OSDN Git Service
Matthias Braun [Wed, 7 Nov 2018 02:04:07 +0000 (02:04 +0000)]
RegAllocFast: Rename statistic from NumCopies to NumCoalesced
The metric does not return the number of remaining (or inserted) copies
but the number of copies that were coalesced. Pick a more descriptive
name.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346287
91177308-0d34-0410-b5e6-
96231b3b80d8
Heejin Ahn [Wed, 7 Nov 2018 01:58:50 +0000 (01:58 +0000)]
[WebAssembly] Update test cases after FixFunctionBitcasts
Summary:
This updates generated binaries and corresponding test cases up to date
after applying FixFunctionBitcasts pass.
Reviewers: sbc100
Subscribers: dschuff, jgravelle-google, sunfish, llvm-commits
Differential Revision: https://reviews.llvm.org/D54070
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346286
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Wed, 7 Nov 2018 00:00:42 +0000 (00:00 +0000)]
[IR] add optional parameter for copying IR flags to compare instructions
As shown, this is used to eliminate redundant code in InstCombine,
and there are more cases where we should be using this pattern, but
we're currently unintentionally dropping flags.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346282
91177308-0d34-0410-b5e6-
96231b3b80d8
Reid Kleckner [Tue, 6 Nov 2018 23:39:59 +0000 (23:39 +0000)]
[Windows] Simplify WindowsSupport.h
Sink Windows version detection code from WindowsSupport.h to Path.inc.
These functions don't need to be inlined. I randomly picked Process.inc
for the Windows version helpers, since that's the most related file.
Sink MakeErrMsg to Program.inc since it's the main client.
Move those functions into the llvm namespace, and delete the scoped
handle copy and assignment operators.
Reviewers: zturner, aganea
Differential Revision: https://reviews.llvm.org/D54182
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346280
91177308-0d34-0410-b5e6-
96231b3b80d8
Joel E. Denny [Tue, 6 Nov 2018 22:42:10 +0000 (22:42 +0000)]
[FileCheck] Try to fix windows bots broken by r346272
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346277
91177308-0d34-0410-b5e6-
96231b3b80d8
Jessica Paquette [Tue, 6 Nov 2018 22:23:13 +0000 (22:23 +0000)]
[MachineOutliner][NFC] Remove OccurrenceCount from SuffixTreeNode
After changing the way we find candidates in r346269, this is no longer used.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346275
91177308-0d34-0410-b5e6-
96231b3b80d8
Jessica Paquette [Tue, 6 Nov 2018 22:21:11 +0000 (22:21 +0000)]
[MachineOutliner][NFC] Remove IsInTree from SuffixTreeNode
After changing the way we find repeated substrings in r346269, this
field is no longer used by anything, so it can be removed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346274
91177308-0d34-0410-b5e6-
96231b3b80d8
Evandro Menezes [Tue, 6 Nov 2018 22:17:14 +0000 (22:17 +0000)]
[PATCH] [AArch64] Refactor helper functions (NFC)
Refactor helper functions in AArch64InstrInfo to be static methods.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346273
91177308-0d34-0410-b5e6-
96231b3b80d8
Joel E. Denny [Tue, 6 Nov 2018 22:07:03 +0000 (22:07 +0000)]
[FileCheck] Parse command-line options from FILECHECK_OPTS
This feature makes it easy to tune FileCheck diagnostic output when
running the test suite via ninja, a bot, or an IDE. For example:
```
$ FILECHECK_OPTS='-color -v -dump-input-on-failure' \
LIT_FILTER='OpenMP/for_codegen.cpp' ninja check-clang \
| less -R
```
Reviewed By: probinson
Differential Revision: https://reviews.llvm.org/D53517
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346272
91177308-0d34-0410-b5e6-
96231b3b80d8
Jessica Paquette [Tue, 6 Nov 2018 21:46:41 +0000 (21:46 +0000)]
[MachineOutliner][NFC] Add findRepeatedSubstrings to SuffixTree, kill LeafVector
Instead of iterating over the leaves to find repeated substrings, and walking
collecting leaf children when we don't necessarily need them, let's just
calculate what we need and iterate over that.
By doing this, we don't have to save every leaf. It's easier to read the code
too and understand what's going on.
The goal here, at the end of the day, is to set up to allow us to do something
like
for (RepeatedSubstring &RS : ST) {
... do stuff with RS ...
}
Which would let us perform the cost model stuff and the repeated substring
query at the same time.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346269
91177308-0d34-0410-b5e6-
96231b3b80d8
Reid Kleckner [Tue, 6 Nov 2018 21:40:32 +0000 (21:40 +0000)]
Silence deprecation warning for GetVersionEx with clang-cl
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346268
91177308-0d34-0410-b5e6-
96231b3b80d8
Yaxun Liu [Tue, 6 Nov 2018 21:28:17 +0000 (21:28 +0000)]
AMDGPU: Add an option -disable-promote-alloca-to-lds
Add this option for debugging and providing workaround.
By default it is off so no behavior change in backend.
Differential Revision: https://reviews.llvm.org/D54158
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346267
91177308-0d34-0410-b5e6-
96231b3b80d8
Konstantin Zhuravlyov [Tue, 6 Nov 2018 20:23:53 +0000 (20:23 +0000)]
AMDGPU/Docs: Fix the processor table
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346263
91177308-0d34-0410-b5e6-
96231b3b80d8
Teresa Johnson [Tue, 6 Nov 2018 19:41:35 +0000 (19:41 +0000)]
[ThinLTO] Split NotEligibleToImport into legality and inlinability flags
Summary:
The NotEligibleToImport flag on the GlobalValueSummary was set if it
isn't legal to import (e.g. because it references unpromotable locals)
and when it can't be inlined (in which case importing is pointless).
I split out the inlinable piece into a separate flag on the
FunctionSummary (doesn't make sense for aliases or global variables),
because in the future we may want to import for reasons other than
inlining.
Reviewers: davidxl
Subscribers: mehdi_amini, inglorion, eraman, steven_wu, dexonsmith, arphaman, llvm-commits
Differential Revision: https://reviews.llvm.org/D53345
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346261
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 6 Nov 2018 19:24:21 +0000 (19:24 +0000)]
[X86] Add custom promotion of v2i8/v2i16 fp_to_sint to avoid over promotion to v2i64 which would force scalarization.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346259
91177308-0d34-0410-b5e6-
96231b3b80d8
Vedant Kumar [Tue, 6 Nov 2018 19:06:08 +0000 (19:06 +0000)]
[CodeExtractor] Do not extract calls to eh_typeid_for (PR39545)
The lowering for a call to eh_typeid_for changes when it's moved from
one function to another.
There are several proposals for fixing this issue in llvm.org/PR39545.
Until some solution is in place, do not allow CodeExtractor to extract
calls to eh_typeid_for, as that results in serious miscompilations.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346256
91177308-0d34-0410-b5e6-
96231b3b80d8
Vedant Kumar [Tue, 6 Nov 2018 19:05:53 +0000 (19:05 +0000)]
[CodeExtractor] Erase use-without-def debug intrinsics in parent func
When CodeExtractor moves instructions to a new function, debug
intrinsics referring to those instructions within the parent function
become invalid.
This results in the same verifier failure which motivated r344545, about
function-local metadata being used in the wrong function.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346255
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthias Braun [Tue, 6 Nov 2018 19:00:11 +0000 (19:00 +0000)]
LivePhysRegs/IfConversion: Change some types from unsigned to MCPhysReg; NFC
Change the type in a couple of lists and sets that only store physical
registers from unsigned to MCPhysRegs. The later is only 16bits and
saves us a bit of memory.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346254
91177308-0d34-0410-b5e6-
96231b3b80d8
Volkan Keles [Tue, 6 Nov 2018 18:59:18 +0000 (18:59 +0000)]
[AArch64][GlobalISel] Simplify and autogenerate the legalizer tests
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346253
91177308-0d34-0410-b5e6-
96231b3b80d8
Justin Bogner [Tue, 6 Nov 2018 18:52:30 +0000 (18:52 +0000)]
[utils] Update SmallVector lldb formatter for r337514
SmallVector was changed to store a begin and a size rather than a
begin and an end a while back. Update the formatter to look at the
correct members.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346252
91177308-0d34-0410-b5e6-
96231b3b80d8
Volkan Keles [Tue, 6 Nov 2018 18:31:25 +0000 (18:31 +0000)]
Reland r346166: [GlobalISel] Refactor the artifact combiner a bit by using MIPatternMatch
It was causing a crash because we were trying to get the definition
of a target register. Fixed the issue by adding a check and added
a test case for that.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346251
91177308-0d34-0410-b5e6-
96231b3b80d8
Eli Friedman [Tue, 6 Nov 2018 18:23:32 +0000 (18:23 +0000)]
Disable calls to *_finite and other glibc-only functions on Musl.
Non-GNU environments don't have __finite_*, so treat them as
unavailable.
Differential Revision: https://reviews.llvm.org/D51282
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346250
91177308-0d34-0410-b5e6-
96231b3b80d8
Derek Schuff [Tue, 6 Nov 2018 17:27:25 +0000 (17:27 +0000)]
[WebAssembly] Add shared memory support to limits field
Support the IS_SHARED bit in the memory limits flag word.
The compiler does not create object files with memory definitions,
but the field is used by the linker.
Differential Revision: https://reviews.llvm.org/D54131
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346246
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 17:20:20 +0000 (17:20 +0000)]
[InstCombine] allow vector types for fcmp+fpext fold
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346245
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 17:06:58 +0000 (17:06 +0000)]
[InstCombine] add vector test for fcmp+fpext; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346243
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 16:45:27 +0000 (16:45 +0000)]
[InstCombine] propagate fast-math-flags when folding fcmp+fpext, part 2
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346242
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 16:37:35 +0000 (16:37 +0000)]
[InstCombine] rearrange code for fcmp+fpext; NFCI
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346241
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 16:23:03 +0000 (16:23 +0000)]
[InstCombine] propagate fast-math-flags when folding fcmp+fpext
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346240
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 16:07:39 +0000 (16:07 +0000)]
[InstCombine] adjust tests to show dropping FMF; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346239
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 15:58:57 +0000 (15:58 +0000)]
[InstCombine] propagate fast-math-flags when folding fcmp+fneg, part 2
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346238
91177308-0d34-0410-b5e6-
96231b3b80d8
Elizabeth Andrews [Tue, 6 Nov 2018 15:57:59 +0000 (15:57 +0000)]
[benchmark] Disable exceptions in Microsoft STL
This patch disables exceptions in Microsoft STL when exception
handling is not enabled in Benchmark project. It fixes Windows
builds that were failing due to C4530 warnings thrown by MS STL.
Differential Revision: https://reviews.llvm.org/D52998
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346237
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 15:57:52 +0000 (15:57 +0000)]
[InstCombine] adjust tests to show dropping FMF; NFC
Also, remove some stale FIXME comments ( rL346234 ).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346236
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 15:53:58 +0000 (15:53 +0000)]
[InstCombine] reduce code; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346235
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 15:49:45 +0000 (15:49 +0000)]
[InstCombine] propagate fast-math-flags when folding fcmp+fneg
This is another part of solving PR39475:
https://bugs.llvm.org/show_bug.cgi?id=39475
This might be enough to fix that particular issue, but as noted
with the FIXME, we're still dropping FMF on other folds around here.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346234
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 6 Nov 2018 15:21:44 +0000 (15:21 +0000)]
[InstCombine] add tests for FMF propagation failure; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346232
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Atanasyan [Tue, 6 Nov 2018 14:37:24 +0000 (14:37 +0000)]
[mips] Support sigrie instruction
The `sigrie` instruction signals a Reserved Instruction Exception.
This patch adds support for assembling / disassembling the instruction.
Differential Revision: http://reviews.llvm.org/D53861
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346230
91177308-0d34-0410-b5e6-
96231b3b80d8
Clement Courbet [Tue, 6 Nov 2018 14:11:58 +0000 (14:11 +0000)]
[llvm-exegesis] Ignore X86 pseudo instructions.
Summary: They do not lower to actual MCInsts and have no scheduling info.
Reviewers: gchatelet
Subscribers: llvm-commits, tschuett
Differential Revision: https://reviews.llvm.org/D54147
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346227
91177308-0d34-0410-b5e6-
96231b3b80d8
Clement Courbet [Tue, 6 Nov 2018 13:48:56 +0000 (13:48 +0000)]
[X86][NFC] Fix comment.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346226
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Tue, 6 Nov 2018 11:28:22 +0000 (11:28 +0000)]
[InstCombine] Ensure nested shifts are in range (OSS-Fuzz #9880)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346225
91177308-0d34-0410-b5e6-
96231b3b80d8
Martin Storsjo [Tue, 6 Nov 2018 09:08:20 +0000 (09:08 +0000)]
[Support] Fix `warning: unknown pragma ignored` for mingw target
Differential Revision: https://reviews.llvm.org/D54133
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346218
91177308-0d34-0410-b5e6-
96231b3b80d8
Max Kazantsev [Tue, 6 Nov 2018 09:07:03 +0000 (09:07 +0000)]
[NFC] Turn collectTransitivePredecessors into a static function
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346217
91177308-0d34-0410-b5e6-
96231b3b80d8
Dean Michael Berris [Tue, 6 Nov 2018 08:51:37 +0000 (08:51 +0000)]
[XRay] Update XRayRecord to support Custom/Typed Events
Summary:
This change cuts across LLVM and compiler-rt to add support for
rendering custom events in the XRayRecord type, to allow for including
user-provided annotations in the output YAML (as raw bytes).
This work enables us to add custom event and typed event records into
the `llvm::xray::Trace` type for user-provided events. This can then be
programmatically handled through the C++ API and can be included in some
of the tooling as well. For now we support printing the raw data we
encounter in the custom events in the converted output.
Future work will allow us to start interpreting these custom and typed
events through a yet-to-be-defined API for extending the trace analysis
library.
Reviewers: mboerger
Subscribers: hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D54139
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346214
91177308-0d34-0410-b5e6-
96231b3b80d8
Max Kazantsev [Tue, 6 Nov 2018 04:17:40 +0000 (04:17 +0000)]
[LICM] Remove too conservative IsMustExecute variable
LICM relies on variable `MustExecute` which is conservatively set to `false`
in all non-headers. It is used when we decide whether or not we want to hoist
an instruction or a guard.
For the guards, it might be too conservative to use this variable, we can
instead use a more precise logic from LoopSafetyInfo. Currently it is only NFC
because `IsMemoryNotModified` is also conservatively set to `false` for all
non-headers, and we cannot hoist guards from non-header blocks. However once we
give up using `IsMemoryNotModified` and use a smarter check instead, this will
allow us to hoist guards from all mustexecute non-header blocks.
Differential Revision: https://reviews.llvm.org/D50888
Reveiwed By: fedor.sergeev
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346204
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthias Braun [Tue, 6 Nov 2018 03:15:22 +0000 (03:15 +0000)]
AArch64: Cleanup CCMP code; NFC
Cleanup CCMP pattern matching code in preparation for review/bugfix:
- Rename `isConjunctionDisjunctionTree()` to `canEmitConjunction()`
(it won't accept arbitrary disjunctions and is really about whether we
can transform the subtree into a conjunction that we can emit).
- Rename `emitConjunctionDisjunctionTree()` to `emitConjunction()`
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346203
91177308-0d34-0410-b5e6-
96231b3b80d8
Zi Xuan Wu [Tue, 6 Nov 2018 03:07:03 +0000 (03:07 +0000)]
It's a test commit, which is my first commit and also add my name to CREDITS.TXT
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346202
91177308-0d34-0410-b5e6-
96231b3b80d8
Max Kazantsev [Tue, 6 Nov 2018 02:44:49 +0000 (02:44 +0000)]
[LICM] Use ICFLoopSafetyInfo in LICM
This patch makes LICM use `ICFLoopSafetyInfo` that is a smarter version
of LoopSafetyInfo that leverages power of Implicit Control Flow Tracking
to keep track of throwing instructions and give less pessimistic answers
to queries related to throws.
The ICFLoopSafetyInfo itself has been introduced in rL344601. This patch
enables it in LICM only.
Differential Revision: https://reviews.llvm.org/D50377
Reviewed By: apilipenko
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346201
91177308-0d34-0410-b5e6-
96231b3b80d8
Max Kazantsev [Tue, 6 Nov 2018 02:12:44 +0000 (02:12 +0000)]
[NFC] Add motivating test case for revert in rL346198
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346199
91177308-0d34-0410-b5e6-
96231b3b80d8
Max Kazantsev [Tue, 6 Nov 2018 02:02:05 +0000 (02:02 +0000)]
Revert "[IndVars] Smart hard uses detection"
This reverts commit
2f425e9c7946b9d74e64ebbfa33c1caa36914402.
It seems that the check that we still should do the transform if we
know the result is constant is missing in this code. So the logic that
has been deleted by this change is still sometimes accidentally useful.
I revert the change to see what can be done about it. The motivating
case is the following:
@Y = global [400 x i16] zeroinitializer, align 1
define i16 @foo() {
entry:
br label %for.body
for.body: ; preds = %entry, %for.body
%i = phi i16 [ 0, %entry ], [ %inc, %for.body ]
%arrayidx = getelementptr inbounds [400 x i16], [400 x i16]* @Y, i16 0, i16 %i
store i16 0, i16* %arrayidx, align 1
%inc = add nuw nsw i16 %i, 1
%cmp = icmp ult i16 %inc, 400
br i1 %cmp, label %for.body, label %for.end
for.end: ; preds = %for.body
%inc.lcssa = phi i16 [ %inc, %for.body ]
ret i16 %inc.lcssa
}
We should be able to figure out that the result is constant, but the patch
breaks it.
Differential Revision: https://reviews.llvm.org/D51584
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346198
91177308-0d34-0410-b5e6-
96231b3b80d8
Robert Widmann [Tue, 6 Nov 2018 01:54:12 +0000 (01:54 +0000)]
[LLVM-C] Fix Windows Build of Core
strndup doesn't exist outside of GNU-land and modern macOSes. Use
strdup instead as c_str() is guaranteed to be NUL-terminated.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346197
91177308-0d34-0410-b5e6-
96231b3b80d8
Robert Widmann [Tue, 6 Nov 2018 01:38:14 +0000 (01:38 +0000)]
[LLVM-C] Improve Intrinsics Bindings
Summary:
Improve the intrinsic bindings with operations for
- Retrieving and automatically inserting the declaration of an intrinsic by ID
- Retrieving the name of a non-overloaded intrinsic by ID
- Retrieving the name of an overloaded intrinsic by ID and overloaded parameter types
Improve the echo test to copy non-overloaded intrinsics by ID.
Reviewers: whitequark, deadalnix
Reviewed By: whitequark
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D53626
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346195
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 6 Nov 2018 00:31:27 +0000 (00:31 +0000)]
[X86] Autogenerate complete checks. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346188
91177308-0d34-0410-b5e6-
96231b3b80d8
Sam Clegg [Tue, 6 Nov 2018 00:31:02 +0000 (00:31 +0000)]
Revert "[WebAssembly] Fixup `main` signature by default"
This reverts rL345880. It caused some test failures on the
webassembly waterfall. e.g. binaryen2.test_mainenv fails due
the fact that `envp` ends up being undef rather than 0.
Differential Revision: https://reviews.llvm.org/D54117
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346187
91177308-0d34-0410-b5e6-
96231b3b80d8
Justin Bogner [Tue, 6 Nov 2018 00:16:32 +0000 (00:16 +0000)]
Specify REQUIRES: default_triple in two debuginfo tests
These were failing when specifying LLVM_DEFAULT_TARGET_TRIPLE=''
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346185
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthias Braun [Mon, 5 Nov 2018 23:49:15 +0000 (23:49 +0000)]
TargetMachine: Move lib/CodeGen specific callbacks to LLVMTargetMachine; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346184
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthias Braun [Mon, 5 Nov 2018 23:49:14 +0000 (23:49 +0000)]
MachineFunction: Store more specific reference to LLVMTargetMachine; NFC
MachineFunction can only be used in code using lib/CodeGen, hence we
can keep a more specific reference to LLVMTargetMachine rather than just
TargetMachine around.
Do the same for references in ScheduleDAG and RegUsageInfoCollector.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346183
91177308-0d34-0410-b5e6-
96231b3b80d8
Matthias Braun [Mon, 5 Nov 2018 23:49:13 +0000 (23:49 +0000)]
MachineModuleInfo: Store more specific reference to LLVMTargetMachine; NFC
MachineModuleInfo can only be used in code using lib/CodeGen, hence we
can keep a more specific reference to LLVMTargetMachine rather than just
TargetMachine around.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346182
91177308-0d34-0410-b5e6-
96231b3b80d8
Fangrui Song [Mon, 5 Nov 2018 23:27:53 +0000 (23:27 +0000)]
[DWARF] Support types CU list in .gdb_index dumping
Some executables have non-empty types CU list and -gdb-index would report "<error reporting>" before.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346181
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 5 Nov 2018 23:26:13 +0000 (23:26 +0000)]
[TargetLowering] Change TargetLoweringBase::getPreferredVectorAction to take an MVT instead of an EVT. NFC
The main caller of this already has an MVT and several targets called getSimpleVT inside without checking isSimple. This makes the simpleness explicit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346180
91177308-0d34-0410-b5e6-
96231b3b80d8
Konstantin Zhuravlyov [Mon, 5 Nov 2018 22:44:19 +0000 (22:44 +0000)]
AMDGPU: Add sram-ecc feature
Differential Revision: https://reviews.llvm.org/D53222
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346177
91177308-0d34-0410-b5e6-
96231b3b80d8
Volkan Keles [Mon, 5 Nov 2018 22:25:01 +0000 (22:25 +0000)]
Revert "[GlobalISel] Refactor the artifact combiner a bit by using MIPatternMatch"
This reverts r346166 as it breaks
test-suite-verify-machineinstrs-aarch64-globalisel-O0-g.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346175
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 5 Nov 2018 22:08:17 +0000 (22:08 +0000)]
[X86] Don't turn any_extend from a mask register into a sign_extend during lowering. Add patterns to match any_extend during isel instead.
SimplifyDemandedBits can turn a sign_extend back into an any_extend and trigger an infinite loop. So instead legalize it the same way as a sign_extend, but preserve the opcode. Then just pattern match it the same as sign_extend during isel.
I don't have a reduced test case for such an infinite loop yet.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346170
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 21:51:39 +0000 (21:51 +0000)]
[InstSimplify] fold select (fcmp X, Y), X, Y
This is NFCI for InstCombine because it calls InstSimplify,
so I left the tests for this transform there. As noted in
the code comment, we can allow this fold more often by using
FMF and/or value tracking.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346169
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 21:42:01 +0000 (21:42 +0000)]
[InstSimplify] add tests for select+fcmp; NFC
These are translated from InstCombine's test file with the same name.
We should move the transform from InstCombine to InstSimplify.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346168
91177308-0d34-0410-b5e6-
96231b3b80d8
Volkan Keles [Mon, 5 Nov 2018 20:51:13 +0000 (20:51 +0000)]
[GlobalISel] Refactor the artifact combiner a bit by using MIPatternMatch
Reviewers: aditya_nandakumar
Reviewed By: aditya_nandakumar
Subscribers: rovka, kristof.beyls, llvm-commits
Differential Revision: https://reviews.llvm.org/D54116
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346166
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 5 Nov 2018 19:45:37 +0000 (19:45 +0000)]
[X86] Regenerate test checks in preparation for a patch. NFC
I'm preparing a patch to avoid creating critical edges in cmov expansion. Updating these tests to make the changes by the next patch easier to see.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346161
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexandre Ganea [Mon, 5 Nov 2018 19:20:47 +0000 (19:20 +0000)]
[COFF][LLD] Add link support for Microsoft precompiled headers OBJs
This change allows for link-time merging of debugging information from
Microsoft precompiled types OBJs compiled with cl.exe /Z7 /Yc and /Yu.
This fixes llvm.org/PR34278
Differential Revision: https://reviews.llvm.org/D45213
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346154
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexandre Ganea [Mon, 5 Nov 2018 19:14:10 +0000 (19:14 +0000)]
Only call FlushFileBuffers() when writing executables on Windows
This is a follow-up for "r325274: Call FlushFileBuffers on output files."
Previously, FlushFileBuffers() was called in all cases when writing a file. The objective was to go around a bug in the Windows kernel (as described here: https://randomascii.wordpress.com/2018/02/25/compiler-bug-linker-bug-windows-kernel-bug/). However that is required only when writing EXEs, any other file type doesn't need flushing.
This patch calls FlushFileBuffers() only for EXEs. In addition, we completly disable FlushFileBuffers() for known Windows 10 versions that do not exhibit the original kernel bug.
Differential Revision: https://reviews.llvm.org/D53727
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346152
91177308-0d34-0410-b5e6-
96231b3b80d8
Taewook Oh [Mon, 5 Nov 2018 18:16:32 +0000 (18:16 +0000)]
[MergeICmps] Do not perform the transformation if GEP is used outside of block
Summary:
This patch prevents MergeICmps to performn the transformation if the address operand GEP of the load instruction has a use outside of the load's parent block. Without this patch, compiler crashes with the given test case because the use of `%first.i` is still around when the basic block is erased from https://github.com/llvm-mirror/llvm/blob/master/lib/Transforms/Scalar/MergeICmps.cpp#L620. I think checking `isUsedOutsideOfBlock` with `GEP` is the original intention of the code, as the checking for `LoadI` is already performed in the same function.
This patch is incomplete though, as this makes the pass overly conservative and fails the test `tuple-four-int8.ll`. I believe what needs to be done is checking if GEP has a use outside of block that is not the part of "Comparisons" chain. Submit the patch as of now to prevent compiler crash.
Reviewers: courbet, trentxintong
Reviewed By: courbet
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D54089
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346151
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 18:09:10 +0000 (18:09 +0000)]
[InstCombine] add/adjust tests for fcmp+select substitution; NFC
There was no coverage for at least 2 out of the 4 patterns because
of fcmp canonicalization. The tests and code should be moved to
InstSimplify in a follow-up because this doesn't create any new values.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346150
91177308-0d34-0410-b5e6-
96231b3b80d8
Zaara Syeda [Mon, 5 Nov 2018 17:31:26 +0000 (17:31 +0000)]
[Power9] Add support for stxvw4x.be and stxvd2x.be intrinsics
On Power9, we don't have patterns to select the following intrinsics:
llvm.ppc.vsx.stxvw4x.be
llvm.ppc.vsx.stxvd2x.be
This patch adds support for these.
Differential Revision: https://reviews.llvm.org/D53581
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346148
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 17:26:42 +0000 (17:26 +0000)]
[InstCombine] canonicalize -0.0 to +0.0 in fcmp
As stated in IEEE-754 and discussed in:
https://bugs.llvm.org/show_bug.cgi?id=38086
...the sign of zero does not affect any FP compare predicate.
Known regressions were fixed with:
rL346097 (D54001)
rL346143
The transform will help reduce pattern-matching complexity to solve:
https://bugs.llvm.org/show_bug.cgi?id=39475
...as well as improve CSE and codegen (a zero constant is almost always
easier to produce than 0x80..00).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346147
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 16:50:44 +0000 (16:50 +0000)]
[InstCombine] loosen FP 0.0 constraint for fcmp+select substitution
It looks like we correctly removed edge cases with 0.0 from D50714,
but we were a bit conservative because getBinOpIdentity() doesn't
distinguish between +0.0 and -0.0 and 'nsz' is effectively always
true for fcmp (see discussion in:
https://bugs.llvm.org/show_bug.cgi?id=38086
Without this change, we would get regressions by canonicalizing
to +0.0 in all fcmp, and that's a step towards solving:
https://bugs.llvm.org/show_bug.cgi?id=39475
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346143
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 16:27:03 +0000 (16:27 +0000)]
[InstCombine] adjust tests for select with FP identity op; NFC
These are mislabeled as negative tests.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346142
91177308-0d34-0410-b5e6-
96231b3b80d8
Cameron McInally [Mon, 5 Nov 2018 15:59:49 +0000 (15:59 +0000)]
[FPEnv] Add constrained CEIL/FLOOR/ROUND/TRUNC intrinsics
Differential Revision: https://reviews.llvm.org/D53411
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346141
91177308-0d34-0410-b5e6-
96231b3b80d8
Xin Tong [Mon, 5 Nov 2018 15:49:46 +0000 (15:49 +0000)]
[ThinLTO] Add an option to disable (thin)lto internalization.
Summary:
LTO and ThinLTO optimizes the IR differently.
One source of differences is the amount of internalizations that
can happen.
Add an option to enable/disable internalization so that other
differences can be studied in isolation. e.g. inlining.
There are other things lto and thinlto do differently, I will add
flags to enable/disable them as needed.
Reviewers: tejohnson, pcc, steven_wu
Subscribers: mehdi_amini, inglorion, steven_wu, dexonsmith, dang, llvm-commits
Differential Revision: https://reviews.llvm.org/D53294
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346140
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Mon, 5 Nov 2018 15:49:09 +0000 (15:49 +0000)]
[TargetLowering] Begin generalizing TargetLowering::expandFP_TO_SINT support. NFCI.
Prior to initial work to add vector expansion support, remove assumptions that we're working on scalar types.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346139
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 15:45:01 +0000 (15:45 +0000)]
[InstCombine] add/adjust tests for select with fsub identity op; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346138
91177308-0d34-0410-b5e6-
96231b3b80d8
Cameron McInally [Mon, 5 Nov 2018 15:28:10 +0000 (15:28 +0000)]
[NFCI][FPEnv] Split constrained intrinsic tests
The constrained intrinsic tests have grown in number. Split off
the FMA tests into their own file to reduce double coverage.
Differential Revision: https://reviews.llvm.org/D53932
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346137
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 5 Nov 2018 15:08:36 +0000 (15:08 +0000)]
[InstCombine] add tests for select with FP identity op; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346136
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Mon, 5 Nov 2018 14:54:34 +0000 (14:54 +0000)]
[Inliner] Penalise inlining of calls with loops at Oz
We currently seem to underestimate the size of functions with loops in them,
both in terms of absolute code size and in the difficulties of dealing with
such code. (Calls, for example, can be tail merged to further reduce
codesize). At -Oz, we can then increase code size by inlining small loops
multiple times.
This attempts to penalise functions with loops at -Oz by adding a CallPenalty
for each top level loop in the function. It uses LI (and hence DT) to calculate
the number of loops. As we are dealing with minsize, the inline threshold is
small and functions at this point should be relatively small, making the
construction of these cheap.
Differential Revision: https://reviews.llvm.org/D52716
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346134
91177308-0d34-0410-b5e6-
96231b3b80d8
Stefan Maksimovic [Mon, 5 Nov 2018 14:37:41 +0000 (14:37 +0000)]
[Mips] Supplement long branch pseudo instructions
Expand on LONG_BRANCH_LUi and LONG_BRANCH_(D)ADDiu pseudo
instructions by creating variants which support
less operands/accept GPR64Opnds as their operand in order
to appease the machine verifier pass.
Differential Revision: https://reviews.llvm.org/D53977
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346133
91177308-0d34-0410-b5e6-
96231b3b80d8
Sam Parker [Mon, 5 Nov 2018 14:17:27 +0000 (14:17 +0000)]
[NFC][ARM] Adding extra test for ARM CGP
Added a reproducer that I received a while ago.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346132
91177308-0d34-0410-b5e6-
96231b3b80d8
Neil Henning [Mon, 5 Nov 2018 12:04:48 +0000 (12:04 +0000)]
[AMDGPU] Fix the new atomic optimizer in pixel shaders.
The new atomic optimizer I previously added in D51969 did not work
correctly when a pixel shader was using derivatives, and had helper
lanes active.
To fix this we add an llvm.amdgcn.ps.live call that guards a branch
around the entire atomic operation - ensuring that all helper lanes are
inactive within the wavefront when we compute our atomic results.
I've added a test case that can cause derivatives, and exposes the
problem.
Differential Revision: https://reviews.llvm.org/D53930
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346128
91177308-0d34-0410-b5e6-
96231b3b80d8
Francis Visoiu Mistrih [Mon, 5 Nov 2018 11:57:44 +0000 (11:57 +0000)]
[CMake] Expose opt-remark tooling through libOptRemarks.dylib
* Create an install target for it
* Add it under tools/opt-remarks
* Add an export file for the dylib
* Install the llvm-c/OptRemarks.h header
* Add an API to query its version
rdar://
45458839
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346127
91177308-0d34-0410-b5e6-
96231b3b80d8
Sam Parker [Mon, 5 Nov 2018 11:26:04 +0000 (11:26 +0000)]
[ARM] Turn assert into condition in ARMCGP
Turn the assert in PrepareConstants into a conditon so that we can
handle mul instructions with negative immediates.
Differential Revision: https://reviews.llvm.org/D54094
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346126
91177308-0d34-0410-b5e6-
96231b3b80d8
Sam Parker [Mon, 5 Nov 2018 10:58:37 +0000 (10:58 +0000)]
[ARM][ARMCGP] Remove unecessary zexts and truncs
r345840 slightly changed the way promotion happens which could
result in zext and truncs having the same source and destination
types. This fixes that issue.
We can now also remove the zext and trunc in the following case:
(zext (trunc (promoted op)), i32)
This means that we can no longer treat a value, that is only used by
a sink, to be safe to promote.
I've also added in some extra asserts and replaced a cast for a
dyn_cast.
Differential Revision: https://reviews.llvm.org/D54032
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346125
91177308-0d34-0410-b5e6-
96231b3b80d8
Hans Wennborg [Mon, 5 Nov 2018 09:31:43 +0000 (09:31 +0000)]
Exclude wasm target from Windows packaging due to PR39448
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346122
91177308-0d34-0410-b5e6-
96231b3b80d8
Roman Lebedev [Mon, 5 Nov 2018 09:20:08 +0000 (09:20 +0000)]
[NFC][x86][AArch64] extract-bits.ll: add test with 'ashr'.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346121
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 5 Nov 2018 05:53:06 +0000 (05:53 +0000)]
[DAGCombiner] Use tryFoldToZero to simplify some code and make it work correctly between LegalTypes and LegalOperations.
The original code avoided creating a zero vector after type legalization, but if we're after type legalization the type we have is legal. The real hazard we need to avoid is creating a build vector after op legalization. tryFoldToZero takes care of checking for this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346119
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 5 Nov 2018 05:53:03 +0000 (05:53 +0000)]
[DAGCombiner] Remove an unused argument from tryFoldToZero. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346118
91177308-0d34-0410-b5e6-
96231b3b80d8
Dylan McKay [Mon, 5 Nov 2018 05:49:04 +0000 (05:49 +0000)]
[AVR] Fix a backend bug that left extraneous operands after expansion
This patch fixes a bug in the AVR FRMIDX expansion logic.
The expansion would leave a leftover operand from the original FRMIDX,
but now attached to a MOVWRdRr instruction. The MOVWRdRr instruction
did not expect this operand and so LLVM rejected the machine
instruction.
This would trigger an assertion:
Assertion failed: ((isImpReg || Op.isRegMask() || MCID->isVariadic() ||
OpNo < MCID->getNumOperands() || isMetaDataOp) &&
"Trying to add an operand to a machine instr that is already done!"),
function addOperand, file llvm/lib/CodeGen/MachineInstr.cpp
Tim fixed this so that now the FRMIDX is expanded correctly into
a well-formed MOVWRdRr.
Patch by Tim Neumann
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346117
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 5 Nov 2018 05:02:12 +0000 (05:02 +0000)]
[X86] Custom type legalize v2i8/v2i16/v2i32 mul to use to pmuludq.
v2i8/v2i16/v2i32 are promoted to v2i64. pmuludq takes a v2i64 input and produces a v2i64 output. Since we don't about the upper bits of the type legalized multiply we can use the pmuludq to produce the multiply result for the bits we do care about.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346115
91177308-0d34-0410-b5e6-
96231b3b80d8
Dylan McKay [Mon, 5 Nov 2018 05:00:44 +0000 (05:00 +0000)]
[AVR] Disallow the LDDWRdPtrQ instruction with Z as the destination
This is an AVR-specific workaround for a limitation of the register
allocator that only exposes itself on targets with high register
contention like AVR, which only has three pointer registers.
The three pointer registers are X, Y, and Z.
In most nontrivial functions, Y is reserved for the frame pointer,
as per the calling convention. This leaves X and Z. Some instructions,
such as LPM ("load program memory"), are only defined for the Z
register. Sometimes this just leaves X.
When the backend generates a LDDWRdPtrQ instruction with Z as the
destination pointer, it usually trips up the register allocator
with this error message:
LLVM ERROR: ran out of registers during register allocation
This patch is a hacky workaround. We ban the LDDWRdPtrQ instruction
from ever using the Z register as an operand. This gives the
register allocator a bit more space to allocate, fixing the
regalloc exhaustion error.
Here is a description from the patch author Peter Nimmervoll
As far as I understand the problem occurs when LDDWRdPtrQ uses
the ptrdispregs register class as target register. This should work, but
the allocator can't deal with this for some reason. So from my testing,
it seams like (and I might be totally wrong on this) the allocator reserves
the Z register for the ICALL instruction and then the register class
ptrdispregs only has 1 register left and we can't use Y for source and
destination. Removing the Z register from DREGS fixes the problem but
removing Y register does not.
More information about the bug can be found on the avr-rust issue
tracker at https://github.com/avr-rust/rust/issues/37.
A bug has raised to track the removal of this workaround and a proper
fix; PR39553 at https://bugs.llvm.org/show_bug.cgi?id=39553.
Patch by Peter Nimmervoll
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346114
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 5 Nov 2018 01:21:52 +0000 (01:21 +0000)]
[X86] Fix typo in test comment. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346110
91177308-0d34-0410-b5e6-
96231b3b80d8
Vedant Kumar [Sun, 4 Nov 2018 23:11:57 +0000 (23:11 +0000)]
[HotColdSplitting] Use TTI to inform outlining threshold
Using TargetTransformInfo allows the splitting pass to factor in the
code size cost of instructions as it decides whether or not outlining is
profitable.
This did not regress the overall amount of outlining seen on the handful
of internal frameworks I tested.
Thanks to Jun Bum Lim for suggesting this!
Differential Revision: https://reviews.llvm.org/D53835
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346108
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sun, 4 Nov 2018 21:37:45 +0000 (21:37 +0000)]
[X86] Add nounwind to some tests to remove cfi directives from checks. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346106
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sun, 4 Nov 2018 21:37:43 +0000 (21:37 +0000)]
[X86] Regenerate test checks to merge 32 and 64 bit. Remove stale check prefixes. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346105
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sun, 4 Nov 2018 17:31:27 +0000 (17:31 +0000)]
[X86] Add vector shift by immediate to SimplifyDemandedBitsForTargetNode.
Summary: This also enables some constant folding from KnownBits propagation. This helps on some cases vXi64 case in 32-bit mode where constant vectors appear as vXi32 and a bitcast. This can prevent getNode from constant folding sra/shl/srl.
Reviewers: RKSimon, spatel
Reviewed By: spatel
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D54069
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346102
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Sun, 4 Nov 2018 14:28:48 +0000 (14:28 +0000)]
[ValueTracking] determine sign of 0.0 from select when matching min/max FP
In PR39475:
https://bugs.llvm.org/show_bug.cgi?id=39475
..we may fail to recognize/simplify fabs() in some cases because we do not
canonicalize fcmp with a -0.0 operand.
Adding that canonicalization can cause regressions on min/max FP tests, so
that's this patch: for the purpose of determining whether something is min/max,
let the value returned by the select determine how we treat a 0.0 operand in the fcmp.
This patch doesn't actually change the -0.0 to +0.0. It just changes the analysis, so
we don't fail to recognize equivalent min/max patterns that only differ in the
signbit of 0.0.
Differential Revision: https://reviews.llvm.org/D54001
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346097
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sun, 4 Nov 2018 06:56:32 +0000 (06:56 +0000)]
[DAGCombiner] Remove 'else' after return. NFC
This makes this code consistent with the nearly identical code in visitZERO_EXTEND.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@346090
91177308-0d34-0410-b5e6-
96231b3b80d8