oota-llvm.git
7 years agollvm/unittests/Support/Path.cpp: Use <windows.h> instead of <Windows.h>.
NAKAMURA Takumi [Tue, 16 Jun 2015 06:46:16 +0000 (06:46 +0000)]
llvm/unittests/Support/Path.cpp: Use <windows.h> instead of <Windows.h>.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239804 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoX86: optimized i64 vector multiply with constant
Elena Demikhovsky [Tue, 16 Jun 2015 06:07:24 +0000 (06:07 +0000)]
X86: optimized i64 vector multiply with constant

When we multiply two 64-bit vectors, we extract lower and upper part and use the PMULUDQ instruction.
When one of the operands is a constant, the upper part may be zero, we know this at compile time.
Example: %a = mul <4 x i64> %b, <4 x i64> < i64 5, i64 5, i64 5, i64 5>.
I'm checking the value of the upper part and prevent redundant "multiply", "shift" and "add" operations.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239802 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[TableGen] Remove unused method declaration. NFC
Craig Topper [Tue, 16 Jun 2015 05:44:06 +0000 (05:44 +0000)]
[TableGen] Remove unused method declaration. NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239801 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRevert 239795
Philip Reames [Tue, 16 Jun 2015 01:20:53 +0000 (01:20 +0000)]
Revert 239795

I forgot to update some clang test cases.  I'll fix and resubmit tomorrow.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239800 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[AArch64] Generalize extract-high DUP extension to MOVI/MVNI.
Ahmed Bougacha [Tue, 16 Jun 2015 01:18:14 +0000 (01:18 +0000)]
[AArch64] Generalize extract-high DUP extension to MOVI/MVNI.

These are really immediate DUPs, and suffer from the same problem
with long instructions with a high/2 variant (e.g. smull).

By extending a MOVI (or DUP, before this patch), we can avoid an ext
on the other operand of the long instruction, e.g. turning:
    ext.16b v0, v0, v0, #8
    movi.4h v1, #0x53
    smull.4s  v0, v0, v1
into:
    movi.8h v1, #0x53
    smull2.4s  v0, v0, v1

While there, add a now-necessary combine to fold (VT NVCAST (VT x)).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239799 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[AArch64] Robustize neon-2velem-high test. NFC.
Ahmed Bougacha [Tue, 16 Jun 2015 01:05:39 +0000 (01:05 +0000)]
[AArch64] Robustize neon-2velem-high test. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239798 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMove logic from JumpThreading into LazyValue info to simplify caller.
Philip Reames [Tue, 16 Jun 2015 00:49:59 +0000 (00:49 +0000)]
Move logic from JumpThreading into LazyValue info to simplify caller.

This change is hopefully NFC. The only tricky part is that I changed the context instruction being used to the branch rather than the comparison. I believe both to be correct, but the branch is strictly more powerful. With the moved code, using the branch instruction is required for the basic block comparison test to return the same result. The previous code was able to directly access both the branch and the comparison where the revised code is not.

Differential Revision: http://reviews.llvm.org/D9652

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239797 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agomodules: Add explicit dependency on intrinsics_gen
Duncan P. N. Exon Smith [Tue, 16 Jun 2015 00:44:12 +0000 (00:44 +0000)]
modules: Add explicit dependency on intrinsics_gen

`LLVM_ENABLE_MODULES` builds sometimes fail because `Intrinsics.td`
needs to regenerate `Instrinsics.h` before anyone can include anything
from the LLVM_IR module.  Represent the dependency explicitly to prevent
that.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239796 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[InstCombine] Propagate non-null facts to call parameters
Philip Reames [Tue, 16 Jun 2015 00:43:54 +0000 (00:43 +0000)]
[InstCombine] Propagate non-null facts to call parameters

If a parameter to a function is known non-null, use the existing parameter attributes to record that fact at the call site. This has no optimization benefit by itself - that I know of - but is an enabling change for http://reviews.llvm.org/D9129.

Differential Revision: http://reviews.llvm.org/D9132

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239795 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agomodules: Mark CodeGen/DIEValues.def as a textual inclusion
Duncan P. N. Exon Smith [Tue, 16 Jun 2015 00:43:52 +0000 (00:43 +0000)]
modules: Mark CodeGen/DIEValues.def as a textual inclusion

Mark CodeGen/DIEValues.def as a textual inclusion to fix the
`LLVM_ENABLE_MODULES` build.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239794 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agomodules: Move ProfileKinds to an anonymous namespace
Duncan P. N. Exon Smith [Tue, 16 Jun 2015 00:43:04 +0000 (00:43 +0000)]
modules: Move ProfileKinds to an anonymous namespace

Fix a build failure with `LLVM_ENABLE_MODULES` due to
`ProfileData::instr` conflicting with a function `instr()` in
`<curses.h>`.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239793 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMIR Serialization: Print and parse simple machine function attributes.
Alex Lorenz [Tue, 16 Jun 2015 00:10:47 +0000 (00:10 +0000)]
MIR Serialization: Print and parse simple machine function attributes.

This commit serializes the simple, scalar attributes from the
'MachineFunction' class.

Reviewers: Duncan P. N. Exon Smith

Differential Revision: http://reviews.llvm.org/D10449

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239790 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMIR Serialization: move the MIR printer out of the MIR printing pass.
Alex Lorenz [Mon, 15 Jun 2015 23:52:35 +0000 (23:52 +0000)]
MIR Serialization: move the MIR printer out of the MIR printing pass.

This commit decouples the MIR printer and the MIR printing pass so
that it will be possible to move the MIR printer into a separate
machine IR library later on.

Reviewers: Duncan P. N. Exon Smith

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239788 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[X86] Try to shorten dwarf CFI emission
Reid Kleckner [Mon, 15 Jun 2015 23:45:08 +0000 (23:45 +0000)]
[X86] Try to shorten dwarf CFI emission

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239786 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoDebug Info IR: Switch DIObjCProperty to use DITypeRef.
Adrian Prantl [Mon, 15 Jun 2015 23:18:03 +0000 (23:18 +0000)]
Debug Info IR: Switch DIObjCProperty to use DITypeRef.
This is a prerequisite for turning on ODR type uniquing for ObjC++.

rdar://problem/21377883

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239780 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMIR Serialization: Create dummy functions when the MIR file doesn't have LLVM IR.
Alex Lorenz [Mon, 15 Jun 2015 23:07:38 +0000 (23:07 +0000)]
MIR Serialization: Create dummy functions when the MIR file doesn't have LLVM IR.

This commit creates a dummy LLVM IR function with one basic block and an unreachable
instruction for each parsed machine function when the MIR file doesn't have LLVM IR.
This change is required as the machine function analysis pass creates machine
functions only for the functions that are defined in the current LLVM module.

Reviewers: Duncan P. N. Exon Smith

Differential Revision: http://reviews.llvm.org/D10135

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239778 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMIR Serialization: Report an error when machine functions have the same name.
Alex Lorenz [Mon, 15 Jun 2015 22:23:23 +0000 (22:23 +0000)]
MIR Serialization: Report an error when machine functions have the same name.

This commit reports an error when the MIR parser encounters a machine
function with the name that is the same as the name of a different
machine function.

Reviewers: Duncan P. N. Exon Smith

Differential Revision: http://reviews.llvm.org/D10130

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239774 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAdd safestack attribute to LLVMAttribute enum and Go bindings. Correct
Peter Collingbourne [Mon, 15 Jun 2015 22:16:51 +0000 (22:16 +0000)]
Add safestack attribute to LLVMAttribute enum and Go bindings. Correct
constants in commented-out part of LLVMAttribute enum. Add tests that verify
that the safestack attribute is only allowed as a function attribute.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239772 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Hexagon] Using readobj rather than objdump.
Colin LeMahieu [Mon, 15 Jun 2015 21:57:41 +0000 (21:57 +0000)]
[Hexagon] Using readobj rather than objdump.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239770 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Hexagon] PC-relative offsets are relative to packet start rather than the offset...
Colin LeMahieu [Mon, 15 Jun 2015 21:52:13 +0000 (21:52 +0000)]
[Hexagon] PC-relative offsets are relative to packet start rather than the offset of the relocation.  Set relocation addend and check it's correct in the ELF.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239769 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[X86][SSE] Added tests for vector i8/i16 to f32/f64 conversions
Simon Pilgrim [Mon, 15 Jun 2015 21:49:31 +0000 (21:49 +0000)]
[X86][SSE] Added tests for vector i8/i16 to f32/f64 conversions

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239767 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoProtection against stack-based memory corruption errors using SafeStack
Peter Collingbourne [Mon, 15 Jun 2015 21:07:11 +0000 (21:07 +0000)]
Protection against stack-based memory corruption errors using SafeStack

This patch adds the safe stack instrumentation pass to LLVM, which separates
the program stack into a safe stack, which stores return addresses, register
spills, and local variables that are statically verified to be accessed
in a safe way, and the unsafe stack, which stores everything else. Such
separation makes it much harder for an attacker to corrupt objects on the
safe stack, including function pointers stored in spilled registers and
return addresses. You can find more information about the safe stack, as
well as other parts of or control-flow hijack protection technique in our
OSDI paper on code-pointer integrity (http://dslab.epfl.ch/pubs/cpi.pdf)
and our project website (http://levee.epfl.ch).

The overhead of our implementation of the safe stack is very close to zero
(0.01% on the Phoronix benchmarks). This is lower than the overhead of
stack cookies, which are supported by LLVM and are commonly used today,
yet the security guarantees of the safe stack are strictly stronger than
stack cookies. In some cases, the safe stack improves performance due to
better cache locality.

Our current implementation of the safe stack is stable and robust, we
used it to recompile multiple projects on Linux including Chromium, and
we also recompiled the entire FreeBSD user-space system and more than 100
packages. We ran unit tests on the FreeBSD system and many of the packages
and observed no errors caused by the safe stack. The safe stack is also fully
binary compatible with non-instrumented code and can be applied to parts of
a program selectively.

This patch is our implementation of the safe stack on top of LLVM. The
patches make the following changes:

- Add the safestack function attribute, similar to the ssp, sspstrong and
  sspreq attributes.

- Add the SafeStack instrumentation pass that applies the safe stack to all
  functions that have the safestack attribute. This pass moves all unsafe local
  variables to the unsafe stack with a separate stack pointer, whereas all
  safe variables remain on the regular stack that is managed by LLVM as usual.

- Invoke the pass as the last stage before code generation (at the same time
  the existing cookie-based stack protector pass is invoked).

- Add unit tests for the safe stack.

Original patch by Volodymyr Kuznetsov and others at the Dependable Systems
Lab at EPFL; updates and upstreaming by myself.

Differential Revision: http://reviews.llvm.org/D6094

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239761 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoDon't indent inside a namespace. NFC.
Rafael Espindola [Mon, 15 Jun 2015 21:04:27 +0000 (21:04 +0000)]
Don't indent inside a namespace. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239760 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoReplace @ with the more common \. NFC.
Rafael Espindola [Mon, 15 Jun 2015 21:02:49 +0000 (21:02 +0000)]
Replace @ with the more common \. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239759 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoDon't repeat names in comments and start functions with a lower case letter.
Rafael Espindola [Mon, 15 Jun 2015 20:55:37 +0000 (20:55 +0000)]
Don't repeat names in comments and start functions with a lower case letter.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239756 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMIR Serialization: Connect the machine function analysis pass to the MIR parser.
Alex Lorenz [Mon, 15 Jun 2015 20:30:22 +0000 (20:30 +0000)]
MIR Serialization: Connect the machine function analysis pass to the MIR parser.

This commit connects the machine function analysis pass (which creates machine
functions) to the MIR parser, which will initialize the machine functions
with the state from the MIR file and reconstruct the machine IR.

This commit introduces a new interface called 'MachineFunctionInitializer',
which can be used to provide custom initialization for the machine functions.

This commit also introduces a new diagnostic class called
'DiagnosticInfoMIRParser' which is used for MIR parsing errors.
This commit modifies the default diagnostic handling in LLVMContext - now the
the diagnostics are printed directly into llvm::errs() so that the MIR parsing
errors can be printed with colours.

Reviewers: Justin Bogner

Differential Revision: http://reviews.llvm.org/D9928

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239753 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRemove duplicate conditional in if-stmt.
Eric Christopher [Mon, 15 Jun 2015 20:16:53 +0000 (20:16 +0000)]
Remove duplicate conditional in if-stmt.

Fixes PR23839.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239751 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoCleanup the constructor of BitcodeReader. NFC.
Rafael Espindola [Mon, 15 Jun 2015 20:08:17 +0000 (20:08 +0000)]
Cleanup the constructor of BitcodeReader. NFC.

Use the same argument names as the members.
Use default member initializes.

Extracted from a patch by Karl Schimpf.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239749 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAdd "REQUIRES: asserts" to test case that uses -debug-only
Sanjoy Das [Mon, 15 Jun 2015 20:05:38 +0000 (20:05 +0000)]
Add "REQUIRES: asserts" to test case that uses -debug-only

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239748 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoUnbreak docs build from r239740.
Sanjoy Das [Mon, 15 Jun 2015 19:38:15 +0000 (19:38 +0000)]
Unbreak docs build from r239740.

Add FaultMaps.rst to toctree.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239747 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoUnbreak the build from r239740.
Sanjoy Das [Mon, 15 Jun 2015 19:29:44 +0000 (19:29 +0000)]
Unbreak the build from r239740.

Do not re-use an enum name as a field name.  Some bots don't like this.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239746 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Hexagon] Moving pass declarations out of header and in to implementation files....
Colin LeMahieu [Mon, 15 Jun 2015 19:05:35 +0000 (19:05 +0000)]
[Hexagon] Moving pass declarations out of header and in to implementation files.  Removing unused function getSubtargetInfo from HexagonMCCodeEmitter.cpp  Removing deletion of copy construction and assignment operator since parent already deletes it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239744 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[CodeGen] Add a pass to fold null checks into nearby memory operations.
Sanjoy Das [Mon, 15 Jun 2015 18:44:27 +0000 (18:44 +0000)]
[CodeGen] Add a pass to fold null checks into nearby memory operations.

Summary:
This change adds an "ImplicitNullChecks" target dependent pass.  This
pass folds null checks into memory operation using the FAULTING_LOAD
pseudo-op introduced in previous patches.

Depends on D10197
Depends on D10199
Depends on D10200

Reviewers: reames, rnk, pgavlin, JosephTremoulet, atrick

Reviewed By: atrick

Subscribers: ab, JosephTremoulet, llvm-commits

Differential Revision: http://reviews.llvm.org/D10201

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239743 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[TargetInstrInfo] Add new hook: AnalyzeBranchPredicate.
Sanjoy Das [Mon, 15 Jun 2015 18:44:21 +0000 (18:44 +0000)]
[TargetInstrInfo] Add new hook: AnalyzeBranchPredicate.

Summary:
NFC: no one uses AnalyzeBranchPredicate yet.

Add TargetInstrInfo::AnalyzeBranchPredicate and implement for x86.  A
later change adding support for page-fault based implicit null checks
depends on this.

Reviewers: reames, ab, atrick

Reviewed By: atrick

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D10200

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239742 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[TargetInstrInfo] Rename getLdStBaseRegImmOfs and implement for x86.
Sanjoy Das [Mon, 15 Jun 2015 18:44:14 +0000 (18:44 +0000)]
[TargetInstrInfo] Rename getLdStBaseRegImmOfs and implement for x86.

Summary:

TargetInstrInfo::getLdStBaseRegImmOfs to
TargetInstrInfo::getMemOpBaseRegImmOfs and implement for x86.  The
implementation only handles a few easy cases now and will be made more
sophisticated in the future.

This is NFCI: the only user of `getLdStBaseRegImmOfs` (now
`getmemOpBaseRegImmOfs`) is `LoadClusterMotion` and `LoadClusterMotion`
is disabled for x86.

Reviewers: reames, ab, MatzeB, atrick

Reviewed By: MatzeB, atrick

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D10199

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239741 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[CodeGen] Introduce a FAULTING_LOAD_OP pseudo-op.
Sanjoy Das [Mon, 15 Jun 2015 18:44:08 +0000 (18:44 +0000)]
[CodeGen] Introduce a FAULTING_LOAD_OP pseudo-op.

Summary:
This instruction encodes a loading operation that may fault, and a label
to branch to if the load page-faults.  The locations of potentially
faulting loads and their "handler" destinations are recorded in a
FaultMap section, meant to be consumed by LLVM's clients.

Nothing generates FAULTING_LOAD_OP instructions yet, but they will be
used in a future change.

The documentation (FaultMaps.rst) needs improvement and I will update
this diff with a more expanded version shortly.

Depends on D10196

Reviewers: rnk, reames, AndyAyers, ab, atrick, pgavlin

Reviewed By: atrick, pgavlin

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D10197

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239740 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[NFC] Extract X86MCInstLower::LowerMachineOperand.
Sanjoy Das [Mon, 15 Jun 2015 18:44:01 +0000 (18:44 +0000)]
[NFC] Extract X86MCInstLower::LowerMachineOperand.

Summary: Refactoring-only change that will be used later.

Reviewers: reames, atrick

Reviewed By: atrick

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D10196

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239739 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoDe-duplicate common expression, NFC.
Yaron Keren [Mon, 15 Jun 2015 17:03:35 +0000 (17:03 +0000)]
De-duplicate common expression, NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239736 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRangify several for loops, NFC.
Yaron Keren [Mon, 15 Jun 2015 16:20:16 +0000 (16:20 +0000)]
Rangify several for loops, NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239733 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoOn behalf of Alexandros Lamprineas:
Evgeny Astigeevich [Mon, 15 Jun 2015 15:48:44 +0000 (15:48 +0000)]
On behalf of Alexandros Lamprineas:

LLVM targeting aarch64 doesn't correctly produce aligned accesses for non-aligned
data at -O0/fast-isel (-mno-unaligned-access).
The root cause seems to be in fast-isel not producing unaligned access correctly
for -mno-unaligned-access.

The patch just aborts fast-isel for loads and stores when -mno-unaligned-access is
present.
The regression test is updated to check this new test case (-mno-unaligned-access
together with fast-isel).

Differential Revision: http://reviews.llvm.org/D10360

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239732 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[LinkerTest] Use LLVMDisposeMessage to free error string.
Benjamin Kramer [Mon, 15 Jun 2015 15:42:26 +0000 (15:42 +0000)]
[LinkerTest] Use LLVMDisposeMessage to free error string.

LLVMDisposeMessage is just a thing wrapper around free at the moment, but it's
the proper API to use here.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239731 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAvoid a "always true" warning from gcc.
Rafael Espindola [Mon, 15 Jun 2015 14:49:41 +0000 (14:49 +0000)]
Avoid a "always true" warning from gcc.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239729 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agogold-plugin: save the .o when given -save-temps.
Rafael Espindola [Mon, 15 Jun 2015 13:36:27 +0000 (13:36 +0000)]
gold-plugin: save the .o when given -save-temps.

The plugin now save the bitcode before and after optimizations and the
.o that is passed to the linker.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239726 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRevert r239721 - Replace string GNU Triples with llvm::Triple in InitMCObjectFileInfo...
Daniel Sanders [Mon, 15 Jun 2015 10:34:38 +0000 (10:34 +0000)]
Revert r239721 - Replace string GNU Triples with llvm::Triple in InitMCObjectFileInfo. NFC.

It appears to cause sparc-little-endian.s to assert on Windows and Darwin.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239724 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoReplace string GNU Triples with llvm::Triple in InitMCObjectFileInfo. NFC.
Daniel Sanders [Mon, 15 Jun 2015 09:19:41 +0000 (09:19 +0000)]
Replace string GNU Triples with llvm::Triple in InitMCObjectFileInfo. NFC.

Summary:
This affects other tools so the previous C++ API has been retained as a
deprecated function for the moment. Clang has been updated with a trivial
patch (not covered by the pre-commit review) to avoid breaking -Werror builds.
Other in-tree tools will be fixed with similar trivial patches.

This continues the patch series to eliminate StringRef forms of GNU triples
from the internals of LLVM that began in r239036.

Reviewers: rengolin

Reviewed By: rengolin

Subscribers: llvm-commits, rengolin

Differential Revision: http://reviews.llvm.org/D10366

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239721 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[MachineSink] Improve runtime performance. NFC.
Arnaud A. de Grandmaison [Mon, 15 Jun 2015 09:09:06 +0000 (09:09 +0000)]
[MachineSink] Improve runtime performance. NFC.

This patch fixes a compilation time issue, when MachineSink faces PHIs
with a huge number of operands. This can happen for example in goto table
based interpreters, where some basic blocks can have several of those PHIs,
each one with several hundreds operands. MachineSink was spending a
significant time re-building and re-sorting the list of successors of
the current MachineBasicBlock. The computing and sorting of the current
MachineBasicBlock successors is now cached.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239720 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[ValueTracking] do not overwrite analysis results already computed
Jingyue Wu [Mon, 15 Jun 2015 05:46:29 +0000 (05:46 +0000)]
[ValueTracking] do not overwrite analysis results already computed

Summary:
ValueTracking used to overwrite the analysis results computed from
assumes and dominating conditions. This patch fixes this issue.

Test Plan: test/Analysis/ValueTracking/assume.ll

Reviewers: hfinkel, majnemer

Reviewed By: majnemer

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D10283

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239718 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Support][Endian] Define |= and &= for u{big,little}{16,32,64}_t.
Rui Ueyama [Mon, 15 Jun 2015 03:00:15 +0000 (03:00 +0000)]
[Support][Endian] Define |= and &= for u{big,little}{16,32,64}_t.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239716 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[AArch64] Delete two empty files, which should be removed by r239713.
Hao Liu [Mon, 15 Jun 2015 02:56:40 +0000 (02:56 +0000)]
[AArch64] Delete two empty files, which should be removed by r239713.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239715 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[AArch64] Revert r239711 again. We need to discuss how to share code between AArch64...
Hao Liu [Mon, 15 Jun 2015 01:56:40 +0000 (01:56 +0000)]
[AArch64] Revert r239711 again. We need to discuss how to share code between AArch64 and ARM backend.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239713 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[AArch64] Match interleaved memory accesses into ldN/stN instructions.
Hao Liu [Mon, 15 Jun 2015 01:35:49 +0000 (01:35 +0000)]
[AArch64] Match interleaved memory accesses into ldN/stN instructions.

Re-commit after adding "-aarch64-neon-syntax=generic" to fix the failure on OS X.
This patch was firstly committed in r239514, then reverted in r239544 because of a syntax incompatible failure on OS X.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239711 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[CMake] Try to fix r239612, not to miss resources/windows_version_resource.rc in...
NAKAMURA Takumi [Sun, 14 Jun 2015 21:47:29 +0000 (21:47 +0000)]
[CMake] Try to fix r239612, not to miss resources/windows_version_resource.rc in clang build.

  - Who defines ${LLVM_SOURCE_DIR} ?
  - Would windows_version_resource.rc be available in an *installed* llvm tree?
    I suggest it may be installed in ${PREFIX}/share.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239703 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[InstSimplify] fsub nnan x, x -> 0.0 is valid without ninf
Benjamin Kramer [Sun, 14 Jun 2015 21:01:20 +0000 (21:01 +0000)]
[InstSimplify] fsub nnan x, x -> 0.0 is valid without ninf

Both inf - inf and (-inf) - (-inf) are NaN, so it's already covered by
nnan.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239702 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[InstSimplify] Add self-fdiv identities for -ffinite-math-only.
Benjamin Kramer [Sun, 14 Jun 2015 18:53:58 +0000 (18:53 +0000)]
[InstSimplify] Add self-fdiv identities for -ffinite-math-only.

When NaNs and Infs are ignored we can fold
 X /  X -> 1.0
-X /  X -> -1.0
 X / -X -> -1.0

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239701 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAVX-512: Implemented DAG lowering for shuff62x2/shufi62x2 instuctions ( Shuffle Packe...
Igor Breger [Sun, 14 Jun 2015 13:07:47 +0000 (13:07 +0000)]
AVX-512: Implemented DAG lowering for shuff62x2/shufi62x2 instuctions ( Shuffle Packed Values at 128-bit Granularity )
Tests added , vector-shuffle-512-v8.ll test re-generated.

Differential Revision: http://reviews.llvm.org/D10300

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239697 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAdd support for parsing the XOR operator in Intel syntax inline assembly.
Michael Kuperstein [Sun, 14 Jun 2015 12:59:45 +0000 (12:59 +0000)]
Add support for parsing the XOR operator in Intel syntax inline assembly.

Differential Revision: http://reviews.llvm.org/D10385
Patch by marina.yatsina@intel.com

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239695 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAVX-512: Implemented cvtsi2ss/d cvtusi2ss/d instructions with round control for KNL.
Igor Breger [Sun, 14 Jun 2015 12:44:55 +0000 (12:44 +0000)]
AVX-512: Implemented cvtsi2ss/d cvtusi2ss/d instructions with round control for KNL.
Added intrinsics for cvtsi2ss/d instructions.
Added tests for intrinsics and encoding.

Differential Revision: http://reviews.llvm.org/D10430

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239694 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAsmPrinter.cpp: Avoid crashes for targeting like "arm-mingw32". CurrentFnSym might...
NAKAMURA Takumi [Sun, 14 Jun 2015 00:23:40 +0000 (00:23 +0000)]
AsmPrinter.cpp: Avoid crashes for targeting like "arm-mingw32". CurrentFnSym might not be <MCSymbolELF> here.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239692 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoReformat.
NAKAMURA Takumi [Sun, 14 Jun 2015 00:23:33 +0000 (00:23 +0000)]
Reformat.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239691 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Hexagon] Adding some codegen tests and updating some to match spec.
Colin LeMahieu [Sat, 13 Jun 2015 21:46:39 +0000 (21:46 +0000)]
[Hexagon] Adding some codegen tests and updating some to match spec.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239690 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Statepoints] Skip a vector copy when uniquing values.
Benjamin Kramer [Sat, 13 Jun 2015 19:50:38 +0000 (19:50 +0000)]
[Statepoints] Skip a vector copy when uniquing values.

No functionality change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239688 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[ExecutionEngine] ArrayRefize argument passing.
Benjamin Kramer [Sat, 13 Jun 2015 19:50:29 +0000 (19:50 +0000)]
[ExecutionEngine] ArrayRefize argument passing.

No functionality change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239687 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoC++11 Rangify loops in AssemblyWriter::printModule, NFC.
Yaron Keren [Sat, 13 Jun 2015 17:50:47 +0000 (17:50 +0000)]
C++11 Rangify loops in AssemblyWriter::printModule, NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239686 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoDon't use std::errc.
Rafael Espindola [Sat, 13 Jun 2015 17:23:04 +0000 (17:23 +0000)]
Don't use std::errc.

As noted on Errc.h:

// * std::errc is just marked with is_error_condition_enum. This means that
//   common patters like AnErrorCode == errc::no_such_file_or_directory take
//   4 virtual calls instead of two comparisons.

And on some libstdc++ those virtual functions conclude that

------------------------
int main() {
  std::error_code foo = std::make_error_code(std::errc::no_such_file_or_directory);
  return foo == std::errc::no_such_file_or_directory;
}
-------------------------

should exit with 0.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239683 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[DAGCombiner] Added BSWAP(BSWAP(x)) -> x combine pattern.
Simon Pilgrim [Sat, 13 Jun 2015 16:25:12 +0000 (16:25 +0000)]
[DAGCombiner] Added BSWAP(BSWAP(x)) -> x combine pattern.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239682 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agohoist loop-invariant; NFCI
Sanjay Patel [Sat, 13 Jun 2015 15:33:15 +0000 (15:33 +0000)]
hoist loop-invariant; NFCI

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239681 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoremove function names from comments and clean up; NFC
Sanjay Patel [Sat, 13 Jun 2015 15:32:45 +0000 (15:32 +0000)]
remove function names from comments and clean up; NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239680 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[SelectionDAG] Added assertions + UNDEF handling for BSWAP node creation.
Simon Pilgrim [Sat, 13 Jun 2015 15:23:58 +0000 (15:23 +0000)]
[SelectionDAG] Added assertions + UNDEF handling for BSWAP node creation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239679 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoremove unnecessary casts; NFCI
Sanjay Patel [Sat, 13 Jun 2015 15:06:33 +0000 (15:06 +0000)]
remove unnecessary casts; NFCI

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239678 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[DAGCombiner] Added BSWAP vector constant folding support.
Simon Pilgrim [Sat, 13 Jun 2015 14:08:15 +0000 (14:08 +0000)]
[DAGCombiner] Added BSWAP vector constant folding support.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239675 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoStripped trailing whitespace. NFC.
Simon Pilgrim [Sat, 13 Jun 2015 12:57:36 +0000 (12:57 +0000)]
Stripped trailing whitespace. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239674 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[LinkerTest] Don't leak error string.
Benjamin Kramer [Sat, 13 Jun 2015 12:53:21 +0000 (12:53 +0000)]
[LinkerTest] Don't leak error string.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239673 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoStripped trailing whitespace. NFC.
Simon Pilgrim [Sat, 13 Jun 2015 12:51:39 +0000 (12:51 +0000)]
Stripped trailing whitespace. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239672 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoBring in a BumpPtrStringSaver from lld and simplify the interface.
Rafael Espindola [Sat, 13 Jun 2015 12:49:52 +0000 (12:49 +0000)]
Bring in a BumpPtrStringSaver from lld and simplify the interface.

StringSaver now always saves to a BumpPtrAllocator.

The only reason for having the virtual saveImpl is so lld can have a
thread safe version.

The reason for the distinct BumpPtrStringSaver class is to avoid the
virtual destructor.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239669 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[LIT] Fix failing LIT tests
Eric Fiselier [Sat, 13 Jun 2015 06:55:44 +0000 (06:55 +0000)]
[LIT] Fix failing LIT tests

Summary:
I spend some time trying to get the LIT test suite passing. Here are the changes that I needed to make on my machine.

I made the following changes for the following reasons.

1. google-test.py: The Google test format now checks for "[  PASSED  ] 1 test." to check if a test passes.
2. discovery.py: The output appears in a different order on my machine than it did in the test.
3. unittest-adaptor.py: The output appears in a different order on my machine than it did in the test.
4. The classname is now formed differently in `getJUnitXML(...)`.

I'm not sure what is causing the output order to differ in discovery.py and unittest-adaptor.py. Does anybody have any thoughts?

Reviewers: ddunbar, danalbert, jroelofs

Reviewed By: jroelofs

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D9864

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239663 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAMDGPU: s/R600/AMDGPU/ in the Makefiles
Tom Stellard [Sat, 13 Jun 2015 05:11:14 +0000 (05:11 +0000)]
AMDGPU: s/R600/AMDGPU/ in the Makefiles

Now the library names in the Makefiles match the library names in
LLVMBuild.txt.

This should hopefully fix the remaining bot failures.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239661 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoconfigure: Remove non-portable fall-through operator: ;&
Tom Stellard [Sat, 13 Jun 2015 03:46:48 +0000 (03:46 +0000)]
configure: Remove non-portable fall-through operator: ;&

This was added in r239657.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239660 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRename TargetSubtargetInfo::enablePostMachineScheduler() to enablePostRAScheduler()
Matthias Braun [Sat, 13 Jun 2015 03:42:16 +0000 (03:42 +0000)]
Rename TargetSubtargetInfo::enablePostMachineScheduler() to enablePostRAScheduler()

r213101 changed the behaviour of this method to not only affect the
PostMachineScheduler scheduler but also the PostRAScheduler scheduler,
renaming should make this fact clear. Also document that the preferred
way is to specify this in the scheduling model instead of overriding
this method.

Differential Revision: http://reviews.llvm.org/D10427

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239659 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMachineLICM: Use TargetSchedModel instead of just itineraries
Matthias Braun [Sat, 13 Jun 2015 03:42:11 +0000 (03:42 +0000)]
MachineLICM: Use TargetSchedModel instead of just itineraries

This will use Itinieraries if available, but will also work if just a
MCSchedModel is available.

Differential Revision: http://reviews.llvm.org/D10428

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239658 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoR600 -> AMDGPU rename
Tom Stellard [Sat, 13 Jun 2015 03:28:10 +0000 (03:28 +0000)]
R600 -> AMDGPU rename

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239657 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRevert 239644.
Matt Wala [Sat, 13 Jun 2015 01:08:00 +0000 (01:08 +0000)]
Revert 239644.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239650 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAArch64: map bare-metal arm64-macho triple to MachO MC layer.
Tim Northover [Fri, 12 Jun 2015 23:37:11 +0000 (23:37 +0000)]
AArch64: map bare-metal arm64-macho triple to MachO MC layer.

Far better than an assertion about expecting ELF.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239647 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoFix returning error message in LLVMLinkModules
Eli Bendersky [Fri, 12 Jun 2015 23:26:42 +0000 (23:26 +0000)]
Fix returning error message in LLVMLinkModules

On error, the temporary output stream wouldn't be flushed and therefore the
caller would see an empty error message.

Patch by Antoine Pitrou

Differential Revision: http://reviews.llvm.org/D10241

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239646 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Orc] Tidy up initialization based on review feedback for r239561 from dblaikie.
Lang Hames [Fri, 12 Jun 2015 23:13:06 +0000 (23:13 +0000)]
[Orc] Tidy up initialization based on review feedback for r239561 from dblaikie.

NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239645 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Scalarizer] Fix potential for stale data in Scattered across invocations
Matt Wala [Fri, 12 Jun 2015 22:49:11 +0000 (22:49 +0000)]
[Scalarizer] Fix potential for stale data in Scattered across invocations

Summary:
Scalarizer has two data structures that hold information about changes
to the function, Gathered and Scattered. These are cleared in finish()
at the end of runOnFunction() if finish() detects any changes to the
function.

However, finish() was checking for changes by only checking if
Gathered was non-empty. The function visitStore() only modifies
Scattered without touching Gathered. As a result, Scattered could have
ended up having stale data if Scalarizer only scalarized store
instructions. Since the data in Scattered is used during the execution
of the pass, this introduced dangling pointer errors.

The fix is to check whether both Scattered and Gathered are empty
before deciding what to do in finish().

Reviewers: srhines

Reviewed By: srhines

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D10422

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239644 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Orc] Tidy up the CompileOnDemand layer based on commit review from dblaikie.
Lang Hames [Fri, 12 Jun 2015 22:22:50 +0000 (22:22 +0000)]
[Orc] Tidy up the CompileOnDemand layer based on commit review from dblaikie.

NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239642 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Orc] Fix a bug in the CompileOnDemand layer where stub decls were not cloned
Lang Hames [Fri, 12 Jun 2015 21:31:15 +0000 (21:31 +0000)]
[Orc] Fix a bug in the CompileOnDemand layer where stub decls were not cloned
into partitions. Also, add an option to clone stub definitions (not just decls)
into partitions: these definitions could be inlined in some places to avoid the
overhead of calling via the stub.

Found by inspection - no test case yet, although I plan to add a unit test for
this once the CompileOnDemand layer refactoring settles down.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239640 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoR600/SI: Add assembler support for FLAT instructions
Tom Stellard [Fri, 12 Jun 2015 20:47:06 +0000 (20:47 +0000)]
R600/SI: Add assembler support for FLAT instructions

- Add glc, slc, and tfe operands to flat instructions
- Add missing flat instructions
- Fix the encoding of flat_load_dwordx3 and flat_store_dwordx3.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239637 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRangify several for loops in ValueEnumerator constructor.
Yaron Keren [Fri, 12 Jun 2015 20:18:20 +0000 (20:18 +0000)]
Rangify several for loops in ValueEnumerator constructor.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239636 91177308-0d34-0410-b5e6-96231b3b80d8

7 years ago[Hexagon] Making intrinsic tests agnostic to register allocation. Narrowing intrinsi...
Colin LeMahieu [Fri, 12 Jun 2015 19:57:32 +0000 (19:57 +0000)]
[Hexagon] Making intrinsic tests agnostic to register allocation.  Narrowing intrinsic parameters to appropriate width.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239634 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoWrap some long lines in LLVMBuild files. NFC
Douglas Katzman [Fri, 12 Jun 2015 18:44:57 +0000 (18:44 +0000)]
Wrap some long lines in LLVMBuild files. NFC

As suggested by jroelofs in a prior review (D9752),
it makes sense to generally prefer multi-line format.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239632 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAdd 'shave' processor name to Triple
Douglas Katzman [Fri, 12 Jun 2015 18:31:38 +0000 (18:31 +0000)]
Add 'shave' processor name to Triple

Based on ArchType, Clang's driver can select a non-Clang compiler.
String parsing in Clang would have sufficed if it were only that,
however this change anticipates true llvm support.

Differential Revision: http://reviews.llvm.org/D10413

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239631 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRefix a use of explicit pointer types in GEP constant folding
David Blaikie [Fri, 12 Jun 2015 18:22:03 +0000 (18:22 +0000)]
Refix a use of explicit pointer types in GEP constant folding

In the glorious future of opaque pointer types, it won't be possible to
retrieve the pointee type of a pointer type which is what's being done
in this GEP loop - but the first iteration is always a pointer type and
the loop doesn't care about that case, except whether or not the index
is a constant.

So pull that special case out before the loop and start at the second
iteration (index 1) instead.

Originally committed in r236670 and reverted with a test case in
r239015. This change keeps the test case working while also avoiding
depending on pointee types.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239629 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoFix a typo in a comment in MemCpyOpt (test commit)
Matt Wala [Fri, 12 Jun 2015 18:16:51 +0000 (18:16 +0000)]
Fix a typo in a comment in MemCpyOpt (test commit)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239628 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRangify two for loops in BitcodeReader.cpp.
Yaron Keren [Fri, 12 Jun 2015 18:13:20 +0000 (18:13 +0000)]
Rangify two for loops in BitcodeReader.cpp.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239627 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRemove unnecessary MCExpr.h include from MCSymbol.h
Pete Cooper [Fri, 12 Jun 2015 18:07:34 +0000 (18:07 +0000)]
Remove unnecessary MCExpr.h include from MCSymbol.h

MCSymbol.h already forwards declares MCExpr and only uses MCExpr* so doesn't
need to include the header.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239626 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRemove a bunch of inline keywords from User. NFC.
Pete Cooper [Fri, 12 Jun 2015 17:48:21 +0000 (17:48 +0000)]
Remove a bunch of inline keywords from User.  NFC.

This came up in the patch review for http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20150608/281362.html.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239624 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoMove OperandList to be allocated prior to User for hung off subclasses.
Pete Cooper [Fri, 12 Jun 2015 17:48:18 +0000 (17:48 +0000)]
Move OperandList to be allocated prior to User for hung off subclasses.

For hung off uses, we need a Use* to tell use where the operands are.
This was User::OperandList but we want to remove that to save space
of all subclasses which aren't making use of 'hung off uses'.

Hung off uses now allocate their own 'OperandList' Use* in the
User::new which they call.

getOperandList() now uses the hung off uses bit to work out where the
Use* for the OperandList lives.  If a User has hung off uses, then this
bit tells them to go back a single Use* from the User* and use that
value as the OperandList.

If a User has no hung off uses, then we get the first operand by
subtracting (NumOperands * sizeof(Use)) from the User this pointer.

This saves a pointer from User and all subclasses.  Given the average
size of a subclass of User is 112 or 128 bytes, this saves around 7% of space
With malloc tending to align to 16-bytes the real saving is typically more like 3.5%.

On 'opt -O2 verify-uselistorder.lto.bc', peak memory usage prior to this change
is 149MB and after is 143MB so the savings are around 2.5% of peak.

Looking at some passes which allocate many Instructions and Values, parseIR drops
from 54.25MB to 52.21MB while the Inliner calls to Instruction::clone() drops
from 28.20MB to 27.05MB.

Reviewed by Duncan Exon Smith.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239623 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoAdded a version of User::new for hung off uses.
Pete Cooper [Fri, 12 Jun 2015 17:48:14 +0000 (17:48 +0000)]
Added a version of User::new for hung off uses.

There are now 2 versions of User::new.  The first takes a size_t and is the current
implementation for subclasses which need 0 or more Use's allocated for their operands.

The new version takes no extra arguments to say that this subclass needs 'hung off uses'.
The HungOffUses bool is now set in this version of User::new and we can assert in
allocHungOffUses that we are allowed to have hung off uses.
This ensures we call the correct version of User::new for subclasses which need hung off uses.

A future commit will then allocate space for a single Use* which will be used
in place of User::OperandList once that field has been removed.

Reviewed by Duncan Exon Smith.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239622 91177308-0d34-0410-b5e6-96231b3b80d8

7 years agoRename NumOperands to make it clear its managed by the User. NFC.
Pete Cooper [Fri, 12 Jun 2015 17:48:10 +0000 (17:48 +0000)]
Rename NumOperands to make it clear its managed by the User. NFC.

This is to try make it very clear that subclasses shouldn't be changing
the value directly.  Now that OperandList for normal instructions is computed
using the NumOperands, its critical that the NumOperands is accurate or we
could compute the wrong offset to the first operand.

I looked over all places which update NumOperands and they are all safe.
Hung off use User's don't use NumOperands to compute the OperandList so they
are safe to continue to manipulate it.  The only other User which changed it
was GlobalVariable which has an optional init list but always allocated space
for a single Use.  It was correctly setting NumOperands to 1 before setting an
initializer, and setting it to 0 after clearing the init list, so the order was safe.

Added some comments to that code to make sure that this isn't changed in future
without being aware of this constraint.

Reviewed by Duncan Exon Smith.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@239621 91177308-0d34-0410-b5e6-96231b3b80d8