OSDN Git Service
Alexandre Ganea [Tue, 28 May 2019 14:14:48 +0000 (14:14 +0000)]
[CMake] Default options for faster executables on MSVC
Differential Revision: https://reviews.llvm.org/D55056
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361826
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 28 May 2019 13:54:17 +0000 (13:54 +0000)]
[x86] split 256-bit store of concatenated vectors
This shows up as a side issue to the main problem for the AVX target example from PR37428:
https://bugs.llvm.org/show_bug.cgi?id=37428 - https://godbolt.org/z/7tpRa3
But as we can see in the pile of existing test diffs, it's actually a widespread problem
that affects any AVX or later target. Apart from a couple of oddballs, I think these are
all improvements for the reasons stated in the code comment: we do not want to enable YMM
unnecessarily (avoid vzeroupper and frequency throttling) and some cores split 256-bit
stores anyway.
We could say that MergeConsecutiveStores() is going overboard on some of these examples,
but that won't solve the problem completely. But that is the reason I'm proposing this as
a lowering rather than a combine: we will infinite loop fighting the merge code if we try
this earlier.
Differential Revision: https://reviews.llvm.org/D62498
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361822
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Tue, 28 May 2019 13:46:26 +0000 (13:46 +0000)]
[DAG] LegalizeVectorTypes - reduce scope of local variables. NFCI.
Move the element index/count variables into the block where they are actually used - appeases cppcheck and helps avoid shadow variable warnings.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361821
91177308-0d34-0410-b5e6-
96231b3b80d8
David Stenberg [Tue, 28 May 2019 13:23:25 +0000 (13:23 +0000)]
Stop undef fragments from closing non-overlapping fragments
Summary:
When DwarfDebug::buildLocationList() encountered an undef debug value,
it would truncate all open values, regardless if they were overlapping or
not. This patch fixes so that it only does that for overlapping fragments.
This change unearthed a bug that I had introduced in D57511,
which I have fixed in this patch. The code in DebugHandlerBase that
changes labels for parameter debug values could break DwarfDebug's
assumption that the labels for the entries in the debug value history
are monotonically increasing. Before this patch, that bug could result
in location list entries whose ending address was lower than the
beginning address, and with the changes for undef debug values that this
patch introduces it could trigger an assertion, due to attempting to
emit location list entries with empty ranges. A reproducer for the bug
is added in param-reg-const-mix.mir.
Reviewers: aprantl, jmorse, probinson
Reviewed By: aprantl
Subscribers: javed.absar, llvm-commits
Tags: #debug-info, #llvm
Differential Revision: https://reviews.llvm.org/D62379
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361820
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Tue, 28 May 2019 13:08:31 +0000 (13:08 +0000)]
MIR: Fix printer crashing on dead CSR frame indexes
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361819
91177308-0d34-0410-b5e6-
96231b3b80d8
Sjoerd Meijer [Tue, 28 May 2019 13:04:47 +0000 (13:04 +0000)]
Follow up of r361810: test case fix attempt for Windows builder
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361817
91177308-0d34-0410-b5e6-
96231b3b80d8
Cameron McInally [Tue, 28 May 2019 13:00:52 +0000 (13:00 +0000)]
[IRBuilder] Add CreateUnOp(...) to the IRBuilder to support unary FNeg
Also update UnaryOperator to support isa, cast, and dyn_cast.
Differential Revision: https://reviews.llvm.org/D62417
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361816
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Tue, 28 May 2019 12:58:07 +0000 (12:58 +0000)]
[x86] fix 256-bit vector store splitting to honor 'volatile'
Forking this out of the discussion in D62498
(and assuming that will be committed later, so adding the helper function here).
The LangRef says:
"the backend should never split or merge target-legal volatile load/store instructions."
Differential Revision: https://reviews.llvm.org/D62506
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361815
91177308-0d34-0410-b5e6-
96231b3b80d8
Benjamin Kramer [Tue, 28 May 2019 12:52:57 +0000 (12:52 +0000)]
[X86] Custom lower CONCAT_VECTORS of v2i1
The generic legalizer cannot handle this. Add an assert instead of
silently miscompiling vectors with elements smaller than 8 bits.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361814
91177308-0d34-0410-b5e6-
96231b3b80d8
Graham Hunter [Tue, 28 May 2019 12:36:39 +0000 (12:36 +0000)]
[NFC] Test commit, delete trailing whitespace
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361813
91177308-0d34-0410-b5e6-
96231b3b80d8
Hans Wennborg [Tue, 28 May 2019 12:30:35 +0000 (12:30 +0000)]
Cleanups for r361807 that I somehow failed to commit
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361812
91177308-0d34-0410-b5e6-
96231b3b80d8
Hans Wennborg [Tue, 28 May 2019 12:19:38 +0000 (12:19 +0000)]
Re-commit r357452 (take 2): "SimplifyCFG SinkCommonCodeFromPredecessors: Also sink function calls without used results (PR41259)"
This was reverted in r360086 as it was supected of causing mysterious test
failures internally. However, it was never concluded that this patch was the
root cause.
> The code was previously checking that candidates for sinking had exactly
> one use or were a store instruction (which can't have uses). This meant
> we could sink call instructions only if they had a use.
>
> That limitation seemed a bit arbitrary, so this patch changes it to
> "instruction has zero or one use" which seems more natural and removes
> the need to special-case stores.
>
> Differential revision: https://reviews.llvm.org/D59936
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361811
91177308-0d34-0410-b5e6-
96231b3b80d8
Sjoerd Meijer [Tue, 28 May 2019 12:06:26 +0000 (12:06 +0000)]
[ARM] Use CHECK-NEXT in CodeGen/ARM/O3-pipeline.ll. NFC.
Use CHECK-NEXT, like in other pipeline tests, so that we actually
notice when the pipeline is changed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361810
91177308-0d34-0410-b5e6-
96231b3b80d8
Yevgeny Rouban [Tue, 28 May 2019 11:33:50 +0000 (11:33 +0000)]
[CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst
This patch fixes the CorrelatedValuePropagation pass to keep
prof branch_weights metadata of SwitchInst consistent.
It makes use of SwitchInstProfUpdateWrapper.
New tests are added.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D62126
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361808
91177308-0d34-0410-b5e6-
96231b3b80d8
Hans Wennborg [Tue, 28 May 2019 11:24:20 +0000 (11:24 +0000)]
Fix some llvm-readelf tests after r361633
They were failing on 32-bit Windows. In the cases where I've changed
test expectations, I've checked that they match the output of GNU
readelf.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361807
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Tue, 28 May 2019 11:10:56 +0000 (11:10 +0000)]
[SLPVectorizer][X86] Add broadcast test case from D62427
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361805
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Tue, 28 May 2019 10:53:23 +0000 (10:53 +0000)]
[X86] X86CmovConverterPass::collectCmovCandidates - fix uninitialized variable warnings. NFCI.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361804
91177308-0d34-0410-b5e6-
96231b3b80d8
Cullen Rhodes [Tue, 28 May 2019 09:36:52 +0000 (09:36 +0000)]
[AArch64][SVE2] Asm: support SVE2 Floating Point Convert Group
Summary:
Patch adds support for the following intructions:
SVE2 floating-point convert precision:
* FCVTXNT, FCVTNT, FCVTLT
The specification can be found here:
https://developer.arm.com/docs/ddi0602/latest
Reviewed By: chill
Differential Revision: https://reviews.llvm.org/D62382
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361801
91177308-0d34-0410-b5e6-
96231b3b80d8
Cullen Rhodes [Tue, 28 May 2019 09:13:17 +0000 (09:13 +0000)]
[AArch64][SVE2] Asm: support SVE2 Crypto Extensions Group
Summary:
Patch adds support for the following instructions:
SVE2 crypto constructive binary operations:
* SM4EKEY, RAX1
SVE2 crypto destructive binary operations:
* AESE, AESD, SM4E
SVE2 crypto unary operations:
* AESMC, AESIMC
AESE, AESD, AESMC and AESIMC are enabled with +sve2-aes. SM4E and
SM4EKEY are enabled with +sve2-sm4. RAX1 is enabled with +sve2-sha3.
The specification can be found here:
https://developer.arm.com/docs/ddi0602/latest
Reviewed By: SjoerdMeijer
Differential Revision: https://reviews.llvm.org/D62307
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361797
91177308-0d34-0410-b5e6-
96231b3b80d8
Cullen Rhodes [Tue, 28 May 2019 08:51:59 +0000 (08:51 +0000)]
[AArch64][SVE2] Asm: support SVE2 Histogram Computation Groups
Summary:
Patch adds support for the following instructions:
SVE2 histogram generation (segment):
* HISTSEG
SVE2 histogram generation (vector):
* HISTCNT
The specification can be found here:
https://developer.arm.com/docs/ddi0602/latest
Reviewed By: chill
Differential Revision: https://reviews.llvm.org/D62306
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361796
91177308-0d34-0410-b5e6-
96231b3b80d8
Cullen Rhodes [Tue, 28 May 2019 08:42:22 +0000 (08:42 +0000)]
[AArch64][SVE2] Asm: support SVE2 Misc Group
Summary:
Patch adds support for the following instructions:
SVE2 bitwise exclusive-or interleaved:
* EORBT, EORTB
SVE2 bitwise permute:
* BEXT, BDEP, BGRP
SVE2 bitwise shift left long:
* SSHLLB, SSHLLT, USHLLB, USHLLT
SVE2 integer add/subtract interleaved long:
* SADDLBT, SSUBLBT, SSUBLTB
BDEP, BEXT and BGRP are enabled with SVE2 feature +bitperm, all other
instructions in this group are enabled with +sve2.
Reviewed By: chill
Differential Revision: https://reviews.llvm.org/D62304
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361795
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 28 May 2019 07:25:27 +0000 (07:25 +0000)]
[InlineCost] Fix a couple comments. NFC
Replace "unary operator" with "unary instruction" in visitUnaryInstruction since
we now have a UnaryOperator class which might needs its own visit function.
Fix a copy/paste in visitCastInst that appears to have been copied from
visitPtrToInt.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361794
91177308-0d34-0410-b5e6-
96231b3b80d8
Don Hinton [Tue, 28 May 2019 06:38:16 +0000 (06:38 +0000)]
Revert [test] Fix plugin tests
This reverts r361790 (git commit
fe5eaab2b5b4523886bd63aebcfea8cfce586fa1)
It's causing buildbot breakage, so reverting while I investigate.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361793
91177308-0d34-0410-b5e6-
96231b3b80d8
Don Hinton [Tue, 28 May 2019 06:26:58 +0000 (06:26 +0000)]
[test] Fix plugin tests
Summary:
The following changes were required to fix these tests:
1) Change LLVM_ENABLE_PLUGINS to an option and move it to
llvm/CMakeLists.txt with an appropriate default -- which matches
the original default behavior.
2) Move the plugins directory from clang/test/Analysis
clang/lib/Analysis. It's not enough to add an exclude to the
lit.local.cfg file because add_lit_testsuites recurses the tree and
automatically adds the appropriate `check-` targets, which don't
make sense for the plugins because they aren't tests and don't
have `RUN` statements.
Here's a list of the `clang-check-anlysis*` targets with this
change:
```
$ ninja -t targets all| sed -n "s/.*\/\(check[^:]*\):.*/\1/p" | sort -u | grep clang-analysis
check-clang-analysis
check-clang-analysis-checkers
check-clang-analysis-copypaste
check-clang-analysis-diagnostics
check-clang-analysis-engine
check-clang-analysis-exploration_order
check-clang-analysis-html_diagnostics
check-clang-analysis-html_diagnostics-relevant_lines
check-clang-analysis-inlining
check-clang-analysis-objc
check-clang-analysis-unified-sources
check-clang-analysis-z3
```
3) Simplify the logic and only include the subdirectories under
clang/lib/Analysis/plugins if LLVM_ENABLE_PLUGINS is set.
Reviewed By: NoQ
Tags: #clang, #llvm
Differential Revision: https://reviews.llvm.org/D62445
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361790
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Tue, 28 May 2019 04:09:18 +0000 (04:09 +0000)]
[CostModel] Add really basic support for being able to query the cost of the FNeg instruction.
Summary:
This reuses the getArithmeticInstrCost, but passes dummy values of the second
operand flags.
The X86 costs are wrong and can be improved in a follow up. I just wanted to
stop it from reporting an unknown cost first.
Reviewers: RKSimon, spatel, andrew.w.kaylor, cameron.mcinally
Reviewed By: spatel
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62444
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361788
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Weber [Tue, 28 May 2019 01:20:36 +0000 (01:20 +0000)]
llvm-undname: Remove unreachable statement
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361786
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 27 May 2019 23:56:41 +0000 (23:56 +0000)]
[x86] add test to show volatile store splitting; NFC
From the LangRef:
"the backend should never split or merge target-legal
volatile load/store instructions."
See also:
D62498
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361785
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Weber [Mon, 27 May 2019 23:10:42 +0000 (23:10 +0000)]
llvm-undname: Extract demangleMD5Name() method; no behavior change
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361783
91177308-0d34-0410-b5e6-
96231b3b80d8
Lang Hames [Mon, 27 May 2019 21:34:31 +0000 (21:34 +0000)]
[RuntimeDyld][ARM] Fix an incorrect assertion condition.
Fixes https://llvm.org/PR42036
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361782
91177308-0d34-0410-b5e6-
96231b3b80d8
Matt Arsenault [Mon, 27 May 2019 20:37:31 +0000 (20:37 +0000)]
RegAllocFast: Set MayLiveAcrossBlocks when allocating uses
Setting mayLiveOut based only on use instructions after allocating the
def block did not work if the use block was allocated before the def
block, since the virtual register uses were already removed.
Fixes bug 41973.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361781
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 27 May 2019 20:26:21 +0000 (20:26 +0000)]
[SelectionDAG] fold concat of extract subvectors
This is derived from the related fold for build vectors.
We also have a version of this in DAGCombiner. The benefit of
having this fold at node creation time is (1) efficiency and
(2) preventing infinite looping from creating patterns that
should not exist in the first place.
Currently, the inf-loop could happen with MergeConsecutiveStores()
because it naively creates concat of extracts when forming a wider
vector store. That could fight with target-specific store narrowing.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361780
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Mon, 27 May 2019 18:26:43 +0000 (18:26 +0000)]
[SelectionDAG] fix formatting and redundant comments; NFC
There's a possible missing fold here for extracting from the
same source vector. It's similar to a check that we use to
squash a build vector with all extracted elements from the
same source vector.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361778
91177308-0d34-0410-b5e6-
96231b3b80d8
Michael Liao [Mon, 27 May 2019 18:26:29 +0000 (18:26 +0000)]
[SelectionDAG] Enhance the simplification of `copyto` from `implicit-def`.
Summary:
- The current implementation simplifies the case where the source of
`copyto` is `implicit-def`ed. However, it only works when that
`implicit-def` is single-used since it detects that from
`implicit-def` and cannot determine which destination vreg should be
used if there are multiple uses.
- This patch changes that detection when `copyto` is being emitted. If
that `copyto`'s source is defined from `implicit-def`, it simplifies
it. Hence, it works even that `implicit-def` is multi-used.
- Except it simplifies the internal IR, it won't improve the quality of
code generation. However, it helps to detect 'implicit-def` in a
straight-forward manner in some passes, such as `si-i1-copies`. A test
case is added.
Reviewers: sunfish, nhaehnle
Subscribers: jvesely, hiraditya, asbirlea, llvm-commits, yaxunl
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62342
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361777
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexander Timofeev [Mon, 27 May 2019 18:17:21 +0000 (18:17 +0000)]
[AMDGPU] Fix for the address sanitizer failure. Fixing typo
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361776
91177308-0d34-0410-b5e6-
96231b3b80d8
Jacques Pienaar [Mon, 27 May 2019 17:38:41 +0000 (17:38 +0000)]
NFC: Change usage of 'DenseSet' to 'DenseSetImpl' in DenseSetImpl::ConstIterator.
Summary:
Change usage of 'DenseSet' to 'DenseSetImpl' in a friend declaration within DenseSetImpl::ConstIterator. 'ConstIterator' was never updated when DenseSet was split into an impl when adding support for DenseSetImpl.
This fixes build errors on MSVC when forward declaring DenseSet as this friend decl does not declare the template arguments as well.
Reviewers: jpienaar
Reviewed By: jpienaar
Subscribers: jpienaar, lebedev.ri, dexonsmith, kristina, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D62467
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361775
91177308-0d34-0410-b5e6-
96231b3b80d8
Dmitri Gribenko [Mon, 27 May 2019 17:03:57 +0000 (17:03 +0000)]
Include what you use in AArch64AsmBackend.cpp
AArch64AsmBackend.cpp was not using any APIs from AArch64.h, and was
only including it for transitive dependencies. Doing so is problematic
from include-what-you-use perspective, but it is also a layering issue
(it creates a dependency cycle between the primary AArch64 target
library and the MCTargetDesc library).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361774
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Mon, 27 May 2019 16:39:25 +0000 (16:39 +0000)]
[SelectionDAG] GetDemandedBits - add demanded elements wrapper implementation
The DemandedElts variable is pretty much inert at the moment - the original GetDemandedBits implementation calls it with an 'all ones' DemandedElts value so the function is active and behaves exactly as it used to.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361773
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Mon, 27 May 2019 16:33:15 +0000 (16:33 +0000)]
[LLParser] Fix uninitialized flag variable warnings. NFCI.
Fixes a large number of warnings in the scan-build report on llvm builds.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361772
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexander Timofeev [Mon, 27 May 2019 15:03:29 +0000 (15:03 +0000)]
[AMDGPU] Fix for the address sanitizer failure caused by the ifollowing commit:
1a8b2ea611cf4ca7cb09562e0238cfefa27c05b5 Divergence driven ISel. Assign register class for cross block values according to the divergence.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361770
91177308-0d34-0410-b5e6-
96231b3b80d8
Dmitry Preobrazhensky [Mon, 27 May 2019 14:08:43 +0000 (14:08 +0000)]
[AMDGPU][MC] Enabled constant expressions as operands of s_waitcnt
See bug 40820: https://bugs.llvm.org/show_bug.cgi?id=40820
Reviewers: artem.tamazov, arsenm
Differential Revision: https://reviews.llvm.org/D61017
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361763
91177308-0d34-0410-b5e6-
96231b3b80d8
Xing Xue [Mon, 27 May 2019 13:57:28 +0000 (13:57 +0000)]
[MustExecute] Improve MustExecute to correctly handle loop nest
Summary:
for.outer:
br for.inner
for.inner:
LI <loop invariant load instruction>
for.inner.latch:
br for.inner, for.outer.latch
for.outer.latch:
br for.outer, for.outer.exit
LI is a loop invariant load instruction that post dominate for.outer, so LI should be able to move out of the loop nest. However, there is a bug in allLoopPathsLeadToBlock().
Current algorithm of allLoopPathsLeadToBlock()
1. get all the transitive predecessors of the basic block LI belongs to (for.inner) ==> for.outer, for.inner.latch
2. if any successors of any of the predecessors are not for.inner or for.inner's predecessors, then return false
3. return true
Although for.inner.latch is for.inner's predecessor, but for.inner dominates for.inner.latch, which means if for.inner.latch is ever executed, for.inner should be as well. It should not return false for cases like this.
Author: Whitney (committed by xingxue)
Reviewers: kbarton, jdoerfert, Meinersbur, hfinkel, fhahn
Reviewed By: jdoerfert
Subscribers: hiraditya, jsji, llvm-commits, etiotto, bmahjour
Tags: #LLVM
Differential Revision: https://reviews.llvm.org/D62418
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361762
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikola Prica [Mon, 27 May 2019 13:51:30 +0000 (13:51 +0000)]
Test commit (NFC)
Add blank line.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361761
91177308-0d34-0410-b5e6-
96231b3b80d8
Diana Picus [Mon, 27 May 2019 10:32:34 +0000 (10:32 +0000)]
[ARM GlobalISel] Un-XFAIL some tests. NFC
It turns out we support big endian now (probably since r332449, but I
haven't bisected to confirm).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361756
91177308-0d34-0410-b5e6-
96231b3b80d8
Diana Picus [Mon, 27 May 2019 10:30:33 +0000 (10:30 +0000)]
[ARM GlobalISel] Cleanup CallLowering a bit
We never actually use the Offsets produced by ComputeValueVTs, so remove
them until we need them.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361755
91177308-0d34-0410-b5e6-
96231b3b80d8
Hans Wennborg [Mon, 27 May 2019 09:03:00 +0000 (09:03 +0000)]
Cmake: allow using LLVM_EXTERNAL_PROJECTS with LLVM_ENABLE_PROJECTS
The current code iterates over the combination of LLVM_EXTERNAL_PROJECTS
and LLVM_ENABLE_PROJECTS, but then disables projects that are only in
the former. If a project is in LLVM_EXTERNAL_PROJECTS, it should be
enabled.
See also llvm-commits thread on r354060.
Differential revision: https://reviews.llvm.org/D62289
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361751
91177308-0d34-0410-b5e6-
96231b3b80d8
Serge Guelton [Mon, 27 May 2019 08:24:06 +0000 (08:24 +0000)]
Make llvm-as --help great again
This is a follow-up to https://reviews.llvm.org/D60411, but for llvm-as.
New output:
OVERVIEW: llvm .ll -> .bc assembler
USAGE: llvm-as [options] <input .llvm file>
OPTIONS:
Generic Options:
-help - Display available options (-help-hidden for more)
-help-list - Display list of available options (-help-list-hidden for more)
-version - Display the version of this program
llvm-as Options:
-data-layout=<layout-string> - data layout string to use
-disable-output - Disable output
-f - Enable binary output on terminals
-module-hash - Emit module hash
-o=<filename> - Override output filename
Differential Revision: https://reviews.llvm.org/D60603
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361750
91177308-0d34-0410-b5e6-
96231b3b80d8
Djordje Todorovic [Mon, 27 May 2019 07:48:28 +0000 (07:48 +0000)]
[test commit] Add my name to the CREDITS.TXT
This is my test commit. (NFC)
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361748
91177308-0d34-0410-b5e6-
96231b3b80d8
David L. Jones [Mon, 27 May 2019 06:00:00 +0000 (06:00 +0000)]
Revert r361356: "[MIR] Add simple PRE pass to MachineCSE"
This is problematic on buildbots, as discussed here: https://reviews.llvm.org/rL361356
It seems like the plan already was to revert, but that hasn't happened yet.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361746
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Mon, 27 May 2019 05:27:57 +0000 (05:27 +0000)]
[X86] Add test cases for D62444. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361745
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Weber [Mon, 27 May 2019 00:48:59 +0000 (00:48 +0000)]
llvm-undname: Make demangling of MD5 names more robust
Demangler::parse() for MD5 names would:
1. Put all remaining text into the MD5 name sight unseen
2. Not modify MangledName
This meant that if the demangler recursively called parse() (e.g. in
demangleLocallyScopedNamePiece()), every recursive call that started on
an MD5 name would add all remaining bytes to the output buffer but
only advance the input by a byte. For valid inputs, MD5 types are
never (well, see comments for 2 exceptions) nested, but for invalid
input this could cause memory use quadratic in the input size.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361744
91177308-0d34-0410-b5e6-
96231b3b80d8
Florian Hahn [Sun, 26 May 2019 23:38:25 +0000 (23:38 +0000)]
[LoopInterchange] Fix handling of LCSSA nodes defined in headers and latches.
The code to preserve LCSSA PHIs currently only properly supports
reduction PHIs and PHIs for values defined outside the latches.
This patch improves the LCSSA PHI handling to cover PHIs for values
defined in the latches.
Fixes PR41725.
Reviewers: efriedma, mcrosier, davide, jdoerfert
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D61576
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361743
91177308-0d34-0410-b5e6-
96231b3b80d8
Yonghong Song [Sun, 26 May 2019 21:26:06 +0000 (21:26 +0000)]
[BPF] generate R_BPF_NONE relocation for BTF DataSec variables
The variables in BTF DataSec type encode in-section offset.
R_BPF_NONE should be generated instead of R_BPF_64_32.
Signed-off-by: Yonghong Song <yhs@fb.com>
Differential Revision: https://reviews.llvm.org/D62460
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361742
91177308-0d34-0410-b5e6-
96231b3b80d8
Alexander Timofeev [Sun, 26 May 2019 20:33:26 +0000 (20:33 +0000)]
[AMDGPU] Divergence driven ISel. Assign register class for cross block values according to the divergence.
Details: To make instruction selection really divergence driven it is necessary to assign
the correct register classes to the cross block values beforehand. For the divergent targets
same value type requires different register classes dependent on the value divergence.
Reviewers: rampitec, nhaehnle
Differential Revision: https://reviews.llvm.org/D59990
This commit was reverted because of the build failure.
The reason was mlformed patch.
Build failure fixed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361741
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrea Di Biagio [Sun, 26 May 2019 19:50:31 +0000 (19:50 +0000)]
[MCA][Scheduler] Improved critical memory dependency computation.
This fixes a problem where back-pressure increases caused by register
dependencies were not correctly notified if execution was also delayed by memory
dependencies.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361740
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sun, 26 May 2019 18:58:14 +0000 (18:58 +0000)]
[SelectionDAG] GetDemandedBits - cleanup to more closely match SimplifyDemandedBits. NFCI.
Prep work before adding demanded elts support.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361739
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sun, 26 May 2019 18:43:44 +0000 (18:43 +0000)]
[SelectionDAG] MaskedValueIsZero - add demanded elements implementation
Will be used in an upcoming patch but I've updated the original implementation to call this to ensure test coverage.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361738
91177308-0d34-0410-b5e6-
96231b3b80d8
Andrea Di Biagio [Sun, 26 May 2019 18:41:35 +0000 (18:41 +0000)]
[MCA] Refactor the logic that computes the critical memory dependency info. NFCI
CriticalRegDep has been renamed CriticalDependency, and it is now used by class
Instruction to store information about the critical register dependency and the
critical memory dependency. No functional change intendend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361737
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 18:15:51 +0000 (18:15 +0000)]
[SimplifyCFG] back out all SwitchInst commits
They caused the sanitizer builds to fail.
My suspicion is the change the countLeadingZeros().
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361736
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sun, 26 May 2019 16:00:35 +0000 (16:00 +0000)]
[X86][SSE] Add shuffle combining support for ISD::ANY_EXTEND_VECTOR_INREG
Reuses what we already have in place for ISD::ZERO_EXTEND_VECTOR_INREG just with a different sentinel
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361734
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 15:29:10 +0000 (15:29 +0000)]
[SimplifyCFG] NFC, one more fixed test from previous push.
The old test was checking for a stupid subtract one that is a transform that
makes the code woorse.
The constant-islands-jump-table.ll test wants the code a specific way,
that makes sense, so I will submit code to fix that one.
Sorry that I really didn't know how to run the test suite before this.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361733
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sun, 26 May 2019 15:08:45 +0000 (15:08 +0000)]
Revert rL361731 : [LLParser] Fix uninitialized variable warnings. NFCI.
These 3 variables cause quite a few warnings in the scan-build report on llvm.
........
Revert accidental commit.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361732
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sun, 26 May 2019 15:05:12 +0000 (15:05 +0000)]
[LLParser] Fix uninitialized variable warnings. NFCI.
These 3 variables cause quite a few warnings in the scan-build report on llvm.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361731
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 14:44:14 +0000 (14:44 +0000)]
[SimplifyCFG] NFC, fix failing tests from last patches.
No problems with the transforms.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361730
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Sun, 26 May 2019 14:03:50 +0000 (14:03 +0000)]
[InstCombine] prevent crashing with invalid extractelement index
This was found/reduced from a fuzzer report:
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=14956
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361729
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 13:55:52 +0000 (13:55 +0000)]
[SimplifyCFG] ReduceSwitchRange: Improve on the case where the SubThreshold doesn't trigger
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361728
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 13:55:14 +0000 (13:55 +0000)]
[SimplifyCFG] Run ReduceSwitchRange unconditionally, generalize
Rather than gating on "isSwitchDense" (resulting in necessesarily
sparse lookup tables even when they were generated), always run
this quite cheap transform.
This transform is useful not just for generating tables.
LowerSwitch also wants this: read LowerSwitch.cpp:257.
Be careful to not generate worse code, by introducing a
SubThreshold heuristic.
Instead of just sorting by signed, generalize the finding of the
best base.
And now that it is run unconditionally, do not replicate its
functionality in SwitchToLookupTable (which could use a Sub
when having a hole is smaller, hence the SubThreshold
heuristic located in a single place).
This simplifies SwitchToLookupTable, and fixes
some ugly corner cases due to the use of signed numbers,
such as a table containing i16 32768 and 32769, of which
32769 would be interpreted as -32768, and now the code thinks
the table is size 65536.
(We still use unconditional subtraction when building a single-register mask,
but I think this whole block should go when the more general sparse
map is added, which doesn't leave empty holes in the table.)
And the reason test4 and test5 did not trigger was documented wrong:
it was because they were not considered sufficiently "dense".
Also, fix generation of invalid LLVM-IR: shl by bit-width.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361727
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 13:54:04 +0000 (13:54 +0000)]
[SimpligyCFG] NFC, remove GCD that was only used for powers of two
and replace with an equilivent countTrailingZeros.
GCD is much more expensive than this, with repeated division.
This depends on D60823
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361726
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 13:52:41 +0000 (13:52 +0000)]
[SimplifyCFG] NFC, update Switch tests to HEAD so I can see if my changes change anything
Also add baseline tests to show effect of later patches.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361725
91177308-0d34-0410-b5e6-
96231b3b80d8
Shawn Landden [Sun, 26 May 2019 13:49:58 +0000 (13:49 +0000)]
[Support] make countLeadingZeros() and countTrailingZeros() return unsigned
This matches countLeadingOnes() and countTrailingOnes(), and
APInt's countLeadingZeros() and countTrailingZeros().
(as well as __builtin_clzll())
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361724
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sun, 26 May 2019 13:22:01 +0000 (13:22 +0000)]
[ValueTracking] Base computeOverflowForUnsignedMul() on ConstantRange code; NFCI
The implementation in ValueTracking and ConstantRange are equally
powerful, reuse the one in ConstantRange, which will make this easier
to extend.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361723
91177308-0d34-0410-b5e6-
96231b3b80d8
Nico Weber [Sun, 26 May 2019 13:06:48 +0000 (13:06 +0000)]
gn build: Merge r361664
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361722
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sun, 26 May 2019 11:43:37 +0000 (11:43 +0000)]
[InstCombine] Refactor OptimizeOverflowCheck; NFCI
Extract method to compute overflow based on binop and signedness,
and then make the result handling code generic. This extends the
always-overflow handling to signed muls, but has currently no effect,
as we don't compute always overflow for them (thus NFC).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361721
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sun, 26 May 2019 11:43:31 +0000 (11:43 +0000)]
[InstCombine] Remove OverflowCheckFlavor; NFC
Instead pass binary op and signedness. The extra enum only makes
things more complicated in this case.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361720
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Sun, 26 May 2019 11:34:30 +0000 (11:34 +0000)]
[ARM] Select fp16 fma
This adds a pattern for fma, similar to the float and double patterns.
Differential Revision: https://reviews.llvm.org/D62330
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361719
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Sun, 26 May 2019 11:13:00 +0000 (11:13 +0000)]
[ARM] Select a number of fp16 rounding functions
This add patterns for fp16 round and ceil etc. Same as the float and double
patterns.
Differential Revision: https://reviews.llvm.org/D62326
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361718
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Sun, 26 May 2019 10:59:21 +0000 (10:59 +0000)]
[ARM] Promote various fp16 math intrinsics
Promote a number of fp16 math intrinsics to float, so that the relevant float
math routines can be used. Copysign is expanded so as to be handled in-place.
Differential Revision: https://reviews.llvm.org/D62325
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361717
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sun, 26 May 2019 10:54:23 +0000 (10:54 +0000)]
[X86][AVX] combineBitcastvxi1 - peek through bitops to determine size of original vector
We were only testing for direct SETCC results - this allows us to peek through AND/OR/XOR combinations of the comparison results as well.
There's a missing SEXT(PACKSS) fold that I need to investigate for v8i1 cases before I can enable it there as well.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361716
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Sun, 26 May 2019 10:51:58 +0000 (10:51 +0000)]
[ARM] Select fp16 fabs
This adds a pattern for the fabs intrinsic, the same as float and double.
Differential Revision: https://reviews.llvm.org/D62324
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361715
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Sun, 26 May 2019 10:42:24 +0000 (10:42 +0000)]
[ARM] Select fp16 fsqrt
This adds a pattern for the sqrt intrinsic, the same as float and double.
Differential Revision: https://reviews.llvm.org/D62322
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361714
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Sun, 26 May 2019 10:30:22 +0000 (10:30 +0000)]
[ARM] Promote fp16 frem
Promote fp16 frem operations on ARM to floats so they call fmodf.
Differential Revision: https://reviews.llvm.org/D62321
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361713
91177308-0d34-0410-b5e6-
96231b3b80d8
David Green [Sun, 26 May 2019 10:06:40 +0000 (10:06 +0000)]
[ARM] Add some base fullfp16 tests. NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361712
91177308-0d34-0410-b5e6-
96231b3b80d8
Fangrui Song [Sun, 26 May 2019 08:31:00 +0000 (08:31 +0000)]
[PowerPC] Add missing R_PPC_* relocation types
While people mostly care about 64-bit, some systems need basic lib32
support. The plan is to make lld (see PR40888) capable of linking some
applications (PR40888).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361711
91177308-0d34-0410-b5e6-
96231b3b80d8
David Bolvansky [Sat, 25 May 2019 22:34:27 +0000 (22:34 +0000)]
[SimplifyCFG] Added condition assumption for unreachable blocks
Summary: PR41688
Reviewers: spatel, efriedma, craig.topper, hfinkel, reames
Reviewed By: hfinkel
Subscribers: javed.absar, dmgreen, fhahn, hfinkel, reames, nikic, lebedev.ri, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D61409
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361707
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sat, 25 May 2019 18:02:17 +0000 (18:02 +0000)]
[X86] lowerBuildVectorToBitOp - support build_vector(shift()) -> shift(build_vector(),C)
Commonly occurs in sign-extension cases
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361706
91177308-0d34-0410-b5e6-
96231b3b80d8
Robert Widmann [Sat, 25 May 2019 16:47:27 +0000 (16:47 +0000)]
[LLVM-C] Add Accessor for Mach-O Universal Binary Slices
Summary: Allow for retrieving an object file corresponding to an architecture-specific slice in a Mach-O universal binary file.
Reviewers: whitequark, deadalnix
Reviewed By: whitequark
Subscribers: hiraditya, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D60378
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361705
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 16:44:29 +0000 (16:44 +0000)]
[X86] Combine fminnum/fmaxnum with non-nan operand to fmin/fmax
If we have a known non-nan operand, place it in the second operand
of fmin/fmax that is returned if either operand is nan.
Differential Revision: https://reviews.llvm.org/D62448
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361704
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 16:44:14 +0000 (16:44 +0000)]
[LVI][CVP] Add support for saturating add/sub
Adds support for the uadd.sat family of intrinsics in LVI, based on
ConstantRange methods from D60946.
Differential Revision: https://reviews.llvm.org/D62447
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361703
91177308-0d34-0410-b5e6-
96231b3b80d8
Simon Pilgrim [Sat, 25 May 2019 16:33:17 +0000 (16:33 +0000)]
[X86][SSE] vector-sext - cleanup prefix lists
Add X32-SSE common prefix to merge some checks
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361702
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Sat, 25 May 2019 15:28:55 +0000 (15:28 +0000)]
[SelectionDAG] define binops as a superset of commutative binops
The test diffs show improved vector narrowing for integer min/max opcodes because
those were all absent from the list. I'm not sure if we can expose functional diffs
for all of the moved/added opcodes though.
It seems like we are missing an AVX512 opportunity to use 256-bit ops in place of
512-bit ops on some tests/targets, but I think that can be a follow-up.
Preliminary steps to make sure the callers are not misusing these queries:
rL361268
rL361547
Differential Revision: https://reviews.llvm.org/D62191
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361701
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 15:06:54 +0000 (15:06 +0000)]
[X86] Add tests for min/maxnum with const operand; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361700
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 14:33:30 +0000 (14:33 +0000)]
[LoopVectorize] Fix test by regenerating checks
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361699
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 14:11:55 +0000 (14:11 +0000)]
[CVP] Remove unnecessary checks for empty GNWR; NFC
The guaranteed no-wrap region is never empty, it always contains at
least zero, so these optimizations don't ever apply.
To make this more obviously true, replace the conversative return
in makeGNWR with an assertion.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361698
91177308-0d34-0410-b5e6-
96231b3b80d8
David Bolvansky [Sat, 25 May 2019 14:10:20 +0000 (14:10 +0000)]
[NFC] Make tests more robust for new optimizations
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361697
91177308-0d34-0410-b5e6-
96231b3b80d8
Sanjay Patel [Sat, 25 May 2019 13:48:07 +0000 (13:48 +0000)]
[SelectionDAG] soften assertion when legalizing narrow vector FP ops
The test based on PR42010:
https://bugs.llvm.org/show_bug.cgi?id=42010
...may show an inaccuracy for PPC's target defs, but we should not
be so aggressive with an assert here. There's no telling what out-of-tree
targets look like.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361696
91177308-0d34-0410-b5e6-
96231b3b80d8
David Bolvansky [Sat, 25 May 2019 13:11:22 +0000 (13:11 +0000)]
[NFC] Update test checks
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361695
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 09:53:51 +0000 (09:53 +0000)]
[CVP] Add tests for saturating add/sub ranges; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361694
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 09:53:45 +0000 (09:53 +0000)]
[LVI][CVP] Calculate with.overflow result range
In LVI, calculate the range of extractvalue(op.with.overflow(%x, %y), 0)
as the range of op(%x, %y). This is mainly useful in conjunction with
D60650: If the result of the operation is extracted in a branch guarded
against overflow, then the value of %x will be appropriately constrained
and the result range of the operation will be calculated taking that
into account.
Differential Revision: https://reviews.llvm.org/D60656
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361693
91177308-0d34-0410-b5e6-
96231b3b80d8
Nikita Popov [Sat, 25 May 2019 09:53:37 +0000 (09:53 +0000)]
[LVI] Extract helper for binary range calculations; NFC
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361692
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sat, 25 May 2019 06:17:47 +0000 (06:17 +0000)]
[X86FixupLEAs] Turn optIncDec into a generic two address LEA optimizer. Support LEA64_32r properly.
INC/DEC is really a special case of a more generic issue. We should also turn leas into add reg/reg or add reg/imm regardless of the slow lea flags.
This also supports LEA64_32 which has 64 bit input registers and 32 bit output registers. So we need to convert the 64 bit inputs to their 32 bit equivalents to check if they are equal to base reg.
One thing to note, the original code preserved the kill flags by adding operands to the new instruction instead of using addReg. But I think tied operands aren't supposed to have the kill flag set. I dropped the kill flags, but I could probably try to preserve it in the add reg/reg case if we think its important. Not sure which operand its supposed to go on for the LEA64_32r instruction due to the super reg implicit uses. Though I'm also not sure those are needed since they were probably just created by an INSERT_SUBREG from a 32-bit input.
Differential Revision: https://reviews.llvm.org/D61472
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361691
91177308-0d34-0410-b5e6-
96231b3b80d8
Craig Topper [Sat, 25 May 2019 04:47:49 +0000 (04:47 +0000)]
[X86] Add zero idioms to the haswell, broadwell, and skylake schedule models. Add 256-bit fp xor to sandybridge zero idioms
This copies the Sandy Bridge zero idiom support to later CPUs. Adding the AVX2 and AVX512F/VL instructions as appropriate.
Differential Revision: https://reviews.llvm.org/D62360
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@361690
91177308-0d34-0410-b5e6-
96231b3b80d8