OSDN Git Service

android-x86/external-llvm.git
5 years ago[X86] Strengthen type constraints on some specialized X86 ISD opcodes that don't...
Craig Topper [Wed, 15 May 2019 21:16:28 +0000 (21:16 +0000)]
[X86] Strengthen type constraints on some specialized X86 ISD opcodes that don't have any flexibility. NFC

These particular instructions only operate on 128-bit vectors and have no wider equivalents. And the
element size is always known.

One could argue that MOVSS/MOVSD could be merged, but that's probably disruptive to code in
X86ISelLowering and probably low value.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360815 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[NFC][InstCombine] Add some more tests for pulling binops through shifts
Roman Lebedev [Wed, 15 May 2019 21:15:44 +0000 (21:15 +0000)]
[NFC][InstCombine] Add some more tests for pulling binops through shifts

The ashr variant may see relaxation in https://reviews.llvm.org/D61938

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360814 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[codeview] Finish support for reading and writing S_ANNOTATION records
Reid Kleckner [Wed, 15 May 2019 20:53:39 +0000 (20:53 +0000)]
[codeview] Finish support for reading and writing S_ANNOTATION records

Implement dumping via llvm-pdbutil and llvm-readobj.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360813 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoRevert llvm-svn: 360807
Cameron McInally [Wed, 15 May 2019 20:48:50 +0000 (20:48 +0000)]
Revert llvm-svn: 360807

Somehow submitted this patch twice.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360812 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoPre-commit unary fneg tests to InstSimplify
Cameron McInally [Wed, 15 May 2019 20:27:37 +0000 (20:27 +0000)]
Pre-commit unary fneg tests to InstSimplify

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360808 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoAdd unary fneg to InstSimplify/fp-nan.ll
Cameron McInally [Wed, 15 May 2019 20:27:35 +0000 (20:27 +0000)]
Add unary fneg to InstSimplify/fp-nan.ll

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360807 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoUncomment LLVM_FALLTHROUGH.
Pete Couperus [Wed, 15 May 2019 19:46:17 +0000 (19:46 +0000)]
Uncomment LLVM_FALLTHROUGH.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360798 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoAdd unary fneg to InstSimplify/fp-nan.ll
Cameron McInally [Wed, 15 May 2019 19:37:03 +0000 (19:37 +0000)]
Add unary fneg to InstSimplify/fp-nan.ll

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360797 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[PredicateInfo] Do not process unreachable operands.
Taewook Oh [Wed, 15 May 2019 19:35:38 +0000 (19:35 +0000)]
[PredicateInfo] Do not process unreachable operands.

Summary: We should excluded unreachable operands from processing as their DFS visitation order is undefined. When `renameUses` function sorts `OpsToRename` (https://fburl.com/d2wubn60), the comparator assumes that the parent block of the operand has a corresponding dominator tree node. This is not the case for unreachable operands and crashes the compiler.

Reviewers: dberlin, mgrang, davide

Subscribers: efriedma, hiraditya, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61154

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360796 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[MachineOperand] Add a ChangeToGA method
Nicolai Haehnle [Wed, 15 May 2019 17:48:10 +0000 (17:48 +0000)]
[MachineOperand] Add a ChangeToGA method

Summary:
Analogous to the other ChangeToXXX methods. See the next patch for a
use case.

Change-Id: I6548d614706834fb9109ab3c8fe915e9c6ece2a7

Reviewers: arsenm, kzhuravl

Subscribers: wdng, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61651

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360789 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoRegAlloc: try to fail more gracefully when out of registers
Nicolai Haehnle [Wed, 15 May 2019 17:29:58 +0000 (17:29 +0000)]
RegAlloc: try to fail more gracefully when out of registers

Summary:
The emitError path allows the program to continue, unlike report_fatal_error.
This is friendlier to use cases where LLVM is embedded in a larger program,
because the caller may be able to deal with the error somewhat gracefully.

Change the number of requested NOP bytes in the AArch64 and PowerPC
test cases to avoid triggering an unrelated assertion. The compilation
still fails, as verified by the test.

Change-Id: Iafb9ca341002a597b82e59ddc7a1f13c78758e3d

Reviewers: arsenm, MatzeB

Subscribers: qcolombet, nemanjai, wdng, javed.absar, kristof.beyls, kbarton, jsji, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61489

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360786 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[FileCheck] Fix sphinx error: Make input be gas block
Thomas Preud'homme [Wed, 15 May 2019 15:20:45 +0000 (15:20 +0000)]
[FileCheck] Fix sphinx error: Make input be gas block

Summary:
Change example of input text from being llvm block to being gas block
since that text is made-up assembly.

Reviewers: jhenderson, jdenny, probinson, arichardson

Subscribers: llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61893

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360781 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[JumpThreading] A bug fix for stale loop info after unfold select
Hiroshi Yamauchi [Wed, 15 May 2019 15:15:16 +0000 (15:15 +0000)]
[JumpThreading] A bug fix for stale loop info after unfold select

Summary:
The return value of a TryToUnfoldSelect call was not checked, which led to an
incorrectly preserved loop info and some crash.

The original crash was reported on https://reviews.llvm.org/D59514.

Reviewers: davidxl, amehsan

Reviewed By: davidxl

Subscribers: fhahn, brzycki, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61920

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360780 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AMDGPU] Increases available SGPR for Calling Convention
Ryan Taylor [Wed, 15 May 2019 14:43:55 +0000 (14:43 +0000)]
[AMDGPU] Increases available SGPR for Calling Convention

Summary:
SGPR in CC can be either hw initialized or set by other chained shaders
and so this increases the SGPR count availalbe to CC to 105.

Change-Id: I3dfadc750fe4a3e2bd07117a2899fd13f3e2fef3

Subscribers: arsenm, kzhuravl, jvesely, wdng, nhaehnle, yaxunl, dstuttard, tpr, t-tye, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61261

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360778 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoTeach InstSimplify -X + X --> 0.0 about unary FNeg
Cameron McInally [Wed, 15 May 2019 14:31:33 +0000 (14:31 +0000)]
Teach InstSimplify -X + X --> 0.0 about unary FNeg

Differential Revision: https://reviews.llvm.org/D61916

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360777 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoRevert r360771 "[MergeICmps] Simplify the code."
Clement Courbet [Wed, 15 May 2019 14:21:59 +0000 (14:21 +0000)]
Revert r360771 "[MergeICmps] Simplify the code."

Breaks a bunch of builbdots.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360776 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[MergeICmps] Fix r360771.
Clement Courbet [Wed, 15 May 2019 14:00:45 +0000 (14:00 +0000)]
[MergeICmps] Fix r360771.

Twine references a StringRef by reference, not value...

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360775 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoRevert "[Salvage] Change salvage debug info implementation to use DW_OP_LLVM_convert...
Stephen Tozer [Wed, 15 May 2019 13:41:44 +0000 (13:41 +0000)]
Revert "[Salvage] Change salvage debug info implementation to use DW_OP_LLVM_convert where needed"

This reverts r360772 due to build issues.
Reverted commit: 17dd4d7403770bd683675e45f5517e0cdb8f9b2b.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360773 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[Salvage] Change salvage debug info implementation to use DW_OP_LLVM_convert where...
Stephen Tozer [Wed, 15 May 2019 13:15:48 +0000 (13:15 +0000)]
[Salvage] Change salvage debug info implementation to use DW_OP_LLVM_convert where needed

Fixes issue: https://bugs.llvm.org/show_bug.cgi?id=40645

Previously, LLVM had no functional way of performing casts inside of a
DIExpression(), which made salvaging cast instructions other than Noop
casts impossible. With the recent addition of DW_OP_LLVM_convert this
salvaging is now possible, and so can be used to fix the attached bug as
well as any cases where SExt instruction results are lost in the
debugging metadata. This patch introduces this fix by expanding the
salvage debug info method to cover these cases using the new operator.

Differential revision: https://reviews.llvm.org/D61184

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360772 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[MergeICmps] Simplify the code.
Clement Courbet [Wed, 15 May 2019 13:04:24 +0000 (13:04 +0000)]
[MergeICmps] Simplify the code.

Instead of patching the original blocks, we now generate new blocks and
delete the old blocks. This results in simpler code with a less twisted
control flow (see the change in `entry-block-shuffled.ll`).

This will make https://reviews.llvm.org/D60318 simpler by making it more
obvious where control flow created and deleted.

Reviewers: gchatelet

Subscribers: hiraditya, llvm-commits, spatel

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61736

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360771 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoRevert rL360675 : [APFloat] APFloat::Storage::Storage - fix use after move
Simon Pilgrim [Wed, 15 May 2019 13:03:10 +0000 (13:03 +0000)]
Revert rL360675 : [APFloat] APFloat::Storage::Storage - fix use after move

This was mentioned both in https://www.viva64.com/en/b/0629/ and by scan-build checks
........
There's concerns this may just introduce a use-after-free instead.....

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360770 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[ARM] Don't use the Machine Scheduler for cortex-m at minsize
David Green [Wed, 15 May 2019 12:58:02 +0000 (12:58 +0000)]
[ARM] Don't use the Machine Scheduler for cortex-m at minsize

The new cortex-m schedule in rL360768 helps performance, but can increase the
amount of high-registers used. This, on average, ends up increasing the
codesize by a fair amount (because less instructions are converted from T2 to
T1). On cortex-m at -Oz, where we are quite size-paranoid, it is better to use
the existing DAG scheduler with the RegPressure scheduling preference (at least
until the issues around T2 vs T1 instructions can be improved).

I have also made sure that the Sched::RegPressure dag scheduler is always
chosen for MinSize.

The test shows one case where we increase the number of registers used.

Differential Revision: https://reviews.llvm.org/D61882

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360769 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[ARM] Cortex-M4 schedule
David Green [Wed, 15 May 2019 12:41:58 +0000 (12:41 +0000)]
[ARM] Cortex-M4 schedule

This patch adds a simple Cortex-M4 schedule, renaming the existing M3
schedule to M4 and filling in the latencies as-per the Cortex-M4 TRM:
https://developer.arm.com/docs/ddi0439/latest

Most of these are 1, with the important exception being loads taking 2
cycles. A few others are also higher, but I don't believe they make a
large difference. I've repurposed the M3 schedule as the latencies are
mostly the same between the two cores, with the M4 having more FP and
DSP instructions. We also turn on MISched and UseAA for the cores that
now use this.

It also adds some schedule Write's to various instruction to make things
simpler.

Differential Revision: https://reviews.llvm.org/D54142

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360768 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agogn build: Merge r360671
Nico Weber [Wed, 15 May 2019 12:08:45 +0000 (12:08 +0000)]
gn build: Merge r360671

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360766 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[mips] LLVM and GAS now use same instructions for CFA Definition. NFCI
Simon Atanasyan [Wed, 15 May 2019 12:05:27 +0000 (12:05 +0000)]
[mips] LLVM and GAS now use same instructions for CFA Definition. NFCI

LLVM previously used `DW_CFA_def_cfa` instruction in .eh_frame to set
the register and offset for current CFA rule. We change it to
`DW_CFA_def_cfa_register` which is the same one used by GAS that only
changes the register but keeping the old offset.

Patch by Mirko Brkusanin.

Differential Revision: https://reviews.llvm.org/D61899

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360765 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agogn build: Run `git ls-files '*.gn' '*.gni' | xargs llvm/utils/gn/gn.py format`
Nico Weber [Wed, 15 May 2019 12:03:10 +0000 (12:03 +0000)]
gn build: Run `git ls-files '*.gn' '*.gni' | xargs llvm/utils/gn/gn.py format`

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360764 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoarm64_32: add some unittests that were in the wrong commit.
Tim Northover [Wed, 15 May 2019 12:01:04 +0000 (12:01 +0000)]
arm64_32: add some unittests that were in the wrong commit.

Accidentally dropped them when committing the arm64_32 binutils support.
There's no change to real code.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360763 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[NFC][InstCombine] Regenerate trunc.ll test
Roman Lebedev [Wed, 15 May 2019 10:24:38 +0000 (10:24 +0000)]
[NFC][InstCombine] Regenerate trunc.ll test

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360759 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[LV] Move getScalarizationOverhead and vector call cost computations to CM. (NFC)
Florian Hahn [Wed, 15 May 2019 10:05:49 +0000 (10:05 +0000)]
[LV] Move getScalarizationOverhead and vector call cost computations to CM. (NFC)

This reduces the number of parameters we need to pass in and they seem a
natural fit in LoopVectorizationCostModel. Also simplifies things for
D59995.

As a follow up refactoring, we could only expose a expose a
shouldUseVectorIntrinsic() helper in LoopVectorizationCostModel, instead
of calling getVectorCallCost/getVectorIntrinsicCost in
InnerLoopVectorizer/VPRecipeBuilder.

Reviewers: Ayal, hsaito, dcaballe, rengolin

Reviewed By: rengolin

Differential Revision: https://reviews.llvm.org/D61638

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360758 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[[DAGCombiner][NFC] Add a comment.
Clement Courbet [Wed, 15 May 2019 08:21:18 +0000 (08:21 +0000)]
[[DAGCombiner][NFC] Add a comment.

As suggested in D61846.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360755 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[X86] Use OR32mi8Locked instead of LOCK_OR32mi8 in emitLockedStackOp.
Craig Topper [Wed, 15 May 2019 04:15:46 +0000 (04:15 +0000)]
[X86] Use OR32mi8Locked instead of LOCK_OR32mi8 in emitLockedStackOp.

They encode the same way, but OR32mi8Locked sets hasUnmodeledSideEffects set
which should be stronger than the mayLoad/mayStore on LOCK_OR32mi8. I think
this makes sense since we are using it as a fence.

This also seems to hide the operation from the speculative load hardening pass
so I've reverted r360511.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360747 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoFix 2-field llvm.global_ctors `REQUIRES: asserts` tests after rL360742
Fangrui Song [Wed, 15 May 2019 03:08:21 +0000 (03:08 +0000)]
Fix 2-field llvm.global_ctors `REQUIRES: asserts` tests after rL360742

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360743 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[IR] Disallow llvm.global_ctors and llvm.global_dtors of the 2-field form in textual...
Fangrui Song [Wed, 15 May 2019 02:35:32 +0000 (02:35 +0000)]
[IR] Disallow llvm.global_ctors and llvm.global_dtors of the 2-field form in textual format

The 3-field form was introduced by D3499 in 2014 and the legacy 2-field
form was planned to be removed in LLVM 4.0

For the textual format, this patch migrates the existing 2-field form to
use the 3-field form and deletes the compatibility code.
test/Verifier/global-ctors-2.ll checks we have a friendly error message.

For bitcode, lib/IR/AutoUpgrade UpgradeGlobalVariables will upgrade the
2-field form (add i8* null as the third field).

Reviewed By: rnk, dexonsmith

Differential Revision: https://reviews.llvm.org/D61547

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360742 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[NFC] Reuse a helper function to eliminate duplicate code
Philip Reames [Wed, 15 May 2019 01:39:07 +0000 (01:39 +0000)]
[NFC] Reuse a helper function to eliminate duplicate code

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360740 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[XCore] Create a TargetInfo header. NFC
Richard Trieu [Wed, 15 May 2019 01:28:30 +0000 (01:28 +0000)]
[XCore] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360738 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[X86] Create a TargetInfo header. NFC
Richard Trieu [Wed, 15 May 2019 01:17:58 +0000 (01:17 +0000)]
[X86] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360736 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[WebAssembly] Create a TargetInfo header. NFC
Richard Trieu [Wed, 15 May 2019 01:03:00 +0000 (01:03 +0000)]
[WebAssembly] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360735 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[SystemZ] Create a TargetInfo header. NFC
Richard Trieu [Wed, 15 May 2019 00:46:18 +0000 (00:46 +0000)]
[SystemZ] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360734 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[Sparc] Create a TargetInfo header. NFC
Richard Trieu [Wed, 15 May 2019 00:35:37 +0000 (00:35 +0000)]
[Sparc] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360733 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[RISCV] Create a TargetInfo header. NFC
Richard Trieu [Wed, 15 May 2019 00:24:15 +0000 (00:24 +0000)]
[RISCV] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360732 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[PowerPC] Create a TargetInfo header. NFC
Richard Trieu [Wed, 15 May 2019 00:09:58 +0000 (00:09 +0000)]
[PowerPC] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360731 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[NVPTX] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 23:56:18 +0000 (23:56 +0000)]
[NVPTX] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360729 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[MSP430] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 23:45:18 +0000 (23:45 +0000)]
[MSP430] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360728 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[Mips] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 23:34:37 +0000 (23:34 +0000)]
[Mips] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360727 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[Lanai] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 23:17:18 +0000 (23:17 +0000)]
[Lanai] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360726 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[Hexagon] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 23:04:55 +0000 (23:04 +0000)]
[Hexagon] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360724 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[BPF] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 22:54:06 +0000 (22:54 +0000)]
[BPF] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360722 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AVR] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 22:41:58 +0000 (22:41 +0000)]
[AVR] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360721 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoUse an offset from TOS for idempotent rmw locked op lowering
Philip Reames [Tue, 14 May 2019 22:32:42 +0000 (22:32 +0000)]
Use an offset from TOS for idempotent rmw locked op lowering

This was the portion split off D58632 so that it could follow the redzone API cleanup. Note that I changed the offset preferred from -8 to -64. The difference should be very minor, but I thought it might help address one concern which had been previously raised.

Differential Revision: https://reviews.llvm.org/D61862

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360719 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[ARM] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 22:29:50 +0000 (22:29 +0000)]
[ARM] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360718 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[ARC] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 22:06:04 +0000 (22:06 +0000)]
[ARC] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360716 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AMDGPU] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 21:54:37 +0000 (21:54 +0000)]
[AMDGPU] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360713 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AArch64] Create a TargetInfo header. NFC
Richard Trieu [Tue, 14 May 2019 21:33:53 +0000 (21:33 +0000)]
[AArch64] Create a TargetInfo header.  NFC

Move the declarations of getThe<Name>Target() functions into a new header in
TargetInfo and make users of these functions include this new header.
This fixes a layering problem.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360709 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[NewPM] Port HWASan and Kernel HWASan
Leonard Chan [Tue, 14 May 2019 21:17:21 +0000 (21:17 +0000)]
[NewPM] Port HWASan and Kernel HWASan

Port hardware assisted address sanitizer to new PM following the same guidelines as msan and tsan.

Changes:
- Separate HWAddressSanitizer into a pass class and a sanitizer class.
- Create new PM wrapper pass for the sanitizer class.
- Use the getOrINsert pattern for some module level initialization declarations.
- Also enable kernel-kwasan in new PM
- Update llvm tests and add clang test.

Differential Revision: https://reviews.llvm.org/D61709

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360707 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[NFC][CodeGen][X86][AArch64] Add and-const-mask + const-shift pattern tests
Roman Lebedev [Tue, 14 May 2019 20:17:04 +0000 (20:17 +0000)]
[NFC][CodeGen][X86][AArch64] Add and-const-mask + const-shift pattern tests

Unlike instcombine, we currently don't turn and+shift into shift+and.
We probably should, likely unconditionally.

While i'm adding only all-ones (potentially shifted) mask,
this obviously isn't limited to any particular mask pattern:
https://rise4fun.com/Alive/kmX

Related to https://bugs.llvm.org/show_bug.cgi?id=41874

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360706 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[LICM] Allow AliasSetMap to contain top-level loops.
Florian Hahn [Tue, 14 May 2019 19:41:36 +0000 (19:41 +0000)]
[LICM] Allow AliasSetMap to contain top-level loops.

When an outer loop gets deleted by a different pass, before LICM visits
it, we cannot clean up its sub-loops in AliasSetMap, because at the
point we receive the deleteAnalysisLoop callback for the outer loop, the loop
object is already invalid and we cannot access its sub-loops any longer.

Reviewers: asbirlea, sanjoy, chandlerc

Reviewed By: asbirlea

Differential Revision: https://reviews.llvm.org/D61904

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360704 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AMDGPU][GFX8][GFX9] Corrected predicate of v_*_co_u32 aliases
Dmitry Preobrazhensky [Tue, 14 May 2019 19:16:24 +0000 (19:16 +0000)]
[AMDGPU][GFX8][GFX9] Corrected predicate of v_*_co_u32 aliases

Reviewers: rampitec, arsenm

Differential Revision: https://reviews.llvm.org/D61905

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360702 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[LVI][CVP] Add support for abs/nabs select pattern flavor
Nikita Popov [Tue, 14 May 2019 18:53:47 +0000 (18:53 +0000)]
[LVI][CVP] Add support for abs/nabs select pattern flavor

Based on ConstantRange support added in D61084, we can now handle
abs and nabs select pattern flavors in LVI.

Differential Revision: https://reviews.llvm.org/D61794

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360700 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[MemorySSA] LoopSimplify preserves MemorySSA only when flag is flipped.
Alina Sbirlea [Tue, 14 May 2019 18:07:18 +0000 (18:07 +0000)]
[MemorySSA] LoopSimplify preserves MemorySSA only when flag is flipped.

LoopSimplify can preserve MemorySSA after r360270.
But the MemorySSA analysis is retrieved and preserved only when the
EnableMSSALoopDependency is set to true. Use the same conditional to
mark the pass as preserved, otherwise subsequent passes will get an
invalid analysis.
Resolves PR41853.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360697 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoFix a release mode warning introduced in r360694
Philip Reames [Tue, 14 May 2019 17:50:06 +0000 (17:50 +0000)]
Fix a release mode warning introduced in r360694

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360696 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[IndVars] Extend reasoning about loop invariant exits to non-header blocks
Philip Reames [Tue, 14 May 2019 17:20:10 +0000 (17:20 +0000)]
[IndVars] Extend reasoning about loop invariant exits to non-header blocks

Noticed while glancing through the code for other reasons.  The extension is trivial enough, decided to just do it.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360694 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoSupport FNeg in SpeculativeExecution pass
Cameron McInally [Tue, 14 May 2019 16:51:18 +0000 (16:51 +0000)]
Support FNeg in SpeculativeExecution pass

Differential Revision: https://reviews.llvm.org/D61910

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360692 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[Test] Autogen a test for ease of later changing
Philip Reames [Tue, 14 May 2019 16:37:29 +0000 (16:37 +0000)]
[Test] Autogen a test for ease of later changing

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360690 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AMDGPU] Fixed handling of imemdiate i1 literals
Stanislav Mekhanoshin [Tue, 14 May 2019 16:18:00 +0000 (16:18 +0000)]
[AMDGPU] Fixed handling of imemdiate i1 literals

This bug was exposed by the rL360395.

Differential Revision: https://reviews.llvm.org/D61812

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360689 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AMDGPU] Fixed +DumpCode
Tim Renouf [Tue, 14 May 2019 16:17:14 +0000 (16:17 +0000)]
[AMDGPU] Fixed +DumpCode

The +DumpCode attribute is a horrible hack in AMDGPU to embed the
disassembly of the generated code into the elf file. It is used by LLPC
to implement an extension that allows the application to read back the
disassembly of the code. Longer term, we should re-implement that by
using the LLVM disassembler from the Vulkan driver.

Recent LLVM changes broke +DumpCode. With -filetype=asm it crashed, and
with -filetype=obj I think it did not include any instructions, only the
labels. Fixed with this commit: now it has no effect with -filetype=asm,
and works as intended with -filetype=obj.

Differential Revision: https://reviews.llvm.org/D60682

Change-Id: I6436d86fe2ea220d74a643a85e64753747c9366b

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360688 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[ARM][CMSE] Add cmse intrinsics for TT instructions
Javed Absar [Tue, 14 May 2019 16:14:24 +0000 (16:14 +0000)]
[ARM][CMSE] Add cmse intrinsics for TT instructions

Defines intrinsics cmse_TT, cmse_TTT, cmse_TTA, cmse_TTAT.
No tests here as the tests are in patches that uses these.
Reviewed By: Todd Snider, Dave Green
Differential Revision:  https://reviews.llvm.org/D59888

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360687 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AMDGPU] gfx1010 Strengthen some SMEM WAR hazard unit tests. NFC.
Stanislav Mekhanoshin [Tue, 14 May 2019 16:04:03 +0000 (16:04 +0000)]
[AMDGPU] gfx1010 Strengthen some SMEM WAR hazard unit tests. NFC.

Tighten conditions on SMEM WAR hazard unit tests to ensure rejection
of workaround insertion where a s_waitcnt is present in dependency
chain. The current workaround code already conforms to these revise
tests.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360686 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[X86] Disable shouldFoldConstantShiftPairToMask for scalar shifts on AMD targets...
Simon Pilgrim [Tue, 14 May 2019 15:21:28 +0000 (15:21 +0000)]
[X86] Disable shouldFoldConstantShiftPairToMask for scalar shifts on AMD targets (PR40758)

D61068 handled vector shifts, this patch does the same for scalars where there are similar number of pipes for shifts as bit ops - this is true almost entirely for AMD targets where the scalar ALUs are well balanced.

This combine avoids AND immediate mask which usually means we reduce encoding size.

Some tests show use of (slow, scaled) LEA instead of SHL in some cases, but thats due to particular shift immediates - shift+mask generate these just as easily.

Differential Revision: https://reviews.llvm.org/D61830

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360684 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AArch64][SVE2] Asm: add SQRDMLAH/SQRDMLSH instructions
Cullen Rhodes [Tue, 14 May 2019 15:10:16 +0000 (15:10 +0000)]
[AArch64][SVE2] Asm: add SQRDMLAH/SQRDMLSH instructions

Summary:
This patch adds support for the indexed and unpredicated vectors forms of the
SQRDMLAH and SQRDMLSH instructions.

The specification can be found here:
https://developer.arm.com/docs/ddi0602/latest

Reviewed By: rovka

Differential Revision: https://reviews.llvm.org/D61515

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360683 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[AArch64][SVE2] Asm: add integer multiply-add/subtract (indexed) instructions
Cullen Rhodes [Tue, 14 May 2019 15:01:00 +0000 (15:01 +0000)]
[AArch64][SVE2] Asm: add integer multiply-add/subtract (indexed) instructions

Summary:
This patch adds support for the following instructions:

  MLA mul-add, writing addend (Zda = Zda +  Zn * Zm[idx])
  MLS mul-sub, writing addend (Zda = Zda + -Zn * Zm[idx])

Predicated forms of these instructions were added in SVE.

The specification can be found here:
https://developer.arm.com/docs/ddi0602/latest

Reviewed By: rovka

Differential Revision: https://reviews.llvm.org/D61514

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360682 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoReplace lit feature keyword 'not_COFF' with 'uses_COFF'.
Paul Robinson [Tue, 14 May 2019 14:51:54 +0000 (14:51 +0000)]
Replace lit feature keyword 'not_COFF' with 'uses_COFF'.

Differential Revision: https://reviews.llvm.org/D61791

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360680 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoDWARF v5: emit DW_AT_addr_base if DW_AT_low_pc references .debug_addr
Fangrui Song [Tue, 14 May 2019 14:37:26 +0000 (14:37 +0000)]
DWARF v5: emit DW_AT_addr_base if DW_AT_low_pc references .debug_addr

The condition !AddrPool.empty() is tested before attachRangesOrLowHighPC(), which may add an entry to AddrPool. We emit DW_AT_low_pc (DW_FORM_addrx) but may incorrectly omit DW_AT_addr_base for LineTablesOnly. This can be easily reproduced:

clang -gdwarf-5 -gmlt -c a.cc

Fix this by moving !AddrPool.empty() below.

This was discovered while investigating an lld crash (fixed by D61889) on such object files: ld.lld --gdb-index a.o

Reviewed By: probinson

Differential Revision: https://reviews.llvm.org/D61891

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360678 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[PowerPC] Custom lower known CR bit spills
Lei Huang [Tue, 14 May 2019 14:27:06 +0000 (14:27 +0000)]
[PowerPC] Custom lower known CR bit spills

For known CRBit spills, CRSET/CRUNSET, it is more efficient to load and spill
the known value instead of extracting the bit.

eg. This sequence is currently used to spill a CRUNSET:
    crclr   4*cr5+lt
    mfocrf  r3,4
    rlwinm  r3,r3,20,0,0
    stw     r3,132(r1)

This patch custom lower it to:
    li  r3,0
    stw r3,132(r1)

Differential Revision: https://reviews.llvm.org/D61754

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360677 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[llvm-readobj] - Apply clang format. NFC.
George Rimar [Tue, 14 May 2019 14:22:44 +0000 (14:22 +0000)]
[llvm-readobj] - Apply clang format. NFC.

I am a bit tired of the formatting issues.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360676 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[APFloat] APFloat::Storage::Storage - fix use after move
Simon Pilgrim [Tue, 14 May 2019 14:13:30 +0000 (14:13 +0000)]
[APFloat] APFloat::Storage::Storage - fix use after move

This was mentioned both in https://www.viva64.com/en/b/0629/ and by scan-build checks

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360675 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[lit][tests]Add feature libcxx-used and use it in llvm-*-fuzzer tests
Xing Xue [Tue, 14 May 2019 13:54:33 +0000 (13:54 +0000)]
[lit][tests]Add feature libcxx-used and use it in llvm-*-fuzzer tests

When a LLVM binary such as llvm-*-fuzzer is built with libc++, it has dependency on libc++. The path to find shared libraries specified in llvm-*-fuzzer is relative. As a result, these binaries cannot be copied to an arbitrary directory and launched from there. Changes in this patch add a LIT feature to indicate that libc++ is used to build and, based on the feature exclude test cases that test by copying llvm-*-fuzzer binaries to a directory.

Reviewers: hubert.reinterpretcast, dberris, amyk, jasonliu, EricWF

Reviewed By: hubert.reinterpretcast, amyk

Subscribers: javed.absar, jsji, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61265

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360672 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoSave the induction binary operator in IVDescriptors for non FP induction variables.
Kit Barton [Tue, 14 May 2019 13:26:36 +0000 (13:26 +0000)]
Save the induction binary operator in IVDescriptors for non FP induction variables.

Summary:
Currently InductionBinOps are only saved for FP induction variables, the PR extends it with non FP induction variable, so user of IVDescriptors can query the InductionBinOps for integer induction variables.

The changes in hasUnsafeAlgebra() and getUnsafeAlgebraInst() are required for the existing LIT test cases to pass. As described in the comment of the two functions, one of the requirement to return true is it is a FP induction variable. The checks was not needed because InductionBinOp was not set on non FP cases before.

https://reviews.llvm.org/D60565 depends on the patch.

Committed on behalf of @Whitney (Whitney Tsang).

Reviewers: jdoerfert, kbarton, fhahn, hfinkel, dmgreen, Meinersbur

Reviewed By: jdoerfert

Subscribers: mgorny, hiraditya, jsji, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61329

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360671 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoTableGen: support #ifndef in addition to #ifdef.
Tim Northover [Tue, 14 May 2019 13:04:25 +0000 (13:04 +0000)]
TableGen: support #ifndef in addition to #ifdef.

TableGen has a limited preprocessor, which only really supports
easier.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360670 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoReinstate "FileCheck [5/12]: Introduce regular numeric variables"
Thomas Preud'homme [Tue, 14 May 2019 11:58:30 +0000 (11:58 +0000)]
Reinstate "FileCheck [5/12]: Introduce regular numeric variables"

This reinstates r360578 (git e47362c1ec1ea31b626336cc05822035601c3e57),
reverted in r360653 (git 004393681c25e34e921adccc69ae6378090dee54),
with a fix for the list added in FileCheck.rst to build without error.

Copyright:
    - Linaro (changes up to diff 183612 of revision D55940)
    - GraphCore (changes in later versions of revision D55940 and
                 in new revision created off D55940)

Reviewers: jhenderson, chandlerc, jdenny, probinson, grimar,
arichardson, rnk

Subscribers: hiraditya, llvm-commits, probinson, dblaikie, grimar,
arichardson, tra, rnk, kristina, hfinkel, rogfer01, JonChesterfield

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D60385

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360665 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[X86] X86TargetLowering::LowerINTRINSIC_WO_CHAIN - ensure rounding control is initial...
Simon Pilgrim [Tue, 14 May 2019 11:30:39 +0000 (11:30 +0000)]
[X86] X86TargetLowering::LowerINTRINSIC_WO_CHAIN - ensure rounding control is initialized. NFCI.

Fixes scan-build warnings

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360664 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoAArch64: support binutils-like things on arm64_32.
Tim Northover [Tue, 14 May 2019 11:25:44 +0000 (11:25 +0000)]
AArch64: support binutils-like things on arm64_32.

This adds support for the arm64_32 watchOS ABI to LLVM's low level tools,
teaching them about the specific MachO choices and constants needed to
disassemble things.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360663 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoGlobalOpt: do not promote globals used atomically to constants.
Tim Northover [Tue, 14 May 2019 11:03:13 +0000 (11:03 +0000)]
GlobalOpt: do not promote globals used atomically to constants.

Some atomic loads are implemented as cmpxchg (particularly if large or
floating), and that usually requires write access to the memory involved
or it will segfault.

We can still propagate the constant value to users we understand though.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360662 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[llvm-objcopy] Cache gnu_debuglink's target CRC
James Henderson [Tue, 14 May 2019 10:59:04 +0000 (10:59 +0000)]
[llvm-objcopy] Cache gnu_debuglink's target CRC

.gnu_debuglink section contains information regarding file with
debugging symbols, identified by its CRC32. This target file is not
intended to ever change or it would invalidate the stored checksum, yet
the checksum is calculated over and over again for each of the objects
inside the archive, usually hundreds of times.

This patch precomputes the CRC32 of the target once and then reuses the
value where required, saving lots of redundant I/O.

The error message reported should stay the same, although now it might
be reported earlier.

Reviewed by: jhenderson, jakehehrlich, MaskRay

Differential Revision: https://reviews.llvm.org/D61343

Patch by Michal Janiszewski

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360661 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[test]Make test work on Windows
James Henderson [Tue, 14 May 2019 10:53:35 +0000 (10:53 +0000)]
[test]Make test work on Windows

Previously, the test didn't work because '\' characters appeared in the
sed string, causing bogus escape characters to form in the substituted
string literal. Switching to using '%/p' causes the path to be emitted
with '/' characters instead, so that there are are no escaping issues.

Reviewed by: kzhuravl, grimar

Differential Revision: https://reviews.llvm.org/D61856

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360660 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[MemorySanitizer] getMMXVectorTy - assert valid element size. NFCI.
Simon Pilgrim [Tue, 14 May 2019 10:29:18 +0000 (10:29 +0000)]
[MemorySanitizer] getMMXVectorTy - assert valid element size. NFCI.

Fixes scan-build warnings

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360658 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[IRTranslator] Don't hardcode GEP index type
Diana Picus [Tue, 14 May 2019 09:25:17 +0000 (09:25 +0000)]
[IRTranslator] Don't hardcode GEP index type

When breaking up loads and stores of aggregates, the IRTranslator uses
LLT::scalar(64) for the index type of the G_GEP instructions that
compute the addresses. This is unnecessarily large for 32-bit targets.
Use the int ptr type provided by the DataLayout instead.

Note that we're already doing the right thing when translating
getelementptr instructions from the IR. This is just an oversight when
generating new ones while translating loads/stores.

Both x86 and AArch64 already have tests confirming that the old
behaviour is preserved for 64-bit targets.

Differential Revision: https://reviews.llvm.org/D61852

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360656 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoRevert "FileCheck [5/12]: Introduce regular numeric variables"
Thomas Preud'homme [Tue, 14 May 2019 08:43:11 +0000 (08:43 +0000)]
Revert "FileCheck [5/12]: Introduce regular numeric variables"

This reverts r360578 (git e47362c1ec1ea31b626336cc05822035601c3e57) to
solve the sphinx build failure on
http://lab.llvm.org:8011/builders/llvm-sphinx-docs buildbot.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360653 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agoAdd guidelines/recommendations for organizers of LLVM Socials
Alex Denisov [Tue, 14 May 2019 07:20:58 +0000 (07:20 +0000)]
Add guidelines/recommendations for organizers of LLVM Socials

Differential Revision: https://reviews.llvm.org/D61550

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360651 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[X86] Prefer locked stack op over mfence for seq_cst 64-bit stores on 32-bit targets
Philip Reames [Tue, 14 May 2019 04:43:37 +0000 (04:43 +0000)]
[X86] Prefer locked stack op over mfence for seq_cst 64-bit stores on 32-bit targets

This is a follow on to D58632, with the same logic. Given a memory operation which needs ordering, but doesn't need to modify any particular address, prefer to use a locked stack op over an mfence.

Differential Revision: https://reviews.llvm.org/D61863

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360649 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[Object] Change ObjectFile::getSectionContents to return Expected<ArrayRef<uint8_t>>
Fangrui Song [Tue, 14 May 2019 04:22:51 +0000 (04:22 +0000)]
[Object] Change ObjectFile::getSectionContents to return Expected<ArrayRef<uint8_t>>

Change
std::error_code getSectionContents(DataRefImpl, StringRef &) const;
to
Expected<ArrayRef<uint8_t>> getSectionContents(DataRefImpl) const;

Many object formats use ArrayRef<uint8_t> as the underlying type, which
is generally better than StringRef to represent binary data, so change
the type to decrease the number of type conversions.

Reviewed By: ruiu, sbc100

Differential Revision: https://reviews.llvm.org/D61781

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360648 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agogn build: add Hexagon target
David L. Jones [Tue, 14 May 2019 04:13:59 +0000 (04:13 +0000)]
gn build: add Hexagon target

Differential Revision: https://reviews.llvm.org/D61819

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360647 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agogn build: add Sparc target
David L. Jones [Tue, 14 May 2019 04:02:50 +0000 (04:02 +0000)]
gn build: add Sparc target

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360645 91177308-0d34-0410-b5e6-96231b3b80d8

5 years agogn build: add Lanai target
David L. Jones [Tue, 14 May 2019 03:52:33 +0000 (03:52 +0000)]
gn build: add Lanai target

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360644 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[PowerPC][NFC] Fix typos in triples
Jinsong Ji [Tue, 14 May 2019 03:11:24 +0000 (03:11 +0000)]
[PowerPC][NFC] Fix typos in triples

Found by bzEq (Kai Luo).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360643 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[X86] Use X86 instead of X32 as a check prefix in atomic-idempotent.ll. NFC
Craig Topper [Tue, 14 May 2019 03:07:56 +0000 (03:07 +0000)]
[X86] Use X86 instead of X32 as a check prefix in atomic-idempotent.ll. NFC

X32 can refer to a 64-bit ABI that uses 32-bit ints, longs, and pointers.

I plan to add gnux32 command lines to this test so this prepares for that.

Also remove some check lines that have a prefix that is not in any run lines.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360642 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[SDAG] fix unused variable warning and unneeded indirection; NFC
Sanjay Patel [Tue, 14 May 2019 00:57:31 +0000 (00:57 +0000)]
[SDAG] fix unused variable warning and unneeded indirection; NFC

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360640 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[SDAG, x86] allow targets to override test for binop opcodes
Sanjay Patel [Tue, 14 May 2019 00:39:40 +0000 (00:39 +0000)]
[SDAG, x86] allow targets to override test for binop opcodes

This follows the pattern of the existing isCommutativeBinOp().

x86 shows improvements from vector narrowing for the min/max opcodes.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360639 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[coroutines] Fix spills of static array allocas
Gor Nishanov [Mon, 13 May 2019 23:58:24 +0000 (23:58 +0000)]
[coroutines] Fix spills of static array allocas

Summary:
CoroFrame was not considering static array allocas, and was only ever reserving a single element in the coroutine frame.
This meant that stores to the non-zero'th element would corrupt later frame data.

Store static array allocas as field arrays in the coroutine frame.

Added test.

Committed by Gor Nishanov on behalf of ben-clayton
Reviewers: GorNishanov, modocache

Reviewed By: GorNishanov

Subscribers: Orlando, capn, EricWF, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D61372

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360636 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[gn] Fix build
Vitaly Buka [Mon, 13 May 2019 22:30:53 +0000 (22:30 +0000)]
[gn] Fix build

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360629 91177308-0d34-0410-b5e6-96231b3b80d8

5 years ago[X86] Use ISD::MERGE_VALUES to return from lowerAtomicArith instead of calling Replac...
Craig Topper [Mon, 13 May 2019 22:17:13 +0000 (22:17 +0000)]
[X86] Use ISD::MERGE_VALUES to return from lowerAtomicArith instead of calling ReplaceAllUsesOfValueWith and returning SDValue().

Returning SDValue() makes the caller think that nothing happened and it will
end up executing the Expand path. This generates extra nodes that will need to
be pruned as dead code.

Returning an ISD::MERGE_VALUES will tell the caller that we'd like to make a
change and it will take care of replacing uses. This will prevent falling into
the Expand path.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@360627 91177308-0d34-0410-b5e6-96231b3b80d8