drm: bridge: dw-hdmi: enable 3d mode

Change-Id: I45c4b0ded4aeaf24122d5cc6eb49b6bc72afc060
Signed-off-by: algea.cao <algea.cao@rock-chips.com>

camera: rockchip: camsys v0.0x21.0xa

Change-Id: Ic9c5661ba76eaf8aee36e1a1822aa9bcd288e0ac
Signed-off-by: dalon.zhang <dalon.zhang@rock-chips.com>

arm64: dts: rockchip: update thermal config for rk3368

add cpus' dynamic power coefficient and update alert temperature

Change-Id: I502e49d52268b63625e01103b50e6c18fb8da8b0
Signed-off-by: Rocky Hao <rocky.hao@rock-chips.com>

arm64: dts: rockchip: enable tsadc node for rk3368 p9 board

Change-Id: I5d8b84248b0c713068143df303427ef31cb78963
Signed-off-by: Zorro Liu <lyx@rock-chips.com>

arm64: dts: rk3368-android: debug uart id change to uart3

Change earlycon and console to uart3.

Change-Id: I7c6d7322e077b605b209dce4cf51afb26b9147dc
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

arm64: dts: rk3368: hdmi disabled default and remove hdmi node from p9 board

Change-Id: I034f119de02a04841b8b70746ef63c73a170988f
Signed-off-by: Zorro Liu <lyx@rock-chips.com>

arm64: dts: rockchip: rk3368: xin32k use the fixed clk

If xin32k use the rk808_clkout1, rk808 init is too late,
xin32k enable count and prepare count is not match with it's child clk.

Change-Id: I314776c5024fdf3373619968582497e0e2d5666f
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>

arm64: dts: rockchip: enable tsadc node for rk3368 sheep board

Change-Id: I82c8ef62f5e149d07c76e3a3d971e4222d9b8ecf
Signed-off-by: Rocky Hao <rocky.hao@rock-chips.com>

arm64: dts: rockchip: update tsadc node for rk3368

Change-Id: I0c99dcc6b5515639a496b915832542c3b844f4c8
Signed-off-by: Rocky Hao <rocky.hao@rock-chips.com>

arm64: configs: rockchip_defconfig: enable rk3368 thermal

Change-Id: Ied2658479d5cccc404623af39da2c6cbe028bae0
Signed-off-by: Rocky Hao <rocky.hao@rock-chips.com>

thermal: rockchip: add rk3368 support

Change-Id: I970fedca9542c724d777c0bac788300c4fa21303
Signed-off-by: Rocky Hao <rocky.hao@rock-chips.com>

nvmem: rockchip-efuse: Fix dependencies

Not every rockchip efuse depends on ROCKCHIP_SIP, so delete
dependencies in Kconfig. It is more appropriate to add
dependencies for sip_smc_secure_reg_read/write.

Change-Id: I7f551f9fe71ced847657531e3c3cf418766fa3a4
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>

arm64: dts: rockchip: add RK3399 Firefly Board for Linux Opensource

Add Firefly board dts file for Linux Opensource project

Change-Id: Ia525b7ff17f4d74990625e2e02c764996f57e520
Signed-off-by: hero.huang <hero.huang@rock-chips.com>

arm64: dts: rk3368-p9: add usb host

Change-Id: Ief0045ac99d3af4db22042d468b67609d6fddf2f
Signed-off-by: Zorro Liu <lyx@rock-chips.com>

arm64: dts: rk3368-p9&sheep: disable uart2

Disable uart2 since gpio mux on uart with sdmmc, and rk3368 use
fiq debugger, the uart2 could set to be disabled.

Change-Id: I2d784ccd6cf7526afc0f3bae54914e05febf91a6
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

arm64: dts: rk3368-p9: enable sdmmc

Change-Id: Ib2849e2af020c744e33f0ab1baefbe50cecaa80b
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

Merge branch 'linux-linaro-lsk-v4.4' into linux-linaro-lsk-v4.4-android

 Merge tag 'v4.4.55' into linux-linaro-lsk-v4.4

 This is the 4.4.55 stable release

FROMLIST: drm/core: Reuse the reserved member in drm_event_vblank for crtc_id.

When doing a atomic commit affecting multiple crtc's, multiple events
are generated. The user_data member does not allow you to distinguish,
because they all have the same pointer.

I've chosen to use crtc_id, because using pipe would create ambiguity
when pipe = 0. A test for != 0 is easier to implement, and crtc_id
will never be 0.

Change-Id: Ie2daba50f711f298872f15498b8d46dedb38c0ff
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Stone <daniels@collabora.com>
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>
(am from https://patchwork.kernel.org/patch/9272895/)

arm64: dts: rk3368-android: enable rga

Change-Id: I869f4bae54f72dc384c644fe3a0a499db2af3dbf
Signed-off-by: Zikim,Wei <wzq@rock-chips.com>

arm64: dts: rk3368-sheep: fix sheep backlight

Change-Id: Ief2603afd33a65158bf6e86f08e53f96573ce486
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>

video: backlight: pwm_bl: fix backlight polarity

Backlight polarity not works without pwm_adjust_config.

Change-Id: I11e5eefe340f758b6721021f13238306b3721270
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>

arm64: dts: rk3368: enable uboot loader logo

Change-Id: I4807ac2aafd9cf319e37eacd184c1eae3ea36242
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>

Linux 4.4.55

ext4: don't BUG when truncating encrypted inodes on the orphan list

commit 0d06863f903ac5f4f6efb0273079d27de3e53a28 upstream.

Fix a BUG when the kernel tries to mount a file system constructed as
follows:

echo foo > foo.txt
mke2fs -Fq -t ext4 -O encrypt foo.img 100
debugfs -w foo.img << EOF
write foo.txt a
set_inode_field a i_flags 0x80800
set_super_value s_last_orphan 12
quit
EOF

root@kvm-xfstests:~# mount -o loop foo.img /mnt
[  160.238770] ------------[ cut here ]------------
[  160.240106] kernel BUG at /usr/projects/linux/ext4/fs/ext4/inode.c:3874!
[  160.240106] invalid opcode: 0000 [#1] SMP
[  160.240106] Modules linked in:
[  160.240106] CPU: 0 PID: 2547 Comm: mount Tainted: G        W       4.10.0-rc3-00034-gcdd33b941b67 #227
[  160.240106] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.1-1 04/01/2014
[  160.240106] task: f4518000 task.stack: f47b6000
[  160.240106] EIP: ext4_block_zero_page_range+0x1a7/0x2b4
[  160.240106] EFLAGS: 00010246 CPU: 0
[  160.240106] EAX: 00000001 EBX: f7be4b50 ECX: f47b7dc0 EDX: 00000007
[  160.240106] ESI: f43b05a8 EDI: f43babec EBP: f47b7dd0 ESP: f47b7dac
[  160.240106]  DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
[  160.240106] CR0: 80050033 CR2: bfd85b08 CR3: 34a00680 CR4: 000006f0
[  160.240106] Call Trace:
[  160.240106]  ext4_truncate+0x1e9/0x3e5
[  160.240106]  ext4_fill_super+0x286f/0x2b1e
[  160.240106]  ? set_blocksize+0x2e/0x7e
[  160.240106]  mount_bdev+0x114/0x15f
[  160.240106]  ext4_mount+0x15/0x17
[  160.240106]  ? ext4_calculate_overhead+0x39d/0x39d
[  160.240106]  mount_fs+0x58/0x115
[  160.240106]  vfs_kern_mount+0x4b/0xae
[  160.240106]  do_mount+0x671/0x8c3
[  160.240106]  ? _copy_from_user+0x70/0x83
[  160.240106]  ? strndup_user+0x31/0x46
[  160.240106]  SyS_mount+0x57/0x7b
[  160.240106]  do_int80_syscall_32+0x4f/0x61
[  160.240106]  entry_INT80_32+0x2f/0x2f
[  160.240106] EIP: 0xb76b919e
[  160.240106] EFLAGS: 00000246 CPU: 0
[  160.240106] EAX: ffffffda EBX: 08053838 ECX: 08052188 EDX: 080537e8
[  160.240106] ESI: c0ed0000 EDI: 00000000 EBP: 080537e8 ESP: bfa13660
[  160.240106]  DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 007b
[  160.240106] Code: 59 8b 00 a8 01 0f 84 09 01 00 00 8b 07 66 25 00 f0 66 3d 00 80 75 61 89 f8 e8 3e e2 ff ff 84 c0 74 56 83 bf 48 02 00 00 00 75 02 <0f> 0b 81 7d e8 00 10 00 00 74 02 0f 0b 8b 43 04 8b 53 08 31 c9
[  160.240106] EIP: ext4_block_zero_page_range+0x1a7/0x2b4 SS:ESP: 0068:f47b7dac
[  160.317241] ---[ end trace d6a773a375c810a5 ]---

The problem is that when the kernel tries to truncate an inode in
ext4_truncate(), it tries to clear any on-disk data beyond i_size.
Without the encryption key, it can't do that, and so it triggers a
BUG.

E2fsck does *not* provide this service, and in practice most file
systems have their orphan list processed by e2fsck, so to avoid
crashing, this patch skips this step if we don't have access to the
encryption key (which is the case when processing the orphan list; in
all other cases, we will have the encryption key, or the kernel
wouldn't have allowed the file to be opened).

An open question is whether the fact that e2fsck isn't clearing the
bytes beyond i_size causing problems --- and if we've lived with it
not doing it for so long, can we drop this from the kernel replay of
the orphan list in all cases (not just when we don't have the key for
encrypted inodes).

Addresses-Google-Bug: #35209576

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

dm: flush queued bios when process blocks to avoid deadlock

commit d67a5f4b5947aba4bfe9a80a2b86079c215ca755 upstream.

Commit df2cb6daa4 ("block: Avoid deadlocks with bio allocation by
stacking drivers") created a workqueue for every bio set and code
in bio_alloc_bioset() that tries to resolve some low-memory deadlocks
by redirecting bios queued on current->bio_list to the workqueue if the
system is low on memory.  However other deadlocks (see below **) may
happen, without any low memory condition, because generic_make_request
is queuing bios to current->bio_list (rather than submitting them).

** the related dm-snapshot deadlock is detailed here:
https://www.redhat.com/archives/dm-devel/2016-July/msg00065.html

Fix this deadlock by redirecting any bios on current->bio_list to the
bio_set's rescue workqueue on every schedule() call.  Consequently,
when the process blocks on a mutex, the bios queued on
current->bio_list are dispatched to independent workqueus and they can
complete without waiting for the mutex to be available.

The structure blk_plug contains an entry cb_list and this list can contain
arbitrary callback functions that are called when the process blocks.
To implement this fix DM (ab)uses the onstack plug's cb_list interface
to get its flush_current_bio_list() called at schedule() time.

This fixes the snapshot deadlock - if the map method blocks,
flush_current_bio_list() will be called and it redirects bios waiting
on current->bio_list to appropriate workqueues.

Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1267650
Depends-on: df2cb6daa4 ("block: Avoid deadlocks with bio allocation by stacking drivers")
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

nfit, libnvdimm: fix interleave set cookie calculation

commit 86ef58a4e35e8fa66afb5898cf6dec6a3bb29f67 upstream.

The interleave-set cookie is a sum that sanity checks the composition of
an interleave set has not changed from when the namespace was initially
created.  The checksum is calculated by sorting the DIMMs by their
location in the interleave-set. The comparison for the sort must be
64-bit wide, not byte-by-byte as performed by memcmp() in the broken
case.

Fix the implementation to accept correct cookie values in addition to
the Linux "memcmp" order cookies, but only allow correct cookies to be
generated going forward. It does mean that namespaces created by
third-party-tooling, or created by newer kernels with this fix, will not
validate on older kernels. However, there are a couple mitigating
conditions:

    1/ platforms with namespace-label capable NVDIMMs are not widely
       available.

    2/ interleave-sets with a single-dimm are by definition not affected
       (nothing to sort). This covers the QEMU-KVM NVDIMM emulation case.

The cookie stored in the namespace label will be fixed by any write the
namespace label, the most straightforward way to achieve this is to
write to the "alt_name" attribute of a namespace in sysfs.

Fixes: eaf961536e16 ("libnvdimm, nfit: add interleave-set state-tracking infrastructure")
Reported-by: Nicholas Moulin <nicholas.w.moulin@linux.intel.com>
Tested-by: Nicholas Moulin <nicholas.w.moulin@linux.intel.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

s390/kdump: Use "LINUX" ELF note name instead of "CORE"

commit a4a81d8eebdc1d209d034f62a082a5131e4242b5 upstream.

In binutils/libbfd (bfd/elf.c) it is enforced that all s390 specific ELF
notes like e.g. NT_S390_PREFIX or NT_S390_CTRS have "LINUX" specified
as note name. Otherwise the notes are ignored.

For /proc/vmcore we currently use "CORE" for these notes.

Up to now this has not been a real problem because the dump analysis tool
"crash" does not check the note name. But it will break all programs that
use libbfd for processing ELF notes.

So fix this and use "LINUX" for all s390 specific notes to comply with
libbfd.

Reported-by: Philipp Rudo <prudo@linux.vnet.ibm.com>
Reviewed-by: Philipp Rudo <prudo@linux.vnet.ibm.com>
Signed-off-by: Michael Holzheu <holzheu@linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

KVM: s390: Fix guest migration for huge guests resulting in panic

commit 2e4d88009f57057df7672fa69a32b5224af54d37 upstream.

While we can technically not run huge page guests right now, we can
setup a guest with huge pages. Trying to migrate it will trigger a
VM_BUG_ON and, if the kernel is not configured to panic on a BUG, it
will happily try to work on non-existing page table entries.

With this patch, we always return "dirty" if we encounter a large page
when migrating. This at least fixes the immediate problem until we
have proper handling for both kind of pages.

Fixes: 15f36eb ("KVM: s390: Add proper dirty bitmap support to S390 kvm.")
Cc: <stable@vger.kernel.org> # 3.16+
Signed-off-by: Janosch Frank <frankja@linux.vnet.ibm.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

mvsas: fix misleading indentation

commit 7789cd39274c51bf475411fe22a8ee7255082809 upstream.

Fix a smatch warning:
drivers/scsi/mvsas/mv_sas.c:740 mvs_task_prep() warn: curly braces intended?

The code is correct, the indention is misleading. When the device is not
ready we want to return SAS_PHY_DOWN. But current indentation makes it
look like we only do so in the else branch of if (mvi_dev).

Signed-off-by: Luis de Bethencourt <luisbg@osg.samsung.com>
Reviewed-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

serial: samsung: Continue to work if DMA request fails

commit f98c7bce570bdbe344b74ff5daa7dfeef3f22929 upstream.

If DMA is not available (even when configured in DeviceTree), the driver
will fail the startup procedure thus making serial console not
available.

For example this causes boot failure on QEMU ARMv7 (Exynos4210, SMDKC210):
    [    1.302575] OF: amba_device_add() failed (-19) for /amba/pdma@12680000
    ...
    [   11.435732] samsung-uart 13800000.serial: DMA request failed
    [   72.963893] samsung-uart 13800000.serial: DMA request failed
    [   73.143361] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000000

DMA is not necessary for serial to work, so continue with UART startup
after emitting a warning.

Fixes: 62c37eedb74c ("serial: samsung: add dma reqest/release functions")
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: serial: io_ti: fix information leak in completion handler

commit 654b404f2a222f918af9b0cd18ad469d0c941a8e upstream.

Add missing sanity check to the bulk-in completion handler to avoid an
integer underflow that can be triggered by a malicious device.

This avoids leaking 128 kB of memory content from after the URB transfer
buffer to user space.

Fixes: 8c209e6782ca ("USB: make actual_length in struct urb field u32")
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: serial: io_ti: fix NULL-deref in interrupt callback

commit 0b1d250afb8eb9d65afb568bac9b9f9253a82b49 upstream.

Fix a NULL-pointer dereference in the interrupt callback should a
malicious device send data containing a bad port number by adding the
missing sanity check.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: iowarrior: fix NULL-deref in write

commit de46e56653de7b3b54baa625bd582635008b8d05 upstream.

Make sure to verify that we have the required interrupt-out endpoint for
IOWarrior56 devices to avoid dereferencing a NULL-pointer in write
should a malicious device lack such an endpoint.

Fixes: 946b960d13c1 ("USB: add driver for iowarrior devices.")
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: iowarrior: fix NULL-deref at probe

commit b7321e81fc369abe353cf094d4f0dc2fe11ab95f upstream.

Make sure to check for the required interrupt-in endpoint to avoid
dereferencing a NULL-pointer should a malicious device lack such an
endpoint.

Note that a fairly recent change purported to fix this issue, but added
an insufficient test on the number of endpoints only, a test which can
now be removed.

Fixes: 4ec0ef3a8212 ("USB: iowarrior: fix oops with malicious USB descriptors")
Fixes: 946b960d13c1 ("USB: add driver for iowarrior devices.")
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: serial: omninet: fix reference leaks at open

commit 30572418b445d85fcfe6c8fe84c947d2606767d8 upstream.

This driver needlessly took another reference to the tty on open, a
reference which was then never released on close. This lead to not just
a leak of the tty, but also a driver reference leak that prevented the
driver from being unloaded after a port had once been opened.

Fixes: 4a90f09b20f4 ("tty: usb-serial krefs")
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: serial: safe_serial: fix information leak in completion handler

commit 8c76d7cd520ebffc1ea9ea0850d87a224a50c7f2 upstream.

Add missing sanity check to the bulk-in completion handler to avoid an
integer underflow that could be triggered by a malicious device.

This avoids leaking up to 56 bytes from after the URB transfer buffer to
user space.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: host: xhci-plat: Fix timeout on removal of hot pluggable xhci controllers

commit dcc7620cad5ad1326a78f4031a7bf4f0e5b42984 upstream.

Upstream commit 98d74f9ceaef ("xhci: fix 10 second timeout on removal of
PCI hotpluggable xhci controllers") fixes a problem with hot pluggable PCI
xhci controllers which can result in excessive timeouts, to the point where
the system reports a deadlock.

The same problem is seen with hot pluggable xhci controllers using the
xhci-plat driver, such as the driver used for Type-C ports on rk3399.
Similar to hot-pluggable PCI controllers, the driver for this chip
removes the xhci controller from the system when the Type-C cable is
disconnected.

The solution for PCI devices works just as well for non-PCI devices
and avoids the problem.

Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: host: xhci-dbg: HCIVERSION should be a binary number

commit f95e60a7dbecd2de816bb3ad517b3d4fbc20b507 upstream.

According to xHCI spec, HCIVERSION containing a BCD encoding
of the xHCI specification revision number, 0100h corresponds
to xHCI version 1.0. Change "100" as "0x100".

Cc: Lu Baolu <baolu.lu@linux.intel.com>
Fixes: 04abb6de2825 ("xhci: Read and parse new xhci 1.1 capability register")
Signed-off-by: Peter Chen <peter.chen@nxp.com>
Signed-off-by: Mathias Nyman <mathias.nyman@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: gadget: function: f_fs: pass companion descriptor along

commit 2bfa0719ac2a9b2f3c91345873d3cdebd0296ba9 upstream.

If we're dealing with SuperSpeed endpoints, we need
to make sure to pass along the companion descriptor
and initialize fields needed by the Gadget
API. Eventually, f_fs.c should be converted to use
config_ep_by_speed() like all other functions,
though.

Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: dwc3: gadget: make Set Endpoint Configuration macros safe

commit 7369090a9fb57c3fc705ce355d2e4523a5a24716 upstream.

Some gadget drivers are bad, bad boys. We notice
that ADB was passing bad Burst Size which caused top
bits of param0 to be overwritten which confused DWC3
when running this command.

In order to avoid future issues, we're going to make
sure values passed by macros are always safe for the
controller. Note that ADB still needs a fix to *not*
pass bad values.

Reported-by: Mohamed Abbas <mohamed.abbas@intel.com>
Sugested-by: Adam Andruszak <adam.andruszak@intel.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

usb: gadget: dummy_hcd: clear usb_gadget region before registration

commit 5bbc852676ae08e818241cf66a3ffe4be44225c4 upstream.

When the user does device unbind and rebind test, the kernel will
show below dump due to usb_gadget memory region is dirty after unbind.
Clear usb_gadget region for every new probe.

root@imx6qdlsolo:/sys/bus/platform/drivers/dummy_udc# echo dummy_udc.0 > bind
[  102.523312] kobject (eddd78b0): tried to init an initialized object, something is seriously wrong.
[  102.532447] CPU: 0 PID: 734 Comm: sh Not tainted 4.10.0-rc7-00872-g1b2b8e9 #1298
[  102.539866] Hardware name: Freescale i.MX6 SoloX (Device Tree)
[  102.545717] Backtrace:
[  102.548225] [<c010d090>] (dump_backtrace) from [<c010d338>] (show_stack+0x18/0x1c)
[  102.555822]  r7:ede34000 r6:60010013 r5:00000000 r4:c0f29418
[  102.561512] [<c010d320>] (show_stack) from [<c040c2a4>] (dump_stack+0xb4/0xe8)
[  102.568764] [<c040c1f0>] (dump_stack) from [<c040e6d4>] (kobject_init+0x80/0x9c)
[  102.576187]  r10:0000001f r9:eddd7000 r8:eeaf8c10 r7:eddd78a8 r6:c177891c r5:c0f3b060
[  102.584036]  r4:eddd78b0 r3:00000000
[  102.587641] [<c040e654>] (kobject_init) from [<c05359a4>] (device_initialize+0x28/0xf8)
[  102.595665]  r5:eebc4800 r4:eddd78a8
[  102.599268] [<c053597c>] (device_initialize) from [<c05382ac>] (device_register+0x14/0x20)
[  102.607556]  r7:eddd78a8 r6:00000000 r5:eebc4800 r4:eddd78a8
[  102.613256] [<c0538298>] (device_register) from [<c0668ef4>] (usb_add_gadget_udc_release+0x8c/0x1ec)
[  102.622410]  r5:eebc4800 r4:eddd7860
[  102.626015] [<c0668e68>] (usb_add_gadget_udc_release) from [<c0669068>] (usb_add_gadget_udc+0x14/0x18)
[  102.635351]  r10:0000001f r9:eddd7000 r8:eddd788c r7:bf003770 r6:eddd77f8 r5:eddd7818
[  102.643198]  r4:eddd785c r3:eddd7b24
[  102.646834] [<c0669054>] (usb_add_gadget_udc) from [<bf003428>] (dummy_udc_probe+0x170/0x1c4 [dummy_hcd])
[  102.656458] [<bf0032b8>] (dummy_udc_probe [dummy_hcd]) from [<c053d114>] (platform_drv_probe+0x54/0xb8)
[  102.665881]  r10:00000008 r9:c1778960 r8:bf004128 r7:fffffdfb r6:bf004128 r5:eeaf8c10
[  102.673727]  r4:eeaf8c10
[  102.676293] [<c053d0c0>] (platform_drv_probe) from [<c053b160>] (driver_probe_device+0x264/0x474)
[  102.685186]  r7:00000000 r6:00000000 r5:c1778960 r4:eeaf8c10
[  102.690876] [<c053aefc>] (driver_probe_device) from [<c05397c4>] (bind_store+0xb8/0x14c)
[  102.698994]  r10:eeb3bb4c r9:ede34000 r8:0000000c r7:eeaf8c44 r6:bf004128 r5:c0f3b668
[  102.706840]  r4:eeaf8c10
[  102.709402] [<c053970c>] (bind_store) from [<c0538ca8>] (drv_attr_store+0x28/0x34)
[  102.716998]  r9:ede34000 r8:00000000 r7:ee3863c0 r6:ee3863c0 r5:c0538c80 r4:c053970c
[  102.724776] [<c0538c80>] (drv_attr_store) from [<c029c930>] (sysfs_kf_write+0x50/0x54)
[  102.732711]  r5:c0538c80 r4:0000000c
[  102.736313] [<c029c8e0>] (sysfs_kf_write) from [<c029be84>] (kernfs_fop_write+0x100/0x214)
[  102.744599]  r7:ee3863c0 r6:eeb3bb40 r5:00000000 r4:00000000
[  102.750287] [<c029bd84>] (kernfs_fop_write) from [<c0222dd8>] (__vfs_write+0x34/0x120)
[  102.758231]  r10:00000000 r9:ede34000 r8:c0108bc4 r7:0000000c r6:ede35f80 r5:c029bd84
[  102.766077]  r4:ee223780
[  102.768638] [<c0222da4>] (__vfs_write) from [<c0224678>] (vfs_write+0xa8/0x170)
[  102.775974]  r9:ede34000 r8:c0108bc4 r7:ede35f80 r6:01861cb0 r5:ee223780 r4:0000000c
[  102.783743] [<c02245d0>] (vfs_write) from [<c0225498>] (SyS_write+0x4c/0xa8)
[  102.790818]  r9:ede34000 r8:c0108bc4 r7:0000000c r6:01861cb0 r5:ee223780 r4:ee223780
[  102.798595] [<c022544c>] (SyS_write) from [<c0108a20>] (ret_fast_syscall+0x0/0x1c)
[  102.806188]  r7:00000004 r6:b6e83d58 r5:01861cb0 r4:0000000c

Fixes: 90fccb529d24 ("usb: gadget: Gadget directory cleanup - group UDC drivers")
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Peter Chen <peter.chen@nxp.com>
Tested-by: Xiaolong Ye <xiaolong.ye@intel.com>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Felipe Balbi <felipe.balbi@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

powerpc: Emulation support for load/store instructions on LE

commit e148bd17f48bd17fca2f4f089ec879fa6e47e34c upstream.

emulate_step() uses a number of underlying kernel functions that were
initially not enabled for LE. This has been rectified since. So, fix
emulate_step() for LE for the corresponding instructions.

Reported-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

tracing: Add #undef to fix compile error

commit bf7165cfa23695c51998231c4efa080fe1d3548d upstream.

There are several trace include files that define TRACE_INCLUDE_FILE.

Include several of them in the same .c file (as I currently have in
some code I am working on), and the compile will blow up with a
"warning: "TRACE_INCLUDE_FILE" redefined #define TRACE_INCLUDE_FILE syscalls"

Every other include file in include/trace/events/ avoids that issue
by having a #undef TRACE_INCLUDE_FILE before the #define; syscalls.h
should have one, too.

Link: http://lkml.kernel.org/r/20160928225554.13bd7ac6@annuminas.surriel.com
Fixes: b8007ef74222 ("tracing: Separate raw syscall from syscall tracer")
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: Netlogic: Fix CP0_EBASE redefinition warnings

commit 32eb6e8bee147b45e5e59230630d59541ccbb6e5 upstream.

A couple of netlogic assembly files define CP0_EBASE to $15, the same as
CP0_PRID in mipsregs.h, and use it for accessing both CP0_PRId and
CP0_EBase registers. However commit 609cf6f2291a ("MIPS: CPS: Early
debug using an ns16550-compatible UART") added a different definition of
CP0_EBASE to mipsregs.h, which included a register select of 1. This
causes harmless build warnings like the following:

  arch/mips/netlogic/common/reset.S:53:0: warning: "CP0_EBASE" redefined
  #define CP0_EBASE $15
  ^
  In file included from arch/mips/netlogic/common/reset.S:41:0:
  ./arch/mips/include/asm/mipsregs.h:63:0: note: this is the location of the previous definition
  #define CP0_EBASE $15, 1
  ^

Update the code to use the definitions from mipsregs.h for accessing
both registers.

Fixes: 609cf6f2291a ("MIPS: CPS: Early debug using an ns16550-compatible UART")
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Acked-by: Jayachandran C <jchandra@broadcom.com>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/13183/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: DEC: Avoid la pseudo-instruction in delay slots

commit 3021773c7c3e75e20b693931a19362681e744ea9 upstream.

When expanding the la or dla pseudo-instruction in a delay slot the GNU
assembler will complain should the pseudo-instruction expand to multiple
actual instructions, since only the first of them will be in the delay
slot leading to the pseudo-instruction being only partially executed if
the branch is taken. Use of PTR_LA in the dec int-handler.S leads to
such warnings:

  arch/mips/dec/int-handler.S: Assembler messages:
  arch/mips/dec/int-handler.S:149: Warning: macro instruction expanded into multiple instructions in a branch delay slot
  arch/mips/dec/int-handler.S:198: Warning: macro instruction expanded into multiple instructions in a branch delay slot

Avoid this by open coding the PTR_LA macros.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

mm: memcontrol: avoid unused function warning

commit 358c07fcc3b60ab08d77f1684de8bd81bcf49a1a upstream.

A bugfix in v4.8-rc2 introduced a harmless warning when
CONFIG_MEMCG_SWAP is disabled but CONFIG_MEMCG is enabled:

  mm/memcontrol.c:4085:27: error: 'mem_cgroup_id_get_online' defined but not used [-Werror=unused-function]
   static struct mem_cgroup *mem_cgroup_id_get_online(struct mem_cgroup *memcg)

This moves the function inside of the #ifdef block that hides the
calling function, to avoid the warning.

Fixes: 1f47b61fb407 ("mm: memcontrol: fix swap counter leak on swapout from offline cgroup")
Link: http://lkml.kernel.org/r/20160824113733.2776701-1-arnd@arndb.de
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

cpmac: remove hopeless #warning

commit d43e6fb4ac4abfe4ef7c102833ed02330ad701e0 upstream.

The #warning was present 10 years ago when the driver first got merged.
As the platform is rather obsolete by now, it seems very unlikely that
the warning will cause anyone to fix the code properly.

kernelci.org reports the warning for every build in the meantime, so
I think it's better to just turn it into a code comment to reduce
noise.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: ralink: Remove unused rt*_wdt_reset functions

commit 886f9c69fc68f56ddea34d3de51ac1fc2ac8dfbc upstream.

All pointers to these functions were removed, so now they produce
warnings:

arch/mips/ralink/rt305x.c:92:13: error: 'rt305x_wdt_reset' defined but not used [-Werror=unused-function]

This removes the functions. If we need them again, the patch can be
reverted later.

Fixes: f576fb6a0700 ("MIPS: ralink: cleanup the soc specific pinmux data")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: John Crispin <john@phrozen.org>
Cc: Colin Ian King <colin.king@canonical.com>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/15044/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: ralink: Cosmetic change to prom_init().

commit 9c48568b3692f1a56cbf1935e4eea835e6b185b1 upstream.

Over the years the code has been changed various times leading to
argc/argv being defined in a different function to where we actually
use the variables. Clean this up by moving them to prom_init_cmdline().

Signed-off-by: John Crispin <john@phrozen.org>
Cc: linux-mips@linux-mips.org
Patchwork: https://patchwork.linux-mips.org/patch/14902/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

mtd: pmcmsp: use kstrndup instead of kmalloc+strncpy

commit 906b268477bc03daaa04f739844c120fe4dbc991 upstream.

kernelci.org reports a warning for this driver, as it copies a local
variable into a 'const char *' string:

    drivers/mtd/maps/pmcmsp-flash.c:149:30: warning: passing argument 1 of 'strncpy' discards 'const' qualifier from pointer target type [-Wdiscarded-qualifiers]

Using kstrndup() simplifies the code and avoids the warning.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Marek Vasut <marek.vasut@gmail.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: Update lemote2f_defconfig for CPU_FREQ_STAT change

commit b3f6046186ef45acfeebc5a59c9fb45cefc685e7 upstream.

Since linux-4.8, CPU_FREQ_STAT is a bool symbol, causing a warning in
kernelci.org:

arch/mips/configs/lemote2f_defconfig:42:warning: symbol value 'm' invalid for CPU_FREQ_STAT

This updates the defconfig to have the feature built-in.

Fixes: 1aefc75b2449 ("cpufreq: stats: Make the stats code non-modular")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/15000/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: ip22: Fix ip28 build for modern gcc

commit 23ca9b522383d3b9b7991d8586db30118992af4a upstream.

kernelci reports a failure of the ip28_defconfig build after upgrading its
gcc version:

arch/mips/sgi-ip22/Platform:29: *** gcc doesn't support needed option -mr10k-cache-barrier=store.  Stop.

The problem apparently is that the -mr10k-cache-barrier=store option is now
rejected for CPUs other than r10k. Explicitly including the CPU in the
check fixes this and is safe because both options were introduced in
gcc-4.4.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/15049/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: Update ip27_defconfig for SCSI_DH change

commit ea58fca1842a5dc410cae4167b01643db971a4e2 upstream.

Since linux-4.3, SCSI_DH is a bool symbol, causing a warning in
kernelci.org:

arch/mips/configs/ip27_defconfig:136:warning: symbol value 'm' invalid for SCSI_DH

This updates the defconfig to have the feature built-in.

Fixes: 086b91d052eb ("scsi_dh: integrate into the core SCSI code")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/15001/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: ip27: Disable qlge driver in defconfig

commit b617649468390713db1515ea79fc772d2eb897a8 upstream.

One of the last remaining failures in kernelci.org is for a gcc bug:

drivers/net/ethernet/qlogic/qlge/qlge_main.c:4819:1: error: insn does not satisfy its constraints:
drivers/net/ethernet/qlogic/qlge/qlge_main.c:4819:1: internal compiler error: in extract_constrain_insn, at recog.c:2190

This is apparently broken in gcc-6 but fixed in gcc-7, and I cannot
reproduce the problem here. However, it is clear that ip27_defconfig
does not actually need this driver as the platform has only PCI-X but
not PCIe, and the qlge adapter in turn is PCIe-only.

The driver was originally enabled in 2010 along with lots of other
drivers.

Fixes: 59d302b342e5 ("MIPS: IP27: Make defconfig useful again.")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/15197/
Signed-off-by: James Hogan <james.hogan@imgtec.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

MIPS: Update defconfigs for NF_CT_PROTO_DCCP/UDPLITE change

commit 9ddc16ad8e0bc7742fc96d5aaabc5b8698512cd1 upstream.

In linux-4.10-rc, NF_CT_PROTO_UDPLITE and NF_CT_PROTO_DCCP are bool
symbols instead of tristate, and kernelci.org reports a bunch of
warnings for this, like:

arch/mips/configs/malta_kvm_guest_defconfig:63:warning: symbol value 'm' invalid for NF_CT_PROTO_UDPLITE
arch/mips/configs/malta_defconfig:62:warning: symbol value 'm' invalid for NF_CT_PROTO_DCCP
arch/mips/configs/malta_defconfig:63:warning: symbol value 'm' invalid for NF_CT_PROTO_UDPLITE
arch/mips/configs/ip22_defconfig:70:warning: symbol value 'm' invalid for NF_CT_PROTO_DCCP
arch/mips/configs/ip22_defconfig:71:warning: symbol value 'm' invalid for NF_CT_PROTO_UDPLITE

This changes all the MIPS defconfigs with these symbols to have them
built-in.

Fixes: 9b91c96c5d1f ("netfilter: conntrack: built-in support for UDPlite")
Fixes: c51d39010a1b ("netfilter: conntrack: built-in support for DCCP")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Cc: linux-mips@linux-mips.org
Cc: linux-kernel@vger.kernel.org
Patchwork: https://patchwork.linux-mips.org/patch/14999/
Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

crypto: improve gcc optimization flags for serpent and wp512

commit 7d6e9105026788c497f0ab32fa16c82f4ab5ff61 upstream.

An ancient gcc bug (first reported in 2003) has apparently resurfaced
on MIPS, where kernelci.org reports an overly large stack frame in the
whirlpool hash algorithm:

crypto/wp512.c:987:1: warning: the frame size of 1112 bytes is larger than 1024 bytes [-Wframe-larger-than=]

With some testing in different configurations, I'm seeing large
variations in stack frames size up to 1500 bytes for what should have
around 300 bytes at most. I also checked the reference implementation,
which is essentially the same code but also comes with some test and
benchmarking infrastructure.

It seems that recent compiler versions on at least arm, arm64 and powerpc
have a partial fix for this problem, but enabling "-fsched-pressure", but
even with that fix they suffer from the issue to a certain degree. Some
testing on arm64 shows that the time needed to hash a given amount of
data is roughly proportional to the stack frame size here, which makes
sense given that the wp512 implementation is doing lots of loads for
table lookups, and the problem with the overly large stack is a result
of doing a lot more loads and stores for spilled registers (as seen from
inspecting the object code).

Disabling -fschedule-insns consistently fixes the problem for wp512,
in my collection of cross-compilers, the results are consistently better
or identical when comparing the stack sizes in this function, though
some architectures (notable x86) have schedule-insns disabled by
default.

The four columns are:
default: -O2
press:  -O2 -fsched-pressure
nopress: -O2 -fschedule-insns -fno-sched-pressure
nosched: -O2 -no-schedule-insns (disables sched-pressure)

default press nopress nosched
alpha-linux-gcc-4.9.3 1136 848 1136 176
am33_2.0-linux-gcc-4.9.3 2100 2076 2100 2104
arm-linux-gnueabi-gcc-4.9.3 848 848 1048 352
cris-linux-gcc-4.9.3 272 272 272 272
frv-linux-gcc-4.9.3 1128 1000 1128 280
hppa64-linux-gcc-4.9.3 1128 336 1128 184
hppa-linux-gcc-4.9.3 644 308 644 276
i386-linux-gcc-4.9.3 352 352 352 352
m32r-linux-gcc-4.9.3 720 656 720 268
microblaze-linux-gcc-4.9.3 1108 604 1108 256
mips64-linux-gcc-4.9.3 1328 592 1328 208
mips-linux-gcc-4.9.3 1096 624 1096 240
powerpc64-linux-gcc-4.9.3 1088 432 1088 160
powerpc-linux-gcc-4.9.3 1080 584 1080 224
s390-linux-gcc-4.9.3 456 456 624 360
sh3-linux-gcc-4.9.3 292 292 292 292
sparc64-linux-gcc-4.9.3 992 240 992 208
sparc-linux-gcc-4.9.3 680 592 680 312
x86_64-linux-gcc-4.9.3 224 240 272 224
xtensa-linux-gcc-4.9.3 1152 704 1152 304

aarch64-linux-gcc-7.0.0 224 224 1104 208
arm-linux-gnueabi-gcc-7.0.1 824 824 1048 352
mips-linux-gcc-7.0.0 1120 648 1120 272
x86_64-linux-gcc-7.0.1 240 240 304 240

arm-linux-gnueabi-gcc-4.4.7 840 392
arm-linux-gnueabi-gcc-4.5.4 784 728 784 320
arm-linux-gnueabi-gcc-4.6.4 736 728 736 304
arm-linux-gnueabi-gcc-4.7.4 944 784 944 352
arm-linux-gnueabi-gcc-4.8.5 464 464 760 352
arm-linux-gnueabi-gcc-4.9.3 848 848 1048 352
arm-linux-gnueabi-gcc-5.3.1 824 824 1064 336
arm-linux-gnueabi-gcc-6.1.1 808 808 1056 344
arm-linux-gnueabi-gcc-7.0.1 824 824 1048 352

Trying the same test for serpent-generic, the picture is a bit different,
and while -fno-schedule-insns is generally better here than the default,
-fsched-pressure wins overall, so I picked that instead.

default press nopress nosched
alpha-linux-gcc-4.9.3 1392 864 1392 960
am33_2.0-linux-gcc-4.9.3 536 524 536 528
arm-linux-gnueabi-gcc-4.9.3 552 552 776 536
cris-linux-gcc-4.9.3 528 528 528 528
frv-linux-gcc-4.9.3 536 400 536 504
hppa64-linux-gcc-4.9.3 524 208 524 480
hppa-linux-gcc-4.9.3 768 472 768 508
i386-linux-gcc-4.9.3 564 564 564 564
m32r-linux-gcc-4.9.3 712 576 712 532
microblaze-linux-gcc-4.9.3 724 392 724 512
mips64-linux-gcc-4.9.3 720 384 720 496
mips-linux-gcc-4.9.3 728 384 728 496
powerpc64-linux-gcc-4.9.3 704 304 704 480
powerpc-linux-gcc-4.9.3 704 296 704 480
s390-linux-gcc-4.9.3 560 560 592 536
sh3-linux-gcc-4.9.3 540 540 540 540
sparc64-linux-gcc-4.9.3 544 352 544 496
sparc-linux-gcc-4.9.3 544 344 544 496
x86_64-linux-gcc-4.9.3 528 536 576 528
xtensa-linux-gcc-4.9.3 752 544 752 544

aarch64-linux-gcc-7.0.0 432 432 656 480
arm-linux-gnueabi-gcc-7.0.1 616 616 808 536
mips-linux-gcc-7.0.0 720 464 720 488
x86_64-linux-gcc-7.0.1 536 528 600 536

arm-linux-gnueabi-gcc-4.4.7 592 440
arm-linux-gnueabi-gcc-4.5.4 776 448 776 544
arm-linux-gnueabi-gcc-4.6.4 776 448 776 544
arm-linux-gnueabi-gcc-4.7.4 768 448 768 544
arm-linux-gnueabi-gcc-4.8.5 488 488 776 544
arm-linux-gnueabi-gcc-4.9.3 552 552 776 536
arm-linux-gnueabi-gcc-5.3.1 552 552 776 536
arm-linux-gnueabi-gcc-6.1.1 560 560 776 536
arm-linux-gnueabi-gcc-7.0.1 616 616 808 536

I did not do any runtime tests with serpent, so it is possible that stack
frame size does not directly correlate with runtime performance here and
it actually makes things worse, but it's more likely to help here, and
the reduced stack frame size is probably enough reason to apply the patch,
especially given that the crypto code is often used in deep call chains.

Link: https://kernelci.org/build/id/58797d7559b5149efdf6c3a9/logs/
Link: http://www.larc.usp.br/~pbarreto/WhirlpoolPage.html
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=11488
Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79149
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: serial: digi_acceleport: fix OOB-event processing

commit 2e46565cf622dd0534a9d8bffe152a577b48d7aa upstream.

A recent change claimed to fix an off-by-one error in the OOB-port
completion handler, but instead introduced such an error. This could
specifically led to modem-status changes going unnoticed, effectively
breaking TIOCMGET.

Note that the offending commit fixes a loop-condition underflow and is
marked for stable, but should not be backported without this fix.

Reported-by: Ben Hutchings <ben@decadent.org.uk>
Fixes: 2d380889215f ("USB: serial: digi_acceleport: fix OOB data sanity
check")
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

USB: serial: digi_acceleport: fix OOB data sanity check

commit 2d380889215fe20b8523345649dee0579821800c upstream.

Make sure to check for short transfers to avoid underflow in a loop
condition when parsing the receive buffer.

Also fix an off-by-one error in the incomplete sanity check which could
lead to invalid data being parsed.

Fixes: 8c209e6782ca ("USB: make actual_length in struct urb field u32")
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Johan Hovold <johan@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

ARM64: dts: rk3368: p9: fix dc detect failed

Change-Id: Ia4cd3528725dd4de250b9417295e7705eb16c412
Signed-off-by: Weixin Zhou <zwx@rock-chips.com>

arm64: dts: rk3368-sheep: support mipi display

Change-Id: Ibeadd258ccbcd68a6c96fb08e7bfbcea79e5e6c6
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

arm64: dts: rk3368-android: reserve memory for drm-logo

Change-Id: I81b401d5561c67012f4d42d1640a6b1176490ca4
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

ARM64: dts: rk3368: p9: Add MIPI DSI panel support

Change-Id: I0d42d9ccd7ba09338c3074d1328ab5ec18079c27
Signed-off-by: WeiYong Bi <bivvy.bi@rock-chips.com>
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

arm64: dts: rk3368-android: enable display and vop

Change-Id: Iad1bd3544191d3badc8d0d8b8d9be363e3ac6ed6
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

clk: rockchip: rk3368: add CLK_SET_RATE_PARENT flag for dclk_vop

dclk_vop only allowed on NPLL.

Change-Id: I5325a530d3052de1e8685c39b90357291f0f4fb3
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>

nvmem: Fix dependencies for ROCKCHIP_EFUSE

On some rockchip platforms, need use secure interface to access efuse.

Change-Id: I49a4d5e547b689ff1665f1eb29a1dbbba5ef2595
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>

firmware: Kconfig: ROCKCHIP_SIP depends on HAVE_ARM_SMCCC and ARCH_ROCKCHIP

HAVE_ARM_SMCCC is default selected by ARM(if CPU_V7) or ARM64

Change-Id: I4bc64d4c98de5fad3179b3121b0f361d6337732c
Signed-off-by: chenjh <chenjh@rock-chips.com>

arm64: rockchip_defconfig: enable CONFIG_PHY_ROCKCHIP_INNO_MIPI_DPHY

Change-Id: Iba7bd03c86691670990102e2202bf5c4e2a718b8
Signed-off-by: WeiYong Bi <bivvy.bi@rock-chips.com>

clk: rockchip: rk3368: remove the flag ROCKCHIP_PLL_SYNC_RATE for NPLL

NPLL is used for vop dclk, sync rate flag would cause loader display
abnormal.

Change-Id: Ia170a8d0b7d1f39e2c9dcbc10b5d33fd1886d5f7
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>

ARM64: dts: rk3368-android: remove rkfb related nodes

Change-Id: I6a180419aabd705736fa1274c3463bad0cb95304
Signed-off-by: Zorro Liu <lyx@rock-chips.com>
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

arm64: dts: rk3368: don't assign clock rates for display pll

NPLL is used for display pixelclock, assign clock rates would overlap
loader pll setting, cause display abnormal.

Change-Id: Iaf1094c43526c7ca7b364608fa7153d03f84326c
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>

arm64: dts: rk3368: assign clock rates for aclk_vop and hclk_vop

Change-Id: I1d8559f09cd2df516aa8d479aa1b7407418916aa
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>

ARM64: dts: rk3368: Add MIPI DSI support

Change-Id: Ia74bb0726cb23acc914f976acf76849f0e764280
Signed-off-by: WeiYong Bi <bivvy.bi@rock-chips.com>

drm/rockchip/dsi: dw-mipi: Add support for RK3368 MIPI DSI Controller Host

Change-Id: I6c16b5a51451cdfc112a0bdefb44ad5a4b216c4f
Signed-off-by: WeiYong Bi <bivvy.bi@rock-chips.com>

phy: Add support for INNO MIPI D-PHY

The INNO MIPI D-PHY is built in witch a standard digital interface
to talk to any third part Host controller.That is part of Rockchip SoCs,
like rk3368.

Change-Id: I9806882e0e3fb6b20348015d0f34923d1bc46b89
Signed-off-by: WeiYong Bi <bivvy.bi@rock-chips.com>

arm64: dts: rockchip: rk3368 enable pmu node

Change-Id: I031fb437a84b19bb7cc389acb2404777f732cf6c
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

arm64: dts: rockchip: rk3368: add qos node

when pd power on/off, the qos regs need to save and restore.

Change-Id: I34146660e75609517463d679271386b536401b20
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>

firmware: rockchip: fix AARCH32 compile warning

Change-Id: I31924c9a1180d7fe034233c7ebd90413a7fa0fc3
Signed-off-by: chenjh <chenjh@rock-chips.com>

arm64: dts: rk3368: set higher voltage for gpu dvfs

As GPU share voltage domain with DDR and the minimum voltage of DDR
696MHz is 1100mV, GPU's voltage must be equal or greater than 1100mV.

After add ddr frequency scaling support, we can change them again.

Change-Id: I761931675265aac75425bf1cc9c7280a33f91e16
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

drm/bridge: dw_hdmi: clear ih_mute register when system resume

HDMI PD is power off when system suspend, so ih_mute register
bit0 mute_all_interrupt will be reset to 1 when system resume.
HPD interrupt will be mask, that would cause hdmi plugin could
not be detected.

Change-Id: I3bf2e6116e902cd516a7ac69fbe8569ca943e853
Signed-off-by: Bin Yang <yangbin@rock-chips.com>

arm64: dts: rockchip: modify cpu's opp table for rk3368

Change-Id: I2f7f15f9b3a9e6190e5e8895e9e4fe939d284b43
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>

FROMLIST: arm64: dts: rockchip: rk3368 swap clust0 and clust1

Before this patch, clust1 has little core0~3, clust0 has big core0~3
cpu_l | cpu0 |
cpu_l | cpu1 | clust1
cpu_l | cpu2 |
cpu_l | cpu3 |
----------------------
cpu_b | cpu4 |
cpu_b | cpu5 | clust0
cpu_b | cpu6 |
cpu_b | cpu7 |

With this patch, clust0 will have little core0~3, clust1 will have big core0~3
cpu_l | cpu0 |
cpu_l | cpu1 | clust0
cpu_l | cpu2 |
cpu_l | cpu3 |
----------------------
cpu_b | cpu4 |
cpu_b | cpu5 | clust1
cpu_b | cpu6 |
cpu_b | cpu7 |

It makes no other change, just keep same with other SoCs definations.

Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>
(am from https://patchwork.kernel.org/patch/9625109/)

Change-Id: I1beea4d3e75409d3a1f1614b0b86f1a929db4eee

arm64: dts: rockchip: add linux,pci-domain for PCIe

We need this to ask PCIe bus allocater to always
assign 0 to our root bus isntead of increasing it
, otherwise the hierarchy would be wrong if we unbind
and bind the root port.

Change-Id: I4ada61c89e617c7bccd92c5f9fa3334cae40603e
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>

UPSTREAM: PCI: Add pci_unmap_iospace() to unmap I/O resources

Add pci_unmap_iospace() to undo what pci_remap_iospace() did.

This is needed to support hotplug removal of host bridges that use
pci_remap_iospace().

Change-Id: Iee5d778cb8ddfedab59c55c227a8c60825786854
[bhelgaas: changelog]
Signed-off-by: Sinan Kaya <okaya@codeaurora.org>
Signed-off-by: Tomasz Nowicki <tn@semihalf.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Acked-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
(cherry picked from 4d3f13845957a87729a324cce8509fad8826ef5)
Conflicts:
drivers/pci/pci.c

UPSTREAM: PCI: rockchip: Fix rockchip_pcie_probe() error path to free resource list

rockchip_pcie_probe() calls of_pci_get_host_bridge_resources() to parse
resources from DT and build a resource list.  The caller is responsible for
disposing of the resource list.  This is normally done by
pci_release_host_bridge_dev() when the host bridge is removed.

If the host bridge probe fails, dispose of the resource list in the probe
error path.

Change-Id: Iefc17963a6ce99c64f2940d8dc2ba93bd00fe120
[bhelgaas: changelog]
Suggested-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
(cherry picked from f1d722b607d610b66785f7f00d2e1d457260647c)

PCI: rockchip: remove the warning log of 32bit-access only

That prevents me from doing git-am or git-cp patches from
linux-pci, but it was removed by a patch involving some other
platforms as well as a pci-core change, so I won't touch anything
else but pcie-rockchip.

Change-Id: I4268204b4c1d173a385bed8d1b6f7305bef0b5eb
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>

PCI: rockchip: remove redundant platform data assignment

We add it twice, so remove one.

Change-Id: I103ee8ffcc15569d65e9455a5465ac4b7dc5112d
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>

PCI: rockchip: add new dev variable to indent the log

Change-Id: Idc3bf972c7ab47187b115fb1d2efe7847ee82796
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>

clk: rockchip: rk3368: export SCLK_TIMERXX id for timers

Change-Id: I77fa21f29e7ff46e1bd4150845dfafe0a83b84c1
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>

ARM64: dts: rk3368: Update gpu opp table.

Change-Id: Id155a49d96e7533cb8c7f930c63528f452836b0e
Signed-off-by: ZhuangXiaoLiang <zhuangxl@rock-chips.com>

arm64: dts: rockchip: enable ehci/ohci and u2phy for rk3368-sheep

This adds enable ehci/ohci and u2phy configuration for rk3368-sheep.

Change-Id: Icafd7d0606ea199bdd8eac902b40459694b7f5a4
Signed-off-by: Frank Wang <frank.wang@rock-chips.com>

arm64: dts: rockchip: add vcc-host regulator for rk3368-sheep

This adds abstract vbus-host as a vcc-host regulator on rk3368-sheep.

Change-Id: I64deb38a3333346c47a5e2f499cec8d538d18baa
Signed-off-by: Frank Wang <frank.wang@rock-chips.com>

arm64: dts: rockchip: add ehci/ochi and u2phy nodes for rk3368

This adds configure ehci/ohci and u2phy nodes for rk3368 SoC.

Change-Id: I80cc311d7c14abc56084118baccf87501d44263e
Signed-off-by: Frank Wang <frank.wang@rock-chips.com>

phy: rockchip-inno-usb2: add support for rk3368 SoC

This adds support host-port on rk3368 SoC and amend phy Documentation.

Change-Id: I49a2efe37aad8b34505e4dac08336dc4231f4669
Signed-off-by: Frank Wang <frank.wang@rock-chips.com>

phy: rockchip-inno-usb2: amend sm work to support legacy SoC

This adds amend logic of sm work to compatibly support some legacy SoCs,
because _host_utmi_linestate_ and _host_utmi_hostdisconnect_ GRF status
bits which are required for host sm work were not introduced in these
SoCs.

Change-Id: Ib4f499f592618930ac5016a63b7a530674aa6005
Signed-off-by: Frank Wang <frank.wang@rock-chips.com>

clk: rockchip: rk3368: add 216M and 126M for armclkb and armclkl

support 216M/126M for armclkb and armclkl

Change-Id: I047ac24ad5a176923a55bd6934f06afcf272660d
Signed-off-by: Finley Xiao <finley.xiao@rock-chips.com>

ARM64: dts: rk3368: add vop display node

Change-Id: Ie747e90413fbfabe95e9d3c2ae55e02eff2e4708
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>

phy: rockchip-inno-usb2: don't power on otg phy in ls irq handler

The commit c590056b6ab8 ("phy: rockchip-inno-usb2: usb remote
wakeup support") power on otg phy in linestate irq handler,
this will cause usb peripheral fail to connect to PC in the
following case:
1. enable otg linestate irq
2. set system enter deep sleep
3. wakeup system by power key
4. connect usb peripheral to PC, pull up D+ to ~3V, trigger
   linestate irq and power on otg phy.
5. usb peripheral do BC1.2 detect, but PC try to enumerate
   the usb peripheral at the same time and fail at last.

Actually the usb controller drivers (e.g. dwc3 driver)
and otg_sm_work can manage the otg phy power consumption, so
it doesn't need to power on otg phy in linestate irq handler.

Change-Id: Ifd78e4d44ab96f07f75f063ed20af153b4027028
Signed-off-by: William Wu <wulf@rock-chips.com>

arm64: dts: rockchip: rk3368-p9: add ramp-delay for syr82x dcdc

Change-Id: I0a1cca68d6e40a881e153f824ccbeb611d006ff0
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>

usb: dwc_otg_310: pcd: fix isoc in ep transfer issue

When test usb gadget uvc function, we find a isoc in
ep transfer bug that will cause uvc data transfer fail.
The error case is:

1. The current EP request is done, call complete_ep()
to completes the request, and then call start_next_request()
to check the EP request queue, in this error case, the
queue is empty, so it doesn't start next request, just
set ep frame_num to 0xFFFFFFFF.

2. NAK Interrutp is triggered, check isoc ep frame_num
is 0xFFFFFFFF, then reset the frame_num to 0, and then
call start_next_request() to check the EP request queue,
in this error case, the queue is still empty, so set ep
frame_num to 0xFFFFFFFF again.

But afer the above operation, the current code will
modify the ep frame_num in NAK Interrutp handler by
add ep bInterval to frame_num, this cause frame_num
change again, but not keep in 0xFFFFFFFF, so the next
NAK Interrutp handler doesn't start next request any
more.

This patch reset the frame_num to the current frame
number got from DSTS SOFFN register if detect the
frame_num is 0xFFFFFFFF in NAK Interrutp handler.
And modify the frame_num in NAK Interrutp handler
only when the frame_num is not 0xFFFFFFFF.

TEST=Set usb gadget as webcam, use Ubuntu Guvcview
to preview the webcam, observe the preview screen
and the error log "There are no more ISOC requests".

Change-Id: I4403a67b1d5d257d092a2a71d5666c5d6fd5af3c
Signed-off-by: William Wu <wulf@rock-chips.com>