
p_cache, size / 16);
if (ret) {
dev_err(&adapter->pdev->dev, "MS memory write failed\n");
- release_firmware(fw);
- fw_info->fw = NULL;
- return -EIO;
+ goto exit;
}
/* alignment check */
if (fw->size & 0xF) {
addr = dest + size;
for (i = 0; i < (fw->size & 0xF); i++)
- data[i] = fw->data[size + i];
+ data[i] = temp[size + i];
for (; i < 16; i++)
data[i] = 0;
ret = qlcnic_ms_mem_write128(adapter, addr,
if (ret) {
dev_err(&adapter->pdev->dev,
"MS memory write failed\n");
- release_firmware(fw);
- fw_info->fw = NULL;
- return -EIO;
+ goto exit;
}
}
+
+exit:
release_firmware(fw);
fw_info->fw = NULL;
+ kfree(temp);
- return 0;
+ return ret;
}
static void qlcnic_83xx_dump_pause_control_regs(struct qlcnic_adapter *adapter)
{PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, (device)), \
.class = PCI_CLASS_NETWORK_ETHERNET << 8, .class_mask = ~0}
-static DEFINE_PCI_DEVICE_TABLE(qlcnic_pci_tbl) = {
+static const struct pci_device_id qlcnic_pci_tbl[] = {
ENTRY(PCI_DEVICE_ID_QLOGIC_QLE824X),
ENTRY(PCI_DEVICE_ID_QLOGIC_QLE834X),
ENTRY(PCI_DEVICE_ID_QLOGIC_VF_QLE834X),
u32 type;
u32 offset;
u32 cap_size;
+#if defined(__LITTLE_ENDIAN)
u8 mask;
u8 rsvd[2];
u8 flags;
+#else
+ u8 flags;
+ u8 rsvd[2];
+ u8 mask;
+#endif
} __packed;
struct __crb {
u32 addr;
+#if defined(__LITTLE_ENDIAN)
u8 stride;
u8 rsvd1[3];
+#else
+ u8 rsvd1[3];
+ u8 stride;
+#endif
u32 data_size;
u32 no_ops;
u32 rsvd2[4];
struct __ctrl {
u32 addr;
+#if defined(__LITTLE_ENDIAN)
u8 stride;
u8 index_a;
u16 timeout;
+#else
+ u16 timeout;
+ u8 index_a;
+ u8 stride;
+#endif
u32 data_size;
u32 no_ops;
+#if defined(__LITTLE_ENDIAN)
u8 opcode;
u8 index_v;
u8 shl_val;
u8 shr_val;
+#else
+ u8 shr_val;
+ u8 shl_val;
+ u8 index_v;
+ u8 opcode;
+#endif
u32 val1;
u32 val2;
u32 val3;
struct __cache {
u32 addr;
+#if defined(__LITTLE_ENDIAN)
u16 stride;
u16 init_tag_val;
+#else
+ u16 init_tag_val;
+ u16 stride;
+#endif
u32 size;
u32 no_ops;
u32 ctrl_addr;
u32 ctrl_val;
u32 read_addr;
+#if defined(__LITTLE_ENDIAN)
u8 read_addr_stride;
u8 read_addr_num;
u8 rsvd1[2];
+#else
+ u8 rsvd1[2];
+ u8 read_addr_num;
+ u8 read_addr_stride;
+#endif
} __packed;
struct __ocm {
struct __queue {
u32 sel_addr;
+#if defined(__LITTLE_ENDIAN)
u16 stride;
u8 rsvd[2];
+#else
+ u8 rsvd[2];
+ u16 stride;
+#endif
u32 size;
u32 no_ops;
u8 rsvd2[8];
u32 read_addr;
+#if defined(__LITTLE_ENDIAN)
u8 read_addr_stride;
u8 read_addr_cnt;
u8 rsvd3[2];
+#else
+ u8 rsvd3[2];
+ u8 read_addr_cnt;
+ u8 read_addr_stride;
+#endif
} __packed;
struct __pollrd {
u32 sel_addr;
u32 read_addr;
u32 sel_val;
+#if defined(__LITTLE_ENDIAN)
u16 sel_val_stride;
u16 no_ops;
+#else
+ u16 no_ops;
+ u16 sel_val_stride;
+#endif
u32 poll_wait;
u32 poll_mask;
u32 data_size;
u32 no_ops;
u32 sel_val_mask;
u32 read_addr;
+#if defined(__LITTLE_ENDIAN)
u8 sel_val_stride;
u8 data_size;
u8 rsvd[2];
+#else
+ u8 rsvd[2];
+ u8 data_size;
+ u8 sel_val_stride;
+#endif
} __packed;
struct __pollrdmwr {
if (ret != 0)
return ret;
qlcnic_read_crb(adapter, buf, offset, size);
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
return size;
}
if (ret != 0)
return ret;
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
qlcnic_write_crb(adapter, buf, offset, size);
return size;
}
return -EIO;
memcpy(buf, &data, size);
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
return size;
}
if (ret != 0)
return ret;
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
memcpy(&data, buf, size);
if (qlcnic_pci_mem_write_2M(adapter, offset, data))
if (rem)
return QL_STATUS_INVALID_PARAM;
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
pm_cfg = (struct qlcnic_pm_func_cfg *)buf;
ret = validate_pm_config(adapter, pm_cfg, count);
pm_cfg[pci_func].dest_npar = 0;
pm_cfg[pci_func].pci_func = i;
}
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
return size;
}
if (rem)
return QL_STATUS_INVALID_PARAM;
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
esw_cfg = (struct qlcnic_esw_func_cfg *)buf;
ret = validate_esw_config(adapter, esw_cfg, count);
if (ret)
if (qlcnic_get_eswitch_port_config(adapter, &esw_cfg[pci_func]))
return QL_STATUS_INVALID_PARAM;
}
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
return size;
}
if (rem)
return QL_STATUS_INVALID_PARAM;
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
np_cfg = (struct qlcnic_npar_func_cfg *)buf;
ret = validate_npar_config(adapter, np_cfg, count);
if (ret)
np_cfg[pci_func].max_tx_queues = nic_info.max_tx_ques;
np_cfg[pci_func].max_rx_queues = nic_info.max_rx_ques;
}
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
return size;
}
pci_cfg = (struct qlcnic_pci_func_cfg *)buf;
count = size / sizeof(struct qlcnic_pci_func_cfg);
+ qlcnic_swap32_buffer((u32 *)pci_info, size / sizeof(u32));
for (i = 0; i < count; i++) {
pci_cfg[i].pci_func = pci_info[i].id;
pci_cfg[i].func_type = pci_info[i].type;
}
qlcnic_83xx_unlock_flash(adapter);
+ qlcnic_swap32_buffer((u32 *)p_read_buf, count);
memcpy(buf, p_read_buf, size);
kfree(p_read_buf);
if (!p_cache)
return -ENOMEM;
+ count = size / sizeof(u32);
+ qlcnic_swap32_buffer((u32 *)buf, count);
memcpy(p_cache, buf, size);
p_src = p_cache;
- count = size / sizeof(u32);
if (qlcnic_83xx_lock_flash(adapter) != 0) {
kfree(p_cache);
if (!p_cache)
return -ENOMEM;
+ qlcnic_swap32_buffer((u32 *)buf, size / sizeof(u32));
memcpy(p_cache, buf, size);
p_src = p_cache;
count = size / sizeof(u32);
"Option to allow force of firmware core dump. "
"Default is OFF - Do not allow.");
-static DEFINE_PCI_DEVICE_TABLE(qlge_pci_tbl) = {
+static const struct pci_device_id qlge_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, QLGE_DEVICE_ID_8012)},
{PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, QLGE_DEVICE_ID_8000)},
/* required last entry */
}
-static DEFINE_PCI_DEVICE_TABLE(r6040_pci_tbl) = {
+static const struct pci_device_id r6040_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_RDC, 0x6040) },
{ 0 }
};
};
-static DEFINE_PCI_DEVICE_TABLE(rtl8139_pci_tbl) = {
+static const struct pci_device_id rtl8139_pci_tbl[] = {
{0x10ec, 0x8139, PCI_ANY_ID, PCI_ANY_ID, 0, 0, RTL8139 },
{0x10ec, 0x8138, PCI_ANY_ID, PCI_ANY_ID, 0, 0, RTL8139 },
{0x1113, 0x1211, PCI_ANY_ID, PCI_ANY_ID, 0, 0, RTL8139 },
RTL_CFG_2
};
-static DEFINE_PCI_DEVICE_TABLE(rtl8169_pci_tbl) = {
+static const struct pci_device_id rtl8169_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8129), 0, 0, RTL_CFG_0 },
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8136), 0, 0, RTL_CFG_2 },
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8167), 0, 0, RTL_CFG_0 },
**************************************************************************/
/* PCI device ID table */
-static DEFINE_PCI_DEVICE_TABLE(efx_pci_table) = {
+static const struct pci_device_id efx_pci_table[] = {
{PCI_DEVICE(PCI_VENDOR_ID_SOLARFLARE,
PCI_DEVICE_ID_SOLARFLARE_SFC4000A_0),
.driver_data = (unsigned long) &falcon_a1_nic_type},
*/
}
-static DEFINE_PCI_DEVICE_TABLE(ioc3_pci_tbl) = {
+static const struct pci_device_id ioc3_pci_tbl[] = {
{ PCI_VENDOR_ID_SGI, PCI_DEVICE_ID_SGI_IOC3, PCI_ANY_ID, PCI_ANY_ID },
{ 0 }
};
return 0;
}
-static DEFINE_PCI_DEVICE_TABLE(sc92031_pci_device_id_table) = {
+static const struct pci_device_id sc92031_pci_device_id_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_SILAN, 0x2031) },
{ PCI_DEVICE(PCI_VENDOR_ID_SILAN, 0x8139) },
{ PCI_DEVICE(0x1088, 0x2031) },
{ "SiS 191 PCI Gigabit Ethernet adapter" },
};
-static DEFINE_PCI_DEVICE_TABLE(sis190_pci_tbl) = {
+static const struct pci_device_id sis190_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_SI, 0x0190), 0, 0, 0 },
{ PCI_DEVICE(PCI_VENDOR_ID_SI, 0x0191), 0, 0, 1 },
{ 0, },
"SiS 900 PCI Fast Ethernet",
"SiS 7016 PCI Fast Ethernet"
};
-static DEFINE_PCI_DEVICE_TABLE(sis900_pci_tbl) = {
+
+static const struct pci_device_id sis900_pci_tbl[] = {
{PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_900,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, SIS_900},
{PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_7016,
};
-static DEFINE_PCI_DEVICE_TABLE(epic_pci_tbl) = {
+static const struct pci_device_id epic_pci_tbl[] = {
{ 0x10B8, 0x0005, 0x1092, 0x0AB4, 0, 0, SMSC_83C170_0 },
{ 0x10B8, 0x0005, PCI_ANY_ID, PCI_ANY_ID, 0, 0, SMSC_83C170 },
{ 0x10B8, 0x0006, PCI_ANY_ID, PCI_ANY_ID,
int last_carrier;
};
-static DEFINE_PCI_DEVICE_TABLE(smsc9420_id_table) = {
+static const struct pci_device_id smsc9420_id_table[] = {
{ PCI_VENDOR_ID_9420, PCI_DEVICE_ID_9420, PCI_ANY_ID, PCI_ANY_ID, },
{ 0, }
};
#define STMMAC_VENDOR_ID 0x700
#define STMMAC_DEVICE_ID 0x1108
-static DEFINE_PCI_DEVICE_TABLE(stmmac_id_table) = {
+static const struct pci_device_id stmmac_id_table[] = {
{PCI_DEVICE(STMMAC_VENDOR_ID, STMMAC_DEVICE_ID)},
{PCI_DEVICE(PCI_VENDOR_ID_STMICRO, PCI_DEVICE_ID_STMICRO_MAC)},
{}
CAS_BMCR_SPEED1000|BMCR_FULLDPLX /* 5 : 1000bt full duplex */
};
-static DEFINE_PCI_DEVICE_TABLE(cas_pci_tbl) = {
+static const struct pci_device_id cas_pci_tbl[] = {
{ PCI_VENDOR_ID_SUN, PCI_DEVICE_ID_SUN_CASSINI,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
{ PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_SATURN,
}
#endif
-static DEFINE_PCI_DEVICE_TABLE(niu_pci_tbl) = {
+static const struct pci_device_id niu_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_SUN, 0xabcd)},
{}
};
#define GEM_MODULE_NAME "gem"
-static DEFINE_PCI_DEVICE_TABLE(gem_pci_tbl) = {
+static const struct pci_device_id gem_pci_tbl[] = {
{ PCI_VENDOR_ID_SUN, PCI_DEVICE_ID_SUN_GEM,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
free_netdev(net_dev);
}
-static DEFINE_PCI_DEVICE_TABLE(happymeal_pci_ids) = {
+static const struct pci_device_id happymeal_pci_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_SUN, PCI_DEVICE_ID_SUN_HAPPYMEAL) },
{ } /* Terminating entry */
};
MODULE_LICENSE("GPL");
MODULE_VERSION(DRV_MODULE_VERSION);
+/* Heuristic for the number of times to exponentially backoff and
+ * retry sending an LDC trigger when EAGAIN is encountered
+ */
+#define VNET_MAX_RETRIES 10
+
/* Ordered from largest major to lowest */
static struct vio_version vnet_versions[] = {
{ .major = 1, .minor = 0 },
.state = vio_dring_state,
};
int err, delay;
+ int retries = 0;
hdr.seq = dr->snd_nxt;
delay = 1;
udelay(delay);
if ((delay <<= 1) > 128)
delay = 128;
+ if (retries++ > VNET_MAX_RETRIES) {
+ pr_info("ECONNRESET %x:%x:%x:%x:%x:%x\n",
+ port->raddr[0], port->raddr[1],
+ port->raddr[2], port->raddr[3],
+ port->raddr[4], port->raddr[5]);
+ err = -ECONNRESET;
+ }
} while (err == -EAGAIN);
return err;
return 0;
}
-static void maybe_tx_wakeup(struct vnet *vp)
+static void maybe_tx_wakeup(unsigned long param)
{
+ struct vnet *vp = (struct vnet *)param;
struct net_device *dev = vp->dev;
netif_tx_lock(dev);
break;
}
spin_unlock(&vio->lock);
+ /* Kick off a tasklet to wake the queue. We cannot call
+ * maybe_tx_wakeup directly here because we could deadlock on
+ * netif_tx_lock() with dev_watchdog()
+ */
if (unlikely(tx_wakeup && err != -ECONNRESET))
- maybe_tx_wakeup(port->vp);
+ tasklet_schedule(&port->vp->vnet_tx_wakeup);
+
local_irq_restore(flags);
}
.end_idx = (u32) -1,
};
int err, delay;
+ int retries = 0;
hdr.seq = dr->snd_nxt;
delay = 1;
udelay(delay);
if ((delay <<= 1) > 128)
delay = 128;
+ if (retries++ > VNET_MAX_RETRIES)
+ break;
} while (err == -EAGAIN);
return err;
memset(tx_buf+VNET_PACKET_SKIP+skb->len, 0, len - skb->len);
}
- d->hdr.ack = VIO_ACK_ENABLE;
+ /* We don't rely on the ACKs to free the skb in vnet_start_xmit(),
+ * thus it is safe to not set VIO_ACK_ENABLE for each transmission:
+ * the protocol itself does not require it as long as the peer
+ * sends a VIO_SUBTYPE_ACK for VIO_DRING_STOPPED.
+ *
+ * An ACK for every packet in the ring is expensive as the
+ * sending of LDC messages is slow and affects performance.
+ */
+ d->hdr.ack = VIO_ACK_DISABLE;
d->size = len;
d->ncookies = port->tx_bufs[dr->prod].ncookies;
for (i = 0; i < d->ncookies; i++)
vp = netdev_priv(dev);
spin_lock_init(&vp->lock);
+ tasklet_init(&vp->vnet_tx_wakeup, maybe_tx_wakeup, (unsigned long)vp);
vp->dev = dev;
INIT_LIST_HEAD(&vp->port_list);
vp = list_first_entry(&vnet_list, struct vnet, list);
list_del(&vp->list);
dev = vp->dev;
+ tasklet_kill(&vp->vnet_tx_wakeup);
/* vio_unregister_driver() should have cleaned up port_list */
BUG_ON(!list_empty(&vp->port_list));
unregister_netdev(dev);
#ifndef _SUNVNET_H
#define _SUNVNET_H
+#include <linux/interrupt.h>
+
#define DESC_NCOOKIES(entry_size) \
((entry_size) - sizeof(struct vio_net_desc))
struct list_head list;
u64 local_mac;
+
+ struct tasklet_struct vnet_tx_wakeup;
};
#endif /* _SUNVNET_H */
#include "tehuti.h"
-static DEFINE_PCI_DEVICE_TABLE(bdx_pci_tbl) = {
+static const struct pci_device_id bdx_pci_tbl[] = {
{ PCI_VDEVICE(TEHUTI, 0x3009), },
{ PCI_VDEVICE(TEHUTI, 0x3010), },
{ PCI_VDEVICE(TEHUTI, 0x3014), },
TLAN_ADAPTER_ACTIVITY_LED, 0x83 }, /* EISA card */
};
-static DEFINE_PCI_DEVICE_TABLE(tlan_pci_tbl) = {
+static const struct pci_device_id tlan_pci_tbl[] = {
{ PCI_VENDOR_ID_COMPAQ, PCI_DEVICE_ID_COMPAQ_NETEL10,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ PCI_VENDOR_ID_COMPAQ, PCI_DEVICE_ID_COMPAQ_NETEL100,
char spider_net_driver_name[] = "spidernet";
-static DEFINE_PCI_DEVICE_TABLE(spider_net_pci_tbl) = {
+static const struct pci_device_id spider_net_pci_tbl[] = {
{ PCI_VENDOR_ID_TOSHIBA_2, PCI_DEVICE_ID_TOSHIBA_SPIDER_NET,
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0UL },
{ 0, }
{ "TOSHIBA TC35815/TX4939" },
};
-static DEFINE_PCI_DEVICE_TABLE(tc35815_pci_tbl) = {
+static const struct pci_device_id tc35815_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA_2, PCI_DEVICE_ID_TOSHIBA_TC35815CF), .driver_data = TC35815CF },
{PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA_2, PCI_DEVICE_ID_TOSHIBA_TC35815_NWU), .driver_data = TC35815_NWU },
{PCI_DEVICE(PCI_VENDOR_ID_TOSHIBA_2, PCI_DEVICE_ID_TOSHIBA_TC35815_TX4939), .driver_data = TC35815_TX4939 },
/* Beware of PCI posted writes */
#define IOSYNC do { ioread8(ioaddr + StationAddr); } while (0)
-static DEFINE_PCI_DEVICE_TABLE(rhine_pci_tbl) = {
+static const struct pci_device_id rhine_pci_tbl[] = {
{ 0x1106, 0x3043, PCI_ANY_ID, PCI_ANY_ID, }, /* VT86C100A */
{ 0x1106, 0x3065, PCI_ANY_ID, PCI_ANY_ID, }, /* VT6102 */
{ 0x1106, 0x3106, PCI_ANY_ID, PCI_ANY_ID, }, /* 6105{,L,LOM} */
* device driver. Used for hotplug autoloading.
*/
-static DEFINE_PCI_DEVICE_TABLE(velocity_pci_id_table) = {
+static const struct pci_device_id velocity_pci_id_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_612X) },
{ }
};
.probe = temac_of_probe,
.remove = temac_of_remove,
.driver = {
- .owner = THIS_MODULE,
.name = "xilinx_temac",
.of_match_table = temac_of_match,
},
.probe = axienet_of_probe,
.remove = axienet_of_remove,
.driver = {
- .owner = THIS_MODULE,
.name = "xilinx_axienet",
.of_match_table = axienet_of_match,
},
static struct platform_driver xemaclite_of_driver = {
.driver = {
.name = DRIVER_NAME,
- .owner = THIS_MODULE,
.of_match_table = xemaclite_of_match,
},
.probe = xemaclite_of_probe,
static int dfx_pci_register(struct pci_dev *, const struct pci_device_id *);
static void dfx_pci_unregister(struct pci_dev *);
-static DEFINE_PCI_DEVICE_TABLE(dfx_pci_table) = {
+static const struct pci_device_id dfx_pci_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_FDDI) },
{ }
};
extern void mac_drv_clear_rx_queue(struct s_smc *smc);
extern void enable_tx_irq(struct s_smc *smc, u_short queue);
-static DEFINE_PCI_DEVICE_TABLE(skfddi_pci_tbl) = {
+static const struct pci_device_id skfddi_pci_tbl[] = {
{ PCI_VENDOR_ID_SK, PCI_DEVICE_ID_SK_FP, PCI_ANY_ID, PCI_ANY_ID, },
{ } /* Terminating entry */
};
}
}
-static DEFINE_PCI_DEVICE_TABLE(rr_pci_tbl) = {
+static const struct pci_device_id rr_pci_tbl[] = {
{ PCI_VENDOR_ID_ESSENTIAL, PCI_DEVICE_ID_ESSENTIAL_ROADRUNNER,
PCI_ANY_ID, PCI_ANY_ID, },
{ 0,}
#define CONFIG0H_DMA_ON_NORX CONFIG0H_DMA_OFF| OBOE_CONFIG0H_ENDMAC
#define CONFIG0H_DMA_ON CONFIG0H_DMA_ON_NORX | OBOE_CONFIG0H_ENRX
-static DEFINE_PCI_DEVICE_TABLE(toshoboe_pci_tbl) = {
+static const struct pci_device_id toshoboe_pci_tbl[] = {
{ PCI_VENDOR_ID_TOSHIBA, PCI_DEVICE_ID_FIR701, PCI_ANY_ID, PCI_ANY_ID, },
{ PCI_VENDOR_ID_TOSHIBA, PCI_DEVICE_ID_FIRD01, PCI_ANY_ID, PCI_ANY_ID, },
{ } /* Terminating entry */
.resume = toshoboe_wakeup
};
-static int __init
-donauboe_init (void)
-{
- return pci_register_driver(&donauboe_pci_driver);
-}
-
-static void __exit
-donauboe_cleanup (void)
-{
- pci_unregister_driver(&donauboe_pci_driver);
-}
-
-module_init(donauboe_init);
-module_exit(donauboe_cleanup);
+module_pci_driver(donauboe_pci_driver);
}
}
-static DEFINE_PCI_DEVICE_TABLE(via_pci_tbl) = {
+static const struct pci_device_id via_pci_tbl[] = {
{ PCI_VENDOR_ID_VIA, 0x8231, PCI_ANY_ID, PCI_ANY_ID,0,0,0 },
{ PCI_VENDOR_ID_VIA, 0x3109, PCI_ANY_ID, PCI_ANY_ID,0,0,1 },
{ PCI_VENDOR_ID_VIA, 0x3074, PCI_ANY_ID, PCI_ANY_ID,0,0,2 },
static /* const */ char drivername[] = DRIVER_NAME;
-static DEFINE_PCI_DEVICE_TABLE(vlsi_irda_table) = {
+static const struct pci_device_id vlsi_irda_table[] = {
{
.class = PCI_CLASS_WIRELESS_IRDA << 8,
.class_mask = PCI_CLASS_SUBCLASS_MASK << 8,
struct sk_buff_head bc_queue;
struct work_struct bc_work;
bool passthru;
+ int count;
};
-#define MACVLAN_PORT_IS_EMPTY(port) list_empty(&port->vlans)
-
struct macvlan_skb_cb {
const struct macvlan_dev *src;
};
free_percpu(vlan->pcpu_stats);
- if (MACVLAN_PORT_IS_EMPTY(port))
+ port->count -= 1;
+ if (!port->count)
macvlan_port_destroy(port->dev);
}
struct macvlan_dev *vlan = netdev_priv(dev);
int err = -EINVAL;
- if (!vlan->port->passthru)
+ /* Support unicast filter only on passthru devices.
+ * Multicast filter should be allowed on all devices.
+ */
+ if (!vlan->port->passthru && is_unicast_ether_addr(addr))
return -EOPNOTSUPP;
if (flags & NLM_F_REPLACE)
struct macvlan_dev *vlan = netdev_priv(dev);
int err = -EINVAL;
- if (!vlan->port->passthru)
+ /* Support unicast filter only on passthru devices.
+ * Multicast filter should be allowed on all devices.
+ */
+ if (!vlan->port->passthru && is_unicast_ether_addr(addr))
return -EOPNOTSUPP;
if (is_unicast_ether_addr(addr))
vlan->flags = nla_get_u16(data[IFLA_MACVLAN_FLAGS]);
if (vlan->mode == MACVLAN_MODE_PASSTHRU) {
- if (!MACVLAN_PORT_IS_EMPTY(port))
+ if (port->count)
return -EINVAL;
port->passthru = true;
eth_hw_addr_inherit(dev, lowerdev);
}
+ port->count += 1;
err = register_netdevice(dev);
if (err < 0)
goto destroy_port;
unregister_netdev:
unregister_netdevice(dev);
destroy_port:
- if (MACVLAN_PORT_IS_EMPTY(port))
+ port->count -= 1;
+ if (!port->count)
macvlan_port_destroy(lowerdev);
return err;
return bcm7xxx_28nm_afe_config_init(phydev);
}
+static int bcm7xxx_28nm_resume(struct phy_device *phydev)
+{
+ int ret;
+
+ /* Re-apply workarounds coming out suspend/resume */
+ ret = bcm7xxx_28nm_config_init(phydev);
+ if (ret)
+ return ret;
+
+ /* 28nm Gigabit PHYs come out of reset without any half-duplex
+ * or "hub" compliant advertised mode, fix that. This does not
+ * cause any problems with the PHY library since genphy_config_aneg()
+ * gracefully handles auto-negotiated and forced modes.
+ */
+ return genphy_config_aneg(phydev);
+}
+
static int phy_set_clr_bits(struct phy_device *dev, int location,
int set_mask, int clr_mask)
{
}
/* Workaround for putting the PHY in IDDQ mode, required
- * for all BCM7XXX PHYs
+ * for all BCM7XXX 40nm and 65nm PHYs
*/
static int bcm7xxx_suspend(struct phy_device *phydev)
{
.config_init = bcm7xxx_28nm_afe_config_init,
.config_aneg = genphy_config_aneg,
.read_status = genphy_read_status,
- .suspend = bcm7xxx_suspend,
- .resume = bcm7xxx_28nm_afe_config_init,
+ .resume = bcm7xxx_28nm_resume,
.driver = { .owner = THIS_MODULE },
}, {
.phy_id = PHY_ID_BCM7439,
.config_init = bcm7xxx_28nm_afe_config_init,
.config_aneg = genphy_config_aneg,
.read_status = genphy_read_status,
- .suspend = bcm7xxx_suspend,
- .resume = bcm7xxx_28nm_afe_config_init,
+ .resume = bcm7xxx_28nm_resume,
.driver = { .owner = THIS_MODULE },
}, {
.phy_id = PHY_ID_BCM7445,
.config_init = bcm7xxx_28nm_config_init,
.config_aneg = genphy_config_aneg,
.read_status = genphy_read_status,
- .suspend = bcm7xxx_suspend,
- .resume = bcm7xxx_28nm_config_init,
- .driver = { .owner = THIS_MODULE },
-}, {
- .name = "Broadcom BCM7XXX 28nm",
- .phy_id = PHY_ID_BCM7XXX_28,
- .phy_id_mask = PHY_BCM_OUI_MASK,
- .features = PHY_GBIT_FEATURES |
- SUPPORTED_Pause | SUPPORTED_Asym_Pause,
- .flags = PHY_IS_INTERNAL,
- .config_init = bcm7xxx_28nm_config_init,
- .config_aneg = genphy_config_aneg,
- .read_status = genphy_read_status,
- .suspend = bcm7xxx_suspend,
- .resume = bcm7xxx_28nm_config_init,
+ .resume = bcm7xxx_28nm_afe_config_init,
.driver = { .owner = THIS_MODULE },
}, {
.phy_id = PHY_BCM_OUI_4,
{ PHY_ID_BCM7366, 0xfffffff0, },
{ PHY_ID_BCM7439, 0xfffffff0, },
{ PHY_ID_BCM7445, 0xfffffff0, },
- { PHY_ID_BCM7XXX_28, 0xfffffc00 },
{ PHY_BCM_OUI_4, 0xffff0000 },
{ PHY_BCM_OUI_5, 0xffffff00 },
{ }
}
static int smsc_phy_config_init(struct phy_device *phydev)
+{
+ int rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
+
+ if (rc < 0)
+ return rc;
+
+ /* Enable energy detect mode for this SMSC Transceivers */
+ rc = phy_write(phydev, MII_LAN83C185_CTRL_STATUS,
+ rc | MII_LAN83C185_EDPWRDOWN);
+ if (rc < 0)
+ return rc;
+
+ return smsc_phy_ack_interrupt(phydev);
+}
+
+static int smsc_phy_reset(struct phy_device *phydev)
{
int rc = phy_read(phydev, MII_LAN83C185_SPECIAL_MODES);
if (rc < 0)
rc = phy_read(phydev, MII_BMCR);
} while (rc & BMCR_RESET);
}
-
- rc = phy_read(phydev, MII_LAN83C185_CTRL_STATUS);
- if (rc < 0)
- return rc;
-
- /* Enable energy detect mode for this SMSC Transceivers */
- rc = phy_write(phydev, MII_LAN83C185_CTRL_STATUS,
- rc | MII_LAN83C185_EDPWRDOWN);
- if (rc < 0)
- return rc;
-
- return smsc_phy_ack_interrupt (phydev);
+ return 0;
}
static int lan911x_config_init(struct phy_device *phydev)
.config_aneg = genphy_config_aneg,
.read_status = genphy_read_status,
.config_init = smsc_phy_config_init,
+ .soft_reset = smsc_phy_reset,
/* IRQ related */
.ack_interrupt = smsc_phy_ack_interrupt,
.config_aneg = genphy_config_aneg,
.read_status = genphy_read_status,
.config_init = smsc_phy_config_init,
+ .soft_reset = smsc_phy_reset,
/* IRQ related */
.ack_interrupt = smsc_phy_ack_interrupt,
.config_aneg = genphy_config_aneg,
.read_status = genphy_read_status,
.config_init = smsc_phy_config_init,
+ .soft_reset = smsc_phy_reset,
/* IRQ related */
.ack_interrupt = smsc_phy_ack_interrupt,
.config_aneg = genphy_config_aneg,
.read_status = lan87xx_read_status,
.config_init = smsc_phy_config_init,
+ .soft_reset = smsc_phy_reset,
/* IRQ related */
.ack_interrupt = smsc_phy_ack_interrupt,
* PCI Device ID Table
* Last entry must be all 0s
*/
-static DEFINE_PCI_DEVICE_TABLE(vmxnet3_pciid_table) = {
+static const struct pci_device_id vmxnet3_pciid_table[] = {
{PCI_VDEVICE(VMWARE, PCI_DEVICE_ID_VMWARE_VMXNET3)},
{0}
};
__setup("dscc4.setup=", dscc4_setup);
#endif
-static DEFINE_PCI_DEVICE_TABLE(dscc4_pci_tbl) = {
+static const struct pci_device_id dscc4_pci_tbl[] = {
{ PCI_VENDOR_ID_SIEMENS, PCI_DEVICE_ID_SIEMENS_DSCC4,
PCI_ANY_ID, PCI_ANY_ID, },
{ 0,}
/*
* PCI ID lookup table
*/
-static DEFINE_PCI_DEVICE_TABLE(fst_pci_dev_id) = {
+static const struct pci_device_id fst_pci_dev_id[] = {
{PCI_VENDOR_ID_FARSITE, PCI_DEVICE_ID_FARSITE_T2P, PCI_ANY_ID,
PCI_ANY_ID, 0, 0, FST_TYPE_T2P},
static int LMC_PKT_BUF_SZ = 1542;
-static DEFINE_PCI_DEVICE_TABLE(lmc_pci_tbl) = {
+static const struct pci_device_id lmc_pci_tbl[] = {
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP_FAST,
PCI_VENDOR_ID_LMC, PCI_ANY_ID },
{ PCI_VENDOR_ID_DEC, PCI_DEVICE_ID_DEC_TULIP_FAST,
-static DEFINE_PCI_DEVICE_TABLE(pc300_pci_tbl) = {
+static const struct pci_device_id pc300_pci_tbl[] = {
{ PCI_VENDOR_ID_CYCLADES, PCI_DEVICE_ID_PC300_RX_1, PCI_ANY_ID,
PCI_ANY_ID, 0, 0, 0 },
{ PCI_VENDOR_ID_CYCLADES, PCI_DEVICE_ID_PC300_RX_2, PCI_ANY_ID,
-static DEFINE_PCI_DEVICE_TABLE(pci200_pci_tbl) = {
+static const struct pci_device_id pci200_pci_tbl[] = {
{ PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9050, PCI_VENDOR_ID_PLX,
PCI_DEVICE_ID_PLX_PCI200SYN, 0, 0, 0 },
{ 0, }
return 0;
}
-static DEFINE_PCI_DEVICE_TABLE(wanxl_pci_tbl) = {
+static const struct pci_device_id wanxl_pci_tbl[] = {
{ PCI_VENDOR_ID_SBE, PCI_DEVICE_ID_SBE_WANXL100, PCI_ANY_ID,
PCI_ANY_ID, 0, 0, 0 },
{ PCI_VENDOR_ID_SBE, PCI_DEVICE_ID_SBE_WANXL200, PCI_ANY_ID,
module_param(tx_ring_size, uint, 0);
module_param(rx_ring_size, uint, 0);
-static DEFINE_PCI_DEVICE_TABLE(adm8211_pci_id_table) = {
+static const struct pci_device_id adm8211_pci_id_table[] = {
/* ADMtek ADM8211 */
{ PCI_DEVICE(0x10B7, 0x6000) }, /* 3Com 3CRSHPW796 */
{ PCI_DEVICE(0x1200, 0x8201) }, /* ? */
#define DRV_NAME "airo"
#ifdef CONFIG_PCI
-static DEFINE_PCI_DEVICE_TABLE(card_ids) = {
+static const struct pci_device_id card_ids[] = {
{ 0x14b9, 1, PCI_ANY_ID, PCI_ANY_ID, },
{ 0x14b9, 0x4500, PCI_ANY_ID, PCI_ANY_ID },
{ 0x14b9, 0x4800, PCI_ANY_ID, PCI_ANY_ID, },
#define QCA988X_2_0_DEVICE_ID (0x003c)
-static DEFINE_PCI_DEVICE_TABLE(ath10k_pci_id_table) = {
+static const struct pci_device_id ath10k_pci_id_table[] = {
{ PCI_VDEVICE(ATHEROS, QCA988X_2_0_DEVICE_ID) }, /* PCI-E QCA988X V2 */
{0}
};
#define ATH_POLARITY(data) ((data) & 0xff)
/* Devices we match on for LED config info (typically laptops) */
-static DEFINE_PCI_DEVICE_TABLE(ath5k_led_devices) = {
+static const struct pci_device_id ath5k_led_devices[] = {
/* AR5211 */
{ PCI_VDEVICE(ATHEROS, PCI_DEVICE_ID_ATHEROS_AR5211), ATH_LED(0, 0) },
/* HP Compaq nc6xx, nc4000, nx6000 */
#include "reg.h"
/* Known PCI ids */
-static DEFINE_PCI_DEVICE_TABLE(ath5k_pci_id_table) = {
+static const struct pci_device_id ath5k_pci_id_table[] = {
{ PCI_VDEVICE(ATHEROS, 0x0207) }, /* 5210 early */
{ PCI_VDEVICE(ATHEROS, 0x0007) }, /* 5210 */
{ PCI_VDEVICE(ATHEROS, 0x0011) }, /* 5311 - this is on AHB bus !*/
#include <linux/module.h>
#include "ath9k.h"
-static DEFINE_PCI_DEVICE_TABLE(ath_pci_id_table) = {
+static const struct pci_device_id ath_pci_id_table[] = {
{ PCI_VDEVICE(ATHEROS, 0x0023) }, /* PCI */
{ PCI_VDEVICE(ATHEROS, 0x0024) }, /* PCI-E */
{ PCI_VDEVICE(ATHEROS, 0x0027) }, /* PCI */
atomic_t rx_work_urbs;
atomic_t rx_pool_urbs;
kernel_ulong_t features;
+ bool usb_ep_cmd_is_bulk;
/* firmware settings */
struct completion fw_load_wait;
goto err_free;
}
- usb_fill_int_urb(urb, ar->udev, usb_sndintpipe(ar->udev,
- AR9170_USB_EP_CMD), cmd, cmd->hdr.len + 4,
- carl9170_usb_cmd_complete, ar, 1);
+ if (ar->usb_ep_cmd_is_bulk)
+ usb_fill_bulk_urb(urb, ar->udev,
+ usb_sndbulkpipe(ar->udev, AR9170_USB_EP_CMD),
+ cmd, cmd->hdr.len + 4,
+ carl9170_usb_cmd_complete, ar);
+ else
+ usb_fill_int_urb(urb, ar->udev,
+ usb_sndintpipe(ar->udev, AR9170_USB_EP_CMD),
+ cmd, cmd->hdr.len + 4,
+ carl9170_usb_cmd_complete, ar, 1);
if (free_buf)
urb->transfer_flags |= URB_FREE_BUFFER;
static int carl9170_usb_probe(struct usb_interface *intf,
const struct usb_device_id *id)
{
+ struct usb_endpoint_descriptor *ep;
struct ar9170 *ar;
struct usb_device *udev;
- int err;
+ int i, err;
err = usb_reset_device(interface_to_usbdev(intf));
if (err)
ar->intf = intf;
ar->features = id->driver_info;
+ /* We need to remember the type of endpoint 4 because it differs
+ * between high- and full-speed configuration. The high-speed
+ * configuration specifies it as interrupt and the full-speed
+ * configuration as bulk endpoint. This information is required
+ * later when sending urbs to that endpoint.
+ */
+ for (i = 0; i < intf->cur_altsetting->desc.bNumEndpoints; ++i) {
+ ep = &intf->cur_altsetting->endpoint[i].desc;
+
+ if (usb_endpoint_num(ep) == AR9170_USB_EP_CMD &&
+ usb_endpoint_dir_out(ep) &&
+ usb_endpoint_type(ep) == USB_ENDPOINT_XFER_BULK)
+ ar->usb_ep_cmd_is_bulk = true;
+ }
+
usb_set_intfdata(intf, ar);
SET_IEEE80211_DEV(ar->hw, &intf->dev);
MODULE_LICENSE("GPL");
MODULE_SUPPORTED_DEVICE("Atmel at76c506 PCI wireless cards");
-static DEFINE_PCI_DEVICE_TABLE(card_ids) = {
+static const struct pci_device_id card_ids[] = {
{ 0x1114, 0x0506, PCI_ANY_ID, PCI_ANY_ID },
{ 0, }
};
msgbuf->nrof_flowrings = if_msgbuf->nrof_flowrings;
msgbuf->flowring_dma_handle = kzalloc(msgbuf->nrof_flowrings *
sizeof(*msgbuf->flowring_dma_handle), GFP_ATOMIC);
+ if (!msgbuf->flowring_dma_handle)
+ goto fail;
msgbuf->rx_dataoffset = if_msgbuf->rx_dataoffset;
msgbuf->max_rxbufpost = if_msgbuf->max_rxbufpost;
kfree(msgbuf->flow_map);
kfree(msgbuf->txstatus_done_map);
brcmf_msgbuf_release_pktids(msgbuf);
+ kfree(msgbuf->flowring_dma_handle);
if (msgbuf->ioctbuf)
dma_free_coherent(drvr->bus_if->dev,
BRCMF_TX_IOCTL_MAX_MSG_SIZE,
BRCMF_TX_IOCTL_MAX_MSG_SIZE,
msgbuf->ioctbuf, msgbuf->ioctbuf_handle);
brcmf_msgbuf_release_pktids(msgbuf);
+ kfree(msgbuf->flowring_dma_handle);
kfree(msgbuf);
drvr->proto->pd = NULL;
}
}
if (dtoh_mb_data & BRCMF_D2H_DEV_DS_EXIT_NOTE)
brcmf_dbg(PCIE, "D2H_MB_DATA: DEEP SLEEP EXIT\n");
- if (dtoh_mb_data & BRCMF_D2H_DEV_D3_ACK)
+ if (dtoh_mb_data & BRCMF_D2H_DEV_D3_ACK) {
brcmf_dbg(PCIE, "D2H_MB_DATA: D3 ACK\n");
if (waitqueue_active(&devinfo->mbdata_resp_wait)) {
devinfo->mbdata_completed = true;
wake_up(&devinfo->mbdata_resp_wait);
}
+ }
}
/* FIX: do we need mb/wmb/rmb with memory operations? */
-static DEFINE_PCI_DEVICE_TABLE(prism2_pci_id_table) = {
+static const struct pci_device_id prism2_pci_id_table[] = {
/* Intersil Prism3 ISL3872 11Mb/s WLAN Controller */
{ 0x1260, 0x3872, PCI_ANY_ID, PCI_ANY_ID },
/* Intersil Prism2.5 ISL3874 11Mb/s WLAN Controller */
#define PLXDEV(vendor,dev,str) { vendor, dev, PCI_ANY_ID, PCI_ANY_ID }
-static DEFINE_PCI_DEVICE_TABLE(prism2_plx_id_table) = {
+static const struct pci_device_id prism2_plx_id_table[] = {
PLXDEV(0x10b7, 0x7770, "3Com AirConnect PCI 777A"),
PLXDEV(0x111a, 0x1023, "Siemens SpeedStream SS1023"),
PLXDEV(0x126c, 0x8030, "Nortel emobility"),
#define IPW2100_DEV_ID(x) { PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, x }
-static DEFINE_PCI_DEVICE_TABLE(ipw2100_pci_id_table) = {
+static const struct pci_device_id ipw2100_pci_id_table[] = {
IPW2100_DEV_ID(0x2520), /* IN 2100A mPCI 3A */
IPW2100_DEV_ID(0x2521), /* IN 2100A mPCI 3B */
IPW2100_DEV_ID(0x2524), /* IN 2100A mPCI 3B */
strncpy(extra, "unknown", MAX_WX_STRING);
break;
}
+ extra[MAX_WX_STRING - 1] = '\0';
IPW_DEBUG_WX("PRIV GET MODE: %s\n", extra);
}
/* PCI driver stuff */
-static DEFINE_PCI_DEVICE_TABLE(card_ids) = {
+static const struct pci_device_id card_ids[] = {
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2701, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2702, 0, 0, 0},
{PCI_VENDOR_ID_INTEL, 0x1043, 0x8086, 0x2711, 0, 0, 0},
},
};
-DEFINE_PCI_DEVICE_TABLE(il3945_hw_card_ids) = {
+const struct pci_device_id il3945_hw_card_ids[] = {
{IL_PCI_DEVICE(0x4222, 0x1005, il3945_bg_cfg)},
{IL_PCI_DEVICE(0x4222, 0x1034, il3945_bg_cfg)},
{IL_PCI_DEVICE(0x4222, 0x1044, il3945_bg_cfg)},
*****************************************************************************/
/* Hardware specific file defines the PCI IDs table for that hardware module */
-static DEFINE_PCI_DEVICE_TABLE(il4965_hw_card_ids) = {
+static const struct pci_device_id il4965_hw_card_ids[] = {
{IL_PCI_DEVICE(0x4229, PCI_ANY_ID, il4965_cfg)},
{IL_PCI_DEVICE(0x4230, PCI_ANY_ID, il4965_cfg)},
{0}
else
hw->wiphy->flags &= ~WIPHY_FLAG_PS_ON_BY_DEFAULT;
- hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_SCHED_SCAN;
+ /* TODO: enable that only for firmwares that don't crash */
+ /* hw->wiphy->flags |= WIPHY_FLAG_SUPPORTS_SCHED_SCAN; */
hw->wiphy->max_sched_scan_ssids = PROBE_OPTION_MAX;
hw->wiphy->max_match_sets = IWL_SCAN_MAX_PROFILES;
/* we create the 802.11 header and zero length SSID IE. */
.driver_data = (kernel_ulong_t)&(cfg)
/* Hardware specific file defines the PCI IDs table for that hardware module */
-static DEFINE_PCI_DEVICE_TABLE(iwl_hw_card_ids) = {
+static const struct pci_device_id iwl_hw_card_ids[] = {
#if IS_ENABLED(CONFIG_IWLDVM)
{IWL_PCI_DEVICE(0x4232, 0x1201, iwl5100_agn_cfg)}, /* Mini Card */
{IWL_PCI_DEVICE(0x4232, 0x1301, iwl5100_agn_cfg)}, /* Half Mini Card */
return;
}
-static DEFINE_PCI_DEVICE_TABLE(mwifiex_ids) = {
+static const struct pci_device_id mwifiex_ids[] = {
{
PCIE_VENDOR_ID_MARVELL, PCIE_DEVICE_ID_MARVELL_88W8766P,
PCI_ANY_ID, PCI_ANY_ID, 0, 0,
MODULE_FIRMWARE("mwl8k/fmimage_8366.fw");
MODULE_FIRMWARE(MWL8K_8366_AP_FW(MWL8K_8366_AP_FW_API));
-static DEFINE_PCI_DEVICE_TABLE(mwl8k_pci_id_table) = {
+static const struct pci_device_id mwl8k_pci_id_table[] = {
{ PCI_VDEVICE(MARVELL, 0x2a0a), .driver_data = MWL8363, },
{ PCI_VDEVICE(MARVELL, 0x2a0c), .driver_data = MWL8363, },
{ PCI_VDEVICE(MARVELL, 0x2a24), .driver_data = MWL8363, },
pci_disable_device(pdev);
}
-static DEFINE_PCI_DEVICE_TABLE(orinoco_nortel_id_table) = {
+static const struct pci_device_id orinoco_nortel_id_table[] = {
/* Nortel emobility PCI */
{0x126c, 0x8030, PCI_ANY_ID, PCI_ANY_ID,},
/* Symbol LA-4123 PCI */
pci_disable_device(pdev);
}
-static DEFINE_PCI_DEVICE_TABLE(orinoco_pci_id_table) = {
+static const struct pci_device_id orinoco_pci_id_table[] = {
/* Intersil Prism 3 */
{0x1260, 0x3872, PCI_ANY_ID, PCI_ANY_ID,},
/* Intersil Prism 2.5 */
pci_disable_device(pdev);
}
-static DEFINE_PCI_DEVICE_TABLE(orinoco_plx_id_table) = {
+static const struct pci_device_id orinoco_plx_id_table[] = {
{0x111a, 0x1023, PCI_ANY_ID, PCI_ANY_ID,}, /* Siemens SpeedStream SS1023 */
{0x1385, 0x4100, PCI_ANY_ID, PCI_ANY_ID,}, /* Netgear MA301 */
{0x15e8, 0x0130, PCI_ANY_ID, PCI_ANY_ID,}, /* Correga - does this work? */
pci_disable_device(pdev);
}
-static DEFINE_PCI_DEVICE_TABLE(orinoco_tmd_id_table) = {
+static const struct pci_device_id orinoco_tmd_id_table[] = {
{0x15e8, 0x0131, PCI_ANY_ID, PCI_ANY_ID,}, /* NDC and OEMs, e.g. pheecom */
{0,},
};
MODULE_ALIAS("prism54pci");
MODULE_FIRMWARE("isl3886pci");
-static DEFINE_PCI_DEVICE_TABLE(p54p_table) = {
+static const struct pci_device_id p54p_table[] = {
/* Intersil PRISM Duette/Prism GT Wireless LAN adapter */
{ PCI_DEVICE(0x1260, 0x3890) },
/* 3COM 3CRWE154G72 Wireless LAN adapter */
* driver_data
* If you have an update for this please contact prism54-devel@prism54.org
* The latest list can be found at http://wireless.kernel.org/en/users/Drivers/p54 */
-static DEFINE_PCI_DEVICE_TABLE(prism54_id_tbl) = {
+static const struct pci_device_id prism54_id_tbl[] = {
/* Intersil PRISM Duette/Prism GT Wireless LAN adapter */
{
0x1260, 0x3890,
/*
* RT2400pci module information.
*/
-static DEFINE_PCI_DEVICE_TABLE(rt2400pci_device_table) = {
+static const struct pci_device_id rt2400pci_device_table[] = {
{ PCI_DEVICE(0x1814, 0x0101) },
{ 0, }
};
/*
* RT2500pci module information.
*/
-static DEFINE_PCI_DEVICE_TABLE(rt2500pci_device_table) = {
+static const struct pci_device_id rt2500pci_device_table[] = {
{ PCI_DEVICE(0x1814, 0x0201) },
{ 0, }
};
/*
* RT2800pci module information.
*/
-static DEFINE_PCI_DEVICE_TABLE(rt2800pci_device_table) = {
+static const struct pci_device_id rt2800pci_device_table[] = {
{ PCI_DEVICE(0x1814, 0x0601) },
{ PCI_DEVICE(0x1814, 0x0681) },
{ PCI_DEVICE(0x1814, 0x0701) },
/*
* RT61pci module information.
*/
-static DEFINE_PCI_DEVICE_TABLE(rt61pci_device_table) = {
+static const struct pci_device_id rt61pci_device_table[] = {
/* RT2561s */
{ PCI_DEVICE(0x1814, 0x0301) },
/* RT2561 v2 */
MODULE_DESCRIPTION("RTL8180 / RTL8185 / RTL8187SE PCI wireless driver");
MODULE_LICENSE("GPL");
-static DEFINE_PCI_DEVICE_TABLE(rtl8180_table) = {
+static const struct pci_device_id rtl8180_table[] = {
/* rtl8187se */
{ PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8199) },
.maps[RTL_RC_HT_RATEMCS15] = DESC92C_RATEMCS15,
};
-static DEFINE_PCI_DEVICE_TABLE(rtl88ee_pci_ids) = {
+static const struct pci_device_id rtl88ee_pci_ids[] = {
{RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8179, rtl88ee_hal_cfg)},
{},
};
.maps[RTL_RC_HT_RATEMCS15] = DESC92_RATEMCS15,
};
-static DEFINE_PCI_DEVICE_TABLE(rtl92ce_pci_ids) = {
+static const struct pci_device_id rtl92ce_pci_ids[] = {
{RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8191, rtl92ce_hal_cfg)},
{RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8178, rtl92ce_hal_cfg)},
{RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0x8177, rtl92ce_hal_cfg)},
.maps[RTL_RC_HT_RATEMCS15] = DESC92C_RATEMCS15,
};
-static DEFINE_PCI_DEVICE_TABLE(rtl8723be_pci_id) = {
+static const struct pci_device_id rtl8723be_pci_id[] = {
{RTL_PCI_DEVICE(PCI_VENDOR_ID_REALTEK, 0xb723, rtl8723be_hal_cfg)},
{},
};
u16 dealloc_ring[MAX_PENDING_REQS];
struct task_struct *dealloc_task;
wait_queue_head_t dealloc_wq;
+ atomic_t inflight_packets;
/* Use kthread for guest RX */
struct task_struct *task;
extern struct dentry *xen_netback_dbg_root;
#endif
+void xenvif_skb_zerocopy_prepare(struct xenvif_queue *queue,
+ struct sk_buff *skb);
+void xenvif_skb_zerocopy_complete(struct xenvif_queue *queue);
+
#endif /* __XEN_NETBACK__COMMON_H__ */
#define XENVIF_QUEUE_LENGTH 32
#define XENVIF_NAPI_WEIGHT 64
+/* This function is used to set SKBTX_DEV_ZEROCOPY as well as
+ * increasing the inflight counter. We need to increase the inflight
+ * counter because core driver calls into xenvif_zerocopy_callback
+ * which calls xenvif_skb_zerocopy_complete.
+ */
+void xenvif_skb_zerocopy_prepare(struct xenvif_queue *queue,
+ struct sk_buff *skb)
+{
+ skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ atomic_inc(&queue->inflight_packets);
+}
+
+void xenvif_skb_zerocopy_complete(struct xenvif_queue *queue)
+{
+ atomic_dec(&queue->inflight_packets);
+}
+
static inline void xenvif_stop_queue(struct xenvif_queue *queue)
{
struct net_device *dev = queue->vif->dev;
init_timer(&queue->rx_stalled);
- netif_napi_add(queue->vif->dev, &queue->napi, xenvif_poll,
- XENVIF_NAPI_WEIGHT);
-
return 0;
}
init_waitqueue_head(&queue->wq);
init_waitqueue_head(&queue->dealloc_wq);
+ atomic_set(&queue->inflight_packets, 0);
if (tx_evtchn == rx_evtchn) {
/* feature-split-event-channels == 0 */
wake_up_process(queue->task);
wake_up_process(queue->dealloc_task);
+ netif_napi_add(queue->vif->dev, &queue->napi, xenvif_poll,
+ XENVIF_NAPI_WEIGHT);
+
return 0;
err_rx_unbind:
rtnl_unlock();
}
-static void xenvif_wait_unmap_timeout(struct xenvif_queue *queue,
- unsigned int worst_case_skb_lifetime)
-{
- int i, unmap_timeout = 0;
-
- for (i = 0; i < MAX_PENDING_REQS; ++i) {
- if (queue->grant_tx_handle[i] != NETBACK_INVALID_HANDLE) {
- unmap_timeout++;
- schedule_timeout(msecs_to_jiffies(1000));
- if (unmap_timeout > worst_case_skb_lifetime &&
- net_ratelimit())
- netdev_err(queue->vif->dev,
- "Page still granted! Index: %x\n",
- i);
- i = -1;
- }
- }
-}
-
void xenvif_disconnect(struct xenvif *vif)
{
struct xenvif_queue *queue = NULL;
for (queue_index = 0; queue_index < num_queues; ++queue_index) {
queue = &vif->queues[queue_index];
+ netif_napi_del(&queue->napi);
+
if (queue->task) {
del_timer_sync(&queue->rx_stalled);
kthread_stop(queue->task);
void xenvif_deinit_queue(struct xenvif_queue *queue)
{
free_xenballooned_pages(MAX_PENDING_REQS, queue->mmap_pages);
- netif_napi_del(&queue->napi);
}
void xenvif_free(struct xenvif *vif)
struct xenvif_queue *queue = NULL;
unsigned int num_queues = vif->num_queues;
unsigned int queue_index;
- /* Here we want to avoid timeout messages if an skb can be legitimately
- * stuck somewhere else. Realistically this could be an another vif's
- * internal or QDisc queue. That another vif also has this
- * rx_drain_timeout_msecs timeout, so give it time to drain out.
- * Although if that other guest wakes up just before its timeout happens
- * and takes only one skb from QDisc, it can hold onto other skbs for a
- * longer period.
- */
- unsigned int worst_case_skb_lifetime = (rx_drain_timeout_msecs/1000);
unregister_netdev(vif->dev);
for (queue_index = 0; queue_index < num_queues; ++queue_index) {
queue = &vif->queues[queue_index];
- xenvif_wait_unmap_timeout(queue, worst_case_skb_lifetime);
xenvif_deinit_queue(queue);
}
/* remove traces of mapped pages and frag_list */
skb_frag_list_init(skb);
uarg = skb_shinfo(skb)->destructor_arg;
+ /* increase inflight counter to offset decrement in callback */
+ atomic_inc(&queue->inflight_packets);
uarg->callback(uarg, true);
skb_shinfo(skb)->destructor_arg = NULL;
- skb_shinfo(nskb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ xenvif_skb_zerocopy_prepare(queue, nskb);
kfree_skb(nskb);
return 0;
if (net_ratelimit())
netdev_err(queue->vif->dev,
"Not enough memory to consolidate frag_list!\n");
- skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ xenvif_skb_zerocopy_prepare(queue, skb);
kfree_skb(skb);
continue;
}
"Can't setup checksum in net_tx_action\n");
/* We have to set this flag to trigger the callback */
if (skb_shinfo(skb)->destructor_arg)
- skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ xenvif_skb_zerocopy_prepare(queue, skb);
kfree_skb(skb);
continue;
}
* skb. E.g. the __pskb_pull_tail earlier can do such thing.
*/
if (skb_shinfo(skb)->destructor_arg) {
- skb_shinfo(skb)->tx_flags |= SKBTX_DEV_ZEROCOPY;
+ xenvif_skb_zerocopy_prepare(queue, skb);
queue->stats.tx_zerocopy_sent++;
}
queue->stats.tx_zerocopy_success++;
else
queue->stats.tx_zerocopy_fail++;
+ xenvif_skb_zerocopy_complete(queue);
}
static inline void xenvif_tx_dealloc_action(struct xenvif_queue *queue)
return 0;
}
+static bool xenvif_dealloc_kthread_should_stop(struct xenvif_queue *queue)
+{
+ /* Dealloc thread must remain running until all inflight
+ * packets complete.
+ */
+ return kthread_should_stop() &&
+ !atomic_read(&queue->inflight_packets);
+}
+
int xenvif_dealloc_kthread(void *data)
{
struct xenvif_queue *queue = data;
- while (!kthread_should_stop()) {
+ for (;;) {
wait_event_interruptible(queue->dealloc_wq,
tx_dealloc_work_todo(queue) ||
- kthread_should_stop());
- if (kthread_should_stop())
+ xenvif_dealloc_kthread_should_stop(queue));
+ if (xenvif_dealloc_kthread_should_stop(queue))
break;
xenvif_tx_dealloc_action(queue);
}
#define XENVIF_KICK_STR "kick"
+#define BUFFER_SIZE 32
static ssize_t
xenvif_write_io_ring(struct file *filp, const char __user *buf, size_t count,
struct xenvif_queue *queue =
((struct seq_file *)filp->private_data)->private;
int len;
- char write[sizeof(XENVIF_KICK_STR)];
+ char write[BUFFER_SIZE];
/* don't allow partial writes and check the length */
if (*ppos != 0)
return 0;
- if (count < sizeof(XENVIF_KICK_STR) - 1)
+ if (count >= sizeof(write))
return -ENOSPC;
len = simple_write_to_buffer(write,
- sizeof(write),
+ sizeof(write) - 1,
ppos,
buf,
count);
if (len < 0)
return len;
+ write[len] = '\0';
+
if (!strncmp(write, XENVIF_KICK_STR, sizeof(XENVIF_KICK_STR) - 1))
xenvif_interrupt(0, (void *)queue);
else {
.write = xenvif_write_io_ring,
};
-static void xenvif_debugfs_addif(struct xenvif_queue *queue)
+static void xenvif_debugfs_addif(struct xenvif *vif)
{
struct dentry *pfile;
- struct xenvif *vif = queue->vif;
int i;
if (IS_ERR_OR_NULL(xen_netback_dbg_root))
be->vif->num_queues = queue_index;
goto err;
}
+ }
+
#ifdef CONFIG_DEBUG_FS
- xenvif_debugfs_addif(queue);
+ xenvif_debugfs_addif(be->vif);
#endif /* CONFIG_DEBUG_FS */
- }
/* Initialisation completed, tell core driver the number of
* active queues.
config OF_SELFTEST
bool "Device Tree Runtime self tests"
- depends on OF_IRQ
+ depends on OF_IRQ && OF_EARLY_FLATTREE
+ select OF_DYNAMIC
help
This option builds in test cases for the device tree infrastructure
that are executed once at boot time, and the results dumped to the
obj-y = base.o device.o platform.o
+obj-$(CONFIG_OF_DYNAMIC) += dynamic.o
obj-$(CONFIG_OF_FLATTREE) += fdt.o
obj-$(CONFIG_OF_EARLY_FLATTREE) += fdt_address.o
obj-$(CONFIG_OF_PROMTREE) += pdt.o
obj-$(CONFIG_OF_ADDRESS) += address.o
obj-$(CONFIG_OF_IRQ) += irq.o
obj-$(CONFIG_OF_NET) += of_net.o
-obj-$(CONFIG_OF_SELFTEST) += selftest.o
+obj-$(CONFIG_OF_SELFTEST) += of_selftest.o
+of_selftest-objs := selftest.o testcase-data/testcases.dtb.o
obj-$(CONFIG_OF_MDIO) += of_mdio.o
obj-$(CONFIG_OF_PCI) += of_pci.o
obj-$(CONFIG_OF_PCI_IRQ) += of_pci_irq.o
* as published by the Free Software Foundation; either version
* 2 of the License, or (at your option) any later version.
*/
+#include <linux/console.h>
#include <linux/ctype.h>
#include <linux/cpu.h>
#include <linux/module.h>
EXPORT_SYMBOL(of_allnodes);
struct device_node *of_chosen;
struct device_node *of_aliases;
-static struct device_node *of_stdout;
+struct device_node *of_stdout;
-static struct kset *of_kset;
+struct kset *of_kset;
/*
- * Used to protect the of_aliases; but also overloaded to hold off addition of
- * nodes to sysfs
+ * Used to protect the of_aliases, to hold off addition of nodes to sysfs.
+ * This mutex must be held whenever modifications are being made to the
+ * device tree. The of_{attach,detach}_node() and
+ * of_{add,remove,update}_property() helpers make sure this happens.
*/
-DEFINE_MUTEX(of_aliases_mutex);
+DEFINE_MUTEX(of_mutex);
/* use when traversing tree through the allnext, child, sibling,
* or parent members of struct device_node.
}
#endif
-#if defined(CONFIG_OF_DYNAMIC)
-/**
- * of_node_get - Increment refcount of a node
- * @node: Node to inc refcount, NULL is supported to
- * simplify writing of callers
- *
- * Returns node.
- */
-struct device_node *of_node_get(struct device_node *node)
-{
- if (node)
- kobject_get(&node->kobj);
- return node;
-}
-EXPORT_SYMBOL(of_node_get);
-
-static inline struct device_node *kobj_to_device_node(struct kobject *kobj)
-{
- return container_of(kobj, struct device_node, kobj);
-}
-
-/**
- * of_node_release - release a dynamically allocated node
- * @kref: kref element of the node to be released
- *
- * In of_node_put() this function is passed to kref_put()
- * as the destructor.
- */
-static void of_node_release(struct kobject *kobj)
-{
- struct device_node *node = kobj_to_device_node(kobj);
- struct property *prop = node->properties;
-
- /* We should never be releasing nodes that haven't been detached. */
- if (!of_node_check_flag(node, OF_DETACHED)) {
- pr_err("ERROR: Bad of_node_put() on %s\n", node->full_name);
- dump_stack();
- return;
- }
-
- if (!of_node_check_flag(node, OF_DYNAMIC))
- return;
-
- while (prop) {
- struct property *next = prop->next;
- kfree(prop->name);
- kfree(prop->value);
- kfree(prop);
- prop = next;
-
- if (!prop) {
- prop = node->deadprops;
- node->deadprops = NULL;
- }
- }
- kfree(node->full_name);
- kfree(node->data);
- kfree(node);
-}
-
-/**
- * of_node_put - Decrement refcount of a node
- * @node: Node to dec refcount, NULL is supported to
- * simplify writing of callers
- *
- */
-void of_node_put(struct device_node *node)
-{
- if (node)
- kobject_put(&node->kobj);
-}
-EXPORT_SYMBOL(of_node_put);
-#else
+#ifndef CONFIG_OF_DYNAMIC
static void of_node_release(struct kobject *kobj)
{
/* Without CONFIG_OF_DYNAMIC, no nodes gets freed */
return name;
}
-static int __of_add_property_sysfs(struct device_node *np, struct property *pp)
+int __of_add_property_sysfs(struct device_node *np, struct property *pp)
{
int rc;
/* Important: Don't leak passwords */
bool secure = strncmp(pp->name, "security-", 9) == 0;
+ if (!of_kset || !of_node_is_attached(np))
+ return 0;
+
sysfs_bin_attr_init(&pp->attr);
pp->attr.attr.name = safe_name(&np->kobj, pp->name);
pp->attr.attr.mode = secure ? S_IRUSR : S_IRUGO;
return rc;
}
-static int __of_node_add(struct device_node *np)
+int __of_attach_node_sysfs(struct device_node *np)
{
const char *name;
struct property *pp;
int rc;
+ if (!of_kset)
+ return 0;
+
np->kobj.kset = of_kset;
if (!np->parent) {
/* Nodes without parents are new top level trees */
return 0;
}
-int of_node_add(struct device_node *np)
-{
- int rc = 0;
-
- BUG_ON(!of_node_is_initialized(np));
-
- /*
- * Grab the mutex here so that in a race condition between of_init() and
- * of_node_add(), node addition will still be consistent.
- */
- mutex_lock(&of_aliases_mutex);
- if (of_kset)
- rc = __of_node_add(np);
- else
- /* This scenario may be perfectly valid, but report it anyway */
- pr_info("of_node_add(%s) before of_init()\n", np->full_name);
- mutex_unlock(&of_aliases_mutex);
- return rc;
-}
-
-#if defined(CONFIG_OF_DYNAMIC)
-static void of_node_remove(struct device_node *np)
-{
- struct property *pp;
-
- BUG_ON(!of_node_is_initialized(np));
-
- /* only remove properties if on sysfs */
- if (of_node_is_attached(np)) {
- for_each_property_of_node(np, pp)
- sysfs_remove_bin_file(&np->kobj, &pp->attr);
- kobject_del(&np->kobj);
- }
-
- /* finally remove the kobj_init ref */
- of_node_put(np);
-}
-#endif
-
static int __init of_init(void)
{
struct device_node *np;
/* Create the kset, and register existing nodes */
- mutex_lock(&of_aliases_mutex);
+ mutex_lock(&of_mutex);
of_kset = kset_create_and_add("devicetree", NULL, firmware_kobj);
if (!of_kset) {
- mutex_unlock(&of_aliases_mutex);
+ mutex_unlock(&of_mutex);
return -ENOMEM;
}
for_each_of_allnodes(np)
- __of_node_add(np);
- mutex_unlock(&of_aliases_mutex);
+ __of_attach_node_sysfs(np);
+ mutex_unlock(&of_mutex);
/* Symlink in /proc as required by userspace ABI */
if (of_allnodes)
* Find a property with a given name for a given node
* and return the value.
*/
-static const void *__of_get_property(const struct device_node *np,
- const char *name, int *lenp)
+const void *__of_get_property(const struct device_node *np,
+ const char *name, int *lenp)
{
struct property *pp = __of_find_property(np, name, lenp);
}
EXPORT_SYMBOL(of_count_phandle_with_args);
-#if defined(CONFIG_OF_DYNAMIC)
-static int of_property_notify(int action, struct device_node *np,
- struct property *prop)
-{
- struct of_prop_reconfig pr;
-
- /* only call notifiers if the node is attached */
- if (!of_node_is_attached(np))
- return 0;
-
- pr.dn = np;
- pr.prop = prop;
- return of_reconfig_notify(action, &pr);
-}
-#else
-static int of_property_notify(int action, struct device_node *np,
- struct property *prop)
-{
- return 0;
-}
-#endif
-
/**
* __of_add_property - Add a property to a node without lock operations
*/
-static int __of_add_property(struct device_node *np, struct property *prop)
+int __of_add_property(struct device_node *np, struct property *prop)
{
struct property **next;
unsigned long flags;
int rc;
- rc = of_property_notify(OF_RECONFIG_ADD_PROPERTY, np, prop);
- if (rc)
- return rc;
+ mutex_lock(&of_mutex);
raw_spin_lock_irqsave(&devtree_lock, flags);
rc = __of_add_property(np, prop);
raw_spin_unlock_irqrestore(&devtree_lock, flags);
- if (rc)
- return rc;
- if (of_node_is_attached(np))
+ if (!rc)
__of_add_property_sysfs(np, prop);
+ mutex_unlock(&of_mutex);
+
+ if (!rc)
+ of_property_notify(OF_RECONFIG_ADD_PROPERTY, np, prop, NULL);
+
return rc;
}
+int __of_remove_property(struct device_node *np, struct property *prop)
+{
+ struct property **next;
+
+ for (next = &np->properties; *next; next = &(*next)->next) {
+ if (*next == prop)
+ break;
+ }
+ if (*next == NULL)
+ return -ENODEV;
+
+ /* found the node */
+ *next = prop->next;
+ prop->next = np->deadprops;
+ np->deadprops = prop;
+
+ return 0;
+}
+
+void __of_remove_property_sysfs(struct device_node *np, struct property *prop)
+{
+ /* at early boot, bail here and defer setup to of_init() */
+ if (of_kset && of_node_is_attached(np))
+ sysfs_remove_bin_file(&np->kobj, &prop->attr);
+}
+
/**
* of_remove_property - Remove a property from a node.
*
*/
int of_remove_property(struct device_node *np, struct property *prop)
{
- struct property **next;
unsigned long flags;
- int found = 0;
int rc;
- rc = of_property_notify(OF_RECONFIG_REMOVE_PROPERTY, np, prop);
- if (rc)
- return rc;
+ mutex_lock(&of_mutex);
raw_spin_lock_irqsave(&devtree_lock, flags);
- next = &np->properties;
- while (*next) {
- if (*next == prop) {
- /* found the node */
- *next = prop->next;
- prop->next = np->deadprops;
- np->deadprops = prop;
- found = 1;
- break;
- }
- next = &(*next)->next;
- }
+ rc = __of_remove_property(np, prop);
raw_spin_unlock_irqrestore(&devtree_lock, flags);
- if (!found)
- return -ENODEV;
+ if (!rc)
+ __of_remove_property_sysfs(np, prop);
- /* at early boot, bail hear and defer setup to of_init() */
- if (!of_kset)
- return 0;
+ mutex_unlock(&of_mutex);
- sysfs_remove_bin_file(&np->kobj, &prop->attr);
+ if (!rc)
+ of_property_notify(OF_RECONFIG_REMOVE_PROPERTY, np, prop, NULL);
- return 0;
+ return rc;
}
-/*
- * of_update_property - Update a property in a node, if the property does
- * not exist, add it.
- *
- * Note that we don't actually remove it, since we have given out
- * who-knows-how-many pointers to the data using get-property.
- * Instead we just move the property to the "dead properties" list,
- * and add the new property to the property list
- */
-int of_update_property(struct device_node *np, struct property *newprop)
+int __of_update_property(struct device_node *np, struct property *newprop,
+ struct property **oldpropp)
{
struct property **next, *oldprop;
- unsigned long flags;
- int rc;
-
- rc = of_property_notify(OF_RECONFIG_UPDATE_PROPERTY, np, newprop);
- if (rc)
- return rc;
- if (!newprop->name)
- return -EINVAL;
+ for (next = &np->properties; *next; next = &(*next)->next) {
+ if (of_prop_cmp((*next)->name, newprop->name) == 0)
+ break;
+ }
+ *oldpropp = oldprop = *next;
- raw_spin_lock_irqsave(&devtree_lock, flags);
- next = &np->properties;
- oldprop = __of_find_property(np, newprop->name, NULL);
- if (!oldprop) {
- /* add the new node */
- rc = __of_add_property(np, newprop);
- } else while (*next) {
+ if (oldprop) {
/* replace the node */
- if (*next == oldprop) {
- newprop->next = oldprop->next;
- *next = newprop;
- oldprop->next = np->deadprops;
- np->deadprops = oldprop;
- break;
- }
- next = &(*next)->next;
+ newprop->next = oldprop->next;
+ *next = newprop;
+ oldprop->next = np->deadprops;
+ np->deadprops = oldprop;
+ } else {
+ /* new node */
+ newprop->next = NULL;
+ *next = newprop;
}
- raw_spin_unlock_irqrestore(&devtree_lock, flags);
- if (rc)
- return rc;
+ return 0;
+}
+
+void __of_update_property_sysfs(struct device_node *np, struct property *newprop,
+ struct property *oldprop)
+{
/* At early boot, bail out and defer setup to of_init() */
if (!of_kset)
- return 0;
+ return;
- /* Update the sysfs attribute */
if (oldprop)
sysfs_remove_bin_file(&np->kobj, &oldprop->attr);
__of_add_property_sysfs(np, newprop);
-
- return 0;
}
-#if defined(CONFIG_OF_DYNAMIC)
/*
- * Support for dynamic device trees.
+ * of_update_property - Update a property in a node, if the property does
+ * not exist, add it.
*
- * On some platforms, the device tree can be manipulated at runtime.
- * The routines in this section support adding, removing and changing
- * device tree nodes.
- */
-
-static BLOCKING_NOTIFIER_HEAD(of_reconfig_chain);
-
-int of_reconfig_notifier_register(struct notifier_block *nb)
-{
- return blocking_notifier_chain_register(&of_reconfig_chain, nb);
-}
-EXPORT_SYMBOL_GPL(of_reconfig_notifier_register);
-
-int of_reconfig_notifier_unregister(struct notifier_block *nb)
-{
- return blocking_notifier_chain_unregister(&of_reconfig_chain, nb);
-}
-EXPORT_SYMBOL_GPL(of_reconfig_notifier_unregister);
-
-int of_reconfig_notify(unsigned long action, void *p)
-{
- int rc;
-
- rc = blocking_notifier_call_chain(&of_reconfig_chain, action, p);
- return notifier_to_errno(rc);
-}
-
-/**
- * of_attach_node - Plug a device node into the tree and global list.
+ * Note that we don't actually remove it, since we have given out
+ * who-knows-how-many pointers to the data using get-property.
+ * Instead we just move the property to the "dead properties" list,
+ * and add the new property to the property list
*/
-int of_attach_node(struct device_node *np)
+int of_update_property(struct device_node *np, struct property *newprop)
{
+ struct property *oldprop;
unsigned long flags;
int rc;
- rc = of_reconfig_notify(OF_RECONFIG_ATTACH_NODE, np);
- if (rc)
- return rc;
-
- raw_spin_lock_irqsave(&devtree_lock, flags);
- np->sibling = np->parent->child;
- np->allnext = np->parent->allnext;
- np->parent->allnext = np;
- np->parent->child = np;
- of_node_clear_flag(np, OF_DETACHED);
- raw_spin_unlock_irqrestore(&devtree_lock, flags);
-
- of_node_add(np);
- return 0;
-}
-
-/**
- * of_detach_node - "Unplug" a node from the device tree.
- *
- * The caller must hold a reference to the node. The memory associated with
- * the node is not freed until its refcount goes to zero.
- */
-int of_detach_node(struct device_node *np)
-{
- struct device_node *parent;
- unsigned long flags;
- int rc = 0;
+ if (!newprop->name)
+ return -EINVAL;
- rc = of_reconfig_notify(OF_RECONFIG_DETACH_NODE, np);
- if (rc)
- return rc;
+ mutex_lock(&of_mutex);
raw_spin_lock_irqsave(&devtree_lock, flags);
+ rc = __of_update_property(np, newprop, &oldprop);
+ raw_spin_unlock_irqrestore(&devtree_lock, flags);
- if (of_node_check_flag(np, OF_DETACHED)) {
- /* someone already detached it */
- raw_spin_unlock_irqrestore(&devtree_lock, flags);
- return rc;
- }
-
- parent = np->parent;
- if (!parent) {
- raw_spin_unlock_irqrestore(&devtree_lock, flags);
- return rc;
- }
+ if (!rc)
+ __of_update_property_sysfs(np, newprop, oldprop);
- if (of_allnodes == np)
- of_allnodes = np->allnext;
- else {
- struct device_node *prev;
- for (prev = of_allnodes;
- prev->allnext != np;
- prev = prev->allnext)
- ;
- prev->allnext = np->allnext;
- }
+ mutex_unlock(&of_mutex);
- if (parent->child == np)
- parent->child = np->sibling;
- else {
- struct device_node *prevsib;
- for (prevsib = np->parent->child;
- prevsib->sibling != np;
- prevsib = prevsib->sibling)
- ;
- prevsib->sibling = np->sibling;
- }
-
- of_node_set_flag(np, OF_DETACHED);
- raw_spin_unlock_irqrestore(&devtree_lock, flags);
+ if (!rc)
+ of_property_notify(OF_RECONFIG_UPDATE_PROPERTY, np, newprop, oldprop);
- of_node_remove(np);
return rc;
}
-#endif /* defined(CONFIG_OF_DYNAMIC) */
static void of_alias_add(struct alias_prop *ap, struct device_node *np,
int id, const char *stem, int stem_len)
of_chosen = of_find_node_by_path("/chosen@0");
if (of_chosen) {
+ /* linux,stdout-path and /aliases/stdout are for legacy compatibility */
const char *name = of_get_property(of_chosen, "stdout-path", NULL);
if (!name)
name = of_get_property(of_chosen, "linux,stdout-path", NULL);
+ if (IS_ENABLED(CONFIG_PPC) && !name)
+ name = of_get_property(of_aliases, "stdout", NULL);
if (name)
of_stdout = of_find_node_by_path(name);
}
struct alias_prop *app;
int id = -ENODEV;
- mutex_lock(&of_aliases_mutex);
+ mutex_lock(&of_mutex);
list_for_each_entry(app, &aliases_lookup, link) {
if (strcmp(app->stem, stem) != 0)
continue;
break;
}
}
- mutex_unlock(&of_aliases_mutex);
+ mutex_unlock(&of_mutex);
return id;
}
EXPORT_SYMBOL_GPL(of_prop_next_string);
/**
- * of_device_is_stdout_path - check if a device node matches the
- * linux,stdout-path property
- *
- * Check if this device node matches the linux,stdout-path property
- * in the chosen node. return true if yes, false otherwise.
+ * of_console_check() - Test and setup console for DT setup
+ * @dn - Pointer to device node
+ * @name - Name to use for preferred console without index. ex. "ttyS"
+ * @index - Index to use for preferred console.
+ *
+ * Check if the given device node matches the stdout-path property in the
+ * /chosen node. If it does then register it as the preferred console and return
+ * TRUE. Otherwise return FALSE.
*/
-int of_device_is_stdout_path(struct device_node *dn)
+bool of_console_check(struct device_node *dn, char *name, int index)
{
- if (!of_stdout)
+ if (!dn || dn != of_stdout || console_set_on_cmdline)
return false;
-
- return of_stdout == dn;
+ return add_preferred_console(name, index, NULL);
}
-EXPORT_SYMBOL_GPL(of_device_is_stdout_path);
+EXPORT_SYMBOL_GPL(of_console_check);
/**
* of_find_next_cache_node - Find a node's subsidiary cache
add_uevent_var(env, "OF_COMPATIBLE_N=%d", seen);
seen = 0;
- mutex_lock(&of_aliases_mutex);
+ mutex_lock(&of_mutex);
list_for_each_entry(app, &aliases_lookup, link) {
if (dev->of_node == app->np) {
add_uevent_var(env, "OF_ALIAS_%d=%s", seen,
seen++;
}
}
- mutex_unlock(&of_aliases_mutex);
+ mutex_unlock(&of_mutex);
}
int of_device_uevent_modalias(struct device *dev, struct kobj_uevent_env *env)
--- /dev/null
+/*
+ * Support for dynamic device trees.
+ *
+ * On some platforms, the device tree can be manipulated at runtime.
+ * The routines in this section support adding, removing and changing
+ * device tree nodes.
+ */
+
+#include <linux/of.h>
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/proc_fs.h>
+
+#include "of_private.h"
+
+/**
+ * of_node_get() - Increment refcount of a node
+ * @node: Node to inc refcount, NULL is supported to simplify writing of
+ * callers
+ *
+ * Returns node.
+ */
+struct device_node *of_node_get(struct device_node *node)
+{
+ if (node)
+ kobject_get(&node->kobj);
+ return node;
+}
+EXPORT_SYMBOL(of_node_get);
+
+/**
+ * of_node_put() - Decrement refcount of a node
+ * @node: Node to dec refcount, NULL is supported to simplify writing of
+ * callers
+ */
+void of_node_put(struct device_node *node)
+{
+ if (node)
+ kobject_put(&node->kobj);
+}
+EXPORT_SYMBOL(of_node_put);
+
+void __of_detach_node_sysfs(struct device_node *np)
+{
+ struct property *pp;
+
+ BUG_ON(!of_node_is_initialized(np));
+ if (!of_kset)
+ return;
+
+ /* only remove properties if on sysfs */
+ if (of_node_is_attached(np)) {
+ for_each_property_of_node(np, pp)
+ sysfs_remove_bin_file(&np->kobj, &pp->attr);
+ kobject_del(&np->kobj);
+ }
+
+ /* finally remove the kobj_init ref */
+ of_node_put(np);
+}
+
+static BLOCKING_NOTIFIER_HEAD(of_reconfig_chain);
+
+int of_reconfig_notifier_register(struct notifier_block *nb)
+{
+ return blocking_notifier_chain_register(&of_reconfig_chain, nb);
+}
+EXPORT_SYMBOL_GPL(of_reconfig_notifier_register);
+
+int of_reconfig_notifier_unregister(struct notifier_block *nb)
+{
+ return blocking_notifier_chain_unregister(&of_reconfig_chain, nb);
+}
+EXPORT_SYMBOL_GPL(of_reconfig_notifier_unregister);
+
+int of_reconfig_notify(unsigned long action, void *p)
+{
+ int rc;
+
+ rc = blocking_notifier_call_chain(&of_reconfig_chain, action, p);
+ return notifier_to_errno(rc);
+}
+
+int of_property_notify(int action, struct device_node *np,
+ struct property *prop, struct property *oldprop)
+{
+ struct of_prop_reconfig pr;
+
+ /* only call notifiers if the node is attached */
+ if (!of_node_is_attached(np))
+ return 0;
+
+ pr.dn = np;
+ pr.prop = prop;
+ pr.old_prop = oldprop;
+ return of_reconfig_notify(action, &pr);
+}
+
+void __of_attach_node(struct device_node *np)
+{
+ const __be32 *phandle;
+ int sz;
+
+ np->name = __of_get_property(np, "name", NULL) ? : "<NULL>";
+ np->type = __of_get_property(np, "device_type", NULL) ? : "<NULL>";
+
+ phandle = __of_get_property(np, "phandle", &sz);
+ if (!phandle)
+ phandle = __of_get_property(np, "linux,phandle", &sz);
+ if (IS_ENABLED(PPC_PSERIES) && !phandle)
+ phandle = __of_get_property(np, "ibm,phandle", &sz);
+ np->phandle = (phandle && (sz >= 4)) ? be32_to_cpup(phandle) : 0;
+
+ np->child = NULL;
+ np->sibling = np->parent->child;
+ np->allnext = np->parent->allnext;
+ np->parent->allnext = np;
+ np->parent->child = np;
+ of_node_clear_flag(np, OF_DETACHED);
+}
+
+/**
+ * of_attach_node() - Plug a device node into the tree and global list.
+ */
+int of_attach_node(struct device_node *np)
+{
+ unsigned long flags;
+
+ mutex_lock(&of_mutex);
+ raw_spin_lock_irqsave(&devtree_lock, flags);
+ __of_attach_node(np);
+ raw_spin_unlock_irqrestore(&devtree_lock, flags);
+
+ __of_attach_node_sysfs(np);
+ mutex_unlock(&of_mutex);
+
+ of_reconfig_notify(OF_RECONFIG_ATTACH_NODE, np);
+
+ return 0;
+}
+
+void __of_detach_node(struct device_node *np)
+{
+ struct device_node *parent;
+
+ if (WARN_ON(of_node_check_flag(np, OF_DETACHED)))
+ return;
+
+ parent = np->parent;
+ if (WARN_ON(!parent))
+ return;
+
+ if (of_allnodes == np)
+ of_allnodes = np->allnext;
+ else {
+ struct device_node *prev;
+ for (prev = of_allnodes;
+ prev->allnext != np;
+ prev = prev->allnext)
+ ;
+ prev->allnext = np->allnext;
+ }
+
+ if (parent->child == np)
+ parent->child = np->sibling;
+ else {
+ struct device_node *prevsib;
+ for (prevsib = np->parent->child;
+ prevsib->sibling != np;
+ prevsib = prevsib->sibling)
+ ;
+ prevsib->sibling = np->sibling;
+ }
+
+ of_node_set_flag(np, OF_DETACHED);
+}
+
+/**
+ * of_detach_node() - "Unplug" a node from the device tree.
+ *
+ * The caller must hold a reference to the node. The memory associated with
+ * the node is not freed until its refcount goes to zero.
+ */
+int of_detach_node(struct device_node *np)
+{
+ unsigned long flags;
+ int rc = 0;
+
+ mutex_lock(&of_mutex);
+ raw_spin_lock_irqsave(&devtree_lock, flags);
+ __of_detach_node(np);
+ raw_spin_unlock_irqrestore(&devtree_lock, flags);
+
+ __of_detach_node_sysfs(np);
+ mutex_unlock(&of_mutex);
+
+ of_reconfig_notify(OF_RECONFIG_DETACH_NODE, np);
+
+ return rc;
+}
+
+/**
+ * of_node_release() - release a dynamically allocated node
+ * @kref: kref element of the node to be released
+ *
+ * In of_node_put() this function is passed to kref_put() as the destructor.
+ */
+void of_node_release(struct kobject *kobj)
+{
+ struct device_node *node = kobj_to_device_node(kobj);
+ struct property *prop = node->properties;
+
+ /* We should never be releasing nodes that haven't been detached. */
+ if (!of_node_check_flag(node, OF_DETACHED)) {
+ pr_err("ERROR: Bad of_node_put() on %s\n", node->full_name);
+ dump_stack();
+ return;
+ }
+
+ if (!of_node_check_flag(node, OF_DYNAMIC))
+ return;
+
+ while (prop) {
+ struct property *next = prop->next;
+ kfree(prop->name);
+ kfree(prop->value);
+ kfree(prop);
+ prop = next;
+
+ if (!prop) {
+ prop = node->deadprops;
+ node->deadprops = NULL;
+ }
+ }
+ kfree(node->full_name);
+ kfree(node->data);
+ kfree(node);
+}
+
+/**
+ * __of_prop_dup - Copy a property dynamically.
+ * @prop: Property to copy
+ * @allocflags: Allocation flags (typically pass GFP_KERNEL)
+ *
+ * Copy a property by dynamically allocating the memory of both the
+ * property stucture and the property name & contents. The property's
+ * flags have the OF_DYNAMIC bit set so that we can differentiate between
+ * dynamically allocated properties and not.
+ * Returns the newly allocated property or NULL on out of memory error.
+ */
+struct property *__of_prop_dup(const struct property *prop, gfp_t allocflags)
+{
+ struct property *new;
+
+ new = kzalloc(sizeof(*new), allocflags);
+ if (!new)
+ return NULL;
+
+ /*
+ * NOTE: There is no check for zero length value.
+ * In case of a boolean property, this will allocate a value
+ * of zero bytes. We do this to work around the use
+ * of of_get_property() calls on boolean values.
+ */
+ new->name = kstrdup(prop->name, allocflags);
+ new->value = kmemdup(prop->value, prop->length, allocflags);
+ new->length = prop->length;
+ if (!new->name || !new->value)
+ goto err_free;
+
+ /* mark the property as dynamic */
+ of_property_set_flag(new, OF_DYNAMIC);
+
+ return new;
+
+ err_free:
+ kfree(new->name);
+ kfree(new->value);
+ kfree(new);
+ return NULL;
+}
+
+/**
+ * __of_node_alloc() - Create an empty device node dynamically.
+ * @full_name: Full name of the new device node
+ * @allocflags: Allocation flags (typically pass GFP_KERNEL)
+ *
+ * Create an empty device tree node, suitable for further modification.
+ * The node data are dynamically allocated and all the node flags
+ * have the OF_DYNAMIC & OF_DETACHED bits set.
+ * Returns the newly allocated node or NULL on out of memory error.
+ */
+struct device_node *__of_node_alloc(const char *full_name, gfp_t allocflags)
+{
+ struct device_node *node;
+
+ node = kzalloc(sizeof(*node), allocflags);
+ if (!node)
+ return NULL;
+
+ node->full_name = kstrdup(full_name, allocflags);
+ of_node_set_flag(node, OF_DYNAMIC);
+ of_node_set_flag(node, OF_DETACHED);
+ if (!node->full_name)
+ goto err_free;
+
+ of_node_init(node);
+
+ return node;
+
+ err_free:
+ kfree(node->full_name);
+ kfree(node);
+ return NULL;
+}
+
+static void __of_changeset_entry_destroy(struct of_changeset_entry *ce)
+{
+ of_node_put(ce->np);
+ list_del(&ce->node);
+ kfree(ce);
+}
+
+#ifdef DEBUG
+static void __of_changeset_entry_dump(struct of_changeset_entry *ce)
+{
+ switch (ce->action) {
+ case OF_RECONFIG_ADD_PROPERTY:
+ pr_debug("%p: %s %s/%s\n",
+ ce, "ADD_PROPERTY ", ce->np->full_name,
+ ce->prop->name);
+ break;
+ case OF_RECONFIG_REMOVE_PROPERTY:
+ pr_debug("%p: %s %s/%s\n",
+ ce, "REMOVE_PROPERTY", ce->np->full_name,
+ ce->prop->name);
+ break;
+ case OF_RECONFIG_UPDATE_PROPERTY:
+ pr_debug("%p: %s %s/%s\n",
+ ce, "UPDATE_PROPERTY", ce->np->full_name,
+ ce->prop->name);
+ break;
+ case OF_RECONFIG_ATTACH_NODE:
+ pr_debug("%p: %s %s\n",
+ ce, "ATTACH_NODE ", ce->np->full_name);
+ break;
+ case OF_RECONFIG_DETACH_NODE:
+ pr_debug("%p: %s %s\n",
+ ce, "DETACH_NODE ", ce->np->full_name);
+ break;
+ }
+}
+#else
+static inline void __of_changeset_entry_dump(struct of_changeset_entry *ce)
+{
+ /* empty */
+}
+#endif
+
+static void __of_changeset_entry_invert(struct of_changeset_entry *ce,
+ struct of_changeset_entry *rce)
+{
+ memcpy(rce, ce, sizeof(*rce));
+
+ switch (ce->action) {
+ case OF_RECONFIG_ATTACH_NODE:
+ rce->action = OF_RECONFIG_DETACH_NODE;
+ break;
+ case OF_RECONFIG_DETACH_NODE:
+ rce->action = OF_RECONFIG_ATTACH_NODE;
+ break;
+ case OF_RECONFIG_ADD_PROPERTY:
+ rce->action = OF_RECONFIG_REMOVE_PROPERTY;
+ break;
+ case OF_RECONFIG_REMOVE_PROPERTY:
+ rce->action = OF_RECONFIG_ADD_PROPERTY;
+ break;
+ case OF_RECONFIG_UPDATE_PROPERTY:
+ rce->old_prop = ce->prop;
+ rce->prop = ce->old_prop;
+ break;
+ }
+}
+
+static void __of_changeset_entry_notify(struct of_changeset_entry *ce, bool revert)
+{
+ struct of_changeset_entry ce_inverted;
+ int ret;
+
+ if (revert) {
+ __of_changeset_entry_invert(ce, &ce_inverted);
+ ce = &ce_inverted;
+ }
+
+ switch (ce->action) {
+ case OF_RECONFIG_ATTACH_NODE:
+ case OF_RECONFIG_DETACH_NODE:
+ ret = of_reconfig_notify(ce->action, ce->np);
+ break;
+ case OF_RECONFIG_ADD_PROPERTY:
+ case OF_RECONFIG_REMOVE_PROPERTY:
+ case OF_RECONFIG_UPDATE_PROPERTY:
+ ret = of_property_notify(ce->action, ce->np, ce->prop, ce->old_prop);
+ break;
+ default:
+ pr_err("%s: invalid devicetree changeset action: %i\n", __func__,
+ (int)ce->action);
+ return;
+ }
+
+ if (ret)
+ pr_err("%s: notifier error @%s\n", __func__, ce->np->full_name);
+}
+
+static int __of_changeset_entry_apply(struct of_changeset_entry *ce)
+{
+ struct property *old_prop, **propp;
+ unsigned long flags;
+ int ret = 0;
+
+ __of_changeset_entry_dump(ce);
+
+ raw_spin_lock_irqsave(&devtree_lock, flags);
+ switch (ce->action) {
+ case OF_RECONFIG_ATTACH_NODE:
+ __of_attach_node(ce->np);
+ break;
+ case OF_RECONFIG_DETACH_NODE:
+ __of_detach_node(ce->np);
+ break;
+ case OF_RECONFIG_ADD_PROPERTY:
+ /* If the property is in deadprops then it must be removed */
+ for (propp = &ce->np->deadprops; *propp; propp = &(*propp)->next) {
+ if (*propp == ce->prop) {
+ *propp = ce->prop->next;
+ ce->prop->next = NULL;
+ break;
+ }
+ }
+
+ ret = __of_add_property(ce->np, ce->prop);
+ if (ret) {
+ pr_err("%s: add_property failed @%s/%s\n",
+ __func__, ce->np->full_name,
+ ce->prop->name);
+ break;
+ }
+ break;
+ case OF_RECONFIG_REMOVE_PROPERTY:
+ ret = __of_remove_property(ce->np, ce->prop);
+ if (ret) {
+ pr_err("%s: remove_property failed @%s/%s\n",
+ __func__, ce->np->full_name,
+ ce->prop->name);
+ break;
+ }
+ break;
+
+ case OF_RECONFIG_UPDATE_PROPERTY:
+ /* If the property is in deadprops then it must be removed */
+ for (propp = &ce->np->deadprops; *propp; propp = &(*propp)->next) {
+ if (*propp == ce->prop) {
+ *propp = ce->prop->next;
+ ce->prop->next = NULL;
+ break;
+ }
+ }
+
+ ret = __of_update_property(ce->np, ce->prop, &old_prop);
+ if (ret) {
+ pr_err("%s: update_property failed @%s/%s\n",
+ __func__, ce->np->full_name,
+ ce->prop->name);
+ break;
+ }
+ break;
+ default:
+ ret = -EINVAL;
+ }
+ raw_spin_unlock_irqrestore(&devtree_lock, flags);
+
+ if (ret)
+ return ret;
+
+ switch (ce->action) {
+ case OF_RECONFIG_ATTACH_NODE:
+ __of_attach_node_sysfs(ce->np);
+ break;
+ case OF_RECONFIG_DETACH_NODE:
+ __of_detach_node_sysfs(ce->np);
+ break;
+ case OF_RECONFIG_ADD_PROPERTY:
+ /* ignore duplicate names */
+ __of_add_property_sysfs(ce->np, ce->prop);
+ break;
+ case OF_RECONFIG_REMOVE_PROPERTY:
+ __of_remove_property_sysfs(ce->np, ce->prop);
+ break;
+ case OF_RECONFIG_UPDATE_PROPERTY:
+ __of_update_property_sysfs(ce->np, ce->prop, ce->old_prop);
+ break;
+ }
+
+ return 0;
+}
+
+static inline int __of_changeset_entry_revert(struct of_changeset_entry *ce)
+{
+ struct of_changeset_entry ce_inverted;
+
+ __of_changeset_entry_invert(ce, &ce_inverted);
+ return __of_changeset_entry_apply(&ce_inverted);
+}
+
+/**
+ * of_changeset_init - Initialize a changeset for use
+ *
+ * @ocs: changeset pointer
+ *
+ * Initialize a changeset structure
+ */
+void of_changeset_init(struct of_changeset *ocs)
+{
+ memset(ocs, 0, sizeof(*ocs));
+ INIT_LIST_HEAD(&ocs->entries);
+}
+
+/**
+ * of_changeset_destroy - Destroy a changeset
+ *
+ * @ocs: changeset pointer
+ *
+ * Destroys a changeset. Note that if a changeset is applied,
+ * its changes to the tree cannot be reverted.
+ */
+void of_changeset_destroy(struct of_changeset *ocs)
+{
+ struct of_changeset_entry *ce, *cen;
+
+ list_for_each_entry_safe_reverse(ce, cen, &ocs->entries, node)
+ __of_changeset_entry_destroy(ce);
+}
+
+/**
+ * of_changeset_apply - Applies a changeset
+ *
+ * @ocs: changeset pointer
+ *
+ * Applies a changeset to the live tree.
+ * Any side-effects of live tree state changes are applied here on
+ * sucess, like creation/destruction of devices and side-effects
+ * like creation of sysfs properties and directories.
+ * Returns 0 on success, a negative error value in case of an error.
+ * On error the partially applied effects are reverted.
+ */
+int of_changeset_apply(struct of_changeset *ocs)
+{
+ struct of_changeset_entry *ce;
+ int ret;
+
+ /* perform the rest of the work */
+ pr_debug("of_changeset: applying...\n");
+ list_for_each_entry(ce, &ocs->entries, node) {
+ ret = __of_changeset_entry_apply(ce);
+ if (ret) {
+ pr_err("%s: Error applying changeset (%d)\n", __func__, ret);
+ list_for_each_entry_continue_reverse(ce, &ocs->entries, node)
+ __of_changeset_entry_revert(ce);
+ return ret;
+ }
+ }
+ pr_debug("of_changeset: applied, emitting notifiers.\n");
+
+ /* drop the global lock while emitting notifiers */
+ mutex_unlock(&of_mutex);
+ list_for_each_entry(ce, &ocs->entries, node)
+ __of_changeset_entry_notify(ce, 0);
+ mutex_lock(&of_mutex);
+ pr_debug("of_changeset: notifiers sent.\n");
+
+ return 0;
+}
+
+/**
+ * of_changeset_revert - Reverts an applied changeset
+ *
+ * @ocs: changeset pointer
+ *
+ * Reverts a changeset returning the state of the tree to what it
+ * was before the application.
+ * Any side-effects like creation/destruction of devices and
+ * removal of sysfs properties and directories are applied.
+ * Returns 0 on success, a negative error value in case of an error.
+ */
+int of_changeset_revert(struct of_changeset *ocs)
+{
+ struct of_changeset_entry *ce;
+ int ret;
+
+ pr_debug("of_changeset: reverting...\n");
+ list_for_each_entry_reverse(ce, &ocs->entries, node) {
+ ret = __of_changeset_entry_revert(ce);
+ if (ret) {
+ pr_err("%s: Error reverting changeset (%d)\n", __func__, ret);
+ list_for_each_entry_continue(ce, &ocs->entries, node)
+ __of_changeset_entry_apply(ce);
+ return ret;
+ }
+ }
+ pr_debug("of_changeset: reverted, emitting notifiers.\n");
+
+ /* drop the global lock while emitting notifiers */
+ mutex_unlock(&of_mutex);
+ list_for_each_entry_reverse(ce, &ocs->entries, node)
+ __of_changeset_entry_notify(ce, 1);
+ mutex_lock(&of_mutex);
+ pr_debug("of_changeset: notifiers sent.\n");
+
+ return 0;
+}
+
+/**
+ * of_changeset_action - Perform a changeset action
+ *
+ * @ocs: changeset pointer
+ * @action: action to perform
+ * @np: Pointer to device node
+ * @prop: Pointer to property
+ *
+ * On action being one of:
+ * + OF_RECONFIG_ATTACH_NODE
+ * + OF_RECONFIG_DETACH_NODE,
+ * + OF_RECONFIG_ADD_PROPERTY
+ * + OF_RECONFIG_REMOVE_PROPERTY,
+ * + OF_RECONFIG_UPDATE_PROPERTY
+ * Returns 0 on success, a negative error value in case of an error.
+ */
+int of_changeset_action(struct of_changeset *ocs, unsigned long action,
+ struct device_node *np, struct property *prop)
+{
+ struct of_changeset_entry *ce;
+
+ ce = kzalloc(sizeof(*ce), GFP_KERNEL);
+ if (!ce) {
+ pr_err("%s: Failed to allocate\n", __func__);
+ return -ENOMEM;
+ }
+ /* get a reference to the node */
+ ce->action = action;
+ ce->np = of_node_get(np);
+ ce->prop = prop;
+
+ if (action == OF_RECONFIG_UPDATE_PROPERTY && prop)
+ ce->old_prop = of_find_property(np, prop->name, NULL);
+
+ /* add it to the list */
+ list_add_tail(&ce->node, &ocs->entries);
+ return 0;
+}
base = dt_mem_next_cell(dt_root_addr_cells, &prop);
size = dt_mem_next_cell(dt_root_size_cells, &prop);
- if (base && size &&
+ if (size &&
early_init_dt_reserve_memory_arch(base, size, nomap) == 0)
pr_debug("Reserved memory: reserved region for node '%s': base %pa, size %ld MiB\n",
uname, &base, (unsigned long)size / SZ_1M);
}
#ifdef CONFIG_HAVE_MEMBLOCK
+#define MAX_PHYS_ADDR ((phys_addr_t)~0)
+
void __init __weak early_init_dt_add_memory_arch(u64 base, u64 size)
{
const u64 phys_offset = __pa(PAGE_OFFSET);
base &= PAGE_MASK;
size &= PAGE_MASK;
- if (sizeof(phys_addr_t) < sizeof(u64)) {
- if (base > ULONG_MAX) {
- pr_warning("Ignoring memory block 0x%llx - 0x%llx\n",
- base, base + size);
- return;
- }
+ if (base > MAX_PHYS_ADDR) {
+ pr_warning("Ignoring memory block 0x%llx - 0x%llx\n",
+ base, base + size);
+ return;
+ }
- if (base + size > ULONG_MAX) {
- pr_warning("Ignoring memory range 0x%lx - 0x%llx\n",
- ULONG_MAX, base + size);
- size = ULONG_MAX - base;
- }
+ if (base + size > MAX_PHYS_ADDR) {
+ pr_warning("Ignoring memory range 0x%lx - 0x%llx\n",
+ ULONG_MAX, base + size);
+ size = MAX_PHYS_ADDR - base;
}
if (base + size < phys_offset) {
/* Get the reg property (if any) */
addr = of_get_property(device, "reg", NULL);
+ /* Try the new-style interrupts-extended first */
+ res = of_parse_phandle_with_args(device, "interrupts-extended",
+ "#interrupt-cells", index, out_irq);
+ if (!res)
+ return of_irq_parse_raw(addr, out_irq);
+
/* Get the interrupts property */
intspec = of_get_property(device, "interrupts", &intlen);
- if (intspec == NULL) {
- /* Try the new-style interrupts-extended */
- res = of_parse_phandle_with_args(device, "interrupts-extended",
- "#interrupt-cells", index, out_irq);
- if (res)
- return -EINVAL;
- return of_irq_parse_raw(addr, out_irq);
- }
+ if (intspec == NULL)
+ return -EINVAL;
+
intlen /= sizeof(*intspec);
pr_debug(" intspec=%d intlen=%d\n", be32_to_cpup(intspec), intlen);
char stem[0];
};
-extern struct mutex of_aliases_mutex;
+extern struct mutex of_mutex;
extern struct list_head aliases_lookup;
+extern struct kset *of_kset;
+
+
+static inline struct device_node *kobj_to_device_node(struct kobject *kobj)
+{
+ return container_of(kobj, struct device_node, kobj);
+}
+
+#if defined(CONFIG_OF_DYNAMIC)
+extern int of_property_notify(int action, struct device_node *np,
+ struct property *prop, struct property *old_prop);
+extern void of_node_release(struct kobject *kobj);
+#else /* CONFIG_OF_DYNAMIC */
+static inline int of_property_notify(int action, struct device_node *np,
+ struct property *prop, struct property *old_prop)
+{
+ return 0;
+}
+#endif /* CONFIG_OF_DYNAMIC */
+
+/**
+ * General utilities for working with live trees.
+ *
+ * All functions with two leading underscores operate
+ * without taking node references, so you either have to
+ * own the devtree lock or work on detached trees only.
+ */
+struct property *__of_prop_dup(const struct property *prop, gfp_t allocflags);
+struct device_node *__of_node_alloc(const char *full_name, gfp_t allocflags);
+
+extern const void *__of_get_property(const struct device_node *np,
+ const char *name, int *lenp);
+extern int __of_add_property(struct device_node *np, struct property *prop);
+extern int __of_add_property_sysfs(struct device_node *np,
+ struct property *prop);
+extern int __of_remove_property(struct device_node *np, struct property *prop);
+extern void __of_remove_property_sysfs(struct device_node *np,
+ struct property *prop);
+extern int __of_update_property(struct device_node *np,
+ struct property *newprop, struct property **oldprop);
+extern void __of_update_property_sysfs(struct device_node *np,
+ struct property *newprop, struct property *oldprop);
+
+extern void __of_attach_node(struct device_node *np);
+extern int __of_attach_node_sysfs(struct device_node *np);
+extern void __of_detach_node(struct device_node *np);
+extern void __of_detach_node_sysfs(struct device_node *np);
+
+/* iterators for transactions, used for overlays */
+/* forward iterator */
+#define for_each_transaction_entry(_oft, _te) \
+ list_for_each_entry(_te, &(_oft)->te_list, node)
+
+/* reverse iterator */
+#define for_each_transaction_entry_reverse(_oft, _te) \
+ list_for_each_entry_reverse(_te, &(_oft)->te_list, node)
+
#endif /* _LINUX_OF_PRIVATE_H */
for (i = 0; i < reserved_mem_count; i++) {
struct reserved_mem *rmem = &reserved_mem[i];
unsigned long node = rmem->fdt_node;
+ int len;
+ const __be32 *prop;
int err = 0;
+ prop = of_get_flat_dt_prop(node, "phandle", &len);
+ if (!prop)
+ prop = of_get_flat_dt_prop(node, "linux,phandle", &len);
+ if (prop)
+ rmem->phandle = of_read_number(prop, len/4);
+
if (rmem->size == 0)
err = __reserved_mem_alloc_size(node, rmem->name,
&rmem->base, &rmem->size);
__reserved_mem_init_node(rmem);
}
}
+
+static inline struct reserved_mem *__find_rmem(struct device_node *node)
+{
+ unsigned int i;
+
+ if (!node->phandle)
+ return NULL;
+
+ for (i = 0; i < reserved_mem_count; i++)
+ if (reserved_mem[i].phandle == node->phandle)
+ return &reserved_mem[i];
+ return NULL;
+}
+
+/**
+ * of_reserved_mem_device_init() - assign reserved memory region to given device
+ *
+ * This function assign memory region pointed by "memory-region" device tree
+ * property to the given device.
+ */
+void of_reserved_mem_device_init(struct device *dev)
+{
+ struct reserved_mem *rmem;
+ struct device_node *np;
+
+ np = of_parse_phandle(dev->of_node, "memory-region", 0);
+ if (!np)
+ return;
+
+ rmem = __find_rmem(np);
+ of_node_put(np);
+
+ if (!rmem || !rmem->ops || !rmem->ops->device_init)
+ return;
+
+ rmem->ops->device_init(rmem, dev);
+ dev_info(dev, "assigned reserved memory node %s\n", rmem->name);
+}
+
+/**
+ * of_reserved_mem_device_release() - release reserved memory device structures
+ *
+ * This function releases structures allocated for memory region handling for
+ * the given device.
+ */
+void of_reserved_mem_device_release(struct device *dev)
+{
+ struct reserved_mem *rmem;
+ struct device_node *np;
+
+ np = of_parse_phandle(dev->of_node, "memory-region", 0);
+ if (!np)
+ return;
+
+ rmem = __find_rmem(np);
+ of_node_put(np);
+
+ if (!rmem || !rmem->ops || !rmem->ops->device_release)
+ return;
+
+ rmem->ops->device_release(rmem, dev);
+}
break;
}
}
+ of_node_set_flag(bus, OF_POPULATED_BUS);
return rc;
}
static int of_platform_device_destroy(struct device *dev, void *data)
{
- bool *children_left = data;
-
/* Do not touch devices not populated from the device tree */
- if (!dev->of_node || !of_node_check_flag(dev->of_node, OF_POPULATED)) {
- *children_left = true;
+ if (!dev->of_node || !of_node_check_flag(dev->of_node, OF_POPULATED))
return 0;
- }
- /* Recurse, but don't touch this device if it has any children left */
- if (of_platform_depopulate(dev) != 0) {
- *children_left = true;
- return 0;
- }
+ /* Recurse for any nodes that were treated as busses */
+ if (of_node_check_flag(dev->of_node, OF_POPULATED_BUS))
+ device_for_each_child(dev, NULL, of_platform_device_destroy);
if (dev->bus == &platform_bus_type)
platform_device_unregister(to_platform_device(dev));
else if (dev->bus == &amba_bustype)
amba_device_unregister(to_amba_device(dev));
#endif
- else {
- *children_left = true;
- return 0;
- }
of_node_clear_flag(dev->of_node, OF_POPULATED);
-
+ of_node_clear_flag(dev->of_node, OF_POPULATED_BUS);
return 0;
}
/**
* of_platform_depopulate() - Remove devices populated from device tree
- * @parent: device which childred will be removed
+ * @parent: device which children will be removed
*
* Complementary to of_platform_populate(), this function removes children
* of the given device (and, recurrently, their children) that have been
* Returns 0 when all children devices have been removed or
* -EBUSY when some children remained.
*/
-int of_platform_depopulate(struct device *parent)
+void of_platform_depopulate(struct device *parent)
{
- bool children_left = false;
-
- device_for_each_child(parent, &children_left,
- of_platform_device_destroy);
-
- return children_left ? -EBUSY : 0;
+ device_for_each_child(parent, NULL, of_platform_device_destroy);
}
EXPORT_SYMBOL_GPL(of_platform_depopulate);
#include <linux/errno.h>
#include <linux/module.h>
#include <linux/of.h>
+#include <linux/of_fdt.h>
#include <linux/of_irq.h>
#include <linux/of_platform.h>
#include <linux/list.h>
#include <linux/slab.h>
#include <linux/device.h>
+#include "of_private.h"
+
static struct selftest_results {
int passed;
int failed;
} selftest_results;
+#define NO_OF_NODES 2
+static struct device_node *nodes[NO_OF_NODES];
+static int last_node_index;
+static bool selftest_live_tree;
+
#define selftest(result, fmt, ...) { \
if (!(result)) { \
selftest_results.failed++; \
selftest(rc == -EILSEQ, "unterminated string; rc=%i", rc);
}
+#define propcmp(p1, p2) (((p1)->length == (p2)->length) && \
+ (p1)->value && (p2)->value && \
+ !memcmp((p1)->value, (p2)->value, (p1)->length) && \
+ !strcmp((p1)->name, (p2)->name))
+static void __init of_selftest_property_copy(void)
+{
+#ifdef CONFIG_OF_DYNAMIC
+ struct property p1 = { .name = "p1", .length = 0, .value = "" };
+ struct property p2 = { .name = "p2", .length = 5, .value = "abcd" };
+ struct property *new;
+
+ new = __of_prop_dup(&p1, GFP_KERNEL);
+ selftest(new && propcmp(&p1, new), "empty property didn't copy correctly\n");
+ kfree(new->value);
+ kfree(new->name);
+ kfree(new);
+
+ new = __of_prop_dup(&p2, GFP_KERNEL);
+ selftest(new && propcmp(&p2, new), "non-empty property didn't copy correctly\n");
+ kfree(new->value);
+ kfree(new->name);
+ kfree(new);
+#endif
+}
+
+static void __init of_selftest_changeset(void)
+{
+#ifdef CONFIG_OF_DYNAMIC
+ struct property *ppadd, padd = { .name = "prop-add", .length = 0, .value = "" };
+ struct property *ppupdate, pupdate = { .name = "prop-update", .length = 5, .value = "abcd" };
+ struct property *ppremove;
+ struct device_node *n1, *n2, *n21, *nremove, *parent;
+ struct of_changeset chgset;
+
+ of_changeset_init(&chgset);
+ n1 = __of_node_alloc("/testcase-data/changeset/n1", GFP_KERNEL);
+ selftest(n1, "testcase setup failure\n");
+ n2 = __of_node_alloc("/testcase-data/changeset/n2", GFP_KERNEL);
+ selftest(n2, "testcase setup failure\n");
+ n21 = __of_node_alloc("/testcase-data/changeset/n2/n21", GFP_KERNEL);
+ selftest(n21, "testcase setup failure %p\n", n21);
+ nremove = of_find_node_by_path("/testcase-data/changeset/node-remove");
+ selftest(nremove, "testcase setup failure\n");
+ ppadd = __of_prop_dup(&padd, GFP_KERNEL);
+ selftest(ppadd, "testcase setup failure\n");
+ ppupdate = __of_prop_dup(&pupdate, GFP_KERNEL);
+ selftest(ppupdate, "testcase setup failure\n");
+ parent = nremove->parent;
+ n1->parent = parent;
+ n2->parent = parent;
+ n21->parent = n2;
+ n2->child = n21;
+ ppremove = of_find_property(parent, "prop-remove", NULL);
+ selftest(ppremove, "failed to find removal prop");
+
+ of_changeset_init(&chgset);
+ selftest(!of_changeset_attach_node(&chgset, n1), "fail attach n1\n");
+ selftest(!of_changeset_attach_node(&chgset, n2), "fail attach n2\n");
+ selftest(!of_changeset_detach_node(&chgset, nremove), "fail remove node\n");
+ selftest(!of_changeset_attach_node(&chgset, n21), "fail attach n21\n");
+ selftest(!of_changeset_add_property(&chgset, parent, ppadd), "fail add prop\n");
+ selftest(!of_changeset_update_property(&chgset, parent, ppupdate), "fail update prop\n");
+ selftest(!of_changeset_remove_property(&chgset, parent, ppremove), "fail remove prop\n");
+ mutex_lock(&of_mutex);
+ selftest(!of_changeset_apply(&chgset), "apply failed\n");
+ mutex_unlock(&of_mutex);
+
+ mutex_lock(&of_mutex);
+ selftest(!of_changeset_revert(&chgset), "revert failed\n");
+ mutex_unlock(&of_mutex);
+
+ of_changeset_destroy(&chgset);
+#endif
+}
+
static void __init of_selftest_parse_interrupts(void)
{
struct device_node *np;
}
}
+/**
+ * update_node_properties - adds the properties
+ * of np into dup node (present in live tree) and
+ * updates parent of children of np to dup.
+ *
+ * @np: node already present in live tree
+ * @dup: node present in live tree to be updated
+ */
+static void update_node_properties(struct device_node *np,
+ struct device_node *dup)
+{
+ struct property *prop;
+ struct device_node *child;
+
+ for_each_property_of_node(np, prop)
+ of_add_property(dup, prop);
+
+ for_each_child_of_node(np, child)
+ child->parent = dup;
+}
+
+/**
+ * attach_node_and_children - attaches nodes
+ * and its children to live tree
+ *
+ * @np: Node to attach to live tree
+ */
+static int attach_node_and_children(struct device_node *np)
+{
+ struct device_node *next, *root = np, *dup;
+
+ /* skip root node */
+ np = np->child;
+ /* storing a copy in temporary node */
+ dup = np;
+
+ while (dup) {
+ nodes[last_node_index++] = dup;
+ dup = dup->sibling;
+ }
+ dup = NULL;
+
+ while (np) {
+ next = np->allnext;
+ dup = of_find_node_by_path(np->full_name);
+ if (dup)
+ update_node_properties(np, dup);
+ else {
+ np->child = NULL;
+ if (np->parent == root)
+ np->parent = of_allnodes;
+ of_attach_node(np);
+ }
+ np = next;
+ }
+
+ return 0;
+}
+
+/**
+ * selftest_data_add - Reads, copies data from
+ * linked tree and attaches it to the live tree
+ */
+static int __init selftest_data_add(void)
+{
+ void *selftest_data;
+ struct device_node *selftest_data_node, *np;
+ extern uint8_t __dtb_testcases_begin[];
+ extern uint8_t __dtb_testcases_end[];
+ const int size = __dtb_testcases_end - __dtb_testcases_begin;
+
+ if (!size) {
+ pr_warn("%s: No testcase data to attach; not running tests\n",
+ __func__);
+ return -ENODATA;
+ }
+
+ /* creating copy */
+ selftest_data = kmemdup(__dtb_testcases_begin, size, GFP_KERNEL);
+
+ if (!selftest_data) {
+ pr_warn("%s: Failed to allocate memory for selftest_data; "
+ "not running tests\n", __func__);
+ return -ENOMEM;
+ }
+ of_fdt_unflatten_tree(selftest_data, &selftest_data_node);
+ if (!selftest_data_node) {
+ pr_warn("%s: No tree to attach; not running tests\n", __func__);
+ return -ENODATA;
+ }
+
+ if (!of_allnodes) {
+ /* enabling flag for removing nodes */
+ selftest_live_tree = true;
+ of_allnodes = selftest_data_node;
+
+ for_each_of_allnodes(np)
+ __of_attach_node_sysfs(np);
+ of_aliases = of_find_node_by_path("/aliases");
+ of_chosen = of_find_node_by_path("/chosen");
+ return 0;
+ }
+
+ /* attach the sub-tree to live tree */
+ return attach_node_and_children(selftest_data_node);
+}
+
+/**
+ * detach_node_and_children - detaches node
+ * and its children from live tree
+ *
+ * @np: Node to detach from live tree
+ */
+static void detach_node_and_children(struct device_node *np)
+{
+ while (np->child)
+ detach_node_and_children(np->child);
+
+ while (np->sibling)
+ detach_node_and_children(np->sibling);
+
+ of_detach_node(np);
+}
+
+/**
+ * selftest_data_remove - removes the selftest data
+ * nodes from the live tree
+ */
+static void selftest_data_remove(void)
+{
+ struct device_node *np;
+ struct property *prop;
+
+ if (selftest_live_tree) {
+ of_node_put(of_aliases);
+ of_node_put(of_chosen);
+ of_aliases = NULL;
+ of_chosen = NULL;
+ for_each_child_of_node(of_allnodes, np)
+ detach_node_and_children(np);
+ __of_detach_node_sysfs(of_allnodes);
+ of_allnodes = NULL;
+ return;
+ }
+
+ while (last_node_index >= 0) {
+ if (nodes[last_node_index]) {
+ np = of_find_node_by_path(nodes[last_node_index]->full_name);
+ if (strcmp(np->full_name, "/aliases") != 0) {
+ detach_node_and_children(np->child);
+ of_detach_node(np);
+ } else {
+ for_each_property_of_node(np, prop) {
+ if (strcmp(prop->name, "testcase-alias") == 0)
+ of_remove_property(np, prop);
+ }
+ }
+ }
+ last_node_index--;
+ }
+}
+
static int __init of_selftest(void)
{
struct device_node *np;
+ int res;
+
+ /* adding data for selftest */
+ res = selftest_data_add();
+ if (res)
+ return res;
np = of_find_node_by_path("/testcase-data/phandle-tests/consumer-a");
if (!np) {
of_selftest_dynamic();
of_selftest_parse_phandle_with_args();
of_selftest_property_match_string();
+ of_selftest_property_copy();
+ of_selftest_changeset();
of_selftest_parse_interrupts();
of_selftest_parse_interrupts_extended();
of_selftest_match_node();
of_selftest_platform_populate();
pr_info("end of selftest - %i passed, %i failed\n",
selftest_results.passed, selftest_results.failed);
+
+ /* removing selftest data from live tree */
+ selftest_data_remove();
+
return 0;
}
late_initcall(of_selftest);
--- /dev/null
+/dts-v1/;
+/ {
+ testcase-data {
+ changeset {
+ prop-update = "hello";
+ prop-remove = "world";
+ node-remove {
+ };
+ };
+ };
+};
+#include "tests-phandle.dtsi"
+#include "tests-interrupts.dtsi"
+#include "tests-match.dtsi"
+#include "tests-platform.dtsi"
+++ /dev/null
-#include "tests-phandle.dtsi"
-#include "tests-interrupts.dtsi"
-#include "tests-match.dtsi"
-#include "tests-platform.dtsi"
menu "PCI host controller drivers"
depends on PCI
+config PCI_DRA7XX
+ bool "TI DRA7xx PCIe controller"
+ select PCIE_DW
+ depends on OF && HAS_IOMEM && TI_PIPE3
+ help
+ Enables support for the PCIe controller in the DRA7xx SoC. There
+ are two instances of PCIe controller in DRA7xx. This controller can
+ act both as EP and RC. This reuses the Designware core.
+
config PCI_MVEBU
bool "Marvell EBU PCIe controller"
- depends on ARCH_MVEBU || ARCH_DOVE || ARCH_KIRKWOOD
+ depends on ARCH_MVEBU || ARCH_DOVE
depends on OF
config PCIE_DW
obj-$(CONFIG_PCIE_DW) += pcie-designware.o
+obj-$(CONFIG_PCI_DRA7XX) += pci-dra7xx.o
obj-$(CONFIG_PCI_EXYNOS) += pci-exynos.o
obj-$(CONFIG_PCI_IMX6) += pci-imx6.o
obj-$(CONFIG_PCI_MVEBU) += pci-mvebu.o
--- /dev/null
+/*
+ * pcie-dra7xx - PCIe controller driver for TI DRA7xx SoCs
+ *
+ * Copyright (C) 2013-2014 Texas Instruments Incorporated - http://www.ti.com
+ *
+ * Authors: Kishon Vijay Abraham I <kishon@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
+#include <linux/irqdomain.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/phy/phy.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/resource.h>
+#include <linux/types.h>
+
+#include "pcie-designware.h"
+
+/* PCIe controller wrapper DRA7XX configuration registers */
+
+#define PCIECTRL_DRA7XX_CONF_IRQSTATUS_MAIN 0x0024
+#define PCIECTRL_DRA7XX_CONF_IRQENABLE_SET_MAIN 0x0028
+#define ERR_SYS BIT(0)
+#define ERR_FATAL BIT(1)
+#define ERR_NONFATAL BIT(2)
+#define ERR_COR BIT(3)
+#define ERR_AXI BIT(4)
+#define ERR_ECRC BIT(5)
+#define PME_TURN_OFF BIT(8)
+#define PME_TO_ACK BIT(9)
+#define PM_PME BIT(10)
+#define LINK_REQ_RST BIT(11)
+#define LINK_UP_EVT BIT(12)
+#define CFG_BME_EVT BIT(13)
+#define CFG_MSE_EVT BIT(14)
+#define INTERRUPTS (ERR_SYS | ERR_FATAL | ERR_NONFATAL | ERR_COR | ERR_AXI | \
+ ERR_ECRC | PME_TURN_OFF | PME_TO_ACK | PM_PME | \
+ LINK_REQ_RST | LINK_UP_EVT | CFG_BME_EVT | CFG_MSE_EVT)
+
+#define PCIECTRL_DRA7XX_CONF_IRQSTATUS_MSI 0x0034
+#define PCIECTRL_DRA7XX_CONF_IRQENABLE_SET_MSI 0x0038
+#define INTA BIT(0)
+#define INTB BIT(1)
+#define INTC BIT(2)
+#define INTD BIT(3)
+#define MSI BIT(4)
+#define LEG_EP_INTERRUPTS (INTA | INTB | INTC | INTD)
+
+#define PCIECTRL_DRA7XX_CONF_DEVICE_CMD 0x0104
+#define LTSSM_EN 0x1
+
+#define PCIECTRL_DRA7XX_CONF_PHY_CS 0x010C
+#define LINK_UP BIT(16)
+
+struct dra7xx_pcie {
+ void __iomem *base;
+ struct phy **phy;
+ int phy_count;
+ struct device *dev;
+ struct pcie_port pp;
+};
+
+#define to_dra7xx_pcie(x) container_of((x), struct dra7xx_pcie, pp)
+
+static inline u32 dra7xx_pcie_readl(struct dra7xx_pcie *pcie, u32 offset)
+{
+ return readl(pcie->base + offset);
+}
+
+static inline void dra7xx_pcie_writel(struct dra7xx_pcie *pcie, u32 offset,
+ u32 value)
+{
+ writel(value, pcie->base + offset);
+}
+
+static int dra7xx_pcie_link_up(struct pcie_port *pp)
+{
+ struct dra7xx_pcie *dra7xx = to_dra7xx_pcie(pp);
+ u32 reg = dra7xx_pcie_readl(dra7xx, PCIECTRL_DRA7XX_CONF_PHY_CS);
+
+ return !!(reg & LINK_UP);
+}
+
+static int dra7xx_pcie_establish_link(struct pcie_port *pp)
+{
+ u32 reg;
+ unsigned int retries = 1000;
+ struct dra7xx_pcie *dra7xx = to_dra7xx_pcie(pp);
+
+ if (dw_pcie_link_up(pp)) {
+ dev_err(pp->dev, "link is already up\n");
+ return 0;
+ }
+
+ reg = dra7xx_pcie_readl(dra7xx, PCIECTRL_DRA7XX_CONF_DEVICE_CMD);
+ reg |= LTSSM_EN;
+ dra7xx_pcie_writel(dra7xx, PCIECTRL_DRA7XX_CONF_DEVICE_CMD, reg);
+
+ while (retries--) {
+ reg = dra7xx_pcie_readl(dra7xx, PCIECTRL_DRA7XX_CONF_PHY_CS);
+ if (reg & LINK_UP)
+ break;
+ usleep_range(10, 20);
+ }
+
+ if (retries == 0) {
+ dev_err(pp->dev, "link is not up\n");
+ return -ETIMEDOUT;
+ }
+
+ return 0;
+}
+
+static void dra7xx_pcie_enable_interrupts(struct pcie_port *pp)
+{
+ struct dra7xx_pcie *dra7xx = to_dra7xx_pcie(pp);
+
+ dra7xx_pcie_writel(dra7xx, PCIECTRL_DRA7XX_CONF_IRQSTATUS_MAIN,
+ ~INTERRUPTS);
+ dra7xx_pcie_writel(dra7xx,
+ PCIECTRL_DRA7XX_CONF_IRQENABLE_SET_MAIN, INTERRUPTS);
+ dra7xx_pcie_writel(dra7xx, PCIECTRL_DRA7XX_CONF_IRQSTATUS_MSI,
+ ~LEG_EP_INTERRUPTS & ~MSI);
+
+ if (IS_ENABLED(CONFIG_PCI_MSI))
+ dra7xx_pcie_writel(dra7xx,
+ PCIECTRL_DRA7XX_CONF_IRQENABLE_SET_MSI, MSI);
+ else
+ dra7xx_pcie_writel(dra7xx,
+ PCIECTRL_DRA7XX_CONF_IRQENABLE_SET_MSI,
+ LEG_EP_INTERRUPTS);
+}
+
+static void dra7xx_pcie_host_init(struct pcie_port *pp)
+{
+ dw_pcie_setup_rc(pp);
+ dra7xx_pcie_establish_link(pp);
+ if (IS_ENABLED(CONFIG_PCI_MSI))
+ dw_pcie_msi_init(pp);
+ dra7xx_pcie_enable_interrupts(pp);
+}
+
+static struct pcie_host_ops dra7xx_pcie_host_ops = {
+ .link_up = dra7xx_pcie_link_up,
+ .host_init = dra7xx_pcie_host_init,
+};
+
+static int dra7xx_pcie_intx_map(struct irq_domain *domain, unsigned int irq,
+ irq_hw_number_t hwirq)
+{
+ irq_set_chip_and_handler(irq, &dummy_irq_chip, handle_simple_irq);
+ irq_set_chip_data(irq, domain->host_data);
+ set_irq_flags(irq, IRQF_VALID);
+
+ return 0;
+}
+
+static const struct irq_domain_ops intx_domain_ops = {
+ .map = dra7xx_pcie_intx_map,
+};
+
+static int dra7xx_pcie_init_irq_domain(struct pcie_port *pp)
+{
+ struct device *dev = pp->dev;
+ struct device_node *node = dev->of_node;
+ struct device_node *pcie_intc_node = of_get_next_child(node, NULL);
+
+ if (!pcie_intc_node) {
+ dev_err(dev, "No PCIe Intc node found\n");
+ return PTR_ERR(pcie_intc_node);
+ }
+
+ pp->irq_domain = irq_domain_add_linear(pcie_intc_node, 4,
+ &intx_domain_ops, pp);
+ if (!pp->irq_domain) {
+ dev_err(dev, "Failed to get a INTx IRQ domain\n");
+ return PTR_ERR(pp->irq_domain);
+ }
+
+ return 0;
+}
+
+static irqreturn_t dra7xx_pcie_msi_irq_handler(int irq, void *arg)
+{
+ struct pcie_port *pp = arg;
+ struct dra7xx_pcie *dra7xx = to_dra7xx_pcie(pp);
+ u32 reg;
+
+ reg = dra7xx_pcie_readl(dra7xx, PCIECTRL_DRA7XX_CONF_IRQSTATUS_MSI);
+
+ switch (reg) {
+ case MSI:
+ dw_handle_msi_irq(pp);
+ break;
+ case INTA:
+ case INTB:
+ case INTC:
+ case INTD:
+ generic_handle_irq(irq_find_mapping(pp->irq_domain, ffs(reg)));
+ break;
+ }
+
+ dra7xx_pcie_writel(dra7xx, PCIECTRL_DRA7XX_CONF_IRQSTATUS_MSI, reg);
+
+ return IRQ_HANDLED;
+}
+
+
+static irqreturn_t dra7xx_pcie_irq_handler(int irq, void *arg)
+{
+ struct dra7xx_pcie *dra7xx = arg;
+ u32 reg;
+
+ reg = dra7xx_pcie_readl(dra7xx, PCIECTRL_DRA7XX_CONF_IRQSTATUS_MAIN);
+
+ if (reg & ERR_SYS)
+ dev_dbg(dra7xx->dev, "System Error\n");
+
+ if (reg & ERR_FATAL)
+ dev_dbg(dra7xx->dev, "Fatal Error\n");
+
+ if (reg & ERR_NONFATAL)
+ dev_dbg(dra7xx->dev, "Non Fatal Error\n");
+
+ if (reg & ERR_COR)
+ dev_dbg(dra7xx->dev, "Correctable Error\n");
+
+ if (reg & ERR_AXI)
+ dev_dbg(dra7xx->dev, "AXI tag lookup fatal Error\n");
+
+ if (reg & ERR_ECRC)
+ dev_dbg(dra7xx->dev, "ECRC Error\n");
+
+ if (reg & PME_TURN_OFF)
+ dev_dbg(dra7xx->dev,
+ "Power Management Event Turn-Off message received\n");
+
+ if (reg & PME_TO_ACK)
+ dev_dbg(dra7xx->dev,
+ "Power Management Turn-Off Ack message received\n");
+
+ if (reg & PM_PME)
+ dev_dbg(dra7xx->dev,
+ "PM Power Management Event message received\n");
+
+ if (reg & LINK_REQ_RST)
+ dev_dbg(dra7xx->dev, "Link Request Reset\n");
+
+ if (reg & LINK_UP_EVT)
+ dev_dbg(dra7xx->dev, "Link-up state change\n");
+
+ if (reg & CFG_BME_EVT)
+ dev_dbg(dra7xx->dev, "CFG 'Bus Master Enable' change\n");
+
+ if (reg & CFG_MSE_EVT)
+ dev_dbg(dra7xx->dev, "CFG 'Memory Space Enable' change\n");
+
+ dra7xx_pcie_writel(dra7xx, PCIECTRL_DRA7XX_CONF_IRQSTATUS_MAIN, reg);
+
+ return IRQ_HANDLED;
+}
+
+static int add_pcie_port(struct dra7xx_pcie *dra7xx,
+ struct platform_device *pdev)
+{
+ int ret;
+ struct pcie_port *pp;
+ struct resource *res;
+ struct device *dev = &pdev->dev;
+
+ pp = &dra7xx->pp;
+ pp->dev = dev;
+ pp->ops = &dra7xx_pcie_host_ops;
+
+ pp->irq = platform_get_irq(pdev, 1);
+ if (pp->irq < 0) {
+ dev_err(dev, "missing IRQ resource\n");
+ return -EINVAL;
+ }
+
+ ret = devm_request_irq(&pdev->dev, pp->irq,
+ dra7xx_pcie_msi_irq_handler, IRQF_SHARED,
+ "dra7-pcie-msi", pp);
+ if (ret) {
+ dev_err(&pdev->dev, "failed to request irq\n");
+ return ret;
+ }
+
+ if (!IS_ENABLED(CONFIG_PCI_MSI)) {
+ ret = dra7xx_pcie_init_irq_domain(pp);
+ if (ret < 0)
+ return ret;
+ }
+
+ res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "rc_dbics");
+ pp->dbi_base = devm_ioremap(dev, res->start, resource_size(res));
+ if (!pp->dbi_base)
+ return -ENOMEM;
+
+ ret = dw_pcie_host_init(pp);
+ if (ret) {
+ dev_err(dra7xx->dev, "failed to initialize host\n");
+ return ret;
+ }
+
+ return 0;
+}
+
+static int __init dra7xx_pcie_probe(struct platform_device *pdev)
+{
+ u32 reg;
+ int ret;
+ int irq;
+ int i;
+ int phy_count;
+ struct phy **phy;
+ void __iomem *base;
+ struct resource *res;
+ struct dra7xx_pcie *dra7xx;
+ struct device *dev = &pdev->dev;
+ struct device_node *np = dev->of_node;
+ char name[10];
+
+ dra7xx = devm_kzalloc(dev, sizeof(*dra7xx), GFP_KERNEL);
+ if (!dra7xx)
+ return -ENOMEM;
+
+ irq = platform_get_irq(pdev, 0);
+ if (irq < 0) {
+ dev_err(dev, "missing IRQ resource\n");
+ return -EINVAL;
+ }
+
+ ret = devm_request_irq(dev, irq, dra7xx_pcie_irq_handler,
+ IRQF_SHARED, "dra7xx-pcie-main", dra7xx);
+ if (ret) {
+ dev_err(dev, "failed to request irq\n");
+ return ret;
+ }
+
+ res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "ti_conf");
+ base = devm_ioremap_nocache(dev, res->start, resource_size(res));
+ if (!base)
+ return -ENOMEM;
+
+ phy_count = of_property_count_strings(np, "phy-names");
+ if (phy_count < 0) {
+ dev_err(dev, "unable to find the strings\n");
+ return phy_count;
+ }
+
+ phy = devm_kzalloc(dev, sizeof(*phy) * phy_count, GFP_KERNEL);
+ if (!phy)
+ return -ENOMEM;
+
+ for (i = 0; i < phy_count; i++) {
+ snprintf(name, sizeof(name), "pcie-phy%d", i);
+ phy[i] = devm_phy_get(dev, name);
+ if (IS_ERR(phy[i]))
+ return PTR_ERR(phy[i]);
+
+ ret = phy_init(phy[i]);
+ if (ret < 0)
+ goto err_phy;
+
+ ret = phy_power_on(phy[i]);
+ if (ret < 0) {
+ phy_exit(phy[i]);
+ goto err_phy;
+ }
+ }
+
+ dra7xx->base = base;
+ dra7xx->phy = phy;
+ dra7xx->dev = dev;
+ dra7xx->phy_count = phy_count;
+
+ pm_runtime_enable(dev);
+ ret = pm_runtime_get_sync(dev);
+ if (IS_ERR_VALUE(ret)) {
+ dev_err(dev, "pm_runtime_get_sync failed\n");
+ goto err_phy;
+ }
+
+ reg = dra7xx_pcie_readl(dra7xx, PCIECTRL_DRA7XX_CONF_DEVICE_CMD);
+ reg &= ~LTSSM_EN;
+ dra7xx_pcie_writel(dra7xx, PCIECTRL_DRA7XX_CONF_DEVICE_CMD, reg);
+
+ platform_set_drvdata(pdev, dra7xx);
+
+ ret = add_pcie_port(dra7xx, pdev);
+ if (ret < 0)
+ goto err_add_port;
+
+ return 0;
+
+err_add_port:
+ pm_runtime_put(dev);
+ pm_runtime_disable(dev);
+
+err_phy:
+ while (--i >= 0) {
+ phy_power_off(phy[i]);
+ phy_exit(phy[i]);
+ }
+
+ return ret;
+}
+
+static int __exit dra7xx_pcie_remove(struct platform_device *pdev)
+{
+ struct dra7xx_pcie *dra7xx = platform_get_drvdata(pdev);
+ struct pcie_port *pp = &dra7xx->pp;
+ struct device *dev = &pdev->dev;
+ int count = dra7xx->phy_count;
+
+ if (pp->irq_domain)
+ irq_domain_remove(pp->irq_domain);
+ pm_runtime_put(dev);
+ pm_runtime_disable(dev);
+ while (count--) {
+ phy_power_off(dra7xx->phy[count]);
+ phy_exit(dra7xx->phy[count]);
+ }
+
+ return 0;
+}
+
+static const struct of_device_id of_dra7xx_pcie_match[] = {
+ { .compatible = "ti,dra7-pcie", },
+ {},
+};
+MODULE_DEVICE_TABLE(of, of_dra7xx_pcie_match);
+
+static struct platform_driver dra7xx_pcie_driver = {
+ .remove = __exit_p(dra7xx_pcie_remove),
+ .driver = {
+ .name = "dra7-pcie",
+ .owner = THIS_MODULE,
+ .of_match_table = of_dra7xx_pcie_match,
+ },
+};
+
+module_platform_driver_probe(dra7xx_pcie_driver, dra7xx_pcie_probe);
+
+MODULE_AUTHOR("Kishon Vijay Abraham I <kishon@ti.com>");
+MODULE_DESCRIPTION("TI PCIe controller driver");
+MODULE_LICENSE("GPL v2");
*/
#include <linux/clk.h>
+#include <linux/debugfs.h>
#include <linux/delay.h>
#include <linux/export.h>
#include <linux/interrupt.h>
unsigned int num_supplies;
const struct tegra_pcie_soc_data *soc_data;
+ struct dentry *debugfs;
};
struct tegra_pcie_port {
};
MODULE_DEVICE_TABLE(of, tegra_pcie_of_match);
+static void *tegra_pcie_ports_seq_start(struct seq_file *s, loff_t *pos)
+{
+ struct tegra_pcie *pcie = s->private;
+
+ if (list_empty(&pcie->ports))
+ return NULL;
+
+ seq_printf(s, "Index Status\n");
+
+ return seq_list_start(&pcie->ports, *pos);
+}
+
+static void *tegra_pcie_ports_seq_next(struct seq_file *s, void *v, loff_t *pos)
+{
+ struct tegra_pcie *pcie = s->private;
+
+ return seq_list_next(v, &pcie->ports, pos);
+}
+
+static void tegra_pcie_ports_seq_stop(struct seq_file *s, void *v)
+{
+}
+
+static int tegra_pcie_ports_seq_show(struct seq_file *s, void *v)
+{
+ bool up = false, active = false;
+ struct tegra_pcie_port *port;
+ unsigned int value;
+
+ port = list_entry(v, struct tegra_pcie_port, list);
+
+ value = readl(port->base + RP_VEND_XP);
+
+ if (value & RP_VEND_XP_DL_UP)
+ up = true;
+
+ value = readl(port->base + RP_LINK_CONTROL_STATUS);
+
+ if (value & RP_LINK_CONTROL_STATUS_DL_LINK_ACTIVE)
+ active = true;
+
+ seq_printf(s, "%2u ", port->index);
+
+ if (up)
+ seq_printf(s, "up");
+
+ if (active) {
+ if (up)
+ seq_printf(s, ", ");
+
+ seq_printf(s, "active");
+ }
+
+ seq_printf(s, "\n");
+ return 0;
+}
+
+static const struct seq_operations tegra_pcie_ports_seq_ops = {
+ .start = tegra_pcie_ports_seq_start,
+ .next = tegra_pcie_ports_seq_next,
+ .stop = tegra_pcie_ports_seq_stop,
+ .show = tegra_pcie_ports_seq_show,
+};
+
+static int tegra_pcie_ports_open(struct inode *inode, struct file *file)
+{
+ struct tegra_pcie *pcie = inode->i_private;
+ struct seq_file *s;
+ int err;
+
+ err = seq_open(file, &tegra_pcie_ports_seq_ops);
+ if (err)
+ return err;
+
+ s = file->private_data;
+ s->private = pcie;
+
+ return 0;
+}
+
+static const struct file_operations tegra_pcie_ports_ops = {
+ .owner = THIS_MODULE,
+ .open = tegra_pcie_ports_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = seq_release,
+};
+
+static int tegra_pcie_debugfs_init(struct tegra_pcie *pcie)
+{
+ struct dentry *file;
+
+ pcie->debugfs = debugfs_create_dir("pcie", NULL);
+ if (!pcie->debugfs)
+ return -ENOMEM;
+
+ file = debugfs_create_file("ports", S_IFREG | S_IRUGO, pcie->debugfs,
+ pcie, &tegra_pcie_ports_ops);
+ if (!file)
+ goto remove;
+
+ return 0;
+
+remove:
+ debugfs_remove_recursive(pcie->debugfs);
+ pcie->debugfs = NULL;
+ return -ENOMEM;
+}
+
static int tegra_pcie_probe(struct platform_device *pdev)
{
const struct of_device_id *match;
goto disable_msi;
}
+ if (IS_ENABLED(CONFIG_DEBUG_FS)) {
+ err = tegra_pcie_debugfs_init(pcie);
+ if (err < 0)
+ dev_err(&pdev->dev, "failed to setup debugfs: %d\n",
+ err);
+ }
+
platform_set_drvdata(pdev, pcie);
return 0;
#include <linux/of_pci.h>
#include <linux/pci.h>
#include <linux/pci_regs.h>
+#include <linux/platform_device.h>
#include <linux/types.h>
#include "pcie-designware.h"
return 0;
}
+static void dw_pcie_msi_clear_irq(struct pcie_port *pp, int irq)
+{
+ unsigned int res, bit, val;
+
+ res = (irq / 32) * 12;
+ bit = irq % 32;
+ dw_pcie_rd_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, &val);
+ val &= ~(1 << bit);
+ dw_pcie_wr_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, val);
+}
+
static void clear_irq_range(struct pcie_port *pp, unsigned int irq_base,
unsigned int nvec, unsigned int pos)
{
- unsigned int i, res, bit, val;
+ unsigned int i;
for (i = 0; i < nvec; i++) {
irq_set_msi_desc_off(irq_base, i, NULL);
clear_bit(pos + i, pp->msi_irq_in_use);
/* Disable corresponding interrupt on MSI controller */
- res = ((pos + i) / 32) * 12;
- bit = (pos + i) % 32;
- dw_pcie_rd_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, &val);
- val &= ~(1 << bit);
- dw_pcie_wr_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, val);
+ if (pp->ops->msi_clear_irq)
+ pp->ops->msi_clear_irq(pp, pos + i);
+ else
+ dw_pcie_msi_clear_irq(pp, pos + i);
}
}
+static void dw_pcie_msi_set_irq(struct pcie_port *pp, int irq)
+{
+ unsigned int res, bit, val;
+
+ res = (irq / 32) * 12;
+ bit = irq % 32;
+ dw_pcie_rd_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, &val);
+ val |= 1 << bit;
+ dw_pcie_wr_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, val);
+}
+
static int assign_irq(int no_irqs, struct msi_desc *desc, int *pos)
{
- int res, bit, irq, pos0, pos1, i;
- u32 val;
+ int irq, pos0, pos1, i;
struct pcie_port *pp = sys_to_pcie(desc->dev->bus->sysdata);
if (!pp) {
}
set_bit(pos0 + i, pp->msi_irq_in_use);
/*Enable corresponding interrupt in MSI interrupt controller */
- res = ((pos0 + i) / 32) * 12;
- bit = (pos0 + i) % 32;
- dw_pcie_rd_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, &val);
- val |= 1 << bit;
- dw_pcie_wr_own_conf(pp, PCIE_MSI_INTR0_ENABLE + res, 4, val);
+ if (pp->ops->msi_set_irq)
+ pp->ops->msi_set_irq(pp, pos0 + i);
+ else
+ dw_pcie_msi_set_irq(pp, pos0 + i);
}
*pos = pos0;
*/
desc->msi_attrib.multiple = msgvec;
- msg.address_lo = virt_to_phys((void *)pp->msi_data);
+ if (pp->ops->get_msi_data)
+ msg.address_lo = pp->ops->get_msi_data(pp);
+ else
+ msg.address_lo = virt_to_phys((void *)pp->msi_data);
msg.address_hi = 0x0;
msg.data = pos;
write_msi_msg(irq, &msg);
int __init dw_pcie_host_init(struct pcie_port *pp)
{
struct device_node *np = pp->dev->of_node;
+ struct platform_device *pdev = to_platform_device(pp->dev);
struct of_pci_range range;
struct of_pci_range_parser parser;
- u32 val;
- int i;
+ struct resource *cfg_res;
+ u32 val, na, ns;
+ const __be32 *addrp;
+ int i, index;
+
+ /* Find the address cell size and the number of cells in order to get
+ * the untranslated address.
+ */
+ of_property_read_u32(np, "#address-cells", &na);
+ ns = of_n_size_cells(np);
+
+ cfg_res = platform_get_resource_byname(pdev, IORESOURCE_MEM, "config");
+ if (cfg_res) {
+ pp->config.cfg0_size = resource_size(cfg_res)/2;
+ pp->config.cfg1_size = resource_size(cfg_res)/2;
+ pp->cfg0_base = cfg_res->start;
+ pp->cfg1_base = cfg_res->start + pp->config.cfg0_size;
+
+ /* Find the untranslated configuration space address */
+ index = of_property_match_string(np, "reg-names", "config");
+ addrp = of_get_address(np, index, false, false);
+ pp->cfg0_mod_base = of_read_number(addrp, ns);
+ pp->cfg1_mod_base = pp->cfg0_mod_base + pp->config.cfg0_size;
+ } else {
+ dev_err(pp->dev, "missing *config* reg space\n");
+ }
if (of_pci_range_parser_init(&parser, np)) {
dev_err(pp->dev, "missing ranges property\n");
pp->config.io_size = resource_size(&pp->io);
pp->config.io_bus_addr = range.pci_addr;
pp->io_base = range.cpu_addr;
+
+ /* Find the untranslated IO space address */
+ pp->io_mod_base = of_read_number(parser.range -
+ parser.np + na, ns);
}
if (restype == IORESOURCE_MEM) {
of_pci_range_to_resource(&range, np, &pp->mem);
pp->mem.name = "MEM";
pp->config.mem_size = resource_size(&pp->mem);
pp->config.mem_bus_addr = range.pci_addr;
+
+ /* Find the untranslated MEM space address */
+ pp->mem_mod_base = of_read_number(parser.range -
+ parser.np + na, ns);
}
if (restype == 0) {
of_pci_range_to_resource(&range, np, &pp->cfg);
pp->config.cfg0_size = resource_size(&pp->cfg)/2;
pp->config.cfg1_size = resource_size(&pp->cfg)/2;
+ pp->cfg0_base = pp->cfg.start;
+ pp->cfg1_base = pp->cfg.start + pp->config.cfg0_size;
+
+ /* Find the untranslated configuration space address */
+ pp->cfg0_mod_base = of_read_number(parser.range -
+ parser.np + na, ns);
+ pp->cfg1_mod_base = pp->cfg0_mod_base +
+ pp->config.cfg0_size;
}
}
}
}
- pp->cfg0_base = pp->cfg.start;
- pp->cfg1_base = pp->cfg.start + pp->config.cfg0_size;
pp->mem_base = pp->mem.start;
pp->va_cfg0_base = devm_ioremap(pp->dev, pp->cfg0_base,
/* Program viewport 0 : OUTBOUND : CFG0 */
dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
PCIE_ATU_VIEWPORT);
- dw_pcie_writel_rc(pp, pp->cfg0_base, PCIE_ATU_LOWER_BASE);
- dw_pcie_writel_rc(pp, (pp->cfg0_base >> 32), PCIE_ATU_UPPER_BASE);
- dw_pcie_writel_rc(pp, pp->cfg0_base + pp->config.cfg0_size - 1,
+ dw_pcie_writel_rc(pp, pp->cfg0_mod_base, PCIE_ATU_LOWER_BASE);
+ dw_pcie_writel_rc(pp, (pp->cfg0_mod_base >> 32), PCIE_ATU_UPPER_BASE);
+ dw_pcie_writel_rc(pp, pp->cfg0_mod_base + pp->config.cfg0_size - 1,
PCIE_ATU_LIMIT);
dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET);
dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET);
dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
PCIE_ATU_VIEWPORT);
dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_CFG1, PCIE_ATU_CR1);
- dw_pcie_writel_rc(pp, pp->cfg1_base, PCIE_ATU_LOWER_BASE);
- dw_pcie_writel_rc(pp, (pp->cfg1_base >> 32), PCIE_ATU_UPPER_BASE);
- dw_pcie_writel_rc(pp, pp->cfg1_base + pp->config.cfg1_size - 1,
+ dw_pcie_writel_rc(pp, pp->cfg1_mod_base, PCIE_ATU_LOWER_BASE);
+ dw_pcie_writel_rc(pp, (pp->cfg1_mod_base >> 32), PCIE_ATU_UPPER_BASE);
+ dw_pcie_writel_rc(pp, pp->cfg1_mod_base + pp->config.cfg1_size - 1,
PCIE_ATU_LIMIT);
dw_pcie_writel_rc(pp, busdev, PCIE_ATU_LOWER_TARGET);
dw_pcie_writel_rc(pp, 0, PCIE_ATU_UPPER_TARGET);
dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX0,
PCIE_ATU_VIEWPORT);
dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_MEM, PCIE_ATU_CR1);
- dw_pcie_writel_rc(pp, pp->mem_base, PCIE_ATU_LOWER_BASE);
- dw_pcie_writel_rc(pp, (pp->mem_base >> 32), PCIE_ATU_UPPER_BASE);
- dw_pcie_writel_rc(pp, pp->mem_base + pp->config.mem_size - 1,
+ dw_pcie_writel_rc(pp, pp->mem_mod_base, PCIE_ATU_LOWER_BASE);
+ dw_pcie_writel_rc(pp, (pp->mem_mod_base >> 32), PCIE_ATU_UPPER_BASE);
+ dw_pcie_writel_rc(pp, pp->mem_mod_base + pp->config.mem_size - 1,
PCIE_ATU_LIMIT);
dw_pcie_writel_rc(pp, pp->config.mem_bus_addr, PCIE_ATU_LOWER_TARGET);
dw_pcie_writel_rc(pp, upper_32_bits(pp->config.mem_bus_addr),
dw_pcie_writel_rc(pp, PCIE_ATU_REGION_OUTBOUND | PCIE_ATU_REGION_INDEX1,
PCIE_ATU_VIEWPORT);
dw_pcie_writel_rc(pp, PCIE_ATU_TYPE_IO, PCIE_ATU_CR1);
- dw_pcie_writel_rc(pp, pp->io_base, PCIE_ATU_LOWER_BASE);
- dw_pcie_writel_rc(pp, (pp->io_base >> 32), PCIE_ATU_UPPER_BASE);
- dw_pcie_writel_rc(pp, pp->io_base + pp->config.io_size - 1,
+ dw_pcie_writel_rc(pp, pp->io_mod_base, PCIE_ATU_LOWER_BASE);
+ dw_pcie_writel_rc(pp, (pp->io_mod_base >> 32), PCIE_ATU_UPPER_BASE);
+ dw_pcie_writel_rc(pp, pp->io_mod_base + pp->config.io_size - 1,
PCIE_ATU_LIMIT);
dw_pcie_writel_rc(pp, pp->config.io_bus_addr, PCIE_ATU_LOWER_TARGET);
dw_pcie_writel_rc(pp, upper_32_bits(pp->config.io_bus_addr),
}
if (bus->number != pp->root_bus_nr)
- ret = dw_pcie_rd_other_conf(pp, bus, devfn,
+ if (pp->ops->rd_other_conf)
+ ret = pp->ops->rd_other_conf(pp, bus, devfn,
+ where, size, val);
+ else
+ ret = dw_pcie_rd_other_conf(pp, bus, devfn,
where, size, val);
else
ret = dw_pcie_rd_own_conf(pp, where, size, val);
return PCIBIOS_DEVICE_NOT_FOUND;
if (bus->number != pp->root_bus_nr)
- ret = dw_pcie_wr_other_conf(pp, bus, devfn,
+ if (pp->ops->wr_other_conf)
+ ret = pp->ops->wr_other_conf(pp, bus, devfn,
+ where, size, val);
+ else
+ ret = dw_pcie_wr_other_conf(pp, bus, devfn,
where, size, val);
else
ret = dw_pcie_wr_own_conf(pp, where, size, val);
u8 root_bus_nr;
void __iomem *dbi_base;
u64 cfg0_base;
+ u64 cfg0_mod_base;
void __iomem *va_cfg0_base;
u64 cfg1_base;
+ u64 cfg1_mod_base;
void __iomem *va_cfg1_base;
u64 io_base;
+ u64 io_mod_base;
u64 mem_base;
+ u64 mem_mod_base;
struct resource cfg;
struct resource io;
struct resource mem;
u32 val, void __iomem *dbi_base);
int (*rd_own_conf)(struct pcie_port *pp, int where, int size, u32 *val);
int (*wr_own_conf)(struct pcie_port *pp, int where, int size, u32 val);
+ int (*rd_other_conf)(struct pcie_port *pp, struct pci_bus *bus,
+ unsigned int devfn, int where, int size, u32 *val);
+ int (*wr_other_conf)(struct pcie_port *pp, struct pci_bus *bus,
+ unsigned int devfn, int where, int size, u32 val);
int (*link_up)(struct pcie_port *pp);
void (*host_init)(struct pcie_port *pp);
+ void (*msi_set_irq)(struct pcie_port *pp, int irq);
+ void (*msi_clear_irq)(struct pcie_port *pp, int irq);
+ u32 (*get_msi_data)(struct pcie_port *pp);
};
int dw_pcie_cfg_read(void __iomem *addr, int where, int size, u32 *val);
static int __init rpaphp_init(void)
{
- struct device_node *dn = NULL;
+ struct device_node *dn;
info(DRIVER_DESC " version: " DRIVER_VERSION "\n");
- while ((dn = of_find_node_by_name(dn, "pci")))
+ for_each_node_by_name(dn, "pci")
rpaphp_add_slot(dn);
return 0;
}
-static DEFINE_PCI_DEVICE_TABLE(ioapic_devices) = {
+static const struct pci_device_id ioapic_devices[] = {
{ PCI_DEVICE_CLASS(PCI_CLASS_SYSTEM_PIC_IOAPIC, ~0) },
{ PCI_DEVICE_CLASS(PCI_CLASS_SYSTEM_PIC_IOXAPIC, ~0) },
{ }
If you have a modern Toshiba laptop with a Bluetooth and an
RFKill switch (such as the Portege R500), say Y.
+config TOSHIBA_HAPS
+ tristate "Toshiba HDD Active Protection Sensor"
+ depends on ACPI
+ ---help---
+ This driver adds support for the built-in accelerometer
+ found on recent Toshiba laptops equiped with HID TOS620A
+ device.
+
+ This driver receives ACPI notify events 0x80 when the sensor
+ detects a sudden move or a harsh vibration, as well as an
+ ACPI notify event 0x81 whenever the movement or vibration has
+ been stabilized.
+
+ Also provides sysfs entries to get/set the desired protection
+ level and reseting the HDD protection interface.
+
+ If you have a recent Toshiba laptop with a built-in accelerometer
+ device, say Y.
+
config ACPI_CMPC
tristate "CMPC Laptop Extras"
depends on X86 && ACPI
obj-$(CONFIG_ACPI_TOSHIBA) += toshiba_acpi.o
obj-$(CONFIG_TOSHIBA_BT_RFKILL) += toshiba_bluetooth.o
+obj-$(CONFIG_TOSHIBA_HAPS) += toshiba_haps.o
obj-$(CONFIG_INTEL_SCU_IPC) += intel_scu_ipc.o
obj-$(CONFIG_INTEL_SCU_IPC_UTIL) += intel_scu_ipcutil.o
obj-$(CONFIG_INTEL_MFLD_THERMAL) += intel_mid_thermal.o
WMID_ACCEL_EVENT = 0x5,
};
-static const struct key_entry acer_wmi_keymap[] = {
+static const struct key_entry acer_wmi_keymap[] __initconst = {
{KE_KEY, 0x01, {KEY_WLAN} }, /* WiFi */
{KE_KEY, 0x03, {KEY_WLAN} }, /* WiFi */
{KE_KEY, 0x04, {KEY_WLAN} }, /* WiFi */
static struct quirk_entry *quirks;
-static void set_quirks(void)
+static void __init set_quirks(void)
{
if (!interface)
return;
interface->capability |= ACER_CAP_BRIGHTNESS;
}
-static int dmi_matched(const struct dmi_system_id *dmi)
+static int __init dmi_matched(const struct dmi_system_id *dmi)
{
quirks = dmi->driver_data;
return 1;
};
/* The Aspire One has a dummy ACPI-WMI interface - disable it */
-static struct dmi_system_id acer_blacklist[] = {
+static const struct dmi_system_id acer_blacklist[] __initconst = {
{
.ident = "Acer Aspire One (SSD)",
.matches = {
{}
};
-static struct dmi_system_id acer_quirks[] = {
+static const struct dmi_system_id acer_quirks[] __initconst = {
{
.callback = dmi_matched,
.ident = "Acer Aspire 1360",
{}
};
-static int video_set_backlight_video_vendor(const struct dmi_system_id *d)
+static int __init
+video_set_backlight_video_vendor(const struct dmi_system_id *d)
{
interface->capability &= ~ACER_CAP_BRIGHTNESS;
pr_info("Brightness must be controlled by generic video driver\n");
return 0;
}
-static const struct dmi_system_id video_vendor_dmi_table[] = {
+static const struct dmi_system_id video_vendor_dmi_table[] __initconst = {
{
.callback = video_set_backlight_video_vendor,
.ident = "Acer TravelMate 4750",
};
/* Find which quirks are needed for a particular vendor/ model pair */
-static void find_quirks(void)
+static void __init find_quirks(void)
{
if (!force_series) {
dmi_check_system(acer_quirks);
return wmab_execute(&args, NULL);
}
-static acpi_status AMW0_find_mailled(void)
+static acpi_status __init AMW0_find_mailled(void)
{
struct wmab_args args;
struct wmab_ret ret;
return AE_OK;
}
-static int AMW0_set_cap_acpi_check_device_found;
+static int AMW0_set_cap_acpi_check_device_found __initdata;
-static acpi_status AMW0_set_cap_acpi_check_device_cb(acpi_handle handle,
+static acpi_status __init AMW0_set_cap_acpi_check_device_cb(acpi_handle handle,
u32 level, void *context, void **retval)
{
AMW0_set_cap_acpi_check_device_found = 1;
return AE_OK;
}
-static const struct acpi_device_id norfkill_ids[] = {
+static const struct acpi_device_id norfkill_ids[] __initconst = {
{ "VPC2004", 0},
{ "IBM0068", 0},
{ "LEN0068", 0},
{ "", 0},
};
-static int AMW0_set_cap_acpi_check_device(void)
+static int __init AMW0_set_cap_acpi_check_device(void)
{
const struct acpi_device_id *id;
return AMW0_set_cap_acpi_check_device_found;
}
-static acpi_status AMW0_set_capabilities(void)
+static acpi_status __init AMW0_set_capabilities(void)
{
struct wmab_args args;
struct wmab_ret ret;
return wmid3_set_device_status(value, device);
}
-static void type_aa_dmi_decode(const struct dmi_header *header, void *dummy)
+static void __init type_aa_dmi_decode(const struct dmi_header *header, void *d)
{
struct hotkey_function_type_aa *type_aa;
commun_fn_key_number = type_aa->commun_fn_key_number;
}
-static acpi_status WMID_set_capabilities(void)
+static acpi_status __init WMID_set_capabilities(void)
{
struct acpi_buffer out = {ACPI_ALLOCATE_BUFFER, NULL};
union acpi_object *obj;
u32 result; \
acpi_status status;
- pr_info("This threeg sysfs will be removed in 2012 - used by: %s\n",
+ pr_info("This threeg sysfs will be removed in 2014 - used by: %s\n",
current->comm);
status = get_u32(&result, ACER_CAP_THREEG);
if (ACPI_SUCCESS(status))
{
u32 tmp = simple_strtoul(buf, NULL, 10);
acpi_status status = set_u32(tmp, ACER_CAP_THREEG);
- pr_info("This threeg sysfs will be removed in 2012 - used by: %s\n",
+ pr_info("This threeg sysfs will be removed in 2014 - used by: %s\n",
current->comm);
if (ACPI_FAILURE(status))
return -EINVAL;
static ssize_t show_interface(struct device *dev, struct device_attribute *attr,
char *buf)
{
- pr_info("This interface sysfs will be removed in 2012 - used by: %s\n",
+ pr_info("This interface sysfs will be removed in 2014 - used by: %s\n",
current->comm);
switch (interface->type) {
case ACER_AMW0:
}
}
-static acpi_status
+static acpi_status __init
wmid3_set_lm_mode(struct lm_input_params *params,
struct lm_return_value *return_value)
{
return status;
}
-static int acer_wmi_enable_ec_raw(void)
+static int __init acer_wmi_enable_ec_raw(void)
{
struct lm_return_value return_value;
acpi_status status;
return status;
}
-static int acer_wmi_enable_lm(void)
+static int __init acer_wmi_enable_lm(void)
{
struct lm_return_value return_value;
acpi_status status;
return 0;
}
+#ifdef CONFIG_PM_SLEEP
static int acer_suspend(struct device *dev)
{
u32 value;
return 0;
}
+#else
+#define acer_suspend NULL
+#define acer_resume NULL
+#endif
static SIMPLE_DEV_PM_OPS(acer_pm, acer_suspend, acer_resume);
return 0;
}
-static int create_sysfs(void)
+static int __init create_sysfs(void)
{
int retval = -ENOMEM;
debugfs_remove(interface->debug.root);
}
-static int create_debugfs(void)
+static int __init create_debugfs(void)
{
interface->debug.root = debugfs_create_dir("acer-wmi", NULL);
if (!interface->debug.root) {
struct quirk_entry {
u8 num_zones;
+ u8 hdmi_mux;
};
static struct quirk_entry *quirks;
static struct quirk_entry quirk_unknown = {
.num_zones = 2,
+ .hdmi_mux = 0,
};
static struct quirk_entry quirk_x51_family = {
.num_zones = 3,
+ .hdmi_mux = 0.
};
-static int dmi_matched(const struct dmi_system_id *dmi)
+static struct quirk_entry quirk_asm100 = {
+ .num_zones = 2,
+ .hdmi_mux = 1,
+};
+
+static int __init dmi_matched(const struct dmi_system_id *dmi)
{
quirks = dmi->driver_data;
return 1;
}
-static struct dmi_system_id alienware_quirks[] = {
+static const struct dmi_system_id alienware_quirks[] __initconst = {
{
.callback = dmi_matched,
.ident = "Alienware X51 R1",
},
.driver_data = &quirk_x51_family,
},
+ {
+ .callback = dmi_matched,
+ .ident = "Alienware ASM100",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Alienware"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "ASM100"),
+ },
+ .driver_data = &quirk_asm100,
+ },
{}
};
static void remove_hdmi(struct platform_device *dev)
{
- sysfs_remove_group(&dev->dev.kobj, &hdmi_attribute_group);
+ if (quirks->hdmi_mux > 0)
+ sysfs_remove_group(&dev->dev.kobj, &hdmi_attribute_group);
}
static int create_hdmi(struct platform_device *dev)
if (ret)
goto fail_platform_device2;
- if (interface == WMAX) {
+ if (quirks->hdmi_mux > 0) {
ret = create_hdmi(platform_device);
if (ret)
goto fail_prep_hdmi;
.no_display_toggle = true,
};
-static struct quirk_entry quirk_asus_x401u = {
+static struct quirk_entry quirk_asus_wapf4 = {
.wapf = 4,
};
+static struct quirk_entry quirk_asus_x200ca = {
+ .wapf = 2,
+};
+
static int dmi_matched(const struct dmi_system_id *dmi)
{
quirks = dmi->driver_data;
return 1;
}
-static struct dmi_system_id asus_quirks[] = {
+static const struct dmi_system_id asus_quirks[] = {
+ {
+ .callback = dmi_matched,
+ .ident = "ASUSTeK COMPUTER INC. U32U",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK Computer Inc."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "U32U"),
+ },
+ /*
+ * Note this machine has a Brazos APU, and most Brazos Asus
+ * machines need quirk_asus_x55u / wmi_backlight_power but
+ * here acpi-video seems to work fine for backlight control.
+ */
+ .driver_data = &quirk_asus_wapf4,
+ },
{
.callback = dmi_matched,
.ident = "ASUSTeK COMPUTER INC. X401U",
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X401A"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X401A1"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X501A"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X501A1"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X550CA"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
+ },
+ {
+ .callback = dmi_matched,
+ .ident = "ASUSTeK COMPUTER INC. X550CC",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "X550CC"),
+ },
+ .driver_data = &quirk_asus_wapf4,
+ },
+ {
+ .callback = dmi_matched,
+ .ident = "ASUSTeK COMPUTER INC. X550CL",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "X550CL"),
+ },
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X55A"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X55C"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X55VD"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "X75A"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
+ },
+ {
+ .callback = dmi_matched,
+ .ident = "ASUSTeK COMPUTER INC. X75VBP",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "X75VBP"),
+ },
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "1015E"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
},
{
.callback = dmi_matched,
DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
DMI_MATCH(DMI_PRODUCT_NAME, "1015U"),
},
- .driver_data = &quirk_asus_x401u,
+ .driver_data = &quirk_asus_wapf4,
+ },
+ {
+ .callback = dmi_matched,
+ .ident = "ASUSTeK COMPUTER INC. X200CA",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "X200CA"),
+ },
+ .driver_data = &quirk_asus_x200ca,
},
{},
};
#include <linux/platform_device.h>
#include <linux/thermal.h>
#include <linux/acpi.h>
+#include <linux/dmi.h>
#include <acpi/video.h>
#include "asus-wmi.h"
goto error;
}
- if (wlan_led_presence(asus) && (asus->driver->quirks->wapf == 4)) {
+ if (wlan_led_presence(asus) && (asus->driver->quirks->wapf > 0)) {
INIT_WORK(&asus->wlan_led_work, wlan_led_update);
asus->wlan_led.name = "asus::wlan";
return -EINVAL;
if ((dev_id == ASUS_WMI_DEVID_WLAN) &&
- (asus->driver->quirks->wapf == 4))
+ (asus->driver->quirks->wapf > 0))
rfkill_set_led_trigger_name(*rfkill, "asus-wlan");
rfkill_init_sw_state(*rfkill, !result);
int power;
max = read_brightness_max(asus);
-
- if (max == -ENODEV)
- max = 0;
- else if (max < 0)
+ if (max < 0)
return max;
power = read_backlight_power(asus);
struct platform_driver *pdrv = to_platform_driver(pdev->dev.driver);
struct asus_wmi_driver *wdrv = to_asus_wmi_driver(pdrv);
struct asus_wmi *asus;
+ const char *chassis_type;
acpi_status status;
int err;
u32 result;
if (err)
goto fail_rfkill;
+ /* Some Asus desktop boards export an acpi-video backlight interface,
+ stop this from showing up */
+ chassis_type = dmi_get_system_info(DMI_CHASSIS_TYPE);
+ if (chassis_type && !strcmp(chassis_type, "3"))
+ acpi_video_dmi_promote_vendor();
if (asus->driver->quirks->wmi_backlight_power)
acpi_video_dmi_promote_vendor();
if (!acpi_video_backlight_support()) {
return err;
hwmon_dev = hwmon_device_register_with_groups(&pdev->dev,
- DRIVER_NAME, data,
+ "compal", data,
compal_hwmon_groups);
if (IS_ERR(hwmon_dev)) {
err = PTR_ERR(hwmon_dev);
.touchpad_led = 1,
};
-static int dmi_matched(const struct dmi_system_id *dmi)
+static int __init dmi_matched(const struct dmi_system_id *dmi)
{
quirks = dmi->driver_data;
return 1;
};
MODULE_DEVICE_TABLE(dmi, dell_device_table);
-static struct dmi_system_id dell_quirks[] = {
+static const struct dmi_system_id dell_quirks[] __initconst = {
{
.callback = dmi_matched,
.ident = "Dell Vostro V130",
.flags = LED_CORE_SUSPENDRESUME,
};
-static int touchpad_led_init(struct device *dev)
+static int __init touchpad_led_init(struct device *dev)
{
return led_classdev_register(dev, &touchpad_led);
}
return sprintf(buf, "%d\n", get());
}
-#define EEEPC_CREATE_SENSOR_ATTR(_name, _mode, _set, _get) \
+#define EEEPC_CREATE_SENSOR_ATTR(_name, _mode, _get, _set) \
static ssize_t show_##_name(struct device *dev, \
struct device_attribute *attr, \
char *buf) \
{ \
- return show_sys_hwmon(_set, buf); \
+ return show_sys_hwmon(_get, buf); \
} \
static ssize_t store_##_name(struct device *dev, \
struct device_attribute *attr, \
const char *buf, size_t count) \
{ \
- return store_sys_hwmon(_get, buf, count); \
+ return store_sys_hwmon(_set, buf, count); \
} \
- static DEVICE_ATTR(_name, _mode, show_##_name, store_##_name);
+ static DEVICE_ATTR(_name, _mode, show_##_name, store_##_name)
EEEPC_CREATE_SENSOR_ATTR(fan1_input, S_IRUGO, eeepc_get_fan_rpm, NULL);
EEEPC_CREATE_SENSOR_ATTR(pwm1, S_IRUGO | S_IWUSR,
return 1;
}
-static struct dmi_system_id asus_quirks[] = {
+static const struct dmi_system_id asus_quirks[] = {
{
.callback = dmi_matched,
.ident = "ASUSTeK Computer INC. 1000H",
#define FUJLAPTOP_DBG_INFO 0x0004
#define FUJLAPTOP_DBG_TRACE 0x0008
-#define dbg_printk(a_dbg_level, format, arg...) \
+#ifdef CONFIG_FUJITSU_LAPTOP_DEBUG
+#define vdbg_printk(a_dbg_level, format, arg...) \
do { if (dbg_level & a_dbg_level) \
printk(FUJLAPTOP_DEBUG "%s: " format, __func__ , ## arg); \
} while (0)
-#ifdef CONFIG_FUJITSU_LAPTOP_DEBUG
-#define vdbg_printk(a_dbg_level, format, arg...) \
- dbg_printk(a_dbg_level, format, ## arg)
#else
-#define vdbg_printk(a_dbg_level, format, arg...)
+#define vdbg_printk(a_dbg_level, format, arg...) \
+ do { } while (0)
#endif
/* Device controlling the backlight and associated keys */
}
};
-static void dmi_check_cb_common(const struct dmi_system_id *id)
+static void __init dmi_check_cb_common(const struct dmi_system_id *id)
{
pr_info("Identified laptop model '%s'\n", id->ident);
if (use_alt_lcd_levels == -1) {
}
}
-static int dmi_check_cb_s6410(const struct dmi_system_id *id)
+static int __init dmi_check_cb_s6410(const struct dmi_system_id *id)
{
dmi_check_cb_common(id);
fujitsu->keycode1 = KEY_SCREENLOCK; /* "Lock" */
return 1;
}
-static int dmi_check_cb_s6420(const struct dmi_system_id *id)
+static int __init dmi_check_cb_s6420(const struct dmi_system_id *id)
{
dmi_check_cb_common(id);
fujitsu->keycode1 = KEY_SCREENLOCK; /* "Lock" */
return 1;
}
-static int dmi_check_cb_p8010(const struct dmi_system_id *id)
+static int __init dmi_check_cb_p8010(const struct dmi_system_id *id)
{
dmi_check_cb_common(id);
fujitsu->keycode1 = KEY_HELP; /* "Support" */
return 1;
}
-static struct dmi_system_id fujitsu_dmi_table[] = {
+static const struct dmi_system_id fujitsu_dmi_table[] __initconst = {
{
.ident = "Fujitsu Siemens S6410",
.matches = {
return IRQ_HANDLED;
}
-static void fujitsu_dmi_common(const struct dmi_system_id *dmi)
+static void __init fujitsu_dmi_common(const struct dmi_system_id *dmi)
{
pr_info("%s\n", dmi->ident);
memcpy(fujitsu.config.keymap, dmi->driver_data,
sizeof(fujitsu.config.keymap));
}
-static int fujitsu_dmi_lifebook(const struct dmi_system_id *dmi)
+static int __init fujitsu_dmi_lifebook(const struct dmi_system_id *dmi)
{
fujitsu_dmi_common(dmi);
fujitsu.config.quirks |= INVERT_TABLET_MODE_BIT;
return 1;
}
-static int fujitsu_dmi_stylistic(const struct dmi_system_id *dmi)
+static int __init fujitsu_dmi_stylistic(const struct dmi_system_id *dmi)
{
fujitsu_dmi_common(dmi);
fujitsu.config.quirks |= FORCE_TABLET_MODE_IF_UNDOCK;
return (state & 0x4) ? 1 : 0;
}
-static int hp_wmi_bios_2009_later(void)
+static int __init hp_wmi_bios_2009_later(void)
{
int state = 0;
int ret = hp_wmi_perform_query(HPWMI_FEATURE_QUERY, 0, &state,
device_remove_file(&device->dev, &dev_attr_postcode);
}
-static int hp_wmi_rfkill_setup(struct platform_device *device)
+static int __init hp_wmi_rfkill_setup(struct platform_device *device)
{
int err;
int wireless = 0;
return err;
}
-static int hp_wmi_rfkill2_setup(struct platform_device *device)
+static int __init hp_wmi_rfkill2_setup(struct platform_device *device)
{
int err, i;
struct bios_rfkill2_state state;
/* HP-specific accelerometer driver ------------------------------------ */
/* For automatic insertion of the module */
-static struct acpi_device_id lis3lv02d_device_ids[] = {
+static const struct acpi_device_id lis3lv02d_device_ids[] = {
{"HPQ0004", 0}, /* HP Mobile Data Protection System PNP */
{"HPQ6000", 0}, /* HP Mobile Data Protection System PNP */
{"HPQ6007", 0}, /* HP Mobile Data Protection System PNP */
}, \
.driver_data = &lis3lv02d_axis_##_axis \
}
-static struct dmi_system_id lis3lv02d_dmi_ids[] = {
+static const struct dmi_system_id lis3lv02d_dmi_ids[] = {
/* product names are truncated to match all kinds of a same model */
AXIS_DMI_MATCH("NC64x0", "HP Compaq nc64", x_inverted),
AXIS_DMI_MATCH("NC84x0", "HP Compaq nc84", z_inverted),
struct backlight_device *blightdev;
struct dentry *debug;
unsigned long cfg;
+ bool has_hw_rfkill_switch;
+ bool has_touchpad_control;
};
static bool no_bt_rfkill;
return supported ? attr->mode : 0;
}
-static struct attribute_group ideapad_attribute_group = {
+static const struct attribute_group ideapad_attribute_group = {
.is_visible = ideapad_is_visible,
.attrs = ideapad_attributes
};
int type;
};
-const struct ideapad_rfk_data ideapad_rfk_data[] = {
+const const struct ideapad_rfk_data ideapad_rfk_data[] = {
{ "ideapad_wlan", CFG_WIFI_BIT, VPCCMD_W_WIFI, RFKILL_TYPE_WLAN },
{ "ideapad_bluetooth", CFG_BT_BIT, VPCCMD_W_BT, RFKILL_TYPE_BLUETOOTH },
{ "ideapad_3g", CFG_3G_BIT, VPCCMD_W_3G, RFKILL_TYPE_WWAN },
static void ideapad_sync_rfk_state(struct ideapad_private *priv)
{
- unsigned long hw_blocked;
+ unsigned long hw_blocked = 0;
int i;
- if (read_ec_data(priv->adev->handle, VPCCMD_R_RF, &hw_blocked))
- return;
- hw_blocked = !hw_blocked;
+ if (priv->has_hw_rfkill_switch) {
+ if (read_ec_data(priv->adev->handle, VPCCMD_R_RF, &hw_blocked))
+ return;
+ hw_blocked = !hw_blocked;
+ }
for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++)
if (priv->rfk[i])
{
unsigned long value;
+ if (!priv->has_touchpad_control)
+ return;
+
/* Without reading from EC touchpad LED doesn't switch state */
if (!read_ec_data(priv->adev->handle, VPCCMD_R_TOUCHPAD, &value)) {
/* Some IdeaPads don't really turn off touchpad - they only
}
}
-/* Blacklist for devices where the ideapad rfkill interface does not work */
-static struct dmi_system_id rfkill_blacklist[] = {
- /* The Lenovo Yoga 2 11 always reports everything as blocked */
+/*
+ * Some ideapads don't have a hardware rfkill switch, reading VPCCMD_R_RF
+ * always results in 0 on these models, causing ideapad_laptop to wrongly
+ * report all radios as hardware-blocked.
+ */
+static struct dmi_system_id no_hw_rfkill_list[] = {
{
- .ident = "Lenovo Yoga 2 11",
+ .ident = "Lenovo Yoga 2 11 / 13 / Pro",
.matches = {
DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
- DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo Yoga 2 11"),
+ DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo Yoga 2"),
+ },
+ },
+ {}
+};
+
+/*
+ * Some models don't offer touchpad ctrl through the ideapad interface, causing
+ * ideapad_sync_touchpad_state to send wrong touchpad enable/disable events.
+ */
+static struct dmi_system_id no_touchpad_ctrl_list[] = {
+ {
+ .ident = "Lenovo Yoga 1 series",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+ DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo IdeaPad Yoga"),
+ },
+ },
+ {
+ .ident = "Lenovo Yoga 2 11 / 13 / Pro",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+ DMI_MATCH(DMI_PRODUCT_VERSION, "Lenovo Yoga 2"),
},
},
{}
priv->cfg = cfg;
priv->adev = adev;
priv->platform_device = pdev;
+ priv->has_hw_rfkill_switch = !dmi_check_system(no_hw_rfkill_list);
+ priv->has_touchpad_control = !dmi_check_system(no_touchpad_ctrl_list);
ret = ideapad_sysfs_init(priv);
if (ret)
if (ret)
goto input_failed;
- if (!dmi_check_system(rfkill_blacklist)) {
- for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++)
- if (test_bit(ideapad_rfk_data[i].cfgbit, &priv->cfg))
- ideapad_register_rfkill(priv, i);
- }
+ /*
+ * On some models without a hw-switch (the yoga 2 13 at least)
+ * VPCCMD_W_RF must be explicitly set to 1 for the wifi to work.
+ */
+ if (!priv->has_hw_rfkill_switch)
+ write_ec_cmd(priv->adev->handle, VPCCMD_W_RF, 1);
+
+ for (i = 0; i < IDEAPAD_RFKILL_DEV_NUM; i++)
+ if (test_bit(ideapad_rfk_data[i].cfgbit, &priv->cfg))
+ ideapad_register_rfkill(priv, i);
+
ideapad_sync_rfk_state(priv);
ideapad_sync_touchpad_state(priv);
/* Max temps are -10 degrees C to avoid PROCHOT# */
-struct ips_mcp_limits ips_sv_limits = {
+static struct ips_mcp_limits ips_sv_limits = {
.mcp_power_limit = 35000,
.core_power_limit = 29000,
.mch_power_limit = 20000,
.mch_temp_limit = 90
};
-struct ips_mcp_limits ips_lv_limits = {
+static struct ips_mcp_limits ips_lv_limits = {
.mcp_power_limit = 25000,
.core_power_limit = 21000,
.mch_power_limit = 13000,
.mch_temp_limit = 90
};
-struct ips_mcp_limits ips_ulv_limits = {
+static struct ips_mcp_limits ips_ulv_limits = {
.mcp_power_limit = 18000,
.core_power_limit = 14000,
.mch_power_limit = 11000,
}
EXPORT_SYMBOL_GPL(ips_link_to_i915_driver);
-static DEFINE_PCI_DEVICE_TABLE(ips_id_table) = {
+static const struct pci_device_id ips_id_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL,
PCI_DEVICE_ID_INTEL_THERMAL_SENSOR), },
{ 0, }
intel_scu_devices_destroy();
}
-static DEFINE_PCI_DEVICE_TABLE(pci_ids) = {
+static const struct pci_device_id pci_ids[] = {
{
PCI_VDEVICE(INTEL, PCI_DEVICE_ID_LINCROFT),
(kernel_ulong_t)&intel_scu_ipc_lincroft_pdata,
lid_ctl->attrs[LID_RESUME_S3].store = sony_nc_lid_resume_store;
}
for (i = 0; i < LID_RESUME_MAX &&
- lid_ctl->attrs[LID_RESUME_S3].attr.name; i++) {
+ lid_ctl->attrs[i].attr.name; i++) {
result = device_create_file(&pd->dev, &lid_ctl->attrs[i]);
if (result)
goto liderror;
KEY_UNKNOWN,
/* Extra keys in use since the X240 / T440 / T540 */
- KEY_CONFIG, KEY_SEARCH, KEY_SCALE, KEY_COMPUTER,
+ KEY_CONFIG, KEY_SEARCH, KEY_SCALE, KEY_FILE,
},
};
{
int res;
- if (value > bright_maxlvl || value < 0)
+ if (value > bright_maxlvl)
return -EINVAL;
vdbg_printk(TPACPI_DBG_BRGHT,
return volume_alsa_set_mute(!ucontrol->value.integer.value[0]);
}
-static struct snd_kcontrol_new volume_alsa_control_vol = {
+static struct snd_kcontrol_new volume_alsa_control_vol __initdata = {
.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
.name = "Console Playback Volume",
.index = 0,
.get = volume_alsa_vol_get,
};
-static struct snd_kcontrol_new volume_alsa_control_mute = {
+static struct snd_kcontrol_new volume_alsa_control_mute __initdata = {
.iface = SNDRV_CTL_ELEM_IFACE_MIXER,
.name = "Console Playback Switch",
.index = 0,
DMI_MATCH(DMI_PRODUCT_NAME, "Satellite M840"),
},
},
+ {
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "TOSHIBA"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "Qosmio X75-A"),
+ },
+ },
{}
};
{ KE_KEY, 0x157, { KEY_MUTE } },
{ KE_KEY, 0x102, { KEY_ZOOMOUT } },
{ KE_KEY, 0x103, { KEY_ZOOMIN } },
+ { KE_KEY, 0x12c, { KEY_KBDILLUMTOGGLE } },
{ KE_KEY, 0x139, { KEY_ZOOMRESET } },
{ KE_KEY, 0x13e, { KEY_SWITCHVIDEOMODE } },
{ KE_KEY, 0x13c, { KEY_BRIGHTNESSDOWN } },
static int set_lcd_brightness(struct toshiba_acpi_dev *dev, int value)
{
- u32 hci_result;
+ u32 in[HCI_WORDS] = { HCI_SET, HCI_LCD_BRIGHTNESS, 0, 0, 0, 0 };
+ u32 out[HCI_WORDS];
+ acpi_status status;
if (dev->tr_backlight_supported) {
bool enable = !value;
value--;
}
- value = value << HCI_LCD_BRIGHTNESS_SHIFT;
- hci_write1(dev, HCI_LCD_BRIGHTNESS, value, &hci_result);
- return hci_result == HCI_SUCCESS ? 0 : -EIO;
+ in[2] = value << HCI_LCD_BRIGHTNESS_SHIFT;
+ status = hci_raw(dev, in, out);
+ if (ACPI_FAILURE(status) || out[0] == HCI_FAILURE) {
+ pr_err("ACPI call to set brightness failed");
+ return -EIO;
+ }
+ /* Extra check for "incomplete" backlight method, where the AML code
+ * doesn't check for HCI_SET or HCI_GET and returns HCI_SUCCESS,
+ * the actual brightness, and in some cases the max brightness.
+ */
+ if (out[2] > 0 || out[3] == 0xE000)
+ return -ENODEV;
+
+ return out[0] == HCI_SUCCESS ? 0 : -EIO;
}
static int set_lcd_status(struct backlight_device *bd)
--- /dev/null
+/*
+ * Toshiba HDD Active Protection Sensor (HAPS) driver
+ *
+ * Copyright (C) 2014 Azael Avalos <coproscefalo@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/acpi.h>
+
+MODULE_AUTHOR("Azael Avalos <coproscefalo@gmail.com>");
+MODULE_DESCRIPTION("Toshiba HDD Active Protection Sensor");
+MODULE_LICENSE("GPL");
+
+struct toshiba_haps_dev {
+ struct acpi_device *acpi_dev;
+
+ int protection_level;
+};
+
+static struct toshiba_haps_dev *toshiba_haps;
+
+/* HAPS functions */
+static int toshiba_haps_reset_protection(acpi_handle handle)
+{
+ acpi_status status;
+
+ status = acpi_evaluate_object(handle, "RSSS", NULL, NULL);
+ if (ACPI_FAILURE(status)) {
+ pr_err("Unable to reset the HDD protection\n");
+ return -EIO;
+ }
+
+ return 0;
+}
+
+static int toshiba_haps_protection_level(acpi_handle handle, int level)
+{
+ acpi_status status;
+
+ status = acpi_execute_simple_method(handle, "PTLV", level);
+ if (ACPI_FAILURE(status)) {
+ pr_err("Error while setting the protection level\n");
+ return -EIO;
+ }
+
+ pr_info("HDD protection level set to: %d\n", level);
+
+ return 0;
+}
+
+/* sysfs files */
+static ssize_t protection_level_show(struct device *dev,
+ struct device_attribute *attr, char *buf)
+{
+ struct toshiba_haps_dev *haps = dev_get_drvdata(dev);
+
+ return sprintf(buf, "%i\n", haps->protection_level);
+}
+
+static ssize_t protection_level_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct toshiba_haps_dev *haps = dev_get_drvdata(dev);
+ int level, ret;
+
+ if (sscanf(buf, "%d", &level) != 1 || level < 0 || level > 3)
+ return -EINVAL;
+
+ /* Set the sensor level.
+ * Acceptable levels are:
+ * 0 - Disabled | 1 - Low | 2 - Medium | 3 - High
+ */
+ ret = toshiba_haps_protection_level(haps->acpi_dev->handle, level);
+ if (ret != 0)
+ return ret;
+
+ haps->protection_level = level;
+
+ return count;
+}
+
+static ssize_t reset_protection_store(struct device *dev,
+ struct device_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct toshiba_haps_dev *haps = dev_get_drvdata(dev);
+ int reset, ret;
+
+ if (sscanf(buf, "%d", &reset) != 1 || reset != 1)
+ return -EINVAL;
+
+ /* Reset the protection interface */
+ ret = toshiba_haps_reset_protection(haps->acpi_dev->handle);
+ if (ret != 0)
+ return ret;
+
+ return count;
+}
+
+static DEVICE_ATTR(protection_level, S_IRUGO | S_IWUSR,
+ protection_level_show, protection_level_store);
+static DEVICE_ATTR(reset_protection, S_IWUSR, NULL, reset_protection_store);
+
+static struct attribute *haps_attributes[] = {
+ &dev_attr_protection_level.attr,
+ &dev_attr_reset_protection.attr,
+ NULL,
+};
+
+static struct attribute_group haps_attr_group = {
+ .attrs = haps_attributes,
+};
+
+/*
+ * ACPI stuff
+ */
+static void toshiba_haps_notify(struct acpi_device *device, u32 event)
+{
+ pr_info("Received event: 0x%x", event);
+
+ acpi_bus_generate_netlink_event(device->pnp.device_class,
+ dev_name(&device->dev),
+ event, 0);
+}
+
+static int toshiba_haps_remove(struct acpi_device *device)
+{
+ sysfs_remove_group(&device->dev.kobj, &haps_attr_group);
+
+ if (toshiba_haps)
+ toshiba_haps = NULL;
+
+ return 0;
+}
+
+/* Helper function */
+static int toshiba_haps_available(acpi_handle handle)
+{
+ acpi_status status;
+ u64 hdd_present;
+
+ /*
+ * A non existent device as well as having (only)
+ * Solid State Drives can cause the call to fail.
+ */
+ status = acpi_evaluate_integer(handle, "_STA", NULL,
+ &hdd_present);
+ if (ACPI_FAILURE(status) || !hdd_present) {
+ pr_info("HDD protection not available or using SSD\n");
+ return 0;
+ }
+
+ return 1;
+}
+
+static int toshiba_haps_add(struct acpi_device *acpi_dev)
+{
+ struct toshiba_haps_dev *haps;
+ int ret;
+
+ if (toshiba_haps)
+ return -EBUSY;
+
+ if (!toshiba_haps_available(acpi_dev->handle))
+ return -ENODEV;
+
+ pr_info("Toshiba HDD Active Protection Sensor device\n");
+
+ haps = kzalloc(sizeof(struct toshiba_haps_dev), GFP_KERNEL);
+ if (!haps)
+ return -ENOMEM;
+
+ haps->acpi_dev = acpi_dev;
+ haps->protection_level = 2;
+ acpi_dev->driver_data = haps;
+ dev_set_drvdata(&acpi_dev->dev, haps);
+
+ /* Set the protection level, currently at level 2 (Medium) */
+ ret = toshiba_haps_protection_level(acpi_dev->handle, 2);
+ if (ret != 0)
+ return ret;
+
+ ret = sysfs_create_group(&acpi_dev->dev.kobj, &haps_attr_group);
+ if (ret)
+ return ret;
+
+ toshiba_haps = haps;
+
+ return 0;
+}
+
+#ifdef CONFIG_PM_SLEEP
+static int toshiba_haps_suspend(struct device *device)
+{
+ struct toshiba_haps_dev *haps;
+ int ret;
+
+ haps = acpi_driver_data(to_acpi_device(device));
+
+ /* Deactivate the protection on suspend */
+ ret = toshiba_haps_protection_level(haps->acpi_dev->handle, 0);
+
+ return ret;
+}
+
+static int toshiba_haps_resume(struct device *device)
+{
+ struct toshiba_haps_dev *haps;
+ int ret;
+
+ haps = acpi_driver_data(to_acpi_device(device));
+
+ /* Set the stored protection level */
+ ret = toshiba_haps_protection_level(haps->acpi_dev->handle,
+ haps->protection_level);
+
+ /* Reset the protection on resume */
+ ret = toshiba_haps_reset_protection(haps->acpi_dev->handle);
+ if (ret != 0)
+ return ret;
+
+ return ret;
+}
+#endif
+
+static SIMPLE_DEV_PM_OPS(toshiba_haps_pm,
+ toshiba_haps_suspend, toshiba_haps_resume);
+
+static const struct acpi_device_id haps_device_ids[] = {
+ {"TOS620A", 0},
+ {"", 0},
+};
+MODULE_DEVICE_TABLE(acpi, haps_device_ids);
+
+static struct acpi_driver toshiba_haps_driver = {
+ .name = "Toshiba HAPS",
+ .owner = THIS_MODULE,
+ .ids = haps_device_ids,
+ .flags = ACPI_DRIVER_ALL_NOTIFY_EVENTS,
+ .ops = {
+ .add = toshiba_haps_add,
+ .remove = toshiba_haps_remove,
+ .notify = toshiba_haps_notify,
+ },
+ .drv.pm = &toshiba_haps_pm,
+};
+
+module_acpi_driver(toshiba_haps_driver);
block = &wblock->gblock;
handle = wblock->handle;
- if (!block)
- return AE_NOT_EXIST;
-
-
snprintf(method, 5, "WE%02X", block->notify_id);
status = acpi_execute_simple_method(handle, method, enable);
return ret;
}
-static DEFINE_PCI_DEVICE_TABLE(pch_ieee1588_pcidev_id) = {
+static const struct pci_device_id pch_ieee1588_pcidev_id[] = {
{
.vendor = PCI_VENDOR_ID_INTEL,
.device = PCI_DEVICE_ID_PCH_1588
return err;
}
-static DEFINE_PCI_DEVICE_TABLE(tsi721_pci_tbl) = {
+static const struct pci_device_id tsi721_pci_tbl[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_IDT, PCI_DEVICE_ID_TSI721) },
{ 0, } /* terminate list */
};
PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
{ }
};*/
-static DEFINE_PCI_DEVICE_TABLE(blogic_pci_tbl) = {
+static const struct pci_device_id blogic_pci_tbl[] = {
{PCI_DEVICE(PCI_VENDOR_ID_BUSLOGIC, PCI_DEVICE_ID_BUSLOGIC_MULTIMASTER)},
{PCI_DEVICE(PCI_VENDOR_ID_BUSLOGIC, PCI_DEVICE_ID_BUSLOGIC_MULTIMASTER_NC)},
{PCI_DEVICE(PCI_VENDOR_ID_BUSLOGIC, PCI_DEVICE_ID_BUSLOGIC_FLASHPOINT)},
}
/*------------------- PCI Driver operations and data ----------------- */
-static DEFINE_PCI_DEVICE_TABLE(beiscsi_pci_id_table) = {
+static const struct pci_device_id beiscsi_pci_id_table[] = {
{ PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID1) },
{ PCI_DEVICE(BE_VENDOR_ID, BE_DEVICE_ID2) },
{ PCI_DEVICE(BE_VENDOR_ID, OC_DEVICE_ID1) },
.resume = csio_pci_resume,
};
-static DEFINE_PCI_DEVICE_TABLE(csio_pci_tbl) = {
+static const struct pci_device_id csio_pci_tbl[] = {
CSIO_DEVICE(CSIO_DEVID_T440DBG_FCOE, 0), /* T4 DEBUG FCOE */
CSIO_DEVICE(CSIO_DEVID_T420CR_FCOE, 0), /* T420CR FCOE */
CSIO_DEVICE(CSIO_DEVID_T422CR_FCOE, 0), /* T422CR FCOE */
config SCSI_CXGB3_ISCSI
tristate "Chelsio T3 iSCSI support"
- depends on PCI && INET
+ depends on PCI && INET && (IPV6 || IPV6=n)
select NETDEVICES
select ETHERNET
select NET_VENDOR_CHELSIO
config SCSI_CXGB4_ISCSI
tristate "Chelsio T4 iSCSI support"
- depends on PCI && INET
+ depends on PCI && INET && (IPV6 || IPV6=n)
select NETDEVICES
select ETHERNET
select NET_VENDOR_CHELSIO
static struct scsi_transport_template *isci_transport_template;
-static DEFINE_PCI_DEVICE_TABLE(isci_id_table) = {
+static const struct pci_device_id isci_id_table[] = {
{ PCI_VDEVICE(INTEL, 0x1D61),},
{ PCI_VDEVICE(INTEL, 0x1D63),},
{ PCI_VDEVICE(INTEL, 0x1D65),},
conn->ping_timeout, conn->recv_timeout,
last_recv, conn->last_ping, jiffies);
spin_unlock(&session->frwd_lock);
- iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED);
+ iscsi_conn_failure(conn, ISCSI_ERR_NOP_TIMEDOUT);
return;
}
/*
* The pci device ids are defined in mpi/mpi2_cnfg.h.
*/
-static DEFINE_PCI_DEVICE_TABLE(scsih_pci_table) = {
+static const struct pci_device_id scsih_pci_table[] = {
/* Fury ~ 3004 and 3008 */
{ MPI2_MFGPAGE_VENDORID_LSI, MPI25_MFGPAGE_DEVID_SAS3004,
PCI_ANY_ID, PCI_ANY_ID },
MODULE_AUTHOR("jyli@marvell.com");
MODULE_DESCRIPTION("Marvell UMI Driver");
-static DEFINE_PCI_DEVICE_TABLE(mvumi_pci_table) = {
+static const struct pci_device_id mvumi_pci_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, PCI_DEVICE_ID_MARVELL_MV9143) },
{ PCI_DEVICE(PCI_VENDOR_ID_MARVELL_EXT, PCI_DEVICE_ID_MARVELL_MV9580) },
{ 0 }
{
struct pm8001_ioctl_payload *payload;
DECLARE_COMPLETION_ONSTACK(completion);
- u8 *ioctlbuffer = NULL;
- u32 length = 0;
- u32 ret = 0;
+ u8 *ioctlbuffer;
+ u32 ret;
+ u32 length = 1024 * 5 + sizeof(*payload) - 1;
+
+ if (pm8001_ha->fw_image->size > 4096) {
+ pm8001_ha->fw_status = FAIL_FILE_SIZE;
+ return -EFAULT;
+ }
- length = 1024 * 5 + sizeof(*payload) - 1;
ioctlbuffer = kzalloc(length, GFP_KERNEL);
- if (!ioctlbuffer)
+ if (!ioctlbuffer) {
+ pm8001_ha->fw_status = FAIL_OUT_MEMORY;
return -ENOMEM;
- if ((pm8001_ha->fw_image->size <= 0) ||
- (pm8001_ha->fw_image->size > 4096)) {
- ret = FAIL_FILE_SIZE;
- goto out;
}
payload = (struct pm8001_ioctl_payload *)ioctlbuffer;
memcpy((u8 *)&payload->func_specific, (u8 *)pm8001_ha->fw_image->data,
payload->minor_function = 0x1;
pm8001_ha->nvmd_completion = &completion;
ret = PM8001_CHIP_DISP->set_nvmd_req(pm8001_ha, payload);
+ if (ret) {
+ pm8001_ha->fw_status = FAIL_OUT_MEMORY;
+ goto out;
+ }
wait_for_completion(&completion);
out:
kfree(ioctlbuffer);
{
struct pm8001_ioctl_payload *payload;
DECLARE_COMPLETION_ONSTACK(completion);
- u8 *ioctlbuffer = NULL;
- u32 length = 0;
+ u8 *ioctlbuffer;
struct fw_control_info *fwControl;
- u32 loopNumber, loopcount = 0;
- u32 sizeRead = 0;
u32 partitionSize, partitionSizeTmp;
- u32 ret = 0;
- u32 partitionNumber = 0;
+ u32 loopNumber, loopcount;
struct pm8001_fw_image_header *image_hdr;
+ u32 sizeRead = 0;
+ u32 ret = 0;
+ u32 length = 1024 * 16 + sizeof(*payload) - 1;
- length = 1024 * 16 + sizeof(*payload) - 1;
+ if (pm8001_ha->fw_image->size < 28) {
+ pm8001_ha->fw_status = FAIL_FILE_SIZE;
+ return -EFAULT;
+ }
ioctlbuffer = kzalloc(length, GFP_KERNEL);
- image_hdr = (struct pm8001_fw_image_header *)pm8001_ha->fw_image->data;
- if (!ioctlbuffer)
+ if (!ioctlbuffer) {
+ pm8001_ha->fw_status = FAIL_OUT_MEMORY;
return -ENOMEM;
- if (pm8001_ha->fw_image->size < 28) {
- ret = FAIL_FILE_SIZE;
- goto out;
}
-
+ image_hdr = (struct pm8001_fw_image_header *)pm8001_ha->fw_image->data;
while (sizeRead < pm8001_ha->fw_image->size) {
partitionSizeTmp =
*(u32 *)((u8 *)&image_hdr->image_length + sizeRead);
partitionSize = be32_to_cpu(partitionSizeTmp);
- loopcount = (partitionSize + HEADER_LEN)/IOCTL_BUF_SIZE;
- if (loopcount % IOCTL_BUF_SIZE)
- loopcount++;
- if (loopcount == 0)
- loopcount++;
+ loopcount = DIV_ROUND_UP(partitionSize + HEADER_LEN,
+ IOCTL_BUF_SIZE);
for (loopNumber = 0; loopNumber < loopcount; loopNumber++) {
payload = (struct pm8001_ioctl_payload *)ioctlbuffer;
payload->length = 1024*16;
pm8001_ha->nvmd_completion = &completion;
ret = PM8001_CHIP_DISP->fw_flash_update_req(pm8001_ha, payload);
- if (ret)
- break;
+ if (ret) {
+ pm8001_ha->fw_status = FAIL_OUT_MEMORY;
+ goto out;
+ }
wait_for_completion(&completion);
if (fwControl->retcode > FLASH_UPDATE_IN_PROGRESS) {
- ret = fwControl->retcode;
- break;
+ pm8001_ha->fw_status = fwControl->retcode;
+ ret = -EFAULT;
+ goto out;
+ }
}
}
- if (ret)
- break;
- partitionNumber++;
-}
out:
kfree(ioctlbuffer);
return ret;
char *cmd_ptr, *filename_ptr;
int res, i;
int flash_command = FLASH_CMD_NONE;
- int err = 0;
+ int ret;
+
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
- cmd_ptr = kzalloc(count*2, GFP_KERNEL);
+ /* this test protects us from running two flash processes at once,
+ * so we should start with this test */
+ if (pm8001_ha->fw_status == FLASH_IN_PROGRESS)
+ return -EINPROGRESS;
+ pm8001_ha->fw_status = FLASH_IN_PROGRESS;
+ cmd_ptr = kzalloc(count*2, GFP_KERNEL);
if (!cmd_ptr) {
- err = FAIL_OUT_MEMORY;
- goto out;
+ pm8001_ha->fw_status = FAIL_OUT_MEMORY;
+ return -ENOMEM;
}
filename_ptr = cmd_ptr + count;
res = sscanf(buf, "%s %s", cmd_ptr, filename_ptr);
if (res != 2) {
- err = FAIL_PARAMETERS;
- goto out1;
+ pm8001_ha->fw_status = FAIL_PARAMETERS;
+ ret = -EINVAL;
+ goto out;
}
for (i = 0; flash_command_table[i].code != FLASH_CMD_NONE; i++) {
}
}
if (flash_command == FLASH_CMD_NONE) {
- err = FAIL_PARAMETERS;
- goto out1;
+ pm8001_ha->fw_status = FAIL_PARAMETERS;
+ ret = -EINVAL;
+ goto out;
}
- if (pm8001_ha->fw_status == FLASH_IN_PROGRESS) {
- err = FLASH_IN_PROGRESS;
- goto out1;
- }
- err = request_firmware(&pm8001_ha->fw_image,
+ ret = request_firmware(&pm8001_ha->fw_image,
filename_ptr,
pm8001_ha->dev);
- if (err) {
+ if (ret) {
PM8001_FAIL_DBG(pm8001_ha,
- pm8001_printk("Failed to load firmware image file %s,"
- " error %d\n", filename_ptr, err));
- err = FAIL_OPEN_BIOS_FILE;
- goto out1;
+ pm8001_printk(
+ "Failed to load firmware image file %s, error %d\n",
+ filename_ptr, ret));
+ pm8001_ha->fw_status = FAIL_OPEN_BIOS_FILE;
+ goto out;
}
- switch (flash_command) {
- case FLASH_CMD_UPDATE:
- pm8001_ha->fw_status = FLASH_IN_PROGRESS;
- err = pm8001_update_flash(pm8001_ha);
- break;
- case FLASH_CMD_SET_NVMD:
- pm8001_ha->fw_status = FLASH_IN_PROGRESS;
- err = pm8001_set_nvmd(pm8001_ha);
- break;
- default:
- pm8001_ha->fw_status = FAIL_PARAMETERS;
- err = FAIL_PARAMETERS;
- break;
- }
+ if (FLASH_CMD_UPDATE == flash_command)
+ ret = pm8001_update_flash(pm8001_ha);
+ else
+ ret = pm8001_set_nvmd(pm8001_ha);
+
release_firmware(pm8001_ha->fw_image);
-out1:
- kfree(cmd_ptr);
out:
- pm8001_ha->fw_status = err;
+ kfree(cmd_ptr);
- if (!err)
- return count;
- else
- return -err;
+ if (ret)
+ return ret;
+
+ pm8001_ha->fw_status = FLASH_OK;
+ return count;
}
static ssize_t pm8001_show_update_fw(struct device *cdev,
rc = pm8001_tag_alloc(pm8001_ha, &tag);
if (rc) {
kfree(fw_control_context);
- return rc;
+ return -EBUSY;
}
ccb = &pm8001_ha->ccb_info[tag];
ccb->fw_control_context = fw_control_context;
rc = pm8001_tag_alloc(pm8001_ha, &tag);
if (rc) {
kfree(fw_control_context);
- return rc;
+ return -EBUSY;
}
ccb = &pm8001_ha->ccb_info[tag];
ccb->fw_control_context = fw_control_context;
sizeof(pm8001_ha->msix_entries[0]);
for (i = 0; i < max_entry ; i++)
pm8001_ha->msix_entries[i].entry = i;
- rc = pci_enable_msix(pm8001_ha->pdev, pm8001_ha->msix_entries,
+ rc = pci_enable_msix_exact(pm8001_ha->pdev, pm8001_ha->msix_entries,
number_of_intr);
pm8001_ha->number_of_intr = number_of_intr;
- if (!rc) {
- PM8001_INIT_DBG(pm8001_ha, pm8001_printk(
- "pci_enable_msix request ret:%d no of intr %d\n",
- rc, pm8001_ha->number_of_intr));
+ if (rc)
+ return rc;
+ PM8001_INIT_DBG(pm8001_ha, pm8001_printk(
+ "pci_enable_msix_exact request ret:%d no of intr %d\n",
+ rc, pm8001_ha->number_of_intr));
- for (i = 0; i < number_of_intr; i++) {
- snprintf(intr_drvname[i], sizeof(intr_drvname[0]),
- DRV_NAME"%d", i);
- pm8001_ha->irq_vector[i].irq_id = i;
- pm8001_ha->irq_vector[i].drv_inst = pm8001_ha;
+ for (i = 0; i < number_of_intr; i++) {
+ snprintf(intr_drvname[i], sizeof(intr_drvname[0]),
+ DRV_NAME"%d", i);
+ pm8001_ha->irq_vector[i].irq_id = i;
+ pm8001_ha->irq_vector[i].drv_inst = pm8001_ha;
- rc = request_irq(pm8001_ha->msix_entries[i].vector,
- pm8001_interrupt_handler_msix, flag,
- intr_drvname[i], &(pm8001_ha->irq_vector[i]));
- if (rc) {
- for (j = 0; j < i; j++)
- free_irq(
- pm8001_ha->msix_entries[j].vector,
+ rc = request_irq(pm8001_ha->msix_entries[i].vector,
+ pm8001_interrupt_handler_msix, flag,
+ intr_drvname[i], &(pm8001_ha->irq_vector[i]));
+ if (rc) {
+ for (j = 0; j < i; j++) {
+ free_irq(pm8001_ha->msix_entries[j].vector,
&(pm8001_ha->irq_vector[i]));
- pci_disable_msix(pm8001_ha->pdev);
- break;
}
+ pci_disable_msix(pm8001_ha->pdev);
+ break;
}
}
+
return rc;
}
#endif
/* Allocate memory for saving the template */
md_tmp = dma_alloc_coherent(&ha->pdev->dev, ha->fw_dump_tmplt_size,
&md_tmp_dma, GFP_KERNEL);
+ if (!md_tmp) {
+ ql4_printk(KERN_INFO, ha,
+ "scsi%ld: Failed to allocate DMA memory\n",
+ ha->host_no);
+ return;
+ }
/* Request template */
status = qla4xxx_get_minidump_template(ha, md_tmp_dma);
goto exit_get_chap;
}
- strncpy(password, chap_table->secret, QL4_CHAP_MAX_SECRET_LEN);
- strncpy(username, chap_table->name, QL4_CHAP_MAX_NAME_LEN);
+ strlcpy(password, chap_table->secret, QL4_CHAP_MAX_SECRET_LEN);
+ strlcpy(username, chap_table->name, QL4_CHAP_MAX_NAME_LEN);
chap_table->cookie = __constant_cpu_to_le16(CHAP_VALID_COOKIE);
exit_get_chap:
else
chap_table->flags |= BIT_7; /* local */
chap_table->secret_len = strlen(password);
- strncpy(chap_table->secret, password, MAX_CHAP_SECRET_LEN);
- strncpy(chap_table->name, username, MAX_CHAP_NAME_LEN);
+ strncpy(chap_table->secret, password, MAX_CHAP_SECRET_LEN - 1);
+ strncpy(chap_table->name, username, MAX_CHAP_NAME_LEN - 1);
chap_table->cookie = __constant_cpu_to_le16(CHAP_VALID_COOKIE);
if (is_qla40XX(ha)) {
goto exit_unlock_uni_chap;
}
- strncpy(password, chap_table->secret, MAX_CHAP_SECRET_LEN);
- strncpy(username, chap_table->name, MAX_CHAP_NAME_LEN);
+ strlcpy(password, chap_table->secret, MAX_CHAP_SECRET_LEN);
+ strlcpy(username, chap_table->name, MAX_CHAP_NAME_LEN);
rval = QLA_SUCCESS;
if (param == SET_DRVR_VERSION) {
mbox_cmd[1] = SET_DRVR_VERSION;
strncpy((char *)&mbox_cmd[2], QLA4XXX_DRIVER_VERSION,
- MAX_DRVR_VER_LEN);
+ MAX_DRVR_VER_LEN - 1);
} else {
ql4_printk(KERN_ERR, ha, "%s: invalid parameter 0x%x\n",
__func__, param);
for (i = 0; i < QLA_MSIX_ENTRIES; i++)
entries[i].entry = qla4_8xxx_msix_entries[i].entry;
- ret = pci_enable_msix(ha->pdev, entries, ARRAY_SIZE(entries));
+ ret = pci_enable_msix_exact(ha->pdev, entries, ARRAY_SIZE(entries));
if (ret) {
ql4_printk(KERN_WARNING, ha,
"MSI-X: Failed to enable support -- %d/%d\n",
continue;
chap_rec->chap_tbl_idx = i;
- strncpy(chap_rec->username, chap_table->name,
+ strlcpy(chap_rec->username, chap_table->name,
ISCSI_CHAP_AUTH_NAME_MAX_LEN);
- strncpy(chap_rec->password, chap_table->secret,
+ strlcpy(chap_rec->password, chap_table->secret,
QL4_CHAP_MAX_SECRET_LEN);
chap_rec->password_length = chap_table->secret_len;
if (!ql_iscsi_stats) {
ql4_printk(KERN_ERR, ha,
"Unable to allocate memory for iscsi stats\n");
+ ret = -ENOMEM;
goto exit_host_stats;
}
if (ret != QLA_SUCCESS) {
ql4_printk(KERN_ERR, ha,
"Unable to retrieve iscsi stats\n");
+ ret = -EIO;
goto exit_host_stats;
}
host_stats->mactx_frames = le64_to_cpu(ql_iscsi_stats->mac_tx_frames);
if (!(chap_table->flags & BIT_6)) /* Not BIDI */
continue;
- strncpy(password, chap_table->secret, QL4_CHAP_MAX_SECRET_LEN);
- strncpy(username, chap_table->name, QL4_CHAP_MAX_NAME_LEN);
+ strlcpy(password, chap_table->secret, QL4_CHAP_MAX_SECRET_LEN);
+ strlcpy(username, chap_table->name, QL4_CHAP_MAX_NAME_LEN);
ret = 0;
break;
}
tddb->tpgt = sess->tpgt;
tddb->port = conn->persistent_port;
- strncpy(tddb->iscsi_name, sess->targetname, ISCSI_NAME_SIZE);
- strncpy(tddb->ip_addr, conn->persistent_address, DDB_IPADDR_LEN);
+ strlcpy(tddb->iscsi_name, sess->targetname, ISCSI_NAME_SIZE);
+ strlcpy(tddb->ip_addr, conn->persistent_address, DDB_IPADDR_LEN);
}
static void qla4xxx_convert_param_ddb(struct dev_db_entry *fw_ddb_entry,
goto exit_ddb_logout;
}
- strncpy(flash_tddb->iscsi_name, fnode_sess->targetname,
+ strlcpy(flash_tddb->iscsi_name, fnode_sess->targetname,
ISCSI_NAME_SIZE);
if (!strncmp(fnode_sess->portal_type, PORTAL_TYPE_IPV6, 4))
pool->slab_flags |= SLAB_CACHE_DMA;
pool->gfp_mask = __GFP_DMA;
}
+
+ if (hostt->cmd_size)
+ hostt->cmd_pool = pool;
+
return pool;
}
out_free_slab:
kmem_cache_destroy(pool->cmd_slab);
out_free_pool:
- if (hostt->cmd_size)
+ if (hostt->cmd_size) {
scsi_free_host_cmd_pool(pool);
+ hostt->cmd_pool = NULL;
+ }
goto out;
}
if (!--pool->users) {
kmem_cache_destroy(pool->cmd_slab);
kmem_cache_destroy(pool->sense_slab);
- if (hostt->cmd_size)
+ if (hostt->cmd_size) {
scsi_free_host_cmd_pool(pool);
+ hostt->cmd_pool = NULL;
+ }
}
mutex_unlock(&host_cmd_pool_mutex);
}
blk_requeue_request(q, req);
atomic_dec(&sdev->device_busy);
out_delay:
- if (atomic_read(&sdev->device_busy) && !scsi_device_blocked(sdev))
+ if (!atomic_read(&sdev->device_busy) && !scsi_device_blocked(sdev))
blk_delay_queue(q, SCSI_QUEUE_DELAY);
}
char *buf;
if (!transport->get_host_stats)
- return -EINVAL;
+ return -ENOSYS;
priv = iscsi_if_transport_lookup(transport);
if (!priv)
memset(buf, 0, host_stats_size);
err = transport->get_host_stats(shost, buf, host_stats_size);
+ if (err) {
+ kfree_skb(skbhost_stats);
+ goto exit_host_stats;
+ }
actual_size = nlmsg_total_size(sizeof(*ev) + host_stats_size);
skb_trim(skbhost_stats, NLMSG_ALIGN(actual_size));
if (delay > 0)
queue_delayed_work(system_long_wq, &rport->reconnect_work,
1UL * delay * HZ);
- if (srp_rport_set_state(rport, SRP_RPORT_BLOCKED) == 0) {
+ if ((fast_io_fail_tmo >= 0 || dev_loss_tmo >= 0) &&
+ srp_rport_set_state(rport, SRP_RPORT_BLOCKED) == 0) {
pr_debug("%s new state: %d\n", dev_name(&shost->shost_gendev),
rport->state);
scsi_target_block(&shost->shost_gendev);
sh[j]->irq, dma_name, sh[j]->sg_tablesize, sh[j]->can_queue);
if (sh[j]->max_id > 8 || sh[j]->max_lun > 8)
- printk("%s: wide SCSI support enabled, max_id %u, max_lun %u.\n",
+ printk("%s: wide SCSI support enabled, max_id %u, max_lun %llu.\n",
BN(j), sh[j]->max_id, sh[j]->max_lun);
for (i = 0; i <= sh[j]->max_channel; i++)
cpp->cpp_index = i;
SCpnt->host_scribble = (unsigned char *) &cpp->cpp_index;
- if (do_trace) printk("%s: qcomm, mbox %d, target %d.%d:%llu.\n",
+ if (do_trace) printk("%s: qcomm, mbox %d, target %d.%d:%u.\n",
BN(j), i, SCpnt->device->channel, SCpnt->device->id,
(u8)SCpnt->device->lun);
.runtime_idle = ufshcd_pci_runtime_idle,
};
-static DEFINE_PCI_DEVICE_TABLE(ufshcd_pci_tbl) = {
+static const struct pci_device_id ufshcd_pci_tbl[] = {
{ PCI_VENDOR_ID_SAMSUNG, 0xC00C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 },
{ } /* terminate list */
};
*
* The byte channel to be used for the console is specified via a "stdout"
* property in the /chosen node.
- *
- * For compatible with legacy device trees, we also look for a "stdout" alias.
*/
static int find_console_handle(void)
{
- struct device_node *np, *np2;
+ struct device_node *np = of_stdout;
const char *sprop = NULL;
const uint32_t *iprop;
- np = of_find_node_by_path("/chosen");
- if (np)
- sprop = of_get_property(np, "stdout-path", NULL);
-
- if (!np || !sprop) {
- of_node_put(np);
- np = of_find_node_by_name(NULL, "aliases");
- if (np)
- sprop = of_get_property(np, "stdout", NULL);
- }
-
- if (!sprop) {
- of_node_put(np);
- return 0;
- }
-
/* We don't care what the aliased node is actually called. We only
* care if it's compatible with "epapr,hv-byte-channel", because that
- * indicates that it's a byte channel node. We use a temporary
- * variable, 'np2', because we can't release 'np' until we're done with
- * 'sprop'.
+ * indicates that it's a byte channel node.
*/
- np2 = of_find_node_by_path(sprop);
- of_node_put(np);
- np = np2;
- if (!np) {
- pr_warning("ehv-bc: stdout node '%s' does not exist\n", sprop);
- return 0;
- }
-
- /* Is it a byte channel? */
- if (!of_device_is_compatible(np, "epapr,hv-byte-channel")) {
- of_node_put(np);
+ if (!np || !of_device_is_compatible(np, "epapr,hv-byte-channel"))
return 0;
- }
stdout_irq = irq_of_parse_and_map(np, 0);
if (stdout_irq == NO_IRQ) {
- pr_err("ehv-bc: no 'interrupts' property in %s node\n", sprop);
- of_node_put(np);
+ pr_err("ehv-bc: no 'interrupts' property in %s node\n", np->full_name);
return 0;
}
if (!iprop) {
pr_err("ehv-bc: no 'hv-handle' property in %s node\n",
np->name);
- of_node_put(np);
return 0;
}
stdout_bc = be32_to_cpu(*iprop);
-
- of_node_put(np);
return 1;
}
void __init hvc_opal_init_early(void)
{
- struct device_node *stdout_node = NULL;
+ struct device_node *stdout_node = of_node_get(of_stdout);
const __be32 *termno;
- const char *name = NULL;
const struct hv_ops *ops;
u32 index;
- /* find the boot console from /chosen/stdout */
- if (of_chosen)
- name = of_get_property(of_chosen, "linux,stdout-path", NULL);
- if (name) {
- stdout_node = of_find_node_by_path(name);
- if (!stdout_node) {
- pr_err("hvc_opal: Failed to locate default console!\n");
- return;
- }
- } else {
+ /* If the console wasn't in /chosen, try /ibm,opal */
+ if (!stdout_node) {
struct device_node *opal, *np;
/* Current OPAL takeover doesn't provide the stdout
void __init hvc_vio_init_early(void)
{
- struct device_node *stdout_node;
const __be32 *termno;
const char *name;
const struct hv_ops *ops;
/* find the boot console from /chosen/stdout */
- if (!of_chosen)
+ if (!of_stdout)
return;
- name = of_get_property(of_chosen, "linux,stdout-path", NULL);
- if (name == NULL)
- return;
- stdout_node = of_find_node_by_path(name);
- if (!stdout_node)
- return;
- name = of_get_property(stdout_node, "name", NULL);
+ name = of_get_property(of_stdout, "name", NULL);
if (!name) {
printk(KERN_WARNING "stdout node missing 'name' property!\n");
- goto out;
+ return;
}
/* Check if it's a virtual terminal */
if (strncmp(name, "vty", 3) != 0)
- goto out;
- termno = of_get_property(stdout_node, "reg", NULL);
+ return;
+ termno = of_get_property(of_stdout, "reg", NULL);
if (termno == NULL)
- goto out;
+ return;
hvterm_priv0.termno = of_read_number(termno, 1);
spin_lock_init(&hvterm_priv0.buf_lock);
hvterm_privs[0] = &hvterm_priv0;
/* Check the protocol */
- if (of_device_is_compatible(stdout_node, "hvterm1")) {
+ if (of_device_is_compatible(of_stdout, "hvterm1")) {
hvterm_priv0.proto = HV_PROTOCOL_RAW;
ops = &hvterm_raw_ops;
}
- else if (of_device_is_compatible(stdout_node, "hvterm-protocol")) {
+ else if (of_device_is_compatible(of_stdout, "hvterm-protocol")) {
hvterm_priv0.proto = HV_PROTOCOL_HVSI;
ops = &hvterm_hvsi_ops;
hvsilib_init(&hvterm_priv0.hvsi, hvc_get_chars, hvc_put_chars,
/* HVSI, perform the handshake now */
hvsilib_establish(&hvterm_priv0.hvsi);
} else
- goto out;
+ return;
udbg_putc = udbg_hvc_putc;
udbg_getc = udbg_hvc_getc;
udbg_getc_poll = udbg_hvc_getc_poll;
* backend for HVSI, only do udbg
*/
if (hvterm_priv0.proto == HV_PROTOCOL_HVSI)
- goto out;
+ return;
#endif
/* Check whether the user has requested a different console. */
if (!strstr(cmd_line, "console="))
add_preferred_console("hvc", 0, NULL);
hvc_instantiate(0, 0, ops);
-out:
- of_node_put(stdout_node);
}
/* call this from early_init() for a working debug console on
/*
* Find all escc chips in the system
*/
- node_p = of_find_node_by_name(NULL, "escc");
- while (node_p) {
+ for_each_node_by_name(node_p, "escc") {
/*
* First get channel A/B node pointers
*
of_node_put(node_b);
printk(KERN_ERR "pmac_zilog: missing node %c for escc %s\n",
(!node_a) ? 'a' : 'b', node_p->full_name);
- goto next;
+ continue;
}
/*
of_node_put(node_b);
memset(&pmz_ports[count], 0, sizeof(struct uart_pmac_port));
memset(&pmz_ports[count+1], 0, sizeof(struct uart_pmac_port));
- goto next;
+ continue;
}
count += 2;
-next:
- node_p = of_find_node_by_name(node_p, "escc");
}
pmz_ports_count = count;
#include <linux/slab.h>
#include <linux/init.h>
#include <linux/console.h>
+#include <linux/of.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/device.h>
spin_lock_init(&uport->lock);
lockdep_set_class(&uport->lock, &port_lock_key);
}
+ if (uport->cons && uport->dev)
+ of_console_check(uport->dev->of_node, uport->cons->name, uport->line);
uart_configure_port(drv, state, uport);
depends on VFIO && SPAPR_TCE_IOMMU
default n
+config VFIO_SPAPR_EEH
+ tristate
+ depends on EEH && VFIO_IOMMU_SPAPR_TCE
+ default n
+
menuconfig VFIO
tristate "VFIO Non-Privileged userspace driver framework"
depends on IOMMU_API
select VFIO_IOMMU_TYPE1 if X86
select VFIO_IOMMU_SPAPR_TCE if (PPC_POWERNV || PPC_PSERIES)
+ select VFIO_SPAPR_EEH if (PPC_POWERNV || PPC_PSERIES)
select ANON_INODES
help
VFIO provides a framework for secure userspace device drivers.
obj-$(CONFIG_VFIO) += vfio.o
obj-$(CONFIG_VFIO_IOMMU_TYPE1) += vfio_iommu_type1.o
obj-$(CONFIG_VFIO_IOMMU_SPAPR_TCE) += vfio_iommu_spapr_tce.o
-obj-$(CONFIG_EEH) += vfio_spapr_eeh.o
+obj-$(CONFIG_VFIO_SPAPR_EEH) += vfio_spapr_eeh.o
obj-$(CONFIG_VFIO_PCI) += pci/
MODULE_PARM_DESC(nointxmask,
"Disable support for PCI 2.3 style INTx masking. If this resolves problems for specific devices, report lspci -vvvxxx to linux-pci@vger.kernel.org so the device can be fixed automatically via the broken_intx_masking flag.");
+static DEFINE_MUTEX(driver_lock);
+
+static void vfio_pci_try_bus_reset(struct vfio_pci_device *vdev);
+
static int vfio_pci_enable(struct vfio_pci_device *vdev)
{
struct pci_dev *pdev = vdev->pdev;
u16 cmd;
u8 msix_pos;
+ /* Don't allow our initial saved state to include busmaster */
+ pci_clear_master(pdev);
+
ret = pci_enable_device(pdev);
if (ret)
return ret;
struct pci_dev *pdev = vdev->pdev;
int bar;
- pci_disable_device(pdev);
+ /* Stop the device from further DMA */
+ pci_clear_master(pdev);
vfio_pci_set_irqs_ioctl(vdev, VFIO_IRQ_SET_DATA_NONE |
VFIO_IRQ_SET_ACTION_TRIGGER,
vdev->barmap[bar] = NULL;
}
+ vdev->needs_reset = true;
+
/*
* If we have saved state, restore it. If we can reset the device,
* even better. Resetting with current state seems better than
__func__, dev_name(&pdev->dev));
if (!vdev->reset_works)
- return;
+ goto out;
pci_save_state(pdev);
}
if (ret)
pr_warn("%s: Failed to reset device %s (%d)\n",
__func__, dev_name(&pdev->dev), ret);
+ else
+ vdev->needs_reset = false;
}
pci_restore_state(pdev);
+out:
+ pci_disable_device(pdev);
+
+ vfio_pci_try_bus_reset(vdev);
}
static void vfio_pci_release(void *device_data)
{
struct vfio_pci_device *vdev = device_data;
- if (atomic_dec_and_test(&vdev->refcnt)) {
+ mutex_lock(&driver_lock);
+
+ if (!(--vdev->refcnt)) {
vfio_spapr_pci_eeh_release(vdev->pdev);
vfio_pci_disable(vdev);
}
+ mutex_unlock(&driver_lock);
+
module_put(THIS_MODULE);
}
static int vfio_pci_open(void *device_data)
{
struct vfio_pci_device *vdev = device_data;
- int ret;
+ int ret = 0;
if (!try_module_get(THIS_MODULE))
return -ENODEV;
- if (atomic_inc_return(&vdev->refcnt) == 1) {
+ mutex_lock(&driver_lock);
+
+ if (!vdev->refcnt) {
ret = vfio_pci_enable(vdev);
if (ret)
goto error;
- ret = vfio_spapr_pci_eeh_open(vdev->pdev);
- if (ret) {
- vfio_pci_disable(vdev);
- goto error;
- }
+ vfio_spapr_pci_eeh_open(vdev->pdev);
}
-
- return 0;
+ vdev->refcnt++;
error:
- module_put(THIS_MODULE);
+ mutex_unlock(&driver_lock);
+ if (ret)
+ module_put(THIS_MODULE);
return ret;
}
vdev->irq_type = VFIO_PCI_NUM_IRQS;
mutex_init(&vdev->igate);
spin_lock_init(&vdev->irqlock);
- atomic_set(&vdev->refcnt, 0);
ret = vfio_add_group_dev(&pdev->dev, &vfio_pci_ops, vdev);
if (ret) {
{
struct vfio_pci_device *vdev;
+ mutex_lock(&driver_lock);
+
vdev = vfio_del_group_dev(&pdev->dev);
- if (!vdev)
- return;
+ if (vdev) {
+ iommu_group_put(pdev->dev.iommu_group);
+ kfree(vdev);
+ }
- iommu_group_put(pdev->dev.iommu_group);
- kfree(vdev);
+ mutex_unlock(&driver_lock);
}
static pci_ers_result_t vfio_pci_aer_err_detected(struct pci_dev *pdev,
.err_handler = &vfio_err_handlers,
};
+/*
+ * Test whether a reset is necessary and possible. We mark devices as
+ * needs_reset when they are released, but don't have a function-local reset
+ * available. If any of these exist in the affected devices, we want to do
+ * a bus/slot reset. We also need all of the affected devices to be unused,
+ * so we abort if any device has a non-zero refcnt. driver_lock prevents a
+ * device from being opened during the scan or unbound from vfio-pci.
+ */
+static int vfio_pci_test_bus_reset(struct pci_dev *pdev, void *data)
+{
+ bool *needs_reset = data;
+ struct pci_driver *pci_drv = ACCESS_ONCE(pdev->driver);
+ int ret = -EBUSY;
+
+ if (pci_drv == &vfio_pci_driver) {
+ struct vfio_device *device;
+ struct vfio_pci_device *vdev;
+
+ device = vfio_device_get_from_dev(&pdev->dev);
+ if (!device)
+ return ret;
+
+ vdev = vfio_device_data(device);
+ if (vdev) {
+ if (vdev->needs_reset)
+ *needs_reset = true;
+
+ if (!vdev->refcnt)
+ ret = 0;
+ }
+
+ vfio_device_put(device);
+ }
+
+ /*
+ * TODO: vfio-core considers groups to be viable even if some devices
+ * are attached to known drivers, like pci-stub or pcieport. We can't
+ * freeze devices from being unbound to those drivers like we can
+ * here though, so it would be racy to test for them. We also can't
+ * use device_lock() to prevent changes as that would interfere with
+ * PCI-core taking device_lock during bus reset. For now, we require
+ * devices to be bound to vfio-pci to get a bus/slot reset on release.
+ */
+
+ return ret;
+}
+
+/* Clear needs_reset on all affected devices after successful bus/slot reset */
+static int vfio_pci_clear_needs_reset(struct pci_dev *pdev, void *data)
+{
+ struct pci_driver *pci_drv = ACCESS_ONCE(pdev->driver);
+
+ if (pci_drv == &vfio_pci_driver) {
+ struct vfio_device *device;
+ struct vfio_pci_device *vdev;
+
+ device = vfio_device_get_from_dev(&pdev->dev);
+ if (!device)
+ return 0;
+
+ vdev = vfio_device_data(device);
+ if (vdev)
+ vdev->needs_reset = false;
+
+ vfio_device_put(device);
+ }
+
+ return 0;
+}
+
+/*
+ * Attempt to do a bus/slot reset if there are devices affected by a reset for
+ * this device that are needs_reset and all of the affected devices are unused
+ * (!refcnt). Callers of this function are required to hold driver_lock such
+ * that devices can not be unbound from vfio-pci or opened by a user while we
+ * test for and perform a bus/slot reset.
+ */
+static void vfio_pci_try_bus_reset(struct vfio_pci_device *vdev)
+{
+ bool needs_reset = false, slot = false;
+ int ret;
+
+ if (!pci_probe_reset_slot(vdev->pdev->slot))
+ slot = true;
+ else if (pci_probe_reset_bus(vdev->pdev->bus))
+ return;
+
+ if (vfio_pci_for_each_slot_or_bus(vdev->pdev,
+ vfio_pci_test_bus_reset,
+ &needs_reset, slot) || !needs_reset)
+ return;
+
+ if (slot)
+ ret = pci_try_reset_slot(vdev->pdev->slot);
+ else
+ ret = pci_try_reset_bus(vdev->pdev->bus);
+
+ if (ret)
+ return;
+
+ vfio_pci_for_each_slot_or_bus(vdev->pdev,
+ vfio_pci_clear_needs_reset, NULL, slot);
+}
+
static void __exit vfio_pci_cleanup(void)
{
pci_unregister_driver(&vfio_pci_driver);
bool extended_caps;
bool bardirty;
bool has_vga;
+ bool needs_reset;
struct pci_saved_state *pci_saved_state;
- atomic_t refcnt;
+ int refcnt;
struct eventfd_ctx *err_trigger;
};
* published by the Free Software Foundation.
*/
+#include <linux/module.h>
#include <linux/uaccess.h>
#include <linux/vfio.h>
#include <asm/eeh.h>
+#define DRIVER_VERSION "0.1"
+#define DRIVER_AUTHOR "Gavin Shan, IBM Corporation"
+#define DRIVER_DESC "VFIO IOMMU SPAPR EEH"
+
/* We might build address mapping here for "fast" path later */
-int vfio_spapr_pci_eeh_open(struct pci_dev *pdev)
+void vfio_spapr_pci_eeh_open(struct pci_dev *pdev)
{
- return eeh_dev_open(pdev);
+ eeh_dev_open(pdev);
}
+EXPORT_SYMBOL_GPL(vfio_spapr_pci_eeh_open);
void vfio_spapr_pci_eeh_release(struct pci_dev *pdev)
{
eeh_dev_release(pdev);
}
+EXPORT_SYMBOL_GPL(vfio_spapr_pci_eeh_release);
long vfio_spapr_iommu_eeh_ioctl(struct iommu_group *group,
unsigned int cmd, unsigned long arg)
return ret;
}
+EXPORT_SYMBOL(vfio_spapr_iommu_eeh_ioctl);
+
+MODULE_VERSION(DRIVER_VERSION);
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
}
-static DEFINE_PCI_DEVICE_TABLE(pci_stub_id_table) = {
+static const struct pci_device_id pci_stub_id_table[] = {
{
.vendor = PCI_VENDOR_ID_MICROSOFT,
.device = PCI_DEVICE_ID_HYPERV_VIDEO,
#define I740_ID_PCI 0x00d1
#define I740_ID_AGP 0x7800
-static DEFINE_PCI_DEVICE_TABLE(i740fb_id_table) = {
+static const struct pci_device_id i740fb_id_table[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, I740_ID_PCI) },
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, I740_ID_AGP) },
{ 0 }
}
}
-static DEFINE_PCI_DEVICE_TABLE(pcistub_ids) = {
+static const struct pci_device_id pcistub_ids[] = {
{
.vendor = PCI_ANY_ID,
.device = PCI_ANY_ID,
}
file->f_flags = O_RDWR;
- file->private_data = ctx;
return file;
}
static const struct dentry_operations ops = {
.d_dname = simple_dname,
};
- return mount_pseudo(fs_type, "aio:", NULL, &ops, 0xa10a10a1);
+ return mount_pseudo(fs_type, "aio:", NULL, &ops, AIO_RING_MAGIC);
}
/* aio_setup
struct aio_ring *ring;
spin_lock(&mm->ioctx_lock);
- rcu_read_lock();
- table = rcu_dereference(mm->ioctx_table);
+ table = rcu_dereference_raw(mm->ioctx_table);
while (1) {
if (table)
if (!table->table[i]) {
ctx->id = i;
table->table[i] = ctx;
- rcu_read_unlock();
spin_unlock(&mm->ioctx_lock);
/* While kioctx setup is in progress,
}
new_nr = (table ? table->nr : 1) * 4;
-
- rcu_read_unlock();
spin_unlock(&mm->ioctx_lock);
table = kzalloc(sizeof(*table) + sizeof(struct kioctx *) *
table->nr = new_nr;
spin_lock(&mm->ioctx_lock);
- rcu_read_lock();
- old = rcu_dereference(mm->ioctx_table);
+ old = rcu_dereference_raw(mm->ioctx_table);
if (!old) {
rcu_assign_pointer(mm->ioctx_table, table);
spin_lock(&mm->ioctx_lock);
- rcu_read_lock();
- table = rcu_dereference(mm->ioctx_table);
-
+ table = rcu_dereference_raw(mm->ioctx_table);
WARN_ON(ctx != table->table[ctx->id]);
table->table[ctx->id] = NULL;
- rcu_read_unlock();
spin_unlock(&mm->ioctx_lock);
/* percpu_ref_kill() will do the necessary call_rcu() */
*/
void exit_aio(struct mm_struct *mm)
{
- struct kioctx_table *table;
- struct kioctx *ctx;
- unsigned i = 0;
-
- while (1) {
- rcu_read_lock();
- table = rcu_dereference(mm->ioctx_table);
-
- do {
- if (!table || i >= table->nr) {
- rcu_read_unlock();
- rcu_assign_pointer(mm->ioctx_table, NULL);
- if (table)
- kfree(table);
- return;
- }
+ struct kioctx_table *table = rcu_dereference_raw(mm->ioctx_table);
+ int i;
- ctx = table->table[i++];
- } while (!ctx);
+ if (!table)
+ return;
- rcu_read_unlock();
+ for (i = 0; i < table->nr; ++i) {
+ struct kioctx *ctx = table->table[i];
+ if (!ctx)
+ continue;
/*
- * We don't need to bother with munmap() here -
- * exit_mmap(mm) is coming and it'll unmap everything.
- * Since aio_free_ring() uses non-zero ->mmap_size
- * as indicator that it needs to unmap the area,
- * just set it to 0; aio_free_ring() is the only
- * place that uses ->mmap_size, so it's safe.
+ * We don't need to bother with munmap() here - exit_mmap(mm)
+ * is coming and it'll unmap everything. And we simply can't,
+ * this is not necessarily our ->mm.
+ * Since kill_ioctx() uses non-zero ->mmap_size as indicator
+ * that it needs to unmap the area, just set it to 0.
*/
ctx->mmap_size = 0;
-
kill_ioctx(mm, ctx, NULL);
}
+
+ RCU_INIT_POINTER(mm->ioctx_table, NULL);
+ kfree(table);
}
static void put_reqs_available(struct kioctx *ctx, unsigned nr)
struct kioctx_cpu *kcpu;
unsigned long flags;
- preempt_disable();
- kcpu = this_cpu_ptr(ctx->cpu);
-
local_irq_save(flags);
+ kcpu = this_cpu_ptr(ctx->cpu);
kcpu->reqs_available += nr;
while (kcpu->reqs_available >= ctx->req_batch * 2) {
}
local_irq_restore(flags);
- preempt_enable();
}
static bool get_reqs_available(struct kioctx *ctx)
bool ret = false;
unsigned long flags;
- preempt_disable();
- kcpu = this_cpu_ptr(ctx->cpu);
-
local_irq_save(flags);
+ kcpu = this_cpu_ptr(ctx->cpu);
if (!kcpu->reqs_available) {
int old, avail = atomic_read(&ctx->reqs_available);
kcpu->reqs_available--;
out:
local_irq_restore(flags);
- preempt_enable();
return ret;
}
}
EXPORT_SYMBOL(aio_complete);
-/* aio_read_events
+/* aio_read_events_ring
* Pull an event off of the ioctx's event ring. Returns the number of
* events fetched
*/
if (compat)
ret = compat_rw_copy_check_uvector(rw,
(struct compat_iovec __user *)buf,
- *nr_segs, 1, *iovec, iovec);
+ *nr_segs, UIO_FASTIOV, *iovec, iovec);
else
#endif
ret = rw_copy_check_uvector(rw,
(struct iovec __user *)buf,
- *nr_segs, 1, *iovec, iovec);
+ *nr_segs, UIO_FASTIOV, *iovec, iovec);
if (ret < 0)
return ret;
}
/*
- * aio_setup_iocb:
- * Performs the initial checks and aio retry method
- * setup for the kiocb at the time of io submission.
+ * aio_run_iocb:
+ * Performs the initial checks and io submission.
*/
static ssize_t aio_run_iocb(struct kiocb *req, unsigned opcode,
char __user *buf, bool compat)
fmode_t mode;
aio_rw_op *rw_op;
rw_iter_op *iter_op;
- struct iovec inline_vec, *iovec = &inline_vec;
+ struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs;
struct iov_iter iter;
switch (opcode) {
if (!ret)
ret = rw_verify_area(rw, file, &req->ki_pos, req->ki_nbytes);
if (ret < 0) {
- if (iovec != &inline_vec)
+ if (iovec != inline_vecs)
kfree(iovec);
return ret;
}
return -EINVAL;
}
- if (iovec != &inline_vec)
+ if (iovec != inline_vecs)
kfree(iovec);
if (ret != -EIOCBQUEUED) {
}
if (ret > 0)
goto next;
- ret = ulist_add_merge(parents, eb->start,
- (uintptr_t)eie,
- (u64 *)&old, GFP_NOFS);
+ ret = ulist_add_merge_ptr(parents, eb->start,
+ eie, (void **)&old, GFP_NOFS);
if (ret < 0)
break;
if (!ret && extent_item_pos) {
ret = -EIO;
goto out;
}
+ btrfs_tree_read_lock(eb);
+ btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
ret = find_extent_in_eb(eb, bytenr,
*extent_item_pos, &eie);
+ btrfs_tree_read_unlock_blocking(eb);
free_extent_buffer(eb);
if (ret < 0)
goto out;
ref->inode_list = eie;
}
- ret = ulist_add_merge(refs, ref->parent,
- (uintptr_t)ref->inode_list,
- (u64 *)&eie, GFP_NOFS);
+ ret = ulist_add_merge_ptr(refs, ref->parent,
+ ref->inode_list,
+ (void **)&eie, GFP_NOFS);
if (ret < 0)
goto out;
if (!ret && extent_item_pos) {
*/
struct list_head delalloc_inodes;
- /*
- * list for tracking inodes that must be sent to disk before a
- * rename or truncate commit
- */
- struct list_head ordered_operations;
-
/* node for the red-black tree that links inodes in subvolume root */
struct rb_node rb_node;
WARN_ON(btrfs_header_generation(buf) > trans->transid);
if (new_root_objectid == BTRFS_TREE_RELOC_OBJECTID)
- ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+ ret = btrfs_inc_ref(trans, root, cow, 1);
else
- ret = btrfs_inc_ref(trans, root, cow, 0, 1);
+ ret = btrfs_inc_ref(trans, root, cow, 0);
if (ret)
return ret;
if ((owner == root->root_key.objectid ||
root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) &&
!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)) {
- ret = btrfs_inc_ref(trans, root, buf, 1, 1);
+ ret = btrfs_inc_ref(trans, root, buf, 1);
BUG_ON(ret); /* -ENOMEM */
if (root->root_key.objectid ==
BTRFS_TREE_RELOC_OBJECTID) {
- ret = btrfs_dec_ref(trans, root, buf, 0, 1);
+ ret = btrfs_dec_ref(trans, root, buf, 0);
BUG_ON(ret); /* -ENOMEM */
- ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+ ret = btrfs_inc_ref(trans, root, cow, 1);
BUG_ON(ret); /* -ENOMEM */
}
new_flags |= BTRFS_BLOCK_FLAG_FULL_BACKREF;
if (root->root_key.objectid ==
BTRFS_TREE_RELOC_OBJECTID)
- ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+ ret = btrfs_inc_ref(trans, root, cow, 1);
else
- ret = btrfs_inc_ref(trans, root, cow, 0, 1);
+ ret = btrfs_inc_ref(trans, root, cow, 0);
BUG_ON(ret); /* -ENOMEM */
}
if (new_flags != 0) {
if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF) {
if (root->root_key.objectid ==
BTRFS_TREE_RELOC_OBJECTID)
- ret = btrfs_inc_ref(trans, root, cow, 1, 1);
+ ret = btrfs_inc_ref(trans, root, cow, 1);
else
- ret = btrfs_inc_ref(trans, root, cow, 0, 1);
+ ret = btrfs_inc_ref(trans, root, cow, 0);
BUG_ON(ret); /* -ENOMEM */
- ret = btrfs_dec_ref(trans, root, buf, 1, 1);
+ ret = btrfs_dec_ref(trans, root, buf, 1);
BUG_ON(ret); /* -ENOMEM */
}
clean_tree_block(trans, root, buf);
u64 min_alloc_size, u64 empty_size, u64 hint_byte,
struct btrfs_key *ins, int is_data, int delalloc);
int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct extent_buffer *buf, int full_backref, int no_quota);
+ struct extent_buffer *buf, int full_backref);
int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct extent_buffer *buf, int full_backref, int no_quota);
+ struct extent_buffer *buf, int full_backref);
int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 bytenr, u64 num_bytes, u64 flags,
static void free_fs_root(struct btrfs_root *root);
static int btrfs_check_super_valid(struct btrfs_fs_info *fs_info,
int read_only);
-static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
- struct btrfs_root *root);
static void btrfs_destroy_ordered_extents(struct btrfs_root *root);
static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans,
struct btrfs_root *root);
btrfs_cleanup_transaction(root);
}
-static void btrfs_destroy_ordered_operations(struct btrfs_transaction *t,
- struct btrfs_root *root)
-{
- struct btrfs_inode *btrfs_inode;
- struct list_head splice;
-
- INIT_LIST_HEAD(&splice);
-
- mutex_lock(&root->fs_info->ordered_operations_mutex);
- spin_lock(&root->fs_info->ordered_root_lock);
-
- list_splice_init(&t->ordered_operations, &splice);
- while (!list_empty(&splice)) {
- btrfs_inode = list_entry(splice.next, struct btrfs_inode,
- ordered_operations);
-
- list_del_init(&btrfs_inode->ordered_operations);
- spin_unlock(&root->fs_info->ordered_root_lock);
-
- btrfs_invalidate_inodes(btrfs_inode->root);
-
- spin_lock(&root->fs_info->ordered_root_lock);
- }
-
- spin_unlock(&root->fs_info->ordered_root_lock);
- mutex_unlock(&root->fs_info->ordered_operations_mutex);
-}
-
static void btrfs_destroy_ordered_extents(struct btrfs_root *root)
{
struct btrfs_ordered_extent *ordered;
void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans,
struct btrfs_root *root)
{
- btrfs_destroy_ordered_operations(cur_trans, root);
-
btrfs_destroy_delayed_refs(cur_trans, root);
cur_trans->state = TRANS_STATE_COMMIT_START;
static int __btrfs_mod_ref(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct extent_buffer *buf,
- int full_backref, int inc, int no_quota)
+ int full_backref, int inc)
{
u64 bytenr;
u64 num_bytes;
key.offset -= btrfs_file_extent_offset(buf, fi);
ret = process_func(trans, root, bytenr, num_bytes,
parent, ref_root, key.objectid,
- key.offset, no_quota);
+ key.offset, 1);
if (ret)
goto fail;
} else {
num_bytes = btrfs_level_size(root, level - 1);
ret = process_func(trans, root, bytenr, num_bytes,
parent, ref_root, level - 1, 0,
- no_quota);
+ 1);
if (ret)
goto fail;
}
}
int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct extent_buffer *buf, int full_backref, int no_quota)
+ struct extent_buffer *buf, int full_backref)
{
- return __btrfs_mod_ref(trans, root, buf, full_backref, 1, no_quota);
+ return __btrfs_mod_ref(trans, root, buf, full_backref, 1);
}
int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root,
- struct extent_buffer *buf, int full_backref, int no_quota)
+ struct extent_buffer *buf, int full_backref)
{
- return __btrfs_mod_ref(trans, root, buf, full_backref, 0, no_quota);
+ return __btrfs_mod_ref(trans, root, buf, full_backref, 0);
}
static int write_one_cache_group(struct btrfs_trans_handle *trans,
wc->reada_slot = slot;
}
+static int account_leaf_items(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct extent_buffer *eb)
+{
+ int nr = btrfs_header_nritems(eb);
+ int i, extent_type, ret;
+ struct btrfs_key key;
+ struct btrfs_file_extent_item *fi;
+ u64 bytenr, num_bytes;
+
+ for (i = 0; i < nr; i++) {
+ btrfs_item_key_to_cpu(eb, &key, i);
+
+ if (key.type != BTRFS_EXTENT_DATA_KEY)
+ continue;
+
+ fi = btrfs_item_ptr(eb, i, struct btrfs_file_extent_item);
+ /* filter out non qgroup-accountable extents */
+ extent_type = btrfs_file_extent_type(eb, fi);
+
+ if (extent_type == BTRFS_FILE_EXTENT_INLINE)
+ continue;
+
+ bytenr = btrfs_file_extent_disk_bytenr(eb, fi);
+ if (!bytenr)
+ continue;
+
+ num_bytes = btrfs_file_extent_disk_num_bytes(eb, fi);
+
+ ret = btrfs_qgroup_record_ref(trans, root->fs_info,
+ root->objectid,
+ bytenr, num_bytes,
+ BTRFS_QGROUP_OPER_SUB_SUBTREE, 0);
+ if (ret)
+ return ret;
+ }
+ return 0;
+}
+
+/*
+ * Walk up the tree from the bottom, freeing leaves and any interior
+ * nodes which have had all slots visited. If a node (leaf or
+ * interior) is freed, the node above it will have it's slot
+ * incremented. The root node will never be freed.
+ *
+ * At the end of this function, we should have a path which has all
+ * slots incremented to the next position for a search. If we need to
+ * read a new node it will be NULL and the node above it will have the
+ * correct slot selected for a later read.
+ *
+ * If we increment the root nodes slot counter past the number of
+ * elements, 1 is returned to signal completion of the search.
+ */
+static int adjust_slots_upwards(struct btrfs_root *root,
+ struct btrfs_path *path, int root_level)
+{
+ int level = 0;
+ int nr, slot;
+ struct extent_buffer *eb;
+
+ if (root_level == 0)
+ return 1;
+
+ while (level <= root_level) {
+ eb = path->nodes[level];
+ nr = btrfs_header_nritems(eb);
+ path->slots[level]++;
+ slot = path->slots[level];
+ if (slot >= nr || level == 0) {
+ /*
+ * Don't free the root - we will detect this
+ * condition after our loop and return a
+ * positive value for caller to stop walking the tree.
+ */
+ if (level != root_level) {
+ btrfs_tree_unlock_rw(eb, path->locks[level]);
+ path->locks[level] = 0;
+
+ free_extent_buffer(eb);
+ path->nodes[level] = NULL;
+ path->slots[level] = 0;
+ }
+ } else {
+ /*
+ * We have a valid slot to walk back down
+ * from. Stop here so caller can process these
+ * new nodes.
+ */
+ break;
+ }
+
+ level++;
+ }
+
+ eb = path->nodes[root_level];
+ if (path->slots[root_level] >= btrfs_header_nritems(eb))
+ return 1;
+
+ return 0;
+}
+
+/*
+ * root_eb is the subtree root and is locked before this function is called.
+ */
+static int account_shared_subtree(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct extent_buffer *root_eb,
+ u64 root_gen,
+ int root_level)
+{
+ int ret = 0;
+ int level;
+ struct extent_buffer *eb = root_eb;
+ struct btrfs_path *path = NULL;
+
+ BUG_ON(root_level < 0 || root_level > BTRFS_MAX_LEVEL);
+ BUG_ON(root_eb == NULL);
+
+ if (!root->fs_info->quota_enabled)
+ return 0;
+
+ if (!extent_buffer_uptodate(root_eb)) {
+ ret = btrfs_read_buffer(root_eb, root_gen);
+ if (ret)
+ goto out;
+ }
+
+ if (root_level == 0) {
+ ret = account_leaf_items(trans, root, root_eb);
+ goto out;
+ }
+
+ path = btrfs_alloc_path();
+ if (!path)
+ return -ENOMEM;
+
+ /*
+ * Walk down the tree. Missing extent blocks are filled in as
+ * we go. Metadata is accounted every time we read a new
+ * extent block.
+ *
+ * When we reach a leaf, we account for file extent items in it,
+ * walk back up the tree (adjusting slot pointers as we go)
+ * and restart the search process.
+ */
+ extent_buffer_get(root_eb); /* For path */
+ path->nodes[root_level] = root_eb;
+ path->slots[root_level] = 0;
+ path->locks[root_level] = 0; /* so release_path doesn't try to unlock */
+walk_down:
+ level = root_level;
+ while (level >= 0) {
+ if (path->nodes[level] == NULL) {
+ int child_bsize = root->nodesize;
+ int parent_slot;
+ u64 child_gen;
+ u64 child_bytenr;
+
+ /* We need to get child blockptr/gen from
+ * parent before we can read it. */
+ eb = path->nodes[level + 1];
+ parent_slot = path->slots[level + 1];
+ child_bytenr = btrfs_node_blockptr(eb, parent_slot);
+ child_gen = btrfs_node_ptr_generation(eb, parent_slot);
+
+ eb = read_tree_block(root, child_bytenr, child_bsize,
+ child_gen);
+ if (!eb || !extent_buffer_uptodate(eb)) {
+ ret = -EIO;
+ goto out;
+ }
+
+ path->nodes[level] = eb;
+ path->slots[level] = 0;
+
+ btrfs_tree_read_lock(eb);
+ btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+ path->locks[level] = BTRFS_READ_LOCK_BLOCKING;
+
+ ret = btrfs_qgroup_record_ref(trans, root->fs_info,
+ root->objectid,
+ child_bytenr,
+ child_bsize,
+ BTRFS_QGROUP_OPER_SUB_SUBTREE,
+ 0);
+ if (ret)
+ goto out;
+
+ }
+
+ if (level == 0) {
+ ret = account_leaf_items(trans, root, path->nodes[level]);
+ if (ret)
+ goto out;
+
+ /* Nonzero return here means we completed our search */
+ ret = adjust_slots_upwards(root, path, root_level);
+ if (ret)
+ break;
+
+ /* Restart search with new slots */
+ goto walk_down;
+ }
+
+ level--;
+ }
+
+ ret = 0;
+out:
+ btrfs_free_path(path);
+
+ return ret;
+}
+
/*
* helper to process tree block while walking down the tree.
*
/* wc->stage == UPDATE_BACKREF */
if (!(wc->flags[level] & flag)) {
BUG_ON(!path->locks[level]);
- ret = btrfs_inc_ref(trans, root, eb, 1, wc->for_reloc);
+ ret = btrfs_inc_ref(trans, root, eb, 1);
BUG_ON(ret); /* -ENOMEM */
- ret = btrfs_dec_ref(trans, root, eb, 0, wc->for_reloc);
+ ret = btrfs_dec_ref(trans, root, eb, 0);
BUG_ON(ret); /* -ENOMEM */
ret = btrfs_set_disk_extent_flags(trans, root, eb->start,
eb->len, flag,
int level = wc->level;
int reada = 0;
int ret = 0;
+ bool need_account = false;
generation = btrfs_node_ptr_generation(path->nodes[level],
path->slots[level]);
if (wc->stage == DROP_REFERENCE) {
if (wc->refs[level - 1] > 1) {
+ need_account = true;
if (level == 1 &&
(wc->flags[0] & BTRFS_BLOCK_FLAG_FULL_BACKREF))
goto skip;
parent = 0;
}
+ if (need_account) {
+ ret = account_shared_subtree(trans, root, next,
+ generation, level - 1);
+ if (ret) {
+ printk_ratelimited(KERN_ERR "BTRFS: %s Error "
+ "%d accounting shared subtree. Quota "
+ "is out of sync, rescan required.\n",
+ root->fs_info->sb->s_id, ret);
+ }
+ }
ret = btrfs_free_extent(trans, root, bytenr, blocksize, parent,
root->root_key.objectid, level - 1, 0, 0);
BUG_ON(ret); /* -ENOMEM */
if (wc->refs[level] == 1) {
if (level == 0) {
if (wc->flags[level] & BTRFS_BLOCK_FLAG_FULL_BACKREF)
- ret = btrfs_dec_ref(trans, root, eb, 1,
- wc->for_reloc);
+ ret = btrfs_dec_ref(trans, root, eb, 1);
else
- ret = btrfs_dec_ref(trans, root, eb, 0,
- wc->for_reloc);
+ ret = btrfs_dec_ref(trans, root, eb, 0);
BUG_ON(ret); /* -ENOMEM */
+ ret = account_leaf_items(trans, root, eb);
+ if (ret) {
+ printk_ratelimited(KERN_ERR "BTRFS: %s Error "
+ "%d accounting leaf items. Quota "
+ "is out of sync, rescan required.\n",
+ root->fs_info->sb->s_id, ret);
+ }
}
/* make block locked assertion in clean_tree_block happy */
if (!path->locks[level] &&
int level;
bool root_dropped = false;
+ btrfs_debug(root->fs_info, "Drop subvolume %llu", root->objectid);
+
path = btrfs_alloc_path();
if (!path) {
err = -ENOMEM;
goto out_end_trans;
}
+ /*
+ * Qgroup update accounting is run from
+ * delayed ref handling. This usually works
+ * out because delayed refs are normally the
+ * only way qgroup updates are added. However,
+ * we may have added updates during our tree
+ * walk so run qgroups here to make sure we
+ * don't lose any updates.
+ */
+ ret = btrfs_delayed_qgroup_accounting(trans,
+ root->fs_info);
+ if (ret)
+ printk_ratelimited(KERN_ERR "BTRFS: Failure %d "
+ "running qgroup updates "
+ "during snapshot delete. "
+ "Quota is out of sync, "
+ "rescan required.\n", ret);
+
btrfs_end_transaction_throttle(trans, tree_root);
if (!for_reloc && btrfs_need_cleaner_sleep(root)) {
pr_debug("BTRFS: drop snapshot early exit\n");
}
root_dropped = true;
out_end_trans:
+ ret = btrfs_delayed_qgroup_accounting(trans, tree_root->fs_info);
+ if (ret)
+ printk_ratelimited(KERN_ERR "BTRFS: Failure %d "
+ "running qgroup updates "
+ "during snapshot delete. "
+ "Quota is out of sync, "
+ "rescan required.\n", ret);
+
btrfs_end_transaction_throttle(trans, tree_root);
out_free:
kfree(wc);
found_next = 1;
if (ret != 0)
goto insert;
- slot = 0;
+ slot = path->slots[0];
}
btrfs_item_key_to_cpu(path->nodes[0], &found_key, slot);
if (found_key.objectid != BTRFS_EXTENT_CSUM_OBJECTID ||
int btrfs_release_file(struct inode *inode, struct file *filp)
{
- /*
- * ordered_data_close is set by settattr when we are about to truncate
- * a file from a non-zero size to a zero size. This tries to
- * flush down new bytes that may have been written if the
- * application were using truncate to replace a file in place.
- */
- if (test_and_clear_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
- &BTRFS_I(inode)->runtime_flags)) {
- struct btrfs_trans_handle *trans;
- struct btrfs_root *root = BTRFS_I(inode)->root;
-
- /*
- * We need to block on a committing transaction to keep us from
- * throwing a ordered operation on to the list and causing
- * something like sync to deadlock trying to flush out this
- * inode.
- */
- trans = btrfs_start_transaction(root, 0);
- if (IS_ERR(trans))
- return PTR_ERR(trans);
- btrfs_add_ordered_operation(trans, BTRFS_I(inode)->root, inode);
- btrfs_end_transaction(trans, root);
- if (inode->i_size > BTRFS_ORDERED_OPERATIONS_FLUSH_LIMIT)
- filemap_flush(inode->i_mapping);
- }
if (filp->private_data)
btrfs_ioctl_trans_end(filp);
+ filemap_flush(inode->i_mapping);
return 0;
}
unlock_extent(io_tree, async_extent->start,
async_extent->start +
async_extent->ram_size - 1);
+
+ /*
+ * we need to redirty the pages if we decide to
+ * fallback to uncompressed IO, otherwise we
+ * will not submit these pages down to lower
+ * layers.
+ */
+ extent_range_redirty_for_io(inode,
+ async_extent->start,
+ async_extent->start +
+ async_extent->ram_size - 1);
+
goto retry;
}
goto out_free;
min_size);
BUG_ON(ret);
- /*
- * setattr is responsible for setting the ordered_data_close flag,
- * but that is only tested during the last file release. That
- * could happen well after the next commit, leaving a great big
- * window where new writes may get lost if someone chooses to write
- * to this file after truncating to zero
- *
- * The inode doesn't have any dirty data here, and so if we commit
- * this is a noop. If someone immediately starts writing to the inode
- * it is very likely we'll catch some of their writes in this
- * transaction, and the commit will find this file on the ordered
- * data list with good things to send down.
- *
- * This is a best effort solution, there is still a window where
- * using truncate to replace the contents of the file will
- * end up with a zero length file after a crash.
- */
- if (inode->i_size == 0 && test_bit(BTRFS_INODE_ORDERED_DATA_CLOSE,
- &BTRFS_I(inode)->runtime_flags))
- btrfs_add_ordered_operation(trans, root, inode);
-
/*
* So if we truncate and then write and fsync we normally would just
* write the extents that changed, which is a problem if we need to
mutex_init(&ei->delalloc_mutex);
btrfs_ordered_inode_tree_init(&ei->ordered_tree);
INIT_LIST_HEAD(&ei->delalloc_inodes);
- INIT_LIST_HEAD(&ei->ordered_operations);
RB_CLEAR_NODE(&ei->rb_node);
return inode;
if (!root)
goto free;
- /*
- * Make sure we're properly removed from the ordered operation
- * lists.
- */
- smp_mb();
- if (!list_empty(&BTRFS_I(inode)->ordered_operations)) {
- spin_lock(&root->fs_info->ordered_root_lock);
- list_del_init(&BTRFS_I(inode)->ordered_operations);
- spin_unlock(&root->fs_info->ordered_root_lock);
- }
-
if (test_bit(BTRFS_INODE_HAS_ORPHAN_ITEM,
&BTRFS_I(inode)->runtime_flags)) {
btrfs_info(root->fs_info, "inode %llu still on the orphan list",
ret = 0;
/*
- * we're using rename to replace one file with another.
- * and the replacement file is large. Start IO on it now so
- * we don't add too much work to the end of the transaction
+ * we're using rename to replace one file with another. Start IO on it
+ * now so we don't add too much work to the end of the transaction
*/
- if (new_inode && S_ISREG(old_inode->i_mode) && new_inode->i_size &&
- old_inode->i_size > BTRFS_ORDERED_OPERATIONS_FLUSH_LIMIT)
+ if (new_inode && S_ISREG(old_inode->i_mode) && new_inode->i_size)
filemap_flush(old_inode->i_mapping);
/* close the racy window with snapshot create/destroy ioctl */
*/
btrfs_pin_log_trans(root);
}
- /*
- * make sure the inode gets flushed if it is replacing
- * something.
- */
- if (new_inode && new_inode->i_size && S_ISREG(old_inode->i_mode))
- btrfs_add_ordered_operation(trans, root, old_inode);
inode_inc_iversion(old_dir);
inode_inc_iversion(new_dir);
trace_btrfs_ordered_extent_remove(inode, entry);
- /*
- * we have no more ordered extents for this inode and
- * no dirty pages. We can safely remove it from the
- * list of ordered extents
- */
- if (RB_EMPTY_ROOT(&tree->tree) &&
- !mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) {
- spin_lock(&root->fs_info->ordered_root_lock);
- list_del_init(&BTRFS_I(inode)->ordered_operations);
- spin_unlock(&root->fs_info->ordered_root_lock);
- }
-
if (!root->nr_ordered_extents) {
spin_lock(&root->fs_info->ordered_root_lock);
BUG_ON(list_empty(&root->ordered_root));
mutex_unlock(&fs_info->ordered_operations_mutex);
}
-/*
- * this is used during transaction commit to write all the inodes
- * added to the ordered operation list. These files must be fully on
- * disk before the transaction commits.
- *
- * we have two modes here, one is to just start the IO via filemap_flush
- * and the other is to wait for all the io. When we wait, we have an
- * extra check to make sure the ordered operation list really is empty
- * before we return
- */
-int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, int wait)
-{
- struct btrfs_inode *btrfs_inode;
- struct inode *inode;
- struct btrfs_transaction *cur_trans = trans->transaction;
- struct list_head splice;
- struct list_head works;
- struct btrfs_delalloc_work *work, *next;
- int ret = 0;
-
- INIT_LIST_HEAD(&splice);
- INIT_LIST_HEAD(&works);
-
- mutex_lock(&root->fs_info->ordered_extent_flush_mutex);
- spin_lock(&root->fs_info->ordered_root_lock);
- list_splice_init(&cur_trans->ordered_operations, &splice);
- while (!list_empty(&splice)) {
- btrfs_inode = list_entry(splice.next, struct btrfs_inode,
- ordered_operations);
- inode = &btrfs_inode->vfs_inode;
-
- list_del_init(&btrfs_inode->ordered_operations);
-
- /*
- * the inode may be getting freed (in sys_unlink path).
- */
- inode = igrab(inode);
- if (!inode)
- continue;
-
- if (!wait)
- list_add_tail(&BTRFS_I(inode)->ordered_operations,
- &cur_trans->ordered_operations);
- spin_unlock(&root->fs_info->ordered_root_lock);
-
- work = btrfs_alloc_delalloc_work(inode, wait, 1);
- if (!work) {
- spin_lock(&root->fs_info->ordered_root_lock);
- if (list_empty(&BTRFS_I(inode)->ordered_operations))
- list_add_tail(&btrfs_inode->ordered_operations,
- &splice);
- list_splice_tail(&splice,
- &cur_trans->ordered_operations);
- spin_unlock(&root->fs_info->ordered_root_lock);
- ret = -ENOMEM;
- goto out;
- }
- list_add_tail(&work->list, &works);
- btrfs_queue_work(root->fs_info->flush_workers,
- &work->work);
-
- cond_resched();
- spin_lock(&root->fs_info->ordered_root_lock);
- }
- spin_unlock(&root->fs_info->ordered_root_lock);
-out:
- list_for_each_entry_safe(work, next, &works, list) {
- list_del_init(&work->list);
- btrfs_wait_and_free_delalloc_work(work);
- }
- mutex_unlock(&root->fs_info->ordered_extent_flush_mutex);
- return ret;
-}
-
/*
* Used to start IO or wait for a given ordered extent to finish.
*
return index;
}
-
-/*
- * add a given inode to the list of inodes that must be fully on
- * disk before a transaction commit finishes.
- *
- * This basically gives us the ext3 style data=ordered mode, and it is mostly
- * used to make sure renamed files are fully on disk.
- *
- * It is a noop if the inode is already fully on disk.
- *
- * If trans is not null, we'll do a friendly check for a transaction that
- * is already flushing things and force the IO down ourselves.
- */
-void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode)
-{
- struct btrfs_transaction *cur_trans = trans->transaction;
- u64 last_mod;
-
- last_mod = max(BTRFS_I(inode)->generation, BTRFS_I(inode)->last_trans);
-
- /*
- * if this file hasn't been changed since the last transaction
- * commit, we can safely return without doing anything
- */
- if (last_mod <= root->fs_info->last_trans_committed)
- return;
-
- spin_lock(&root->fs_info->ordered_root_lock);
- if (list_empty(&BTRFS_I(inode)->ordered_operations)) {
- list_add_tail(&BTRFS_I(inode)->ordered_operations,
- &cur_trans->ordered_operations);
- }
- spin_unlock(&root->fs_info->ordered_root_lock);
-}
-
int __init ordered_data_init(void)
{
btrfs_ordered_extent_cache = kmem_cache_create("btrfs_ordered_extent",
struct btrfs_ordered_extent *ordered);
int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr,
u32 *sum, int len);
-int btrfs_run_ordered_operations(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, int wait);
-void btrfs_add_ordered_operation(struct btrfs_trans_handle *trans,
- struct btrfs_root *root,
- struct inode *inode);
int btrfs_wait_ordered_extents(struct btrfs_root *root, int nr);
void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, int nr);
void btrfs_get_logged_extents(struct inode *inode,
mutex_unlock(&fs_info->qgroup_ioctl_lock);
return ret;
}
+
+static int comp_oper_exist(struct btrfs_qgroup_operation *oper1,
+ struct btrfs_qgroup_operation *oper2)
+{
+ /*
+ * Ignore seq and type here, we're looking for any operation
+ * at all related to this extent on that root.
+ */
+ if (oper1->bytenr < oper2->bytenr)
+ return -1;
+ if (oper1->bytenr > oper2->bytenr)
+ return 1;
+ if (oper1->ref_root < oper2->ref_root)
+ return -1;
+ if (oper1->ref_root > oper2->ref_root)
+ return 1;
+ return 0;
+}
+
+static int qgroup_oper_exists(struct btrfs_fs_info *fs_info,
+ struct btrfs_qgroup_operation *oper)
+{
+ struct rb_node *n;
+ struct btrfs_qgroup_operation *cur;
+ int cmp;
+
+ spin_lock(&fs_info->qgroup_op_lock);
+ n = fs_info->qgroup_op_tree.rb_node;
+ while (n) {
+ cur = rb_entry(n, struct btrfs_qgroup_operation, n);
+ cmp = comp_oper_exist(cur, oper);
+ if (cmp < 0) {
+ n = n->rb_right;
+ } else if (cmp) {
+ n = n->rb_left;
+ } else {
+ spin_unlock(&fs_info->qgroup_op_lock);
+ return -EEXIST;
+ }
+ }
+ spin_unlock(&fs_info->qgroup_op_lock);
+ return 0;
+}
+
static int comp_oper(struct btrfs_qgroup_operation *oper1,
struct btrfs_qgroup_operation *oper2)
{
oper->seq = atomic_inc_return(&fs_info->qgroup_op_seq);
INIT_LIST_HEAD(&oper->elem.list);
oper->elem.seq = 0;
+
+ if (type == BTRFS_QGROUP_OPER_SUB_SUBTREE) {
+ /*
+ * If any operation for this bytenr/ref_root combo
+ * exists, then we know it's not exclusively owned and
+ * shouldn't be queued up.
+ *
+ * This also catches the case where we have a cloned
+ * extent that gets queued up multiple times during
+ * drop snapshot.
+ */
+ if (qgroup_oper_exists(fs_info, oper)) {
+ kfree(oper);
+ return 0;
+ }
+ }
+
ret = insert_qgroup_oper(fs_info, oper);
if (ret) {
/* Shouldn't happen so have an assert for developers */
return ret;
}
+/*
+ * Process a reference to a shared subtree. This type of operation is
+ * queued during snapshot removal when we encounter extents which are
+ * shared between more than one root.
+ */
+static int qgroup_subtree_accounting(struct btrfs_trans_handle *trans,
+ struct btrfs_fs_info *fs_info,
+ struct btrfs_qgroup_operation *oper)
+{
+ struct ulist *roots = NULL;
+ struct ulist_node *unode;
+ struct ulist_iterator uiter;
+ struct btrfs_qgroup_list *glist;
+ struct ulist *parents;
+ int ret = 0;
+ int err;
+ struct btrfs_qgroup *qg;
+ u64 root_obj = 0;
+ struct seq_list elem = {};
+
+ parents = ulist_alloc(GFP_NOFS);
+ if (!parents)
+ return -ENOMEM;
+
+ btrfs_get_tree_mod_seq(fs_info, &elem);
+ ret = btrfs_find_all_roots(trans, fs_info, oper->bytenr,
+ elem.seq, &roots);
+ btrfs_put_tree_mod_seq(fs_info, &elem);
+ if (ret < 0)
+ return ret;
+
+ if (roots->nnodes != 1)
+ goto out;
+
+ ULIST_ITER_INIT(&uiter);
+ unode = ulist_next(roots, &uiter); /* Only want 1 so no need to loop */
+ /*
+ * If we find our ref root then that means all refs
+ * this extent has to the root have not yet been
+ * deleted. In that case, we do nothing and let the
+ * last ref for this bytenr drive our update.
+ *
+ * This can happen for example if an extent is
+ * referenced multiple times in a snapshot (clone,
+ * etc). If we are in the middle of snapshot removal,
+ * queued updates for such an extent will find the
+ * root if we have not yet finished removing the
+ * snapshot.
+ */
+ if (unode->val == oper->ref_root)
+ goto out;
+
+ root_obj = unode->val;
+ BUG_ON(!root_obj);
+
+ spin_lock(&fs_info->qgroup_lock);
+ qg = find_qgroup_rb(fs_info, root_obj);
+ if (!qg)
+ goto out_unlock;
+
+ qg->excl += oper->num_bytes;
+ qg->excl_cmpr += oper->num_bytes;
+ qgroup_dirty(fs_info, qg);
+
+ /*
+ * Adjust counts for parent groups. First we find all
+ * parents, then in the 2nd loop we do the adjustment
+ * while adding parents of the parents to our ulist.
+ */
+ list_for_each_entry(glist, &qg->groups, next_group) {
+ err = ulist_add(parents, glist->group->qgroupid,
+ ptr_to_u64(glist->group), GFP_ATOMIC);
+ if (err < 0) {
+ ret = err;
+ goto out_unlock;
+ }
+ }
+
+ ULIST_ITER_INIT(&uiter);
+ while ((unode = ulist_next(parents, &uiter))) {
+ qg = u64_to_ptr(unode->aux);
+ qg->excl += oper->num_bytes;
+ qg->excl_cmpr += oper->num_bytes;
+ qgroup_dirty(fs_info, qg);
+
+ /* Add any parents of the parents */
+ list_for_each_entry(glist, &qg->groups, next_group) {
+ err = ulist_add(parents, glist->group->qgroupid,
+ ptr_to_u64(glist->group), GFP_ATOMIC);
+ if (err < 0) {
+ ret = err;
+ goto out_unlock;
+ }
+ }
+ }
+
+out_unlock:
+ spin_unlock(&fs_info->qgroup_lock);
+
+out:
+ ulist_free(roots);
+ ulist_free(parents);
+ return ret;
+}
+
/*
* btrfs_qgroup_account_ref is called for every ref that is added to or deleted
* from the fs. First, all roots referencing the extent are searched, and
case BTRFS_QGROUP_OPER_SUB_SHARED:
ret = qgroup_shared_accounting(trans, fs_info, oper);
break;
+ case BTRFS_QGROUP_OPER_SUB_SUBTREE:
+ ret = qgroup_subtree_accounting(trans, fs_info, oper);
+ break;
default:
ASSERT(0);
}
BTRFS_QGROUP_OPER_ADD_SHARED,
BTRFS_QGROUP_OPER_SUB_EXCL,
BTRFS_QGROUP_OPER_SUB_SHARED,
+ BTRFS_QGROUP_OPER_SUB_SUBTREE,
};
struct btrfs_qgroup_operation {
return 0;
}
+/*
+ * Calculate numbers for 'df', pessimistic in case of mixed raid profiles.
+ *
+ * If there's a redundant raid level at DATA block groups, use the respective
+ * multiplier to scale the sizes.
+ *
+ * Unused device space usage is based on simulating the chunk allocator
+ * algorithm that respects the device sizes, order of allocations and the
+ * 'alloc_start' value, this is a close approximation of the actual use but
+ * there are other factors that may change the result (like a new metadata
+ * chunk).
+ *
+ * FIXME: not accurate for mixed block groups, total and free/used are ok,
+ * available appears slightly larger.
+ */
static int btrfs_statfs(struct dentry *dentry, struct kstatfs *buf)
{
struct btrfs_fs_info *fs_info = btrfs_sb(dentry->d_sb);
u64 total_free_data = 0;
int bits = dentry->d_sb->s_blocksize_bits;
__be32 *fsid = (__be32 *)fs_info->fsid;
+ unsigned factor = 1;
+ struct btrfs_block_rsv *block_rsv = &fs_info->global_block_rsv;
int ret;
/* holding chunk_muext to avoid allocating new chunks */
rcu_read_lock();
list_for_each_entry_rcu(found, head, list) {
if (found->flags & BTRFS_BLOCK_GROUP_DATA) {
+ int i;
+
total_free_data += found->disk_total - found->disk_used;
total_free_data -=
btrfs_account_ro_block_groups_free_space(found);
+
+ for (i = 0; i < BTRFS_NR_RAID_TYPES; i++) {
+ if (!list_empty(&found->block_groups[i])) {
+ switch (i) {
+ case BTRFS_RAID_DUP:
+ case BTRFS_RAID_RAID1:
+ case BTRFS_RAID_RAID10:
+ factor = 2;
+ }
+ }
+ }
}
total_used += found->disk_used;
}
+
rcu_read_unlock();
- buf->f_namelen = BTRFS_NAME_LEN;
- buf->f_blocks = btrfs_super_total_bytes(disk_super) >> bits;
- buf->f_bfree = buf->f_blocks - (total_used >> bits);
- buf->f_bsize = dentry->d_sb->s_blocksize;
- buf->f_type = BTRFS_SUPER_MAGIC;
+ buf->f_blocks = div_u64(btrfs_super_total_bytes(disk_super), factor);
+ buf->f_blocks >>= bits;
+ buf->f_bfree = buf->f_blocks - (div_u64(total_used, factor) >> bits);
+
+ /* Account global block reserve as used, it's in logical size already */
+ spin_lock(&block_rsv->lock);
+ buf->f_bfree -= block_rsv->size >> bits;
+ spin_unlock(&block_rsv->lock);
+
buf->f_bavail = total_free_data;
ret = btrfs_calc_avail_data_space(fs_info->tree_root, &total_free_data);
if (ret) {
mutex_unlock(&fs_info->chunk_mutex);
return ret;
}
- buf->f_bavail += total_free_data;
+ buf->f_bavail += div_u64(total_free_data, factor);
buf->f_bavail = buf->f_bavail >> bits;
mutex_unlock(&fs_info->chunk_mutex);
+ buf->f_type = BTRFS_SUPER_MAGIC;
+ buf->f_bsize = dentry->d_sb->s_blocksize;
+ buf->f_namelen = BTRFS_NAME_LEN;
+
/* We treat it as constant endianness (it doesn't matter _which_)
because we want the fsid to come out the same whether mounted
on a big-endian or little-endian host */
spin_lock_init(&cur_trans->delayed_refs.lock);
INIT_LIST_HEAD(&cur_trans->pending_snapshots);
- INIT_LIST_HEAD(&cur_trans->ordered_operations);
INIT_LIST_HEAD(&cur_trans->pending_chunks);
INIT_LIST_HEAD(&cur_trans->switch_commits);
list_add_tail(&cur_trans->list, &fs_info->trans_list);
kmem_cache_free(btrfs_trans_handle_cachep, trans);
}
-static int btrfs_flush_all_pending_stuffs(struct btrfs_trans_handle *trans,
- struct btrfs_root *root)
-{
- int ret;
-
- ret = btrfs_run_delayed_items(trans, root);
- if (ret)
- return ret;
-
- /*
- * rename don't use btrfs_join_transaction, so, once we
- * set the transaction to blocked above, we aren't going
- * to get any new ordered operations. We can safely run
- * it here and no for sure that nothing new will be added
- * to the list
- */
- ret = btrfs_run_ordered_operations(trans, root, 1);
-
- return ret;
-}
-
static inline int btrfs_start_delalloc_flush(struct btrfs_fs_info *fs_info)
{
if (btrfs_test_opt(fs_info->tree_root, FLUSHONCOMMIT))
struct btrfs_transaction *prev_trans = NULL;
int ret;
- ret = btrfs_run_ordered_operations(trans, root, 0);
- if (ret) {
- btrfs_abort_transaction(trans, root, ret);
- btrfs_end_transaction(trans, root);
- return ret;
- }
-
/* Stop the commit early if ->aborted is set */
if (unlikely(ACCESS_ONCE(cur_trans->aborted))) {
ret = cur_trans->aborted;
if (ret)
goto cleanup_transaction;
- ret = btrfs_flush_all_pending_stuffs(trans, root);
+ ret = btrfs_run_delayed_items(trans, root);
if (ret)
goto cleanup_transaction;
extwriter_counter_read(cur_trans) == 0);
/* some pending stuffs might be added after the previous flush. */
- ret = btrfs_flush_all_pending_stuffs(trans, root);
+ ret = btrfs_run_delayed_items(trans, root);
if (ret)
goto cleanup_transaction;
wait_queue_head_t writer_wait;
wait_queue_head_t commit_wait;
struct list_head pending_snapshots;
- struct list_head ordered_operations;
struct list_head pending_chunks;
struct list_head switch_commits;
struct btrfs_delayed_ref_root delayed_refs;
int ulist_add(struct ulist *ulist, u64 val, u64 aux, gfp_t gfp_mask);
int ulist_add_merge(struct ulist *ulist, u64 val, u64 aux,
u64 *old_aux, gfp_t gfp_mask);
+
+/* just like ulist_add_merge() but take a pointer for the aux data */
+static inline int ulist_add_merge_ptr(struct ulist *ulist, u64 val, void *aux,
+ void **old_aux, gfp_t gfp_mask)
+{
+#if BITS_PER_LONG == 32
+ u64 old64 = (uintptr_t)*old_aux;
+ int ret = ulist_add_merge(ulist, val, (uintptr_t)aux, &old64, gfp_mask);
+ *old_aux = (void *)((uintptr_t)old64);
+ return ret;
+#else
+ return ulist_add_merge(ulist, val, (u64)aux, (u64 *)old_aux, gfp_mask);
+#endif
+}
+
struct ulist_node *ulist_next(struct ulist *ulist,
struct ulist_iterator *uiter);
return 0;
}
+static long cifs_fallocate(struct file *file, int mode, loff_t off, loff_t len)
+{
+ struct super_block *sb = file->f_path.dentry->d_sb;
+ struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
+ struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb);
+ struct TCP_Server_Info *server = tcon->ses->server;
+
+ if (server->ops->fallocate)
+ return server->ops->fallocate(file, tcon, mode, off, len);
+
+ return -EOPNOTSUPP;
+}
+
static int cifs_permission(struct inode *inode, int mask)
{
struct cifs_sb_info *cifs_sb;
if (!(S_ISREG(inode->i_mode)))
return -EINVAL;
- /* check if file is oplocked */
- if (((arg == F_RDLCK) && CIFS_CACHE_READ(CIFS_I(inode))) ||
+ /* Check if file is oplocked if this is request for new lease */
+ if (arg == F_UNLCK ||
+ ((arg == F_RDLCK) && CIFS_CACHE_READ(CIFS_I(inode))) ||
((arg == F_WRLCK) && CIFS_CACHE_WRITE(CIFS_I(inode))))
return generic_setlease(file, arg, lease);
else if (tlink_tcon(cfile->tlink)->local_lease &&
.unlocked_ioctl = cifs_ioctl,
#endif /* CONFIG_CIFS_POSIX */
.setlease = cifs_setlease,
+ .fallocate = cifs_fallocate,
};
const struct file_operations cifs_file_strict_ops = {
.unlocked_ioctl = cifs_ioctl,
#endif /* CONFIG_CIFS_POSIX */
.setlease = cifs_setlease,
+ .fallocate = cifs_fallocate,
};
const struct file_operations cifs_file_direct_ops = {
#endif /* CONFIG_CIFS_POSIX */
.llseek = cifs_llseek,
.setlease = cifs_setlease,
+ .fallocate = cifs_fallocate,
};
const struct file_operations cifs_file_nobrl_ops = {
.unlocked_ioctl = cifs_ioctl,
#endif /* CONFIG_CIFS_POSIX */
.setlease = cifs_setlease,
+ .fallocate = cifs_fallocate,
};
const struct file_operations cifs_file_strict_nobrl_ops = {
.unlocked_ioctl = cifs_ioctl,
#endif /* CONFIG_CIFS_POSIX */
.setlease = cifs_setlease,
+ .fallocate = cifs_fallocate,
};
const struct file_operations cifs_file_direct_nobrl_ops = {
#endif /* CONFIG_CIFS_POSIX */
.llseek = cifs_llseek,
.setlease = cifs_setlease,
+ .fallocate = cifs_fallocate,
};
const struct file_operations cifs_dir_ops = {
/* get mtu credits */
int (*wait_mtu_credits)(struct TCP_Server_Info *, unsigned int,
unsigned int *, unsigned int *);
+ /* check if we need to issue closedir */
+ bool (*dir_needs_close)(struct cifsFileInfo *);
+ long (*fallocate)(struct file *, struct cifs_tcon *, int, loff_t,
+ loff_t);
};
struct smb_version_values {
for this mount even if server would support */
bool local_lease:1; /* check leases (only) on local system not remote */
bool broken_posix_open; /* e.g. Samba server versions < 3.3.2, 3.2.9 */
+ bool broken_sparse_sup; /* if server or share does not support sparse */
bool need_reconnect:1; /* connection reset, tid now invalid */
#ifdef CONFIG_CIFS_SMB2
bool print:1; /* set if connection to printer share */
/* minimum includes first three fields, and empty FS Name */
#define MIN_FS_ATTR_INFO_SIZE 12
+
+/* List of FileSystemAttributes - see 2.5.1 of MS-FSCC */
+#define FILE_SUPPORT_INTEGRITY_STREAMS 0x04000000
+#define FILE_SUPPORTS_USN_JOURNAL 0x02000000
+#define FILE_SUPPORTS_OPEN_BY_FILE_ID 0x01000000
+#define FILE_SUPPORTS_EXTENDED_ATTRIBUTES 0x00800000
+#define FILE_SUPPORTS_HARD_LINKS 0x00400000
+#define FILE_SUPPORTS_TRANSACTIONS 0x00200000
+#define FILE_SEQUENTIAL_WRITE_ONCE 0x00100000
+#define FILE_READ_ONLY_VOLUME 0x00080000
+#define FILE_NAMED_STREAMS 0x00040000
+#define FILE_SUPPORTS_ENCRYPTION 0x00020000
+#define FILE_SUPPORTS_OBJECT_IDS 0x00010000
+#define FILE_VOLUME_IS_COMPRESSED 0x00008000
+#define FILE_SUPPORTS_REMOTE_STORAGE 0x00000100
+#define FILE_SUPPORTS_REPARSE_POINTS 0x00000080
+#define FILE_SUPPORTS_SPARSE_FILES 0x00000040
+#define FILE_VOLUME_QUOTAS 0x00000020
+#define FILE_FILE_COMPRESSION 0x00000010
+#define FILE_PERSISTENT_ACLS 0x00000008
+#define FILE_UNICODE_ON_DISK 0x00000004
+#define FILE_CASE_PRESERVED_NAMES 0x00000002
+#define FILE_CASE_SENSITIVE_SEARCH 0x00000001
typedef struct {
__le32 Attributes;
__le32 MaxPathNameComponentLength;
cifs_dbg(FYI, "Freeing private data in close dir\n");
spin_lock(&cifs_file_list_lock);
- if (!cfile->srch_inf.endOfSearch && !cfile->invalidHandle) {
+ if (server->ops->dir_needs_close(cfile)) {
cfile->invalidHandle = true;
spin_unlock(&cifs_file_list_lock);
if (server->ops->close_dir)
target_dentry, to_name);
}
+ /* force revalidate to go get info when needed */
+ CIFS_I(source_dir)->time = CIFS_I(target_dir)->time = 0;
+
+ source_dir->i_ctime = source_dir->i_mtime = target_dir->i_ctime =
+ target_dir->i_mtime = current_fs_time(source_dir->i_sb);
+
cifs_rename_exit:
kfree(info_buf_source);
kfree(from_name);
cinode->oplock = 0;
}
-static int
-cifs_oplock_break_wait(void *unused)
-{
- schedule();
- return signal_pending(current) ? -ERESTARTSYS : 0;
-}
-
/*
* We wait for oplock breaks to be processed before we attempt to perform
* writes.
/* close and restart search */
cifs_dbg(FYI, "search backing up - close and restart search\n");
spin_lock(&cifs_file_list_lock);
- if (!cfile->srch_inf.endOfSearch && !cfile->invalidHandle) {
+ if (server->ops->dir_needs_close(cfile)) {
cfile->invalidHandle = true;
spin_unlock(&cifs_file_list_lock);
if (server->ops->close)
return CIFS_SB(inode->i_sb)->wsize;
}
+static bool
+cifs_dir_needs_close(struct cifsFileInfo *cfile)
+{
+ return !cfile->srch_inf.endOfSearch && !cfile->invalidHandle;
+}
+
struct smb_version_operations smb1_operations = {
.send_cancel = send_nt_cancel,
.compare_fids = cifs_compare_fids,
.create_mf_symlink = cifs_create_mf_symlink,
.is_read_op = cifs_is_read_op,
.wp_retry_size = cifs_wp_retry_size,
+ .dir_needs_close = cifs_dir_needs_close,
#ifdef CONFIG_CIFS_XATTR
.query_all_EAs = CIFSSMBQAllEAs,
.set_EA = CIFSSMBSetEA,
{STATUS_BREAKPOINT, -EIO, "STATUS_BREAKPOINT"},
{STATUS_SINGLE_STEP, -EIO, "STATUS_SINGLE_STEP"},
{STATUS_BUFFER_OVERFLOW, -EIO, "STATUS_BUFFER_OVERFLOW"},
- {STATUS_NO_MORE_FILES, -EIO, "STATUS_NO_MORE_FILES"},
+ {STATUS_NO_MORE_FILES, -ENODATA, "STATUS_NO_MORE_FILES"},
{STATUS_WAKE_SYSTEM_DEBUGGER, -EIO, "STATUS_WAKE_SYSTEM_DEBUGGER"},
{STATUS_HANDLES_CLOSED, -EIO, "STATUS_HANDLES_CLOSED"},
{STATUS_NO_INHERITANCE, -EIO, "STATUS_NO_INHERITANCE"},
{STATUS_INVALID_PARAMETER, -EINVAL, "STATUS_INVALID_PARAMETER"},
{STATUS_NO_SUCH_DEVICE, -ENODEV, "STATUS_NO_SUCH_DEVICE"},
{STATUS_NO_SUCH_FILE, -ENOENT, "STATUS_NO_SUCH_FILE"},
- {STATUS_INVALID_DEVICE_REQUEST, -EIO, "STATUS_INVALID_DEVICE_REQUEST"},
+ {STATUS_INVALID_DEVICE_REQUEST, -EOPNOTSUPP, "STATUS_INVALID_DEVICE_REQUEST"},
{STATUS_END_OF_FILE, -ENODATA, "STATUS_END_OF_FILE"},
{STATUS_WRONG_VOLUME, -EIO, "STATUS_WRONG_VOLUME"},
{STATUS_NO_MEDIA_IN_DEVICE, -EIO, "STATUS_NO_MEDIA_IN_DEVICE"},
/* Windows 7 server returns 24 bytes more */
if (clc_len + 20 == len && command == SMB2_OPLOCK_BREAK_HE)
return 0;
- /* server can return one byte more */
+ /* server can return one byte more due to implied bcc[0] */
if (clc_len == 4 + len + 1)
return 0;
+
+ /*
+ * MacOS server pads after SMB2.1 write response with 3 bytes
+ * of junk. Other servers match RFC1001 len to actual
+ * SMB2/SMB3 frame length (header + smb2 response specific data)
+ * Log the server error (once), but allow it and continue
+ * since the frame is parseable.
+ */
+ if (clc_len < 4 /* RFC1001 header size */ + len) {
+ printk_once(KERN_WARNING
+ "SMB2 server sent bad RFC1001 len %d not %d\n",
+ len, clc_len - 4);
+ return 0;
+ }
+
return 1;
}
return 0;
return SMB2_write(xid, parms, written, iov, nr_segs);
}
+/* Set or clear the SPARSE_FILE attribute based on value passed in setsparse */
+static bool smb2_set_sparse(const unsigned int xid, struct cifs_tcon *tcon,
+ struct cifsFileInfo *cfile, struct inode *inode, __u8 setsparse)
+{
+ struct cifsInodeInfo *cifsi;
+ int rc;
+
+ cifsi = CIFS_I(inode);
+
+ /* if file already sparse don't bother setting sparse again */
+ if ((cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) && setsparse)
+ return true; /* already sparse */
+
+ if (!(cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) && !setsparse)
+ return true; /* already not sparse */
+
+ /*
+ * Can't check for sparse support on share the usual way via the
+ * FS attribute info (FILE_SUPPORTS_SPARSE_FILES) on the share
+ * since Samba server doesn't set the flag on the share, yet
+ * supports the set sparse FSCTL and returns sparse correctly
+ * in the file attributes. If we fail setting sparse though we
+ * mark that server does not support sparse files for this share
+ * to avoid repeatedly sending the unsupported fsctl to server
+ * if the file is repeatedly extended.
+ */
+ if (tcon->broken_sparse_sup)
+ return false;
+
+ rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
+ cfile->fid.volatile_fid, FSCTL_SET_SPARSE,
+ true /* is_fctl */, &setsparse, 1, NULL, NULL);
+ if (rc) {
+ tcon->broken_sparse_sup = true;
+ cifs_dbg(FYI, "set sparse rc = %d\n", rc);
+ return false;
+ }
+
+ if (setsparse)
+ cifsi->cifsAttrs |= FILE_ATTRIBUTE_SPARSE_FILE;
+ else
+ cifsi->cifsAttrs &= (~FILE_ATTRIBUTE_SPARSE_FILE);
+
+ return true;
+}
+
static int
smb2_set_file_size(const unsigned int xid, struct cifs_tcon *tcon,
struct cifsFileInfo *cfile, __u64 size, bool set_alloc)
{
__le64 eof = cpu_to_le64(size);
+ struct inode *inode;
+
+ /*
+ * If extending file more than one page make sparse. Many Linux fs
+ * make files sparse by default when extending via ftruncate
+ */
+ inode = cfile->dentry->d_inode;
+
+ if (!set_alloc && (size > inode->i_size + 8192)) {
+ __u8 set_sparse = 1;
+
+ /* whether set sparse succeeds or not, extend the file */
+ smb2_set_sparse(xid, tcon, cfile, inode, set_sparse);
+ }
+
return SMB2_set_eof(xid, tcon, cfile->fid.persistent_fid,
cfile->fid.volatile_fid, cfile->pid, &eof, false);
}
return rc;
}
+static long smb3_zero_range(struct file *file, struct cifs_tcon *tcon,
+ loff_t offset, loff_t len, bool keep_size)
+{
+ struct inode *inode;
+ struct cifsInodeInfo *cifsi;
+ struct cifsFileInfo *cfile = file->private_data;
+ struct file_zero_data_information fsctl_buf;
+ long rc;
+ unsigned int xid;
+
+ xid = get_xid();
+
+ inode = cfile->dentry->d_inode;
+ cifsi = CIFS_I(inode);
+
+ /* if file not oplocked can't be sure whether asking to extend size */
+ if (!CIFS_CACHE_READ(cifsi))
+ if (keep_size == false)
+ return -EOPNOTSUPP;
+
+ /*
+ * Must check if file sparse since fallocate -z (zero range) assumes
+ * non-sparse allocation
+ */
+ if (!(cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE))
+ return -EOPNOTSUPP;
+
+ /*
+ * need to make sure we are not asked to extend the file since the SMB3
+ * fsctl does not change the file size. In the future we could change
+ * this to zero the first part of the range then set the file size
+ * which for a non sparse file would zero the newly extended range
+ */
+ if (keep_size == false)
+ if (i_size_read(inode) < offset + len)
+ return -EOPNOTSUPP;
+
+ cifs_dbg(FYI, "offset %lld len %lld", offset, len);
+
+ fsctl_buf.FileOffset = cpu_to_le64(offset);
+ fsctl_buf.BeyondFinalZero = cpu_to_le64(offset + len);
+
+ rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
+ cfile->fid.volatile_fid, FSCTL_SET_ZERO_DATA,
+ true /* is_fctl */, (char *)&fsctl_buf,
+ sizeof(struct file_zero_data_information), NULL, NULL);
+ free_xid(xid);
+ return rc;
+}
+
+static long smb3_punch_hole(struct file *file, struct cifs_tcon *tcon,
+ loff_t offset, loff_t len)
+{
+ struct inode *inode;
+ struct cifsInodeInfo *cifsi;
+ struct cifsFileInfo *cfile = file->private_data;
+ struct file_zero_data_information fsctl_buf;
+ long rc;
+ unsigned int xid;
+ __u8 set_sparse = 1;
+
+ xid = get_xid();
+
+ inode = cfile->dentry->d_inode;
+ cifsi = CIFS_I(inode);
+
+ /* Need to make file sparse, if not already, before freeing range. */
+ /* Consider adding equivalent for compressed since it could also work */
+ if (!smb2_set_sparse(xid, tcon, cfile, inode, set_sparse))
+ return -EOPNOTSUPP;
+
+ cifs_dbg(FYI, "offset %lld len %lld", offset, len);
+
+ fsctl_buf.FileOffset = cpu_to_le64(offset);
+ fsctl_buf.BeyondFinalZero = cpu_to_le64(offset + len);
+
+ rc = SMB2_ioctl(xid, tcon, cfile->fid.persistent_fid,
+ cfile->fid.volatile_fid, FSCTL_SET_ZERO_DATA,
+ true /* is_fctl */, (char *)&fsctl_buf,
+ sizeof(struct file_zero_data_information), NULL, NULL);
+ free_xid(xid);
+ return rc;
+}
+
+static long smb3_fallocate(struct file *file, struct cifs_tcon *tcon, int mode,
+ loff_t off, loff_t len)
+{
+ /* KEEP_SIZE already checked for by do_fallocate */
+ if (mode & FALLOC_FL_PUNCH_HOLE)
+ return smb3_punch_hole(file, tcon, off, len);
+ else if (mode & FALLOC_FL_ZERO_RANGE) {
+ if (mode & FALLOC_FL_KEEP_SIZE)
+ return smb3_zero_range(file, tcon, off, len, true);
+ return smb3_zero_range(file, tcon, off, len, false);
+ }
+
+ return -EOPNOTSUPP;
+}
+
static void
smb2_downgrade_oplock(struct TCP_Server_Info *server,
struct cifsInodeInfo *cinode, bool set_level2)
SMB2_MAX_BUFFER_SIZE);
}
+static bool
+smb2_dir_needs_close(struct cifsFileInfo *cfile)
+{
+ return !cfile->invalidHandle;
+}
+
struct smb_version_operations smb20_operations = {
.compare_fids = smb2_compare_fids,
.setup_request = smb2_setup_request,
.parse_lease_buf = smb2_parse_lease_buf,
.clone_range = smb2_clone_range,
.wp_retry_size = smb2_wp_retry_size,
+ .dir_needs_close = smb2_dir_needs_close,
};
struct smb_version_operations smb21_operations = {
.parse_lease_buf = smb2_parse_lease_buf,
.clone_range = smb2_clone_range,
.wp_retry_size = smb2_wp_retry_size,
+ .dir_needs_close = smb2_dir_needs_close,
};
struct smb_version_operations smb30_operations = {
.clone_range = smb2_clone_range,
.validate_negotiate = smb3_validate_negotiate,
.wp_retry_size = smb2_wp_retry_size,
+ .dir_needs_close = smb2_dir_needs_close,
+ .fallocate = smb3_fallocate,
};
struct smb_version_values smb20_values = {
tcon_error_exit:
if (rsp->hdr.Status == STATUS_BAD_NETWORK_NAME) {
cifs_dbg(VFS, "BAD_NETWORK_NAME: %s\n", tree);
- tcon->bad_network_name = true;
+ if (tcon)
+ tcon->bad_network_name = true;
}
goto tcon_exit;
}
cifs_dbg(FYI, "SMB2 IOCTL\n");
- *out_data = NULL;
+ if (out_data != NULL)
+ *out_data = NULL;
+
/* zero out returned data len, in case of error */
if (plen)
*plen = 0;
rsp = (struct smb2_query_directory_rsp *)iov[0].iov_base;
if (rc) {
+ if (rc == -ENODATA && rsp->hdr.Status == STATUS_NO_MORE_FILES) {
+ srch_inf->endOfSearch = true;
+ rc = 0;
+ }
cifs_stats_fail_inc(tcon, SMB2_QUERY_DIRECTORY_HE);
goto qdir_exit;
}
else
cifs_dbg(VFS, "illegal search buffer type\n");
- if (rsp->hdr.Status == STATUS_NO_MORE_FILES)
- srch_inf->endOfSearch = 1;
- else
- srch_inf->endOfSearch = 0;
-
return rc;
qdir_exit:
__u32 Reserved2;
} __packed;
+/* this goes in the ioctl buffer when doing FSCTL_SET_ZERO_DATA */
+struct file_zero_data_information {
+ __le64 FileOffset;
+ __le64 BeyondFinalZero;
+} __packed;
+
struct copychunk_ioctl_rsp {
__le32 ChunksWritten;
__le32 ChunkBytesWritten;
#define FSCTL_SET_OBJECT_ID_EXTENDED 0x000900BC /* BB add struct */
#define FSCTL_CREATE_OR_GET_OBJECT_ID 0x000900C0 /* BB add struct */
#define FSCTL_SET_SPARSE 0x000900C4 /* BB add struct */
-#define FSCTL_SET_ZERO_DATA 0x000900C8 /* BB add struct */
+#define FSCTL_SET_ZERO_DATA 0x000980C8
#define FSCTL_SET_ENCRYPTION 0x000900D7 /* BB add struct */
#define FSCTL_ENCRYPTION_FSCTL_IO 0x000900DB /* BB add struct */
#define FSCTL_WRITE_RAW_ENCRYPTED 0x000900DF /* BB add struct */
*/
overhead += ngroups * (2 + sbi->s_itb_per_group);
- /* Add the journal blocks as well */
- overhead += sbi->s_journal->j_maxlen;
+ /* Add the internal journal blocks as well */
+ if (sbi->s_journal && !sbi->journal_bdev)
+ overhead += sbi->s_journal->j_maxlen;
sbi->s_overhead_last = overhead;
smp_wmb();
return;
}
-static int isofs_read_inode(struct inode *);
+static int isofs_read_inode(struct inode *, int relocated);
static int isofs_statfs (struct dentry *, struct kstatfs *);
static struct kmem_cache *isofs_inode_cachep;
goto out;
}
-static int isofs_read_inode(struct inode *inode)
+static int isofs_read_inode(struct inode *inode, int relocated)
{
struct super_block *sb = inode->i_sb;
struct isofs_sb_info *sbi = ISOFS_SB(sb);
*/
if (!high_sierra) {
- parse_rock_ridge_inode(de, inode);
+ parse_rock_ridge_inode(de, inode, relocated);
/* if we want uid/gid set, override the rock ridge setting */
if (sbi->s_uid_set)
inode->i_uid = sbi->s_uid;
* offset that point to the underlying meta-data for the inode. The
* code below is otherwise similar to the iget() code in
* include/linux/fs.h */
-struct inode *isofs_iget(struct super_block *sb,
- unsigned long block,
- unsigned long offset)
+struct inode *__isofs_iget(struct super_block *sb,
+ unsigned long block,
+ unsigned long offset,
+ int relocated)
{
unsigned long hashval;
struct inode *inode;
return ERR_PTR(-ENOMEM);
if (inode->i_state & I_NEW) {
- ret = isofs_read_inode(inode);
+ ret = isofs_read_inode(inode, relocated);
if (ret < 0) {
iget_failed(inode);
inode = ERR_PTR(ret);
struct inode; /* To make gcc happy */
-extern int parse_rock_ridge_inode(struct iso_directory_record *, struct inode *);
+extern int parse_rock_ridge_inode(struct iso_directory_record *, struct inode *, int relocated);
extern int get_rock_ridge_filename(struct iso_directory_record *, char *, struct inode *);
extern int isofs_name_translate(struct iso_directory_record *, char *, struct inode *);
extern struct buffer_head *isofs_bread(struct inode *, sector_t);
extern int isofs_get_blocks(struct inode *, sector_t, struct buffer_head **, unsigned long);
-extern struct inode *isofs_iget(struct super_block *sb,
- unsigned long block,
- unsigned long offset);
+struct inode *__isofs_iget(struct super_block *sb,
+ unsigned long block,
+ unsigned long offset,
+ int relocated);
+
+static inline struct inode *isofs_iget(struct super_block *sb,
+ unsigned long block,
+ unsigned long offset)
+{
+ return __isofs_iget(sb, block, offset, 0);
+}
+
+static inline struct inode *isofs_iget_reloc(struct super_block *sb,
+ unsigned long block,
+ unsigned long offset)
+{
+ return __isofs_iget(sb, block, offset, 1);
+}
/* Because the inode number is no longer relevant to finding the
* underlying meta-data for an inode, we are free to choose a more
goto out;
}
+#define RR_REGARD_XA 1
+#define RR_RELOC_DE 2
+
static int
parse_rock_ridge_inode_internal(struct iso_directory_record *de,
- struct inode *inode, int regard_xa)
+ struct inode *inode, int flags)
{
int symlink_len = 0;
int cnt, sig;
+ unsigned int reloc_block;
struct inode *reloc;
struct rock_ridge *rr;
int rootflag;
init_rock_state(&rs, inode);
setup_rock_ridge(de, inode, &rs);
- if (regard_xa) {
+ if (flags & RR_REGARD_XA) {
rs.chr += 14;
rs.len -= 14;
if (rs.len < 0)
"relocated directory\n");
goto out;
case SIG('C', 'L'):
- ISOFS_I(inode)->i_first_extent =
- isonum_733(rr->u.CL.location);
- reloc =
- isofs_iget(inode->i_sb,
- ISOFS_I(inode)->i_first_extent,
- 0);
+ if (flags & RR_RELOC_DE) {
+ printk(KERN_ERR
+ "ISOFS: Recursive directory relocation "
+ "is not supported\n");
+ goto eio;
+ }
+ reloc_block = isonum_733(rr->u.CL.location);
+ if (reloc_block == ISOFS_I(inode)->i_iget5_block &&
+ ISOFS_I(inode)->i_iget5_offset == 0) {
+ printk(KERN_ERR
+ "ISOFS: Directory relocation points to "
+ "itself\n");
+ goto eio;
+ }
+ ISOFS_I(inode)->i_first_extent = reloc_block;
+ reloc = isofs_iget_reloc(inode->i_sb, reloc_block, 0);
if (IS_ERR(reloc)) {
ret = PTR_ERR(reloc);
goto out;
return rpnt;
}
-int parse_rock_ridge_inode(struct iso_directory_record *de, struct inode *inode)
+int parse_rock_ridge_inode(struct iso_directory_record *de, struct inode *inode,
+ int relocated)
{
- int result = parse_rock_ridge_inode_internal(de, inode, 0);
+ int flags = relocated ? RR_RELOC_DE : 0;
+ int result = parse_rock_ridge_inode_internal(de, inode, flags);
/*
* if rockridge flag was reset and we didn't look for attributes
*/
if ((ISOFS_SB(inode->i_sb)->s_rock_offset == -1)
&& (ISOFS_SB(inode->i_sb)->s_rock == 2)) {
- result = parse_rock_ridge_inode_internal(de, inode, 14);
+ result = parse_rock_ridge_inode_internal(de, inode,
+ flags | RR_REGARD_XA);
}
return result;
}
}
EXPORT_SYMBOL(locks_free_lock);
+static void
+locks_dispose_list(struct list_head *dispose)
+{
+ struct file_lock *fl;
+
+ while (!list_empty(dispose)) {
+ fl = list_first_entry(dispose, struct file_lock, fl_block);
+ list_del_init(&fl->fl_block);
+ locks_free_lock(fl);
+ }
+}
+
void locks_init_lock(struct file_lock *fl)
{
memset(fl, 0, sizeof(struct file_lock));
void locks_copy_lock(struct file_lock *new, struct file_lock *fl)
{
- locks_release_private(new);
+ /* "new" must be a freshly-initialized lock */
+ WARN_ON_ONCE(new->fl_ops);
__locks_copy_lock(new, fl);
new->fl_file = fl->fl_file;
*
* Must be called with i_lock held!
*/
-static void locks_delete_lock(struct file_lock **thisfl_p)
+static void locks_delete_lock(struct file_lock **thisfl_p,
+ struct list_head *dispose)
{
struct file_lock *fl = *thisfl_p;
locks_unlink_lock(thisfl_p);
- locks_free_lock(fl);
+ if (dispose)
+ list_add(&fl->fl_block, dispose);
+ else
+ locks_free_lock(fl);
}
/* Determine if lock sys_fl blocks lock caller_fl. Common functionality
struct inode * inode = file_inode(filp);
int error = 0;
int found = 0;
+ LIST_HEAD(dispose);
if (!(request->fl_flags & FL_ACCESS) && (request->fl_type != F_UNLCK)) {
new_fl = locks_alloc_lock();
if (request->fl_type == fl->fl_type)
goto out;
found = 1;
- locks_delete_lock(before);
+ locks_delete_lock(before, &dispose);
break;
}
spin_unlock(&inode->i_lock);
if (new_fl)
locks_free_lock(new_fl);
+ locks_dispose_list(&dispose);
return error;
}
struct file_lock **before;
int error;
bool added = false;
+ LIST_HEAD(dispose);
/*
* We may need two file_lock structures for this operation,
else
request->fl_end = fl->fl_end;
if (added) {
- locks_delete_lock(before);
+ locks_delete_lock(before, &dispose);
continue;
}
request = fl;
* one (This may happen several times).
*/
if (added) {
- locks_delete_lock(before);
+ locks_delete_lock(before, &dispose);
continue;
}
- /* Replace the old lock with the new one.
- * Wake up anybody waiting for the old one,
- * as the change in lock type might satisfy
- * their needs.
+ /*
+ * Replace the old lock with new_fl, and
+ * remove the old one. It's safe to do the
+ * insert here since we know that we won't be
+ * using new_fl later, and that the lock is
+ * just replacing an existing lock.
*/
- locks_wake_up_blocks(fl);
- fl->fl_start = request->fl_start;
- fl->fl_end = request->fl_end;
- fl->fl_type = request->fl_type;
- locks_release_private(fl);
- locks_copy_private(fl, request);
- request = fl;
+ error = -ENOLCK;
+ if (!new_fl)
+ goto out;
+ locks_copy_lock(new_fl, request);
+ request = new_fl;
+ new_fl = NULL;
+ locks_delete_lock(before, &dispose);
+ locks_insert_lock(before, request);
added = true;
}
}
locks_free_lock(new_fl);
if (new_fl2)
locks_free_lock(new_fl2);
+ locks_dispose_list(&dispose);
return error;
}
printk(KERN_ERR "locks_delete_lock: fasync == %p\n", fl->fl_fasync);
fl->fl_fasync = NULL;
}
- locks_delete_lock(before);
+ locks_delete_lock(before, NULL);
}
return 0;
}
ret = fl;
spin_lock(&inode->i_lock);
error = __vfs_setlease(filp, arg, &ret);
- if (error) {
- spin_unlock(&inode->i_lock);
- locks_free_lock(fl);
- goto out_free_fasync;
- }
- if (ret != fl)
- locks_free_lock(fl);
+ if (error)
+ goto out_unlock;
+ if (ret == fl)
+ fl = NULL;
/*
* fasync_insert_entry() returns the old entry if any.
new = NULL;
error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0);
+out_unlock:
spin_unlock(&inode->i_lock);
-
-out_free_fasync:
+ if (fl)
+ locks_free_lock(fl);
if (new)
fasync_free(new);
return error;
struct inode * inode = file_inode(filp);
struct file_lock *fl;
struct file_lock **before;
+ LIST_HEAD(dispose);
if (!inode->i_flock)
return;
fl->fl_type, fl->fl_flags,
fl->fl_start, fl->fl_end);
- locks_delete_lock(before);
+ locks_delete_lock(before, &dispose);
continue;
}
before = &fl->fl_next;
}
spin_unlock(&inode->i_lock);
+ locks_dispose_list(&dispose);
}
/**
seq_puts(f, "FLOCK ADVISORY ");
}
} else if (IS_LEASE(fl)) {
- seq_puts(f, "LEASE ");
+ if (fl->fl_flags & FL_DELEG)
+ seq_puts(f, "DELEG ");
+ else
+ seq_puts(f, "LEASE ");
+
if (lease_breaking(fl))
seq_puts(f, "BREAKING ");
else if (fl->fl_file)
fi = udf_add_entry(dir, dentry, &fibh, &cfi, &err);
if (!fi)
- goto out_no_entry;
+ goto out_fail;
cfi.icb.extLength = cpu_to_le32(sb->s_blocksize);
cfi.icb.extLocation = cpu_to_lelb(iinfo->i_location);
if (UDF_SB(inode->i_sb)->s_lvid_bh) {
out_no_entry:
up_write(&iinfo->i_data_sem);
+out_fail:
inode_dec_link_count(inode);
iput(inode);
goto out;
unsigned short bip_slab; /* slab the bip came from */
unsigned short bip_vcnt; /* # of integrity bio_vecs */
+ unsigned short bip_max_vcnt; /* integrity bio_vec slots */
unsigned bip_owns_buf:1; /* should free bip_buf */
struct work_struct bip_work; /* I/O completion */
#include <linux/bsg.h>
#include <linux/smp.h>
#include <linux/rcupdate.h>
+#include <linux/percpu-refcount.h>
#include <asm/scatterlist.h>
struct mutex sysfs_lock;
int bypass_depth;
+ int mq_freeze_depth;
#if defined(CONFIG_BLK_DEV_BSG)
bsg_job_fn *bsg_job_fn;
#endif
struct rcu_head rcu_head;
wait_queue_head_t mq_freeze_wq;
- struct percpu_counter mq_usage_counter;
+ struct percpu_ref mq_usage_counter;
struct list_head all_q_node;
struct blk_mq_tag_set *tag_set;
#define PHY_ID_BCM7366 0x600d8490
#define PHY_ID_BCM7439 0x600d8480
#define PHY_ID_BCM7445 0x600d8510
-#define PHY_ID_BCM7XXX_28 0x600d8400
#define PHY_BCM_OUI_MASK 0xfffffc00
#define PHY_BCM_OUI_1 0x00206000
#endif
extern const char *drbd_buildtag(void);
-#define REL_VERSION "8.4.3"
+#define REL_VERSION "8.4.5"
#define API_VERSION 1
#define PRO_VERSION_MIN 86
#define PRO_VERSION_MAX 101
D_DISKLESS,
D_ATTACHING, /* In the process of reading the meta-data */
D_FAILED, /* Becomes D_DISKLESS as soon as we told it the peer */
- /* when >= D_FAILED it is legal to access mdev->bc */
+ /* when >= D_FAILED it is legal to access mdev->ldev */
D_NEGOTIATING, /* Late attaching state, we need to talk to the peer */
D_INCONSISTENT,
D_OUTDATED,
__flg_field(28, DRBD_GENLA_F_MANDATORY | DRBD_F_INVARIANT, tentative)
__flg_field_def(29, DRBD_GENLA_F_MANDATORY, use_rle, DRBD_USE_RLE_DEF)
/* 9: __u32_field_def(30, DRBD_GENLA_F_MANDATORY, fencing_policy, DRBD_FENCING_DEF) */
+ /* 9: __str_field_def(31, DRBD_GENLA_F_MANDATORY, name, SHARED_SECRET_MAX) */
+ /* 9: __u32_field(32, DRBD_F_REQUIRED | DRBD_F_INVARIANT, peer_node_id) */
+ __flg_field_def(33, 0 /* OPTIONAL */, csums_after_crash_only, DRBD_CSUMS_AFTER_CRASH_ONLY_DEF)
+ __u32_field_def(34, 0 /* OPTIONAL */, sock_check_timeo, DRBD_SOCKET_CHECK_TIMEO_DEF)
)
GENL_struct(DRBD_NLA_SET_ROLE_PARMS, 6, set_role_parms,
#define DRBD_ALLOW_TWO_PRIMARIES_DEF 0
#define DRBD_ALWAYS_ASBP_DEF 0
#define DRBD_USE_RLE_DEF 1
+#define DRBD_CSUMS_AFTER_CRASH_ONLY_DEF 0
#define DRBD_AL_STRIPES_MIN 1
#define DRBD_AL_STRIPES_MAX 1024
#define DRBD_AL_STRIPE_SIZE_MAX 16777216
#define DRBD_AL_STRIPE_SIZE_DEF 32
#define DRBD_AL_STRIPE_SIZE_SCALE 'k' /* kilobytes */
+
+#define DRBD_SOCKET_CHECK_TIMEO_MIN 0
+#define DRBD_SOCKET_CHECK_TIMEO_MAX DRBD_PING_TIMEO_MAX
+#define DRBD_SOCKET_CHECK_TIMEO_DEF 0
+#define DRBD_SOCKET_CHECK_TIMEO_SCALE '1'
#endif
* @MEM_DDR3: DDR3 RAM
* @MEM_RDDR3: Registered DDR3 RAM
* This is a variant of the DDR3 memories.
+ * @MEM_DDR4: DDR4 RAM
+ * @MEM_RDDR4: Registered DDR4 RAM
+ * This is a variant of the DDR4 memories.
*/
enum mem_type {
MEM_EMPTY = 0,
MEM_XDR,
MEM_DDR3,
MEM_RDDR3,
+ MEM_DDR4,
+ MEM_RDDR4,
};
#define MEM_FLAG_EMPTY BIT(MEM_EMPTY)
}
#endif /* CONFIG_OF */
-#ifdef CONFIG_I2C_ACPI
-int acpi_i2c_install_space_handler(struct i2c_adapter *adapter);
-void acpi_i2c_remove_space_handler(struct i2c_adapter *adapter);
+#ifdef CONFIG_ACPI
void acpi_i2c_register_devices(struct i2c_adapter *adap);
#else
static inline void acpi_i2c_register_devices(struct i2c_adapter *adap) { }
+#endif /* CONFIG_ACPI */
+
+#ifdef CONFIG_ACPI_I2C_OPREGION
+int acpi_i2c_install_space_handler(struct i2c_adapter *adapter);
+void acpi_i2c_remove_space_handler(struct i2c_adapter *adapter);
+#else
static inline void acpi_i2c_remove_space_handler(struct i2c_adapter *adapter)
{ }
static inline int acpi_i2c_install_space_handler(struct i2c_adapter *adapter)
{ return 0; }
-#endif
+#endif /* CONFIG_ACPI_I2C_OPREGION */
#endif /* _LINUX_I2C_H */
void input_mt_report_finger_count(struct input_dev *dev, int count);
void input_mt_report_pointer_emulation(struct input_dev *dev, bool use_count);
+void input_mt_drop_unused(struct input_dev *dev);
void input_mt_sync_frame(struct input_dev *dev);
/* special QP and management commands */
MLX4_CMD_CONF_SPECIAL_QP = 0x23,
MLX4_CMD_MAD_IFC = 0x24,
+ MLX4_CMD_MAD_DEMUX = 0x203,
/* multicast commands */
MLX4_CMD_READ_MCG = 0x25,
MLX4_SET_PORT_VXLAN = 0xB
};
+enum {
+ MLX4_CMD_MAD_DEMUX_CONFIG = 0,
+ MLX4_CMD_MAD_DEMUX_QUERY_STATE = 1,
+ MLX4_CMD_MAD_DEMUX_QUERY_RESTR = 2, /* Query mad demux restrictions */
+};
+
enum {
MLX4_CMD_WRAPPED,
MLX4_CMD_NATIVE
MLX4_DEV_CAP_FLAG2_UPDATE_QP = 1LL << 8,
MLX4_DEV_CAP_FLAG2_DMFS_IPOIB = 1LL << 9,
MLX4_DEV_CAP_FLAG2_VXLAN_OFFLOADS = 1LL << 10,
+ MLX4_DEV_CAP_FLAG2_MAD_DEMUX = 1LL << 11,
};
enum {
MLX4_PERM_REMOTE_WRITE = 1 << 13,
MLX4_PERM_ATOMIC = 1 << 14,
MLX4_PERM_BIND_MW = 1 << 15,
+ MLX4_PERM_MASK = 0xFC00
};
enum {
int mlx4_vf_get_enable_smi_admin(struct mlx4_dev *dev, int slave, int port);
int mlx4_vf_set_enable_smi_admin(struct mlx4_dev *dev, int slave, int port,
int enable);
+int mlx4_mr_hw_get_mpt(struct mlx4_dev *dev, struct mlx4_mr *mmr,
+ struct mlx4_mpt_entry ***mpt_entry);
+int mlx4_mr_hw_write_mpt(struct mlx4_dev *dev, struct mlx4_mr *mmr,
+ struct mlx4_mpt_entry **mpt_entry);
+int mlx4_mr_hw_change_pd(struct mlx4_dev *dev, struct mlx4_mpt_entry *mpt_entry,
+ u32 pdn);
+int mlx4_mr_hw_change_access(struct mlx4_dev *dev,
+ struct mlx4_mpt_entry *mpt_entry,
+ u32 access);
+void mlx4_mr_hw_put_mpt(struct mlx4_dev *dev,
+ struct mlx4_mpt_entry **mpt_entry);
+void mlx4_mr_rereg_mem_cleanup(struct mlx4_dev *dev, struct mlx4_mr *mr);
+int mlx4_mr_rereg_mem_write(struct mlx4_dev *dev, struct mlx4_mr *mr,
+ u64 iova, u64 size, int npages,
+ int page_shift, struct mlx4_mpt_entry *mpt_entry);
/* Returns true if running in low memory profile (kdump kernel) */
static inline bool mlx4_low_memory_profile(void)
#define DW_MCI_QUIRK_HIGHSPEED BIT(2)
/* Unreliable card detection */
#define DW_MCI_QUIRK_BROKEN_CARD_DETECTION BIT(3)
+/* No write protect */
+#define DW_MCI_QUIRK_NO_WRITE_PROTECT BIT(4)
/* Slot level quirks */
/* This slot has no write protect */
const struct sdhci_ops *ops; /* Low level hw interface */
- struct regulator *vmmc; /* Power regulator (vmmc) */
- struct regulator *vqmmc; /* Signaling regulator (vccq) */
-
/* Internal data */
struct mmc_host *mmc; /* MMC structure */
u64 dma_mask; /* custom DMA mask */
uint32_t args[MAX_PHANDLE_ARGS];
};
-extern int of_node_add(struct device_node *node);
-
/* initialize a node */
extern struct kobj_type of_node_ktype;
static inline void of_node_init(struct device_node *node)
extern struct device_node *of_allnodes;
extern struct device_node *of_chosen;
extern struct device_node *of_aliases;
+extern struct device_node *of_stdout;
extern raw_spinlock_t devtree_lock;
static inline bool of_have_populated_dt(void)
#define OF_DYNAMIC 1 /* node and properties were allocated via kmalloc */
#define OF_DETACHED 2 /* node has been detached from the device tree */
#define OF_POPULATED 3 /* device already created for the node */
+#define OF_POPULATED_BUS 4 /* of_platform_populate recursed to children of this node */
#define OF_IS_DYNAMIC(x) test_bit(OF_DYNAMIC, &x->_flags)
#define OF_MARK_DYNAMIC(x) set_bit(OF_DYNAMIC, &x->_flags)
struct of_prop_reconfig {
struct device_node *dn;
struct property *prop;
+ struct property *old_prop;
};
extern int of_reconfig_notifier_register(struct notifier_block *);
*/
const char *of_prop_next_string(struct property *prop, const char *cur);
-int of_device_is_stdout_path(struct device_node *dn);
+bool of_console_check(struct device_node *dn, char *name, int index);
#else /* CONFIG_OF */
return 0;
}
-static inline int of_device_is_stdout_path(struct device_node *dn)
+static inline bool of_console_check(const struct device_node *dn, const char *name, int index)
{
- return 0;
+ return false;
}
static inline const __be32 *of_prop_next_u32(struct property *prop,
#define OF_DECLARE_2(table, name, compat, fn) \
_OF_DECLARE(table, name, compat, fn, of_init_fn_2)
+/**
+ * struct of_changeset_entry - Holds a changeset entry
+ *
+ * @node: list_head for the log list
+ * @action: notifier action
+ * @np: pointer to the device node affected
+ * @prop: pointer to the property affected
+ * @old_prop: hold a pointer to the original property
+ *
+ * Every modification of the device tree during a changeset
+ * is held in a list of of_changeset_entry structures.
+ * That way we can recover from a partial application, or we can
+ * revert the changeset
+ */
+struct of_changeset_entry {
+ struct list_head node;
+ unsigned long action;
+ struct device_node *np;
+ struct property *prop;
+ struct property *old_prop;
+};
+
+/**
+ * struct of_changeset - changeset tracker structure
+ *
+ * @entries: list_head for the changeset entries
+ *
+ * changesets are a convenient way to apply bulk changes to the
+ * live tree. In case of an error, changes are rolled-back.
+ * changesets live on after initial application, and if not
+ * destroyed after use, they can be reverted in one single call.
+ */
+struct of_changeset {
+ struct list_head entries;
+};
+
+#ifdef CONFIG_OF_DYNAMIC
+extern void of_changeset_init(struct of_changeset *ocs);
+extern void of_changeset_destroy(struct of_changeset *ocs);
+extern int of_changeset_apply(struct of_changeset *ocs);
+extern int of_changeset_revert(struct of_changeset *ocs);
+extern int of_changeset_action(struct of_changeset *ocs,
+ unsigned long action, struct device_node *np,
+ struct property *prop);
+
+static inline int of_changeset_attach_node(struct of_changeset *ocs,
+ struct device_node *np)
+{
+ return of_changeset_action(ocs, OF_RECONFIG_ATTACH_NODE, np, NULL);
+}
+
+static inline int of_changeset_detach_node(struct of_changeset *ocs,
+ struct device_node *np)
+{
+ return of_changeset_action(ocs, OF_RECONFIG_DETACH_NODE, np, NULL);
+}
+
+static inline int of_changeset_add_property(struct of_changeset *ocs,
+ struct device_node *np, struct property *prop)
+{
+ return of_changeset_action(ocs, OF_RECONFIG_ADD_PROPERTY, np, prop);
+}
+
+static inline int of_changeset_remove_property(struct of_changeset *ocs,
+ struct device_node *np, struct property *prop)
+{
+ return of_changeset_action(ocs, OF_RECONFIG_REMOVE_PROPERTY, np, prop);
+}
+
+static inline int of_changeset_update_property(struct of_changeset *ocs,
+ struct device_node *np, struct property *prop)
+{
+ return of_changeset_action(ocs, OF_RECONFIG_UPDATE_PROPERTY, np, prop);
+}
+#endif
+
#endif /* _LINUX_OF_H */
const struct of_device_id *matches,
const struct of_dev_auxdata *lookup,
struct device *parent);
-extern int of_platform_depopulate(struct device *parent);
+extern void of_platform_depopulate(struct device *parent);
#else
static inline int of_platform_populate(struct device_node *root,
const struct of_device_id *matches,
{
return -ENODEV;
}
-static inline int of_platform_depopulate(struct device *parent)
-{
- return -ENODEV;
-}
+static inline void of_platform_depopulate(struct device *parent) { }
#endif
#endif /* _LINUX_OF_PLATFORM_H */
struct reserved_mem {
const char *name;
unsigned long fdt_node;
+ unsigned long phandle;
const struct reserved_mem_ops *ops;
phys_addr_t base;
phys_addr_t size;
_OF_DECLARE(reservedmem, name, compat, init, reservedmem_of_init_fn)
#ifdef CONFIG_OF_RESERVED_MEM
+void of_reserved_mem_device_init(struct device *dev);
+void of_reserved_mem_device_release(struct device *dev);
+
void fdt_init_reserved_mem(void);
void fdt_reserved_mem_save_node(unsigned long node, const char *uname,
phys_addr_t base, phys_addr_t size);
#else
+static inline void of_reserved_mem_device_init(struct device *dev) { }
+static inline void of_reserved_mem_device_release(struct device *pdev) { }
+
static inline void fdt_init_reserved_mem(void) { }
static inline void fdt_reserved_mem_save_node(unsigned long node,
const char *uname, phys_addr_t base, phys_addr_t size) { }
*/
#define OMAP_HSMMC_SUPPORTS_DUAL_VOLT BIT(0)
#define OMAP_HSMMC_BROKEN_MULTIBLOCK_READ BIT(1)
+#define OMAP_HSMMC_SWAKEUP_MISSING BIT(2)
struct mmc_card;
extern const char linux_banner[];
extern const char linux_proc_banner[];
+extern char *log_buf_addr_get(void);
+extern u32 log_buf_len_get(void);
+
static inline int printk_get_level(const char *buffer)
{
if (buffer[0] == KERN_SOH_ASCII && buffer[1]) {
#include <linux/rculist.h>
struct rhash_head {
- struct rhash_head *next;
+ struct rhash_head __rcu *next;
};
#define INIT_HASH_HEAD(ptr) ((ptr)->next = NULL)
void rhashtable_insert(struct rhashtable *ht, struct rhash_head *node, gfp_t);
bool rhashtable_remove(struct rhashtable *ht, struct rhash_head *node, gfp_t);
void rhashtable_remove_pprev(struct rhashtable *ht, struct rhash_head *obj,
- struct rhash_head **pprev, gfp_t flags);
+ struct rhash_head __rcu **pprev, gfp_t flags);
bool rht_grow_above_75(const struct rhashtable *ht, size_t new_size);
bool rht_shrink_below_30(const struct rhashtable *ht, size_t new_size);
#define rht_dereference_rcu(p, ht) \
rcu_dereference_check(p, lockdep_rht_mutex_is_held(ht))
-/* Internal, use rht_obj() instead */
#define rht_entry(ptr, type, member) container_of(ptr, type, member)
#define rht_entry_safe(ptr, type, member) \
({ \
typeof(ptr) __ptr = (ptr); \
__ptr ? rht_entry(__ptr, type, member) : NULL; \
})
-#define rht_entry_safe_rcu(ptr, type, member) \
-({ \
- typeof(*ptr) __rcu *__ptr = (typeof(*ptr) __rcu __force *)ptr; \
- __ptr ? container_of((typeof(ptr))rcu_dereference_raw(__ptr), type, member) : NULL; \
-})
#define rht_next_entry_safe(pos, ht, member) \
({ \
* traversal is guarded by rcu_read_lock().
*/
#define rht_for_each_entry_rcu(pos, head, member) \
- for (pos = rht_entry_safe_rcu(head, typeof(*(pos)), member); \
+ for (pos = rht_entry_safe(rcu_dereference_raw(head), \
+ typeof(*(pos)), member); \
pos; \
- pos = rht_entry_safe_rcu((pos)->member.next, \
- typeof(*(pos)), member))
+ pos = rht_entry_safe(rcu_dereference_raw((pos)->member.next), \
+ typeof(*(pos)), member))
#endif /* _LINUX_RHASHTABLE_H */
extern unsigned long nr_running(void);
extern unsigned long nr_iowait(void);
extern unsigned long nr_iowait_cpu(int cpu);
-extern unsigned long this_cpu_load(void);
-
+extern void get_iowait_load(unsigned long *nr_waiters, unsigned long *load);
extern void calc_global_load(unsigned long ticks);
extern void update_cpu_load_nohz(void);
extern long vfio_external_check_extension(struct vfio_group *group,
unsigned long arg);
+struct pci_dev;
#ifdef CONFIG_EEH
-extern int vfio_spapr_pci_eeh_open(struct pci_dev *pdev);
+extern void vfio_spapr_pci_eeh_open(struct pci_dev *pdev);
extern void vfio_spapr_pci_eeh_release(struct pci_dev *pdev);
extern long vfio_spapr_iommu_eeh_ioctl(struct iommu_group *group,
unsigned int cmd,
unsigned long arg);
#else
-static inline int vfio_spapr_pci_eeh_open(struct pci_dev *pdev)
+static inline void vfio_spapr_pci_eeh_open(struct pci_dev *pdev)
{
- return 0;
}
static inline void vfio_spapr_pci_eeh_release(struct pci_dev *pdev)
void (*addr2sockaddr)(struct sock *sk, struct sockaddr *);
int (*bind_conflict)(const struct sock *sk,
const struct inet_bind_bucket *tb, bool relax);
+ void (*mtu_reduced)(struct sock *sk);
};
/** inet_connection_sock - INET connection oriented sock
struct sk_buff *skb);
void (*release_cb)(struct sock *sk);
- void (*mtu_reduced)(struct sock *sk);
/* Keeping track of sk's, looking them up, and port selection methods. */
void (*hash)(struct sock *sk);
void tcp_init_metrics(struct sock *sk);
void tcp_metrics_init(void);
bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst,
- bool paws_check);
+ bool paws_check, bool timestamps);
bool tcp_remember_stamp(struct sock *sk);
bool tcp_tw_remember_stamp(struct inet_timewait_sock *tw);
void tcp_fetch_timewait_stamp(struct sock *sk, struct dst_entry *dst);
*/
void tcp_v4_send_check(struct sock *sk, struct sk_buff *skb);
+void tcp_v4_mtu_reduced(struct sock *sk);
int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb);
struct sock *tcp_create_openreq_child(struct sock *sk,
struct request_sock *req,
#define TCPCB_SACKED_RETRANS 0x02 /* SKB retransmitted */
#define TCPCB_LOST 0x04 /* SKB is lost */
#define TCPCB_TAGBITS 0x07 /* All tag bits */
+#define TCPCB_REPAIRED 0x10 /* SKB repaired (no skb_mstamp) */
#define TCPCB_EVER_RETRANS 0x80 /* Ever retransmitted frame */
-#define TCPCB_RETRANS (TCPCB_SACKED_RETRANS|TCPCB_EVER_RETRANS)
+#define TCPCB_RETRANS (TCPCB_SACKED_RETRANS|TCPCB_EVER_RETRANS| \
+ TCPCB_REPAIRED)
__u8 ip_dsfield; /* IPv4 tos or IPv6 dsfield */
/* 1 byte hole */
#include <linux/list.h>
#include <rdma/ib_verbs.h>
+#include <uapi/rdma/ib_user_mad.h>
/* Management base version */
#define IB_MGMT_BASE_VERSION 1
* @hi_tid: Access layer assigned transaction ID for this client.
* Unsolicited MADs sent by this client will have the upper 32-bits
* of their TID set to this value.
+ * @flags: registration flags
* @port_num: Port number on which QP is registered
* @rmpp_version: If set, indicates the RMPP version used by this agent.
*/
+enum {
+ IB_MAD_USER_RMPP = IB_USER_MAD_USER_RMPP,
+};
struct ib_mad_agent {
struct ib_device *device;
struct ib_qp *qp;
ib_mad_snoop_handler snoop_handler;
void *context;
u32 hi_tid;
+ u32 flags;
u8 port_num;
u8 rmpp_version;
};
* in the range from 0x30 to 0x4f. Otherwise not used.
* @method_mask: The caller will receive unsolicited MADs for any method
* where @method_mask = 1.
+ *
*/
struct ib_mad_reg_req {
u8 mgmt_class;
* @recv_handler: The completion callback routine invoked for a received
* MAD.
* @context: User specified context associated with the registration.
+ * @registration_flags: Registration flags to set for this agent
*/
struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device,
u8 port_num,
u8 rmpp_version,
ib_mad_send_handler send_handler,
ib_mad_recv_handler recv_handler,
- void *context);
+ void *context,
+ u32 registration_flags);
enum ib_mad_snoop_flags {
/*IB_MAD_SNOOP_POSTED_SENDS = 1,*/
*/
void ib_free_send_mad(struct ib_mad_send_buf *send_buf);
+/**
+ * ib_mad_kernel_rmpp_agent - Returns if the agent is performing RMPP.
+ * @agent: the agent in question
+ * @return: true if agent is performing rmpp, false otherwise.
+ */
+int ib_mad_kernel_rmpp_agent(struct ib_mad_agent *agent);
+
#endif /* IB_MAD_H */
enum ib_mr_rereg_flags {
IB_MR_REREG_TRANS = 1,
IB_MR_REREG_PD = (1<<1),
- IB_MR_REREG_ACCESS = (1<<2)
+ IB_MR_REREG_ACCESS = (1<<2),
+ IB_MR_REREG_SUPPORTED = ((IB_MR_REREG_ACCESS << 1) - 1)
};
/**
u64 virt_addr,
int mr_access_flags,
struct ib_udata *udata);
+ int (*rereg_user_mr)(struct ib_mr *mr,
+ int flags,
+ u64 start, u64 length,
+ u64 virt_addr,
+ int mr_access_flags,
+ struct ib_pd *pd,
+ struct ib_udata *udata);
int (*query_mr)(struct ib_mr *mr,
struct ib_mr_attr *mr_attr);
int (*dereg_mr)(struct ib_mr *mr);
ISCSI_ERR_XMIT_FAILED = ISCSI_ERR_BASE + 19,
ISCSI_ERR_TCP_CONN_CLOSE = ISCSI_ERR_BASE + 20,
ISCSI_ERR_SCSI_EH_SESSION_RST = ISCSI_ERR_BASE + 21,
+ ISCSI_ERR_NOP_TIMEDOUT = ISCSI_ERR_BASE + 22,
};
/*
#define SG_FLAG_MMAP_IO 4 /* request memory mapped IO */
#define SG_FLAG_NO_DXFER 0x10000 /* no transfer of kernel buffers to/from */
/* user space (debug indirect IO) */
-#define SG_FLAG_Q_AT_TAIL 0x10 /* default is Q_AT_HEAD */
+/* defaults:: for sg driver: Q_AT_HEAD; for block layer: Q_AT_TAIL */
+#define SG_FLAG_Q_AT_TAIL 0x10
+#define SG_FLAG_Q_AT_HEAD 0x20
/* following 'info' values are "or"-ed together */
#define SG_INFO_OK_MASK 0x1
);
TRACE_EVENT(bcache_write,
- TP_PROTO(struct bio *bio, bool writeback, bool bypass),
- TP_ARGS(bio, writeback, bypass),
+ TP_PROTO(struct cache_set *c, u64 inode, struct bio *bio,
+ bool writeback, bool bypass),
+ TP_ARGS(c, inode, bio, writeback, bypass),
TP_STRUCT__entry(
- __field(dev_t, dev )
+ __array(char, uuid, 16 )
+ __field(u64, inode )
__field(sector_t, sector )
__field(unsigned int, nr_sector )
__array(char, rwbs, 6 )
),
TP_fast_assign(
- __entry->dev = bio->bi_bdev->bd_dev;
+ memcpy(__entry->uuid, c->sb.set_uuid, 16);
+ __entry->inode = inode;
__entry->sector = bio->bi_iter.bi_sector;
__entry->nr_sector = bio->bi_iter.bi_size >> 9;
blk_fill_rwbs(__entry->rwbs, bio->bi_rw, bio->bi_iter.bi_size);
__entry->bypass = bypass;
),
- TP_printk("%d,%d %s %llu + %u hit %u bypass %u",
- MAJOR(__entry->dev), MINOR(__entry->dev),
+ TP_printk("%pU inode %llu %s %llu + %u hit %u bypass %u",
+ __entry->uuid, __entry->inode,
__entry->rwbs, (unsigned long long)__entry->sector,
__entry->nr_sector, __entry->writeback, __entry->bypass)
);
TP_ARGS(b)
);
-DEFINE_EVENT(btree_node, bcache_btree_node_alloc_fail,
- TP_PROTO(struct btree *b),
- TP_ARGS(b)
+DEFINE_EVENT(cache_set, bcache_btree_node_alloc_fail,
+ TP_PROTO(struct cache_set *c),
+ TP_ARGS(c)
);
DEFINE_EVENT(btree_node, bcache_btree_node_free,
--- /dev/null
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM thp
+
+#if !defined(_TRACE_THP_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_THP_H
+
+#include <linux/types.h>
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(hugepage_invalidate,
+
+ TP_PROTO(unsigned long addr, unsigned long pte),
+ TP_ARGS(addr, pte),
+ TP_STRUCT__entry(
+ __field(unsigned long, addr)
+ __field(unsigned long, pte)
+ ),
+
+ TP_fast_assign(
+ __entry->addr = addr;
+ __entry->pte = pte;
+ ),
+
+ TP_printk("hugepage invalidate at addr 0x%lx and pte = 0x%lx",
+ __entry->addr, __entry->pte)
+);
+
+TRACE_EVENT(hugepage_set_pmd,
+
+ TP_PROTO(unsigned long addr, unsigned long pmd),
+ TP_ARGS(addr, pmd),
+ TP_STRUCT__entry(
+ __field(unsigned long, addr)
+ __field(unsigned long, pmd)
+ ),
+
+ TP_fast_assign(
+ __entry->addr = addr;
+ __entry->pmd = pmd;
+ ),
+
+ TP_printk("Set pmd with 0x%lx with 0x%lx", __entry->addr, __entry->pmd)
+);
+
+
+TRACE_EVENT(hugepage_update,
+
+ TP_PROTO(unsigned long addr, unsigned long pte, unsigned long clr, unsigned long set),
+ TP_ARGS(addr, pte, clr, set),
+ TP_STRUCT__entry(
+ __field(unsigned long, addr)
+ __field(unsigned long, pte)
+ __field(unsigned long, clr)
+ __field(unsigned long, set)
+ ),
+
+ TP_fast_assign(
+ __entry->addr = addr;
+ __entry->pte = pte;
+ __entry->clr = clr;
+ __entry->set = set;
+
+ ),
+
+ TP_printk("hugepage update at addr 0x%lx and pte = 0x%lx clr = 0x%lx, set = 0x%lx", __entry->addr, __entry->pte, __entry->clr, __entry->set)
+);
+TRACE_EVENT(hugepage_splitting,
+
+ TP_PROTO(unsigned long addr, unsigned long pte),
+ TP_ARGS(addr, pte),
+ TP_STRUCT__entry(
+ __field(unsigned long, addr)
+ __field(unsigned long, pte)
+ ),
+
+ TP_fast_assign(
+ __entry->addr = addr;
+ __entry->pte = pte;
+ ),
+
+ TP_printk("hugepage splitting at addr 0x%lx and pte = 0x%lx",
+ __entry->addr, __entry->pte)
+);
+
+#endif /* _TRACE_THP_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
__SYSCALL(__NR_seccomp, sys_seccomp)
#define __NR_getrandom 278
__SYSCALL(__NR_getrandom, sys_getrandom)
+#define __NR_memfd_create 279
+__SYSCALL(__NR_memfd_create, sys_memfd_create)
#undef __NR_syscalls
-#define __NR_syscalls 279
+#define __NR_syscalls 280
/*
* All syscalls below here should go away really,
#define BSG_SUB_PROTOCOL_SCSI_TRANSPORT 2
/*
- * For flags member below
- * sg.h sg_io_hdr also has bits defined for it's flags member. However
- * none of these bits are implemented/used by bsg. The bits below are
- * allocated to not conflict with sg.h ones anyway.
+ * For flag constants below:
+ * sg.h sg_io_hdr also has bits defined for it's flags member. These
+ * two flag values (0x10 and 0x20) have the same meaning in sg.h . For
+ * bsg the BSG_FLAG_Q_AT_HEAD flag is ignored since it is the deafult.
*/
-#define BSG_FLAG_Q_AT_TAIL 0x10 /* default, == 0 at this bit, is Q_AT_HEAD */
+#define BSG_FLAG_Q_AT_TAIL 0x10 /* default is Q_AT_HEAD */
+#define BSG_FLAG_Q_AT_HEAD 0x20
struct sg_io_v4 {
__s32 guard; /* [i] 'Q' to differentiate from v3 */
#define VIRTIO_BLK_F_WCE 9 /* Writeback mode enabled after reset */
#define VIRTIO_BLK_F_TOPOLOGY 10 /* Topology information is available */
#define VIRTIO_BLK_F_CONFIG_WCE 11 /* Writeback mode available in config */
+#define VIRTIO_BLK_F_MQ 12 /* support more than one vq */
#ifndef __KERNEL__
/* Old (deprecated) name for VIRTIO_BLK_F_WCE. */
/* writeback mode (if VIRTIO_BLK_F_CONFIG_WCE) */
__u8 wce;
+ __u8 unused;
+
+ /* number of vqs, only available when VIRTIO_BLK_F_MQ is set */
+ __u16 num_queues;
} __attribute__((packed));
/*
__u8 rmpp_version;
};
+/**
+ * ib_user_mad_reg_req2 - MAD registration request
+ *
+ * @id - Set by the _kernel_; used by userspace to identify the
+ * registered agent in future requests.
+ * @qpn - Queue pair number; must be 0 or 1.
+ * @mgmt_class - Indicates which management class of MADs should be
+ * receive by the caller. This field is only required if
+ * the user wishes to receive unsolicited MADs, otherwise
+ * it should be 0.
+ * @mgmt_class_version - Indicates which version of MADs for the given
+ * management class to receive.
+ * @res - Ignored.
+ * @flags - additional registration flags; Must be in the set of
+ * flags defined in IB_USER_MAD_REG_FLAGS_CAP
+ * @method_mask - The caller wishes to receive unsolicited MADs for the
+ * methods whose bit(s) is(are) set.
+ * @oui - Indicates IEEE OUI to use when mgmt_class is a vendor
+ * class in the range from 0x30 to 0x4f. Otherwise not
+ * used.
+ * @rmpp_version - If set, indicates the RMPP version to use.
+ */
+enum {
+ IB_USER_MAD_USER_RMPP = (1 << 0),
+};
+#define IB_USER_MAD_REG_FLAGS_CAP (IB_USER_MAD_USER_RMPP)
+struct ib_user_mad_reg_req2 {
+ __u32 id;
+ __u32 qpn;
+ __u8 mgmt_class;
+ __u8 mgmt_class_version;
+ __u16 res;
+ __u32 flags;
+ __u64 method_mask[2];
+ __u32 oui;
+ __u8 rmpp_version;
+ __u8 reserved[3];
+};
+
#define IB_IOCTL_MAGIC 0x1b
#define IB_USER_MAD_REGISTER_AGENT _IOWR(IB_IOCTL_MAGIC, 1, \
#define IB_USER_MAD_ENABLE_PKEY _IO(IB_IOCTL_MAGIC, 3)
+#define IB_USER_MAD_REGISTER_AGENT2 _IOWR(IB_IOCTL_MAGIC, 4, \
+ struct ib_user_mad_reg_req2)
+
#endif /* IB_USER_MAD_H */
__u32 rkey;
};
+struct ib_uverbs_rereg_mr {
+ __u64 response;
+ __u32 mr_handle;
+ __u32 flags;
+ __u64 start;
+ __u64 length;
+ __u64 hca_va;
+ __u32 pd_handle;
+ __u32 access_flags;
+};
+
+struct ib_uverbs_rereg_mr_resp {
+ __u32 lkey;
+ __u32 rkey;
+};
+
struct ib_uverbs_dereg_mr {
__u32 mr_handle;
};
#define RDMA_USER_CM_H
#include <linux/types.h>
+#include <linux/socket.h>
#include <linux/in6.h>
#include <rdma/ib_user_verbs.h>
#include <rdma/ib_user_sa.h>
help
Enabling this option adds the system calls process_vm_readv and
process_vm_writev which allow a process with the correct privileges
- to directly read from or write to to another process's address space.
+ to directly read from or write to another process' address space.
See the man page for more details.
config FHANDLE
* needed because this new task is not yet running and cannot
* be racing exec.
*/
- BUG_ON(!spin_is_locked(¤t->sighand->siglock));
+ assert_spin_locked(¤t->sighand->siglock);
/* Ref-count the new filter user, and assign it. */
get_seccomp_filter(current);
mutex_lock(&module_mutex);
module_bug_cleanup(mod);
mutex_unlock(&module_mutex);
+
+ /* we can't deallocate the module until we clear memory protection */
+ unset_module_init_ro_nx(mod);
+ unset_module_core_ro_nx(mod);
+
ddebug_cleanup:
dynamic_debug_remove(info->debug);
synchronize_sched();
}
}
+static bool is_nosave_page(unsigned long pfn)
+{
+ struct nosave_region *region;
+
+ list_for_each_entry(region, &nosave_regions, list) {
+ if (pfn >= region->start_pfn && pfn < region->end_pfn) {
+ pr_err("PM: %#010llx in e820 nosave region: "
+ "[mem %#010llx-%#010llx]\n",
+ (unsigned long long) pfn << PAGE_SHIFT,
+ (unsigned long long) region->start_pfn << PAGE_SHIFT,
+ ((unsigned long long) region->end_pfn << PAGE_SHIFT)
+ - 1);
+ return true;
+ }
+ }
+
+ return false;
+}
+
/**
* create_basic_memory_bitmaps - create bitmaps needed for marking page
* frames that should not be saved and free page frames. The pointers
do {
pfn = memory_bm_next_pfn(bm);
if (likely(pfn != BM_END_OF_MAP)) {
- if (likely(pfn_valid(pfn)))
+ if (likely(pfn_valid(pfn)) && !is_nosave_page(pfn))
swsusp_set_page_free(pfn_to_page(pfn));
else
return -EFAULT;
static char *log_buf = __log_buf;
static u32 log_buf_len = __LOG_BUF_LEN;
+/* Return log buffer address */
+char *log_buf_addr_get(void)
+{
+ return log_buf;
+}
+
+/* Return log buffer size */
+u32 log_buf_len_get(void)
+{
+ return log_buf_len;
+}
+
/* human readable text of the record */
static char *log_text(const struct printk_log *msg)
{
return atomic_read(&this->nr_iowait);
}
+void get_iowait_load(unsigned long *nr_waiters, unsigned long *load)
+{
+ struct rq *this = this_rq();
+ *nr_waiters = atomic_read(&this->nr_iowait);
+ *load = this->cpu_load[0];
+}
+
#ifdef CONFIG_SMP
/*
#include "sched.h"
-unsigned long this_cpu_load(void)
-{
- struct rq *this = this_rq();
- return this->cpu_load[0];
-}
-
-
/*
* Global load-average calculations
*
static inline bool seccomp_may_assign_mode(unsigned long seccomp_mode)
{
- BUG_ON(!spin_is_locked(¤t->sighand->siglock));
+ assert_spin_locked(¤t->sighand->siglock);
if (current->seccomp.mode && current->seccomp.mode != seccomp_mode)
return false;
static inline void seccomp_assign_mode(struct task_struct *task,
unsigned long seccomp_mode)
{
- BUG_ON(!spin_is_locked(&task->sighand->siglock));
+ assert_spin_locked(&task->sighand->siglock);
task->seccomp.mode = seccomp_mode;
/*
struct task_struct *thread, *caller;
BUG_ON(!mutex_is_locked(¤t->signal->cred_guard_mutex));
- BUG_ON(!spin_is_locked(¤t->sighand->siglock));
+ assert_spin_locked(¤t->sighand->siglock);
/* Validate all threads being eligible for synchronization. */
caller = current;
struct task_struct *thread, *caller;
BUG_ON(!mutex_is_locked(¤t->signal->cred_guard_mutex));
- BUG_ON(!spin_is_locked(¤t->sighand->siglock));
+ assert_spin_locked(¤t->sighand->siglock);
/* Synchronize all threads. */
caller = current;
unsigned long total_insns;
struct seccomp_filter *walker;
- BUG_ON(!spin_is_locked(¤t->sighand->siglock));
+ assert_spin_locked(¤t->sighand->siglock);
/* Validate resulting filter length. */
total_insns = filter->prog->len;
static inline void update_vsyscall(struct timekeeper *tk)
{
- struct timespec xt;
+ struct timespec xt, wm;
xt = timespec64_to_timespec(tk_xtime(tk));
- update_vsyscall_old(&xt, &tk->wall_to_monotonic, tk->tkr.clock, tk->tkr.mult,
+ wm = timespec64_to_timespec(tk->wall_to_monotonic);
+ update_vsyscall_old(&xt, &wm, tk->tkr.clock, tk->tkr.mult,
tk->tkr.cycle_last);
}
DEBUG_INFO build and compile times are reduced too.
Only works with newer gcc versions.
+config DEBUG_INFO_SPLIT
+ bool "Produce split debuginfo in .dwo files"
+ depends on DEBUG_INFO
+ help
+ Generate debug info into separate .dwo files. This significantly
+ reduces the build directory size for builds with DEBUG_INFO,
+ because it stores the information only once on disk in .dwo
+ files instead of multiple times in object files and executables.
+ In addition the debug information is also compressed.
+
+ Requires recent gcc (4.7+) and recent gdb/binutils.
+ Any tool that packages or reads debug information would need
+ to know about the .dwo files and include them.
+ Incompatible with older versions of ccache.
+
+config DEBUG_INFO_DWARF4
+ bool "Generate dwarf4 debuginfo"
+ depends on DEBUG_INFO
+ help
+ Generate dwarf4 debug info. This requires recent versions
+ of gcc and gdb. It makes the debug information larger.
+ But it significantly improves the success of resolving
+ variables in gdb on optimized code.
+
config ENABLE_WARN_DEPRECATED
bool "Enable __deprecated logic"
default y
return NULL;
}
-void lc_free_by_index(struct lru_cache *lc, unsigned i)
+static void lc_free_by_index(struct lru_cache *lc, unsigned i)
{
void *p = lc->lc_element[i];
WARN_ON(!p);
* lc_dump - Dump a complete LRU cache to seq in textual form.
* @lc: the lru cache to operate on
* @seq: the &struct seq_file pointer to seq_printf into
- * @utext: user supplied "heading" or other info
+ * @utext: user supplied additional "heading" or other info
* @detail: function pointer the user may provide to dump further details
- * of the object the lc_element is embedded in.
+ * of the object the lc_element is embedded in. May be NULL.
+ * Note: a leading space ' ' and trailing newline '\n' is implied.
*/
void lc_seq_dump_details(struct seq_file *seq, struct lru_cache *lc, char *utext,
void (*detail) (struct seq_file *, struct lc_element *))
struct lc_element *e;
int i;
- seq_printf(seq, "\tnn: lc_number refcnt %s\n ", utext);
+ seq_printf(seq, "\tnn: lc_number (new nr) refcnt %s\n ", utext);
for (i = 0; i < nr_elements; i++) {
e = lc_element_by_index(lc, i);
- if (e->lc_number == LC_FREE) {
- seq_printf(seq, "\t%2d: FREE\n", i);
- } else {
- seq_printf(seq, "\t%2d: %4u %4u ", i,
- e->lc_number, e->refcnt);
+ if (e->lc_number != e->lc_new_number)
+ seq_printf(seq, "\t%5d: %6d %8d %6d ",
+ i, e->lc_number, e->lc_new_number, e->refcnt);
+ else
+ seq_printf(seq, "\t%5d: %6d %-8s %6d ",
+ i, e->lc_number, "-\"-", e->refcnt);
+ if (detail)
detail(seq, e);
- }
+ seq_putc(seq, '\n');
}
}
EXPORT_SYMBOL_GPL(lockdep_rht_mutex_is_held);
#endif
-/**
- * rht_obj - cast hash head to outer object
- * @ht: hash table
- * @he: hashed node
- */
-void *rht_obj(const struct rhashtable *ht, const struct rhash_head *he)
+static void *rht_obj(const struct rhashtable *ht, const struct rhash_head *he)
{
return (void *) he - ht->p.head_offset;
}
-EXPORT_SYMBOL_GPL(rht_obj);
static u32 __hashfn(const struct rhashtable *ht, const void *key,
u32 len, u32 hsize)
* deletion when combined with walking or lookup.
*/
void rhashtable_remove_pprev(struct rhashtable *ht, struct rhash_head *obj,
- struct rhash_head **pprev, gfp_t flags)
+ struct rhash_head __rcu **pprev, gfp_t flags)
{
struct bucket_table *tbl = rht_dereference(ht->tbl, ht);
ret = res_counter_memparse_write_strategy(buf, &val);
if (ret)
break;
+ val = ALIGN(val, 1ULL << huge_page_shift(&hstates[idx]));
ret = res_counter_set_limit(&h_cg->hugepage[idx], val);
break;
default:
priv->lane2_ops = NULL;
if (priv->lane_version > 1)
priv->lane2_ops = &lane2_ops;
+ rtnl_lock();
if (dev_set_mtu(dev, mesg->content.config.mtu))
pr_info("%s: change_mtu to %d failed\n",
dev->name, mesg->content.config.mtu);
+ rtnl_unlock();
priv->is_proxy = mesg->content.config.is_proxy;
break;
case l_flush_tran_id:
loff_t *l)
{
struct hlist_node *e = state->node;
- struct lec_arp_table *tmp;
if (!e)
e = tbl->first;
--*l;
}
- tmp = container_of(e, struct lec_arp_table, next);
-
- hlist_for_each_entry_from(tmp, next) {
+ for (; e; e = e->next) {
if (--*l < 0)
break;
}
pr_debug("%p\n", vcc);
if (test_bit(ATM_VF_REGIS, &vcc->flags)) {
- prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
sigd_enq(vcc, as_close, NULL, NULL, NULL);
- while (!test_bit(ATM_VF_RELEASED, &vcc->flags) && sigd) {
+ for (;;) {
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
+ if (test_bit(ATM_VF_RELEASED, &vcc->flags) || !sigd)
+ break;
schedule();
- prepare_to_wait(sk_sleep(sk), &wait,
- TASK_UNINTERRUPTIBLE);
}
finish_wait(sk_sleep(sk), &wait);
}
}
vcc->local = *addr;
set_bit(ATM_VF_WAITING, &vcc->flags);
- prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
sigd_enq(vcc, as_bind, NULL, NULL, &vcc->local);
- while (test_bit(ATM_VF_WAITING, &vcc->flags) && sigd) {
- schedule();
+ for (;;) {
prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
+ if (!test_bit(ATM_VF_WAITING, &vcc->flags) || !sigd)
+ break;
+ schedule();
}
finish_wait(sk_sleep(sk), &wait);
clear_bit(ATM_VF_REGIS, &vcc->flags); /* doesn't count */
}
vcc->remote = *addr;
set_bit(ATM_VF_WAITING, &vcc->flags);
- prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
sigd_enq(vcc, as_connect, NULL, NULL, &vcc->remote);
if (flags & O_NONBLOCK) {
- finish_wait(sk_sleep(sk), &wait);
sock->state = SS_CONNECTING;
error = -EINPROGRESS;
goto out;
}
error = 0;
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
while (test_bit(ATM_VF_WAITING, &vcc->flags) && sigd) {
schedule();
if (!signal_pending(current)) {
goto out;
}
set_bit(ATM_VF_WAITING, &vcc->flags);
- prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
sigd_enq(vcc, as_listen, NULL, NULL, &vcc->local);
- while (test_bit(ATM_VF_WAITING, &vcc->flags) && sigd) {
- schedule();
+ for (;;) {
prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
+ if (!test_bit(ATM_VF_WAITING, &vcc->flags) || !sigd)
+ break;
+ schedule();
}
finish_wait(sk_sleep(sk), &wait);
if (!sigd) {
}
/* wait should be short, so we ignore the non-blocking flag */
set_bit(ATM_VF_WAITING, &new_vcc->flags);
- prepare_to_wait(sk_sleep(sk_atm(new_vcc)), &wait,
- TASK_UNINTERRUPTIBLE);
sigd_enq(new_vcc, as_accept, old_vcc, NULL, NULL);
- while (test_bit(ATM_VF_WAITING, &new_vcc->flags) && sigd) {
+ for (;;) {
+ prepare_to_wait(sk_sleep(sk_atm(new_vcc)), &wait,
+ TASK_UNINTERRUPTIBLE);
+ if (!test_bit(ATM_VF_WAITING, &new_vcc->flags) || !sigd)
+ break;
release_sock(sk);
schedule();
lock_sock(sk);
- prepare_to_wait(sk_sleep(sk_atm(new_vcc)), &wait,
- TASK_UNINTERRUPTIBLE);
}
finish_wait(sk_sleep(sk_atm(new_vcc)), &wait);
if (!sigd) {
DEFINE_WAIT(wait);
set_bit(ATM_VF_WAITING, &vcc->flags);
- prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
sigd_enq2(vcc, as_modify, NULL, NULL, &vcc->local, qos, 0);
- while (test_bit(ATM_VF_WAITING, &vcc->flags) &&
- !test_bit(ATM_VF_RELEASED, &vcc->flags) && sigd) {
- schedule();
+ for (;;) {
prepare_to_wait(sk_sleep(sk), &wait, TASK_UNINTERRUPTIBLE);
+ if (!test_bit(ATM_VF_WAITING, &vcc->flags) ||
+ test_bit(ATM_VF_RELEASED, &vcc->flags) || !sigd) {
+ break;
+ }
+ schedule();
}
finish_wait(sk_sleep(sk), &wait);
if (!sigd)
lock_sock(sk);
set_bit(ATM_VF_WAITING, &vcc->flags);
- prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
sigd_enq(vcc, as_addparty, NULL, NULL,
(struct sockaddr_atmsvc *) sockaddr);
if (flags & O_NONBLOCK) {
- finish_wait(sk_sleep(sk), &wait);
error = -EINPROGRESS;
goto out;
}
pr_debug("added wait queue\n");
- while (test_bit(ATM_VF_WAITING, &vcc->flags) && sigd) {
- schedule();
+ for (;;) {
prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+ if (!test_bit(ATM_VF_WAITING, &vcc->flags) || !sigd)
+ break;
+ schedule();
}
finish_wait(sk_sleep(sk), &wait);
error = xchg(&sk->sk_err_soft, 0);
lock_sock(sk);
set_bit(ATM_VF_WAITING, &vcc->flags);
- prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
sigd_enq2(vcc, as_dropparty, NULL, NULL, NULL, NULL, ep_ref);
- while (test_bit(ATM_VF_WAITING, &vcc->flags) && sigd) {
- schedule();
+ for (;;) {
prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+ if (!test_bit(ATM_VF_WAITING, &vcc->flags) || !sigd)
+ break;
+ schedule();
}
finish_wait(sk_sleep(sk), &wait);
if (!sigd) {
/* Reached the end of the list, so insert after 'frag_entry_last'. */
if (likely(frag_entry_last)) {
- hlist_add_behind(&frag_entry_last->list, &frag_entry_new->list);
+ hlist_add_behind(&frag_entry_new->list, &frag_entry_last->list);
chain->size += skb->len - hdr_size;
chain->timestamp = jiffies;
ret = true;
if (!skb)
goto wait_for_memory;
- /*
- * All packets are restored as if they have
- * already been sent.
- */
- if (tp->repair)
- TCP_SKB_CB(skb)->when = tcp_time_stamp;
-
/*
* Check whether we can use HW checksum.
*/
skb_entail(sk, skb);
copy = size_goal;
max = size_goal;
+
+ /* All packets are restored as if they have
+ * already been sent. skb_mstamp isn't set to
+ * avoid wrong rtt estimation.
+ */
+ if (tp->repair)
+ TCP_SKB_CB(skb)->sacked |= TCPCB_REPAIRED;
}
/* Try to append data to the end of skb. */
*/
static void tcp_process_loss(struct sock *sk, int flag, bool is_dupack)
{
- struct inet_connection_sock *icsk = inet_csk(sk);
struct tcp_sock *tp = tcp_sk(sk);
bool recovered = !before(tp->snd_una, tp->high_seq);
if (recovered) {
/* F-RTO RFC5682 sec 3.1 step 2.a and 1st part of step 3.a */
- icsk->icsk_retransmits = 0;
tcp_try_undo_recovery(sk);
return;
}
- if (flag & FLAG_DATA_ACKED)
- icsk->icsk_retransmits = 0;
if (tcp_is_reno(tp)) {
/* A Reno DUPACK means new data in F-RTO step 2.b above are
* delivered. Lower inflight to clock out (re)tranmissions.
first_ackt.v64 = 0;
while ((skb = tcp_write_queue_head(sk)) && skb != tcp_send_head(sk)) {
+ struct skb_shared_info *shinfo = skb_shinfo(skb);
struct tcp_skb_cb *scb = TCP_SKB_CB(skb);
u8 sacked = scb->sacked;
u32 acked_pcount;
+ if (unlikely(shinfo->tx_flags & SKBTX_ACK_TSTAMP) &&
+ between(shinfo->tskey, prior_snd_una, tp->snd_una - 1))
+ __skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK);
+
/* Determine how many packets and what bytes were acked, tso and else */
if (after(scb->end_seq, tp->snd_una)) {
if (tcp_skb_pcount(skb) == 1 ||
tp->retrans_stamp = 0;
}
- if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_ACK_TSTAMP) &&
- between(skb_shinfo(skb)->tskey, prior_snd_una,
- tp->snd_una + 1))
- __skb_tstamp_tx(skb, NULL, sk, SCM_TSTAMP_ACK);
-
if (!fully_acked)
break;
icsk->icsk_pending == ICSK_TIME_LOSS_PROBE)
tcp_rearm_rto(sk);
- if (after(ack, prior_snd_una))
+ if (after(ack, prior_snd_una)) {
flag |= FLAG_SND_UNA_ADVANCED;
+ icsk->icsk_retransmits = 0;
+ }
prior_fackets = tp->fackets_out;
* timewait bucket, so that all the necessary checks
* are made in the function processing timewait state.
*/
- if (tmp_opt.saw_tstamp && tcp_death_row.sysctl_tw_recycle) {
+ if (tcp_death_row.sysctl_tw_recycle) {
bool strict;
dst = af_ops->route_req(sk, &fl, req, &strict);
+
if (dst && strict &&
- !tcp_peer_is_proven(req, dst, true)) {
+ !tcp_peer_is_proven(req, dst, true,
+ tmp_opt.saw_tstamp)) {
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_PAWSPASSIVEREJECTED);
goto drop_and_release;
}
else if (!sysctl_tcp_syncookies &&
(sysctl_max_syn_backlog - inet_csk_reqsk_queue_len(sk) <
(sysctl_max_syn_backlog >> 2)) &&
- !tcp_peer_is_proven(req, dst, false)) {
+ !tcp_peer_is_proven(req, dst, false,
+ tmp_opt.saw_tstamp)) {
/* Without syncookies last quarter of
* backlog is filled with destinations,
* proven to be alive.
* It can be called through tcp_release_cb() if socket was owned by user
* at the time tcp_v4_err() was called to handle ICMP message.
*/
-static void tcp_v4_mtu_reduced(struct sock *sk)
+void tcp_v4_mtu_reduced(struct sock *sk)
{
struct dst_entry *dst;
struct inet_sock *inet = inet_sk(sk);
tcp_simple_retransmit(sk);
} /* else let the usual retransmit timer handle it */
}
+EXPORT_SYMBOL(tcp_v4_mtu_reduced);
static void do_redirect(struct sk_buff *skb, struct sock *sk)
{
.compat_setsockopt = compat_ip_setsockopt,
.compat_getsockopt = compat_ip_getsockopt,
#endif
+ .mtu_reduced = tcp_v4_mtu_reduced,
};
EXPORT_SYMBOL(ipv4_specific);
.sendpage = tcp_sendpage,
.backlog_rcv = tcp_v4_do_rcv,
.release_cb = tcp_release_cb,
- .mtu_reduced = tcp_v4_mtu_reduced,
.hash = inet_hash,
.unhash = inet_unhash,
.get_port = inet_csk_get_port,
tp->snd_cwnd_stamp = tcp_time_stamp;
}
-bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst, bool paws_check)
+bool tcp_peer_is_proven(struct request_sock *req, struct dst_entry *dst,
+ bool paws_check, bool timestamps)
{
struct tcp_metrics_block *tm;
bool ret;
if (paws_check) {
if (tm &&
(u32)get_seconds() - tm->tcpm_ts_stamp < TCP_PAWS_MSL &&
- (s32)(tm->tcpm_ts - req->ts_recent) > TCP_PAWS_WINDOW)
+ ((s32)(tm->tcpm_ts - req->ts_recent) > TCP_PAWS_WINDOW ||
+ !timestamps))
ret = false;
else
ret = true;
__sock_put(sk);
}
if (flags & (1UL << TCP_MTU_REDUCED_DEFERRED)) {
- sk->sk_prot->mtu_reduced(sk);
+ inet_csk(sk)->icsk_af_ops->mtu_reduced(sk);
__sock_put(sk);
}
}
tcp_verify_left_out(tp);
}
+static void tcp_fragment_tstamp(struct sk_buff *skb, struct sk_buff *skb2)
+{
+ struct skb_shared_info *shinfo = skb_shinfo(skb);
+
+ if (unlikely(shinfo->tx_flags & SKBTX_ANY_TSTAMP) &&
+ !before(shinfo->tskey, TCP_SKB_CB(skb2)->seq)) {
+ struct skb_shared_info *shinfo2 = skb_shinfo(skb2);
+ u8 tsflags = shinfo->tx_flags & SKBTX_ANY_TSTAMP;
+
+ shinfo->tx_flags &= ~tsflags;
+ shinfo2->tx_flags |= tsflags;
+ swap(shinfo->tskey, shinfo2->tskey);
+ }
+}
+
/* Function to create two new TCP segments. Shrinks the given segment
* to the specified size and appends a new segment with the rest of the
* packet to the list. This won't be called frequently, I hope.
*/
TCP_SKB_CB(buff)->when = TCP_SKB_CB(skb)->when;
buff->tstamp = skb->tstamp;
+ tcp_fragment_tstamp(skb, buff);
old_factor = tcp_skb_pcount(skb);
buff->ip_summed = skb->ip_summed = CHECKSUM_PARTIAL;
skb_split(skb, buff, len);
+ tcp_fragment_tstamp(skb, buff);
/* Fix up tso_factor for both original and new SKB. */
tcp_set_skb_tso_segs(sk, skb, mss_now);
tso_segs = tcp_init_tso_segs(sk, skb, mss_now);
BUG_ON(!tso_segs);
- if (unlikely(tp->repair) && tp->repair_queue == TCP_SEND_QUEUE)
+ if (unlikely(tp->repair) && tp->repair_queue == TCP_SEND_QUEUE) {
+ /* "when" is used as a start point for the retransmit timer */
+ TCP_SKB_CB(skb)->when = tcp_time_stamp;
goto repair; /* Skip network transmission */
+ }
cwnd_quota = tcp_cwnd_test(tp, skb);
if (!cwnd_quota) {
if (dst->flags & DST_HOST) {
mp = dst_metrics_write_ptr(dst);
} else {
- mp = kzalloc(sizeof(u32) * RTAX_MAX, GFP_KERNEL);
+ mp = kzalloc(sizeof(u32) * RTAX_MAX, GFP_ATOMIC);
if (!mp)
return -ENOMEM;
dst_init_metrics(dst, mp, 0);
for_each_ip_tunnel_rcu(t, sitn->tunnels_r_l[h0 ^ h1]) {
if (local == t->parms.iph.saddr &&
remote == t->parms.iph.daddr &&
- (!dev || !t->parms.link || dev->iflink == t->parms.link) &&
+ (!dev || !t->parms.link || dev->ifindex == t->parms.link) &&
(t->dev->flags & IFF_UP))
return t;
}
for_each_ip_tunnel_rcu(t, sitn->tunnels_r[h0]) {
if (remote == t->parms.iph.daddr &&
- (!dev || !t->parms.link || dev->iflink == t->parms.link) &&
+ (!dev || !t->parms.link || dev->ifindex == t->parms.link) &&
(t->dev->flags & IFF_UP))
return t;
}
for_each_ip_tunnel_rcu(t, sitn->tunnels_l[h1]) {
if (local == t->parms.iph.saddr &&
- (!dev || !t->parms.link || dev->iflink == t->parms.link) &&
+ (!dev || !t->parms.link || dev->ifindex == t->parms.link) &&
(t->dev->flags & IFF_UP))
return t;
}
.compat_setsockopt = compat_ipv6_setsockopt,
.compat_getsockopt = compat_ipv6_getsockopt,
#endif
+ .mtu_reduced = tcp_v6_mtu_reduced,
};
#ifdef CONFIG_TCP_MD5SIG
.compat_setsockopt = compat_ipv6_setsockopt,
.compat_getsockopt = compat_ipv6_getsockopt,
#endif
+ .mtu_reduced = tcp_v4_mtu_reduced,
};
#ifdef CONFIG_TCP_MD5SIG
.sendpage = tcp_sendpage,
.backlog_rcv = tcp_v6_do_rcv,
.release_cb = tcp_release_cb,
- .mtu_reduced = tcp_v6_mtu_reduced,
.hash = tcp_v6_hash,
.unhash = inet_unhash,
.get_port = inet_csk_get_port,
frame = (struct rd_frame *)skb_put(tx_skb, 2);
frame->caddr = self->caddr;
- frame->caddr = RD_RSP | PF_BIT;
+ frame->control = RD_RSP | PF_BIT;
irlap_queue_xmit(self, tx_skb);
}
list_del(&sdata->reserved_chanctx_list);
list_move(&sdata->assigned_chanctx_list,
- &new_ctx->assigned_vifs);
+ &ctx->assigned_vifs);
sdata->reserved_chanctx = NULL;
ieee80211_vif_chanctx_reservation_complete(sdata);
}
static void *netlink_seq_start(struct seq_file *seq, loff_t *pos)
+ __acquires(RCU)
{
rcu_read_lock();
return *pos ? netlink_seq_socket_idx(seq, *pos - 1) : SEQ_START_TOKEN;
}
static void netlink_seq_stop(struct seq_file *seq, void *v)
+ __releases(RCU)
{
rcu_read_unlock();
}
static int make_writable(struct sk_buff *skb, int write_len)
{
+ if (!pskb_may_pull(skb, write_len))
+ return -ENOMEM;
+
if (!skb_cloned(skb) || skb_clone_writable(skb, write_len))
return 0;
vlan_set_encap_proto(skb, vhdr);
skb->mac_header += VLAN_HLEN;
+ if (skb_network_offset(skb) < ETH_HLEN)
+ skb_set_network_header(skb, ETH_HLEN);
skb_reset_mac_len(skb);
return 0;
vport->ops = ops;
INIT_HLIST_NODE(&vport->dp_hash_node);
- if (ovs_vport_set_upcall_portids(vport, parms->upcall_portids))
+ if (ovs_vport_set_upcall_portids(vport, parms->upcall_portids)) {
+ kfree(vport);
return ERR_PTR(-EINVAL);
+ }
vport->percpu_stats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
if (!vport->percpu_stats) {
p1->tov_in_jiffies = msecs_to_jiffies(p1->retire_blk_tov);
p1->blk_sizeof_priv = req_u->req3.tp_sizeof_priv;
+ p1->max_frame_len = p1->kblk_size - BLK_PLUS_PRIV(p1->blk_sizeof_priv);
prb_init_ft_ops(p1, req_u);
prb_setup_retire_blk_timer(po, tx_ring);
prb_open_block(p1, pbd);
if ((int)snaplen < 0)
snaplen = 0;
}
+ } else if (unlikely(macoff + snaplen >
+ GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len)) {
+ u32 nval;
+
+ nval = GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len - macoff;
+ pr_err_once("tpacket_rcv: packet too big, clamped from %u to %u. macoff=%u\n",
+ snaplen, nval, macoff);
+ snaplen = nval;
+ if (unlikely((int)snaplen < 0)) {
+ snaplen = 0;
+ macoff = GET_PBDQC_FROM_RB(&po->rx_ring)->max_frame_len;
+ }
}
spin_lock(&sk->sk_receive_queue.lock);
h.raw = packet_current_rx_frame(po, skb,
goto out;
if (unlikely(req->tp_block_size & (PAGE_SIZE - 1)))
goto out;
+ if (po->tp_version >= TPACKET_V3 &&
+ (int)(req->tp_block_size -
+ BLK_PLUS_PRIV(req_u->req3.tp_sizeof_priv)) <= 0)
+ goto out;
if (unlikely(req->tp_frame_size < po->tp_hdrlen +
po->tp_reserve))
goto out;
char *pkblk_start;
char *pkblk_end;
int kblk_size;
+ unsigned int max_frame_len;
unsigned int knum_blocks;
uint64_t knxt_seq_num;
char *prev;
struct cbq_class *tx_borrowed;
int tx_len;
psched_time_t now; /* Cached timestamp */
- psched_time_t now_rt; /* Cached real time */
unsigned int pmask;
struct hrtimer delay_timer;
int toplevel = q->toplevel;
if (toplevel > cl->level && !(qdisc_is_throttled(cl->q))) {
- psched_time_t now;
- psched_tdiff_t incr;
-
- now = psched_get_time();
- incr = now - q->now_rt;
- now = q->now + incr;
+ psched_time_t now = psched_get_time();
do {
if (cl->undertime < now) {
struct cbq_class *this = q->tx_class;
struct cbq_class *cl = this;
int len = q->tx_len;
+ psched_time_t now;
q->tx_class = NULL;
+ /* Time integrator. We calculate EOS time
+ * by adding expected packet transmission time.
+ */
+ now = q->now + L2T(&q->link, len);
for ( ; cl; cl = cl->share) {
long avgidle = cl->avgidle;
* idle = (now - last) - last_pktlen/rate
*/
- idle = q->now - cl->last;
+ idle = now - cl->last;
if ((unsigned long)idle > 128*1024*1024) {
avgidle = cl->maxidle;
} else {
idle -= L2T(&q->link, len);
idle += L2T(cl, len);
- cl->undertime = q->now + idle;
+ cl->undertime = now + idle;
} else {
/* Underlimit */
else
cl->avgidle = avgidle;
}
- cl->last = q->now;
+ if ((s64)(now - cl->last) > 0)
+ cl->last = now;
}
cbq_update_toplevel(q, this, q->tx_borrowed);
struct sk_buff *skb;
struct cbq_sched_data *q = qdisc_priv(sch);
psched_time_t now;
- psched_tdiff_t incr;
now = psched_get_time();
- incr = now - q->now_rt;
-
- if (q->tx_class) {
- psched_tdiff_t incr2;
- /* Time integrator. We calculate EOS time
- * by adding expected packet transmission time.
- * If real time is greater, we warp artificial clock,
- * so that:
- *
- * cbq_time = max(real_time, work);
- */
- incr2 = L2T(&q->link, q->tx_len);
- q->now += incr2;
+
+ if (q->tx_class)
cbq_update(q);
- if ((incr -= incr2) < 0)
- incr = 0;
- q->now += incr;
- } else {
- if (now > q->now)
- q->now = now;
- }
- q->now_rt = now;
+
+ q->now = now;
for (;;) {
q->wd_expires = 0;
hrtimer_cancel(&q->delay_timer);
q->toplevel = TC_CBQ_MAXLEVEL;
q->now = psched_get_time();
- q->now_rt = q->now;
for (prio = 0; prio <= TC_CBQ_MAXPRIO; prio++)
q->active[prio] = NULL;
q->delay_timer.function = cbq_undelay;
q->toplevel = TC_CBQ_MAXLEVEL;
q->now = psched_get_time();
- q->now_rt = q->now;
cbq_link_class(&q->link);
else {
dst_release(transport->dst);
transport->dst = NULL;
+ ulp_notify = false;
}
spc_state = SCTP_ADDR_UNREACHABLE;
{
u8 score_curr, score_best;
- if (best == NULL)
+ if (best == NULL || curr == best)
return curr;
score_curr = sctp_trans_score(curr);
trans_sec = trans_pri;
/* If we failed to find a usable transport, just camp on the
- * primary or retran, even if they are inactive, if possible
- * pick a PF iff it's the better choice.
+ * active or pick a PF iff it's the better choice.
*/
if (trans_pri == NULL) {
- trans_pri = sctp_trans_elect_best(asoc->peer.primary_path,
- asoc->peer.retran_path);
- trans_pri = sctp_trans_elect_best(trans_pri, trans_pf);
- trans_sec = asoc->peer.primary_path;
+ trans_pri = sctp_trans_elect_best(asoc->peer.active_path, trans_pf);
+ trans_sec = trans_pri;
}
/* Set the active and retran transports. */
return msg_importance(&port->phdr);
}
-static inline void tipc_port_set_importance(struct tipc_port *port, int imp)
+static inline int tipc_port_set_importance(struct tipc_port *port, int imp)
{
+ if (imp > TIPC_CRITICAL_IMPORTANCE)
+ return -EINVAL;
msg_set_importance(&port->phdr, (u32)imp);
+ return 0;
}
#endif
switch (opt) {
case TIPC_IMPORTANCE:
- tipc_port_set_importance(port, value);
+ res = tipc_port_set_importance(port, value);
break;
case TIPC_SRC_DROPPABLE:
if (sock->type != SOCK_STREAM)
arg-check = $(if $(strip $(cmd_$@)),,1)
endif
-# >'< substitution is for echo to work,
-# >$< substitution to preserve $ when reloading .cmd file
-# note: when using inline perl scripts [perl -e '...$$t=1;...']
-# in $(cmd_xxx) double $$ your perl vars
-make-cmd = $(subst \\,\\\\,$(subst \#,\\\#,$(subst $$,$$$$,$(call escsq,$(cmd_$(1))))))
+# Replace >$< with >$$< to preserve $ when reloading the .cmd file
+# (needed for make)
+# Replace >#< with >\#< to avoid starting a comment in the .cmd file
+# (needed for make)
+# Replace >'< with >'\''< to be able to enclose the whole string in '...'
+# (needed for the shell)
+make-cmd = $(call escsq,$(subst \#,\\\#,$(subst $$,$$$$,$(cmd_$(1)))))
# Find any prerequisites that is newer than target or that does not exist.
# PHONY targets skipped in both cases.
if_changed = $(if $(strip $(any-prereq) $(arg-check)), \
@set -e; \
$(echo-cmd) $(cmd_$(1)); \
- echo 'cmd_$@ := $(make-cmd)' > $(dot-target).cmd)
+ printf '%s\n' 'cmd_$@ := $(make-cmd)' > $(dot-target).cmd)
# Execute the command and also postprocess generated .d dependencies file.
if_changed_dep = $(if $(strip $(any-prereq) $(arg-check) ), \
# build a list of files to remove, usually relative to the current
# directory
-__clean-files := $(extra-y) $(always) \
- $(targets) $(clean-files) \
+__clean-files := $(extra-y) $(extra-m) $(extra-) \
+ $(always) $(targets) $(clean-files) \
$(host-progs) \
$(hostprogs-y) $(hostprogs-m) $(hostprogs-)
warning-1 += $(call cc-option, -Wunused-but-set-variable)
warning-1 += $(call cc-disable-warning, missing-field-initializers)
-# Clang
-warning-1 += $(call cc-disable-warning, initializer-overrides)
-warning-1 += $(call cc-disable-warning, unused-value)
-warning-1 += $(call cc-disable-warning, format)
-warning-1 += $(call cc-disable-warning, unknown-warning-option)
-warning-1 += $(call cc-disable-warning, sign-compare)
-warning-1 += $(call cc-disable-warning, format-zero-length)
-warning-1 += $(call cc-disable-warning, uninitialized)
-warning-1 += $(call cc-option, -fcatch-undefined-behavior)
-
warning-2 := -Waggregate-return
warning-2 += -Wcast-align
warning-2 += -Wdisabled-optimization
endif
KBUILD_CFLAGS += $(warning)
+else
+
+ifeq ($(COMPILER),clang)
+KBUILD_CFLAGS += $(call cc-disable-warning, initializer-overrides)
+KBUILD_CFLAGS += $(call cc-disable-warning, unused-value)
+KBUILD_CFLAGS += $(call cc-disable-warning, format)
+KBUILD_CFLAGS += $(call cc-disable-warning, unknown-warning-option)
+KBUILD_CFLAGS += $(call cc-disable-warning, sign-compare)
+KBUILD_CFLAGS += $(call cc-disable-warning, format-zero-length)
+KBUILD_CFLAGS += $(call cc-disable-warning, uninitialized)
+endif
endif
# Will compile qconf as a C++ program, and menu as a C program.
# They are linked as C++ code to the executable qconf
-# hostprogs-y := conf
-# conf-objs := conf.o libkconfig.so
-# libkconfig-objs := expr.o type.o
-# Will create a shared library named libkconfig.so that consists of
-# expr.o and type.o (they are both compiled as C code and the object files
-# are made as position independent code).
-# conf.c is compiled as a C program, and conf.o is linked together with
-# libkconfig.so as the executable conf.
-# Note: Shared libraries consisting of C++ files are not supported
-
__hostprogs := $(sort $(hostprogs-y) $(hostprogs-m))
# C code
# Executables compiled from a single .c file
-host-csingle := $(foreach m,$(__hostprogs),$(if $($(m)-objs),,$(m)))
+host-csingle := $(foreach m,$(__hostprogs), \
+ $(if $($(m)-objs)$($(m)-cxxobjs),,$(m)))
# C executables linked based on several .o files
host-cmulti := $(foreach m,$(__hostprogs),\
host-cobjs := $(sort $(foreach m,$(__hostprogs),$($(m)-objs)))
# C++ code
-# C++ executables compiled from at least on .cc file
+# C++ executables compiled from at least one .cc file
# and zero or more .c files
host-cxxmulti := $(foreach m,$(__hostprogs),$(if $($(m)-cxxobjs),$(m)))
# C++ Object (.o) files compiled from .cc files
host-cxxobjs := $(sort $(foreach m,$(host-cxxmulti),$($(m)-cxxobjs)))
-# Shared libaries (only .c supported)
-# Shared libraries (.so) - all .so files referenced in "xxx-objs"
-host-cshlib := $(sort $(filter %.so, $(host-cobjs)))
-# Remove .so files from "xxx-objs"
-host-cobjs := $(filter-out %.so,$(host-cobjs))
-
-#Object (.o) files used by the shared libaries
-host-cshobjs := $(sort $(foreach m,$(host-cshlib),$($(m:.so=-objs))))
-
# output directory for programs/.o files
-# hostprogs-y := tools/build may have been specified. Retrieve directory
-host-objdirs := $(foreach f,$(__hostprogs), $(if $(dir $(f)),$(dir $(f))))
-# directory of .o files from prog-objs notation
-host-objdirs += $(foreach f,$(host-cmulti), \
- $(foreach m,$($(f)-objs), \
- $(if $(dir $(m)),$(dir $(m)))))
-# directory of .o files from prog-cxxobjs notation
-host-objdirs += $(foreach f,$(host-cxxmulti), \
- $(foreach m,$($(f)-cxxobjs), \
- $(if $(dir $(m)),$(dir $(m)))))
+# hostprogs-y := tools/build may have been specified.
+# Retrieve also directory of .o files from prog-objs or prog-cxxobjs notation
+host-objdirs := $(dir $(__hostprogs) $(host-cobjs) $(host-cxxobjs))
host-objdirs := $(strip $(sort $(filter-out ./,$(host-objdirs))))
host-cobjs := $(addprefix $(obj)/,$(host-cobjs))
host-cxxmulti := $(addprefix $(obj)/,$(host-cxxmulti))
host-cxxobjs := $(addprefix $(obj)/,$(host-cxxobjs))
-host-cshlib := $(addprefix $(obj)/,$(host-cshlib))
-host-cshobjs := $(addprefix $(obj)/,$(host-cshobjs))
host-objdirs := $(addprefix $(obj)/,$(host-objdirs))
obj-dirs += $(host-objdirs)
cmd_host-cmulti = $(HOSTCC) $(HOSTLDFLAGS) -o $@ \
$(addprefix $(obj)/,$($(@F)-objs)) \
$(HOST_LOADLIBES) $(HOSTLOADLIBES_$(@F))
-$(host-cmulti): $(obj)/%: $(host-cobjs) $(host-cshlib) FORCE
+$(host-cmulti): $(obj)/%: $(host-cobjs) FORCE
$(call if_changed,host-cmulti)
# Create .o file from a single .c file
$(foreach o,objs cxxobjs,\
$(addprefix $(obj)/,$($(@F)-$(o)))) \
$(HOST_LOADLIBES) $(HOSTLOADLIBES_$(@F))
-$(host-cxxmulti): $(obj)/%: $(host-cobjs) $(host-cxxobjs) $(host-cshlib) FORCE
+$(host-cxxmulti): $(obj)/%: $(host-cobjs) $(host-cxxobjs) FORCE
$(call if_changed,host-cxxmulti)
# Create .o file from a single .cc (C++) file
$(host-cxxobjs): $(obj)/%.o: $(src)/%.cc FORCE
$(call if_changed_dep,host-cxxobjs)
-# Compile .c file, create position independent .o file
-# host-cshobjs -> .o
-quiet_cmd_host-cshobjs = HOSTCC -fPIC $@
- cmd_host-cshobjs = $(HOSTCC) $(hostc_flags) -fPIC -c -o $@ $<
-$(host-cshobjs): $(obj)/%.o: $(src)/%.c FORCE
- $(call if_changed_dep,host-cshobjs)
-
-# Link a shared library, based on position independent .o files
-# *.o -> .so shared library (host-cshlib)
-quiet_cmd_host-cshlib = HOSTLLD -shared $@
- cmd_host-cshlib = $(HOSTCC) $(HOSTLDFLAGS) -shared -o $@ \
- $(addprefix $(obj)/,$($(@F:.so=-objs))) \
- $(HOST_LOADLIBES) $(HOSTLOADLIBES_$(@F))
-$(host-cshlib): $(obj)/%: $(host-cshobjs) FORCE
- $(call if_changed,host-cshlib)
-
targets += $(host-csingle) $(host-cmulti) $(host-cobjs)\
- $(host-cxxmulti) $(host-cxxobjs) $(host-cshlib) $(host-cshobjs)
+ $(host-cxxmulti) $(host-cxxobjs)
# CONFIG_FUNCTION_TRACER=y
# CONFIG_FUNCTION_GRAPH_TRACER=y
#
+# For kernel versions older than 3.15:
# The following additional kernel parameters are required:
# (e.g. in file /etc/default/grub)
# GRUB_CMDLINE_LINUX_DEFAULT="... initcall_debug log_buf_len=16M ..."
#
+# ----------------- LIBRARIES --------------------
+
import sys
import time
import os
import string
import re
-import array
import platform
-import datetime
+from datetime import datetime
import struct
-# -- classes --
+# ----------------- CLASSES --------------------
+# Class: SystemValues
+# Description:
+# A global, single-instance container used to
+# store system values and test parameters
class SystemValues:
- testdir = "."
- tpath = "/sys/kernel/debug/tracing/"
- mempath = "/dev/mem"
- powerfile = "/sys/power/state"
- suspendmode = "mem"
- prefix = "test"
- teststamp = ""
- dmesgfile = ""
- ftracefile = ""
- htmlfile = ""
+ version = 3.0
+ verbose = False
+ testdir = '.'
+ tpath = '/sys/kernel/debug/tracing/'
+ fpdtpath = '/sys/firmware/acpi/tables/FPDT'
+ epath = '/sys/kernel/debug/tracing/events/power/'
+ traceevents = [
+ 'suspend_resume',
+ 'device_pm_callback_end',
+ 'device_pm_callback_start'
+ ]
+ modename = {
+ 'freeze': 'Suspend-To-Idle (S0)',
+ 'standby': 'Power-On Suspend (S1)',
+ 'mem': 'Suspend-to-RAM (S3)',
+ 'disk': 'Suspend-to-disk (S4)'
+ }
+ mempath = '/dev/mem'
+ powerfile = '/sys/power/state'
+ suspendmode = 'mem'
+ hostname = 'localhost'
+ prefix = 'test'
+ teststamp = ''
+ dmesgfile = ''
+ ftracefile = ''
+ htmlfile = ''
rtcwake = False
+ rtcwaketime = 10
+ rtcpath = ''
+ android = False
+ adb = 'adb'
+ devicefilter = []
+ stamp = 0
+ execcount = 1
+ x2delay = 0
+ usecallgraph = False
+ usetraceevents = False
+ usetraceeventsonly = False
+ notestrun = False
+ altdevname = dict()
+ postresumetime = 0
+ tracertypefmt = '# tracer: (?P<t>.*)'
+ firmwarefmt = '# fwsuspend (?P<s>[0-9]*) fwresume (?P<r>[0-9]*)$'
+ postresumefmt = '# post resume time (?P<t>[0-9]*)$'
+ stampfmt = '# suspend-(?P<m>[0-9]{2})(?P<d>[0-9]{2})(?P<y>[0-9]{2})-'+\
+ '(?P<H>[0-9]{2})(?P<M>[0-9]{2})(?P<S>[0-9]{2})'+\
+ ' (?P<host>.*) (?P<mode>.*) (?P<kernel>.*)$'
+ def __init__(self):
+ self.hostname = platform.node()
+ if(self.hostname == ''):
+ self.hostname = 'localhost'
+ rtc = "rtc0"
+ if os.path.exists('/dev/rtc'):
+ rtc = os.readlink('/dev/rtc')
+ rtc = '/sys/class/rtc/'+rtc
+ if os.path.exists(rtc) and os.path.exists(rtc+'/date') and \
+ os.path.exists(rtc+'/time') and os.path.exists(rtc+'/wakealarm'):
+ self.rtcpath = rtc
def setOutputFile(self):
- if((self.htmlfile == "") and (self.dmesgfile != "")):
- m = re.match(r"(?P<name>.*)_dmesg\.txt$", self.dmesgfile)
+ if((self.htmlfile == '') and (self.dmesgfile != '')):
+ m = re.match('(?P<name>.*)_dmesg\.txt$', self.dmesgfile)
if(m):
- self.htmlfile = m.group("name")+".html"
- if((self.htmlfile == "") and (self.ftracefile != "")):
- m = re.match(r"(?P<name>.*)_ftrace\.txt$", self.ftracefile)
+ self.htmlfile = m.group('name')+'.html'
+ if((self.htmlfile == '') and (self.ftracefile != '')):
+ m = re.match('(?P<name>.*)_ftrace\.txt$', self.ftracefile)
if(m):
- self.htmlfile = m.group("name")+".html"
- if(self.htmlfile == ""):
- self.htmlfile = "output.html"
- def initTestOutput(self):
- hostname = platform.node()
- if(hostname != ""):
- self.prefix = hostname
- v = os.popen("cat /proc/version").read().strip()
- kver = string.split(v)[2]
- self.testdir = os.popen("date \"+suspend-%m%d%y-%H%M%S\"").read().strip()
- self.teststamp = "# "+self.testdir+" "+self.prefix+" "+self.suspendmode+" "+kver
- self.dmesgfile = self.testdir+"/"+self.prefix+"_"+self.suspendmode+"_dmesg.txt"
- self.ftracefile = self.testdir+"/"+self.prefix+"_"+self.suspendmode+"_ftrace.txt"
- self.htmlfile = self.testdir+"/"+self.prefix+"_"+self.suspendmode+".html"
+ self.htmlfile = m.group('name')+'.html'
+ if(self.htmlfile == ''):
+ self.htmlfile = 'output.html'
+ def initTestOutput(self, subdir):
+ if(not self.android):
+ self.prefix = self.hostname
+ v = open('/proc/version', 'r').read().strip()
+ kver = string.split(v)[2]
+ else:
+ self.prefix = 'android'
+ v = os.popen(self.adb+' shell cat /proc/version').read().strip()
+ kver = string.split(v)[2]
+ testtime = datetime.now().strftime('suspend-%m%d%y-%H%M%S')
+ if(subdir != "."):
+ self.testdir = subdir+"/"+testtime
+ else:
+ self.testdir = testtime
+ self.teststamp = \
+ '# '+testtime+' '+self.prefix+' '+self.suspendmode+' '+kver
+ self.dmesgfile = \
+ self.testdir+'/'+self.prefix+'_'+self.suspendmode+'_dmesg.txt'
+ self.ftracefile = \
+ self.testdir+'/'+self.prefix+'_'+self.suspendmode+'_ftrace.txt'
+ self.htmlfile = \
+ self.testdir+'/'+self.prefix+'_'+self.suspendmode+'.html'
os.mkdir(self.testdir)
+ def setDeviceFilter(self, devnames):
+ self.devicefilter = string.split(devnames)
+ def rtcWakeAlarm(self):
+ os.system('echo 0 > '+self.rtcpath+'/wakealarm')
+ outD = open(self.rtcpath+'/date', 'r').read().strip()
+ outT = open(self.rtcpath+'/time', 'r').read().strip()
+ mD = re.match('^(?P<y>[0-9]*)-(?P<m>[0-9]*)-(?P<d>[0-9]*)', outD)
+ mT = re.match('^(?P<h>[0-9]*):(?P<m>[0-9]*):(?P<s>[0-9]*)', outT)
+ if(mD and mT):
+ # get the current time from hardware
+ utcoffset = int((datetime.now() - datetime.utcnow()).total_seconds())
+ dt = datetime(\
+ int(mD.group('y')), int(mD.group('m')), int(mD.group('d')),
+ int(mT.group('h')), int(mT.group('m')), int(mT.group('s')))
+ nowtime = int(dt.strftime('%s')) + utcoffset
+ else:
+ # if hardware time fails, use the software time
+ nowtime = int(datetime.now().strftime('%s'))
+ alarm = nowtime + self.rtcwaketime
+ os.system('echo %d > %s/wakealarm' % (alarm, self.rtcpath))
+sysvals = SystemValues()
+
+# Class: DeviceNode
+# Description:
+# A container used to create a device hierachy, with a single root node
+# and a tree of child nodes. Used by Data.deviceTopology()
+class DeviceNode:
+ name = ''
+ children = 0
+ depth = 0
+ def __init__(self, nodename, nodedepth):
+ self.name = nodename
+ self.children = []
+ self.depth = nodedepth
+
+# Class: Data
+# Description:
+# The primary container for suspend/resume test data. There is one for
+# each test run. The data is organized into a cronological hierarchy:
+# Data.dmesg {
+# root structure, started as dmesg & ftrace, but now only ftrace
+# contents: times for suspend start/end, resume start/end, fwdata
+# phases {
+# 10 sequential, non-overlapping phases of S/R
+# contents: times for phase start/end, order/color data for html
+# devlist {
+# device callback or action list for this phase
+# device {
+# a single device callback or generic action
+# contents: start/stop times, pid/cpu/driver info
+# parents/children, html id for timeline/callgraph
+# optionally includes an ftrace callgraph
+# optionally includes intradev trace events
+# }
+# }
+# }
+# }
+#
class Data:
- altdevname = dict()
- usedmesg = False
- useftrace = False
- notestrun = False
- verbose = False
- phases = []
- dmesg = {} # root data structure
- start = 0.0
- end = 0.0
- stamp = {'time': "", 'host': "", 'mode': ""}
- id = 0
- tSuspended = 0.0
- fwValid = False
- fwSuspend = 0
- fwResume = 0
- def initialize(self):
- self.dmesg = { # dmesg log data
- 'suspend_general': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "#CCFFCC", 'order': 0},
- 'suspend_early': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "green", 'order': 1},
+ dmesg = {} # root data structure
+ phases = [] # ordered list of phases
+ start = 0.0 # test start
+ end = 0.0 # test end
+ tSuspended = 0.0 # low-level suspend start
+ tResumed = 0.0 # low-level resume start
+ tLow = 0.0 # time spent in low-level suspend (standby/freeze)
+ fwValid = False # is firmware data available
+ fwSuspend = 0 # time spent in firmware suspend
+ fwResume = 0 # time spent in firmware resume
+ dmesgtext = [] # dmesg text file in memory
+ testnumber = 0
+ idstr = ''
+ html_device_id = 0
+ stamp = 0
+ outfile = ''
+ def __init__(self, num):
+ idchar = 'abcdefghijklmnopqrstuvwxyz'
+ self.testnumber = num
+ self.idstr = idchar[num]
+ self.dmesgtext = []
+ self.phases = []
+ self.dmesg = { # fixed list of 10 phases
+ 'suspend_prepare': {'list': dict(), 'start': -1.0, 'end': -1.0,
+ 'row': 0, 'color': '#CCFFCC', 'order': 0},
+ 'suspend': {'list': dict(), 'start': -1.0, 'end': -1.0,
+ 'row': 0, 'color': '#88FF88', 'order': 1},
+ 'suspend_late': {'list': dict(), 'start': -1.0, 'end': -1.0,
+ 'row': 0, 'color': '#00AA00', 'order': 2},
'suspend_noirq': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "#00FFFF", 'order': 2},
- 'suspend_cpu': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "blue", 'order': 3},
- 'resume_cpu': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "red", 'order': 4},
+ 'row': 0, 'color': '#008888', 'order': 3},
+ 'suspend_machine': {'list': dict(), 'start': -1.0, 'end': -1.0,
+ 'row': 0, 'color': '#0000FF', 'order': 4},
+ 'resume_machine': {'list': dict(), 'start': -1.0, 'end': -1.0,
+ 'row': 0, 'color': '#FF0000', 'order': 5},
'resume_noirq': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "orange", 'order': 5},
+ 'row': 0, 'color': '#FF9900', 'order': 6},
'resume_early': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "yellow", 'order': 6},
- 'resume_general': {'list': dict(), 'start': -1.0, 'end': -1.0,
- 'row': 0, 'color': "#FFFFCC", 'order': 7}
+ 'row': 0, 'color': '#FFCC00', 'order': 7},
+ 'resume': {'list': dict(), 'start': -1.0, 'end': -1.0,
+ 'row': 0, 'color': '#FFFF88', 'order': 8},
+ 'resume_complete': {'list': dict(), 'start': -1.0, 'end': -1.0,
+ 'row': 0, 'color': '#FFFFCC', 'order': 9}
}
self.phases = self.sortedPhases()
- def normalizeTime(self):
- tSus = tRes = self.tSuspended
- if self.fwValid:
- tSus -= -self.fwSuspend / 1000000000.0
- tRes -= self.fwResume / 1000000000.0
- self.tSuspended = 0.0
- self.start -= tSus
- self.end -= tRes
+ def getStart(self):
+ return self.dmesg[self.phases[0]]['start']
+ def setStart(self, time):
+ self.start = time
+ self.dmesg[self.phases[0]]['start'] = time
+ def getEnd(self):
+ return self.dmesg[self.phases[-1]]['end']
+ def setEnd(self, time):
+ self.end = time
+ self.dmesg[self.phases[-1]]['end'] = time
+ def isTraceEventOutsideDeviceCalls(self, pid, time):
+ for phase in self.phases:
+ list = self.dmesg[phase]['list']
+ for dev in list:
+ d = list[dev]
+ if(d['pid'] == pid and time >= d['start'] and
+ time <= d['end']):
+ return False
+ return True
+ def addIntraDevTraceEvent(self, action, name, pid, time):
+ if(action == 'mutex_lock_try'):
+ color = 'red'
+ elif(action == 'mutex_lock_pass'):
+ color = 'green'
+ elif(action == 'mutex_unlock'):
+ color = 'blue'
+ else:
+ # create separate colors based on the name
+ v1 = len(name)*10 % 256
+ v2 = string.count(name, 'e')*100 % 256
+ v3 = ord(name[0])*20 % 256
+ color = '#%06X' % ((v1*0x10000) + (v2*0x100) + v3)
+ for phase in self.phases:
+ list = self.dmesg[phase]['list']
+ for dev in list:
+ d = list[dev]
+ if(d['pid'] == pid and time >= d['start'] and
+ time <= d['end']):
+ e = TraceEvent(action, name, color, time)
+ if('traceevents' not in d):
+ d['traceevents'] = []
+ d['traceevents'].append(e)
+ return d
+ break
+ return 0
+ def capIntraDevTraceEvent(self, action, name, pid, time):
+ for phase in self.phases:
+ list = self.dmesg[phase]['list']
+ for dev in list:
+ d = list[dev]
+ if(d['pid'] == pid and time >= d['start'] and
+ time <= d['end']):
+ if('traceevents' not in d):
+ return
+ for e in d['traceevents']:
+ if(e.action == action and
+ e.name == name and not e.ready):
+ e.length = time - e.time
+ e.ready = True
+ break
+ return
+ def trimTimeVal(self, t, t0, dT, left):
+ if left:
+ if(t > t0):
+ if(t - dT < t0):
+ return t0
+ return t - dT
+ else:
+ return t
+ else:
+ if(t < t0 + dT):
+ if(t > t0):
+ return t0 + dT
+ return t + dT
+ else:
+ return t
+ def trimTime(self, t0, dT, left):
+ self.tSuspended = self.trimTimeVal(self.tSuspended, t0, dT, left)
+ self.tResumed = self.trimTimeVal(self.tResumed, t0, dT, left)
+ self.start = self.trimTimeVal(self.start, t0, dT, left)
+ self.end = self.trimTimeVal(self.end, t0, dT, left)
for phase in self.phases:
- zero = tRes
- if "suspend" in phase:
- zero = tSus
p = self.dmesg[phase]
- p['start'] -= zero
- p['end'] -= zero
+ p['start'] = self.trimTimeVal(p['start'], t0, dT, left)
+ p['end'] = self.trimTimeVal(p['end'], t0, dT, left)
list = p['list']
for name in list:
d = list[name]
- d['start'] -= zero
- d['end'] -= zero
+ d['start'] = self.trimTimeVal(d['start'], t0, dT, left)
+ d['end'] = self.trimTimeVal(d['end'], t0, dT, left)
if('ftrace' in d):
cg = d['ftrace']
- cg.start -= zero
- cg.end -= zero
+ cg.start = self.trimTimeVal(cg.start, t0, dT, left)
+ cg.end = self.trimTimeVal(cg.end, t0, dT, left)
for line in cg.list:
- line.time -= zero
- if self.fwValid:
- fws = -self.fwSuspend / 1000000000.0
- fwr = self.fwResume / 1000000000.0
- list = dict()
- self.id += 1
- devid = "dc%d" % self.id
- list["firmware-suspend"] = \
- {'start': fws, 'end': 0, 'pid': 0, 'par': "",
- 'length': -fws, 'row': 0, 'id': devid };
- self.id += 1
- devid = "dc%d" % self.id
- list["firmware-resume"] = \
- {'start': 0, 'end': fwr, 'pid': 0, 'par': "",
- 'length': fwr, 'row': 0, 'id': devid };
- self.dmesg['BIOS'] = \
- {'list': list, 'start': fws, 'end': fwr,
- 'row': 0, 'color': "purple", 'order': 4}
- self.dmesg['resume_cpu']['order'] += 1
- self.dmesg['resume_noirq']['order'] += 1
- self.dmesg['resume_early']['order'] += 1
- self.dmesg['resume_general']['order'] += 1
- self.phases = self.sortedPhases()
- def vprint(self, msg):
- if(self.verbose):
- print(msg)
+ line.time = self.trimTimeVal(line.time, t0, dT, left)
+ if('traceevents' in d):
+ for e in d['traceevents']:
+ e.time = self.trimTimeVal(e.time, t0, dT, left)
+ def normalizeTime(self, tZero):
+ # first trim out any standby or freeze clock time
+ if(self.tSuspended != self.tResumed):
+ if(self.tResumed > tZero):
+ self.trimTime(self.tSuspended, \
+ self.tResumed-self.tSuspended, True)
+ else:
+ self.trimTime(self.tSuspended, \
+ self.tResumed-self.tSuspended, False)
+ # shift the timeline so that tZero is the new 0
+ self.tSuspended -= tZero
+ self.tResumed -= tZero
+ self.start -= tZero
+ self.end -= tZero
+ for phase in self.phases:
+ p = self.dmesg[phase]
+ p['start'] -= tZero
+ p['end'] -= tZero
+ list = p['list']
+ for name in list:
+ d = list[name]
+ d['start'] -= tZero
+ d['end'] -= tZero
+ if('ftrace' in d):
+ cg = d['ftrace']
+ cg.start -= tZero
+ cg.end -= tZero
+ for line in cg.list:
+ line.time -= tZero
+ if('traceevents' in d):
+ for e in d['traceevents']:
+ e.time -= tZero
+ def newPhaseWithSingleAction(self, phasename, devname, start, end, color):
+ for phase in self.phases:
+ self.dmesg[phase]['order'] += 1
+ self.html_device_id += 1
+ devid = '%s%d' % (self.idstr, self.html_device_id)
+ list = dict()
+ list[devname] = \
+ {'start': start, 'end': end, 'pid': 0, 'par': '',
+ 'length': (end-start), 'row': 0, 'id': devid, 'drv': '' };
+ self.dmesg[phasename] = \
+ {'list': list, 'start': start, 'end': end,
+ 'row': 0, 'color': color, 'order': 0}
+ self.phases = self.sortedPhases()
+ def newPhase(self, phasename, start, end, color, order):
+ if(order < 0):
+ order = len(self.phases)
+ for phase in self.phases[order:]:
+ self.dmesg[phase]['order'] += 1
+ if(order > 0):
+ p = self.phases[order-1]
+ self.dmesg[p]['end'] = start
+ if(order < len(self.phases)):
+ p = self.phases[order]
+ self.dmesg[p]['start'] = end
+ list = dict()
+ self.dmesg[phasename] = \
+ {'list': list, 'start': start, 'end': end,
+ 'row': 0, 'color': color, 'order': order}
+ self.phases = self.sortedPhases()
+ def setPhase(self, phase, ktime, isbegin):
+ if(isbegin):
+ self.dmesg[phase]['start'] = ktime
+ else:
+ self.dmesg[phase]['end'] = ktime
def dmesgSortVal(self, phase):
return self.dmesg[phase]['order']
def sortedPhases(self):
dev = phaselist[devname]
if(dev['end'] < 0):
dev['end'] = end
- self.vprint("%s (%s): callback didn't return" % (devname, phase))
+ vprint('%s (%s): callback didnt return' % (devname, phase))
+ def deviceFilter(self, devicefilter):
+ # remove all by the relatives of the filter devnames
+ filter = []
+ for phase in self.phases:
+ list = self.dmesg[phase]['list']
+ for name in devicefilter:
+ dev = name
+ while(dev in list):
+ if(dev not in filter):
+ filter.append(dev)
+ dev = list[dev]['par']
+ children = self.deviceDescendants(name, phase)
+ for dev in children:
+ if(dev not in filter):
+ filter.append(dev)
+ for phase in self.phases:
+ list = self.dmesg[phase]['list']
+ rmlist = []
+ for name in list:
+ pid = list[name]['pid']
+ if(name not in filter and pid >= 0):
+ rmlist.append(name)
+ for name in rmlist:
+ del list[name]
def fixupInitcallsThatDidntReturn(self):
# if any calls never returned, clip them at system resume end
for phase in self.phases:
- self.fixupInitcalls(phase, self.dmesg['resume_general']['end'])
- if(phase == "resume_general"):
- break
- def newAction(self, phase, name, pid, parent, start, end):
- self.id += 1
- devid = "dc%d" % self.id
+ self.fixupInitcalls(phase, self.getEnd())
+ def newActionGlobal(self, name, start, end):
+ # which phase is this device callback or action "in"
+ targetphase = "none"
+ overlap = 0.0
+ for phase in self.phases:
+ pstart = self.dmesg[phase]['start']
+ pend = self.dmesg[phase]['end']
+ o = max(0, min(end, pend) - max(start, pstart))
+ if(o > overlap):
+ targetphase = phase
+ overlap = o
+ if targetphase in self.phases:
+ self.newAction(targetphase, name, -1, '', start, end, '')
+ return True
+ return False
+ def newAction(self, phase, name, pid, parent, start, end, drv):
+ # new device callback for a specific phase
+ self.html_device_id += 1
+ devid = '%s%d' % (self.idstr, self.html_device_id)
list = self.dmesg[phase]['list']
length = -1.0
if(start >= 0 and end >= 0):
length = end - start
list[name] = {'start': start, 'end': end, 'pid': pid, 'par': parent,
- 'length': length, 'row': 0, 'id': devid }
+ 'length': length, 'row': 0, 'id': devid, 'drv': drv }
def deviceIDs(self, devlist, phase):
idlist = []
- for p in self.phases:
- if(p[0] != phase[0]):
- continue
- list = data.dmesg[p]['list']
- for devname in list:
- if devname in devlist:
- idlist.append(list[devname]['id'])
+ list = self.dmesg[phase]['list']
+ for devname in list:
+ if devname in devlist:
+ idlist.append(list[devname]['id'])
return idlist
def deviceParentID(self, devname, phase):
- pdev = ""
- pdevid = ""
- for p in self.phases:
- if(p[0] != phase[0]):
- continue
- list = data.dmesg[p]['list']
- if devname in list:
- pdev = list[devname]['par']
- for p in self.phases:
- if(p[0] != phase[0]):
- continue
- list = data.dmesg[p]['list']
- if pdev in list:
- return list[pdev]['id']
+ pdev = ''
+ pdevid = ''
+ list = self.dmesg[phase]['list']
+ if devname in list:
+ pdev = list[devname]['par']
+ if pdev in list:
+ return list[pdev]['id']
return pdev
- def deviceChildrenIDs(self, devname, phase):
+ def deviceChildren(self, devname, phase):
devlist = []
- for p in self.phases:
- if(p[0] != phase[0]):
- continue
- list = data.dmesg[p]['list']
- for child in list:
- if(list[child]['par'] == devname):
- devlist.append(child)
+ list = self.dmesg[phase]['list']
+ for child in list:
+ if(list[child]['par'] == devname):
+ devlist.append(child)
+ return devlist
+ def deviceDescendants(self, devname, phase):
+ children = self.deviceChildren(devname, phase)
+ family = children
+ for child in children:
+ family += self.deviceDescendants(child, phase)
+ return family
+ def deviceChildrenIDs(self, devname, phase):
+ devlist = self.deviceChildren(devname, phase)
return self.deviceIDs(devlist, phase)
-
+ def printDetails(self):
+ vprint(' test start: %f' % self.start)
+ for phase in self.phases:
+ dc = len(self.dmesg[phase]['list'])
+ vprint(' %16s: %f - %f (%d devices)' % (phase, \
+ self.dmesg[phase]['start'], self.dmesg[phase]['end'], dc))
+ vprint(' test end: %f' % self.end)
+ def masterTopology(self, name, list, depth):
+ node = DeviceNode(name, depth)
+ for cname in list:
+ clist = self.deviceChildren(cname, 'resume')
+ cnode = self.masterTopology(cname, clist, depth+1)
+ node.children.append(cnode)
+ return node
+ def printTopology(self, node):
+ html = ''
+ if node.name:
+ info = ''
+ drv = ''
+ for phase in self.phases:
+ list = self.dmesg[phase]['list']
+ if node.name in list:
+ s = list[node.name]['start']
+ e = list[node.name]['end']
+ if list[node.name]['drv']:
+ drv = ' {'+list[node.name]['drv']+'}'
+ info += ('<li>%s: %.3fms</li>' % (phase, (e-s)*1000))
+ html += '<li><b>'+node.name+drv+'</b>'
+ if info:
+ html += '<ul>'+info+'</ul>'
+ html += '</li>'
+ if len(node.children) > 0:
+ html += '<ul>'
+ for cnode in node.children:
+ html += self.printTopology(cnode)
+ html += '</ul>'
+ return html
+ def rootDeviceList(self):
+ # list of devices graphed
+ real = []
+ for phase in self.dmesg:
+ list = self.dmesg[phase]['list']
+ for dev in list:
+ if list[dev]['pid'] >= 0 and dev not in real:
+ real.append(dev)
+ # list of top-most root devices
+ rootlist = []
+ for phase in self.dmesg:
+ list = self.dmesg[phase]['list']
+ for dev in list:
+ pdev = list[dev]['par']
+ if(re.match('[0-9]*-[0-9]*\.[0-9]*[\.0-9]*\:[\.0-9]*$', pdev)):
+ continue
+ if pdev and pdev not in real and pdev not in rootlist:
+ rootlist.append(pdev)
+ return rootlist
+ def deviceTopology(self):
+ rootlist = self.rootDeviceList()
+ master = self.masterTopology('', rootlist, 0)
+ return self.printTopology(master)
+
+# Class: TraceEvent
+# Description:
+# A container for trace event data found in the ftrace file
+class TraceEvent:
+ ready = False
+ name = ''
+ time = 0.0
+ color = '#FFFFFF'
+ length = 0.0
+ action = ''
+ def __init__(self, a, n, c, t):
+ self.action = a
+ self.name = n
+ self.color = c
+ self.time = t
+
+# Class: FTraceLine
+# Description:
+# A container for a single line of ftrace data. There are six basic types:
+# callgraph line:
+# call: " dpm_run_callback() {"
+# return: " }"
+# leaf: " dpm_run_callback();"
+# trace event:
+# tracing_mark_write: SUSPEND START or RESUME COMPLETE
+# suspend_resume: phase or custom exec block data
+# device_pm_callback: device callback info
class FTraceLine:
time = 0.0
length = 0.0
freturn = False
fevent = False
depth = 0
- name = ""
+ name = ''
+ type = ''
def __init__(self, t, m, d):
self.time = float(t)
- # check to see if this is a trace event
- em = re.match(r"^ *\/\* *(?P<msg>.*) \*\/ *$", m)
- if(em):
- self.name = em.group("msg")
+ # is this a trace event
+ if(d == 'traceevent' or re.match('^ *\/\* *(?P<msg>.*) \*\/ *$', m)):
+ if(d == 'traceevent'):
+ # nop format trace event
+ msg = m
+ else:
+ # function_graph format trace event
+ em = re.match('^ *\/\* *(?P<msg>.*) \*\/ *$', m)
+ msg = em.group('msg')
+
+ emm = re.match('^(?P<call>.*?): (?P<msg>.*)', msg)
+ if(emm):
+ self.name = emm.group('msg')
+ self.type = emm.group('call')
+ else:
+ self.name = msg
self.fevent = True
return
# convert the duration to seconds
if(d):
self.length = float(d)/1000000
# the indentation determines the depth
- match = re.match(r"^(?P<d> *)(?P<o>.*)$", m)
+ match = re.match('^(?P<d> *)(?P<o>.*)$', m)
if(not match):
return
self.depth = self.getDepth(match.group('d'))
self.freturn = True
if(len(m) > 1):
# includes comment with function name
- match = re.match(r"^} *\/\* *(?P<n>.*) *\*\/$", m)
+ match = re.match('^} *\/\* *(?P<n>.*) *\*\/$', m)
if(match):
self.name = match.group('n')
# function call
self.fcall = True
# function call with children
if(m[-1] == '{'):
- match = re.match(r"^(?P<n>.*) *\(.*", m)
+ match = re.match('^(?P<n>.*) *\(.*', m)
if(match):
self.name = match.group('n')
# function call with no children (leaf)
elif(m[-1] == ';'):
self.freturn = True
- match = re.match(r"^(?P<n>.*) *\(.*", m)
+ match = re.match('^(?P<n>.*) *\(.*', m)
if(match):
self.name = match.group('n')
# something else (possibly a trace marker)
self.name = m
def getDepth(self, str):
return len(str)/2
+ def debugPrint(self, dev):
+ if(self.freturn and self.fcall):
+ print('%s -- %f (%02d): %s(); (%.3f us)' % (dev, self.time, \
+ self.depth, self.name, self.length*1000000))
+ elif(self.freturn):
+ print('%s -- %f (%02d): %s} (%.3f us)' % (dev, self.time, \
+ self.depth, self.name, self.length*1000000))
+ else:
+ print('%s -- %f (%02d): %s() { (%.3f us)' % (dev, self.time, \
+ self.depth, self.name, self.length*1000000))
+# Class: FTraceCallGraph
+# Description:
+# A container for the ftrace callgraph of a single recursive function.
+# This can be a dpm_run_callback, dpm_prepare, or dpm_complete callgraph
+# Each instance is tied to a single device in a single phase, and is
+# comprised of an ordered list of FTraceLine objects
class FTraceCallGraph:
start = -1.0
end = -1.0
if(not self.invalid):
self.setDepth(line)
if(line.depth == 0 and line.freturn):
+ if(self.start < 0):
+ self.start = line.time
self.end = line.time
self.list.append(line)
return True
if(self.invalid):
return False
if(len(self.list) >= 1000000 or self.depth < 0):
- first = self.list[0]
- self.list = []
- self.list.append(first)
- self.invalid = True
- id = "task %s cpu %s" % (match.group("pid"), match.group("cpu"))
- window = "(%f - %f)" % (self.start, line.time)
- data.vprint("Too much data for "+id+" "+window+", ignoring this callback")
- return False
+ if(len(self.list) > 0):
+ first = self.list[0]
+ self.list = []
+ self.list.append(first)
+ self.invalid = True
+ if(not match):
+ return False
+ id = 'task %s cpu %s' % (match.group('pid'), match.group('cpu'))
+ window = '(%f - %f)' % (self.start, line.time)
+ if(self.depth < 0):
+ print('Too much data for '+id+\
+ ' (buffer overflow), ignoring this callback')
+ else:
+ print('Too much data for '+id+\
+ ' '+window+', ignoring this callback')
+ return False
self.list.append(line)
if(self.start < 0):
self.start = line.time
return False
+ def slice(self, t0, tN):
+ minicg = FTraceCallGraph()
+ count = -1
+ firstdepth = 0
+ for l in self.list:
+ if(l.time < t0 or l.time > tN):
+ continue
+ if(count < 0):
+ if(not l.fcall or l.name == 'dev_driver_string'):
+ continue
+ firstdepth = l.depth
+ count = 0
+ l.depth -= firstdepth
+ minicg.addLine(l, 0)
+ if((count == 0 and l.freturn and l.fcall) or
+ (count > 0 and l.depth <= 0)):
+ break
+ count += 1
+ return minicg
def sanityCheck(self):
stack = dict()
cnt = 0
stack[l.depth] = l
cnt += 1
elif(l.freturn and not l.fcall):
- if(not stack[l.depth]):
+ if(l.depth not in stack):
return False
stack[l.depth].length = l.length
stack[l.depth] = 0
return True
return False
def debugPrint(self, filename):
- if(filename == "stdout"):
- print("[%f - %f]") % (self.start, self.end)
+ if(filename == 'stdout'):
+ print('[%f - %f]') % (self.start, self.end)
for l in self.list:
if(l.freturn and l.fcall):
- print("%f (%02d): %s(); (%.3f us)" % (l.time, l.depth, l.name, l.length*1000000))
+ print('%f (%02d): %s(); (%.3f us)' % (l.time, \
+ l.depth, l.name, l.length*1000000))
elif(l.freturn):
- print("%f (%02d): %s} (%.3f us)" % (l.time, l.depth, l.name, l.length*1000000))
+ print('%f (%02d): %s} (%.3f us)' % (l.time, \
+ l.depth, l.name, l.length*1000000))
else:
- print("%f (%02d): %s() { (%.3f us)" % (l.time, l.depth, l.name, l.length*1000000))
- print(" ")
+ print('%f (%02d): %s() { (%.3f us)' % (l.time, \
+ l.depth, l.name, l.length*1000000))
+ print(' ')
else:
fp = open(filename, 'w')
print(filename)
for l in self.list:
if(l.freturn and l.fcall):
- fp.write("%f (%02d): %s(); (%.3f us)\n" % (l.time, l.depth, l.name, l.length*1000000))
+ fp.write('%f (%02d): %s(); (%.3f us)\n' % (l.time, \
+ l.depth, l.name, l.length*1000000))
elif(l.freturn):
- fp.write("%f (%02d): %s} (%.3f us)\n" % (l.time, l.depth, l.name, l.length*1000000))
+ fp.write('%f (%02d): %s} (%.3f us)\n' % (l.time, \
+ l.depth, l.name, l.length*1000000))
else:
- fp.write("%f (%02d): %s() { (%.3f us)\n" % (l.time, l.depth, l.name, l.length*1000000))
+ fp.write('%f (%02d): %s() { (%.3f us)\n' % (l.time, \
+ l.depth, l.name, l.length*1000000))
fp.close()
+# Class: Timeline
+# Description:
+# A container for a suspend/resume html timeline. In older versions
+# of the script there were multiple timelines, but in the latest
+# there is only one.
class Timeline:
html = {}
- scaleH = 0.0 # height of the timescale row as a percent of the timeline height
+ scaleH = 0.0 # height of the row as a percent of the timeline height
rowH = 0.0 # height of each row in percent of the timeline height
row_height_pixels = 30
maxrows = 0
height = 0
def __init__(self):
self.html = {
- 'timeline': "",
- 'legend': "",
- 'scale': ""
+ 'timeline': '',
+ 'legend': '',
+ 'scale': ''
}
def setRows(self, rows):
self.maxrows = int(rows)
r = 1.0
self.rowH = (100.0 - self.scaleH)/r
-# -- global objects --
+# Class: TestRun
+# Description:
+# A container for a suspend/resume test run. This is necessary as
+# there could be more than one, and they need to be separate.
+class TestRun:
+ ftrace_line_fmt_fg = \
+ '^ *(?P<time>[0-9\.]*) *\| *(?P<cpu>[0-9]*)\)'+\
+ ' *(?P<proc>.*)-(?P<pid>[0-9]*) *\|'+\
+ '[ +!]*(?P<dur>[0-9\.]*) .*\| (?P<msg>.*)'
+ ftrace_line_fmt_nop = \
+ ' *(?P<proc>.*)-(?P<pid>[0-9]*) *\[(?P<cpu>[0-9]*)\] *'+\
+ '(?P<flags>.{4}) *(?P<time>[0-9\.]*): *'+\
+ '(?P<msg>.*)'
+ ftrace_line_fmt = ftrace_line_fmt_nop
+ cgformat = False
+ ftemp = dict()
+ ttemp = dict()
+ inthepipe = False
+ tracertype = ''
+ data = 0
+ def __init__(self, dataobj):
+ self.data = dataobj
+ self.ftemp = dict()
+ self.ttemp = dict()
+ def isReady(self):
+ if(tracertype == '' or not data):
+ return False
+ return True
+ def setTracerType(self, tracer):
+ self.tracertype = tracer
+ if(tracer == 'function_graph'):
+ self.cgformat = True
+ self.ftrace_line_fmt = self.ftrace_line_fmt_fg
+ elif(tracer == 'nop'):
+ self.ftrace_line_fmt = self.ftrace_line_fmt_nop
+ else:
+ doError('Invalid tracer format: [%s]' % tracer, False)
-sysvals = SystemValues()
-data = Data()
+# ----------------- FUNCTIONS --------------------
-# -- functions --
+# Function: vprint
+# Description:
+# verbose print (prints only with -verbose option)
+# Arguments:
+# msg: the debug/log message to print
+def vprint(msg):
+ global sysvals
+ if(sysvals.verbose):
+ print(msg)
# Function: initFtrace
# Description:
-# Configure ftrace to capture a function trace during suspend/resume
+# Configure ftrace to use trace events and/or a callgraph
def initFtrace():
global sysvals
- print("INITIALIZING FTRACE...")
- # turn trace off
- os.system("echo 0 > "+sysvals.tpath+"tracing_on")
- # set the trace clock to global
- os.system("echo global > "+sysvals.tpath+"trace_clock")
- # set trace buffer to a huge value
- os.system("echo nop > "+sysvals.tpath+"current_tracer")
- os.system("echo 100000 > "+sysvals.tpath+"buffer_size_kb")
- # clear the trace buffer
- os.system("echo \"\" > "+sysvals.tpath+"trace")
- # set trace type
- os.system("echo function_graph > "+sysvals.tpath+"current_tracer")
- os.system("echo \"\" > "+sysvals.tpath+"set_ftrace_filter")
- # set trace format options
- os.system("echo funcgraph-abstime > "+sysvals.tpath+"trace_options")
- os.system("echo funcgraph-proc > "+sysvals.tpath+"trace_options")
- # focus only on device suspend and resume
- os.system("cat "+sysvals.tpath+"available_filter_functions | grep dpm_run_callback > "+sysvals.tpath+"set_graph_function")
+ tp = sysvals.tpath
+ cf = 'dpm_run_callback'
+ if(sysvals.usetraceeventsonly):
+ cf = '-e dpm_prepare -e dpm_complete -e dpm_run_callback'
+ if(sysvals.usecallgraph or sysvals.usetraceevents):
+ print('INITIALIZING FTRACE...')
+ # turn trace off
+ os.system('echo 0 > '+tp+'tracing_on')
+ # set the trace clock to global
+ os.system('echo global > '+tp+'trace_clock')
+ # set trace buffer to a huge value
+ os.system('echo nop > '+tp+'current_tracer')
+ os.system('echo 100000 > '+tp+'buffer_size_kb')
+ # initialize the callgraph trace, unless this is an x2 run
+ if(sysvals.usecallgraph and sysvals.execcount == 1):
+ # set trace type
+ os.system('echo function_graph > '+tp+'current_tracer')
+ os.system('echo "" > '+tp+'set_ftrace_filter')
+ # set trace format options
+ os.system('echo funcgraph-abstime > '+tp+'trace_options')
+ os.system('echo funcgraph-proc > '+tp+'trace_options')
+ # focus only on device suspend and resume
+ os.system('cat '+tp+'available_filter_functions | grep '+\
+ cf+' > '+tp+'set_graph_function')
+ if(sysvals.usetraceevents):
+ # turn trace events on
+ events = iter(sysvals.traceevents)
+ for e in events:
+ os.system('echo 1 > '+sysvals.epath+e+'/enable')
+ # clear the trace buffer
+ os.system('echo "" > '+tp+'trace')
+
+# Function: initFtraceAndroid
+# Description:
+# Configure ftrace to capture trace events
+def initFtraceAndroid():
+ global sysvals
+
+ tp = sysvals.tpath
+ if(sysvals.usetraceevents):
+ print('INITIALIZING FTRACE...')
+ # turn trace off
+ os.system(sysvals.adb+" shell 'echo 0 > "+tp+"tracing_on'")
+ # set the trace clock to global
+ os.system(sysvals.adb+" shell 'echo global > "+tp+"trace_clock'")
+ # set trace buffer to a huge value
+ os.system(sysvals.adb+" shell 'echo nop > "+tp+"current_tracer'")
+ os.system(sysvals.adb+" shell 'echo 10000 > "+tp+"buffer_size_kb'")
+ # turn trace events on
+ events = iter(sysvals.traceevents)
+ for e in events:
+ os.system(sysvals.adb+" shell 'echo 1 > "+\
+ sysvals.epath+e+"/enable'")
+ # clear the trace buffer
+ os.system(sysvals.adb+" shell 'echo \"\" > "+tp+"trace'")
# Function: verifyFtrace
# Description:
# Check that ftrace is working on the system
+# Output:
+# True or False
def verifyFtrace():
global sysvals
- files = ["available_filter_functions", "buffer_size_kb",
- "current_tracer", "set_ftrace_filter",
- "trace", "trace_marker"]
+ # files needed for any trace data
+ files = ['buffer_size_kb', 'current_tracer', 'trace', 'trace_clock',
+ 'trace_marker', 'trace_options', 'tracing_on']
+ # files needed for callgraph trace data
+ tp = sysvals.tpath
+ if(sysvals.usecallgraph):
+ files += [
+ 'available_filter_functions',
+ 'set_ftrace_filter',
+ 'set_graph_function'
+ ]
for f in files:
- if(os.path.exists(sysvals.tpath+f) == False):
- return False
+ if(sysvals.android):
+ out = os.popen(sysvals.adb+' shell ls '+tp+f).read().strip()
+ if(out != tp+f):
+ return False
+ else:
+ if(os.path.exists(tp+f) == False):
+ return False
return True
-def parseStamp(line):
- global data, sysvals
- stampfmt = r"# suspend-(?P<m>[0-9]{2})(?P<d>[0-9]{2})(?P<y>[0-9]{2})-"+\
- "(?P<H>[0-9]{2})(?P<M>[0-9]{2})(?P<S>[0-9]{2})"+\
- " (?P<host>.*) (?P<mode>.*) (?P<kernel>.*)$"
- m = re.match(stampfmt, line)
- if(m):
- dt = datetime.datetime(int(m.group("y"))+2000, int(m.group("m")),
- int(m.group("d")), int(m.group("H")), int(m.group("M")),
- int(m.group("S")))
- data.stamp['time'] = dt.strftime("%B %d %Y, %I:%M:%S %p")
- data.stamp['host'] = m.group("host")
- data.stamp['mode'] = m.group("mode")
- data.stamp['kernel'] = m.group("kernel")
- sysvals.suspendmode = data.stamp['mode']
-
-# Function: analyzeTraceLog
+# Function: parseStamp
# Description:
-# Analyse an ftrace log output file generated from this app during
-# the execution phase. Create an "ftrace" structure in memory for
-# subsequent formatting in the html output file
-def analyzeTraceLog():
- global sysvals, data
-
- # the ftrace data is tied to the dmesg data
- if(not data.usedmesg):
- return
-
- # read through the ftrace and parse the data
- data.vprint("Analyzing the ftrace data...")
- ftrace_line_fmt = r"^ *(?P<time>[0-9\.]*) *\| *(?P<cpu>[0-9]*)\)"+\
- " *(?P<proc>.*)-(?P<pid>[0-9]*) *\|"+\
- "[ +!]*(?P<dur>[0-9\.]*) .*\| (?P<msg>.*)"
- ftemp = dict()
- inthepipe = False
+# Pull in the stamp comment line from the data file(s),
+# create the stamp, and add it to the global sysvals object
+# Arguments:
+# m: the valid re.match output for the stamp line
+def parseStamp(m, data):
+ global sysvals
+ data.stamp = {'time': '', 'host': '', 'mode': ''}
+ dt = datetime(int(m.group('y'))+2000, int(m.group('m')),
+ int(m.group('d')), int(m.group('H')), int(m.group('M')),
+ int(m.group('S')))
+ data.stamp['time'] = dt.strftime('%B %d %Y, %I:%M:%S %p')
+ data.stamp['host'] = m.group('host')
+ data.stamp['mode'] = m.group('mode')
+ data.stamp['kernel'] = m.group('kernel')
+ sysvals.suspendmode = data.stamp['mode']
+ if not sysvals.stamp:
+ sysvals.stamp = data.stamp
+
+# Function: diffStamp
+# Description:
+# compare the host, kernel, and mode fields in 3 stamps
+# Arguments:
+# stamp1: string array with mode, kernel, and host
+# stamp2: string array with mode, kernel, and host
+# Return:
+# True if stamps differ, False if they're the same
+def diffStamp(stamp1, stamp2):
+ if 'host' in stamp1 and 'host' in stamp2:
+ if stamp1['host'] != stamp2['host']:
+ return True
+ if 'kernel' in stamp1 and 'kernel' in stamp2:
+ if stamp1['kernel'] != stamp2['kernel']:
+ return True
+ if 'mode' in stamp1 and 'mode' in stamp2:
+ if stamp1['mode'] != stamp2['mode']:
+ return True
+ return False
+
+# Function: doesTraceLogHaveTraceEvents
+# Description:
+# Quickly determine if the ftrace log has some or all of the trace events
+# required for primary parsing. Set the usetraceevents and/or
+# usetraceeventsonly flags in the global sysvals object
+def doesTraceLogHaveTraceEvents():
+ global sysvals
+
+ sysvals.usetraceeventsonly = True
+ sysvals.usetraceevents = False
+ for e in sysvals.traceevents:
+ out = os.popen('cat '+sysvals.ftracefile+' | grep "'+e+': "').read()
+ if(not out):
+ sysvals.usetraceeventsonly = False
+ if(e == 'suspend_resume' and out):
+ sysvals.usetraceevents = True
+
+# Function: appendIncompleteTraceLog
+# Description:
+# [deprecated for kernel 3.15 or newer]
+# Legacy support of ftrace outputs that lack the device_pm_callback
+# and/or suspend_resume trace events. The primary data should be
+# taken from dmesg, and this ftrace is used only for callgraph data
+# or custom actions in the timeline. The data is appended to the Data
+# objects provided.
+# Arguments:
+# testruns: the array of Data objects obtained from parseKernelLog
+def appendIncompleteTraceLog(testruns):
+ global sysvals
+
+ # create TestRun vessels for ftrace parsing
+ testcnt = len(testruns)
+ testidx = -1
+ testrun = []
+ for data in testruns:
+ testrun.append(TestRun(data))
+
+ # extract the callgraph and traceevent data
+ vprint('Analyzing the ftrace data...')
tf = open(sysvals.ftracefile, 'r')
- count = 0
for line in tf:
- count = count + 1
- # grab the time stamp if it's valid
- if(count == 1):
- parseStamp(line)
+ # remove any latent carriage returns
+ line = line.replace('\r\n', '')
+ # grab the time stamp first (signifies the start of the test run)
+ m = re.match(sysvals.stampfmt, line)
+ if(m):
+ testidx += 1
+ parseStamp(m, testrun[testidx].data)
+ continue
+ # pull out any firmware data
+ if(re.match(sysvals.firmwarefmt, line)):
+ continue
+ # if we havent found a test time stamp yet keep spinning til we do
+ if(testidx < 0):
+ continue
+ # determine the trace data type (required for further parsing)
+ m = re.match(sysvals.tracertypefmt, line)
+ if(m):
+ tracer = m.group('t')
+ testrun[testidx].setTracerType(tracer)
continue
- # parse only valid lines
- m = re.match(ftrace_line_fmt, line)
+ # parse only valid lines, if this isnt one move on
+ m = re.match(testrun[testidx].ftrace_line_fmt, line)
if(not m):
continue
- m_time = m.group("time")
- m_pid = m.group("pid")
- m_msg = m.group("msg")
- m_dur = m.group("dur")
+ # gather the basic message data from the line
+ m_time = m.group('time')
+ m_pid = m.group('pid')
+ m_msg = m.group('msg')
+ if(testrun[testidx].cgformat):
+ m_param3 = m.group('dur')
+ else:
+ m_param3 = 'traceevent'
if(m_time and m_pid and m_msg):
- t = FTraceLine(m_time, m_msg, m_dur)
+ t = FTraceLine(m_time, m_msg, m_param3)
pid = int(m_pid)
else:
continue
if(not t.fcall and not t.freturn and not t.fevent):
continue
# only parse the ftrace data during suspend/resume
- if(not inthepipe):
+ data = testrun[testidx].data
+ if(not testrun[testidx].inthepipe):
# look for the suspend start marker
if(t.fevent):
- if(t.name == "SUSPEND START"):
- data.vprint("SUSPEND START %f %s:%d" % (t.time, sysvals.ftracefile, count))
- inthepipe = True
+ if(t.name == 'SUSPEND START'):
+ testrun[testidx].inthepipe = True
+ data.setStart(t.time)
continue
else:
- # look for the resume end marker
+ # trace event processing
if(t.fevent):
- if(t.name == "RESUME COMPLETE"):
- data.vprint("RESUME COMPLETE %f %s:%d" % (t.time, sysvals.ftracefile, count))
- inthepipe = False
- break
+ if(t.name == 'RESUME COMPLETE'):
+ testrun[testidx].inthepipe = False
+ data.setEnd(t.time)
+ if(testidx == testcnt - 1):
+ break
+ continue
+ # general trace events have two types, begin and end
+ if(re.match('(?P<name>.*) begin$', t.name)):
+ isbegin = True
+ elif(re.match('(?P<name>.*) end$', t.name)):
+ isbegin = False
+ else:
+ continue
+ m = re.match('(?P<name>.*)\[(?P<val>[0-9]*)\] .*', t.name)
+ if(m):
+ val = m.group('val')
+ if val == '0':
+ name = m.group('name')
+ else:
+ name = m.group('name')+'['+val+']'
+ else:
+ m = re.match('(?P<name>.*) .*', t.name)
+ name = m.group('name')
+ # special processing for trace events
+ if re.match('dpm_prepare\[.*', name):
+ continue
+ elif re.match('machine_suspend.*', name):
+ continue
+ elif re.match('suspend_enter\[.*', name):
+ if(not isbegin):
+ data.dmesg['suspend_prepare']['end'] = t.time
+ continue
+ elif re.match('dpm_suspend\[.*', name):
+ if(not isbegin):
+ data.dmesg['suspend']['end'] = t.time
+ continue
+ elif re.match('dpm_suspend_late\[.*', name):
+ if(isbegin):
+ data.dmesg['suspend_late']['start'] = t.time
+ else:
+ data.dmesg['suspend_late']['end'] = t.time
+ continue
+ elif re.match('dpm_suspend_noirq\[.*', name):
+ if(isbegin):
+ data.dmesg['suspend_noirq']['start'] = t.time
+ else:
+ data.dmesg['suspend_noirq']['end'] = t.time
+ continue
+ elif re.match('dpm_resume_noirq\[.*', name):
+ if(isbegin):
+ data.dmesg['resume_machine']['end'] = t.time
+ data.dmesg['resume_noirq']['start'] = t.time
+ else:
+ data.dmesg['resume_noirq']['end'] = t.time
+ continue
+ elif re.match('dpm_resume_early\[.*', name):
+ if(isbegin):
+ data.dmesg['resume_early']['start'] = t.time
+ else:
+ data.dmesg['resume_early']['end'] = t.time
+ continue
+ elif re.match('dpm_resume\[.*', name):
+ if(isbegin):
+ data.dmesg['resume']['start'] = t.time
+ else:
+ data.dmesg['resume']['end'] = t.time
+ continue
+ elif re.match('dpm_complete\[.*', name):
+ if(isbegin):
+ data.dmesg['resume_complete']['start'] = t.time
+ else:
+ data.dmesg['resume_complete']['end'] = t.time
+ continue
+ # is this trace event outside of the devices calls
+ if(data.isTraceEventOutsideDeviceCalls(pid, t.time)):
+ # global events (outside device calls) are simply graphed
+ if(isbegin):
+ # store each trace event in ttemp
+ if(name not in testrun[testidx].ttemp):
+ testrun[testidx].ttemp[name] = []
+ testrun[testidx].ttemp[name].append(\
+ {'begin': t.time, 'end': t.time})
+ else:
+ # finish off matching trace event in ttemp
+ if(name in testrun[testidx].ttemp):
+ testrun[testidx].ttemp[name][-1]['end'] = t.time
+ else:
+ if(isbegin):
+ data.addIntraDevTraceEvent('', name, pid, t.time)
+ else:
+ data.capIntraDevTraceEvent('', name, pid, t.time)
+ # call/return processing
+ elif sysvals.usecallgraph:
+ # create a callgraph object for the data
+ if(pid not in testrun[testidx].ftemp):
+ testrun[testidx].ftemp[pid] = []
+ testrun[testidx].ftemp[pid].append(FTraceCallGraph())
+ # when the call is finished, see which device matches it
+ cg = testrun[testidx].ftemp[pid][-1]
+ if(cg.addLine(t, m)):
+ testrun[testidx].ftemp[pid].append(FTraceCallGraph())
+ tf.close()
+
+ for test in testrun:
+ # add the traceevent data to the device hierarchy
+ if(sysvals.usetraceevents):
+ for name in test.ttemp:
+ for event in test.ttemp[name]:
+ begin = event['begin']
+ end = event['end']
+ # if event starts before timeline start, expand timeline
+ if(begin < test.data.start):
+ test.data.setStart(begin)
+ # if event ends after timeline end, expand the timeline
+ if(end > test.data.end):
+ test.data.setEnd(end)
+ test.data.newActionGlobal(name, begin, end)
+
+ # add the callgraph data to the device hierarchy
+ for pid in test.ftemp:
+ for cg in test.ftemp[pid]:
+ if(not cg.sanityCheck()):
+ id = 'task %s cpu %s' % (pid, m.group('cpu'))
+ vprint('Sanity check failed for '+\
+ id+', ignoring this callback')
+ continue
+ callstart = cg.start
+ callend = cg.end
+ for p in test.data.phases:
+ if(test.data.dmesg[p]['start'] <= callstart and
+ callstart <= test.data.dmesg[p]['end']):
+ list = test.data.dmesg[p]['list']
+ for devname in list:
+ dev = list[devname]
+ if(pid == dev['pid'] and
+ callstart <= dev['start'] and
+ callend >= dev['end']):
+ dev['ftrace'] = cg
+ break
+
+ if(sysvals.verbose):
+ test.data.printDetails()
+
+
+ # add the time in between the tests as a new phase so we can see it
+ if(len(testruns) > 1):
+ t1e = testruns[0].getEnd()
+ t2s = testruns[-1].getStart()
+ testruns[-1].newPhaseWithSingleAction('user mode', \
+ 'user mode', t1e, t2s, '#FF9966')
+
+# Function: parseTraceLog
+# Description:
+# Analyze an ftrace log output file generated from this app during
+# the execution phase. Used when the ftrace log is the primary data source
+# and includes the suspend_resume and device_pm_callback trace events
+# The ftrace filename is taken from sysvals
+# Output:
+# An array of Data objects
+def parseTraceLog():
+ global sysvals
+
+ vprint('Analyzing the ftrace data...')
+ if(os.path.exists(sysvals.ftracefile) == False):
+ doError('%s doesnt exist' % sysvals.ftracefile, False)
+
+ # extract the callgraph and traceevent data
+ testruns = []
+ testdata = []
+ testrun = 0
+ data = 0
+ tf = open(sysvals.ftracefile, 'r')
+ phase = 'suspend_prepare'
+ for line in tf:
+ # remove any latent carriage returns
+ line = line.replace('\r\n', '')
+ # stamp line: each stamp means a new test run
+ m = re.match(sysvals.stampfmt, line)
+ if(m):
+ data = Data(len(testdata))
+ testdata.append(data)
+ testrun = TestRun(data)
+ testruns.append(testrun)
+ parseStamp(m, data)
+ continue
+ if(not data):
+ continue
+ # firmware line: pull out any firmware data
+ m = re.match(sysvals.firmwarefmt, line)
+ if(m):
+ data.fwSuspend = int(m.group('s'))
+ data.fwResume = int(m.group('r'))
+ if(data.fwSuspend > 0 or data.fwResume > 0):
+ data.fwValid = True
+ continue
+ # tracer type line: determine the trace data type
+ m = re.match(sysvals.tracertypefmt, line)
+ if(m):
+ tracer = m.group('t')
+ testrun.setTracerType(tracer)
+ continue
+ # post resume time line: did this test run include post-resume data
+ m = re.match(sysvals.postresumefmt, line)
+ if(m):
+ t = int(m.group('t'))
+ if(t > 0):
+ sysvals.postresumetime = t
+ continue
+ # ftrace line: parse only valid lines
+ m = re.match(testrun.ftrace_line_fmt, line)
+ if(not m):
+ continue
+ # gather the basic message data from the line
+ m_time = m.group('time')
+ m_pid = m.group('pid')
+ m_msg = m.group('msg')
+ if(testrun.cgformat):
+ m_param3 = m.group('dur')
+ else:
+ m_param3 = 'traceevent'
+ if(m_time and m_pid and m_msg):
+ t = FTraceLine(m_time, m_msg, m_param3)
+ pid = int(m_pid)
+ else:
+ continue
+ # the line should be a call, return, or event
+ if(not t.fcall and not t.freturn and not t.fevent):
+ continue
+ # only parse the ftrace data during suspend/resume
+ if(not testrun.inthepipe):
+ # look for the suspend start marker
+ if(t.fevent):
+ if(t.name == 'SUSPEND START'):
+ testrun.inthepipe = True
+ data.setStart(t.time)
+ continue
+ # trace event processing
+ if(t.fevent):
+ if(t.name == 'RESUME COMPLETE'):
+ if(sysvals.postresumetime > 0):
+ phase = 'post_resume'
+ data.newPhase(phase, t.time, t.time, '#FF9966', -1)
+ else:
+ testrun.inthepipe = False
+ data.setEnd(t.time)
+ continue
+ if(phase == 'post_resume'):
+ data.setEnd(t.time)
+ if(t.type == 'suspend_resume'):
+ # suspend_resume trace events have two types, begin and end
+ if(re.match('(?P<name>.*) begin$', t.name)):
+ isbegin = True
+ elif(re.match('(?P<name>.*) end$', t.name)):
+ isbegin = False
+ else:
+ continue
+ m = re.match('(?P<name>.*)\[(?P<val>[0-9]*)\] .*', t.name)
+ if(m):
+ val = m.group('val')
+ if val == '0':
+ name = m.group('name')
+ else:
+ name = m.group('name')+'['+val+']'
+ else:
+ m = re.match('(?P<name>.*) .*', t.name)
+ name = m.group('name')
+ # ignore these events
+ if(re.match('acpi_suspend\[.*', t.name) or
+ re.match('suspend_enter\[.*', name)):
+ continue
+ # -- phase changes --
+ # suspend_prepare start
+ if(re.match('dpm_prepare\[.*', t.name)):
+ phase = 'suspend_prepare'
+ if(not isbegin):
+ data.dmesg[phase]['end'] = t.time
+ continue
+ # suspend start
+ elif(re.match('dpm_suspend\[.*', t.name)):
+ phase = 'suspend'
+ data.setPhase(phase, t.time, isbegin)
+ continue
+ # suspend_late start
+ elif(re.match('dpm_suspend_late\[.*', t.name)):
+ phase = 'suspend_late'
+ data.setPhase(phase, t.time, isbegin)
+ continue
+ # suspend_noirq start
+ elif(re.match('dpm_suspend_noirq\[.*', t.name)):
+ phase = 'suspend_noirq'
+ data.setPhase(phase, t.time, isbegin)
+ if(not isbegin):
+ phase = 'suspend_machine'
+ data.dmesg[phase]['start'] = t.time
+ continue
+ # suspend_machine/resume_machine
+ elif(re.match('machine_suspend\[.*', t.name)):
+ if(isbegin):
+ phase = 'suspend_machine'
+ data.dmesg[phase]['end'] = t.time
+ data.tSuspended = t.time
+ else:
+ if(sysvals.suspendmode in ['mem', 'disk']):
+ data.dmesg['suspend_machine']['end'] = t.time
+ data.tSuspended = t.time
+ phase = 'resume_machine'
+ data.dmesg[phase]['start'] = t.time
+ data.tResumed = t.time
+ data.tLow = data.tResumed - data.tSuspended
+ continue
+ # resume_noirq start
+ elif(re.match('dpm_resume_noirq\[.*', t.name)):
+ phase = 'resume_noirq'
+ data.setPhase(phase, t.time, isbegin)
+ if(isbegin):
+ data.dmesg['resume_machine']['end'] = t.time
+ continue
+ # resume_early start
+ elif(re.match('dpm_resume_early\[.*', t.name)):
+ phase = 'resume_early'
+ data.setPhase(phase, t.time, isbegin)
+ continue
+ # resume start
+ elif(re.match('dpm_resume\[.*', t.name)):
+ phase = 'resume'
+ data.setPhase(phase, t.time, isbegin)
+ continue
+ # resume complete start
+ elif(re.match('dpm_complete\[.*', t.name)):
+ phase = 'resume_complete'
+ if(isbegin):
+ data.dmesg[phase]['start'] = t.time
+ continue
+
+ # is this trace event outside of the devices calls
+ if(data.isTraceEventOutsideDeviceCalls(pid, t.time)):
+ # global events (outside device calls) are simply graphed
+ if(name not in testrun.ttemp):
+ testrun.ttemp[name] = []
+ if(isbegin):
+ # create a new list entry
+ testrun.ttemp[name].append(\
+ {'begin': t.time, 'end': t.time})
+ else:
+ if(len(testrun.ttemp[name]) > 0):
+ # if an antry exists, assume this is its end
+ testrun.ttemp[name][-1]['end'] = t.time
+ elif(phase == 'post_resume'):
+ # post resume events can just have ends
+ testrun.ttemp[name].append({
+ 'begin': data.dmesg[phase]['start'],
+ 'end': t.time})
+ else:
+ if(isbegin):
+ data.addIntraDevTraceEvent('', name, pid, t.time)
+ else:
+ data.capIntraDevTraceEvent('', name, pid, t.time)
+ # device callback start
+ elif(t.type == 'device_pm_callback_start'):
+ m = re.match('(?P<drv>.*) (?P<d>.*), parent: *(?P<p>.*), .*',\
+ t.name);
+ if(not m):
+ continue
+ drv = m.group('drv')
+ n = m.group('d')
+ p = m.group('p')
+ if(n and p):
+ data.newAction(phase, n, pid, p, t.time, -1, drv)
+ # device callback finish
+ elif(t.type == 'device_pm_callback_end'):
+ m = re.match('(?P<drv>.*) (?P<d>.*), err.*', t.name);
+ if(not m):
+ continue
+ n = m.group('d')
+ list = data.dmesg[phase]['list']
+ if(n in list):
+ dev = list[n]
+ dev['length'] = t.time - dev['start']
+ dev['end'] = t.time
+ # callgraph processing
+ elif sysvals.usecallgraph:
+ # this shouldn't happen, but JIC, ignore callgraph data post-res
+ if(phase == 'post_resume'):
continue
# create a callgraph object for the data
- if(pid not in ftemp):
- ftemp[pid] = FTraceCallGraph()
+ if(pid not in testrun.ftemp):
+ testrun.ftemp[pid] = []
+ testrun.ftemp[pid].append(FTraceCallGraph())
# when the call is finished, see which device matches it
- if(ftemp[pid].addLine(t, m)):
- if(not ftemp[pid].sanityCheck()):
- id = "task %s cpu %s" % (pid, m.group("cpu"))
- data.vprint("Sanity check failed for "+id+", ignoring this callback")
+ cg = testrun.ftemp[pid][-1]
+ if(cg.addLine(t, m)):
+ testrun.ftemp[pid].append(FTraceCallGraph())
+ tf.close()
+
+ for test in testruns:
+ # add the traceevent data to the device hierarchy
+ if(sysvals.usetraceevents):
+ for name in test.ttemp:
+ for event in test.ttemp[name]:
+ begin = event['begin']
+ end = event['end']
+ # if event starts before timeline start, expand timeline
+ if(begin < test.data.start):
+ test.data.setStart(begin)
+ # if event ends after timeline end, expand the timeline
+ if(end > test.data.end):
+ test.data.setEnd(end)
+ test.data.newActionGlobal(name, begin, end)
+
+ # add the callgraph data to the device hierarchy
+ borderphase = {
+ 'dpm_prepare': 'suspend_prepare',
+ 'dpm_complete': 'resume_complete'
+ }
+ for pid in test.ftemp:
+ for cg in test.ftemp[pid]:
+ if len(cg.list) < 2:
+ continue
+ if(not cg.sanityCheck()):
+ id = 'task %s cpu %s' % (pid, m.group('cpu'))
+ vprint('Sanity check failed for '+\
+ id+', ignoring this callback')
+ continue
+ callstart = cg.start
+ callend = cg.end
+ if(cg.list[0].name in borderphase):
+ p = borderphase[cg.list[0].name]
+ list = test.data.dmesg[p]['list']
+ for devname in list:
+ dev = list[devname]
+ if(pid == dev['pid'] and
+ callstart <= dev['start'] and
+ callend >= dev['end']):
+ dev['ftrace'] = cg.slice(dev['start'], dev['end'])
continue
- callstart = ftemp[pid].start
- callend = ftemp[pid].end
- for p in data.phases:
- if(data.dmesg[p]['start'] <= callstart and callstart <= data.dmesg[p]['end']):
- list = data.dmesg[p]['list']
+ if(cg.list[0].name != 'dpm_run_callback'):
+ continue
+ for p in test.data.phases:
+ if(test.data.dmesg[p]['start'] <= callstart and
+ callstart <= test.data.dmesg[p]['end']):
+ list = test.data.dmesg[p]['list']
for devname in list:
dev = list[devname]
- if(pid == dev['pid'] and callstart <= dev['start'] and callend >= dev['end']):
- data.vprint("%15s [%f - %f] %s(%d)" % (p, callstart, callend, devname, pid))
- dev['ftrace'] = ftemp[pid]
+ if(pid == dev['pid'] and
+ callstart <= dev['start'] and
+ callend >= dev['end']):
+ dev['ftrace'] = cg
break
- ftemp[pid] = FTraceCallGraph()
- tf.close()
-# Function: sortKernelLog
+ # fill in any missing phases
+ for data in testdata:
+ lp = data.phases[0]
+ for p in data.phases:
+ if(data.dmesg[p]['start'] < 0 and data.dmesg[p]['end'] < 0):
+ print('WARNING: phase "%s" is missing!' % p)
+ if(data.dmesg[p]['start'] < 0):
+ data.dmesg[p]['start'] = data.dmesg[lp]['end']
+ if(p == 'resume_machine'):
+ data.tSuspended = data.dmesg[lp]['end']
+ data.tResumed = data.dmesg[lp]['end']
+ data.tLow = 0
+ if(data.dmesg[p]['end'] < 0):
+ data.dmesg[p]['end'] = data.dmesg[p]['start']
+ lp = p
+
+ if(len(sysvals.devicefilter) > 0):
+ data.deviceFilter(sysvals.devicefilter)
+ data.fixupInitcallsThatDidntReturn()
+ if(sysvals.verbose):
+ data.printDetails()
+
+ # add the time in between the tests as a new phase so we can see it
+ if(len(testdata) > 1):
+ t1e = testdata[0].getEnd()
+ t2s = testdata[-1].getStart()
+ testdata[-1].newPhaseWithSingleAction('user mode', \
+ 'user mode', t1e, t2s, '#FF9966')
+ return testdata
+
+# Function: loadKernelLog
# Description:
-# The dmesg output log sometimes comes with with lines that have
-# timestamps out of order. This could cause issues since a call
-# could accidentally end up in the wrong phase
-def sortKernelLog():
- global sysvals, data
+# [deprecated for kernel 3.15.0 or newer]
+# load the dmesg file into memory and fix up any ordering issues
+# The dmesg filename is taken from sysvals
+# Output:
+# An array of empty Data objects with only their dmesgtext attributes set
+def loadKernelLog():
+ global sysvals
+
+ vprint('Analyzing the dmesg data...')
+ if(os.path.exists(sysvals.dmesgfile) == False):
+ doError('%s doesnt exist' % sysvals.dmesgfile, False)
+
+ # there can be multiple test runs in a single file delineated by stamps
+ testruns = []
+ data = 0
lf = open(sysvals.dmesgfile, 'r')
- dmesglist = []
- count = 0
for line in lf:
- line = line.replace("\r\n", "")
- if(count == 0):
- parseStamp(line)
- elif(count == 1):
- m = re.match(r"# fwsuspend (?P<s>[0-9]*) fwresume (?P<r>[0-9]*)$", line)
- if(m):
- data.fwSuspend = int(m.group("s"))
- data.fwResume = int(m.group("r"))
- if(data.fwSuspend > 0 or data.fwResume > 0):
- data.fwValid = True
- if(re.match(r".*(\[ *)(?P<ktime>[0-9\.]*)(\]) (?P<msg>.*)", line)):
- dmesglist.append(line)
- count += 1
+ line = line.replace('\r\n', '')
+ idx = line.find('[')
+ if idx > 1:
+ line = line[idx:]
+ m = re.match(sysvals.stampfmt, line)
+ if(m):
+ if(data):
+ testruns.append(data)
+ data = Data(len(testruns))
+ parseStamp(m, data)
+ continue
+ if(not data):
+ continue
+ m = re.match(sysvals.firmwarefmt, line)
+ if(m):
+ data.fwSuspend = int(m.group('s'))
+ data.fwResume = int(m.group('r'))
+ if(data.fwSuspend > 0 or data.fwResume > 0):
+ data.fwValid = True
+ continue
+ m = re.match('[ \t]*(\[ *)(?P<ktime>[0-9\.]*)(\]) (?P<msg>.*)', line)
+ if(m):
+ data.dmesgtext.append(line)
+ if(re.match('ACPI: resume from mwait', m.group('msg'))):
+ print('NOTE: This suspend appears to be freeze rather than'+\
+ ' %s, it will be treated as such' % sysvals.suspendmode)
+ sysvals.suspendmode = 'freeze'
+ else:
+ vprint('ignoring dmesg line: %s' % line.replace('\n', ''))
+ testruns.append(data)
lf.close()
- last = ""
-
- # fix lines with the same time stamp and function with the call and return swapped
- for line in dmesglist:
- mc = re.match(r".*(\[ *)(?P<t>[0-9\.]*)(\]) calling (?P<f>.*)\+ @ .*, parent: .*", line)
- mr = re.match(r".*(\[ *)(?P<t>[0-9\.]*)(\]) call (?P<f>.*)\+ returned .* after (?P<dt>.*) usecs", last)
- if(mc and mr and (mc.group("t") == mr.group("t")) and (mc.group("f") == mr.group("f"))):
- i = dmesglist.index(last)
- j = dmesglist.index(line)
- dmesglist[i] = line
- dmesglist[j] = last
- last = line
- return dmesglist
-
-# Function: analyzeKernelLog
+
+ if(not data):
+ print('ERROR: analyze_suspend header missing from dmesg log')
+ sys.exit()
+
+ # fix lines with same timestamp/function with the call and return swapped
+ for data in testruns:
+ last = ''
+ for line in data.dmesgtext:
+ mc = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) calling '+\
+ '(?P<f>.*)\+ @ .*, parent: .*', line)
+ mr = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) call '+\
+ '(?P<f>.*)\+ returned .* after (?P<dt>.*) usecs', last)
+ if(mc and mr and (mc.group('t') == mr.group('t')) and
+ (mc.group('f') == mr.group('f'))):
+ i = data.dmesgtext.index(last)
+ j = data.dmesgtext.index(line)
+ data.dmesgtext[i] = line
+ data.dmesgtext[j] = last
+ last = line
+ return testruns
+
+# Function: parseKernelLog
# Description:
+# [deprecated for kernel 3.15.0 or newer]
# Analyse a dmesg log output file generated from this app during
# the execution phase. Create a set of device structures in memory
# for subsequent formatting in the html output file
-def analyzeKernelLog():
- global sysvals, data
+# This call is only for legacy support on kernels where the ftrace
+# data lacks the suspend_resume or device_pm_callbacks trace events.
+# Arguments:
+# data: an empty Data object (with dmesgtext) obtained from loadKernelLog
+# Output:
+# The filled Data object
+def parseKernelLog(data):
+ global sysvals
- print("PROCESSING DATA")
- data.vprint("Analyzing the dmesg data...")
- if(os.path.exists(sysvals.dmesgfile) == False):
- print("ERROR: %s doesn't exist") % sysvals.dmesgfile
- return False
+ phase = 'suspend_runtime'
- lf = sortKernelLog()
- phase = "suspend_runtime"
+ if(data.fwValid):
+ vprint('Firmware Suspend = %u ns, Firmware Resume = %u ns' % \
+ (data.fwSuspend, data.fwResume))
+ # dmesg phase match table
dm = {
- 'suspend_general': r"PM: Syncing filesystems.*",
- 'suspend_early': r"PM: suspend of devices complete after.*",
- 'suspend_noirq': r"PM: late suspend of devices complete after.*",
- 'suspend_cpu': r"PM: noirq suspend of devices complete after.*",
- 'resume_cpu': r"ACPI: Low-level resume complete.*",
- 'resume_noirq': r"ACPI: Waking up from system sleep state.*",
- 'resume_early': r"PM: noirq resume of devices complete after.*",
- 'resume_general': r"PM: early resume of devices complete after.*",
- 'resume_complete': r".*Restarting tasks \.\.\..*",
+ 'suspend_prepare': 'PM: Syncing filesystems.*',
+ 'suspend': 'PM: Entering [a-z]* sleep.*',
+ 'suspend_late': 'PM: suspend of devices complete after.*',
+ 'suspend_noirq': 'PM: late suspend of devices complete after.*',
+ 'suspend_machine': 'PM: noirq suspend of devices complete after.*',
+ 'resume_machine': 'ACPI: Low-level resume complete.*',
+ 'resume_noirq': 'ACPI: Waking up from system sleep state.*',
+ 'resume_early': 'PM: noirq resume of devices complete after.*',
+ 'resume': 'PM: early resume of devices complete after.*',
+ 'resume_complete': 'PM: resume of devices complete after.*',
+ 'post_resume': '.*Restarting tasks \.\.\..*',
}
- if(sysvals.suspendmode == "standby"):
- dm['resume_cpu'] = r"PM: Restoring platform NVS memory"
- elif(sysvals.suspendmode == "disk"):
- dm['suspend_early'] = r"PM: freeze of devices complete after.*"
- dm['suspend_noirq'] = r"PM: late freeze of devices complete after.*"
- dm['suspend_cpu'] = r"PM: noirq freeze of devices complete after.*"
- dm['resume_cpu'] = r"PM: Restoring platform NVS memory"
- dm['resume_early'] = r"PM: noirq restore of devices complete after.*"
- dm['resume_general'] = r"PM: early restore of devices complete after.*"
-
- action_start = 0.0
- for line in lf:
+ if(sysvals.suspendmode == 'standby'):
+ dm['resume_machine'] = 'PM: Restoring platform NVS memory'
+ elif(sysvals.suspendmode == 'disk'):
+ dm['suspend_late'] = 'PM: freeze of devices complete after.*'
+ dm['suspend_noirq'] = 'PM: late freeze of devices complete after.*'
+ dm['suspend_machine'] = 'PM: noirq freeze of devices complete after.*'
+ dm['resume_machine'] = 'PM: Restoring platform NVS memory'
+ dm['resume_early'] = 'PM: noirq restore of devices complete after.*'
+ dm['resume'] = 'PM: early restore of devices complete after.*'
+ dm['resume_complete'] = 'PM: restore of devices complete after.*'
+ elif(sysvals.suspendmode == 'freeze'):
+ dm['resume_machine'] = 'ACPI: resume from mwait'
+
+ # action table (expected events that occur and show up in dmesg)
+ at = {
+ 'sync_filesystems': {
+ 'smsg': 'PM: Syncing filesystems.*',
+ 'emsg': 'PM: Preparing system for mem sleep.*' },
+ 'freeze_user_processes': {
+ 'smsg': 'Freezing user space processes .*',
+ 'emsg': 'Freezing remaining freezable tasks.*' },
+ 'freeze_tasks': {
+ 'smsg': 'Freezing remaining freezable tasks.*',
+ 'emsg': 'PM: Entering (?P<mode>[a-z,A-Z]*) sleep.*' },
+ 'ACPI prepare': {
+ 'smsg': 'ACPI: Preparing to enter system sleep state.*',
+ 'emsg': 'PM: Saving platform NVS memory.*' },
+ 'PM vns': {
+ 'smsg': 'PM: Saving platform NVS memory.*',
+ 'emsg': 'Disabling non-boot CPUs .*' },
+ }
+
+ t0 = -1.0
+ cpu_start = -1.0
+ prevktime = -1.0
+ actions = dict()
+ for line in data.dmesgtext:
# -- preprocessing --
# parse each dmesg line into the time and message
- m = re.match(r".*(\[ *)(?P<ktime>[0-9\.]*)(\]) (?P<msg>.*)", line)
+ m = re.match('[ \t]*(\[ *)(?P<ktime>[0-9\.]*)(\]) (?P<msg>.*)', line)
if(m):
- ktime = float(m.group("ktime"))
- msg = m.group("msg")
+ val = m.group('ktime')
+ try:
+ ktime = float(val)
+ except:
+ doWarning('INVALID DMESG LINE: '+\
+ line.replace('\n', ''), 'dmesg')
+ continue
+ msg = m.group('msg')
+ # initialize data start to first line time
+ if t0 < 0:
+ data.setStart(ktime)
+ t0 = ktime
else:
- print line
continue
+ # hack for determining resume_machine end for freeze
+ if(not sysvals.usetraceevents and sysvals.suspendmode == 'freeze' \
+ and phase == 'resume_machine' and \
+ re.match('calling (?P<f>.*)\+ @ .*, parent: .*', msg)):
+ data.dmesg['resume_machine']['end'] = ktime
+ phase = 'resume_noirq'
+ data.dmesg[phase]['start'] = ktime
+
# -- phase changes --
- # suspend_general start
- if(re.match(dm['suspend_general'], msg)):
- phase = "suspend_general"
+ # suspend start
+ if(re.match(dm['suspend_prepare'], msg)):
+ phase = 'suspend_prepare'
+ data.dmesg[phase]['start'] = ktime
+ data.setStart(ktime)
+ # suspend start
+ elif(re.match(dm['suspend'], msg)):
+ data.dmesg['suspend_prepare']['end'] = ktime
+ phase = 'suspend'
data.dmesg[phase]['start'] = ktime
- data.start = ktime
- # action start: syncing filesystems
- action_start = ktime
- # suspend_early start
- elif(re.match(dm['suspend_early'], msg)):
- data.dmesg["suspend_general"]['end'] = ktime
- phase = "suspend_early"
+ # suspend_late start
+ elif(re.match(dm['suspend_late'], msg)):
+ data.dmesg['suspend']['end'] = ktime
+ phase = 'suspend_late'
data.dmesg[phase]['start'] = ktime
# suspend_noirq start
elif(re.match(dm['suspend_noirq'], msg)):
- data.dmesg["suspend_early"]['end'] = ktime
- phase = "suspend_noirq"
+ data.dmesg['suspend_late']['end'] = ktime
+ phase = 'suspend_noirq'
data.dmesg[phase]['start'] = ktime
- # suspend_cpu start
- elif(re.match(dm['suspend_cpu'], msg)):
- data.dmesg["suspend_noirq"]['end'] = ktime
- phase = "suspend_cpu"
+ # suspend_machine start
+ elif(re.match(dm['suspend_machine'], msg)):
+ data.dmesg['suspend_noirq']['end'] = ktime
+ phase = 'suspend_machine'
data.dmesg[phase]['start'] = ktime
- # resume_cpu start
- elif(re.match(dm['resume_cpu'], msg)):
- data.tSuspended = ktime
- data.dmesg["suspend_cpu"]['end'] = ktime
- phase = "resume_cpu"
+ # resume_machine start
+ elif(re.match(dm['resume_machine'], msg)):
+ if(sysvals.suspendmode in ['freeze', 'standby']):
+ data.tSuspended = prevktime
+ data.dmesg['suspend_machine']['end'] = prevktime
+ else:
+ data.tSuspended = ktime
+ data.dmesg['suspend_machine']['end'] = ktime
+ phase = 'resume_machine'
+ data.tResumed = ktime
+ data.tLow = data.tResumed - data.tSuspended
data.dmesg[phase]['start'] = ktime
# resume_noirq start
elif(re.match(dm['resume_noirq'], msg)):
- data.dmesg["resume_cpu"]['end'] = ktime
- phase = "resume_noirq"
+ data.dmesg['resume_machine']['end'] = ktime
+ phase = 'resume_noirq'
data.dmesg[phase]['start'] = ktime
- # action end: ACPI resume
- data.newAction("resume_cpu", "ACPI", -1, "", action_start, ktime)
# resume_early start
elif(re.match(dm['resume_early'], msg)):
- data.dmesg["resume_noirq"]['end'] = ktime
- phase = "resume_early"
+ data.dmesg['resume_noirq']['end'] = ktime
+ phase = 'resume_early'
data.dmesg[phase]['start'] = ktime
- # resume_general start
- elif(re.match(dm['resume_general'], msg)):
- data.dmesg["resume_early"]['end'] = ktime
- phase = "resume_general"
+ # resume start
+ elif(re.match(dm['resume'], msg)):
+ data.dmesg['resume_early']['end'] = ktime
+ phase = 'resume'
data.dmesg[phase]['start'] = ktime
# resume complete start
elif(re.match(dm['resume_complete'], msg)):
- data.dmesg["resume_general"]['end'] = ktime
- data.end = ktime
- phase = "resume_runtime"
+ data.dmesg['resume']['end'] = ktime
+ phase = 'resume_complete'
+ data.dmesg[phase]['start'] = ktime
+ # post resume start
+ elif(re.match(dm['post_resume'], msg)):
+ data.dmesg['resume_complete']['end'] = ktime
+ data.setEnd(ktime)
+ phase = 'post_resume'
break
# -- device callbacks --
if(phase in data.phases):
# device init call
- if(re.match(r"calling (?P<f>.*)\+ @ .*, parent: .*", msg)):
- sm = re.match(r"calling (?P<f>.*)\+ @ (?P<n>.*), parent: (?P<p>.*)", msg);
- f = sm.group("f")
- n = sm.group("n")
- p = sm.group("p")
+ if(re.match('calling (?P<f>.*)\+ @ .*, parent: .*', msg)):
+ sm = re.match('calling (?P<f>.*)\+ @ '+\
+ '(?P<n>.*), parent: (?P<p>.*)', msg);
+ f = sm.group('f')
+ n = sm.group('n')
+ p = sm.group('p')
if(f and n and p):
- data.newAction(phase, f, int(n), p, ktime, -1)
+ data.newAction(phase, f, int(n), p, ktime, -1, '')
# device init return
- elif(re.match(r"call (?P<f>.*)\+ returned .* after (?P<t>.*) usecs", msg)):
- sm = re.match(r"call (?P<f>.*)\+ returned .* after (?P<t>.*) usecs(?P<a>.*)", msg);
- f = sm.group("f")
- t = sm.group("t")
+ elif(re.match('call (?P<f>.*)\+ returned .* after '+\
+ '(?P<t>.*) usecs', msg)):
+ sm = re.match('call (?P<f>.*)\+ returned .* after '+\
+ '(?P<t>.*) usecs(?P<a>.*)', msg);
+ f = sm.group('f')
+ t = sm.group('t')
list = data.dmesg[phase]['list']
if(f in list):
dev = list[f]
dev['length'] = int(t)
dev['end'] = ktime
- data.vprint("%15s [%f - %f] %s(%d) %s" %
- (phase, dev['start'], dev['end'], f, dev['pid'], dev['par']))
-
- # -- phase specific actions --
- if(phase == "suspend_general"):
- if(re.match(r"PM: Preparing system for mem sleep.*", msg)):
- data.newAction(phase, "filesystem-sync", -1, "", action_start, ktime)
- elif(re.match(r"Freezing user space processes .*", msg)):
- action_start = ktime
- elif(re.match(r"Freezing remaining freezable tasks.*", msg)):
- data.newAction(phase, "freeze-user-processes", -1, "", action_start, ktime)
- action_start = ktime
- elif(re.match(r"PM: Entering (?P<mode>[a-z,A-Z]*) sleep.*", msg)):
- data.newAction(phase, "freeze-tasks", -1, "", action_start, ktime)
- elif(phase == "suspend_cpu"):
- m = re.match(r"smpboot: CPU (?P<cpu>[0-9]*) is now offline", msg)
- if(m):
- cpu = "CPU"+m.group("cpu")
- data.newAction(phase, cpu, -1, "", action_start, ktime)
- action_start = ktime
- elif(re.match(r"ACPI: Preparing to enter system sleep state.*", msg)):
- action_start = ktime
- elif(re.match(r"Disabling non-boot CPUs .*", msg)):
- data.newAction(phase, "ACPI", -1, "", action_start, ktime)
- action_start = ktime
- elif(phase == "resume_cpu"):
- m = re.match(r"CPU(?P<cpu>[0-9]*) is up", msg)
- if(m):
- cpu = "CPU"+m.group("cpu")
- data.newAction(phase, cpu, -1, "", action_start, ktime)
- action_start = ktime
- elif(re.match(r"Enabling non-boot CPUs .*", msg)):
- action_start = ktime
+
+ # -- non-devicecallback actions --
+ # if trace events are not available, these are better than nothing
+ if(not sysvals.usetraceevents):
+ # look for known actions
+ for a in at:
+ if(re.match(at[a]['smsg'], msg)):
+ if(a not in actions):
+ actions[a] = []
+ actions[a].append({'begin': ktime, 'end': ktime})
+ if(re.match(at[a]['emsg'], msg)):
+ actions[a][-1]['end'] = ktime
+ # now look for CPU on/off events
+ if(re.match('Disabling non-boot CPUs .*', msg)):
+ # start of first cpu suspend
+ cpu_start = ktime
+ elif(re.match('Enabling non-boot CPUs .*', msg)):
+ # start of first cpu resume
+ cpu_start = ktime
+ elif(re.match('smpboot: CPU (?P<cpu>[0-9]*) is now offline', msg)):
+ # end of a cpu suspend, start of the next
+ m = re.match('smpboot: CPU (?P<cpu>[0-9]*) is now offline', msg)
+ cpu = 'CPU'+m.group('cpu')
+ if(cpu not in actions):
+ actions[cpu] = []
+ actions[cpu].append({'begin': cpu_start, 'end': ktime})
+ cpu_start = ktime
+ elif(re.match('CPU(?P<cpu>[0-9]*) is up', msg)):
+ # end of a cpu resume, start of the next
+ m = re.match('CPU(?P<cpu>[0-9]*) is up', msg)
+ cpu = 'CPU'+m.group('cpu')
+ if(cpu not in actions):
+ actions[cpu] = []
+ actions[cpu].append({'begin': cpu_start, 'end': ktime})
+ cpu_start = ktime
+ prevktime = ktime
# fill in any missing phases
- lp = "suspend_general"
+ lp = data.phases[0]
for p in data.phases:
- if(p == "suspend_general"):
- continue
+ if(data.dmesg[p]['start'] < 0 and data.dmesg[p]['end'] < 0):
+ print('WARNING: phase "%s" is missing, something went wrong!' % p)
+ print(' In %s, this dmesg line denotes the start of %s:' % \
+ (sysvals.suspendmode, p))
+ print(' "%s"' % dm[p])
if(data.dmesg[p]['start'] < 0):
data.dmesg[p]['start'] = data.dmesg[lp]['end']
- if(p == "resume_cpu"):
+ if(p == 'resume_machine'):
data.tSuspended = data.dmesg[lp]['end']
+ data.tResumed = data.dmesg[lp]['end']
+ data.tLow = 0
if(data.dmesg[p]['end'] < 0):
data.dmesg[p]['end'] = data.dmesg[p]['start']
lp = p
+ # fill in any actions we've found
+ for name in actions:
+ for event in actions[name]:
+ begin = event['begin']
+ end = event['end']
+ # if event starts before timeline start, expand timeline
+ if(begin < data.start):
+ data.setStart(begin)
+ # if event ends after timeline end, expand the timeline
+ if(end > data.end):
+ data.setEnd(end)
+ data.newActionGlobal(name, begin, end)
+
+ if(sysvals.verbose):
+ data.printDetails()
+ if(len(sysvals.devicefilter) > 0):
+ data.deviceFilter(sysvals.devicefilter)
data.fixupInitcallsThatDidntReturn()
return True
# Function: setTimelineRows
# Description:
-# Organize the device or thread lists into the smallest
+# Organize the timeline entries into the smallest
# number of rows possible, with no entry overlapping
# Arguments:
-# list: the list to sort (dmesg or ftrace)
-# sortedkeys: sorted key list to use
+# list: the list of devices/actions for a single phase
+# sortedkeys: cronologically sorted key list to use
+# Output:
+# The total number of rows needed to display this phase of the timeline
def setTimelineRows(list, sortedkeys):
- global data
# clear all rows and set them to undefined
remaining = len(list)
for ritem in rowdata[row]:
rs = ritem['start']
re = ritem['end']
- if(not (((s <= rs) and (e <= rs)) or ((s >= re) and (e >= re)))):
+ if(not (((s <= rs) and (e <= rs)) or
+ ((s >= re) and (e >= re)))):
valid = False
break
if(valid):
# Function: createTimeScale
# Description:
-# Create timescale lines for the dmesg and ftrace timelines
+# Create the timescale header for the html timeline
# Arguments:
# t0: start time (suspend begin)
# tMax: end time (resume end)
-# tSuspend: time when suspend occurs
+# tSuspend: time when suspend occurs, i.e. the zero time
+# Output:
+# The html code needed to display the time scale
def createTimeScale(t0, tMax, tSuspended):
- global data
- timescale = "<div class=\"t\" style=\"right:{0}%\">{1}</div>\n"
+ timescale = '<div class="t" style="right:{0}%">{1}</div>\n'
output = '<div id="timescale">\n'
# set scale for timeline
tS = 1
if(tSuspended < 0):
for i in range(int(tTotal/tS)+1):
- pos = "%0.3f" % (100 - ((float(i)*tS*100)/tTotal))
+ pos = '%0.3f' % (100 - ((float(i)*tS*100)/tTotal))
if(i > 0):
- val = "%0.f" % (float(i)*tS*1000)
+ val = '%0.fms' % (float(i)*tS*1000)
else:
- val = ""
+ val = ''
output += timescale.format(pos, val)
else:
tSuspend = tSuspended - t0
divSuspend = int(tSuspend/tS)
s0 = (tSuspend - tS*divSuspend)*100/tTotal
for i in range(divTotal):
- pos = "%0.3f" % (100 - ((float(i)*tS*100)/tTotal) - s0)
+ pos = '%0.3f' % (100 - ((float(i)*tS*100)/tTotal) - s0)
if((i == 0) and (s0 < 3)):
- val = ""
+ val = ''
elif(i == divSuspend):
- val = "S/R"
+ val = 'S/R'
else:
- val = "%0.f" % (float(i-divSuspend)*tS*1000)
+ val = '%0.fms' % (float(i-divSuspend)*tS*1000)
output += timescale.format(pos, val)
output += '</div>\n'
return output
+# Function: createHTMLSummarySimple
+# Description:
+# Create summary html file for a series of tests
+# Arguments:
+# testruns: array of Data objects from parseTraceLog
+def createHTMLSummarySimple(testruns, htmlfile):
+ global sysvals
+
+ # print out the basic summary of all the tests
+ hf = open(htmlfile, 'w')
+
+ # write the html header first (html head, css code, up to body start)
+ html = '<!DOCTYPE html>\n<html>\n<head>\n\
+ <meta http-equiv="content-type" content="text/html; charset=UTF-8">\n\
+ <title>AnalyzeSuspend Summary</title>\n\
+ <style type=\'text/css\'>\n\
+ body {overflow-y: scroll;}\n\
+ .stamp {width: 100%;text-align:center;background-color:#495E09;line-height:30px;color:white;font: 25px Arial;}\n\
+ table {width:100%;border-collapse: collapse;}\n\
+ .summary {font: 22px Arial;border:1px solid;}\n\
+ th {border: 1px solid black;background-color:#A7C942;color:white;}\n\
+ td {text-align: center;}\n\
+ tr.alt td {background-color:#EAF2D3;}\n\
+ tr.avg td {background-color:#BDE34C;}\n\
+ a:link {color: #90B521;}\n\
+ a:visited {color: #495E09;}\n\
+ a:hover {color: #B1DF28;}\n\
+ a:active {color: #FFFFFF;}\n\
+ </style>\n</head>\n<body>\n'
+
+ # group test header
+ count = len(testruns)
+ headline_stamp = '<div class="stamp">{0} {1} {2} {3} ({4} tests)</div>\n'
+ html += headline_stamp.format(sysvals.stamp['host'],
+ sysvals.stamp['kernel'], sysvals.stamp['mode'],
+ sysvals.stamp['time'], count)
+
+ # check to see if all the tests have the same value
+ stampcolumns = False
+ for data in testruns:
+ if diffStamp(sysvals.stamp, data.stamp):
+ stampcolumns = True
+ break
+
+ th = '\t<th>{0}</th>\n'
+ td = '\t<td>{0}</td>\n'
+ tdlink = '\t<td><a href="{0}">Click Here</a></td>\n'
+
+ # table header
+ html += '<table class="summary">\n<tr>\n'
+ html += th.format("Test #")
+ if stampcolumns:
+ html += th.format("Hostname")
+ html += th.format("Kernel Version")
+ html += th.format("Suspend Mode")
+ html += th.format("Test Time")
+ html += th.format("Suspend Time")
+ html += th.format("Resume Time")
+ html += th.format("Detail")
+ html += '</tr>\n'
+
+ # test data, 1 row per test
+ sTimeAvg = 0.0
+ rTimeAvg = 0.0
+ num = 1
+ for data in testruns:
+ # data.end is the end of post_resume
+ resumeEnd = data.dmesg['resume_complete']['end']
+ if num % 2 == 1:
+ html += '<tr class="alt">\n'
+ else:
+ html += '<tr>\n'
+
+ # test num
+ html += td.format("test %d" % num)
+ num += 1
+ if stampcolumns:
+ # host name
+ val = "unknown"
+ if('host' in data.stamp):
+ val = data.stamp['host']
+ html += td.format(val)
+ # host kernel
+ val = "unknown"
+ if('kernel' in data.stamp):
+ val = data.stamp['kernel']
+ html += td.format(val)
+ # suspend mode
+ val = "unknown"
+ if('mode' in data.stamp):
+ val = data.stamp['mode']
+ html += td.format(val)
+ # test time
+ val = "unknown"
+ if('time' in data.stamp):
+ val = data.stamp['time']
+ html += td.format(val)
+ # suspend time
+ sTime = (data.tSuspended - data.start)*1000
+ sTimeAvg += sTime
+ html += td.format("%3.3f ms" % sTime)
+ # resume time
+ rTime = (resumeEnd - data.tResumed)*1000
+ rTimeAvg += rTime
+ html += td.format("%3.3f ms" % rTime)
+ # link to the output html
+ html += tdlink.format(data.outfile)
+
+ html += '</tr>\n'
+
+ # last line: test average
+ if(count > 0):
+ sTimeAvg /= count
+ rTimeAvg /= count
+ html += '<tr class="avg">\n'
+ html += td.format('Average') # name
+ if stampcolumns:
+ html += td.format('') # host
+ html += td.format('') # kernel
+ html += td.format('') # mode
+ html += td.format('') # time
+ html += td.format("%3.3f ms" % sTimeAvg) # suspend time
+ html += td.format("%3.3f ms" % rTimeAvg) # resume time
+ html += td.format('') # output link
+ html += '</tr>\n'
+
+ # flush the data to file
+ hf.write(html+'</table>\n')
+ hf.write('</body>\n</html>\n')
+ hf.close()
+
# Function: createHTML
# Description:
-# Create the output html file.
-def createHTML():
- global sysvals, data
+# Create the output html file from the resident test data
+# Arguments:
+# testruns: array of Data objects from parseKernelLog or parseTraceLog
+# Output:
+# True if the html file was created, false if it failed
+def createHTML(testruns):
+ global sysvals
- data.normalizeTime()
+ for data in testruns:
+ data.normalizeTime(testruns[-1].tSuspended)
+ x2changes = ['', 'absolute']
+ if len(testruns) > 1:
+ x2changes = ['1', 'relative']
# html function templates
headline_stamp = '<div class="stamp">{0} {1} {2} {3}</div>\n'
- html_zoombox = '<center><button id="zoomin">ZOOM IN</button><button id="zoomout">ZOOM OUT</button><button id="zoomdef">ZOOM 1:1</button></center>\n<div id="dmesgzoombox" class="zoombox">\n'
- html_timeline = '<div id="{0}" class="timeline" style="height:{1}px">\n'
+ html_devlist1 = '<button id="devlist1" class="devlist" style="float:left;">Device Detail%s</button>' % x2changes[0]
+ html_zoombox = '<center><button id="zoomin">ZOOM IN</button><button id="zoomout">ZOOM OUT</button><button id="zoomdef">ZOOM 1:1</button></center>\n'
+ html_devlist2 = '<button id="devlist2" class="devlist" style="float:right;">Device Detail2</button>\n'
+ html_timeline = '<div id="dmesgzoombox" class="zoombox">\n<div id="{0}" class="timeline" style="height:{1}px">\n'
html_device = '<div id="{0}" title="{1}" class="thread" style="left:{2}%;top:{3}%;height:{4}%;width:{5}%;">{6}</div>\n'
+ html_traceevent = '<div title="{0}" class="traceevent" style="left:{1}%;top:{2}%;height:{3}%;width:{4}%;border:1px solid {5};background-color:{5}">{6}</div>\n'
html_phase = '<div class="phase" style="left:{0}%;width:{1}%;top:{2}%;height:{3}%;background-color:{4}">{5}</div>\n'
+ html_phaselet = '<div id="{0}" class="phaselet" style="left:{1}%;width:{2}%;background-color:{3}"></div>\n'
html_legend = '<div class="square" style="left:{0}%;background-color:{1}"> {2}</div>\n'
html_timetotal = '<table class="time1">\n<tr>'\
- '<td class="gray">{2} Suspend Time: <b>{0} ms</b></td>'\
- '<td class="gray">{2} Resume Time: <b>{1} ms</b></td>'\
+ '<td class="green">{2} Suspend Time: <b>{0} ms</b></td>'\
+ '<td class="yellow">{2} Resume Time: <b>{1} ms</b></td>'\
+ '</tr>\n</table>\n'
+ html_timetotal2 = '<table class="time1">\n<tr>'\
+ '<td class="green">{3} Suspend Time: <b>{0} ms</b></td>'\
+ '<td class="gray">'+sysvals.suspendmode+' time: <b>{1} ms</b></td>'\
+ '<td class="yellow">{3} Resume Time: <b>{2} ms</b></td>'\
'</tr>\n</table>\n'
html_timegroups = '<table class="time2">\n<tr>'\
- '<td class="green">Kernel Suspend: {0} ms</td>'\
- '<td class="purple">Firmware Suspend: {1} ms</td>'\
- '<td class="purple">Firmware Resume: {2} ms</td>'\
- '<td class="yellow">Kernel Resume: {3} ms</td>'\
+ '<td class="green">{4}Kernel Suspend: {0} ms</td>'\
+ '<td class="purple">{4}Firmware Suspend: {1} ms</td>'\
+ '<td class="purple">{4}Firmware Resume: {2} ms</td>'\
+ '<td class="yellow">{4}Kernel Resume: {3} ms</td>'\
'</tr>\n</table>\n'
- # device timeline (dmesg)
- if(data.usedmesg):
- data.vprint("Creating Device Timeline...")
- devtl = Timeline()
+ # device timeline
+ vprint('Creating Device Timeline...')
+ devtl = Timeline()
- # Generate the header for this timeline
- t0 = data.start
- tMax = data.end
- tTotal = tMax - t0
+ # Generate the header for this timeline
+ textnum = ['First', 'Second']
+ for data in testruns:
+ tTotal = data.end - data.start
+ tEnd = data.dmesg['resume_complete']['end']
if(tTotal == 0):
- print("ERROR: No timeline data")
+ print('ERROR: No timeline data')
sys.exit()
- suspend_time = "%.0f"%(-data.start*1000)
- resume_time = "%.0f"%(data.end*1000)
+ if(data.tLow > 0):
+ low_time = '%.0f'%(data.tLow*1000)
if data.fwValid:
- devtl.html['timeline'] = html_timetotal.format(suspend_time, resume_time, "Total")
- sktime = "%.3f"%((data.dmesg['suspend_cpu']['end'] - data.dmesg['suspend_general']['start'])*1000)
- sftime = "%.3f"%(data.fwSuspend / 1000000.0)
- rftime = "%.3f"%(data.fwResume / 1000000.0)
- rktime = "%.3f"%((data.dmesg['resume_general']['end'] - data.dmesg['resume_cpu']['start'])*1000)
- devtl.html['timeline'] += html_timegroups.format(sktime, sftime, rftime, rktime)
+ suspend_time = '%.0f'%((data.tSuspended-data.start)*1000 + \
+ (data.fwSuspend/1000000.0))
+ resume_time = '%.0f'%((tEnd-data.tSuspended)*1000 + \
+ (data.fwResume/1000000.0))
+ testdesc1 = 'Total'
+ testdesc2 = ''
+ if(len(testruns) > 1):
+ testdesc1 = testdesc2 = textnum[data.testnumber]
+ testdesc2 += ' '
+ if(data.tLow == 0):
+ thtml = html_timetotal.format(suspend_time, \
+ resume_time, testdesc1)
+ else:
+ thtml = html_timetotal2.format(suspend_time, low_time, \
+ resume_time, testdesc1)
+ devtl.html['timeline'] += thtml
+ sktime = '%.3f'%((data.dmesg['suspend_machine']['end'] - \
+ data.getStart())*1000)
+ sftime = '%.3f'%(data.fwSuspend / 1000000.0)
+ rftime = '%.3f'%(data.fwResume / 1000000.0)
+ rktime = '%.3f'%((data.getEnd() - \
+ data.dmesg['resume_machine']['start'])*1000)
+ devtl.html['timeline'] += html_timegroups.format(sktime, \
+ sftime, rftime, rktime, testdesc2)
else:
- devtl.html['timeline'] = html_timetotal.format(suspend_time, resume_time, "Kernel")
+ suspend_time = '%.0f'%((data.tSuspended-data.start)*1000)
+ resume_time = '%.0f'%((tEnd-data.tSuspended)*1000)
+ testdesc = 'Kernel'
+ if(len(testruns) > 1):
+ testdesc = textnum[data.testnumber]+' '+testdesc
+ if(data.tLow == 0):
+ thtml = html_timetotal.format(suspend_time, \
+ resume_time, testdesc)
+ else:
+ thtml = html_timetotal2.format(suspend_time, low_time, \
+ resume_time, testdesc)
+ devtl.html['timeline'] += thtml
+
+ # time scale for potentially multiple datasets
+ t0 = testruns[0].start
+ tMax = testruns[-1].end
+ tSuspended = testruns[-1].tSuspended
+ tTotal = tMax - t0
- # determine the maximum number of rows we need to draw
- timelinerows = 0
+ # determine the maximum number of rows we need to draw
+ timelinerows = 0
+ for data in testruns:
for phase in data.dmesg:
list = data.dmesg[phase]['list']
rows = setTimelineRows(list, list)
if(rows > timelinerows):
timelinerows = rows
- # calculate the timeline height and create its bounding box
- devtl.setRows(timelinerows + 1)
- devtl.html['timeline'] += html_zoombox;
- devtl.html['timeline'] += html_timeline.format("dmesg", devtl.height);
+ # calculate the timeline height and create bounding box, add buttons
+ devtl.setRows(timelinerows + 1)
+ devtl.html['timeline'] += html_devlist1
+ if len(testruns) > 1:
+ devtl.html['timeline'] += html_devlist2
+ devtl.html['timeline'] += html_zoombox
+ devtl.html['timeline'] += html_timeline.format('dmesg', devtl.height)
- # draw the colored boxes for each of the phases
+ # draw the colored boxes for each of the phases
+ for data in testruns:
for b in data.dmesg:
phase = data.dmesg[b]
- left = "%.3f" % (((phase['start']-data.start)*100)/tTotal)
- width = "%.3f" % (((phase['end']-phase['start'])*100)/tTotal)
- devtl.html['timeline'] += html_phase.format(left, width, "%.3f"%devtl.scaleH, "%.3f"%(100-devtl.scaleH), data.dmesg[b]['color'], "")
-
- # draw the time scale, try to make the number of labels readable
- devtl.html['scale'] = createTimeScale(t0, tMax, data.tSuspended)
- devtl.html['timeline'] += devtl.html['scale']
+ length = phase['end']-phase['start']
+ left = '%.3f' % (((phase['start']-t0)*100.0)/tTotal)
+ width = '%.3f' % ((length*100.0)/tTotal)
+ devtl.html['timeline'] += html_phase.format(left, width, \
+ '%.3f'%devtl.scaleH, '%.3f'%(100-devtl.scaleH), \
+ data.dmesg[b]['color'], '')
+
+ # draw the time scale, try to make the number of labels readable
+ devtl.html['scale'] = createTimeScale(t0, tMax, tSuspended)
+ devtl.html['timeline'] += devtl.html['scale']
+ for data in testruns:
for b in data.dmesg:
phaselist = data.dmesg[b]['list']
for d in phaselist:
name = d
- if(d in data.altdevname):
- name = data.altdevname[d]
+ drv = ''
dev = phaselist[d]
+ if(d in sysvals.altdevname):
+ name = sysvals.altdevname[d]
+ if('drv' in dev and dev['drv']):
+ drv = ' {%s}' % dev['drv']
height = (100.0 - devtl.scaleH)/data.dmesg[b]['row']
- top = "%.3f" % ((dev['row']*height) + devtl.scaleH)
- left = "%.3f" % (((dev['start']-data.start)*100)/tTotal)
- width = "%.3f" % (((dev['end']-dev['start'])*100)/tTotal)
- len = " (%0.3f ms) " % ((dev['end']-dev['start'])*1000)
- color = "rgba(204,204,204,0.5)"
- devtl.html['timeline'] += html_device.format(dev['id'], name+len+b, left, top, "%.3f"%height, width, name)
-
- # timeline is finished
- devtl.html['timeline'] += "</div>\n</div>\n"
-
- # draw a legend which describes the phases by color
- devtl.html['legend'] = "<div class=\"legend\">\n"
- pdelta = 100.0/data.phases.__len__()
- pmargin = pdelta / 4.0
- for phase in data.phases:
- order = "%.2f" % ((data.dmesg[phase]['order'] * pdelta) + pmargin)
- name = string.replace(phase, "_", " ")
- devtl.html['legend'] += html_legend.format(order, data.dmesg[phase]['color'], name)
- devtl.html['legend'] += "</div>\n"
+ top = '%.3f' % ((dev['row']*height) + devtl.scaleH)
+ left = '%.3f' % (((dev['start']-t0)*100)/tTotal)
+ width = '%.3f' % (((dev['end']-dev['start'])*100)/tTotal)
+ length = ' (%0.3f ms) ' % ((dev['end']-dev['start'])*1000)
+ color = 'rgba(204,204,204,0.5)'
+ devtl.html['timeline'] += html_device.format(dev['id'], \
+ d+drv+length+b, left, top, '%.3f'%height, width, name+drv)
+
+ # draw any trace events found
+ for data in testruns:
+ for b in data.dmesg:
+ phaselist = data.dmesg[b]['list']
+ for name in phaselist:
+ dev = phaselist[name]
+ if('traceevents' in dev):
+ vprint('Debug trace events found for device %s' % name)
+ vprint('%20s %20s %10s %8s' % ('action', \
+ 'name', 'time(ms)', 'length(ms)'))
+ for e in dev['traceevents']:
+ vprint('%20s %20s %10.3f %8.3f' % (e.action, \
+ e.name, e.time*1000, e.length*1000))
+ height = (100.0 - devtl.scaleH)/data.dmesg[b]['row']
+ top = '%.3f' % ((dev['row']*height) + devtl.scaleH)
+ left = '%.3f' % (((e.time-t0)*100)/tTotal)
+ width = '%.3f' % (e.length*100/tTotal)
+ color = 'rgba(204,204,204,0.5)'
+ devtl.html['timeline'] += \
+ html_traceevent.format(e.action+' '+e.name, \
+ left, top, '%.3f'%height, \
+ width, e.color, '')
+
+ # timeline is finished
+ devtl.html['timeline'] += '</div>\n</div>\n'
+
+ # draw a legend which describes the phases by color
+ data = testruns[-1]
+ devtl.html['legend'] = '<div class="legend">\n'
+ pdelta = 100.0/len(data.phases)
+ pmargin = pdelta / 4.0
+ for phase in data.phases:
+ order = '%.2f' % ((data.dmesg[phase]['order'] * pdelta) + pmargin)
+ name = string.replace(phase, '_', ' ')
+ devtl.html['legend'] += html_legend.format(order, \
+ data.dmesg[phase]['color'], name)
+ devtl.html['legend'] += '</div>\n'
hf = open(sysvals.htmlfile, 'w')
thread_height = 0
- # write the html header first (html head, css code, everything up to the start of body)
- html_header = "<!DOCTYPE html>\n<html>\n<head>\n\
- <meta http-equiv=\"content-type\" content=\"text/html; charset=UTF-8\">\n\
+ # write the html header first (html head, css code, up to body start)
+ html_header = '<!DOCTYPE html>\n<html>\n<head>\n\
+ <meta http-equiv="content-type" content="text/html; charset=UTF-8">\n\
<title>AnalyzeSuspend</title>\n\
- <style type='text/css'>\n\
+ <style type=\'text/css\'>\n\
body {overflow-y: scroll;}\n\
.stamp {width: 100%;text-align:center;background-color:gray;line-height:30px;color:white;font: 25px Arial;}\n\
.callgraph {margin-top: 30px;box-shadow: 5px 5px 20px black;}\n\
.callgraph article * {padding-left: 28px;}\n\
h1 {color:black;font: bold 30px Times;}\n\
+ t0 {color:black;font: bold 30px Times;}\n\
+ t1 {color:black;font: 30px Times;}\n\
+ t2 {color:black;font: 25px Times;}\n\
+ t3 {color:black;font: 20px Times;white-space:nowrap;}\n\
+ t4 {color:black;font: bold 30px Times;line-height:60px;white-space:nowrap;}\n\
table {width:100%;}\n\
.gray {background-color:rgba(80,80,80,0.1);}\n\
.green {background-color:rgba(204,255,204,0.4);}\n\
.time1 {font: 22px Arial;border:1px solid;}\n\
.time2 {font: 15px Arial;border-bottom:1px solid;border-left:1px solid;border-right:1px solid;}\n\
td {text-align: center;}\n\
+ r {color:#500000;font:15px Tahoma;}\n\
+ n {color:#505050;font:15px Tahoma;}\n\
.tdhl {color: red;}\n\
.hide {display: none;}\n\
.pf {display: none;}\n\
- .pf:checked + label {background: url(\'data:image/svg+xml;utf,<?xml version=\"1.0\" standalone=\"no\"?><svg xmlns=\"http://www.w3.org/2000/svg\" height=\"18\" width=\"18\" version=\"1.1\"><circle cx=\"9\" cy=\"9\" r=\"8\" stroke=\"black\" stroke-width=\"1\" fill=\"white\"/><rect x=\"4\" y=\"8\" width=\"10\" height=\"2\" style=\"fill:black;stroke-width:0\"/><rect x=\"8\" y=\"4\" width=\"2\" height=\"10\" style=\"fill:black;stroke-width:0\"/></svg>\') no-repeat left center;}\n\
- .pf:not(:checked) ~ label {background: url(\'data:image/svg+xml;utf,<?xml version=\"1.0\" standalone=\"no\"?><svg xmlns=\"http://www.w3.org/2000/svg\" height=\"18\" width=\"18\" version=\"1.1\"><circle cx=\"9\" cy=\"9\" r=\"8\" stroke=\"black\" stroke-width=\"1\" fill=\"white\"/><rect x=\"4\" y=\"8\" width=\"10\" height=\"2\" style=\"fill:black;stroke-width:0\"/></svg>\') no-repeat left center;}\n\
+ .pf:checked + label {background: url(\'data:image/svg+xml;utf,<?xml version="1.0" standalone="no"?><svg xmlns="http://www.w3.org/2000/svg" height="18" width="18" version="1.1"><circle cx="9" cy="9" r="8" stroke="black" stroke-width="1" fill="white"/><rect x="4" y="8" width="10" height="2" style="fill:black;stroke-width:0"/><rect x="8" y="4" width="2" height="10" style="fill:black;stroke-width:0"/></svg>\') no-repeat left center;}\n\
+ .pf:not(:checked) ~ label {background: url(\'data:image/svg+xml;utf,<?xml version="1.0" standalone="no"?><svg xmlns="http://www.w3.org/2000/svg" height="18" width="18" version="1.1"><circle cx="9" cy="9" r="8" stroke="black" stroke-width="1" fill="white"/><rect x="4" y="8" width="10" height="2" style="fill:black;stroke-width:0"/></svg>\') no-repeat left center;}\n\
.pf:checked ~ *:not(:nth-child(2)) {display: none;}\n\
.zoombox {position: relative; width: 100%; overflow-x: scroll;}\n\
.timeline {position: relative; font-size: 14px;cursor: pointer;width: 100%; overflow: hidden; background-color:#dddddd;}\n\
- .thread {position: absolute; height: "+"%.3f"%thread_height+"%; overflow: hidden; line-height: 30px; border:1px solid;text-align:center;white-space:nowrap;background-color:rgba(204,204,204,0.5);}\n\
+ .thread {position: absolute; height: '+'%.3f'%thread_height+'%; overflow: hidden; line-height: 30px; border:1px solid;text-align:center;white-space:nowrap;background-color:rgba(204,204,204,0.5);}\n\
.thread:hover {background-color:white;border:1px solid red;z-index:10;}\n\
+ .hover {background-color:white;border:1px solid red;z-index:10;}\n\
+ .traceevent {position: absolute;opacity: 0.3;height: '+'%.3f'%thread_height+'%;width:0;overflow:hidden;line-height:30px;text-align:center;white-space:nowrap;}\n\
.phase {position: absolute;overflow: hidden;border:0px;text-align:center;}\n\
- .t {position: absolute; top: 0%; height: 100%; border-right:1px solid black;}\n\
+ .phaselet {position:absolute;overflow:hidden;border:0px;text-align:center;height:100px;font-size:24px;}\n\
+ .t {position:absolute;top:0%;height:100%;border-right:1px solid black;}\n\
.legend {position: relative; width: 100%; height: 40px; text-align: center;margin-bottom:20px}\n\
.legend .square {position:absolute;top:10px; width: 0px;height: 20px;border:1px solid;padding-left:20px;}\n\
button {height:40px;width:200px;margin-bottom:20px;margin-top:20px;font-size:24px;}\n\
- </style>\n</head>\n<body>\n"
+ .devlist {position:'+x2changes[1]+';width:190px;}\n\
+ #devicedetail {height:100px;box-shadow: 5px 5px 20px black;}\n\
+ </style>\n</head>\n<body>\n'
hf.write(html_header)
# write the test title and general info header
- if(data.stamp['time'] != ""):
- hf.write(headline_stamp.format(data.stamp['host'],
- data.stamp['kernel'], data.stamp['mode'], data.stamp['time']))
-
- # write the dmesg data (device timeline)
- if(data.usedmesg):
- hf.write(devtl.html['timeline'])
- hf.write(devtl.html['legend'])
- hf.write('<div id="devicedetail"></div>\n')
- hf.write('<div id="devicetree"></div>\n')
+ if(sysvals.stamp['time'] != ""):
+ hf.write(headline_stamp.format(sysvals.stamp['host'],
+ sysvals.stamp['kernel'], sysvals.stamp['mode'], \
+ sysvals.stamp['time']))
+
+ # write the device timeline
+ hf.write(devtl.html['timeline'])
+ hf.write(devtl.html['legend'])
+ hf.write('<div id="devicedetailtitle"></div>\n')
+ hf.write('<div id="devicedetail" style="display:none;">\n')
+ # draw the colored boxes for the device detail section
+ for data in testruns:
+ hf.write('<div id="devicedetail%d">\n' % data.testnumber)
+ for b in data.phases:
+ phase = data.dmesg[b]
+ length = phase['end']-phase['start']
+ left = '%.3f' % (((phase['start']-t0)*100.0)/tTotal)
+ width = '%.3f' % ((length*100.0)/tTotal)
+ hf.write(html_phaselet.format(b, left, width, \
+ data.dmesg[b]['color']))
+ hf.write('</div>\n')
+ hf.write('</div>\n')
# write the ftrace data (callgraph)
- if(data.useftrace):
+ data = testruns[-1]
+ if(sysvals.usecallgraph):
hf.write('<section id="callgraphs" class="callgraph">\n')
# write out the ftrace data converted to html
html_func_top = '<article id="{0}" class="atop" style="background-color:{1}">\n<input type="checkbox" class="pf" id="f{2}" checked/><label for="f{2}">{3} {4}</label>\n'
if('ftrace' not in list[devname]):
continue
name = devname
- if(devname in data.altdevname):
- name = data.altdevname[devname]
+ if(devname in sysvals.altdevname):
+ name = sysvals.altdevname[devname]
devid = list[devname]['id']
cg = list[devname]['ftrace']
- flen = "(%.3f ms)" % ((cg.end - cg.start)*1000)
- hf.write(html_func_top.format(devid, data.dmesg[p]['color'], num, name+" "+p, flen))
+ flen = '<r>(%.3f ms @ %.3f to %.3f)</r>' % \
+ ((cg.end - cg.start)*1000, cg.start*1000, cg.end*1000)
+ hf.write(html_func_top.format(devid, data.dmesg[p]['color'], \
+ num, name+' '+p, flen))
num += 1
for line in cg.list:
if(line.length < 0.000000001):
- flen = ""
+ flen = ''
else:
- flen = "(%.3f ms)" % (line.length*1000)
+ flen = '<n>(%.3f ms @ %.3f)</n>' % (line.length*1000, \
+ line.time*1000)
if(line.freturn and line.fcall):
hf.write(html_func_leaf.format(line.name, flen))
elif(line.freturn):
hf.write(html_func_start.format(num, line.name, flen))
num += 1
hf.write(html_func_end)
- hf.write("\n\n </section>\n")
+ hf.write('\n\n </section>\n')
# write the footer and close
- addScriptCode(hf)
- hf.write("</body>\n</html>\n")
+ addScriptCode(hf, testruns)
+ hf.write('</body>\n</html>\n')
hf.close()
return True
-def addScriptCode(hf):
- global data
-
- t0 = (data.start - data.tSuspended) * 1000
- tMax = (data.end - data.tSuspended) * 1000
+# Function: addScriptCode
+# Description:
+# Adds the javascript code to the output html
+# Arguments:
+# hf: the open html file pointer
+# testruns: array of Data objects from parseKernelLog or parseTraceLog
+def addScriptCode(hf, testruns):
+ t0 = (testruns[0].start - testruns[-1].tSuspended) * 1000
+ tMax = (testruns[-1].end - testruns[-1].tSuspended) * 1000
# create an array in javascript memory with the device details
- detail = ' var bounds = [%f,%f];\n' % (t0, tMax)
- detail += ' var d = [];\n'
- dfmt = ' d["%s"] = { n:"%s", p:"%s", c:[%s] };\n';
- for p in data.dmesg:
- list = data.dmesg[p]['list']
- for d in list:
- parent = data.deviceParentID(d, p)
- idlist = data.deviceChildrenIDs(d, p)
- idstr = ""
- for i in idlist:
- if(idstr == ""):
- idstr += '"'+i+'"'
- else:
- idstr += ', '+'"'+i+'"'
- detail += dfmt % (list[d]['id'], d, parent, idstr)
-
+ detail = ' var devtable = [];\n'
+ for data in testruns:
+ topo = data.deviceTopology()
+ detail += ' devtable[%d] = "%s";\n' % (data.testnumber, topo)
+ detail += ' var bounds = [%f,%f];\n' % (t0, tMax)
# add the code which will manipulate the data in the browser
script_code = \
'<script type="text/javascript">\n'+detail+\
- ' var filter = [];\n'\
- ' var table = [];\n'\
- ' function deviceParent(devid) {\n'\
- ' var devlist = [];\n'\
- ' if(filter.indexOf(devid) < 0) filter[filter.length] = devid;\n'\
- ' if(d[devid].p in d)\n'\
- ' devlist = deviceParent(d[devid].p);\n'\
- ' else if(d[devid].p != "")\n'\
- ' devlist = [d[devid].p];\n'\
- ' devlist[devlist.length] = d[devid].n;\n'\
- ' return devlist;\n'\
- ' }\n'\
- ' function deviceChildren(devid, column, row) {\n'\
- ' if(!(devid in d)) return;\n'\
- ' if(filter.indexOf(devid) < 0) filter[filter.length] = devid;\n'\
- ' var cell = {name: d[devid].n, span: 1};\n'\
- ' var span = 0;\n'\
- ' if(column >= table.length) table[column] = [];\n'\
- ' table[column][row] = cell;\n'\
- ' for(var i = 0; i < d[devid].c.length; i++) {\n'\
- ' var cid = d[devid].c[i];\n'\
- ' span += deviceChildren(cid, column+1, row+span);\n'\
- ' }\n'\
- ' if(span == 0) span = 1;\n'\
- ' table[column][row].span = span;\n'\
- ' return span;\n'\
- ' }\n'\
- ' function deviceTree(devid, resume) {\n'\
- ' var html = "<table border=1>";\n'\
- ' filter = [];\n'\
- ' table = [];\n'\
- ' plist = deviceParent(devid);\n'\
- ' var devidx = plist.length - 1;\n'\
- ' for(var i = 0; i < devidx; i++)\n'\
- ' table[i] = [{name: plist[i], span: 1}];\n'\
- ' deviceChildren(devid, devidx, 0);\n'\
- ' for(var i = 0; i < devidx; i++)\n'\
- ' table[i][0].span = table[devidx][0].span;\n'\
- ' for(var row = 0; row < table[0][0].span; row++) {\n'\
- ' html += "<tr>";\n'\
- ' for(var col = 0; col < table.length; col++)\n'\
- ' if(row in table[col]) {\n'\
- ' var cell = table[col][row];\n'\
- ' var args = "";\n'\
- ' if(cell.span > 1)\n'\
- ' args += " rowspan="+cell.span;\n'\
- ' if((col == devidx) && (row == 0))\n'\
- ' args += " class=tdhl";\n'\
- ' if(resume)\n'\
- ' html += "<td"+args+">"+cell.name+" →</td>";\n'\
- ' else\n'\
- ' html += "<td"+args+">← "+cell.name+"</td>";\n'\
- ' }\n'\
- ' html += "</tr>";\n'\
- ' }\n'\
- ' html += "</table>";\n'\
- ' return html;\n'\
- ' }\n'\
' function zoomTimeline() {\n'\
' var timescale = document.getElementById("timescale");\n'\
' var dmesg = document.getElementById("dmesg");\n'\
' for(var s = ((t0 / tS)|0) * tS; s < tMax; s += tS) {\n'\
' var pos = (tMax - s) * 100.0 / tTotal;\n'\
' var name = (s == 0)?"S/R":(s+"ms");\n'\
- ' html += \"<div class=\\\"t\\\" style=\\\"right:\"+pos+\"%\\\">\"+name+\"</div>\";\n'\
+ ' html += "<div class=\\"t\\" style=\\"right:"+pos+"%\\">"+name+"</div>";\n'\
' }\n'\
' timescale.innerHTML = html;\n'\
' }\n'\
+ ' function deviceHover() {\n'\
+ ' var name = this.title.slice(0, this.title.indexOf(" ("));\n'\
+ ' var dmesg = document.getElementById("dmesg");\n'\
+ ' var dev = dmesg.getElementsByClassName("thread");\n'\
+ ' var cpu = -1;\n'\
+ ' if(name.match("CPU_ON\[[0-9]*\]"))\n'\
+ ' cpu = parseInt(name.slice(7));\n'\
+ ' else if(name.match("CPU_OFF\[[0-9]*\]"))\n'\
+ ' cpu = parseInt(name.slice(8));\n'\
+ ' for (var i = 0; i < dev.length; i++) {\n'\
+ ' dname = dev[i].title.slice(0, dev[i].title.indexOf(" ("));\n'\
+ ' if((cpu >= 0 && dname.match("CPU_O[NF]*\\\[*"+cpu+"\\\]")) ||\n'\
+ ' (name == dname))\n'\
+ ' {\n'\
+ ' dev[i].className = "thread hover";\n'\
+ ' } else {\n'\
+ ' dev[i].className = "thread";\n'\
+ ' }\n'\
+ ' }\n'\
+ ' }\n'\
+ ' function deviceUnhover() {\n'\
+ ' var dmesg = document.getElementById("dmesg");\n'\
+ ' var dev = dmesg.getElementsByClassName("thread");\n'\
+ ' for (var i = 0; i < dev.length; i++) {\n'\
+ ' dev[i].className = "thread";\n'\
+ ' }\n'\
+ ' }\n'\
+ ' function deviceTitle(title, total, cpu) {\n'\
+ ' var prefix = "Total";\n'\
+ ' if(total.length > 3) {\n'\
+ ' prefix = "Average";\n'\
+ ' total[1] = (total[1]+total[3])/2;\n'\
+ ' total[2] = (total[2]+total[4])/2;\n'\
+ ' }\n'\
+ ' var devtitle = document.getElementById("devicedetailtitle");\n'\
+ ' var name = title.slice(0, title.indexOf(" "));\n'\
+ ' if(cpu >= 0) name = "CPU"+cpu;\n'\
+ ' var driver = "";\n'\
+ ' var tS = "<t2>(</t2>";\n'\
+ ' var tR = "<t2>)</t2>";\n'\
+ ' if(total[1] > 0)\n'\
+ ' tS = "<t2>("+prefix+" Suspend:</t2><t0> "+total[1].toFixed(3)+" ms</t0> ";\n'\
+ ' if(total[2] > 0)\n'\
+ ' tR = " <t2>"+prefix+" Resume:</t2><t0> "+total[2].toFixed(3)+" ms<t2>)</t2></t0>";\n'\
+ ' var s = title.indexOf("{");\n'\
+ ' var e = title.indexOf("}");\n'\
+ ' if((s >= 0) && (e >= 0))\n'\
+ ' driver = title.slice(s+1, e) + " <t1>@</t1> ";\n'\
+ ' if(total[1] > 0 && total[2] > 0)\n'\
+ ' devtitle.innerHTML = "<t0>"+driver+name+"</t0> "+tS+tR;\n'\
+ ' else\n'\
+ ' devtitle.innerHTML = "<t0>"+title+"</t0>";\n'\
+ ' return name;\n'\
+ ' }\n'\
' function deviceDetail() {\n'\
- ' var devtitle = document.getElementById("devicedetail");\n'\
- ' devtitle.innerHTML = "<h1>"+this.title+"</h1>";\n'\
- ' var devtree = document.getElementById("devicetree");\n'\
- ' devtree.innerHTML = deviceTree(this.id, (this.title.indexOf("resume") >= 0));\n'\
+ ' var devinfo = document.getElementById("devicedetail");\n'\
+ ' devinfo.style.display = "block";\n'\
+ ' var name = this.title.slice(0, this.title.indexOf(" ("));\n'\
+ ' var cpu = -1;\n'\
+ ' if(name.match("CPU_ON\[[0-9]*\]"))\n'\
+ ' cpu = parseInt(name.slice(7));\n'\
+ ' else if(name.match("CPU_OFF\[[0-9]*\]"))\n'\
+ ' cpu = parseInt(name.slice(8));\n'\
+ ' var dmesg = document.getElementById("dmesg");\n'\
+ ' var dev = dmesg.getElementsByClassName("thread");\n'\
+ ' var idlist = [];\n'\
+ ' var pdata = [[]];\n'\
+ ' var pd = pdata[0];\n'\
+ ' var total = [0.0, 0.0, 0.0];\n'\
+ ' for (var i = 0; i < dev.length; i++) {\n'\
+ ' dname = dev[i].title.slice(0, dev[i].title.indexOf(" ("));\n'\
+ ' if((cpu >= 0 && dname.match("CPU_O[NF]*\\\[*"+cpu+"\\\]")) ||\n'\
+ ' (name == dname))\n'\
+ ' {\n'\
+ ' idlist[idlist.length] = dev[i].id;\n'\
+ ' var tidx = 1;\n'\
+ ' if(dev[i].id[0] == "a") {\n'\
+ ' pd = pdata[0];\n'\
+ ' } else {\n'\
+ ' if(pdata.length == 1) pdata[1] = [];\n'\
+ ' if(total.length == 3) total[3]=total[4]=0.0;\n'\
+ ' pd = pdata[1];\n'\
+ ' tidx = 3;\n'\
+ ' }\n'\
+ ' var info = dev[i].title.split(" ");\n'\
+ ' var pname = info[info.length-1];\n'\
+ ' pd[pname] = parseFloat(info[info.length-3].slice(1));\n'\
+ ' total[0] += pd[pname];\n'\
+ ' if(pname.indexOf("suspend") >= 0)\n'\
+ ' total[tidx] += pd[pname];\n'\
+ ' else\n'\
+ ' total[tidx+1] += pd[pname];\n'\
+ ' }\n'\
+ ' }\n'\
+ ' var devname = deviceTitle(this.title, total, cpu);\n'\
+ ' var left = 0.0;\n'\
+ ' for (var t = 0; t < pdata.length; t++) {\n'\
+ ' pd = pdata[t];\n'\
+ ' devinfo = document.getElementById("devicedetail"+t);\n'\
+ ' var phases = devinfo.getElementsByClassName("phaselet");\n'\
+ ' for (var i = 0; i < phases.length; i++) {\n'\
+ ' if(phases[i].id in pd) {\n'\
+ ' var w = 100.0*pd[phases[i].id]/total[0];\n'\
+ ' var fs = 32;\n'\
+ ' if(w < 8) fs = 4*w | 0;\n'\
+ ' var fs2 = fs*3/4;\n'\
+ ' phases[i].style.width = w+"%";\n'\
+ ' phases[i].style.left = left+"%";\n'\
+ ' phases[i].title = phases[i].id+" "+pd[phases[i].id]+" ms";\n'\
+ ' left += w;\n'\
+ ' var time = "<t4 style=\\"font-size:"+fs+"px\\">"+pd[phases[i].id]+" ms<br></t4>";\n'\
+ ' var pname = "<t3 style=\\"font-size:"+fs2+"px\\">"+phases[i].id.replace("_", " ")+"</t3>";\n'\
+ ' phases[i].innerHTML = time+pname;\n'\
+ ' } else {\n'\
+ ' phases[i].style.width = "0%";\n'\
+ ' phases[i].style.left = left+"%";\n'\
+ ' }\n'\
+ ' }\n'\
+ ' }\n'\
' var cglist = document.getElementById("callgraphs");\n'\
' if(!cglist) return;\n'\
' var cg = cglist.getElementsByClassName("atop");\n'\
' for (var i = 0; i < cg.length; i++) {\n'\
- ' if(filter.indexOf(cg[i].id) >= 0) {\n'\
+ ' if(idlist.indexOf(cg[i].id) >= 0) {\n'\
' cg[i].style.display = "block";\n'\
' } else {\n'\
' cg[i].style.display = "none";\n'\
' }\n'\
' }\n'\
' }\n'\
+ ' function devListWindow(e) {\n'\
+ ' var sx = e.clientX;\n'\
+ ' if(sx > window.innerWidth - 440)\n'\
+ ' sx = window.innerWidth - 440;\n'\
+ ' var cfg="top="+e.screenY+", left="+sx+", width=440, height=720, scrollbars=yes";\n'\
+ ' var win = window.open("", "_blank", cfg);\n'\
+ ' if(window.chrome) win.moveBy(sx, 0);\n'\
+ ' var html = "<title>"+e.target.innerHTML+"</title>"+\n'\
+ ' "<style type=\\"text/css\\">"+\n'\
+ ' " ul {list-style-type:circle;padding-left:10px;margin-left:10px;}"+\n'\
+ ' "</style>"\n'\
+ ' var dt = devtable[0];\n'\
+ ' if(e.target.id != "devlist1")\n'\
+ ' dt = devtable[1];\n'\
+ ' win.document.write(html+dt);\n'\
+ ' }\n'\
' window.addEventListener("load", function () {\n'\
' var dmesg = document.getElementById("dmesg");\n'\
' dmesg.style.width = "100%"\n'\
' document.getElementById("zoomin").onclick = zoomTimeline;\n'\
' document.getElementById("zoomout").onclick = zoomTimeline;\n'\
' document.getElementById("zoomdef").onclick = zoomTimeline;\n'\
+ ' var devlist = document.getElementsByClassName("devlist");\n'\
+ ' for (var i = 0; i < devlist.length; i++)\n'\
+ ' devlist[i].onclick = devListWindow;\n'\
' var dev = dmesg.getElementsByClassName("thread");\n'\
' for (var i = 0; i < dev.length; i++) {\n'\
' dev[i].onclick = deviceDetail;\n'\
+ ' dev[i].onmouseover = deviceHover;\n'\
+ ' dev[i].onmouseout = deviceUnhover;\n'\
' }\n'\
' zoomTimeline();\n'\
' });\n'\
# Function: executeSuspend
# Description:
-# Execute system suspend through the sysfs interface
+# Execute system suspend through the sysfs interface, then copy the output
+# dmesg and ftrace files to the test output directory.
def executeSuspend():
- global sysvals, data
-
- detectUSB()
- pf = open(sysvals.powerfile, 'w')
- # clear the kernel ring buffer just as we start
- os.system("dmesg -C")
- # start ftrace
- if(data.useftrace):
- print("START TRACING")
- os.system("echo 1 > "+sysvals.tpath+"tracing_on")
- os.system("echo SUSPEND START > "+sysvals.tpath+"trace_marker")
- # initiate suspend
- if(sysvals.rtcwake):
- print("SUSPEND START")
- os.system("rtcwake -s 10 -m "+sysvals.suspendmode)
- else:
- print("SUSPEND START (press a key to resume)")
+ global sysvals
+
+ detectUSB(False)
+ t0 = time.time()*1000
+ tp = sysvals.tpath
+ # execute however many s/r runs requested
+ for count in range(1,sysvals.execcount+1):
+ # clear the kernel ring buffer just as we start
+ os.system('dmesg -C')
+ # enable callgraph ftrace only for the second run
+ if(sysvals.usecallgraph and count == 2):
+ # set trace type
+ os.system('echo function_graph > '+tp+'current_tracer')
+ os.system('echo "" > '+tp+'set_ftrace_filter')
+ # set trace format options
+ os.system('echo funcgraph-abstime > '+tp+'trace_options')
+ os.system('echo funcgraph-proc > '+tp+'trace_options')
+ # focus only on device suspend and resume
+ os.system('cat '+tp+'available_filter_functions | '+\
+ 'grep dpm_run_callback > '+tp+'set_graph_function')
+ # if this is test2 and there's a delay, start here
+ if(count > 1 and sysvals.x2delay > 0):
+ tN = time.time()*1000
+ while (tN - t0) < sysvals.x2delay:
+ tN = time.time()*1000
+ time.sleep(0.001)
+ # start ftrace
+ if(sysvals.usecallgraph or sysvals.usetraceevents):
+ print('START TRACING')
+ os.system('echo 1 > '+tp+'tracing_on')
+ # initiate suspend
+ if(sysvals.usecallgraph or sysvals.usetraceevents):
+ os.system('echo SUSPEND START > '+tp+'trace_marker')
+ if(sysvals.rtcwake):
+ print('SUSPEND START')
+ print('will autoresume in %d seconds' % sysvals.rtcwaketime)
+ sysvals.rtcWakeAlarm()
+ else:
+ print('SUSPEND START (press a key to resume)')
+ pf = open(sysvals.powerfile, 'w')
pf.write(sysvals.suspendmode)
- # execution will pause here
- pf.close()
- # return from suspend
- print("RESUME COMPLETE")
- # stop ftrace
- if(data.useftrace):
- os.system("echo RESUME COMPLETE > "+sysvals.tpath+"trace_marker")
- os.system("echo 0 > "+sysvals.tpath+"tracing_on")
- print("CAPTURING FTRACE")
- os.system("echo \""+sysvals.teststamp+"\" > "+sysvals.ftracefile)
- os.system("cat "+sysvals.tpath+"trace >> "+sysvals.ftracefile)
- # grab a copy of the dmesg output
- print("CAPTURING DMESG")
- os.system("echo \""+sysvals.teststamp+"\" > "+sysvals.dmesgfile)
- os.system("dmesg -c >> "+sysvals.dmesgfile)
+ # execution will pause here
+ pf.close()
+ t0 = time.time()*1000
+ # return from suspend
+ print('RESUME COMPLETE')
+ if(sysvals.usecallgraph or sysvals.usetraceevents):
+ os.system('echo RESUME COMPLETE > '+tp+'trace_marker')
+ # see if there's firmware timing data to be had
+ t = sysvals.postresumetime
+ if(t > 0):
+ print('Waiting %d seconds for POST-RESUME trace events...' % t)
+ time.sleep(t)
+ # stop ftrace
+ if(sysvals.usecallgraph or sysvals.usetraceevents):
+ os.system('echo 0 > '+tp+'tracing_on')
+ print('CAPTURING TRACE')
+ writeDatafileHeader(sysvals.ftracefile)
+ os.system('cat '+tp+'trace >> '+sysvals.ftracefile)
+ os.system('echo "" > '+tp+'trace')
+ # grab a copy of the dmesg output
+ print('CAPTURING DMESG')
+ writeDatafileHeader(sysvals.dmesgfile)
+ os.system('dmesg -c >> '+sysvals.dmesgfile)
+
+def writeDatafileHeader(filename):
+ global sysvals
+
+ fw = getFPDT(False)
+ prt = sysvals.postresumetime
+ fp = open(filename, 'a')
+ fp.write(sysvals.teststamp+'\n')
+ if(fw):
+ fp.write('# fwsuspend %u fwresume %u\n' % (fw[0], fw[1]))
+ if(prt > 0):
+ fp.write('# post resume time %u\n' % prt)
+ fp.close()
+
+# Function: executeAndroidSuspend
+# Description:
+# Execute system suspend through the sysfs interface
+# on a remote android device, then transfer the output
+# dmesg and ftrace files to the local output directory.
+def executeAndroidSuspend():
+ global sysvals
+
+ # check to see if the display is currently off
+ tp = sysvals.tpath
+ out = os.popen(sysvals.adb+\
+ ' shell dumpsys power | grep mScreenOn').read().strip()
+ # if so we need to turn it on so we can issue a new suspend
+ if(out.endswith('false')):
+ print('Waking the device up for the test...')
+ # send the KEYPAD_POWER keyevent to wake it up
+ os.system(sysvals.adb+' shell input keyevent 26')
+ # wait a few seconds so the user can see the device wake up
+ time.sleep(3)
+ # execute however many s/r runs requested
+ for count in range(1,sysvals.execcount+1):
+ # clear the kernel ring buffer just as we start
+ os.system(sysvals.adb+' shell dmesg -c > /dev/null 2>&1')
+ # start ftrace
+ if(sysvals.usetraceevents):
+ print('START TRACING')
+ os.system(sysvals.adb+" shell 'echo 1 > "+tp+"tracing_on'")
+ # initiate suspend
+ for count in range(1,sysvals.execcount+1):
+ if(sysvals.usetraceevents):
+ os.system(sysvals.adb+\
+ " shell 'echo SUSPEND START > "+tp+"trace_marker'")
+ print('SUSPEND START (press a key on the device to resume)')
+ os.system(sysvals.adb+" shell 'echo "+sysvals.suspendmode+\
+ " > "+sysvals.powerfile+"'")
+ # execution will pause here, then adb will exit
+ while(True):
+ check = os.popen(sysvals.adb+\
+ ' shell pwd 2>/dev/null').read().strip()
+ if(len(check) > 0):
+ break
+ time.sleep(1)
+ if(sysvals.usetraceevents):
+ os.system(sysvals.adb+" shell 'echo RESUME COMPLETE > "+tp+\
+ "trace_marker'")
+ # return from suspend
+ print('RESUME COMPLETE')
+ # stop ftrace
+ if(sysvals.usetraceevents):
+ os.system(sysvals.adb+" shell 'echo 0 > "+tp+"tracing_on'")
+ print('CAPTURING TRACE')
+ os.system('echo "'+sysvals.teststamp+'" > '+sysvals.ftracefile)
+ os.system(sysvals.adb+' shell cat '+tp+\
+ 'trace >> '+sysvals.ftracefile)
+ # grab a copy of the dmesg output
+ print('CAPTURING DMESG')
+ os.system('echo "'+sysvals.teststamp+'" > '+sysvals.dmesgfile)
+ os.system(sysvals.adb+' shell dmesg >> '+sysvals.dmesgfile)
+
+# Function: setUSBDevicesAuto
+# Description:
+# Set the autosuspend control parameter of all USB devices to auto
+# This can be dangerous, so use at your own risk, most devices are set
+# to always-on since the kernel cant determine if the device can
+# properly autosuspend
+def setUSBDevicesAuto():
+ global sysvals
+
+ rootCheck()
+ for dirname, dirnames, filenames in os.walk('/sys/devices'):
+ if(re.match('.*/usb[0-9]*.*', dirname) and
+ 'idVendor' in filenames and 'idProduct' in filenames):
+ os.system('echo auto > %s/power/control' % dirname)
+ name = dirname.split('/')[-1]
+ desc = os.popen('cat %s/product 2>/dev/null' % \
+ dirname).read().replace('\n', '')
+ ctrl = os.popen('cat %s/power/control 2>/dev/null' % \
+ dirname).read().replace('\n', '')
+ print('control is %s for %6s: %s' % (ctrl, name, desc))
+
+# Function: yesno
+# Description:
+# Print out an equivalent Y or N for a set of known parameter values
+# Output:
+# 'Y', 'N', or ' ' if the value is unknown
+def yesno(val):
+ yesvals = ['auto', 'enabled', 'active', '1']
+ novals = ['on', 'disabled', 'suspended', 'forbidden', 'unsupported']
+ if val in yesvals:
+ return 'Y'
+ elif val in novals:
+ return 'N'
+ return ' '
+
+# Function: ms2nice
+# Description:
+# Print out a very concise time string in minutes and seconds
+# Output:
+# The time string, e.g. "1901m16s"
+def ms2nice(val):
+ ms = 0
+ try:
+ ms = int(val)
+ except:
+ return 0.0
+ m = ms / 60000
+ s = (ms / 1000) - (m * 60)
+ return '%3dm%2ds' % (m, s)
# Function: detectUSB
# Description:
-# Detect all the USB hosts and devices currently connected
-def detectUSB():
- global sysvals, data
-
- for dirname, dirnames, filenames in os.walk("/sys/devices"):
- if(re.match(r".*/usb[0-9]*.*", dirname) and
- "idVendor" in filenames and "idProduct" in filenames):
- vid = os.popen("cat %s/idVendor 2>/dev/null" % dirname).read().replace('\n', '')
- pid = os.popen("cat %s/idProduct 2>/dev/null" % dirname).read().replace('\n', '')
- product = os.popen("cat %s/product 2>/dev/null" % dirname).read().replace('\n', '')
+# Detect all the USB hosts and devices currently connected and add
+# a list of USB device names to sysvals for better timeline readability
+# Arguments:
+# output: True to output the info to stdout, False otherwise
+def detectUSB(output):
+ global sysvals
+
+ field = {'idVendor':'', 'idProduct':'', 'product':'', 'speed':''}
+ power = {'async':'', 'autosuspend':'', 'autosuspend_delay_ms':'',
+ 'control':'', 'persist':'', 'runtime_enabled':'',
+ 'runtime_status':'', 'runtime_usage':'',
+ 'runtime_active_time':'',
+ 'runtime_suspended_time':'',
+ 'active_duration':'',
+ 'connected_duration':''}
+ if(output):
+ print('LEGEND')
+ print('---------------------------------------------------------------------------------------------')
+ print(' A = async/sync PM queue Y/N D = autosuspend delay (seconds)')
+ print(' S = autosuspend Y/N rACTIVE = runtime active (min/sec)')
+ print(' P = persist across suspend Y/N rSUSPEN = runtime suspend (min/sec)')
+ print(' E = runtime suspend enabled/forbidden Y/N ACTIVE = active duration (min/sec)')
+ print(' R = runtime status active/suspended Y/N CONNECT = connected duration (min/sec)')
+ print(' U = runtime usage count')
+ print('---------------------------------------------------------------------------------------------')
+ print(' NAME ID DESCRIPTION SPEED A S P E R U D rACTIVE rSUSPEN ACTIVE CONNECT')
+ print('---------------------------------------------------------------------------------------------')
+
+ for dirname, dirnames, filenames in os.walk('/sys/devices'):
+ if(re.match('.*/usb[0-9]*.*', dirname) and
+ 'idVendor' in filenames and 'idProduct' in filenames):
+ for i in field:
+ field[i] = os.popen('cat %s/%s 2>/dev/null' % \
+ (dirname, i)).read().replace('\n', '')
name = dirname.split('/')[-1]
- if(len(product) > 0):
- data.altdevname[name] = "%s [%s]" % (product, name)
+ if(len(field['product']) > 0):
+ sysvals.altdevname[name] = \
+ '%s [%s]' % (field['product'], name)
else:
- data.altdevname[name] = "%s:%s [%s]" % (vid, pid, name)
-
+ sysvals.altdevname[name] = \
+ '%s:%s [%s]' % (field['idVendor'], \
+ field['idProduct'], name)
+ if(output):
+ for i in power:
+ power[i] = os.popen('cat %s/power/%s 2>/dev/null' % \
+ (dirname, i)).read().replace('\n', '')
+ if(re.match('usb[0-9]*', name)):
+ first = '%-8s' % name
+ else:
+ first = '%8s' % name
+ print('%s [%s:%s] %-20s %-4s %1s %1s %1s %1s %1s %1s %1s %s %s %s %s' % \
+ (first, field['idVendor'], field['idProduct'], \
+ field['product'][0:20], field['speed'], \
+ yesno(power['async']), \
+ yesno(power['control']), \
+ yesno(power['persist']), \
+ yesno(power['runtime_enabled']), \
+ yesno(power['runtime_status']), \
+ power['runtime_usage'], \
+ power['autosuspend'], \
+ ms2nice(power['runtime_active_time']), \
+ ms2nice(power['runtime_suspended_time']), \
+ ms2nice(power['active_duration']), \
+ ms2nice(power['connected_duration'])))
+
+# Function: getModes
+# Description:
+# Determine the supported power modes on this system
+# Output:
+# A string list of the available modes
def getModes():
global sysvals
- modes = ""
- if(os.path.exists(sysvals.powerfile)):
- fp = open(sysvals.powerfile, 'r')
- modes = string.split(fp.read())
- fp.close()
+ modes = ''
+ if(not sysvals.android):
+ if(os.path.exists(sysvals.powerfile)):
+ fp = open(sysvals.powerfile, 'r')
+ modes = string.split(fp.read())
+ fp.close()
+ else:
+ line = os.popen(sysvals.adb+' shell cat '+\
+ sysvals.powerfile).read().strip()
+ modes = string.split(line)
return modes
+# Function: getFPDT
+# Description:
+# Read the acpi bios tables and pull out FPDT, the firmware data
+# Arguments:
+# output: True to output the info to stdout, False otherwise
+def getFPDT(output):
+ global sysvals
+
+ rectype = {}
+ rectype[0] = 'Firmware Basic Boot Performance Record'
+ rectype[1] = 'S3 Performance Table Record'
+ prectype = {}
+ prectype[0] = 'Basic S3 Resume Performance Record'
+ prectype[1] = 'Basic S3 Suspend Performance Record'
+
+ rootCheck()
+ if(not os.path.exists(sysvals.fpdtpath)):
+ if(output):
+ doError('file doesnt exist: %s' % sysvals.fpdtpath, False)
+ return False
+ if(not os.access(sysvals.fpdtpath, os.R_OK)):
+ if(output):
+ doError('file isnt readable: %s' % sysvals.fpdtpath, False)
+ return False
+ if(not os.path.exists(sysvals.mempath)):
+ if(output):
+ doError('file doesnt exist: %s' % sysvals.mempath, False)
+ return False
+ if(not os.access(sysvals.mempath, os.R_OK)):
+ if(output):
+ doError('file isnt readable: %s' % sysvals.mempath, False)
+ return False
+
+ fp = open(sysvals.fpdtpath, 'rb')
+ buf = fp.read()
+ fp.close()
+
+ if(len(buf) < 36):
+ if(output):
+ doError('Invalid FPDT table data, should '+\
+ 'be at least 36 bytes', False)
+ return False
+
+ table = struct.unpack('4sIBB6s8sI4sI', buf[0:36])
+ if(output):
+ print('')
+ print('Firmware Performance Data Table (%s)' % table[0])
+ print(' Signature : %s' % table[0])
+ print(' Table Length : %u' % table[1])
+ print(' Revision : %u' % table[2])
+ print(' Checksum : 0x%x' % table[3])
+ print(' OEM ID : %s' % table[4])
+ print(' OEM Table ID : %s' % table[5])
+ print(' OEM Revision : %u' % table[6])
+ print(' Creator ID : %s' % table[7])
+ print(' Creator Revision : 0x%x' % table[8])
+ print('')
+
+ if(table[0] != 'FPDT'):
+ if(output):
+ doError('Invalid FPDT table')
+ return False
+ if(len(buf) <= 36):
+ return False
+ i = 0
+ fwData = [0, 0]
+ records = buf[36:]
+ fp = open(sysvals.mempath, 'rb')
+ while(i < len(records)):
+ header = struct.unpack('HBB', records[i:i+4])
+ if(header[0] not in rectype):
+ continue
+ if(header[1] != 16):
+ continue
+ addr = struct.unpack('Q', records[i+8:i+16])[0]
+ try:
+ fp.seek(addr)
+ first = fp.read(8)
+ except:
+ doError('Bad address 0x%x in %s' % (addr, sysvals.mempath), False)
+ rechead = struct.unpack('4sI', first)
+ recdata = fp.read(rechead[1]-8)
+ if(rechead[0] == 'FBPT'):
+ record = struct.unpack('HBBIQQQQQ', recdata)
+ if(output):
+ print('%s (%s)' % (rectype[header[0]], rechead[0]))
+ print(' Reset END : %u ns' % record[4])
+ print(' OS Loader LoadImage Start : %u ns' % record[5])
+ print(' OS Loader StartImage Start : %u ns' % record[6])
+ print(' ExitBootServices Entry : %u ns' % record[7])
+ print(' ExitBootServices Exit : %u ns' % record[8])
+ elif(rechead[0] == 'S3PT'):
+ if(output):
+ print('%s (%s)' % (rectype[header[0]], rechead[0]))
+ j = 0
+ while(j < len(recdata)):
+ prechead = struct.unpack('HBB', recdata[j:j+4])
+ if(prechead[0] not in prectype):
+ continue
+ if(prechead[0] == 0):
+ record = struct.unpack('IIQQ', recdata[j:j+prechead[1]])
+ fwData[1] = record[2]
+ if(output):
+ print(' %s' % prectype[prechead[0]])
+ print(' Resume Count : %u' % \
+ record[1])
+ print(' FullResume : %u ns' % \
+ record[2])
+ print(' AverageResume : %u ns' % \
+ record[3])
+ elif(prechead[0] == 1):
+ record = struct.unpack('QQ', recdata[j+4:j+prechead[1]])
+ fwData[0] = record[1] - record[0]
+ if(output):
+ print(' %s' % prectype[prechead[0]])
+ print(' SuspendStart : %u ns' % \
+ record[0])
+ print(' SuspendEnd : %u ns' % \
+ record[1])
+ print(' SuspendTime : %u ns' % \
+ fwData[0])
+ j += prechead[1]
+ if(output):
+ print('')
+ i += header[1]
+ fp.close()
+ return fwData
+
# Function: statusCheck
# Description:
-# Verify that the requested command and options will work
-def statusCheck(dryrun):
- global sysvals, data
- res = dict()
+# Verify that the requested command and options will work, and
+# print the results to the terminal
+# Output:
+# True if the test will work, False if not
+def statusCheck():
+ global sysvals
+ status = True
- if(data.notestrun):
- print("SUCCESS: The command should run!")
- return
+ if(sysvals.android):
+ print('Checking the android system ...')
+ else:
+ print('Checking this system (%s)...' % platform.node())
+
+ # check if adb is connected to a device
+ if(sysvals.android):
+ res = 'NO'
+ out = os.popen(sysvals.adb+' get-state').read().strip()
+ if(out == 'device'):
+ res = 'YES'
+ print(' is android device connected: %s' % res)
+ if(res != 'YES'):
+ print(' Please connect the device before using this tool')
+ return False
# check we have root access
- check = "YES"
- if(os.environ['USER'] != "root"):
- if(not dryrun):
- doError("root access is required", False)
- check = "NO"
- res[" have root access: "] = check
+ res = 'NO (No features of this tool will work!)'
+ if(sysvals.android):
+ out = os.popen(sysvals.adb+' shell id').read().strip()
+ if('root' in out):
+ res = 'YES'
+ else:
+ if(os.environ['USER'] == 'root'):
+ res = 'YES'
+ print(' have root access: %s' % res)
+ if(res != 'YES'):
+ if(sysvals.android):
+ print(' Try running "adb root" to restart the daemon as root')
+ else:
+ print(' Try running this script with sudo')
+ return False
# check sysfs is mounted
- check = "YES"
- if(not os.path.exists(sysvals.powerfile)):
- if(not dryrun):
- doError("sysfs must be mounted", False)
- check = "NO"
- res[" is sysfs mounted: "] = check
+ res = 'NO (No features of this tool will work!)'
+ if(sysvals.android):
+ out = os.popen(sysvals.adb+' shell ls '+\
+ sysvals.powerfile).read().strip()
+ if(out == sysvals.powerfile):
+ res = 'YES'
+ else:
+ if(os.path.exists(sysvals.powerfile)):
+ res = 'YES'
+ print(' is sysfs mounted: %s' % res)
+ if(res != 'YES'):
+ return False
# check target mode is a valid mode
- check = "YES"
+ res = 'NO'
modes = getModes()
- if(sysvals.suspendmode not in modes):
- if(not dryrun):
- doError("%s is not a value power mode" % sysvals.suspendmode, False)
- check = "NO"
- res[" is "+sysvals.suspendmode+" a power mode: "] = check
+ if(sysvals.suspendmode in modes):
+ res = 'YES'
+ else:
+ status = False
+ print(' is "%s" a valid power mode: %s' % (sysvals.suspendmode, res))
+ if(res == 'NO'):
+ print(' valid power modes are: %s' % modes)
+ print(' please choose one with -m')
+
+ # check if the tool can unlock the device
+ if(sysvals.android):
+ res = 'YES'
+ out1 = os.popen(sysvals.adb+\
+ ' shell dumpsys power | grep mScreenOn').read().strip()
+ out2 = os.popen(sysvals.adb+\
+ ' shell input').read().strip()
+ if(not out1.startswith('mScreenOn') or not out2.startswith('usage')):
+ res = 'NO (wake the android device up before running the test)'
+ print(' can I unlock the screen: %s' % res)
# check if ftrace is available
- if(data.useftrace):
- check = "YES"
- if(not verifyFtrace()):
- if(not dryrun):
- doError("ftrace is not configured", False)
- check = "NO"
- res[" is ftrace usable: "] = check
+ res = 'NO'
+ ftgood = verifyFtrace()
+ if(ftgood):
+ res = 'YES'
+ elif(sysvals.usecallgraph):
+ status = False
+ print(' is ftrace supported: %s' % res)
+
+ # what data source are we using
+ res = 'DMESG'
+ if(ftgood):
+ sysvals.usetraceeventsonly = True
+ sysvals.usetraceevents = False
+ for e in sysvals.traceevents:
+ check = False
+ if(sysvals.android):
+ out = os.popen(sysvals.adb+' shell ls -d '+\
+ sysvals.epath+e).read().strip()
+ if(out == sysvals.epath+e):
+ check = True
+ else:
+ if(os.path.exists(sysvals.epath+e)):
+ check = True
+ if(not check):
+ sysvals.usetraceeventsonly = False
+ if(e == 'suspend_resume' and check):
+ sysvals.usetraceevents = True
+ if(sysvals.usetraceevents and sysvals.usetraceeventsonly):
+ res = 'FTRACE (all trace events found)'
+ elif(sysvals.usetraceevents):
+ res = 'DMESG and FTRACE (suspend_resume trace event found)'
+ print(' timeline data source: %s' % res)
# check if rtcwake
- if(sysvals.rtcwake):
- check = "YES"
- version = os.popen("rtcwake -V 2>/dev/null").read()
- if(not version.startswith("rtcwake")):
- if(not dryrun):
- doError("rtcwake is not installed", False)
- check = "NO"
- res[" is rtcwake usable: "] = check
-
- if(dryrun):
- status = True
- print("Checking if system can run the current command:")
- for r in res:
- print("%s\t%s" % (r, res[r]))
- if(res[r] != "YES"):
- status = False
- if(status):
- print("SUCCESS: The command should run!")
- else:
- print("FAILURE: The command won't run!")
+ res = 'NO'
+ if(sysvals.rtcpath != ''):
+ res = 'YES'
+ elif(sysvals.rtcwake):
+ status = False
+ print(' is rtcwake supported: %s' % res)
-def printHelp():
- global sysvals
- modes = getModes()
-
- print("")
- print("AnalyzeSuspend")
- print("Usage: sudo analyze_suspend.py <options>")
- print("")
- print("Description:")
- print(" Initiates a system suspend/resume while capturing dmesg")
- print(" and (optionally) ftrace data to analyze device timing")
- print("")
- print(" Generates output files in subdirectory: suspend-mmddyy-HHMMSS")
- print(" HTML output: <hostname>_<mode>.html")
- print(" raw dmesg output: <hostname>_<mode>_dmesg.txt")
- print(" raw ftrace output (with -f): <hostname>_<mode>_ftrace.txt")
- print("")
- print("Options:")
- print(" [general]")
- print(" -h Print this help text")
- print(" -verbose Print extra information during execution and analysis")
- print(" -status Test to see if the system is enabled to run this tool")
- print(" -modes List available suspend modes")
- print(" -m mode Mode to initiate for suspend %s (default: %s)") % (modes, sysvals.suspendmode)
- print(" -rtcwake Use rtcwake to autoresume after 10 seconds (default: disabled)")
- print(" -f Use ftrace to create device callgraphs (default: disabled)")
- print(" [re-analyze data from previous runs]")
- print(" -dmesg dmesgfile Create HTML timeline from dmesg file")
- print(" -ftrace ftracefile Create HTML callgraph from ftrace file")
- print("")
- return True
+ return status
+# Function: doError
+# Description:
+# generic error function for catastrphic failures
+# Arguments:
+# msg: the error message to print
+# help: True if printHelp should be called after, False otherwise
def doError(msg, help):
- print("ERROR: %s") % msg
if(help == True):
printHelp()
+ print('ERROR: %s\n') % msg
sys.exit()
-# -- script main --
-# loop through the command line arguments
-cmd = ""
-args = iter(sys.argv[1:])
-for arg in args:
- if(arg == "-m"):
- try:
- val = args.next()
- except:
- doError("No mode supplied", True)
- sysvals.suspendmode = val
- elif(arg == "-f"):
- data.useftrace = True
- elif(arg == "-modes"):
- cmd = "modes"
- elif(arg == "-status"):
- cmd = "status"
- elif(arg == "-verbose"):
- data.verbose = True
- elif(arg == "-rtcwake"):
- sysvals.rtcwake = True
- elif(arg == "-dmesg"):
- try:
- val = args.next()
- except:
- doError("No dmesg file supplied", True)
- data.notestrun = True
- data.usedmesg = True
- sysvals.dmesgfile = val
- elif(arg == "-ftrace"):
- try:
- val = args.next()
- except:
- doError("No ftrace file supplied", True)
- data.notestrun = True
- data.useftrace = True
- sysvals.ftracefile = val
- elif(arg == "-h"):
- printHelp()
- sys.exit()
- else:
- doError("Invalid argument: "+arg, True)
-
-# just run a utility command and exit
-if(cmd != ""):
- if(cmd == "status"):
- statusCheck(True)
- elif(cmd == "modes"):
- modes = getModes()
- print modes
- sys.exit()
+# Function: doWarning
+# Description:
+# generic warning function for non-catastrophic anomalies
+# Arguments:
+# msg: the warning message to print
+# file: If not empty, a filename to request be sent to the owner for debug
+def doWarning(msg, file):
+ print('/* %s */') % msg
+ if(file):
+ print('/* For a fix, please send this'+\
+ ' %s file to <todd.e.brandt@intel.com> */' % file)
+
+# Function: rootCheck
+# Description:
+# quick check to see if we have root access
+def rootCheck():
+ if(os.environ['USER'] != 'root'):
+ doError('This script must be run as root', False)
-data.initialize()
+# Function: getArgInt
+# Description:
+# pull out an integer argument from the command line with checks
+def getArgInt(name, args, min, max):
+ try:
+ arg = args.next()
+ except:
+ doError(name+': no argument supplied', True)
+ try:
+ val = int(arg)
+ except:
+ doError(name+': non-integer value given', True)
+ if(val < min or val > max):
+ doError(name+': value should be between %d and %d' % (min, max), True)
+ return val
+
+# Function: rerunTest
+# Description:
+# generate an output from an existing set of ftrace/dmesg logs
+def rerunTest():
+ global sysvals
-# if instructed, re-analyze existing data files
-if(data.notestrun):
+ if(sysvals.ftracefile != ''):
+ doesTraceLogHaveTraceEvents()
+ if(sysvals.dmesgfile == '' and not sysvals.usetraceeventsonly):
+ doError('recreating this html output '+\
+ 'requires a dmesg file', False)
sysvals.setOutputFile()
- data.vprint("Output file: %s" % sysvals.htmlfile)
- if(sysvals.dmesgfile != ""):
- analyzeKernelLog()
- if(sysvals.ftracefile != ""):
- analyzeTraceLog()
- createHTML()
- sys.exit()
+ vprint('Output file: %s' % sysvals.htmlfile)
+ print('PROCESSING DATA')
+ if(sysvals.usetraceeventsonly):
+ testruns = parseTraceLog()
+ else:
+ testruns = loadKernelLog()
+ for data in testruns:
+ parseKernelLog(data)
+ if(sysvals.ftracefile != ''):
+ appendIncompleteTraceLog(testruns)
+ createHTML(testruns)
+
+# Function: runTest
+# Description:
+# execute a suspend/resume, gather the logs, and generate the output
+def runTest(subdir):
+ global sysvals
+
+ # prepare for the test
+ if(not sysvals.android):
+ initFtrace()
+ else:
+ initFtraceAndroid()
+ sysvals.initTestOutput(subdir)
+
+ vprint('Output files:\n %s' % sysvals.dmesgfile)
+ if(sysvals.usecallgraph or
+ sysvals.usetraceevents or
+ sysvals.usetraceeventsonly):
+ vprint(' %s' % sysvals.ftracefile)
+ vprint(' %s' % sysvals.htmlfile)
+
+ # execute the test
+ if(not sysvals.android):
+ executeSuspend()
+ else:
+ executeAndroidSuspend()
+
+ # analyze the data and create the html output
+ print('PROCESSING DATA')
+ if(sysvals.usetraceeventsonly):
+ # data for kernels 3.15 or newer is entirely in ftrace
+ testruns = parseTraceLog()
+ else:
+ # data for kernels older than 3.15 is primarily in dmesg
+ testruns = loadKernelLog()
+ for data in testruns:
+ parseKernelLog(data)
+ if(sysvals.usecallgraph or sysvals.usetraceevents):
+ appendIncompleteTraceLog(testruns)
+ createHTML(testruns)
+
+# Function: runSummary
+# Description:
+# create a summary of tests in a sub-directory
+def runSummary(subdir, output):
+ global sysvals
+
+ # get a list of ftrace output files
+ files = []
+ for dirname, dirnames, filenames in os.walk(subdir):
+ for filename in filenames:
+ if(re.match('.*_ftrace.txt', filename)):
+ files.append("%s/%s" % (dirname, filename))
+
+ # process the files in order and get an array of data objects
+ testruns = []
+ for file in sorted(files):
+ if output:
+ print("Test found in %s" % os.path.dirname(file))
+ sysvals.ftracefile = file
+ sysvals.dmesgfile = file.replace('_ftrace.txt', '_dmesg.txt')
+ doesTraceLogHaveTraceEvents()
+ sysvals.usecallgraph = False
+ if not sysvals.usetraceeventsonly:
+ if(not os.path.exists(sysvals.dmesgfile)):
+ print("Skipping %s: not a valid test input" % file)
+ continue
+ else:
+ if output:
+ f = os.path.basename(sysvals.ftracefile)
+ d = os.path.basename(sysvals.dmesgfile)
+ print("\tInput files: %s and %s" % (f, d))
+ testdata = loadKernelLog()
+ data = testdata[0]
+ parseKernelLog(data)
+ testdata = [data]
+ appendIncompleteTraceLog(testdata)
+ else:
+ if output:
+ print("\tInput file: %s" % os.path.basename(sysvals.ftracefile))
+ testdata = parseTraceLog()
+ data = testdata[0]
+ data.normalizeTime(data.tSuspended)
+ link = file.replace(subdir+'/', '').replace('_ftrace.txt', '.html')
+ data.outfile = link
+ testruns.append(data)
+
+ createHTMLSummarySimple(testruns, subdir+'/summary.html')
+
+# Function: printHelp
+# Description:
+# print out the help text
+def printHelp():
+ global sysvals
+ modes = getModes()
+
+ print('')
+ print('AnalyzeSuspend v%.1f' % sysvals.version)
+ print('Usage: sudo analyze_suspend.py <options>')
+ print('')
+ print('Description:')
+ print(' This tool is designed to assist kernel and OS developers in optimizing')
+ print(' their linux stack\'s suspend/resume time. Using a kernel image built')
+ print(' with a few extra options enabled, the tool will execute a suspend and')
+ print(' capture dmesg and ftrace data until resume is complete. This data is')
+ print(' transformed into a device timeline and an optional callgraph to give')
+ print(' a detailed view of which devices/subsystems are taking the most')
+ print(' time in suspend/resume.')
+ print('')
+ print(' Generates output files in subdirectory: suspend-mmddyy-HHMMSS')
+ print(' HTML output: <hostname>_<mode>.html')
+ print(' raw dmesg output: <hostname>_<mode>_dmesg.txt')
+ print(' raw ftrace output: <hostname>_<mode>_ftrace.txt')
+ print('')
+ print('Options:')
+ print(' [general]')
+ print(' -h Print this help text')
+ print(' -v Print the current tool version')
+ print(' -verbose Print extra information during execution and analysis')
+ print(' -status Test to see if the system is enabled to run this tool')
+ print(' -modes List available suspend modes')
+ print(' -m mode Mode to initiate for suspend %s (default: %s)') % (modes, sysvals.suspendmode)
+ print(' -rtcwake t Use rtcwake to autoresume after <t> seconds (default: disabled)')
+ print(' [advanced]')
+ print(' -f Use ftrace to create device callgraphs (default: disabled)')
+ print(' -filter "d1 d2 ..." Filter out all but this list of dev names')
+ print(' -x2 Run two suspend/resumes back to back (default: disabled)')
+ print(' -x2delay t Minimum millisecond delay <t> between the two test runs (default: 0 ms)')
+ print(' -postres t Time after resume completion to wait for post-resume events (default: 0 S)')
+ print(' -multi n d Execute <n> consecutive tests at <d> seconds intervals. The outputs will')
+ print(' be created in a new subdirectory with a summary page.')
+ print(' [utilities]')
+ print(' -fpdt Print out the contents of the ACPI Firmware Performance Data Table')
+ print(' -usbtopo Print out the current USB topology with power info')
+ print(' -usbauto Enable autosuspend for all connected USB devices')
+ print(' [android testing]')
+ print(' -adb binary Use the given adb binary to run the test on an android device.')
+ print(' The device should already be connected and with root access.')
+ print(' Commands will be executed on the device using "adb shell"')
+ print(' [re-analyze data from previous runs]')
+ print(' -ftrace ftracefile Create HTML output using ftrace input')
+ print(' -dmesg dmesgfile Create HTML output using dmesg (not needed for kernel >= 3.15)')
+ print(' -summary directory Create a summary of all test in this dir')
+ print('')
+ return True
-# verify that we can run a test
-data.usedmesg = True
-statusCheck(False)
-
-# prepare for the test
-if(data.useftrace):
- initFtrace()
-sysvals.initTestOutput()
-
-data.vprint("Output files:\n %s" % sysvals.dmesgfile)
-if(data.useftrace):
- data.vprint(" %s" % sysvals.ftracefile)
-data.vprint(" %s" % sysvals.htmlfile)
-
-# execute the test
-executeSuspend()
-analyzeKernelLog()
-if(data.useftrace):
- analyzeTraceLog()
-createHTML()
+# ----------------- MAIN --------------------
+# exec start (skipped if script is loaded as library)
+if __name__ == '__main__':
+ cmd = ''
+ cmdarg = ''
+ multitest = {'run': False, 'count': 0, 'delay': 0}
+ # loop through the command line arguments
+ args = iter(sys.argv[1:])
+ for arg in args:
+ if(arg == '-m'):
+ try:
+ val = args.next()
+ except:
+ doError('No mode supplied', True)
+ sysvals.suspendmode = val
+ elif(arg == '-adb'):
+ try:
+ val = args.next()
+ except:
+ doError('No adb binary supplied', True)
+ if(not os.path.exists(val)):
+ doError('file doesnt exist: %s' % val, False)
+ if(not os.access(val, os.X_OK)):
+ doError('file isnt executable: %s' % val, False)
+ try:
+ check = os.popen(val+' version').read().strip()
+ except:
+ doError('adb version failed to execute', False)
+ if(not re.match('Android Debug Bridge .*', check)):
+ doError('adb version failed to execute', False)
+ sysvals.adb = val
+ sysvals.android = True
+ elif(arg == '-x2'):
+ if(sysvals.postresumetime > 0):
+ doError('-x2 is not compatible with -postres', False)
+ sysvals.execcount = 2
+ elif(arg == '-x2delay'):
+ sysvals.x2delay = getArgInt('-x2delay', args, 0, 60000)
+ elif(arg == '-postres'):
+ if(sysvals.execcount != 1):
+ doError('-x2 is not compatible with -postres', False)
+ sysvals.postresumetime = getArgInt('-postres', args, 0, 3600)
+ elif(arg == '-f'):
+ sysvals.usecallgraph = True
+ elif(arg == '-modes'):
+ cmd = 'modes'
+ elif(arg == '-fpdt'):
+ cmd = 'fpdt'
+ elif(arg == '-usbtopo'):
+ cmd = 'usbtopo'
+ elif(arg == '-usbauto'):
+ cmd = 'usbauto'
+ elif(arg == '-status'):
+ cmd = 'status'
+ elif(arg == '-verbose'):
+ sysvals.verbose = True
+ elif(arg == '-v'):
+ print("Version %.1f" % sysvals.version)
+ sys.exit()
+ elif(arg == '-rtcwake'):
+ sysvals.rtcwake = True
+ sysvals.rtcwaketime = getArgInt('-rtcwake', args, 0, 3600)
+ elif(arg == '-multi'):
+ multitest['run'] = True
+ multitest['count'] = getArgInt('-multi n (exec count)', args, 2, 1000000)
+ multitest['delay'] = getArgInt('-multi d (delay between tests)', args, 0, 3600)
+ elif(arg == '-dmesg'):
+ try:
+ val = args.next()
+ except:
+ doError('No dmesg file supplied', True)
+ sysvals.notestrun = True
+ sysvals.dmesgfile = val
+ if(os.path.exists(sysvals.dmesgfile) == False):
+ doError('%s doesnt exist' % sysvals.dmesgfile, False)
+ elif(arg == '-ftrace'):
+ try:
+ val = args.next()
+ except:
+ doError('No ftrace file supplied', True)
+ sysvals.notestrun = True
+ sysvals.usecallgraph = True
+ sysvals.ftracefile = val
+ if(os.path.exists(sysvals.ftracefile) == False):
+ doError('%s doesnt exist' % sysvals.ftracefile, False)
+ elif(arg == '-summary'):
+ try:
+ val = args.next()
+ except:
+ doError('No directory supplied', True)
+ cmd = 'summary'
+ cmdarg = val
+ sysvals.notestrun = True
+ if(os.path.isdir(val) == False):
+ doError('%s isnt accesible' % val, False)
+ elif(arg == '-filter'):
+ try:
+ val = args.next()
+ except:
+ doError('No devnames supplied', True)
+ sysvals.setDeviceFilter(val)
+ elif(arg == '-h'):
+ printHelp()
+ sys.exit()
+ else:
+ doError('Invalid argument: '+arg, True)
+
+ # just run a utility command and exit
+ if(cmd != ''):
+ if(cmd == 'status'):
+ statusCheck()
+ elif(cmd == 'fpdt'):
+ if(sysvals.android):
+ doError('cannot read FPDT on android device', False)
+ getFPDT(True)
+ elif(cmd == 'usbtopo'):
+ if(sysvals.android):
+ doError('cannot read USB topology '+\
+ 'on an android device', False)
+ detectUSB(True)
+ elif(cmd == 'modes'):
+ modes = getModes()
+ print modes
+ elif(cmd == 'usbauto'):
+ setUSBDevicesAuto()
+ elif(cmd == 'summary'):
+ print("Generating a summary of folder \"%s\"" % cmdarg)
+ runSummary(cmdarg, True)
+ sys.exit()
+
+ # run test on android device
+ if(sysvals.android):
+ if(sysvals.usecallgraph):
+ doError('ftrace (-f) is not yet supported '+\
+ 'in the android kernel', False)
+ if(sysvals.notestrun):
+ doError('cannot analyze test files on the '+\
+ 'android device', False)
+
+ # if instructed, re-analyze existing data files
+ if(sysvals.notestrun):
+ rerunTest()
+ sys.exit()
+
+ # verify that we can run a test
+ if(not statusCheck()):
+ print('Check FAILED, aborting the test run!')
+ sys.exit()
+
+ if multitest['run']:
+ # run multiple tests in a separte subdirectory
+ s = 'x%d' % multitest['count']
+ subdir = datetime.now().strftime('suspend-'+s+'-%m%d%y-%H%M%S')
+ os.mkdir(subdir)
+ for i in range(multitest['count']):
+ if(i != 0):
+ print('Waiting %d seconds...' % (multitest['delay']))
+ time.sleep(multitest['delay'])
+ print('TEST (%d/%d) START' % (i+1, multitest['count']))
+ runTest(subdir)
+ print('TEST (%d/%d) COMPLETE' % (i+1, multitest['count']))
+ runSummary(subdir, False)
+ else:
+ # run the test in the current directory
+ runTest(".")
--- /dev/null
+/// Remove casting the values returned by memory allocation functions
+/// like kmalloc, kzalloc, kmem_cache_alloc, kmem_cache_zalloc etc.
+///
+//# This makes an effort to find cases of casting of values returned by
+//# kmalloc, kzalloc, kcalloc, kmem_cache_alloc, kmem_cache_zalloc,
+//# kmem_cache_alloc_node, kmalloc_node and kzalloc_node and removes
+//# the casting as it is not required. The result in the patch case may
+//#need some reformatting.
+//
+// Confidence: High
+// Copyright: 2014, Himangi Saraogi GPLv2.
+// Comments:
+// Options: --no-includes --include-headers
+//
+
+virtual context
+virtual patch
+virtual org
+virtual report
+
+//----------------------------------------------------------
+// For context mode
+//----------------------------------------------------------
+
+@depends on context@
+type T;
+@@
+
+* (T *)
+ \(kmalloc\|kzalloc\|kcalloc\|kmem_cache_alloc\|kmem_cache_zalloc\|
+ kmem_cache_alloc_node\|kmalloc_node\|kzalloc_node\)(...)
+
+//----------------------------------------------------------
+// For patch mode
+//----------------------------------------------------------
+
+@depends on patch@
+type T;
+@@
+
+- (T *)
+ (\(kmalloc\|kzalloc\|kcalloc\|kmem_cache_alloc\|kmem_cache_zalloc\|
+ kmem_cache_alloc_node\|kmalloc_node\|kzalloc_node\)(...))
+
+//----------------------------------------------------------
+// For org and report mode
+//----------------------------------------------------------
+
+@r depends on org || report@
+type T;
+position p;
+@@
+
+ (T@p *)\(kmalloc\|kzalloc\|kcalloc\|kmem_cache_alloc\|kmem_cache_zalloc\|
+ kmem_cache_alloc_node\|kmalloc_node\|kzalloc_node\)(...)
+
+@script:python depends on org@
+p << r.p;
+t << r.T;
+@@
+
+coccilib.org.print_safe_todo(p[0], t)
+
+@script:python depends on report@
+p << r.p;
+t << r.T;
+@@
+
+msg="WARNING: casting value returned by memory allocation function to (%s *) is useless." % (t)
+coccilib.report.print_report(p[0], msg)
+
+
+++ /dev/null
-///
-/// Casting (void *) value returned by kmalloc is useless
-/// as mentioned in Documentation/CodingStyle, Chap 14.
-///
-// Confidence: High
-// Copyright: 2009,2010 Nicolas Palix, DIKU. GPLv2.
-// URL: http://coccinelle.lip6.fr/
-// Options: --no-includes --include-headers
-//
-// Keywords: kmalloc, kzalloc, kcalloc
-// Version min: < 2.6.12 kmalloc
-// Version min: < 2.6.12 kcalloc
-// Version min: 2.6.14 kzalloc
-//
-
-virtual context
-virtual patch
-virtual org
-virtual report
-
-//----------------------------------------------------------
-// For context mode
-//----------------------------------------------------------
-
-@depends on context@
-type T;
-@@
-
-* (T *)
- \(kmalloc\|kzalloc\|kcalloc\)(...)
-
-//----------------------------------------------------------
-// For patch mode
-//----------------------------------------------------------
-
-@depends on patch@
-type T;
-@@
-
-- (T *)
- \(kmalloc\|kzalloc\|kcalloc\)(...)
-
-//----------------------------------------------------------
-// For org and report mode
-//----------------------------------------------------------
-
-@r depends on org || report@
-type T;
-position p;
-@@
-
- (T@p *)\(kmalloc\|kzalloc\|kcalloc\)(...)
-
-@script:python depends on org@
-p << r.p;
-t << r.T;
-@@
-
-coccilib.org.print_safe_todo(p[0], t)
-
-@script:python depends on report@
-p << r.p;
-t << r.T;
-@@
-
-msg="WARNING: casting value returned by k[cmz]alloc to (%s *) is useless." % (t)
-coccilib.report.print_report(p[0], msg)
--- /dev/null
+/// Use ARRAY_SIZE instead of dividing sizeof array with sizeof an element
+///
+//# This makes an effort to find cases where ARRAY_SIZE can be used such as
+//# where there is a division of sizeof the array by the sizeof its first
+//# element or by any indexed element or the element type. It replaces the
+//# division of the two sizeofs by ARRAY_SIZE.
+//
+// Confidence: High
+// Copyright: (C) 2014 Himangi Saraogi. GPLv2.
+// Comments:
+// Options: --no-includes --include-headers
+
+virtual patch
+virtual context
+virtual org
+virtual report
+
+@i@
+@@
+
+#include <linux/kernel.h>
+
+//----------------------------------------------------------
+// For context mode
+//----------------------------------------------------------
+
+@depends on i&&context@
+type T;
+T[] E;
+@@
+(
+* (sizeof(E)/sizeof(*E))
+|
+* (sizeof(E)/sizeof(E[...]))
+|
+* (sizeof(E)/sizeof(T))
+)
+
+//----------------------------------------------------------
+// For patch mode
+//----------------------------------------------------------
+
+@depends on i&&patch@
+type T;
+T[] E;
+@@
+(
+- (sizeof(E)/sizeof(*E))
++ ARRAY_SIZE(E)
+|
+- (sizeof(E)/sizeof(E[...]))
++ ARRAY_SIZE(E)
+|
+- (sizeof(E)/sizeof(T))
++ ARRAY_SIZE(E)
+)
+
+//----------------------------------------------------------
+// For org and report mode
+//----------------------------------------------------------
+
+@r@
+type T;
+T[] E;
+position p;
+@@
+(
+ (sizeof(E)@p /sizeof(*E))
+|
+ (sizeof(E)@p /sizeof(E[...]))
+|
+ (sizeof(E)@p /sizeof(T))
+)
+
+@script:python depends on i&&org@
+p << r.p;
+@@
+
+coccilib.org.print_todo(p[0], "WARNING should use ARRAY_SIZE")
+
+@script:python depends on i&&report@
+p << r.p;
+@@
+
+msg="WARNING: Use ARRAY_SIZE"
+coccilib.report.print_report(p[0], msg)
+
--- /dev/null
+/// Use ARRAY_SIZE instead of dividing sizeof array with sizeof an element
+///
+//# This makes an effort to find cases where the argument to sizeof is wrong
+//# in memory allocation functions by checking the type of the allocated memory
+//# when it is a double pointer and ensuring the sizeof argument takes a pointer
+//# to the the memory being allocated. There are false positives in cases the
+//# sizeof argument is not used in constructing the return value. The result
+//# may need some reformatting.
+//
+// Confidence: Moderate
+// Copyright: (C) 2014 Himangi Saraogi. GPLv2.
+// Comments:
+// Options:
+
+virtual patch
+virtual context
+virtual org
+virtual report
+
+//----------------------------------------------------------
+// For context mode
+//----------------------------------------------------------
+
+@depends on context disable sizeof_type_expr@
+type T;
+T **x;
+@@
+
+ x =
+ <+...sizeof(
+* T
+ )...+>
+
+//----------------------------------------------------------
+// For patch mode
+//----------------------------------------------------------
+
+@depends on patch disable sizeof_type_expr@
+type T;
+T **x;
+@@
+
+ x =
+ <+...sizeof(
+- T
++ *x
+ )...+>
+
+//----------------------------------------------------------
+// For org and report mode
+//----------------------------------------------------------
+
+@r disable sizeof_type_expr@
+type T;
+T **x;
+position p;
+@@
+
+ x =
+ <+...sizeof(
+ T@p
+ )...+>
+
+@script:python depends on org@
+p << r.p;
+@@
+
+coccilib.org.print_todo(p[0], "WARNING sizeof argument should be pointer type, not structure type")
+
+@script:python depends on report@
+p << r.p;
+@@
+
+msg="WARNING: Use correct pointer type argument for sizeof"
+coccilib.report.print_report(p[0], msg)
+
--- /dev/null
+/// Use BUG_ON instead of a if condition followed by BUG.
+///
+//# This makes an effort to find cases where BUG() follows an if
+//# condition on an expression and replaces the if condition and BUG()
+//# with a BUG_ON having the conditional expression of the if statement
+//# as argument.
+//
+// Confidence: High
+// Copyright: (C) 2014 Himangi Saraogi. GPLv2.
+// Comments:
+// Options: --no-includes, --include-headers
+
+virtual patch
+virtual context
+virtual org
+virtual report
+
+//----------------------------------------------------------
+// For context mode
+//----------------------------------------------------------
+
+@depends on context@
+expression e;
+@@
+
+*if (e) BUG();
+
+//----------------------------------------------------------
+// For patch mode
+//----------------------------------------------------------
+
+@depends on patch@
+expression e;
+@@
+
+-if (e) BUG();
++BUG_ON(e);
+
+//----------------------------------------------------------
+// For org and report mode
+//----------------------------------------------------------
+
+@r@
+expression e;
+position p;
+@@
+
+ if (e) BUG@p ();
+
+@script:python depends on org@
+p << r.p;
+@@
+
+coccilib.org.print_todo(p[0], "WARNING use BUG_ON")
+
+@script:python depends on report@
+p << r.p;
+@@
+
+msg="WARNING: Use BUG_ON"
+coccilib.report.print_report(p[0], msg)
+
// Copyright: (C) 2012 Julia Lawall, INRIA/LIP6. GPLv2.
// Copyright: (C) 2012 Gilles Muller, INRIA/LiP6. GPLv2.
// URL: http://coccinelle.lip6.fr/
-// Comments:
+// Comments: Requires Coccinelle version 1.0.0-rc20 or later
// Options:
virtual patch
virtual report
@initialize:ocaml@
+@@
let negtable = Hashtbl.create 101
@depends on patch@
* 5530 only. The 5510/5520 decode is different.
*/
-static DEFINE_PCI_DEVICE_TABLE(id_tbl) = {
+static const struct pci_device_id id_tbl[] = {
{ PCI_VDEVICE(CYRIX, PCI_DEVICE_ID_CYRIX_5530_AUDIO), 0 },
{ }
};
#include "mpu401.h"
-typedef struct uart401_devc
+struct uart401_devc
{
int base;
int irq;
int my_dev;
int share_irq;
spinlock_t lock;
-}
-uart401_devc;
+};
#define DATAPORT (devc->base)
#define COMDPORT (devc->base+1)
#define STATPORT (devc->base+1)
-static int uart401_status(uart401_devc * devc)
+static int uart401_status(struct uart401_devc *devc)
{
return inb(STATPORT);
}
#define input_avail(devc) (!(uart401_status(devc)&INPUT_AVAIL))
#define output_ready(devc) (!(uart401_status(devc)&OUTPUT_READY))
-static void uart401_cmd(uart401_devc * devc, unsigned char cmd)
+static void uart401_cmd(struct uart401_devc *devc, unsigned char cmd)
{
outb((cmd), COMDPORT);
}
-static int uart401_read(uart401_devc * devc)
+static int uart401_read(struct uart401_devc *devc)
{
return inb(DATAPORT);
}
-static void uart401_write(uart401_devc * devc, unsigned char byte)
+static void uart401_write(struct uart401_devc *devc, unsigned char byte)
{
outb((byte), DATAPORT);
}
#define MPU_RESET 0xFF
#define UART_MODE_ON 0x3F
-static int reset_uart401(uart401_devc * devc);
-static void enter_uart_mode(uart401_devc * devc);
+static int reset_uart401(struct uart401_devc *devc);
+static void enter_uart_mode(struct uart401_devc *devc);
-static void uart401_input_loop(uart401_devc * devc)
+static void uart401_input_loop(struct uart401_devc *devc)
{
int work_limit=30000;
irqreturn_t uart401intr(int irq, void *dev_id)
{
- uart401_devc *devc = dev_id;
+ struct uart401_devc *devc = dev_id;
if (devc == NULL)
{
void (*output) (int dev)
)
{
- uart401_devc *devc = (uart401_devc *) midi_devs[dev]->devc;
+ struct uart401_devc *devc = (struct uart401_devc *)
+ midi_devs[dev]->devc;
if (devc->opened)
return -EBUSY;
static void uart401_close(int dev)
{
- uart401_devc *devc = (uart401_devc *) midi_devs[dev]->devc;
+ struct uart401_devc *devc = (struct uart401_devc *)
+ midi_devs[dev]->devc;
reset_uart401(devc);
devc->opened = 0;
{
int timeout;
unsigned long flags;
- uart401_devc *devc = (uart401_devc *) midi_devs[dev]->devc;
+ struct uart401_devc *devc = (struct uart401_devc *)
+ midi_devs[dev]->devc;
if (devc->disabled)
return 1;
.buffer_status = uart401_buffer_status,
};
-static void enter_uart_mode(uart401_devc * devc)
+static void enter_uart_mode(struct uart401_devc *devc)
{
int ok, timeout;
unsigned long flags;
spin_unlock_irqrestore(&devc->lock,flags);
}
-static int reset_uart401(uart401_devc * devc)
+static int reset_uart401(struct uart401_devc *devc)
{
int ok, timeout, n;
int probe_uart401(struct address_info *hw_config, struct module *owner)
{
- uart401_devc *devc;
+ struct uart401_devc *devc;
char *name = "MPU-401 (UART) MIDI";
int ok = 0;
unsigned long flags;
return 0;
}
- devc = kmalloc(sizeof(uart401_devc), GFP_KERNEL);
+ devc = kmalloc(sizeof(struct uart401_devc), GFP_KERNEL);
if (!devc) {
printk(KERN_WARNING "uart401: Can't allocate memory\n");
goto cleanup_region;
void unload_uart401(struct address_info *hw_config)
{
- uart401_devc *devc;
+ struct uart401_devc *devc;
int n=hw_config->slots[4];
/* Not set up */
0x0000 /* Monitor */
};
-typedef struct {
+struct wavnc_info {
struct address_info hw; /* hardware */
char *chip_name;
unsigned int line_mute_state :1;/* set by ioctl or autoselect */
unsigned int use_slider :1;/* use slider setting for o/p vol */
#endif
-} wavnc_info;
+};
/*
* This is the implementation specific mixer information.
unsigned int recording_devs; /* Recordable devies */
unsigned int stereo_devs; /* Stereo devices */
- unsigned int (*select_input)(wavnc_info *, unsigned int,
+ unsigned int (*select_input)(struct wavnc_info *, unsigned int,
unsigned char *, unsigned char *);
- int (*decode_mixer)(wavnc_info *, int,
+ int (*decode_mixer)(struct wavnc_info *, int,
unsigned char, unsigned char);
- int (*get_mixer)(wavnc_info *, int);
+ int (*get_mixer)(struct wavnc_info *, int);
};
-typedef struct wavnc_port_info {
+struct wavnc_port_info {
int open_mode;
int speed;
int channels;
int audio_format;
-} wavnc_port_info;
+};
static int nr_waveartist_devs;
-static wavnc_info adev_info[MAX_AUDIO_DEV];
+static struct wavnc_info adev_info[MAX_AUDIO_DEV];
static DEFINE_SPINLOCK(waveartist_lock);
#ifndef CONFIG_ARCH_NETWINDER
#define machine_is_netwinder() 0
#else
static struct timer_list vnc_timer;
-static void vnc_configure_mixer(wavnc_info *devc, unsigned int input_mask);
+static void vnc_configure_mixer(struct wavnc_info *devc,
+ unsigned int input_mask);
static int vnc_private_ioctl(int dev, unsigned int cmd, int __user *arg);
static void vnc_slider_tick(unsigned long data);
#endif
/* Toggle IRQ acknowledge line
*/
static inline void
-waveartist_iack(wavnc_info *devc)
+waveartist_iack(struct wavnc_info *devc)
{
unsigned int ctlr_port = devc->hw.io_base + CTLR;
int old_ctlr;
}
static int
-waveartist_reset(wavnc_info *devc)
+waveartist_reset(struct wavnc_info *devc)
{
struct address_info *hw = &devc->hw;
unsigned int timeout, res = -1;
* and can send or receive multiple words.
*/
static int
-waveartist_cmd(wavnc_info *devc,
+waveartist_cmd(struct wavnc_info *devc,
int nr_cmd, unsigned int *cmd,
int nr_resp, unsigned int *resp)
{
* Send one command word
*/
static inline int
-waveartist_cmd1(wavnc_info *devc, unsigned int cmd)
+waveartist_cmd1(struct wavnc_info *devc, unsigned int cmd)
{
return waveartist_cmd(devc, 1, &cmd, 0, NULL);
}
* Send one command, receive one word
*/
static inline unsigned int
-waveartist_cmd1_r(wavnc_info *devc, unsigned int cmd)
+waveartist_cmd1_r(struct wavnc_info *devc, unsigned int cmd)
{
unsigned int ret;
* word (and throw it away)
*/
static inline int
-waveartist_cmd2(wavnc_info *devc, unsigned int cmd, unsigned int arg)
+waveartist_cmd2(struct wavnc_info *devc, unsigned int cmd, unsigned int arg)
{
unsigned int vals[2];
* Send a triple command
*/
static inline int
-waveartist_cmd3(wavnc_info *devc, unsigned int cmd,
+waveartist_cmd3(struct wavnc_info *devc, unsigned int cmd,
unsigned int arg1, unsigned int arg2)
{
unsigned int vals[3];
}
static int
-waveartist_getrev(wavnc_info *devc, char *rev)
+waveartist_getrev(struct wavnc_info *devc, char *rev)
{
unsigned int temp[2];
unsigned int cmd = WACMD_GETREV;
static int
waveartist_open(int dev, int mode)
{
- wavnc_info *devc;
- wavnc_port_info *portc;
+ struct wavnc_info *devc;
+ struct wavnc_port_info *portc;
unsigned long flags;
if (dev < 0 || dev >= num_audiodevs)
return -ENXIO;
- devc = (wavnc_info *) audio_devs[dev]->devc;
- portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ devc = (struct wavnc_info *) audio_devs[dev]->devc;
+ portc = (struct wavnc_port_info *) audio_devs[dev]->portc;
spin_lock_irqsave(&waveartist_lock, flags);
if (portc->open_mode || (devc->open_mode & mode)) {
static void
waveartist_close(int dev)
{
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
unsigned long flags;
spin_lock_irqsave(&waveartist_lock, flags);
static void
waveartist_output_block(int dev, unsigned long buf, int __count, int intrflag)
{
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
unsigned long flags;
unsigned int count = __count;
static void
waveartist_start_input(int dev, unsigned long buf, int __count, int intrflag)
{
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
unsigned long flags;
unsigned int count = __count;
}
static unsigned int
-waveartist_get_speed(wavnc_port_info *portc)
+waveartist_get_speed(struct wavnc_port_info *portc)
{
unsigned int speed;
}
static unsigned int
-waveartist_get_bits(wavnc_port_info *portc)
+waveartist_get_bits(struct wavnc_port_info *portc)
{
unsigned int bits;
waveartist_prepare_for_input(int dev, int bsize, int bcount)
{
unsigned long flags;
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
unsigned int speed, bits;
if (devc->audio_mode)
waveartist_prepare_for_output(int dev, int bsize, int bcount)
{
unsigned long flags;
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
unsigned int speed, bits;
/*
static void
waveartist_halt(int dev)
{
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
- wavnc_info *devc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
+ struct wavnc_info *devc;
if (portc->open_mode & OPEN_WRITE)
waveartist_halt_output(dev);
if (portc->open_mode & OPEN_READ)
waveartist_halt_input(dev);
- devc = (wavnc_info *) audio_devs[dev]->devc;
+ devc = (struct wavnc_info *) audio_devs[dev]->devc;
devc->audio_mode = 0;
}
static void
waveartist_halt_input(int dev)
{
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
unsigned long flags;
spin_lock_irqsave(&waveartist_lock, flags);
static void
waveartist_halt_output(int dev)
{
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
unsigned long flags;
spin_lock_irqsave(&waveartist_lock, flags);
static void
waveartist_trigger(int dev, int state)
{
- wavnc_info *devc = (wavnc_info *) audio_devs[dev]->devc;
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ struct wavnc_info *devc = (struct wavnc_info *)
+ audio_devs[dev]->devc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
unsigned long flags;
if (debug_flg & DEBUG_TRIGGER) {
static int
waveartist_set_speed(int dev, int arg)
{
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
if (arg <= 0)
return portc->speed;
static short
waveartist_set_channels(int dev, short arg)
{
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
if (arg != 1 && arg != 2)
return portc->channels;
static unsigned int
waveartist_set_bits(int dev, unsigned int arg)
{
- wavnc_port_info *portc = (wavnc_port_info *) audio_devs[dev]->portc;
+ struct wavnc_port_info *portc = (struct wavnc_port_info *)
+ audio_devs[dev]->portc;
if (arg == 0)
return portc->audio_format;
static irqreturn_t
waveartist_intr(int irq, void *dev_id)
{
- wavnc_info *devc = dev_id;
+ struct wavnc_info *devc = dev_id;
int irqstatus, status;
spin_lock(&waveartist_lock);
};
static void
-waveartist_mixer_update(wavnc_info *devc, int whichDev)
+waveartist_mixer_update(struct wavnc_info *devc, int whichDev)
{
unsigned int lev_left, lev_right;
* relevant *_select_input function has done that for us.
*/
static void
-waveartist_set_adc_mux(wavnc_info *devc, char left_dev, char right_dev)
+waveartist_set_adc_mux(struct wavnc_info *devc, char left_dev,
+ char right_dev)
{
unsigned int reg_08, reg_09;
* SOUND_MASK_MIC Mic Microphone
*/
static unsigned int
-waveartist_select_input(wavnc_info *devc, unsigned int recmask,
+waveartist_select_input(struct wavnc_info *devc, unsigned int recmask,
unsigned char *dev_l, unsigned char *dev_r)
{
unsigned int recdev = ADC_MUX_NONE;
}
static int
-waveartist_decode_mixer(wavnc_info *devc, int dev, unsigned char lev_l,
+waveartist_decode_mixer(struct wavnc_info *devc, int dev,
+ unsigned char lev_l,
unsigned char lev_r)
{
switch (dev) {
return dev;
}
-static int waveartist_get_mixer(wavnc_info *devc, int dev)
+static int waveartist_get_mixer(struct wavnc_info *devc, int dev)
{
return devc->levels[dev];
}
};
static void
-waveartist_set_recmask(wavnc_info *devc, unsigned int recmask)
+waveartist_set_recmask(struct wavnc_info *devc, unsigned int recmask)
{
unsigned char dev_l, dev_r;
}
static int
-waveartist_set_mixer(wavnc_info *devc, int dev, unsigned int level)
+waveartist_set_mixer(struct wavnc_info *devc, int dev, unsigned int level)
{
unsigned int lev_left = level & 0x00ff;
unsigned int lev_right = (level & 0xff00) >> 8;
static int
waveartist_mixer_ioctl(int dev, unsigned int cmd, void __user * arg)
{
- wavnc_info *devc = (wavnc_info *)audio_devs[dev]->devc;
+ struct wavnc_info *devc = (struct wavnc_info *)audio_devs[dev]->devc;
int ret = 0, val, nr;
/*
};
static void
-waveartist_mixer_reset(wavnc_info *devc)
+waveartist_mixer_reset(struct wavnc_info *devc)
{
int i;
waveartist_mixer_update(devc, i);
}
-static int __init waveartist_init(wavnc_info *devc)
+static int __init waveartist_init(struct wavnc_info *devc)
{
- wavnc_port_info *portc;
+ struct wavnc_port_info *portc;
char rev[3], dev_name[64];
int my_dev;
conf_printf2(dev_name, devc->hw.io_base, devc->hw.irq,
devc->hw.dma, devc->hw.dma2);
- portc = kzalloc(sizeof(wavnc_port_info), GFP_KERNEL);
+ portc = kzalloc(sizeof(struct wavnc_port_info), GFP_KERNEL);
if (portc == NULL)
goto nomem;
static int __init probe_waveartist(struct address_info *hw_config)
{
- wavnc_info *devc = &adev_info[nr_waveartist_devs];
+ struct wavnc_info *devc = &adev_info[nr_waveartist_devs];
if (nr_waveartist_devs >= MAX_AUDIO_DEV) {
printk(KERN_WARNING "waveartist: too many audio devices\n");
static void __init
attach_waveartist(struct address_info *hw, const struct waveartist_mixer_info *mix)
{
- wavnc_info *devc = &adev_info[nr_waveartist_devs];
+ struct wavnc_info *devc = &adev_info[nr_waveartist_devs];
/*
* NOTE! If irq < 0, there is another driver which has allocated the
static void __exit unload_waveartist(struct address_info *hw)
{
- wavnc_info *devc = NULL;
+ struct wavnc_info *devc = NULL;
int i;
for (i = 0; i < nr_waveartist_devs; i++)
#define VNC_DISABLE_AUTOSWITCH 0x80
static inline void
-vnc_mute_spkr(wavnc_info *devc)
+vnc_mute_spkr(struct wavnc_info *devc)
{
unsigned long flags;
}
static void
-vnc_mute_lout(wavnc_info *devc)
+vnc_mute_lout(struct wavnc_info *devc)
{
unsigned int left, right;
}
static int
-vnc_volume_slider(wavnc_info *devc)
+vnc_volume_slider(struct wavnc_info *devc)
{
static signed int old_slider_volume;
unsigned long flags;
* SOUND_MASK_MIC Right Mic Builtin microphone
*/
static unsigned int
-netwinder_select_input(wavnc_info *devc, unsigned int recmask,
+netwinder_select_input(struct wavnc_info *devc, unsigned int recmask,
unsigned char *dev_l, unsigned char *dev_r)
{
unsigned int recdev_l = ADC_MUX_NONE, recdev_r = ADC_MUX_NONE;
}
static int
-netwinder_decode_mixer(wavnc_info *devc, int dev, unsigned char lev_l,
+netwinder_decode_mixer(struct wavnc_info *devc, int dev, unsigned char lev_l,
unsigned char lev_r)
{
switch (dev) {
return dev;
}
-static int netwinder_get_mixer(wavnc_info *devc, int dev)
+static int netwinder_get_mixer(struct wavnc_info *devc, int dev)
{
int levels;
};
static void
-vnc_configure_mixer(wavnc_info *devc, unsigned int recmask)
+vnc_configure_mixer(struct wavnc_info *devc, unsigned int recmask)
{
if (!devc->no_autoselect) {
if (devc->handset_detect) {
}
static int
-vnc_slider(wavnc_info *devc)
+vnc_slider(struct wavnc_info *devc)
{
signed int slider_volume;
unsigned int temp, old_hs, old_td;
static int
vnc_private_ioctl(int dev, unsigned int cmd, int __user * arg)
{
- wavnc_info *devc = (wavnc_info *)audio_devs[dev]->devc;
+ struct wavnc_info *devc = (struct wavnc_info *)audio_devs[dev]->devc;
int val;
switch (cmd) {
snd_card_free(pci_get_drvdata(pci));
}
-static DEFINE_PCI_DEVICE_TABLE(snd_ad1889_ids) = {
+static const struct pci_device_id snd_ad1889_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_ANALOG_DEVICES, PCI_DEVICE_ID_AD1889JS) },
{ 0, },
};
#endif
};
-static DEFINE_PCI_DEVICE_TABLE(snd_ali_ids) = {
+static const struct pci_device_id snd_ali_ids[] = {
{PCI_DEVICE(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M5451), 0, 0, 0},
{0, }
};
int block_counter_register;
};
-static DEFINE_PCI_DEVICE_TABLE(snd_als300_ids) = {
+static const struct pci_device_id snd_als300_ids[] = {
{ 0x4005, 0x0300, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_ALS300 },
{ 0x4005, 0x0308, PCI_ANY_ID, PCI_ANY_ID, 0, 0, DEVICE_ALS300_PLUS },
{ 0, }
#endif
};
-static DEFINE_PCI_DEVICE_TABLE(snd_als4000_ids) = {
+static const struct pci_device_id snd_als4000_ids[] = {
{ 0x4005, 0x4000, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, /* ALS4000 */
{ 0, }
};
asihpi_adapter_remove(pci_dev);
}
-static DEFINE_PCI_DEVICE_TABLE(asihpi_pci_tbl) = {
+static const struct pci_device_id asihpi_pci_tbl[] = {
{HPI_PCI_VENDOR_ID_TI, HPI_PCI_DEV_ID_DSP6205,
HPI_PCI_VENDOR_ID_AUDIOSCIENCE, PCI_ANY_ID, 0, 0,
(kernel_ulong_t)HPI_6205},
/*
*/
-static DEFINE_PCI_DEVICE_TABLE(snd_atiixp_ids) = {
+static const struct pci_device_id snd_atiixp_ids[] = {
{ PCI_VDEVICE(ATI, 0x4341), 0 }, /* SB200 */
{ PCI_VDEVICE(ATI, 0x4361), 0 }, /* SB300 */
{ PCI_VDEVICE(ATI, 0x4370), 0 }, /* SB400 */
/*
*/
-static DEFINE_PCI_DEVICE_TABLE(snd_atiixp_ids) = {
+static const struct pci_device_id snd_atiixp_ids[] = {
{ PCI_VDEVICE(ATI, 0x434d), 0 }, /* SB200 */
{ PCI_VDEVICE(ATI, 0x4378), 0 }, /* SB400 */
{ 0, }
#include "au8810.h"
#include "au88x0.h"
-static DEFINE_PCI_DEVICE_TABLE(snd_vortex_ids) = {
+static const struct pci_device_id snd_vortex_ids[] = {
{PCI_VDEVICE(AUREAL, PCI_DEVICE_ID_AUREAL_ADVANTAGE), 1,},
{0,}
};
#include "au8820.h"
#include "au88x0.h"
-static DEFINE_PCI_DEVICE_TABLE(snd_vortex_ids) = {
+static const struct pci_device_id snd_vortex_ids[] = {
{PCI_VDEVICE(AUREAL, PCI_DEVICE_ID_AUREAL_VORTEX_1), 0,},
{0,}
};
#include "au8830.h"
#include "au88x0.h"
-static DEFINE_PCI_DEVICE_TABLE(snd_vortex_ids) = {
+static const struct pci_device_id snd_vortex_ids[] = {
{PCI_VDEVICE(AUREAL, PCI_DEVICE_ID_AUREAL_VORTEX_2), 0,},
{0,}
};
module_param_array(enable, bool, NULL, 0444);
MODULE_PARM_DESC(enable, "Enable Audiowerk2 soundcard.");
-static DEFINE_PCI_DEVICE_TABLE(snd_aw2_ids) = {
+static const struct pci_device_id snd_aw2_ids[] = {
{PCI_VENDOR_ID_PHILIPS, PCI_DEVICE_ID_PHILIPS_SAA7146, 0, 0,
0, 0, 0},
{0}
#endif
};
-static DEFINE_PCI_DEVICE_TABLE(snd_azf3328_ids) = {
+static const struct pci_device_id snd_azf3328_ids[] = {
{ 0x122D, 0x50DC, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* PCI168/3328 */
{ 0x122D, 0x80DA, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* 3328 */
{ 0, }
.driver_data = SND_BT87X_BOARD_ ## id }
/* driver_data is the card id for that device */
-static DEFINE_PCI_DEVICE_TABLE(snd_bt87x_ids) = {
+static const struct pci_device_id snd_bt87x_ids[] = {
/* Hauppauge WinTV series */
BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_878, 0x0070, 0x13eb, GENERIC),
/* Hauppauge WinTV series */
/* default entries for all Bt87x cards - it's not exported */
/* driver_data is set to 0 to call detection */
-static DEFINE_PCI_DEVICE_TABLE(snd_bt87x_default_ids) = {
+static const struct pci_device_id snd_bt87x_default_ids[] = {
BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_878, PCI_ANY_ID, PCI_ANY_ID, UNKNOWN),
BT_DEVICE(PCI_DEVICE_ID_BROOKTREE_879, PCI_ANY_ID, PCI_ANY_ID, UNKNOWN),
{ }
#endif
// PCI IDs
-static DEFINE_PCI_DEVICE_TABLE(snd_ca0106_ids) = {
+static const struct pci_device_id snd_ca0106_ids[] = {
{ PCI_VDEVICE(CREATIVE, 0x0007), 0 }, /* Audigy LS or Live 24bit */
{ 0, }
};
#endif
-static DEFINE_PCI_DEVICE_TABLE(snd_cmipci_ids) = {
+static const struct pci_device_id snd_cmipci_ids[] = {
{PCI_VDEVICE(CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8338A), 0},
{PCI_VDEVICE(CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8338B), 0},
{PCI_VDEVICE(CMEDIA, PCI_DEVICE_ID_CMEDIA_CM8738), 0},
int integrated_midi = 0;
char modelstr[16];
int pcm_index, pcm_spdif_index;
- static DEFINE_PCI_DEVICE_TABLE(intel_82437vx) = {
+ static const struct pci_device_id intel_82437vx[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82437VX) },
{ },
};
static irqreturn_t snd_cs4281_interrupt(int irq, void *dev_id);
-static DEFINE_PCI_DEVICE_TABLE(snd_cs4281_ids) = {
+static const struct pci_device_id snd_cs4281_ids[] = {
{ PCI_VDEVICE(CIRRUS, 0x6005), 0, }, /* CS4281 */
{ 0, }
};
module_param_array(mmap_valid, bool, NULL, 0444);
MODULE_PARM_DESC(mmap_valid, "Support OSS mmap.");
-static DEFINE_PCI_DEVICE_TABLE(snd_cs46xx_ids) = {
+static const struct pci_device_id snd_cs46xx_ids[] = {
{ PCI_VDEVICE(CIRRUS, 0x6001), 0, }, /* CS4280 */
{ PCI_VDEVICE(CIRRUS, 0x6003), 0, }, /* CS4612 */
{ PCI_VDEVICE(CIRRUS, 0x6004), 0, }, /* CS4615 */
unsigned long pci_base;
};
-static DEFINE_PCI_DEVICE_TABLE(snd_cs5530_ids) = {
+static const struct pci_device_id snd_cs5530_ids[] = {
{PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_AUDIO, PCI_ANY_ID,
PCI_ANY_ID, 0, 0},
{0,}
module_param_array(enable, bool, NULL, 0444);
MODULE_PARM_DESC(enable, "Enable " DRIVER_NAME);
-static DEFINE_PCI_DEVICE_TABLE(snd_cs5535audio_ids) = {
+static const struct pci_device_id snd_cs5535audio_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_NS, PCI_DEVICE_ID_NS_CS5535_AUDIO) },
{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_CS5536_AUDIO) },
{}
module_param_array(subsystem, int, NULL, 0444);
MODULE_PARM_DESC(subsystem, "Override subsystem ID for Creative X-Fi driver");
-static DEFINE_PCI_DEVICE_TABLE(ct_pci_dev_ids) = {
+static const struct pci_device_id ct_pci_dev_ids[] = {
/* only X-Fi is supported, so... */
{ PCI_DEVICE(PCI_VENDOR_ID_CREATIVE, PCI_DEVICE_ID_CREATIVE_20K1),
.driver_data = ATC20K1,
{0, "darla20_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x1801, 0xECC0, 0x0010, 0, 0, 0}, /* DSP 56301 Darla20 rev.0 */
{0,}
};
{0, "darla24_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x1801, 0xECC0, 0x0040, 0, 0, 0}, /* DSP 56301 Darla24 rev.0 */
{0x1057, 0x1801, 0xECC0, 0x0041, 0, 0, 0}, /* DSP 56301 Darla24 rev.1 */
{0,}
{0, "3g_asic.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x0100, 0, 0, 0}, /* Echo 3G */
{0,}
};
{0, "gina20_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x1801, 0xECC0, 0x0020, 0, 0, 0}, /* DSP 56301 Gina20 rev.0 */
{0,}
};
{0, "gina24_361_asic.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x1801, 0xECC0, 0x0050, 0, 0, 0}, /* DSP 56301 Gina24 rev.0 */
{0x1057, 0x1801, 0xECC0, 0x0051, 0, 0, 0}, /* DSP 56301 Gina24 rev.1 */
{0x1057, 0x3410, 0xECC0, 0x0050, 0, 0, 0}, /* DSP 56361 Gina24 rev.0 */
{0, "indigo_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x0090, 0, 0, 0}, /* Indigo */
{0,}
};
{0, "indigo_dj_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x00B0, 0, 0, 0}, /* Indigo DJ*/
{0,}
};
{0, "indigo_djx_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x00E0, 0, 0, 0}, /* Indigo DJx*/
{0,}
};
{0, "indigo_io_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x00A0, 0, 0, 0}, /* Indigo IO*/
{0,}
};
{0, "indigo_iox_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x00D0, 0, 0, 0}, /* Indigo IOx */
{0,}
};
{0, "layla20_asic.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x1801, 0xECC0, 0x0030, 0, 0, 0}, /* DSP 56301 Layla20 rev.0 */
{0x1057, 0x1801, 0xECC0, 0x0031, 0, 0, 0}, /* DSP 56301 Layla20 rev.1 */
{0,}
{0, "layla24_2S_asic.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x0060, 0, 0, 0}, /* DSP 56361 Layla24 rev.0 */
{0,}
};
{0, "mia_dsp.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x3410, 0xECC0, 0x0080, 0, 0, 0}, /* DSP 56361 Mia rev.0 */
{0x1057, 0x3410, 0xECC0, 0x0081, 0, 0, 0}, /* DSP 56361 Mia rev.1 */
{0,}
{0, "mona_2_asic.fw"}
};
-static DEFINE_PCI_DEVICE_TABLE(snd_echo_ids) = {
+static const struct pci_device_id snd_echo_ids[] = {
{0x1057, 0x1801, 0xECC0, 0x0070, 0, 0, 0}, /* DSP 56301 Mona rev.0 */
{0x1057, 0x1801, 0xECC0, 0x0071, 0, 0, 0}, /* DSP 56301 Mona rev.1 */
{0x1057, 0x1801, 0xECC0, 0x0072, 0, 0, 0}, /* DSP 56301 Mona rev.2 */
/*
* Class 0401: 1102:0008 (rev 00) Subsystem: 1102:1001 -> Audigy2 Value Model:SB0400
*/
-static DEFINE_PCI_DEVICE_TABLE(snd_emu10k1_ids) = {
+static const struct pci_device_id snd_emu10k1_ids[] = {
{ PCI_VDEVICE(CREATIVE, 0x0002), 0 }, /* EMU10K1 */
{ PCI_VDEVICE(CREATIVE, 0x0004), 1 }, /* Audigy */
{ PCI_VDEVICE(CREATIVE, 0x0008), 1 }, /* Audigy 2 Value SB0400 */
}
// PCI IDs
-static DEFINE_PCI_DEVICE_TABLE(snd_emu10k1x_ids) = {
+static const struct pci_device_id snd_emu10k1x_ids[] = {
{ PCI_VDEVICE(CREATIVE, 0x0006), 0 }, /* Dell OEM version (EMU10K1) */
{ 0, }
};
static irqreturn_t snd_audiopci_interrupt(int irq, void *dev_id);
-static DEFINE_PCI_DEVICE_TABLE(snd_audiopci_ids) = {
+static const struct pci_device_id snd_audiopci_ids[] = {
#ifdef CHIP1370
{ PCI_VDEVICE(ENSONIQ, 0x5000), 0, }, /* ES1370 */
#endif
static irqreturn_t snd_es1938_interrupt(int irq, void *dev_id);
-static DEFINE_PCI_DEVICE_TABLE(snd_es1938_ids) = {
+static const struct pci_device_id snd_es1938_ids[] = {
{ PCI_VDEVICE(ESS, 0x1969), 0, }, /* Solo-1 */
{ 0, }
};
static irqreturn_t snd_es1968_interrupt(int irq, void *dev_id);
-static DEFINE_PCI_DEVICE_TABLE(snd_es1968_ids) = {
+static const struct pci_device_id snd_es1968_ids[] = {
/* Maestro 1 */
{ 0x1285, 0x0100, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, TYPE_MAESTRO },
/* Maestro 2 */
#endif
};
-static DEFINE_PCI_DEVICE_TABLE(snd_fm801_ids) = {
+static const struct pci_device_id snd_fm801_ids[] = {
{ 0x1319, 0x0801, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, 0, }, /* FM801 */
{ 0x5213, 0x0510, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, 0, }, /* Gallant Odyssey Sound 4 */
{ 0, }
AZX_DRIVER_TERA,
AZX_DRIVER_CTX,
AZX_DRIVER_CTHDA,
+ AZX_DRIVER_CMEDIA,
AZX_DRIVER_GENERIC,
AZX_NUM_DRIVERS, /* keep this as last entry */
};
[AZX_DRIVER_TERA] = "HDA Teradici",
[AZX_DRIVER_CTX] = "HDA Creative",
[AZX_DRIVER_CTHDA] = "HDA Creative",
+ [AZX_DRIVER_CMEDIA] = "HDA C-Media",
[AZX_DRIVER_GENERIC] = "HD-Audio Generic",
};
snoop = false;
break;
case AZX_DRIVER_CTHDA:
+ case AZX_DRIVER_CMEDIA:
snoop = false;
break;
}
.driver_data = AZX_DRIVER_CTX | AZX_DCAPS_CTX_WORKAROUND |
AZX_DCAPS_RIRB_PRE_DELAY | AZX_DCAPS_POSFIX_LPIB },
#endif
+ /* CM8888 */
+ { PCI_DEVICE(0x13f6, 0x5011),
+ .driver_data = AZX_DRIVER_CMEDIA |
+ AZX_DCAPS_NO_MSI | AZX_DCAPS_POSFIX_LPIB },
/* Vortex86MX */
{ PCI_DEVICE(0x17f3, 0x3010), .driver_data = AZX_DRIVER_GENERIC },
/* VMware HDAudio */
return; /* NOP */
#endif
+ if (spec->dsp_state == DSP_DOWNLOAD_FAILED)
+ return; /* don't retry failures */
+
chipio_enable_clocks(codec);
spec->dsp_state = DSP_DOWNLOADING;
if (!ca0132_download_dsp_images(codec))
struct auto_pin_cfg *cfg = &spec->autocfg;
int i;
- spec->dsp_state = DSP_DOWNLOAD_INIT;
+ if (spec->dsp_state != DSP_DOWNLOAD_FAILED)
+ spec->dsp_state = DSP_DOWNLOAD_INIT;
spec->curr_chip_addx = INVALID_CHIP_ADDRESS;
snd_hda_power_up(codec);
codec->spec = spec;
spec->codec = codec;
+ spec->dsp_state = DSP_DOWNLOAD_INIT;
spec->num_mixers = 1;
spec->mixers[0] = ca0132_mixer;
return err;
}
+static int patch_cmi8888(struct hda_codec *codec)
+{
+ struct cmi_spec *spec;
+ struct auto_pin_cfg *cfg;
+ int err;
+
+ spec = kzalloc(sizeof(*spec), GFP_KERNEL);
+ if (!spec)
+ return -ENOMEM;
+
+ codec->spec = spec;
+ cfg = &spec->gen.autocfg;
+ snd_hda_gen_spec_init(&spec->gen);
+
+ /* mask NID 0x10 from the playback volume selection;
+ * it's a headphone boost volume handled manually below
+ */
+ spec->gen.out_vol_mask = (1ULL << 0x10);
+
+ err = snd_hda_parse_pin_defcfg(codec, cfg, NULL, 0);
+ if (err < 0)
+ goto error;
+ err = snd_hda_gen_parse_auto_config(codec, cfg);
+ if (err < 0)
+ goto error;
+
+ if (get_defcfg_device(snd_hda_codec_get_pincfg(codec, 0x10)) ==
+ AC_JACK_HP_OUT) {
+ static const struct snd_kcontrol_new amp_kctl =
+ HDA_CODEC_VOLUME("Headphone Amp Playback Volume",
+ 0x10, 0, HDA_OUTPUT);
+ if (!snd_hda_gen_add_kctl(&spec->gen, NULL, &_kctl)) {
+ err = -ENOMEM;
+ goto error;
+ }
+ }
+
+ codec->patch_ops = cmi_auto_patch_ops;
+ return 0;
+
+ error:
+ snd_hda_gen_free(codec);
+ return err;
+}
+
/*
* patch entries
*/
static const struct hda_codec_preset snd_hda_preset_cmedia[] = {
+ { .id = 0x13f68888, .name = "CMI8888", .patch = patch_cmi8888 },
{ .id = 0x13f69880, .name = "CMI9880", .patch = patch_cmi9880 },
{ .id = 0x434d4980, .name = "CMI9880", .patch = patch_cmi9880 },
{} /* terminator */
};
+MODULE_ALIAS("snd-hda-codec-id:13f68888");
MODULE_ALIAS("snd-hda-codec-id:13f69880");
MODULE_ALIAS("snd-hda-codec-id:434d4980");
#include <linux/module.h>
#include <sound/core.h>
#include <sound/jack.h>
+#include <sound/tlv.h>
#include "hda_codec.h"
#include "hda_local.h"
if (err < 0)
goto error;
+ if (codec->vendor_id == 0x14f15051) {
+ /* minimum value is actually mute */
+ spec->gen.vmaster_tlv[3] |= TLV_DB_SCALE_MUTE;
+ }
+
codec->patch_ops = cx_auto_patch_ops;
/* Some laptops with Conexant chips show stalls in S3 resume,
#define is_haswell_plus(codec) (is_haswell(codec) || is_broadwell(codec))
#define is_valleyview(codec) ((codec)->vendor_id == 0x80862882)
+#define is_cherryview(codec) ((codec)->vendor_id == 0x80862883)
+#define is_valleyview_plus(codec) (is_valleyview(codec) || is_cherryview(codec))
struct hdmi_spec_per_cvt {
hda_nid_t cvt_nid;
mux_idx);
/* configure unused pins to choose other converters */
- if (is_haswell_plus(codec) || is_valleyview(codec))
+ if (is_haswell_plus(codec) || is_valleyview_plus(codec))
intel_not_share_assigned_cvt(codec, per_pin->pin_nid, mux_idx);
snd_hda_spdif_ctls_assign(codec, pin_idx, per_cvt->cvt_nid);
* and this can make HW reset converter selection on a pin.
*/
if (eld->eld_valid && !old_eld_valid && per_pin->setup) {
- if (is_haswell_plus(codec) || is_valleyview(codec)) {
+ if (is_haswell_plus(codec) ||
+ is_valleyview_plus(codec)) {
intel_verify_pin_cvt_connect(codec, per_pin);
intel_not_share_assigned_cvt(codec, pin_nid,
per_pin->mux_idx);
bool non_pcm;
int pinctl;
- if (is_haswell_plus(codec) || is_valleyview(codec)) {
+ if (is_haswell_plus(codec) || is_valleyview_plus(codec)) {
/* Verify pin:cvt selections to avoid silent audio after S3.
* After S3, the audio driver restores pin:cvt selections
* but this can happen before gfx is ready and such selection
intel_haswell_fixup_enable_dp12(codec);
}
- if (is_haswell(codec) || is_valleyview(codec)) {
+ if (is_haswell_plus(codec) || is_valleyview_plus(codec))
codec->depop_delay = 0;
- }
if (hdmi_parse_codec(codec) < 0) {
codec->spec = NULL;
spec->pll_coef_idx);
val = snd_hda_codec_read(codec, spec->pll_nid, 0,
AC_VERB_GET_PROC_COEF, 0);
+ if (val == -1)
+ return;
snd_hda_codec_write(codec, spec->pll_nid, 0, AC_VERB_SET_COEF_INDEX,
spec->pll_coef_idx);
snd_hda_codec_write(codec, spec->pll_nid, 0, AC_VERB_SET_PROC_COEF,
return alc_parse_auto_config(codec, alc269_ignore, ssids);
}
+static int find_ext_mic_pin(struct hda_codec *codec);
+
+static void alc286_shutup(struct hda_codec *codec)
+{
+ int i;
+ int mic_pin = find_ext_mic_pin(codec);
+ /* don't shut up pins when unloading the driver; otherwise it breaks
+ * the default pin setup at the next load of the driver
+ */
+ if (codec->bus->shutdown)
+ return;
+ for (i = 0; i < codec->init_pins.used; i++) {
+ struct hda_pincfg *pin = snd_array_elem(&codec->init_pins, i);
+ /* use read here for syncing after issuing each verb */
+ if (pin->nid != mic_pin)
+ snd_hda_codec_read(codec, pin->nid, 0,
+ AC_VERB_SET_PIN_WIDGET_CONTROL, 0);
+ }
+ codec->pins_shutup = 1;
+}
+
static void alc269vb_toggle_power_output(struct hda_codec *codec, int power_up)
{
int val = alc_read_coef_idx(codec, 0x04);
+ if (val == -1)
+ return;
if (power_up)
val |= 1 << 11;
else
snd_hda_codec_resume_cache(codec);
alc_inv_dmic_sync(codec, true);
hda_call_check_power_status(codec, 0x01);
+
+ /* on some machine, the BIOS will clear the codec gpio data when enter
+ * suspend, and won't restore the data after resume, so we restore it
+ * in the driver.
+ */
+ if (spec->gpio_led)
+ snd_hda_codec_write(codec, codec->afg, 0, AC_VERB_SET_GPIO_DATA,
+ spec->gpio_led);
+
if (spec->has_alc5505_dsp)
alc5505_dsp_resume(codec);
/* Avoid pop noises when headphones are plugged in */
if (spec->gen.hp_jack_present)
- if (nid == codec->afg || nid == 0x02)
+ if (nid == codec->afg || nid == 0x02 || nid == 0x15)
return AC_PWRST_D0;
return power_state;
}
{
if (action == HDA_FIXUP_ACT_PROBE) {
struct alc_spec *spec = codec->spec;
+ struct hda_input_mux *imux = &spec->gen.input_mux;
+ int i;
+
spec->shutup = alc_no_shutup;
codec->power_filter = alc_power_filter_xps13;
+
+ /* Make the internal mic the default input source. */
+ for (i = 0; i < imux->num_items; i++) {
+ if (spec->gen.imux_pins[i] == 0x12) {
+ spec->gen.cur_mux[0] = i;
+ break;
+ }
+ }
}
}
if ((alc_get_coef0(codec) & 0x00ff) == 0x017) {
val = alc_read_coef_idx(codec, 0x04);
/* Power up output pin */
- alc_write_coef_idx(codec, 0x04, val | (1<<11));
+ if (val != -1)
+ alc_write_coef_idx(codec, 0x04, val | (1<<11));
}
if ((alc_get_coef0(codec) & 0x00ff) == 0x018) {
val = alc_read_coef_idx(codec, 0xd);
- if ((val & 0x0c00) >> 10 != 0x1) {
+ if (val != -1 && (val & 0x0c00) >> 10 != 0x1) {
/* Capless ramp up clock control */
alc_write_coef_idx(codec, 0xd, val | (1<<10));
}
val = alc_read_coef_idx(codec, 0x17);
- if ((val & 0x01c0) >> 6 != 0x4) {
+ if (val != -1 && (val & 0x01c0) >> 6 != 0x4) {
/* Class D power on reset */
alc_write_coef_idx(codec, 0x17, val | (1<<7));
}
}
val = alc_read_coef_idx(codec, 0xd); /* Class D */
- alc_write_coef_idx(codec, 0xd, val | (1<<14));
+ if (val != -1)
+ alc_write_coef_idx(codec, 0xd, val | (1<<14));
val = alc_read_coef_idx(codec, 0x4); /* HP */
- alc_write_coef_idx(codec, 0x4, val | (1<<11));
+ if (val != -1)
+ alc_write_coef_idx(codec, 0x4, val | (1<<11));
}
/*
case 0x10ec0286:
case 0x10ec0288:
spec->codec_variant = ALC269_TYPE_ALC286;
+ spec->shutup = alc286_shutup;
break;
case 0x10ec0255:
spec->codec_variant = ALC269_TYPE_ALC255;
MODULE_PARM_DESC(dxr_enable, "Enable DXR support for Terratec DMX6FIRE.");
-static DEFINE_PCI_DEVICE_TABLE(snd_ice1712_ids) = {
+static const struct pci_device_id snd_ice1712_ids[] = {
{ PCI_VDEVICE(ICE, PCI_DEVICE_ID_ICE_1712), 0 }, /* ICE1712 */
{ 0, }
};
/* Both VT1720 and VT1724 have the same PCI IDs */
-static DEFINE_PCI_DEVICE_TABLE(snd_vt1724_ids) = {
+static const struct pci_device_id snd_vt1724_ids[] = {
{ PCI_VDEVICE(ICE, PCI_DEVICE_ID_VT1724), 0 },
{ 0, }
};
u32 int_sta_mask; /* interrupt status mask */
};
-static DEFINE_PCI_DEVICE_TABLE(snd_intel8x0_ids) = {
+static const struct pci_device_id snd_intel8x0_ids[] = {
{ PCI_VDEVICE(INTEL, 0x2415), DEVICE_INTEL }, /* 82801AA */
{ PCI_VDEVICE(INTEL, 0x2425), DEVICE_INTEL }, /* 82901AB */
{ PCI_VDEVICE(INTEL, 0x2445), DEVICE_INTEL }, /* 82801BA */
unsigned int pcm_pos_shift;
};
-static DEFINE_PCI_DEVICE_TABLE(snd_intel8x0m_ids) = {
+static const struct pci_device_id snd_intel8x0m_ids[] = {
{ PCI_VDEVICE(INTEL, 0x2416), DEVICE_INTEL }, /* 82801AA */
{ PCI_VDEVICE(INTEL, 0x2426), DEVICE_INTEL }, /* 82901AB */
{ PCI_VDEVICE(INTEL, 0x2446), DEVICE_INTEL }, /* 82801BA */
MODULE_PARM_DESC(enable, "Enable Korg 1212 soundcard.");
MODULE_AUTHOR("Haroldo Gamal <gamal@alternex.com.br>");
-static DEFINE_PCI_DEVICE_TABLE(snd_korg1212_ids) = {
+static const struct pci_device_id snd_korg1212_ids[] = {
{
.vendor = 0x10b5,
.device = 0x906d,
}
/* PCI IDs */
-static DEFINE_PCI_DEVICE_TABLE(lola_ids) = {
+static const struct pci_device_id lola_ids[] = {
{ PCI_VDEVICE(DIGIGRAM, 0x0001) },
{ 0, }
};
#define PCI_DEVICE_ID_PLX_LX6464ES PCI_DEVICE_ID_PLX_9056
-static DEFINE_PCI_DEVICE_TABLE(snd_lx6464es_ids) = {
+static const struct pci_device_id snd_lx6464es_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_LX6464ES),
.subvendor = PCI_VENDOR_ID_DIGIGRAM,
.subdevice = PCI_SUBDEVICE_ID_DIGIGRAM_LX6464ES_SERIAL_SUBSYSTEM
/*
* pci ids
*/
-static DEFINE_PCI_DEVICE_TABLE(snd_m3_ids) = {
+static const struct pci_device_id snd_m3_ids[] = {
{PCI_VENDOR_ID_ESS, PCI_DEVICE_ID_ESS_ALLEGRO_1, PCI_ANY_ID, PCI_ANY_ID,
PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, 0},
{PCI_VENDOR_ID_ESS, PCI_DEVICE_ID_ESS_ALLEGRO, PCI_ANY_ID, PCI_ANY_ID,
/*
*/
-static DEFINE_PCI_DEVICE_TABLE(snd_mixart_ids) = {
+static const struct pci_device_id snd_mixart_ids[] = {
{ PCI_VDEVICE(MOTOROLA, 0x0003), 0, }, /* MC8240 */
{ 0, }
};
/*
* PCI ids
*/
-static DEFINE_PCI_DEVICE_TABLE(snd_nm256_ids) = {
+static const struct pci_device_id snd_nm256_ids[] = {
{PCI_VDEVICE(NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256AV_AUDIO), 0},
{PCI_VDEVICE(NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256ZX_AUDIO), 0},
{PCI_VDEVICE(NEOMAGIC, PCI_DEVICE_ID_NEOMAGIC_NM256XL_PLUS_AUDIO), 0},
MODEL_XONAR_DGX,
};
-static DEFINE_PCI_DEVICE_TABLE(oxygen_ids) = {
+static const struct pci_device_id oxygen_ids[] = {
/* C-Media's reference design */
{ OXYGEN_PCI_SUBID(0x10b0, 0x0216), .driver_data = MODEL_CMEDIA_REF },
{ OXYGEN_PCI_SUBID(0x10b0, 0x0217), .driver_data = MODEL_CMEDIA_REF },
module_param_array(enable, bool, NULL, 0444);
MODULE_PARM_DESC(enable, "enable card");
-static DEFINE_PCI_DEVICE_TABLE(xonar_ids) = {
+static const struct pci_device_id xonar_ids[] = {
{ OXYGEN_PCI_SUBID(0x1043, 0x8269) },
{ OXYGEN_PCI_SUBID(0x1043, 0x8275) },
{ OXYGEN_PCI_SUBID(0x1043, 0x82b7) },
PCI_ID_LAST
};
-static DEFINE_PCI_DEVICE_TABLE(pcxhr_ids) = {
+static const struct pci_device_id pcxhr_ids[] = {
{ 0x10b5, 0x9656, 0x1369, 0xb001, 0, 0, PCI_ID_VX882HR, },
{ 0x10b5, 0x9656, 0x1369, 0xb101, 0, 0, PCI_ID_PCX882HR, },
{ 0x10b5, 0x9656, 0x1369, 0xb201, 0, 0, PCI_ID_VX881HR, },
/*
*/
-static DEFINE_PCI_DEVICE_TABLE(snd_riptide_ids) = {
+static const struct pci_device_id snd_riptide_ids[] = {
{ PCI_DEVICE(0x127a, 0x4310) },
{ PCI_DEVICE(0x127a, 0x4320) },
{ PCI_DEVICE(0x127a, 0x4330) },
};
#ifdef SUPPORT_JOYSTICK
-static DEFINE_PCI_DEVICE_TABLE(snd_riptide_joystick_ids) = {
+static const struct pci_device_id snd_riptide_joystick_ids[] = {
{ PCI_DEVICE(0x127a, 0x4312) },
{ PCI_DEVICE(0x127a, 0x4322) },
{ PCI_DEVICE(0x127a, 0x4332) },
struct snd_kcontrol *spdif_ctl;
};
-static DEFINE_PCI_DEVICE_TABLE(snd_rme32_ids) = {
+static const struct pci_device_id snd_rme32_ids[] = {
{PCI_VDEVICE(XILINX_RME, PCI_DEVICE_ID_RME_DIGI32), 0,},
{PCI_VDEVICE(XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_8), 0,},
{PCI_VDEVICE(XILINX_RME, PCI_DEVICE_ID_RME_DIGI32_PRO), 0,},
struct snd_kcontrol *spdif_ctl;
};
-static DEFINE_PCI_DEVICE_TABLE(snd_rme96_ids) = {
+static const struct pci_device_id snd_rme96_ids[] = {
{ PCI_VDEVICE(XILINX, PCI_DEVICE_ID_RME_DIGI96), 0, },
{ PCI_VDEVICE(XILINX, PCI_DEVICE_ID_RME_DIGI96_8), 0, },
{ PCI_VDEVICE(XILINX, PCI_DEVICE_ID_RME_DIGI96_8_PRO), 0, },
}
-static DEFINE_PCI_DEVICE_TABLE(snd_hdsp_ids) = {
+static const struct pci_device_id snd_hdsp_ids[] = {
{
.vendor = PCI_VENDOR_ID_XILINX,
.device = PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP,
};
-static DEFINE_PCI_DEVICE_TABLE(snd_hdspm_ids) = {
+static const struct pci_device_id snd_hdspm_ids[] = {
{
.vendor = PCI_VENDOR_ID_XILINX,
.device = PCI_DEVICE_ID_XILINX_HAMMERFALL_DSP_MADI,
}
-static DEFINE_PCI_DEVICE_TABLE(snd_rme9652_ids) = {
+static const struct pci_device_id snd_rme9652_ids[] = {
{
.vendor = 0x10ee,
.device = 0x3fc4,
module_param(codecs, int, 0444);
MODULE_PARM_DESC(codecs, "Set bit to indicate that codec number is expected to be present (default 1)");
-static DEFINE_PCI_DEVICE_TABLE(snd_sis7019_ids) = {
+static const struct pci_device_id snd_sis7019_ids[] = {
{ PCI_DEVICE(PCI_VENDOR_ID_SI, 0x7019) },
{ 0, }
};
#endif
};
-static DEFINE_PCI_DEVICE_TABLE(snd_sonic_ids) = {
+static const struct pci_device_id snd_sonic_ids[] = {
{ PCI_VDEVICE(S3, 0xca00), 0, },
{ 0, }
};
module_param_array(wavetable_size, int, NULL, 0444);
MODULE_PARM_DESC(wavetable_size, "Maximum memory size in kB for wavetable synth.");
-static DEFINE_PCI_DEVICE_TABLE(snd_trident_ids) = {
+static const struct pci_device_id snd_trident_ids[] = {
{PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_DX),
PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, 0},
{PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_NX),
#endif
};
-static DEFINE_PCI_DEVICE_TABLE(snd_via82xx_ids) = {
+static const struct pci_device_id snd_via82xx_ids[] = {
/* 0x1106, 0x3058 */
{ PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C686_5), TYPE_CARD_VIA686, }, /* 686A */
/* 0x1106, 0x3059 */
struct snd_info_entry *proc_entry;
};
-static DEFINE_PCI_DEVICE_TABLE(snd_via82xx_modem_ids) = {
+static const struct pci_device_id snd_via82xx_modem_ids[] = {
{ PCI_VDEVICE(VIA, 0x3068), TYPE_CARD_VIA82XX_MODEM, },
{ 0, }
};
VX_PCI_VX222_NEW
};
-static DEFINE_PCI_DEVICE_TABLE(snd_vx222_ids) = {
+static const struct pci_device_id snd_vx222_ids[] = {
{ 0x10b5, 0x9050, 0x1369, PCI_ANY_ID, 0, 0, VX_PCI_VX222_OLD, }, /* PLX */
{ 0x10b5, 0x9030, 0x1369, PCI_ANY_ID, 0, 0, VX_PCI_VX222_NEW, }, /* PLX */
{ 0, }
module_param_array(rear_switch, bool, NULL, 0444);
MODULE_PARM_DESC(rear_switch, "Enable shared rear/line-in switch");
-static DEFINE_PCI_DEVICE_TABLE(snd_ymfpci_ids) = {
+static const struct pci_device_id snd_ymfpci_ids[] = {
{ PCI_VDEVICE(YAMAHA, 0x0004), 0, }, /* YMF724 */
{ PCI_VDEVICE(YAMAHA, 0x000d), 0, }, /* YMF724F */
{ PCI_VDEVICE(YAMAHA, 0x000a), 0, }, /* YMF740 */
return -ENODEV;
if (!sound) {
- sound = of_find_node_by_name(NULL, "sound");
- while (sound && sound->parent != chip->node)
- sound = of_find_node_by_name(sound, "sound");
+ for_each_node_by_name(sound, "sound")
+ if (sound->parent == chip->node)
+ break;
}
if (! sound) {
of_node_put(chip->node);
else
rates = &arizona_48k_bclk_rates[0];
+ wl = snd_pcm_format_width(params_format(params));
+
if (tdm_slots) {
arizona_aif_dbg(dai, "Configuring for %d %d bit TDM slots\n",
tdm_slots, tdm_width);
channels = tdm_slots;
} else {
bclk_target = snd_soc_params_to_bclk(params);
+ tdm_width = wl;
}
if (chan_limit && chan_limit < channels) {
arizona_aif_dbg(dai, "BCLK %dHz LRCLK %dHz\n",
rates[bclk], rates[bclk] / lrclk);
- wl = snd_pcm_format_width(params_format(params));
- frame = wl << ARIZONA_AIF1TX_WL_SHIFT | wl;
+ frame = wl << ARIZONA_AIF1TX_WL_SHIFT | tdm_width;
reconfig = arizona_aif_cfg_changed(codec, base, bclk, lrclk, frame);
pcm512x_ramp_step_text);
static const struct snd_kcontrol_new pcm512x_controls[] = {
-SOC_DOUBLE_R_TLV("Playback Digital Volume", PCM512x_DIGITAL_VOLUME_2,
+SOC_DOUBLE_R_TLV("Digital Playback Volume", PCM512x_DIGITAL_VOLUME_2,
PCM512x_DIGITAL_VOLUME_3, 0, 255, 1, digital_tlv),
SOC_DOUBLE_TLV("Playback Volume", PCM512x_ANALOG_GAIN_CTRL,
PCM512x_LAGN_SHIFT, PCM512x_RAGN_SHIFT, 1, 1, analog_tlv),
SOC_DOUBLE_TLV("Playback Boost Volume", PCM512x_ANALOG_GAIN_BOOST,
PCM512x_AGBL_SHIFT, PCM512x_AGBR_SHIFT, 1, 0, boost_tlv),
-SOC_DOUBLE("Playback Digital Switch", PCM512x_MUTE, PCM512x_RQML_SHIFT,
+SOC_DOUBLE("Digital Playback Switch", PCM512x_MUTE, PCM512x_RQML_SHIFT,
PCM512x_RQMR_SHIFT, 1, 1),
SOC_SINGLE("Deemphasis Switch", PCM512x_DSP, PCM512x_DEMP_SHIFT, 1, 1),
return ret;
}
-static int davinci_mcasp_set_clkdiv(struct snd_soc_dai *dai, int div_id, int div)
+static int __davinci_mcasp_set_clkdiv(struct snd_soc_dai *dai, int div_id,
+ int div, bool explicit)
{
struct davinci_mcasp *mcasp = snd_soc_dai_get_drvdata(dai);
ACLKXDIV(div - 1), ACLKXDIV_MASK);
mcasp_mod_bits(mcasp, DAVINCI_MCASP_ACLKRCTL_REG,
ACLKRDIV(div - 1), ACLKRDIV_MASK);
- mcasp->bclk_div = div;
+ if (explicit)
+ mcasp->bclk_div = div;
break;
case 2: /* BCLK/LRCLK ratio */
return 0;
}
+static int davinci_mcasp_set_clkdiv(struct snd_soc_dai *dai, int div_id,
+ int div)
+{
+ return __davinci_mcasp_set_clkdiv(dai, div_id, div, 1);
+}
+
static int davinci_mcasp_set_sysclk(struct snd_soc_dai *dai, int clk_id,
unsigned int freq, int dir)
{
"Inaccurate BCLK: %u Hz / %u != %u Hz\n",
mcasp->sysclk_freq, div, bclk_freq);
}
- davinci_mcasp_set_clkdiv(cpu_dai, 1, div);
+ __davinci_mcasp_set_clkdiv(cpu_dai, 1, div, 0);
}
ret = mcasp_common_hw_param(mcasp, substream->stream,
tristate "Enhanced Serial Audio Interface (ESAI) module support"
select REGMAP_MMIO
select SND_SOC_IMX_PCM_DMA if SND_IMX_SOC != n
- select SND_SOC_FSL_UTILS
help
Say Y if you want to add Enhanced Synchronous Audio Interface
(ESAI) support for the Freescale CPUs.
#include "fsl_esai.h"
#include "imx-pcm.h"
-#include "fsl_utils.h"
#define FSL_ESAI_RATES SNDRV_PCM_RATE_8000_192000
#define FSL_ESAI_FORMATS (SNDRV_PCM_FMTBIT_S8 | \
.hw_params = fsl_esai_hw_params,
.set_sysclk = fsl_esai_set_dai_sysclk,
.set_fmt = fsl_esai_set_dai_fmt,
- .xlate_tdm_slot_mask = fsl_asoc_xlate_tdm_slot_mask,
.set_tdm_slot = fsl_esai_set_dai_tdm_slot,
};
};
static struct sst_acpi_mach baytrail_machines[] = {
- { "10EC5640", "byt-rt5640", "intel/fw_sst_0f28.bin-i2s_master" },
- { "193C9890", "byt-max98090", "intel/fw_sst_0f28.bin-i2s_master" },
+ { "10EC5640", "byt-rt5640", "intel/fw_sst_0f28.bin-48kHz_i2s_master" },
+ { "193C9890", "byt-max98090", "intel/fw_sst_0f28.bin-48kHz_i2s_master" },
{}
};
.ops = &sst_byt_ops,
};
-int sst_byt_dsp_suspend_noirq(struct device *dev, struct sst_pdata *pdata)
+int sst_byt_dsp_suspend_late(struct device *dev, struct sst_pdata *pdata)
{
struct sst_byt *byt = pdata->dsp;
sst_byt_drop_all(byt);
dev_dbg(byt->dev, "dsp in reset\n");
- return 0;
-}
-EXPORT_SYMBOL_GPL(sst_byt_dsp_suspend_noirq);
-
-int sst_byt_dsp_suspend_late(struct device *dev, struct sst_pdata *pdata)
-{
- struct sst_byt *byt = pdata->dsp;
-
dev_dbg(byt->dev, "free all blocks and unload fw\n");
sst_fw_unload(byt->fw);
int sst_byt_dsp_init(struct device *dev, struct sst_pdata *pdata);
void sst_byt_dsp_free(struct device *dev, struct sst_pdata *pdata);
struct sst_dsp *sst_byt_get_dsp(struct sst_byt *byt);
-int sst_byt_dsp_suspend_noirq(struct device *dev, struct sst_pdata *pdata);
int sst_byt_dsp_suspend_late(struct device *dev, struct sst_pdata *pdata);
int sst_byt_dsp_boot(struct device *dev, struct sst_pdata *pdata);
int sst_byt_dsp_wait_for_ready(struct device *dev, struct sst_pdata *pdata);
/* DAI data */
struct sst_byt_pcm_data pcm[BYT_PCM_COUNT];
+
+ /* flag indicating is stream context restore needed after suspend */
+ bool restore_stream;
};
/* this may get called several times by oss emulation */
sst_byt_stream_start(byt, pcm_data->stream, 0);
break;
case SNDRV_PCM_TRIGGER_RESUME:
- schedule_work(&pcm_data->work);
+ if (pdata->restore_stream == true)
+ schedule_work(&pcm_data->work);
+ else
+ sst_byt_stream_resume(byt, pcm_data->stream);
break;
case SNDRV_PCM_TRIGGER_PAUSE_RELEASE:
sst_byt_stream_resume(byt, pcm_data->stream);
sst_byt_stream_stop(byt, pcm_data->stream);
break;
case SNDRV_PCM_TRIGGER_SUSPEND:
+ pdata->restore_stream = false;
case SNDRV_PCM_TRIGGER_PAUSE_PUSH:
sst_byt_stream_pause(byt, pcm_data->stream);
break;
};
#ifdef CONFIG_PM
-static int sst_byt_pcm_dev_suspend_noirq(struct device *dev)
-{
- struct sst_pdata *sst_pdata = dev_get_platdata(dev);
- int ret;
-
- dev_dbg(dev, "suspending noirq\n");
-
- /* at this point all streams will be stopped and context saved */
- ret = sst_byt_dsp_suspend_noirq(dev, sst_pdata);
- if (ret < 0) {
- dev_err(dev, "failed to suspend %d\n", ret);
- return ret;
- }
-
- return ret;
-}
-
static int sst_byt_pcm_dev_suspend_late(struct device *dev)
{
struct sst_pdata *sst_pdata = dev_get_platdata(dev);
+ struct sst_byt_priv_data *priv_data = dev_get_drvdata(dev);
int ret;
dev_dbg(dev, "suspending late\n");
return ret;
}
+ priv_data->restore_stream = true;
+
return ret;
}
static int sst_byt_pcm_dev_resume_early(struct device *dev)
{
struct sst_pdata *sst_pdata = dev_get_platdata(dev);
+ int ret;
dev_dbg(dev, "resume early\n");
/* load fw and boot DSP */
- return sst_byt_dsp_boot(dev, sst_pdata);
-}
-
-static int sst_byt_pcm_dev_resume(struct device *dev)
-{
- struct sst_pdata *sst_pdata = dev_get_platdata(dev);
-
- dev_dbg(dev, "resume\n");
+ ret = sst_byt_dsp_boot(dev, sst_pdata);
+ if (ret)
+ return ret;
/* wait for FW to finish booting */
return sst_byt_dsp_wait_for_ready(dev, sst_pdata);
}
static const struct dev_pm_ops sst_byt_pm_ops = {
- .suspend_noirq = sst_byt_pcm_dev_suspend_noirq,
.suspend_late = sst_byt_pcm_dev_suspend_late,
.resume_early = sst_byt_pcm_dev_resume_early,
- .resume = sst_byt_pcm_dev_resume,
};
#define SST_BYT_PM_OPS (&sst_byt_pm_ops)
SNDRV_PCM_RATE_48000 | SNDRV_PCM_RATE_64000 | \
SNDRV_PCM_RATE_88200 | SNDRV_PCM_RATE_96000)
-#define PXA_SSP_FORMATS (SNDRV_PCM_FMTBIT_S16_LE |\
- SNDRV_PCM_FMTBIT_S24_LE | \
- SNDRV_PCM_FMTBIT_S32_LE)
+#define PXA_SSP_FORMATS (SNDRV_PCM_FMTBIT_S16_LE | SNDRV_PCM_FMTBIT_S32_LE)
static const struct snd_soc_dai_ops pxa_ssp_dai_ops = {
.startup = pxa_ssp_startup,
struct snd_soc_dapm_context *dapm = snd_soc_dapm_kcontrol_dapm(kcontrol);
struct soc_enum *e = (struct soc_enum *)kcontrol->private_value;
unsigned int reg_val, val;
- int ret = 0;
- if (e->reg != SND_SOC_NOPM)
- ret = soc_dapm_read(dapm, e->reg, ®_val);
- else
+ if (e->reg != SND_SOC_NOPM) {
+ int ret = soc_dapm_read(dapm, e->reg, ®_val);
+ if (ret)
+ return ret;
+ } else {
reg_val = dapm_kcontrol_get_value(kcontrol);
+ }
val = (reg_val >> e->shift_l) & e->mask;
ucontrol->value.enumerated.item[0] = snd_soc_enum_val_to_item(e, val);
ucontrol->value.enumerated.item[1] = val;
}
- return ret;
+ return 0;
}
EXPORT_SYMBOL_GPL(snd_soc_dapm_get_enum_double);
}
}
},
+{
+ /* BOSS ME-25 */
+ USB_DEVICE(0x0582, 0x0113),
+ .driver_info = (unsigned long) & (const struct snd_usb_audio_quirk) {
+ .ifnum = QUIRK_ANY_INTERFACE,
+ .type = QUIRK_COMPOSITE,
+ .data = (const struct snd_usb_audio_quirk[]) {
+ {
+ .ifnum = 0,
+ .type = QUIRK_AUDIO_STANDARD_INTERFACE
+ },
+ {
+ .ifnum = 1,
+ .type = QUIRK_AUDIO_STANDARD_INTERFACE
+ },
+ {
+ .ifnum = 2,
+ .type = QUIRK_MIDI_FIXED_ENDPOINT,
+ .data = & (const struct snd_usb_midi_endpoint_info) {
+ .out_cables = 0x0001,
+ .in_cables = 0x0001
+ }
+ },
+ {
+ .ifnum = -1
+ }
+ }
+ }
+},
{
/* only 44.1 kHz works at the moment */
USB_DEVICE(0x0582, 0x0120),
/*
* Example Format w/ field column widths:
*
- * Package Core CPU Avg_MHz Bzy_MHz TSC_MHz SMI %Busy CPU_%c1 CPU_%c3 CPU_%c6 CPU_%c7 CoreTmp PkgTmp Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 PkgWatt CorWatt GFXWatt
- * 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567 1234567
+ * Package Core CPU Avg_MHz Bzy_MHz TSC_MHz SMI %Busy CPU_%c1 CPU_%c3 CPU_%c6 CPU_%c7 CoreTmp PkgTmp Pkg%pc2 Pkg%pc3 Pkg%pc6 Pkg%pc7 PkgWatt CorWatt GFXWatt
+ * 123456781234567812345678123456781234567812345678123456781234567812345678123456781234567812345678123456781234567812345678123456781234567812345678123456781234567812345678
*/
void print_header(void)
{
if (show_pkg)
- outp += sprintf(outp, "Package ");
+ outp += sprintf(outp, " Package");
if (show_core)
- outp += sprintf(outp, " Core ");
+ outp += sprintf(outp, " Core");
if (show_cpu)
- outp += sprintf(outp, " CPU ");
+ outp += sprintf(outp, " CPU");
if (has_aperf)
- outp += sprintf(outp, "Avg_MHz ");
+ outp += sprintf(outp, " Avg_MHz");
if (do_nhm_cstates)
- outp += sprintf(outp, " %%Busy ");
+ outp += sprintf(outp, " %%Busy");
if (has_aperf)
- outp += sprintf(outp, "Bzy_MHz ");
- outp += sprintf(outp, "TSC_MHz ");
+ outp += sprintf(outp, " Bzy_MHz");
+ outp += sprintf(outp, " TSC_MHz");
if (do_smi)
- outp += sprintf(outp, " SMI ");
+ outp += sprintf(outp, " SMI");
if (extra_delta_offset32)
- outp += sprintf(outp, " count 0x%03X ", extra_delta_offset32);
+ outp += sprintf(outp, " count 0x%03X", extra_delta_offset32);
if (extra_delta_offset64)
- outp += sprintf(outp, " COUNT 0x%03X ", extra_delta_offset64);
+ outp += sprintf(outp, " COUNT 0x%03X", extra_delta_offset64);
if (extra_msr_offset32)
- outp += sprintf(outp, " MSR 0x%03X ", extra_msr_offset32);
+ outp += sprintf(outp, " MSR 0x%03X", extra_msr_offset32);
if (extra_msr_offset64)
- outp += sprintf(outp, " MSR 0x%03X ", extra_msr_offset64);
+ outp += sprintf(outp, " MSR 0x%03X", extra_msr_offset64);
if (do_nhm_cstates)
- outp += sprintf(outp, " CPU%%c1 ");
+ outp += sprintf(outp, " CPU%%c1");
if (do_nhm_cstates && !do_slm_cstates)
- outp += sprintf(outp, " CPU%%c3 ");
+ outp += sprintf(outp, " CPU%%c3");
if (do_nhm_cstates)
- outp += sprintf(outp, " CPU%%c6 ");
+ outp += sprintf(outp, " CPU%%c6");
if (do_snb_cstates)
- outp += sprintf(outp, " CPU%%c7 ");
+ outp += sprintf(outp, " CPU%%c7");
if (do_dts)
- outp += sprintf(outp, "CoreTmp ");
+ outp += sprintf(outp, " CoreTmp");
if (do_ptm)
- outp += sprintf(outp, " PkgTmp ");
+ outp += sprintf(outp, " PkgTmp");
if (do_snb_cstates)
- outp += sprintf(outp, "Pkg%%pc2 ");
+ outp += sprintf(outp, " Pkg%%pc2");
if (do_nhm_cstates && !do_slm_cstates)
- outp += sprintf(outp, "Pkg%%pc3 ");
+ outp += sprintf(outp, " Pkg%%pc3");
if (do_nhm_cstates && !do_slm_cstates)
- outp += sprintf(outp, "Pkg%%pc6 ");
+ outp += sprintf(outp, " Pkg%%pc6");
if (do_snb_cstates)
- outp += sprintf(outp, "Pkg%%pc7 ");
+ outp += sprintf(outp, " Pkg%%pc7");
if (do_c8_c9_c10) {
- outp += sprintf(outp, "Pkg%%pc8 ");
- outp += sprintf(outp, "Pkg%%pc9 ");
- outp += sprintf(outp, "Pk%%pc10 ");
+ outp += sprintf(outp, " Pkg%%pc8");
+ outp += sprintf(outp, " Pkg%%pc9");
+ outp += sprintf(outp, " Pk%%pc10");
}
if (do_rapl && !rapl_joules) {
if (do_rapl & RAPL_PKG)
- outp += sprintf(outp, "PkgWatt ");
+ outp += sprintf(outp, " PkgWatt");
if (do_rapl & RAPL_CORES)
- outp += sprintf(outp, "CorWatt ");
+ outp += sprintf(outp, " CorWatt");
if (do_rapl & RAPL_GFX)
- outp += sprintf(outp, "GFXWatt ");
+ outp += sprintf(outp, " GFXWatt");
if (do_rapl & RAPL_DRAM)
- outp += sprintf(outp, "RAMWatt ");
+ outp += sprintf(outp, " RAMWatt");
if (do_rapl & RAPL_PKG_PERF_STATUS)
- outp += sprintf(outp, " PKG_%% ");
+ outp += sprintf(outp, " PKG_%%");
if (do_rapl & RAPL_DRAM_PERF_STATUS)
- outp += sprintf(outp, " RAM_%% ");
+ outp += sprintf(outp, " RAM_%%");
} else {
if (do_rapl & RAPL_PKG)
- outp += sprintf(outp, " Pkg_J ");
+ outp += sprintf(outp, " Pkg_J");
if (do_rapl & RAPL_CORES)
- outp += sprintf(outp, " Cor_J ");
+ outp += sprintf(outp, " Cor_J");
if (do_rapl & RAPL_GFX)
- outp += sprintf(outp, " GFX_J ");
+ outp += sprintf(outp, " GFX_J");
if (do_rapl & RAPL_DRAM)
- outp += sprintf(outp, " RAM_W ");
+ outp += sprintf(outp, " RAM_W");
if (do_rapl & RAPL_PKG_PERF_STATUS)
- outp += sprintf(outp, " PKG_%% ");
+ outp += sprintf(outp, " PKG_%%");
if (do_rapl & RAPL_DRAM_PERF_STATUS)
- outp += sprintf(outp, " RAM_%% ");
- outp += sprintf(outp, " time ");
+ outp += sprintf(outp, " RAM_%%");
+ outp += sprintf(outp, " time");
}
outp += sprintf(outp, "\n");
dev->irq_requested_type |= guest_irq_type;
if (dev->ack_notifier.gsi != -1)
kvm_register_irq_ack_notifier(kvm, &dev->ack_notifier);
- } else
+ } else {
kvm_free_irq_source_id(kvm, dev->irq_source_id);
+ dev->irq_source_id = -1;
+ }
return r;
}
return pfn;
}
+static void kvm_unpin_pages(struct kvm *kvm, pfn_t pfn, unsigned long npages)
+{
+ unsigned long i;
+
+ for (i = 0; i < npages; ++i)
+ kvm_release_pfn_clean(pfn + i);
+}
+
int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot)
{
gfn_t gfn, end_gfn;
if (r) {
printk(KERN_ERR "kvm_iommu_map_address:"
"iommu failed to map pfn=%llx\n", pfn);
+ kvm_unpin_pages(kvm, pfn, page_size);
goto unmap_pages;
}
return 0;
unmap_pages:
- kvm_iommu_put_pages(kvm, slot->base_gfn, gfn);
+ kvm_iommu_put_pages(kvm, slot->base_gfn, gfn - slot->base_gfn);
return r;
}
return r;
}
-static void kvm_unpin_pages(struct kvm *kvm, pfn_t pfn, unsigned long npages)
-{
- unsigned long i;
-
- for (i = 0; i < npages; ++i)
- kvm_release_pfn_clean(pfn + i);
-}
-
static void kvm_iommu_put_pages(struct kvm *kvm,
gfn_t base_gfn, unsigned long npages)
{
projects / firefly-linux-kernel-4.4.55.git / commitdiff