OSDN Git Service

android-x86/external-llvm.git
8 years ago[X86][SSE] Relax type assertions for matchVectorShuffleAsInsertPS
Simon Pilgrim [Sun, 10 Jul 2016 22:26:05 +0000 (22:26 +0000)]
[X86][SSE] Relax type assertions for matchVectorShuffleAsInsertPS

Calls to matchVectorShuffleAsInsertPS only need to ensure the inputs are 128-bit vectors. Only lowerVectorShuffleAsInsertPS needs to ensure that they are v4f32.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275028 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoLet FuncAttrs infer the 'returned' argument attribute
Hal Finkel [Sun, 10 Jul 2016 22:02:55 +0000 (22:02 +0000)]
Let FuncAttrs infer the 'returned' argument attribute

A function can have one argument with the 'returned' attribute, indicating that
the associated argument is always the return value of the function. Add
FuncAttrs inference logic.

Differential Revision: http://reviews.llvm.org/D22202

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275027 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoUpdate the LangRef description of the 'returned' attribute
Hal Finkel [Sun, 10 Jul 2016 21:52:39 +0000 (21:52 +0000)]
Update the LangRef description of the 'returned' attribute

The description of the 'returned' attribute says that it is only used when
code-generating the caller. I'd like to make the optimizer smarter about
looking through functions with returned arguments (generally, but motivated by
my llvm.noalias work). As David pointed out in the review of D22202, the
LangRef should be updated to make its expanded uses clearer.

Differential Revision: http://reviews.llvm.org/D22205

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275026 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[DAG] make isConstantSplatVector() available to the rest of lowering
Sanjay Patel [Sun, 10 Jul 2016 21:27:06 +0000 (21:27 +0000)]
[DAG] make isConstantSplatVector() available to the rest of lowering

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275025 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU/R600: Add implicitarg.ptr intrinsic
Jan Vesely [Sun, 10 Jul 2016 21:20:29 +0000 (21:20 +0000)]
AMDGPU/R600: Add implicitarg.ptr intrinsic

Differential Revision: http://reviews.llvm.org/D21622

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275024 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86][SSE] Add support for target shuffle combining to PSHUFLW/PSHUFHW
Simon Pilgrim [Sun, 10 Jul 2016 21:02:47 +0000 (21:02 +0000)]
[X86][SSE] Add support for target shuffle combining to PSHUFLW/PSHUFHW

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275022 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agofix documentation comments; NFC
Sanjay Patel [Sun, 10 Jul 2016 21:02:16 +0000 (21:02 +0000)]
fix documentation comments; NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275021 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[x86, SSE, AVX] add tests for icmp+zext (PR28484)
Sanjay Patel [Sun, 10 Jul 2016 20:45:14 +0000 (20:45 +0000)]
[x86, SSE, AVX] add tests for icmp+zext (PR28484)

Note the inconsistent vpbroadcast generation for AVX2; another bug.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275020 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86][SSE] Added tests for combining shuffles to PSHUFLW/PSHUFHW
Simon Pilgrim [Sun, 10 Jul 2016 20:19:56 +0000 (20:19 +0000)]
[X86][SSE] Added tests for combining shuffles to PSHUFLW/PSHUFHW

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275019 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[Support] Make helper function static. NFC.
Benjamin Kramer [Sun, 10 Jul 2016 16:11:53 +0000 (16:11 +0000)]
[Support] Make helper function static. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275017 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[SystemZ] Utilize Test Data Class instructions.
Marcin Koscielnicki [Sun, 10 Jul 2016 14:41:22 +0000 (14:41 +0000)]
[SystemZ] Utilize Test Data Class instructions.

This adds a new SystemZ-specific intrinsic, llvm.s390.tdc.f(32|64|128),
which maps straight to the test data class instructions.  A new IR pass
is added to recognize instructions that can be converted to TDC and
perform the necessary replacements.

Differential Revision: http://reviews.llvm.org/D21949

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275016 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoreformat, fix comments/names; NFCI
Sanjay Patel [Sun, 10 Jul 2016 13:05:57 +0000 (13:05 +0000)]
reformat, fix comments/names; NFCI

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275015 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoGive helper classes/functions internal linkage. NFC.
Benjamin Kramer [Sun, 10 Jul 2016 11:28:51 +0000 (11:28 +0000)]
Give helper classes/functions internal linkage. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275014 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[codeview] Drop unused private inheritance.
Benjamin Kramer [Sun, 10 Jul 2016 10:17:36 +0000 (10:17 +0000)]
[codeview] Drop unused private inheritance.

There is no polymorphism here, and StreamRef already contains a
StreamInterface pointer. Dropping the base class makes StreamRef more
transparent to the compiler, for example it can find unused variables.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275013 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[AVX512] Add support for lowering to 512-bit SHUFPS.
Craig Topper [Sun, 10 Jul 2016 05:55:53 +0000 (05:55 +0000)]
[AVX512] Add support for lowering to 512-bit SHUFPS.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275011 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[pdb] Sanity check the stream map
David Majnemer [Sun, 10 Jul 2016 05:32:05 +0000 (05:32 +0000)]
[pdb] Sanity check the stream map

Some abstractions in LLVM "know" that they are reading in-bounds,
FixedStreamArray, and provide a simple result.  This breaks down if the
stream map is bogus.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275010 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[llvm-pdbdump] Propagate errors a little more consistently
David Majnemer [Sun, 10 Jul 2016 03:34:47 +0000 (03:34 +0000)]
[llvm-pdbdump] Propagate errors a little more consistently

PDBFile::getBlockData didn't really return any indication that it
failed.  It merely returned an empty buffer.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275009 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[cmake] Change lli-child-target to use add_llvm_utility instead of add_llvm_executable.
Michael Gottesman [Sun, 10 Jul 2016 02:43:50 +0000 (02:43 +0000)]
[cmake] Change lli-child-target to use add_llvm_utility instead of add_llvm_executable.

We are currently using add_llvm_utility for executable targets that:

1. Are built by default.
2. Used for testing.
3. Are not installed by default.

Originally, lli-child-target used add_llvm_tool instead of add_llvm_executable
directly. This was changed so that lli-child-target would not be installed. This
was good since this is only used for testing and should never be installed for
users. This also had the unfortunate side effect that one can never turn off the
building of lli-child-target by default, a regression for projects that by
default do not want to compile any LLVM tools beyond tablegen/llvm-config.

This patch changes lli-child-target to use add_llvm_utility. This makes sense
since:

1. lli-child-target matches the semantics of executables created with
add_llvm_utility.
2. We fix the regression since now one can use the flag LLVM_BUILD_UTILS to
eliminate default compilation.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275008 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[cmake] Create the LLVM_BUILD_UTILS option.
Michael Gottesman [Sun, 10 Jul 2016 02:43:47 +0000 (02:43 +0000)]
[cmake] Create the LLVM_BUILD_UTILS option.

This option is the equivalent option to LLVM_BUILD_TOOLS but for executables
created via add_llvm_utility.

This is a useful tool for improving compile time in situations where LLVM is
used as a library and no testing tools are needed.

It follows the exact same implemention model as LLVM_BUILD_TOOLS.

Since the option is by default set to on, no behavior is changed unless one sets
it from the command line to be false.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275007 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[SCCP] Rename undefined -> unknown.
Davide Italiano [Sun, 10 Jul 2016 00:35:15 +0000 (00:35 +0000)]
[SCCP] Rename undefined -> unknown.

In the solver, isUndefined() does really mean "we don't know the
value yet" rather than "this is an UndefinedValue". Discussed with
Eli Friedman.

Differential Revision:  http://reviews.llvm.org/D22192

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275004 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[docs] Fix up a broken link.
Sean Silva [Sat, 9 Jul 2016 23:08:14 +0000 (23:08 +0000)]
[docs] Fix up a broken link.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275002 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[PM] Port LoopVectorize to the new PM.
Sean Silva [Sat, 9 Jul 2016 22:56:50 +0000 (22:56 +0000)]
[PM] Port LoopVectorize to the new PM.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@275000 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoFix up an include guard.
Sean Silva [Sat, 9 Jul 2016 22:56:39 +0000 (22:56 +0000)]
Fix up an include guard.

This should have been done as part of the move in r274960.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274999 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[SCCP] Remove wrong and misleading vector handling code.
Davide Italiano [Sat, 9 Jul 2016 22:49:35 +0000 (22:49 +0000)]
[SCCP] Remove wrong and misleading vector handling code.

This code was already commented out and it made some weird assumptions,
e.g. using isUndefined() as "this value is UndefValue" instead of
"we haven't computed this value is yet". Thanks to Eli Friedman for
pointing out where I was wrong (and where this code was wrong).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274995 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86][SSE] Add support for target shuffle combining to INSERTPS
Simon Pilgrim [Sat, 9 Jul 2016 21:47:55 +0000 (21:47 +0000)]
[X86][SSE] Add support for target shuffle combining to INSERTPS

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274990 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86][SSE] Use scaleShuffleMask helper. NFCI.
Simon Pilgrim [Sat, 9 Jul 2016 21:12:03 +0000 (21:12 +0000)]
[X86][SSE] Use scaleShuffleMask helper. NFCI.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274988 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86][SSE] Regenerate vector shift tests
Simon Pilgrim [Sat, 9 Jul 2016 20:55:20 +0000 (20:55 +0000)]
[X86][SSE] Regenerate vector shift tests

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274987 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[COFF, Dwarf] Don't emit DW_AT_location for dllimported entities
David Majnemer [Sat, 9 Jul 2016 20:47:48 +0000 (20:47 +0000)]
[COFF, Dwarf] Don't emit DW_AT_location for dllimported entities

There exists no relocation which can describe the address of a
dllimported variable: do not try to describe their location.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274986 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[SLSR] Fix crash on handling 128-bit integers.
Jingyue Wu [Sat, 9 Jul 2016 19:13:18 +0000 (19:13 +0000)]
[SLSR] Fix crash on handling 128-bit integers.

ConstantInt::getSExtValue may fail on >64-bit integers. Add checks to call
getSExtValue only on narrow integers.

As a minor aside, simplify slsr-gep.ll to remove unnecessary load instructions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274982 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agofix documentation comments; NFC
Sanjay Patel [Sat, 9 Jul 2016 18:52:07 +0000 (18:52 +0000)]
fix documentation comments; NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274981 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[lanai] Treat .t as optional in assembly parser for RR operands and add predicate...
Jacques Pienaar [Sat, 9 Jul 2016 18:26:04 +0000 (18:26 +0000)]
[lanai] Treat .t as optional in assembly parser for RR operands and add predicate operand to ShiftRR

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274980 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Move R600 only pieces into R600 classes
Matt Arsenault [Sat, 9 Jul 2016 18:11:15 +0000 (18:11 +0000)]
AMDGPU: Move R600 only pieces into R600 classes

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274979 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoRevert "AMDGPU: Remove unused control flow intrinsic"
Matt Arsenault [Sat, 9 Jul 2016 17:18:39 +0000 (17:18 +0000)]
Revert "AMDGPU: Remove unused control flow intrinsic"

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274978 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[ArgPromote] Use function_ref and for-range loops.
Benjamin Kramer [Sat, 9 Jul 2016 10:36:36 +0000 (10:36 +0000)]
[ArgPromote] Use function_ref and for-range loops.

No functionality change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274973 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Merge / reorganize tests
Matt Arsenault [Sat, 9 Jul 2016 08:02:28 +0000 (08:02 +0000)]
AMDGPU: Merge / reorganize tests

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274972 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Simplify tests with per function subtargets
Matt Arsenault [Sat, 9 Jul 2016 07:55:03 +0000 (07:55 +0000)]
AMDGPU: Simplify tests with per function subtargets

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274971 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Prune AMDGPUAsmParser in libdeps.
NAKAMURA Takumi [Sat, 9 Jul 2016 07:54:27 +0000 (07:54 +0000)]
AMDGPU: Prune AMDGPUAsmParser in libdeps.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274970 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Fix fdiv lowering when f32 denormals supported
Matt Arsenault [Sat, 9 Jul 2016 07:48:11 +0000 (07:48 +0000)]
AMDGPU: Fix fdiv lowering when f32 denormals supported

Also fix test not actually using function labels.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274969 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86] Remove sse41 extract intrinsics. They are not used by clang and are not impleme...
Craig Topper [Sat, 9 Jul 2016 04:38:30 +0000 (04:38 +0000)]
[X86] Remove sse41 extract intrinsics. They are not used by clang and are not implemented by the x86 backend.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274967 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86] Remove and autoupgrade 512-bit non-temporal store intrinsics.
Craig Topper [Sat, 9 Jul 2016 04:38:27 +0000 (04:38 +0000)]
[X86] Remove and autoupgrade 512-bit non-temporal store intrinsics.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274966 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[LoopSimplify] Remove a comment which is unlikely to age well.
Davide Italiano [Sat, 9 Jul 2016 03:27:24 +0000 (03:27 +0000)]
[LoopSimplify] Remove a comment which is unlikely to age well.

Chandler pointed out in his review but I forgot to remove before
committing, my bad.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274963 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[PM] Port CrossDSOCFI to the new pass manager.
Davide Italiano [Sat, 9 Jul 2016 03:25:35 +0000 (03:25 +0000)]
[PM] Port CrossDSOCFI to the new pass manager.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274962 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[CFLAA] Make a constant variable `const`. NFC.
George Burgess IV [Sat, 9 Jul 2016 03:21:25 +0000 (03:21 +0000)]
[CFLAA] Make a constant variable `const`. NFC.

`const` was dropped by r274958, and the lack of `const` makes GCC6
(correctly) complain.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274961 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[PM] Fix a think-o. mv {Scalar,Vectorize}/SLPVectorize.h
Sean Silva [Sat, 9 Jul 2016 03:11:29 +0000 (03:11 +0000)]
[PM] Fix a think-o. mv {Scalar,Vectorize}/SLPVectorize.h

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274960 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[PM] Port LoopSimplify to the new pass manager.
Davide Italiano [Sat, 9 Jul 2016 03:03:01 +0000 (03:03 +0000)]
[PM] Port LoopSimplify to the new pass manager.

While here move simplifyLoop() function to the new header, as
suggested by Chandler in the review.

Differential Revision:  http://reviews.llvm.org/D21404

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274959 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[CFLAA] Move the graph builder out from CFLSteens. NFC.
George Burgess IV [Sat, 9 Jul 2016 02:54:42 +0000 (02:54 +0000)]
[CFLAA] Move the graph builder out from CFLSteens. NFC.

Patch by Jia Chen.

Differential Revision: http://reviews.llvm.org/D22022

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274958 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[CFLAA] Simplify CFLGraphBuilder. NFC.
George Burgess IV [Sat, 9 Jul 2016 02:48:56 +0000 (02:48 +0000)]
[CFLAA] Simplify CFLGraphBuilder. NFC.

This removes a few fields from the graph builder by making us compute
things (that we'd always compute anyway) more eagerly.

Patch by Jia Chen.

Differential Revision: http://reviews.llvm.org/D22009

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274957 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoLivePhysRegs: addLiveOuts() can skip addPristines() in ret block
Matthias Braun [Sat, 9 Jul 2016 01:31:36 +0000 (01:31 +0000)]
LivePhysRegs: addLiveOuts() can skip addPristines() in ret block

Drive-by improvement: We would 1) add CSRs, 2) remove callee saved CSRs
and 3) add all CSRs again for the return block.  Just adding CSRs once
obviously gives the same results.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274955 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Improve offset folding for register indexing
Matt Arsenault [Sat, 9 Jul 2016 01:13:56 +0000 (01:13 +0000)]
AMDGPU: Improve offset folding for register indexing

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274954 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Simplify isSchedulingBoundary
Matt Arsenault [Sat, 9 Jul 2016 01:13:51 +0000 (01:13 +0000)]
AMDGPU: Simplify isSchedulingBoundary

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274953 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoVirtRegMap: Replace some identity copies with KILL instructions.
Matthias Braun [Sat, 9 Jul 2016 00:19:07 +0000 (00:19 +0000)]
VirtRegMap: Replace some identity copies with KILL instructions.

An identity COPY like this:
   %AL = COPY %AL, %EAX<imp-def>
has no semantic effect, but encodes liveness information: Further users
of %EAX only depend on this instruction even though it does not define
the full register.

Replace the COPY with a KILL instruction in those cases to maintain this
liveness information. (This reverts a small part of r238588 but this
time adds a comment explaining why a KILL instruction is useful).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274952 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAsmPrinter: Fix emitKill() not flush()ing a raw_string_ostream
Matthias Braun [Sat, 9 Jul 2016 00:18:43 +0000 (00:18 +0000)]
AsmPrinter: Fix emitKill() not flush()ing a raw_string_ostream

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274951 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAdded REQUIRES to TestingGuide documentation
Piotr Padlewski [Fri, 8 Jul 2016 23:47:29 +0000 (23:47 +0000)]
Added REQUIRES to TestingGuide  documentation

Reviewers: alexfh, wolfgangp, rengolin

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D22172

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274949 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAdd 'thinlto_src_module' md with asserts or -enable-import-metadata
Piotr Padlewski [Fri, 8 Jul 2016 23:01:49 +0000 (23:01 +0000)]
Add 'thinlto_src_module' md with asserts or -enable-import-metadata

Summary:
This way the metadata will be only generated when asserts enabled,
or when -enable-import-metadata specified

FIXED missing colon on requires.

Reviewers: tejohnson, eraman, mehdi_amini

Subscribers: mehdi_amini, llvm-commits

Differential Revision: http://reviews.llvm.org/D22167

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274947 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoRevert "Add 'thinlto_src_module' md with asserts or -enable-import-metadata"
Piotr Padlewski [Fri, 8 Jul 2016 22:55:48 +0000 (22:55 +0000)]
Revert "Add 'thinlto_src_module' md with asserts or -enable-import-metadata"

Reverting because of 17463
http://lab.llvm.org:8011/builders/clang-x86_64-linux-selfhost-modules/builds/17463

This reverts commit d20cb431bba2ba43b4c65a8556cff445bfefbb7c.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274946 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[lanai] Update test to use peephole-opt and not peephole-opts
Jacques Pienaar [Fri, 8 Jul 2016 22:28:29 +0000 (22:28 +0000)]
[lanai] Update test to use peephole-opt and not peephole-opts

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274945 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoRevert "InstCombine rule to fold truncs whose value is available"
Anna Thomas [Fri, 8 Jul 2016 22:15:08 +0000 (22:15 +0000)]
Revert "InstCombine rule to fold truncs whose value is available"

This reverts commit r274853.
Caused failure in ppcBE build

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274943 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoLanai: Avoid implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 22:11:30 +0000 (22:11 +0000)]
Lanai: Avoid implicit iterator conversions, NFC

Avoid implicit conversions from MachineInstrBundleIterator to
MachineInstr* in the Lanai backend.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274942 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[MC, COFF] Permit a variable to be redefined
David Majnemer [Fri, 8 Jul 2016 21:54:16 +0000 (21:54 +0000)]
[MC, COFF] Permit a variable to be redefined

Our assertions in WinCOFFStreamer had unexpected side effects resulting
in symbols getting unexpectedly marked as used.

This fixes PR28462.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274941 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[TTI] Expose TTI::getGEPCost and use it in SLSR and NaryReassociate.
Jingyue Wu [Fri, 8 Jul 2016 21:48:05 +0000 (21:48 +0000)]
[TTI] Expose TTI::getGEPCost and use it in SLSR and NaryReassociate.

NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274940 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Remove unused control flow intrinsic
Matt Arsenault [Fri, 8 Jul 2016 21:39:44 +0000 (21:39 +0000)]
AMDGPU: Remove unused control flow intrinsic

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274939 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAdd 'thinlto_src_module' md with asserts or -enable-import-metadata
Piotr Padlewski [Fri, 8 Jul 2016 21:25:39 +0000 (21:25 +0000)]
Add 'thinlto_src_module' md with asserts or -enable-import-metadata

Summary:
This way the metadata will be only generated when asserts enabled,
or when -enable-import-metadata specified

Reviewers: tejohnson, eraman, mehdi_amini

Subscribers: mehdi_amini, llvm-commits

Differential Revision: http://reviews.llvm.org/D22167

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274938 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoReapply r274829 with fix for FP vectors
Matt Arsenault [Fri, 8 Jul 2016 21:25:33 +0000 (21:25 +0000)]
Reapply r274829 with fix for FP vectors

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274937 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[BFI] Minor cleanup. NFC
Adam Nemet [Fri, 8 Jul 2016 21:24:13 +0000 (21:24 +0000)]
[BFI] Minor cleanup. NFC

Use typedef Result in BlockFrequencyAnalysis::run.  Fix typo in comment.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274936 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[LAA] Port test to the new PM
Adam Nemet [Fri, 8 Jul 2016 21:24:06 +0000 (21:24 +0000)]
[LAA] Port test to the new PM

This is a follow-on to r274452.

The LAA with the new PM is a loop pass so we go from inner to outer loops.

Also using a CHECK-NOT didn't make much sense because we print something
in either case; whether an invariant is 'found' or 'not found'.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274935 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[PM] name the new PM LAA class LoopAccessAnalysis (LAA) /NFC
Xinliang David Li [Fri, 8 Jul 2016 21:21:44 +0000 (21:21 +0000)]
[PM] name the new PM LAA class LoopAccessAnalysis (LAA) /NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274934 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoMSP430: Avoid implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 21:19:46 +0000 (21:19 +0000)]
MSP430: Avoid implicit iterator conversions, NFC

Avoid implicit conversions from MachineInstrBundleIIterator to
MachineInstr* in the MSP430 backend by preferring MachineInstr& over
MachineInstr* when a pointer isn't nullable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274933 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[InstCombine] don't form select from bitcasted logic ops if bitcasts have >1 use
Sanjay Patel [Fri, 8 Jul 2016 21:17:51 +0000 (21:17 +0000)]
[InstCombine] don't form select from bitcasted logic ops if bitcasts have >1 use

This isn't a sure thing (are 2 extra bitcasts less expensive than a logic op?),
but we'll try to err on the conservative side by going with the case that has
less IR instructions.

Note: This question came up in http://reviews.llvm.org/D22114 , but this part is
independent of that patch proposal, so I'm making this small change ahead of that
one.

See also:
http://reviews.llvm.org/rL274926

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274932 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoNVPTX: Avoid implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 21:10:58 +0000 (21:10 +0000)]
NVPTX: Avoid implicit iterator conversions, NFC

Avoid implicit conversions from MachineInstrBundleIterator to
MachineInstr* in the NVPTX backend, mainly by preferring MachineInstr&
over MachineInstr* when a pointer isn't nullable and using range-based
for loops.

There was one piece of questionable code in
NVPTXInstrInfo::AnalyzeBranch, where a condition checked a pointer
converted from an iterator for nullptr.  Since this case is impossible
(moreover, the code above guarantees that the iterator is valid), I
removed the check when I changed the pointer to a reference.

Despite that case, there should be no functionality change here.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274931 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoadd another multi-use test for logic->select transform
Sanjay Patel [Fri, 8 Jul 2016 21:08:16 +0000 (21:08 +0000)]
add another multi-use test for logic->select transform

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274929 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAllow dead insts to be kept in DeadRemat only when they are rematerializable.
Wei Mi [Fri, 8 Jul 2016 21:08:09 +0000 (21:08 +0000)]
Allow dead insts to be kept in DeadRemat only when they are rematerializable.

Because isReallyTriviallyReMaterializableGeneric puts many limits on
rematerializable instructions, this fix can prevent instructions with
tied virtual operands and instructions with virtual register uses from
being kept in DeadRemat, so as to workaround the live interval consistency
problem for the dummy instructions kept in DeadRemat.

But we still need to fix the live interval consistency problem. This patch
is just a short time relieve. PR28464 has been filed as a reminder.

Differential Revision: http://reviews.llvm.org/D19486

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274928 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoRename LoopAccessAnalysis to LoopAccessLegacyAnalysis /NFC
Xinliang David Li [Fri, 8 Jul 2016 20:55:26 +0000 (20:55 +0000)]
Rename LoopAccessAnalysis to LoopAccessLegacyAnalysis /NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274927 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[InstCombine] don't form select from logic ops if it's unlikely that we'll eliminate...
Sanjay Patel [Fri, 8 Jul 2016 20:53:29 +0000 (20:53 +0000)]
[InstCombine] don't form select from logic ops if it's unlikely that we'll eliminate any ops

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274926 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoadjust test so it won't completely optimize away
Sanjay Patel [Fri, 8 Jul 2016 20:35:53 +0000 (20:35 +0000)]
adjust test so it won't completely optimize away

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274925 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAArch64: Avoid implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 20:29:42 +0000 (20:29 +0000)]
AArch64: Avoid implicit iterator conversions, NFC

Avoid implicit conversions from MachineInstrBundleInstr to MachineInstr*
in the AArch64 backend, mainly by preferring MachineInstr& over
MachineInstr* when a pointer isn't nullable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274924 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoadd tests for multi-use folding to select
Sanjay Patel [Fri, 8 Jul 2016 20:22:27 +0000 (20:22 +0000)]
add tests for multi-use folding to select

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274922 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoRemove duplicate inclusion /NFC
Xinliang David Li [Fri, 8 Jul 2016 20:21:32 +0000 (20:21 +0000)]
Remove duplicate inclusion /NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274921 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoARM: Remove implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 20:21:17 +0000 (20:21 +0000)]
ARM: Remove implicit iterator conversions, NFC

Remove remaining implicit conversions from MachineInstrBundleIterator to
MachineInstr* from the ARM backend.  In most cases, I made them less attractive
by preferring MachineInstr& or using a ranged-based for loop.

Once all the backends are fixed I'll make the operator explicit so that this
doesn't bitrot back.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274920 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoTableGen: Update style in CodeGenIntrinsics. NFC
Justin Bogner [Fri, 8 Jul 2016 20:14:27 +0000 (20:14 +0000)]
TableGen: Update style in CodeGenIntrinsics. NFC

Ran clang-format to remove the namespace indentation, and stopped
repeating names in doc comments since I was updating every line
anyway.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274919 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoRemove inline hints computation from SampleProfile.cpp
Dehao Chen [Fri, 8 Jul 2016 20:12:44 +0000 (20:12 +0000)]
Remove inline hints computation from SampleProfile.cpp

Summary: As we will move to use uniformed hotness check in inliner, we do not need inline hints in SampleProfile pass any more.

Reviewers: dnovillo, davidxl

Subscribers: eraman, llvm-commits

Differential Revision: http://reviews.llvm.org/D19287

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274918 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoRevert r274829, it caused PR28472.
Nico Weber [Fri, 8 Jul 2016 19:52:19 +0000 (19:52 +0000)]
Revert r274829, it caused PR28472.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274916 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86] Regenerated bitreverse tests to demonstrate what is going on.
Simon Pilgrim [Fri, 8 Jul 2016 19:51:08 +0000 (19:51 +0000)]
[X86] Regenerated bitreverse tests to demonstrate what is going on.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274915 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86] Added bitreverse tests for non-legal types
Simon Pilgrim [Fri, 8 Jul 2016 19:48:33 +0000 (19:48 +0000)]
[X86] Added bitreverse tests for non-legal types

Requested on D21578

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274914 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoSparc: Avoid implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:41:40 +0000 (19:41 +0000)]
Sparc: Avoid implicit iterator conversions, NFC

Remove the only implicit conversions from MachineInstrBundleIterator to
MachineInstr* in the Sparc backend.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274913 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoWebAssembly: Avoid implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:36:40 +0000 (19:36 +0000)]
WebAssembly: Avoid implicit iterator conversions, NFC

Avoid implicit conversions from MachineInstrBundleIterator to
MachineInstr* in the WebAssembly backend by preferring MachineInstr&
over MachineInstr*.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274912 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAsmPrinter: Avoid implicit iterator conversions in DbgValueHistoryCalculator, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:31:47 +0000 (19:31 +0000)]
AsmPrinter: Avoid implicit iterator conversions in DbgValueHistoryCalculator, NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274911 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[CrossDSOCFI] Change the pass so that it doesn't require doInitialization()
Davide Italiano [Fri, 8 Jul 2016 19:30:06 +0000 (19:30 +0000)]
[CrossDSOCFI] Change the pass so that it doesn't require doInitialization()

Differential Revision:  http://reviews.llvm.org/D21357

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274910 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[X86][AVX2] Add support for target shuffle combining to VPERMPD/VPERMQ
Simon Pilgrim [Fri, 8 Jul 2016 19:23:29 +0000 (19:23 +0000)]
[X86][AVX2] Add support for target shuffle combining to VPERMPD/VPERMQ

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274908 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoSelectionDAG: Avoid implicit iterator conversions in SelectionDAGBuilder, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:23:12 +0000 (19:23 +0000)]
SelectionDAG: Avoid implicit iterator conversions in SelectionDAGBuilder, NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274907 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Remove implicit iterator conversions, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:16:05 +0000 (19:16 +0000)]
AMDGPU: Remove implicit iterator conversions, NFC

Remove remaining implicit conversions from MachineInstrBundleIterator to
MachineInstr* from the AMDGPU backend.  In most cases, I made them less
attractive by preferring MachineInstr& or using a ranged-based for loop.

Once all the backends are fixed I'll make the operator explicit so that
this doesn't bitrot back.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274906 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[SCCP] Fold constants as we build them whne visiting cast instructions.
Davide Italiano [Fri, 8 Jul 2016 19:13:40 +0000 (19:13 +0000)]
[SCCP] Fold constants as we build them whne visiting cast instructions.

This should be slightly more efficient and could avoid spurious overdefined
markings, as Eli pointed out.

Differential Revision:  http://reviews.llvm.org/D22122

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274905 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoSelectionDAG: Avoid implicit iterator conversions in SelectionDAGISel, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:11:40 +0000 (19:11 +0000)]
SelectionDAG: Avoid implicit iterator conversions in SelectionDAGISel, NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274904 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoSelectionDAG: Avoid implicit iterator conversions in ScheduleDAGSDNodes, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:07:09 +0000 (19:07 +0000)]
SelectionDAG: Avoid implicit iterator conversions in ScheduleDAGSDNodes, NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274903 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoAMDGPU: Make infinite loop clear, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 19:00:17 +0000 (19:00 +0000)]
AMDGPU: Make infinite loop clear, NFC

Change a while loop that was checking for nullptr on an
iterator-to-pointer conversion to an infinite for loop.  Now it's clear
that the condition doesn't terminate.

The only change in behaviour is if an invalid iterator (holding nullptr)
was passed into AMDGPUCFGStructurizer::reversePredicateSetter.  There
are only two callers, and they both dereference the iterator before
sending it in, so rather than adding an early return to avoid the loop
I've just asserted (using a static_cast, to avoid an implicit conversion
to pointer).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274902 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoSelectionDAG: Avoid implicit iterator conversions in FastISel, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 18:36:41 +0000 (18:36 +0000)]
SelectionDAG: Avoid implicit iterator conversions in FastISel, NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274899 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoTarget: Avoid getFirstTerminator() => pointer, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 18:26:20 +0000 (18:26 +0000)]
Target: Avoid getFirstTerminator() => pointer, NFC

Stop using an implicit conversion from the return of
MachineBasicBlock::getFirstTerminator to MachineInstr*.  In two cases,
directly dereference to a MachineInstr& since later code assumes it's
valid.  In a third case, change to an iterator since later code checks
against MachineBasicBlock::end.

Although the fix for the third case avoids undefined behaviour, I expect
this doesn't cause a functionality change in practice (since the basic
block already has a terminator).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274898 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoCodeGen: Avoid iterator conversions in TwoAddressInstructionPass, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 17:43:08 +0000 (17:43 +0000)]
CodeGen: Avoid iterator conversions in TwoAddressInstructionPass, NFC

Mostly through preferring MachineInstr&, avoid implicit conversions from
iterator to pointer.

Although this may bitrot (since there are other uses blocking me from
removing the implicit operator), this removes the last of the implicit
conversions from MachineInstrBundleIterator to MachineInstr* in the
LLVMCodeGen build target.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274893 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoCodeGen: Use MachineInstr& in StackSlotColoring, NFC
Duncan P. N. Exon Smith [Fri, 8 Jul 2016 17:28:40 +0000 (17:28 +0000)]
CodeGen: Use MachineInstr& in StackSlotColoring, NFC

Avoid implicit iterator to pointer conversions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274892 91177308-0d34-0410-b5e6-96231b3b80d8

8 years ago[InstCombine] check for one-use before turning simple logic op into a select
Sanjay Patel [Fri, 8 Jul 2016 17:26:47 +0000 (17:26 +0000)]
[InstCombine] check for one-use before turning simple logic op into a select

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274891 91177308-0d34-0410-b5e6-96231b3b80d8

8 years agoIR: Set a TargetPrefix for nvvm intrinsics
Justin Bogner [Fri, 8 Jul 2016 17:25:18 +0000 (17:25 +0000)]
IR: Set a TargetPrefix for nvvm intrinsics

Since these are named nvvm_* rather than nvptx_*, we also need to
update getArchTypePrefix. It's a bit unusual for getArchTypePrefix not
to match the backend name, but I think this fits the intent of the
function in this case.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@274890 91177308-0d34-0410-b5e6-96231b3b80d8