oota-llvm.git
9 years ago[Hexagon] Pattern match a CTZ loop into a call to countTrailingZeros.
Benjamin Kramer [Wed, 25 Mar 2015 15:36:57 +0000 (15:36 +0000)]
[Hexagon] Pattern match a CTZ loop into a call to countTrailingZeros.

No functional change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233192 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[ARM] Rewrite .save/.vsave emission with bit math
Benjamin Kramer [Wed, 25 Mar 2015 15:27:58 +0000 (15:27 +0000)]
[ARM] Rewrite .save/.vsave emission with bit math

Hopefully makes it a bit easier to understand what's going on.
No functional change intended.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233191 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoMake exit-code test use same mechanism as existing one.
Daniel Jasper [Wed, 25 Mar 2015 14:35:40 +0000 (14:35 +0000)]
Make exit-code test use same mechanism as existing one.

The other version doesn't properly work with our internal test runner,
which sets pipefail.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233188 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoFix fixup evaluation when deciding what to relocate with.
Rafael Espindola [Wed, 25 Mar 2015 13:16:53 +0000 (13:16 +0000)]
Fix fixup evaluation when deciding what to relocate with.

The previous logic was to first try without relocations at all
and failing that stop on the first defined symbol.

That was inefficient and incorrect in the case part of the
expression could be simplified and another part could not
(see included test).

We now stop the evaluation when we get to a variable whose value
can change (i.e. is weak).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233187 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[Orc] Remove another unnecessary typedef.
Lang Hames [Wed, 25 Mar 2015 12:32:36 +0000 (12:32 +0000)]
[Orc] Remove another unnecessary typedef.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233184 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[optnone] Skip pass Float2Int on optnone functions.
Andrea Di Biagio [Wed, 25 Mar 2015 12:22:37 +0000 (12:22 +0000)]
[optnone] Skip pass Float2Int on optnone functions.

Added test Float2Int/float2int-optnone.ll to verify that pass Float2Int
is not run on optnone functions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233183 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[Orc][lli] Add a very simple Orc-based lazy JIT to lli.
Lang Hames [Wed, 25 Mar 2015 12:11:48 +0000 (12:11 +0000)]
[Orc][lli] Add a very simple Orc-based lazy JIT to lli.

This ensures that we're building and testing the CompileOnDemand layer, at least
in a basic way.

Currently x86-64 only, and with limited to no library calls enabled (depending
on host platform). Patches welcome. ;)

To enable access to the lazy JIT, this patch replaces the '-use-orcmcjit' lli
option with a new option:
'-jit-kind={ mcjit | orc-mcjit | orc-lazy }'.

All regression tests are updated to use the new option, and one trivial test of
the new lazy JIT is added.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233182 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[X86] Simplify check lines in tests. No functional change.
Andrea Di Biagio [Wed, 25 Mar 2015 11:44:19 +0000 (11:44 +0000)]
[X86] Simplify check lines in tests. No functional change.

Also, removed unused check lines from test atomic6432.ll.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233181 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoReapply r233062: "float2int": Add a new pass to demote from float to int where possible.
James Molloy [Wed, 25 Mar 2015 10:03:42 +0000 (10:03 +0000)]
Reapply r233062: "float2int": Add a new pass to demote from float to int where possible.

Now with a fix for PR23008 and extra regression test.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233175 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agotest: Fix the dependencies for the check-llvm-* targets
Justin Bogner [Wed, 25 Mar 2015 08:07:47 +0000 (08:07 +0000)]
test: Fix the dependencies for the check-llvm-* targets

In r233009 we gained specific check-llvm-* build targets for invoking
specific parts of the test suite, but they were copying the
dependencies for check-all, rather than just listing the dependencies
for check-llvm.

This moves the creation of these targets next to the check-llvm
target, and uses that target's configuration rather than the check-all
config.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233174 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoFix build.
Rafael Espindola [Wed, 25 Mar 2015 04:36:20 +0000 (04:36 +0000)]
Fix build.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233171 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[X86] Remove GetCpuIDAndInfo, GetCpuIDAndInfoEx and DetectFamilyModel functions from...
Craig Topper [Wed, 25 Mar 2015 04:16:50 +0000 (04:16 +0000)]
[X86] Remove GetCpuIDAndInfo, GetCpuIDAndInfoEx and DetectFamilyModel functions from X86 MC layer. They haven't been used since CPU autodetection was removed from X86Subtarget.cpp.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233170 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[Orc] Remove extraneous 'typename'. This should cheer the bots up.
Lang Hames [Wed, 25 Mar 2015 04:00:34 +0000 (04:00 +0000)]
[Orc] Remove extraneous 'typename'. This should cheer the bots up.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233168 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[Orc] Refactor JITCompileCallbackManagerBase and CompileOnDemandLayer to support
Lang Hames [Wed, 25 Mar 2015 02:45:50 +0000 (02:45 +0000)]
[Orc] Refactor JITCompileCallbackManagerBase and CompileOnDemandLayer to support
target-independent callback management.

This is a prerequisite for adding orc-based lazy-jitting to lli.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233166 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoLinker: Temporarily disable dwarfdump checks from r233164
Duncan P. N. Exon Smith [Wed, 25 Mar 2015 02:43:04 +0000 (02:43 +0000)]
Linker: Temporarily disable dwarfdump checks from r233164

At least one Linux bot [1] doesn't like my dwarfdump checks, so I've
disable those until I can investigate what's going on there.  I'll
continue to track this in PR22792.

[1]: http://bb.pgr.jp/builders/cmake-llvm-x86_64-linux/builds/22863

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233165 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoLinker: Drop function pointers for overridden subprograms
Duncan P. N. Exon Smith [Wed, 25 Mar 2015 02:26:32 +0000 (02:26 +0000)]
Linker: Drop function pointers for overridden subprograms

Instead of dropping subprograms that have been overridden, just set
their function pointers to `nullptr`.  This is a minor adjustment to the
stop-gap fix for PR21910 committed in r224487, and fixes the crasher
from PR22792.

The problem that r224487 put a band-aid on: how do we find the canonical
subprogram for a `Function`?  Since the backend currently relies on
`DebugInfoFinder` (which does a naive in-order traversal of compile
units and picks the first subprogram) for this, r224487 tried dropping
non-canonical subprograms.

Dropping subprograms fails because the backend *also* builds up a map
from subprogram to compile unit (`DwarfDebug::SPMap`) based on the
subprogram lists.  A missing subprogram causes segfaults later when an
inlined reference (such as in this testcase) is created.

Instead, just drop the `Function` pointer to `nullptr`, which nicely
mirrors what happens when an already-inlined `Function` is optimized
out.  We can't really be sure that it's the same definition anyway, as
the testcase demonstrates.

This still isn't completely satisfactory.  Two flaws at least that I can
think of:

  - I still haven't found a straightforward way to make this symmetric
    in the IR.  (Interestingly, the DWARF output is already symmetric,
    and I've tested for that to be sure we don't regress.)
  - Using `DebugInfoFinder` to find the canonical subprogram for a
    function is kind of crazy.  We should just attach metadata to the
    function, like this:

        define weak i32 @foo(i32, i32) !dbg !MDSubprogram(...) {

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233164 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[ADT] Teach MapVector to support a swap member. Will be used in
Chandler Carruth [Wed, 25 Mar 2015 00:50:21 +0000 (00:50 +0000)]
[ADT] Teach MapVector to support a swap member. Will be used in
a subsequent commit in Clang.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233159 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoFix warning on non-assert build.
Rafael Espindola [Wed, 25 Mar 2015 00:45:41 +0000 (00:45 +0000)]
Fix warning on non-assert build.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233158 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoProduce an error instead of asserting on invalid .sleb128/.uleb128.
Rafael Espindola [Wed, 25 Mar 2015 00:25:37 +0000 (00:25 +0000)]
Produce an error instead of asserting on invalid .sleb128/.uleb128.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233155 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago'optnone' should not disable DAG combiner.
Paul Robinson [Wed, 25 Mar 2015 00:10:24 +0000 (00:10 +0000)]
'optnone' should not disable DAG combiner.

Reverts the code change from r221168 and the relevant test.
It was a mistake to disable the combiner, and based on the ultimate
definition of 'optnone' we shouldn't have considered the test case
as failing in the first place.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233153 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago!invariant.load semantics with potentially clobbering calls
Philip Reames [Tue, 24 Mar 2015 23:54:54 +0000 (23:54 +0000)]
!invariant.load semantics with potentially clobbering calls

A load from an invariant location is assumed to not alias any otherwise potentially aliasing stores. Our implementation only applied this rule to store instructions themselves whereas they it should apply for any memory accessing instruction. This results in both FRE and PRE becoming more effective at eliminating invariant loads.

Note that as a follow on change I will likely move this into AliasAnalysis itself. That's where the TBAA constant flag is handled and the semantics are essentially the same. I'd like to separate the semantic change from the refactoring and thus have extended the hack that's already in MemoryDependenceAnalysis for this change.

Differential Revision: http://reviews.llvm.org/D8591

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233140 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoDon't be over eager in evaluating a subtraction with a weak symbol.
Rafael Espindola [Tue, 24 Mar 2015 23:48:44 +0000 (23:48 +0000)]
Don't be over eager in evaluating a subtraction with a weak symbol.

In a subtraction of the form A - B, if B is weak, there is no way to represent
that on ELF since all relocations add the value of a symbol.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233139 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoX86: Fix frameescape when not using an FP
Reid Kleckner [Tue, 24 Mar 2015 23:46:01 +0000 (23:46 +0000)]
X86: Fix frameescape when not using an FP

We can't use TargetFrameLowering::getFrameIndexOffset directly, because
Win64 really wants the offset from the stack pointer at the end of the
prologue. Instead, use X86FrameLowering::getFrameIndexOffsetFromSP(),
which is a pretty close approximiation of that. It fails to handle cases
with interestingly large stack alignments, which is pretty uncommon on
Win64 and is TODO.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233137 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoUpdate a test I missed in r233132
Justin Bogner [Tue, 24 Mar 2015 23:44:03 +0000 (23:44 +0000)]
Update a test I missed in r233132

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233134 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoDisabling warnings for MSVC build to enable /W4 use.
Andrew Kaylor [Tue, 24 Mar 2015 23:37:10 +0000 (23:37 +0000)]
Disabling warnings for MSVC build to enable /W4 use.

Differential Revision: http://reviews.llvm.org/D8572

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233133 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agollvm-cov: Require a subcommand when invoked as llvm-cov
Justin Bogner [Tue, 24 Mar 2015 23:34:36 +0000 (23:34 +0000)]
llvm-cov: Require a subcommand when invoked as llvm-cov

A while ago llvm-cov gained support for clang's instrumentation based
profiling in addition to its gcov support, and subcommands were added
to choose which behaviour to use. When no subcommand was specified, we
fell back to gcov compatibility with a warning that a subcommand would
be required in the future. Now, we require the subcommand.

Note that if the basename of llvm-cov is gcov (via symlink or
hardlink, for example), we still use the gcov compatible behaviour
with no subcommand required.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233132 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoOpaque Pointer Types: GEP API migrations to specify the gep type explicitly
David Blaikie [Tue, 24 Mar 2015 23:34:31 +0000 (23:34 +0000)]
Opaque Pointer Types: GEP API migrations to specify the gep type explicitly

The changes to InstCombine (& SCEV) do seem a bit silly - it doesn't make
anything obviously better to have the caller access the pointers element
type (the thing I'm trying to remove) than the GEP itself, but it's a
helpful migration step. This will allow me to more obviously lock down
GEP (& Load, etc) API usage, then fix all the code that accesses pointer
element types except the places that need to be removed (most of the
InstCombines) anyway - at which point I'll need to just remove all that
code because it won't be meaningful anymore (there will be no pointer
types, so no bitcasts to combine)

SCEV looks like it'll need some restructuring - we'll have to do a bit
more work for GEP canonicalization, since it'll depend on how it's used
if we can even manage to canonicalize it to a non-ugly GEP. I guess we
can do some fun stuff like voting (do 2 out of 3 load from the GEP with
a certain type that gives a pretty GEP? Does every typed use of the GEP
use either a specific type or a generic type (i8*, etc)?)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233131 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[dsymutil] Temporarily disable some tests on windows.
Frederic Riss [Tue, 24 Mar 2015 23:11:07 +0000 (23:11 +0000)]
[dsymutil] Temporarily disable some tests on windows.

It seems one windows bot fails since I added ilne table linking to
llvm-dsymutil (see r232333 commit thread).
Disable the affected tests until I can figure out what's happening.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233130 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agooptimize the AVX2 (integer) version of vperm2 into a shuffle
Sanjay Patel [Tue, 24 Mar 2015 22:39:29 +0000 (22:39 +0000)]
optimize the AVX2 (integer) version of vperm2 into a shuffle

...because this is what happens when an instruction
set puts its underwear on after its pants.

This is an extension of r232852, r233100, and 233110:
http://llvm.org/viewvc/llvm-project?view=revision&revision=232852
http://llvm.org/viewvc/llvm-project?view=revision&revision=233100
http://llvm.org/viewvc/llvm-project?view=revision&revision=233110

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233127 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoOpaque Pointer Types: GEP API migrations to specify the gep type explicitly
David Blaikie [Tue, 24 Mar 2015 22:38:16 +0000 (22:38 +0000)]
Opaque Pointer Types: GEP API migrations to specify the gep type explicitly

The changes to InstCombine do seem a bit silly - it doesn't make
anything obviously better to have the caller access the pointers element
type (the thing I'm trying to remove) than the GEP itself, but it's a
helpful migration step. This will allow me to more obviously lock down
GEP (& Load, etc) API usage, then fix all the code that accesses pointer
element types except the places that need to be removed (most of the
InstCombines) anyway - at which point I'll need to just remove all that
code because it won't be meaningful anymore (there will be no pointer
types, so no bitcasts to combine)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233126 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoMerge empty landing pads in SimplifyCFG
Philip Reames [Tue, 24 Mar 2015 22:28:45 +0000 (22:28 +0000)]
Merge empty landing pads in SimplifyCFG

This patch tries to merge duplicate landing pads when they branch to a common shared target.

Given IR that looks like this:
lpad1:
  %exn = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
         cleanup
  br label %shared_resume
lpad2:
  %exn2 = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
          cleanup
  br label %shared_resume
shared_resume:
  call void @fn()
  ret void
}

We can rewrite the users of both landing pad blocks to use one of them. This will generally allow the shared_resume block to be merged with the common landing pad as well.

Without this change, tail duplication would likely kick in - creating N (2 in this case) copies of the shared_resume basic block.

Differential Revision: http://reviews.llvm.org/D8297

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233125 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoAdd -m -m elf_x86_64 to gold invocations.
Rafael Espindola [Tue, 24 Mar 2015 22:20:19 +0000 (22:20 +0000)]
Add -m -m elf_x86_64 to gold invocations.

Otherwise the tests would fail if the default was not elf_x86_64.

This fixes PR22966.

Patch by H.J. Lu!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233124 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRevert "Remove an InstCombine that seems to have become redundant."
David Blaikie [Tue, 24 Mar 2015 21:50:35 +0000 (21:50 +0000)]
Revert "Remove an InstCombine that seems to have become redundant."

Assertion fires in compiler-rt. Guess it does fire..

This reverts commit r233116.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233121 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoReset the CFA offset at the start of every FDE.
Rafael Espindola [Tue, 24 Mar 2015 21:47:31 +0000 (21:47 +0000)]
Reset the CFA offset at the start of every FDE.

This fixes PR21515.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233120 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoAArch64: use a different means to determine whether to byte swap relocations.
Peter Collingbourne [Tue, 24 Mar 2015 21:47:03 +0000 (21:47 +0000)]
AArch64: use a different means to determine whether to byte swap relocations.

This code depended on a bug in the FindAssociatedSection function that would
cause it to return the wrong result for certain absolute expressions. Instead,
use EvaluateAsRelocatable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233119 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoMC: Add more stringent symbol checking to test.
Peter Collingbourne [Tue, 24 Mar 2015 21:47:00 +0000 (21:47 +0000)]
MC: Add more stringent symbol checking to test.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233118 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRemove an InstCombine that seems to have become redundant.
David Blaikie [Tue, 24 Mar 2015 21:31:31 +0000 (21:31 +0000)]
Remove an InstCombine that seems to have become redundant.

Assert that this doesn't fire - I'll remove all of this later, but just
leaving it in for a while in case this is firing & we just don't have
test coverage.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233116 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[X86, AVX] instcombine vperm2 intrinsics with zero inputs into shuffles
Sanjay Patel [Tue, 24 Mar 2015 20:36:42 +0000 (20:36 +0000)]
[X86, AVX] instcombine vperm2 intrinsics with zero inputs into shuffles

This is the IR optimizer follow-on patch for D8563: the x86 backend patch
that converts this kind of shuffle back into a vperm2.

This is also a continuation of the transform that started in D8486.
In that patch, Andrea suggested that we could convert vperm2 intrinsics that
use zero masks into a single shuffle.

This is an implementation of that suggestion.

Differential Revision: http://reviews.llvm.org/D8567

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233110 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[llvm-readobj] add support for macho universal binary.
Rafael Espindola [Tue, 24 Mar 2015 20:26:55 +0000 (20:26 +0000)]
[llvm-readobj] add support for macho universal binary.

Patch by Keyue Hu (Chilledheart)!

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233107 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRevert r233062 ""float2int": Add a new pass to demote from float to int where possible."
Hans Wennborg [Tue, 24 Mar 2015 20:07:08 +0000 (20:07 +0000)]
Revert r233062 ""float2int": Add a new pass to demote from float to int where possible."

This caused PR23008, compiles failing with: "Use still stuck around after Def is
destroyed: %.sroa.speculated"

Also reverting follow-up r233064.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233105 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[IRCE] Fix how IRCE checks for no-sign-overflow.
Sanjoy Das [Tue, 24 Mar 2015 19:29:22 +0000 (19:29 +0000)]
[IRCE] Fix how IRCE checks for no-sign-overflow.

IRCE requires the induction variables it handles to not sign-overflow.
The current scheme of checking if sext({X,+,S}) == {sext(X),+,sext(S)}
fails when SCEV simplifies sext(X) too.  After this change we //also//
check no-signed-wrap by looking at the flags set on the SCEVAddRecExpr.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233102 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[IRCE] Fix a regression introduced in r232444.
Sanjoy Das [Tue, 24 Mar 2015 19:29:18 +0000 (19:29 +0000)]
[IRCE] Fix a regression introduced in r232444.

IRCE should not try to eliminate range checks that check an induction
variable against a loop-varying length.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233101 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[X86, AVX] recognize shufflevector with zero input as a vperm2 (PR22984)
Sanjay Patel [Tue, 24 Mar 2015 19:19:07 +0000 (19:19 +0000)]
[X86, AVX] recognize shufflevector with zero input as a vperm2 (PR22984)

vperm2x128 instructions have the special ability (aka free hardware capability)
to shuffle zero values into a vector.

This patch recognizes that type of shuffle and generates the appropriate
control byte.

https://llvm.org/bugs/show_bug.cgi?id=22984

Differential Revision: http://reviews.llvm.org/D8563

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233100 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoDebugInfo: Reorder definitions of MDLocation and MDFile, NFC
Duncan P. N. Exon Smith [Tue, 24 Mar 2015 17:34:33 +0000 (17:34 +0000)]
DebugInfo: Reorder definitions of MDLocation and MDFile, NFC

Move definition of `MDLocation` after `MDLocalScope` so that the latter
is available for casts in the former.  Similarly, move the definition of
`MDFile` as early as possible so that other classes can cast to it in
their definitions.  (Follow-up commits will take advantage of this.)

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233096 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoVerifier: Start recursing into !dbg attachments
Duncan P. N. Exon Smith [Tue, 24 Mar 2015 17:32:19 +0000 (17:32 +0000)]
Verifier: Start recursing into !dbg attachments

The main verifier already recurses through the other entry points, so we
might as well descend here too.

This temporarily duplicates some work already done in
`verifyDebugInfo()`, but eventually I'll be removing the other side.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233095 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoVerifier: !llvm.dbg.cu must point at compile units
Duncan P. N. Exon Smith [Tue, 24 Mar 2015 17:18:03 +0000 (17:18 +0000)]
Verifier: !llvm.dbg.cu must point at compile units

Duplicate this check from `verifyDebugInfo()`.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233094 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoDebugInfo: Add MDLocalScope, a legal scope for locals
Duncan P. N. Exon Smith [Tue, 24 Mar 2015 16:44:29 +0000 (16:44 +0000)]
DebugInfo: Add MDLocalScope, a legal scope for locals

Add a subclass of `MDScope` to explicitly categorize the legal scopes
for locals -- in particular, scopes that are legal for `MDLocation`,
`MDLexicalBlockBase`, and `MDLocalVariable`.  This provides a convenient
`isa<>` target for the verifier, and eventually I'll be changing the
above classes' `getScope()` to specifically return it.  Currently, its
subclasses are `MDSubprogram`, `MDLexicalBlock`, and
`MDLexicalBlockFile`.

I've gone with `MDLocalScope` for now -- a little ambiguous since it's a
scope *for* locals, not a scope that's local -- but I'm open to more
descriptive names if someone can think of something better.  Regardless,
the code docs should make it clear enough.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233092 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor: Simplify boolean expressions in lib/Analysis
David Blaikie [Tue, 24 Mar 2015 16:33:19 +0000 (16:33 +0000)]
Refactor: Simplify boolean expressions in lib/Analysis

Simplify boolean expressions using `true` and `false` with `clang-tidy`

Patch by Richard Thomson.

Reviewed By: nlewycky

Differential Revision: http://reviews.llvm.org/D8528

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233091 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor: Simplify boolean expressions in AArch64 target
David Blaikie [Tue, 24 Mar 2015 16:24:01 +0000 (16:24 +0000)]
Refactor: Simplify boolean expressions in AArch64 target

Simplify boolean expressions using `true` and `false` with `clang-tidy`

Patch by Richard Thomson.

Reviewed By: rengolin

Differential Revision: http://reviews.llvm.org/D8525

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233089 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[mips] Support 16-bit offsets for 'm' inline assembly memory constraint.
Daniel Sanders [Tue, 24 Mar 2015 15:19:14 +0000 (15:19 +0000)]
[mips] Support 16-bit offsets for 'm' inline assembly memory constraint.

Reviewers: vkalintiris

Reviewed By: vkalintiris

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D8435

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233086 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Insert more NOPs after READLANE on VI, don't use NOPs on CI
Marek Olsak [Tue, 24 Mar 2015 13:40:38 +0000 (13:40 +0000)]
R600/SI: Insert more NOPs after READLANE on VI, don't use NOPs on CI

This is a candidate for stable.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233080 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Select V_BFE_U32 for and+shift with a non-literal offset
Marek Olsak [Tue, 24 Mar 2015 13:40:34 +0000 (13:40 +0000)]
R600/SI: Select V_BFE_U32 for and+shift with a non-literal offset

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233079 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Custom-select 32-bit S_BFE from bitwise opcodes
Marek Olsak [Tue, 24 Mar 2015 13:40:27 +0000 (13:40 +0000)]
R600/SI: Custom-select 32-bit S_BFE from bitwise opcodes

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233078 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Improve BFM support
Marek Olsak [Tue, 24 Mar 2015 13:40:21 +0000 (13:40 +0000)]
R600/SI: Improve BFM support

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233077 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Use V_FRACT_F64 for faster 64-bit floor on SI
Marek Olsak [Tue, 24 Mar 2015 13:40:15 +0000 (13:40 +0000)]
R600/SI: Use V_FRACT_F64 for faster 64-bit floor on SI

Other f64 opcodes not supported on SI can be lowered in a similar way.

v2: use complex VOP3 patterns

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233076 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Expand fract to floor, then only select V_FRACT on CI
Marek Olsak [Tue, 24 Mar 2015 13:40:08 +0000 (13:40 +0000)]
R600/SI: Expand fract to floor, then only select V_FRACT on CI

V_FRACT is buggy on SI.

R600-specific code is left intact.

v2: drop the multiclass, use complex VOP3 patterns

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233075 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoInternalize the StackMapLiveness pass.
Benjamin Kramer [Tue, 24 Mar 2015 13:20:54 +0000 (13:20 +0000)]
Internalize the StackMapLiveness pass.

No need to have its own header when it's not used anywhere. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233072 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRevert "Use std::bitset for SubtargetFeatures"
Michael Kuperstein [Tue, 24 Mar 2015 12:56:59 +0000 (12:56 +0000)]
Revert "Use std::bitset for SubtargetFeatures"

This reverts commit r233055.

It still causes buildbot failures (gcc running out of memory on several platforms, and a self-host failure on arm), although less than the previous time.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233068 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoSilencing some MSVC warnings "C4805: '^' : unsafe mix of type 'bool' and type 'unsign...
Aaron Ballman [Tue, 24 Mar 2015 12:47:51 +0000 (12:47 +0000)]
Silencing some MSVC warnings "C4805: '^' : unsafe mix of type 'bool' and type 'unsigned int' in operation"; NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233067 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[mips] Simplify boolean expressions in Mips target with `clang-tidy`
Simon Atanasyan [Tue, 24 Mar 2015 12:24:56 +0000 (12:24 +0000)]
[mips] Simplify boolean expressions in Mips target with `clang-tidy`

No functional changes.

Patch by Richard Thomson.

Differential Revision: http://reviews.llvm.org/D8522

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233065 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[float2int] Sort includes and add missing raw_ostream include.
Benjamin Kramer [Tue, 24 Mar 2015 11:28:47 +0000 (11:28 +0000)]
[float2int] Sort includes and add missing raw_ostream include.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233064 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[mips] Distinguish 'R', 'ZC', and 'm' inline assembly memory constraint.
Daniel Sanders [Tue, 24 Mar 2015 11:26:34 +0000 (11:26 +0000)]
[mips] Distinguish 'R', 'ZC', and 'm' inline assembly memory constraint.

Summary:
Previous behaviour of 'R' and 'm' has been preserved for now. They will be
improved in subsequent commits.

The offset permitted by ZC varies according to the subtarget since it is
intended to match the restrictions of the pref, ll, and sc instructions.

The restrictions on these instructions are:
* For microMIPS: 12-bit signed offset.
* For Mips32r6/Mips64r6: 9-bit signed offset.
* Otherwise: 16-bit signed offset.

Reviewers: vkalintiris

Reviewed By: vkalintiris

Subscribers: llvm-commits

Differential Revision: http://reviews.llvm.org/D8414

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233063 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago"float2int": Add a new pass to demote from float to int where possible.
James Molloy [Tue, 24 Mar 2015 11:15:23 +0000 (11:15 +0000)]
"float2int": Add a new pass to demote from float to int where possible.

It is possible to have code that converts from integer to float, performs operations then converts back, and the result is provably the same as if integers were used.

This can come from different sources, but the most obvious is a helper function that uses floats but the arguments given at an inlined callsites are integers.

This pass considers all integers requiring a bitwidth less than or equal to the bitwidth of the mantissa of a floating point type (23 for floats, 52 for doubles) as exactly representable in floating point.

To reduce the risk of harming efficient code, the pass only attempts to perform complete removal of inttofp/fptoint operations, not just move them around.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233062 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoUse std::bitset for SubtargetFeatures
Michael Kuperstein [Tue, 24 Mar 2015 09:17:25 +0000 (09:17 +0000)]
Use std::bitset for SubtargetFeatures

Previously, subtarget features were a bitfield with the underlying type being uint64_t.
Since several targets (X86 and ARM, in particular) have hit or were very close to hitting this bound, switching the features to use a bitset.
No functional change.

The first time this was committed (r229831), it caused several buildbot failures.
At least some of the ARM ones were due to gcc/binutils issues, and should now be fixed.

Differential Revision: http://reviews.llvm.org/D8542

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233055 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[Orc] Move delta-handling for trampoline sizes into the resolver block.
Lang Hames [Tue, 24 Mar 2015 04:27:02 +0000 (04:27 +0000)]
[Orc] Move delta-handling for trampoline sizes into the resolver block.

This is the first step towards adding a target-independent callback
handler API.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233049 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[Orc] Whitespace fix. NFC.
Lang Hames [Tue, 24 Mar 2015 04:07:28 +0000 (04:07 +0000)]
[Orc] Whitespace fix. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233048 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[Orc] Use std::string to capture name by value.
Lang Hames [Tue, 24 Mar 2015 04:07:01 +0000 (04:07 +0000)]
[Orc] Use std::string to capture name by value.

This just updates the code to reflect the comment, but this bug actually hit the
out-of-tree lazy demo. I'm working on a patch to add the lazy-demo's
functionality to lli so that we can test this in-tree soon.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233047 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[SelectionDAG] Fixed issue with uitofp vector constant folding being treated as sitofp
Simon Pilgrim [Mon, 23 Mar 2015 22:44:55 +0000 (22:44 +0000)]
[SelectionDAG] Fixed issue with uitofp vector constant folding being treated as sitofp

While the uitofp scalar constant folding treats an integer as an unsigned value (from lang ref):

%X = sitofp i8 -1 to double ; yields double:-1.0
%Y = uitofp i8 -1 to double ; yields double:255.0

The vector constant folding was always using sitofp:

%X = sitofp <2 x i8> <i8 -1, i8 -1> to <2 x double> ; yields <double -1.0, double -1.0>
%Y = uitofp <2 x i8> <i8 -1, i8 -1> to <2 x double> ; yields <double -1.0, double -1.0>

This patch fixes this so that the correct opcode is used for sitofp and uitofp.

%X = sitofp <2 x i8> <i8 -1, i8 -1> to <2 x double> ; yields <double -1.0, double -1.0>
%Y = uitofp <2 x i8> <i8 -1, i8 -1> to <2 x double> ; yields <double 255.0, double 255.0>

Differential Revision: http://reviews.llvm.org/D8560

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233033 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRemove dead prototype DebugInfoFinder::processExpression(), NFC
Duncan P. N. Exon Smith [Mon, 23 Mar 2015 22:10:27 +0000 (22:10 +0000)]
Remove dead prototype DebugInfoFinder::processExpression(), NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233031 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoDebugInfo: Overload get() in DIDescriptor subclasses
Duncan P. N. Exon Smith [Mon, 23 Mar 2015 21:54:07 +0000 (21:54 +0000)]
DebugInfo: Overload get() in DIDescriptor subclasses

Continue to simplify the `DIDescriptor` subclasses, so that they behave
more like raw pointers.  Remove `getRaw()`, replace it with an
overloaded `get()`, and overload the arrow and cast operators.  Two
testcases started to crash on the arrow operators with this change
because of `scope:` references that weren't real scopes.  I fixed them.
Soon I'll add verifier checks for them too.

This also adds explicit dereference operators.  Previously, the builtin
dereference against `operator MDNode *()` would have worked, but now the
builtins are ambiguous.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233030 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor how passes get a symbol at the end of a section.
Rafael Espindola [Mon, 23 Mar 2015 21:22:04 +0000 (21:22 +0000)]
Refactor how passes get a symbol at the end of a section.

There is now a canonical symbol at the end of a section that different
passes can request.

This also allows us to assert that we don't switch back to a section whose
end symbol has already been printed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233026 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoCleanup else-after-return and add an early-return to llvm-nm
David Blaikie [Mon, 23 Mar 2015 21:17:43 +0000 (21:17 +0000)]
Cleanup else-after-return and add an early-return to llvm-nm

The loop and error handling in checkMachOAndArchFlags didn't make sense
to me (a loop that only ever executes once? An error path that uses the
element the loop stopped at (which must always be a buffer overrun if
I'm reading that right?)... I'm confused) but I've made a guess at what
was intended.

Based on a patch by Richard Thomson to simplify boolean expressions.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233025 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[AArch64, ARM] Enable GlobalMerge with -O3 rather than -O1.
Ahmed Bougacha [Mon, 23 Mar 2015 21:17:36 +0000 (21:17 +0000)]
[AArch64, ARM] Enable GlobalMerge with -O3 rather than -O1.

The pass used to be enabled by default with CodeGenOpt::Less (-O1).
This is too aggressive, considering the pass indiscriminately merges
all globals together.

Currently, performance doesn't always improve, and, on code that uses
few globals (e.g., the odd file- or function- static), more often than
not is degraded by the optimization.  Lengthy discussion can be found
on llvmdev (AArch64-focused;  ARM has similar problems):
  http://lists.cs.uiuc.edu/pipermail/llvmdev/2015-February/082800.html
Also, it makes tooling and debuggers less useful when dealing with
globals and data sections.

GlobalMerge needs to better identify those cases that benefit, and this
will be done separately.  In the meantime, move the pass to run with
-O3 rather than -O1, on both ARM and AArch64.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233024 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor: Simplify boolean expressions in R600 target
David Blaikie [Mon, 23 Mar 2015 20:56:44 +0000 (20:56 +0000)]
Refactor: Simplify boolean expressions in R600 target

Simplify boolean expressions with `true` and `false` using `clang-tidy`

Patch by Richard Thomson.

Differential Revision: http://reviews.llvm.org/D8520

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233020 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoUpdate variable name and reuse existing variable. NFC.
Rafael Espindola [Mon, 23 Mar 2015 20:25:31 +0000 (20:25 +0000)]
Update variable name and reuse existing variable. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233014 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[AArch64] Add FileCheck that was missing from test in r232967.
Chad Rosier [Mon, 23 Mar 2015 20:25:15 +0000 (20:25 +0000)]
[AArch64] Add FileCheck that was missing from test in r232967.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233013 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRe-land: Generate targets for each lit suite.
Chris Bieneman [Mon, 23 Mar 2015 20:04:00 +0000 (20:04 +0000)]
Re-land: Generate targets for each lit suite.

Summary:
This change makes CMake scan for lit suites and generate a target for each lit test suite. The targets follow the format check-<project>-<suite path>.

For example:
check-llvm-unit - Runs the LLVM unit tests
check-llvm-codegen-arm - Runs the ARM codeine tests

Note: These targets are not generated during multi-configuration generators (i.e. Xcode and Visual Studio) because target clutter impacts UI usability.

* Also fixed a minor issue that Duncan pointed out to me I was passing the suite to lit twice

Reviewers: chandlerc

Subscribers: aemerson, llvm-commits

Differential Revision: http://reviews.llvm.org/D8380

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233009 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRaising minimum required CMake version to 2.8.12.2.
Chris Bieneman [Mon, 23 Mar 2015 20:03:57 +0000 (20:03 +0000)]
Raising minimum required CMake version to 2.8.12.2.

This commit is in reference to the llvm-dev thread: http://lists.cs.uiuc.edu/pipermail/llvmdev/2015-March/083672.html

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233008 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor: Simplify boolean expressions in llvm IR
David Blaikie [Mon, 23 Mar 2015 19:51:23 +0000 (19:51 +0000)]
Refactor: Simplify boolean expressions in llvm IR

Simplify boolean expressions using `true` and `false` with `clang-tidy`

Patch by Richard Thomson with a few other simplifications to fix
else-after-returns in the surrounding code.

Differential Revision: http://reviews.llvm.org/D8527

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233005 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor: Simplify boolean expressions in llvm Support
David Blaikie [Mon, 23 Mar 2015 19:45:40 +0000 (19:45 +0000)]
Refactor: Simplify boolean expressions in llvm Support

Simplify boolean expressions using `true` and `false` with `clang-tidy`

Patch by Richard Thomson - I dropped the parens and != 0 test, for
consistency with other patches/tests like this, but I'm open to the
notion that we should add the explicit non-zero test in all these sort
of cases (non-bool assigned to a bool).

Differential Revision: http://reviews.llvm.org/D8526

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233004 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor: Simplify boolean expressions in x86 target
David Blaikie [Mon, 23 Mar 2015 19:42:36 +0000 (19:42 +0000)]
Refactor: Simplify boolean expressions in x86 target

Simplify boolean expressions with `true` and `false` with `clang-tidy`

Patch by Richard Thomson.

Differential Revision: http://reviews.llvm.org/D8519

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@233002 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRe-sort includes with sort-includes.py and insert raw_ostream.h where it's used.
Benjamin Kramer [Mon, 23 Mar 2015 19:32:43 +0000 (19:32 +0000)]
Re-sort includes with sort-includes.py and insert raw_ostream.h where it's used.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232998 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[ctorutils] Update and sort includes. NFC.
Benjamin Kramer [Mon, 23 Mar 2015 19:06:17 +0000 (19:06 +0000)]
[ctorutils] Update and sort includes. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232995 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[winehprepare] Update and sort includes. NFC.
Benjamin Kramer [Mon, 23 Mar 2015 18:57:17 +0000 (18:57 +0000)]
[winehprepare] Update and sort includes. NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232994 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoAnother set of missing raw_ostream.h. Still no functional change.
Benjamin Kramer [Mon, 23 Mar 2015 18:45:56 +0000 (18:45 +0000)]
Another set of missing raw_ostream.h. Still no functional change.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232993 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Merge tables for commuting
Matt Arsenault [Mon, 23 Mar 2015 18:45:41 +0000 (18:45 +0000)]
R600/SI: Merge tables for commuting

Don't use a separate table for compares anymore,
and use the same VOP2_REV class.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232992 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Only use one range of isCommutable for compares
Matt Arsenault [Mon, 23 Mar 2015 18:45:38 +0000 (18:45 +0000)]
R600/SI: Only use one range of isCommutable for compares

Also don't count the class instructions as isCompare anymore.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232991 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Remove redundant unsetting of hasSideEffects
Matt Arsenault [Mon, 23 Mar 2015 18:45:36 +0000 (18:45 +0000)]
R600/SI: Remove redundant unsetting of hasSideEffects

These are already set in the base class for the instruction.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232990 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Move hasSideEffects setting into VOPCX classes
Matt Arsenault [Mon, 23 Mar 2015 18:45:35 +0000 (18:45 +0000)]
R600/SI: Move hasSideEffects setting into VOPCX classes

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232989 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Allow commuting compares
Matt Arsenault [Mon, 23 Mar 2015 18:45:30 +0000 (18:45 +0000)]
R600/SI: Allow commuting compares

This enables very common cases to switch to the
smaller encoding.

All of the standard LLVM canonicalizations of comparisons
are the opposite of what we want. Compares with constants
are moved to the RHS, but the first operand can be an inline
immediate, literal constant, or SGPR using the 32-bit VOPC
encoding.

There are additional bad canonicalizations that should
also be fixed, such as canonicalizing ge x, k to gt x, (k + 1)
if this makes k no longer an inline immediate value.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232988 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Use right class for cmpsx f64 instructions
Matt Arsenault [Mon, 23 Mar 2015 18:45:23 +0000 (18:45 +0000)]
R600/SI: Use right class for cmpsx f64 instructions

Use VOPCX_F64 to not need the let Defs = [EXEC]

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232987 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoR600/SI: Remove cond operand to VOPCX classes
Matt Arsenault [Mon, 23 Mar 2015 18:45:20 +0000 (18:45 +0000)]
R600/SI: Remove cond operand to VOPCX classes

It isn't used, and these will probably never be directly selected.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232986 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRefactor: simplify boolean expressions in llvm-objdump
David Blaikie [Mon, 23 Mar 2015 18:39:02 +0000 (18:39 +0000)]
Refactor: simplify boolean expressions in llvm-objdump

Simplify boolean expressions involving `true` and `false` with `clang-tidy`.

Actually upon inspection a bunch of these boolean variables could be
factored away entirely anyway - using find_if and then testing the
result before using it. This also helps reduce indentation in the code
anyway - and a bunch of other related simplification fell out nearby so
I just committed all of that.

Patch by Richard Thomson (legalize@xmission.com)

Differential Revision: http://reviews.llvm.org/D8517

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232984 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoAdd missing ELFObjectWriter::reset() override, like other MC classes.
Yaron Keren [Mon, 23 Mar 2015 18:35:01 +0000 (18:35 +0000)]
Add missing ELFObjectWriter::reset() override, like other MC classes.

See detailed discussion at

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20140915/235418.html

and r217907, r217948:

 http://llvm.org/viewvc/llvm-project?view=revision&revision=217907
 http://llvm.org/viewvc/llvm-project?view=revision&revision=217948

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232982 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoMore missing includes only visible to MSVC.
Benjamin Kramer [Mon, 23 Mar 2015 18:23:08 +0000 (18:23 +0000)]
More missing includes only visible to MSVC.

NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232981 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoAdd missing include that MSVC complains about.
Benjamin Kramer [Mon, 23 Mar 2015 18:19:41 +0000 (18:19 +0000)]
Add missing include that MSVC complains about.

Also reorder includes a bit, NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232980 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoPurge unused includes throughout libSupport.
Benjamin Kramer [Mon, 23 Mar 2015 18:07:13 +0000 (18:07 +0000)]
Purge unused includes throughout libSupport.

NFC.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232976 91177308-0d34-0410-b5e6-96231b3b80d8

9 years ago[AArch64] Enable rematerialization of float 0 values.
Chad Rosier [Mon, 23 Mar 2015 17:19:34 +0000 (17:19 +0000)]
[AArch64] Enable rematerialization of float 0 values.

Patch by Geoff Berry<gberry@codeaurora.org>.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232967 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoRevert "[ARM] Add more pattern matching for f16 <-> f64 conversions"
Bradley Smith [Mon, 23 Mar 2015 16:52:52 +0000 (16:52 +0000)]
Revert "[ARM] Add more pattern matching for f16 <-> f64 conversions"

This change is incorrect since it converts double rounding into single rounding,
which can produce different results. Instead this optimization will be done by
modifying Clang's codegen to not produce double rounding in the first place.

This reverts commit r232954.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232962 91177308-0d34-0410-b5e6-96231b3b80d8

9 years agoSimplify boolean expressions with true and false using clang-tidy
Eli Bendersky [Mon, 23 Mar 2015 16:26:23 +0000 (16:26 +0000)]
Simplify boolean expressions with true and false using clang-tidy

Patch by Richard (legalize@xmission.com)

Differential Revision: http://reviews.llvm.org/D8521

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232961 91177308-0d34-0410-b5e6-96231b3b80d8