Peter Collingbourne [Wed, 14 Dec 2016 22:19:22 +0000 (22:19 +0000)]
LibDriver: Reject inputs that are not COFF objects or bitcode files.
Fixes PR31372.
Differential Revision: https://reviews.llvm.org/D27776
llvm-svn: 289726
Dehao Chen [Wed, 14 Dec 2016 22:06:49 +0000 (22:06 +0000)]
Only sets profile summary when it was not preset.
Summary: SampleProfileLoader pass may be invoked twice by LTO. The 2nd pass should not append more summary info as it is already preset by the 1st pass.
Reviewers: eraman, davidxl
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D27733
llvm-svn: 289725
Dehao Chen [Wed, 14 Dec 2016 22:03:08 +0000 (22:03 +0000)]
Fix the bug in r289714 (NFC).
llvm-svn: 289724
Jan Sjodin [Wed, 14 Dec 2016 21:58:42 +0000 (21:58 +0000)]
Revert revision 289721.
llvm-svn: 289723
Davide Italiano [Wed, 14 Dec 2016 21:58:26 +0000 (21:58 +0000)]
[LTO] Catch up with llvm. Modules without DL are now considered invalid.
llvm-svn: 289722
Jan Sjodin [Wed, 14 Dec 2016 21:57:18 +0000 (21:57 +0000)]
Dummy commit.
llvm-svn: 289721
Davide Italiano [Wed, 14 Dec 2016 21:57:14 +0000 (21:57 +0000)]
[LTO] Add the missing datalayout in a test.
llvm-svn: 289720
Davide Italiano [Wed, 14 Dec 2016 21:57:04 +0000 (21:57 +0000)]
[LTO] Reject modules without datalayout.
Also, udpate the ~60 failing tests in the tree which did
not contain a valid datalayout.
This fixes PR31123. lld will be updated in a following patch,
immediately after this is committed.
Differential Revision: https://reviews.llvm.org/D27082
llvm-svn: 289719
Filipe Cabecinhas [Wed, 14 Dec 2016 21:57:04 +0000 (21:57 +0000)]
[asan] Don't skip instrumentation of masked load/store unless we've seen a full load/store on that pointer.
Reviewers: kcc, RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D27625
llvm-svn: 289718
Filipe Cabecinhas [Wed, 14 Dec 2016 21:56:59 +0000 (21:56 +0000)]
[asan] Hook ClInstrumentWrites and ClInstrumentReads to masked operation instrumentation.
Reviewers: kcc
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D27548
llvm-svn: 289717
Eric Fiselier [Wed, 14 Dec 2016 21:44:08 +0000 (21:44 +0000)]
XFAIL test on apple-clang-7.0
llvm-svn: 289716
Dehao Chen [Wed, 14 Dec 2016 21:41:04 +0000 (21:41 +0000)]
Create SampleProfileLoader pass in llvm instead of clang
Summary:
We used to create SampleProfileLoader pass in clang. This makes LTO/ThinLTO unable to add this pass in the linker plugin. This patch moves the SampleProfileLoader pass creation from
clang to llvm pass manager builder.
Reviewers: tejohnson, davidxl, dnovillo
Subscribers: mehdi_amini, cfe-commits
Differential Revision: https://reviews.llvm.org/D27744
llvm-svn: 289715
Dehao Chen [Wed, 14 Dec 2016 21:40:47 +0000 (21:40 +0000)]
Create SampleProfileLoader pass in llvm instead of clang
Summary: We used to create SampleProfileLoader pass in clang. This makes LTO/ThinLTO unable to add this pass in the linker plugin. This patch moves the SampleProfileLoader pass creation from clang to llvm pass manager builder.
Reviewers: tejohnson, davidxl, dnovillo
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D27743
llvm-svn: 289714
Nico Weber [Wed, 14 Dec 2016 21:38:18 +0000 (21:38 +0000)]
Revert 289252 (and follow-up 289285), it caused PR31374
llvm-svn: 289713
Nico Weber [Wed, 14 Dec 2016 21:34:19 +0000 (21:34 +0000)]
Update MSVC compat docs about debug info
https://reviews.llvm.org/D27769
llvm-svn: 289712
Sean Callanan [Wed, 14 Dec 2016 21:31:31 +0000 (21:31 +0000)]
Adopt PrettyStackTrace in LLDB
LLDB needs some minor changes to adopt PrettyStackTrace after https://reviews.llvm.org/D27683.
We remove our own SetCrashDescription() function and use LLVM-provided RAII objects instead.
We also make sure LLDB doesn't define __crashtracer_info__ which would collide with LLVM's definition.
Differential Revision: https://reviews.llvm.org/D27735
llvm-svn: 289711
Eric Fiselier [Wed, 14 Dec 2016 21:29:29 +0000 (21:29 +0000)]
[libcxx] Fix PR24075, PR23841 - Add scoped_allocator_adaptor::construct(pair<T, U>*, ...) overloads.
Summary:
For more information see:
* https://llvm.org/bugs/show_bug.cgi?id=23841
* https://llvm.org/bugs/show_bug.cgi?id=24075
I hope you have as much fun reviewing as I did writing these insane tests!
Reviewers: mclow.lists, AlisdairM, EricWF
Subscribers: AlisdairM, Potatoswatter, cfe-commits
Differential Revision: https://reviews.llvm.org/D27612
llvm-svn: 289710
Eric Fiselier [Wed, 14 Dec 2016 21:22:48 +0000 (21:22 +0000)]
Recommit r286884: P0503R0, adopted in Issaquah, rewords some requirements on nullptr_t and istream_iterator.
No code changes were needed, but I updated a few tests.
Also resolved P0509 and P0521, which required no changes to the library or tests.
This patch was reverted due to llvm.org/PR31016. There is a bug in Clang 3.7
which causes default.pass.cpp to fails. That test is now marked as XFAIL for that
clang version.
This patch was originally authored by Marshall Clow.
llvm-svn: 289708
Michal Gorny [Wed, 14 Dec 2016 20:53:57 +0000 (20:53 +0000)]
Remove unnecessary llvm/Config/config.h includes
Remove the includes of <llvm/Config/config.h> private LLVM header.
The relevant files seem not to use any definitions from that file,
and it is not available when building against installed LLVM.
The use in lib/ReaderWriter/MachO/MachOLinkingContext.cpp originates
from rL218718, and the use in ELF/Strings.cpp from rL274804 (where it
was moved from Symbols.cpp). In both cases, they were added as a part of
demangling support, and they provided HAVE_CXXABI_H.
Since we are now using the LLVM demangler library instead, the code was
removed and the includes and no longer necessary.
Differential Revision: https://reviews.llvm.org/D27757
llvm-svn: 289707
Eli Friedman [Wed, 14 Dec 2016 20:44:38 +0000 (20:44 +0000)]
[ARM] Split 128-bit vectors in BUILD_VECTOR lowering
Given that INSERT_VECTOR_ELT operates on D registers anyway, combining
64-bit vectors into a 128-bit vector is basically free. Therefore, try
to split BUILD_VECTOR nodes before giving up and lowering them to a series
of INSERT_VECTOR_ELT instructions. Sometimes this allows dramatically
better lowerings; see testcases for examples. Inspired by similar code
in the x86 backend for AVX.
Differential Revision: https://reviews.llvm.org/D27624
llvm-svn: 289706
Nico Weber [Wed, 14 Dec 2016 20:33:54 +0000 (20:33 +0000)]
fix gcc warning about a superfluous ;
llvm-svn: 289705
Robert Lougher [Wed, 14 Dec 2016 20:27:22 +0000 (20:27 +0000)]
[InstCombine] Folding of a compare with RHS const should merge debug locations
If all the operands to a phi node are compares that have a RHS constant,
instcombine will try to pull them through the phi node, combining them into
a single operation. When it does this, the debug location of the new op
should be the merged debug locations of the phi node arguments.
Patch 8 of 8 for D26256. Folding of a compare that has a RHS constant.
Differential Revision: https://reviews.llvm.org/D26256
llvm-svn: 289704
Eli Friedman [Wed, 14 Dec 2016 20:25:26 +0000 (20:25 +0000)]
[ARM] Add ARMISD::VLD1DUP to match vld1_dup more consistently.
Currently, there are substantial problems forming vld1_dup even if the
VDUP survives legalization. The lack of an actual node
leads to terrible results: not only can we not form post-increment vld1_dup
instructions, but we form scalar pre-increment and post-increment
loads which force the loaded value into a GPR. This patch fixes that
by combining the vdup+load into an ARMISD node before DAGCombine
messes it up.
Also includes a crash fix for vld2_dup (see testcase @vld2dupi8_postinc_variable).
Differential Revision: https://reviews.llvm.org/D27694
llvm-svn: 289703
Amjad Aboud [Wed, 14 Dec 2016 20:24:54 +0000 (20:24 +0000)]
[DebugInfo] Changed DIBuilder::createCompileUnit() to take DIFile instead of FileName and Directory.
This way it will be easier to expand DIFile (e.g., to contain checksum) without the need to modify the createCompileUnit() API.
Reviewers: llvm-commits, rnk
Differential Revision: https://reviews.llvm.org/D27762
llvm-svn: 289702
Amjad Aboud [Wed, 14 Dec 2016 20:24:40 +0000 (20:24 +0000)]
[DebugInfo] Changed DIBuilder::createCompileUnit() to take DIFile instead of FileName and Directory.
This way it will be easier to expand DIFile (e.g., to contain checksum) without the need to modify the createCompileUnit() API.
Reviewers: cfe-commits, rnk
Differential Revision: https://reviews.llvm.org/D27763
llvm-svn: 289701
Yaxun Liu [Wed, 14 Dec 2016 20:17:47 +0000 (20:17 +0000)]
Fix build failure due to r289674 on certain systems
Removed a useless include which caused conflict.
llvm-svn: 289700
Robert Lougher [Wed, 14 Dec 2016 20:07:49 +0000 (20:07 +0000)]
[InstCombine] Folding of a binop with RHS const should merge the debug locations
If all the operands to a phi node are a binop with a RHS constant, instcombine
will try to pull them through the phi node, combining them into a single
operation. When it does this, the debug location of the new op should be the
merged debug locations of the phi node arguments.
Patch 7 of 8 for D26256. Folding of a binop with RHS constant.
Differential Revision: https://reviews.llvm.org/D26256
llvm-svn: 289699
Richard Smith [Wed, 14 Dec 2016 19:45:03 +0000 (19:45 +0000)]
Remove unused variable found by GCC warning.
llvm-svn: 289698
David Blaikie [Wed, 14 Dec 2016 19:38:39 +0000 (19:38 +0000)]
DebugInfo: Improve type safety and simplify some subprogram finalization code
This probably ended up this way aften the subprogram<>function link
inversion and debug info metadata schema changes.
llvm-svn: 289697
Geoff Berry [Wed, 14 Dec 2016 19:38:22 +0000 (19:38 +0000)]
[GVNHoist] Move GVNHoist to function simplification part of pipeline.
Summary:
Move GVNHoist to later in the optimization pipeline, specifically, to
the function simplification part of the pipeline. The new pipeline
location allows GVNHoist to run on a function after its callees have
been inlined but before the function has been considered for inlining
into its callers, exposing more opportunities for hoisting.
Performance results on AArch64 kryo:
Improvements:
Benchmarks/CoyoteBench/fftbench -24.952%
spec2006/bzip2 -4.071%
internal bmark -3.177%
Benchmarks/PAQ8p/paq8p -1.754%
spec2000/perlbmk -1.328%
spec2006/h264ref -1.140%
Regressions:
internal bmark +1.818%
Benchmarks/mafft/pairlocalalign +1.084%
Reviewers: sebpop, dberlin, hiraditya
Subscribers: aemerson, mehdi_amini, mcrosier, llvm-commits
Differential Revision: https://reviews.llvm.org/D27722
llvm-svn: 289696
Jim Ingham [Wed, 14 Dec 2016 19:35:56 +0000 (19:35 +0000)]
Test num locations >= 1 not == 1.
llvm-svn: 289695
Andrew Kaylor [Wed, 14 Dec 2016 19:30:18 +0000 (19:30 +0000)]
[WinEH] Avoid holding references to BlockColor (DenseMap) entries while inserting new elements
Differential Revision: https://reviews.llvm.org/D27693
llvm-svn: 289694
Robert Lougher [Wed, 14 Dec 2016 19:24:01 +0000 (19:24 +0000)]
[InstCombine] When folding casts through a phi node merge the debug locations
If all the operands to a phi node are a cast, instcombine will try to pull
them through the phi node, combining them into a single cast. When it does
this, the debug location of the new cast should be the merged debug locations
of the phi node arguments.
Patch 6 of 8 for D26256. Folding of a cast operation.
Differential Revision: https://reviews.llvm.org/D26256
llvm-svn: 289693
Tim Northover [Wed, 14 Dec 2016 19:21:30 +0000 (19:21 +0000)]
AArch64: add architecture version feature to Clang invocation.
Otherwise we don't get the correct predefines and so on in the front-end (or
the right features in the backend).
llvm-svn: 289692
Sean Callanan [Wed, 14 Dec 2016 19:19:53 +0000 (19:19 +0000)]
Include <cstdarg> in PrettyStackTrace.cpp, fixing the bots.
llvm-svn: 289691
Kostya Serebryany [Wed, 14 Dec 2016 19:10:17 +0000 (19:10 +0000)]
[sanitizer] intercept bstring functions, patch by Kuang-che Wu (https://reviews.llvm.org/D27659)
llvm-svn: 289690
Sean Callanan [Wed, 14 Dec 2016 19:09:43 +0000 (19:09 +0000)]
Prepare PrettyStackTrace for LLDB adoption
This patch fixes the linkage for __crashtracer_info__, making it have the proper mangling (extern "C") and linkage (private extern).
It also adds a new PrettyStackTrace type, allowing LLDB to adopt this instead of Host::SetCrashDescriptionWithFormat().
Without this patch, CrashTracer on macOS won't pick up pretty stack traces from any LLVM client.
An LLDB commit adopting this API will follow shortly.
Differential Revision: https://reviews.llvm.org/D27683
llvm-svn: 289689
Robert Lougher [Wed, 14 Dec 2016 19:02:14 +0000 (19:02 +0000)]
[InstCombine] Folding loads through a phi node should merge the debug locations
If all the operands to a phi node are a load, instcombine will try to pull
them through the phi node, combining them into a single load. When it does
this, the debug location of the new load should be the merged debug locations
of the phi node arguments.
Patch 5 of 8 for D26256. Folding of a load operation.
Differential Revision: https://reviews.llvm.org/D26256
llvm-svn: 289688
David Gross [Wed, 14 Dec 2016 18:52:33 +0000 (18:52 +0000)]
[DebugInfo] Restore test case for long double constants.
Summary:
D27549 (partial fix for PR26619) emits a constant value in the debug
metadata for a floating-point static const that does not exceed 64
bits in size. Whether or not a long double exceeds 64 bits in size
depends on the target. Modify the test case so that it expects a
constant value for long double if and only if the long double is no
larger than 64 bits.
Reviewers: cfe-commits, probinson
Differential Revision: https://reviews.llvm.org/D27597
llvm-svn: 289686
Devin Coughlin [Wed, 14 Dec 2016 18:46:01 +0000 (18:46 +0000)]
[Driver] Add tests for enabled static analyzer checkers.
The driver passes flags to cc1 that enable various checkers based on
the target triple. This commit adds tests for these flags on Darwin, Linux,
and Windows.
This is a test-only change.
llvm-svn: 289685
Robert Lougher [Wed, 14 Dec 2016 18:37:50 +0000 (18:37 +0000)]
[InstCombine] When folding GEP through a phi node merge the debug locations
If all the operands to a phi node are getelementptr, instcombine
will try to pull them through the phi node, combining them into a single
operation. When it does this, the debug location of the new getelementptr
should be the merged debug locations of the phi node arguments.
Patch 4 of 8 for D26256. Folding of a getelementptr operation.
Differential Revision: https://reviews.llvm.org/D26256
llvm-svn: 289684
Eric Christopher [Wed, 14 Dec 2016 18:29:39 +0000 (18:29 +0000)]
This change does two things:
Adds a "Discriminator" field to struct DILineInfo, which defaults to 0.
Fills out the "Discriminator" field in DILineInfo in DWARFDebugLine::LineTable::getFileLineInfoForAddress().
in order to have a slightly nicer interface in getFileLineInfoForAddress.
Patch by Simon Que!
Differential Revision: https://reviews.llvm.org/D27649
llvm-svn: 289683
Dmitry Vyukov [Wed, 14 Dec 2016 18:20:18 +0000 (18:20 +0000)]
tsan: allow Java VM iterate over allocated objects
Objects may move during the garbage collection, and JVM needs
to notify ThreadAnalyzer about that. The new function
__tsan_java_find eliminates the need to maintain these
objects both in ThreadAnalyzer and JVM.
Author: Alexander Smundak (asmundak)
Reviewed in https://reviews.llvm.org/D27720
llvm-svn: 289682
Robert Lougher [Wed, 14 Dec 2016 18:14:57 +0000 (18:14 +0000)]
[InstCombine] Merge debug locations when folding through a phi node
If all the operands to a phi node are of the same operation, instcombine
will try to pull them through the phi node, combining them into a single
operation. When it does this, the debug location of the operation should
be the merged debug locations of the phi node arguments.
Patch 3 of 8 for D26256. Folding of a compare operation.
Differential Revision: https://reviews.llvm.org/D26256
llvm-svn: 289681
Kostya Serebryany [Wed, 14 Dec 2016 18:13:02 +0000 (18:13 +0000)]
[libFuzzer] disable msan for one more hook that reads target's data that might be uninitialized
llvm-svn: 289680
Robert Lougher [Wed, 14 Dec 2016 17:49:19 +0000 (17:49 +0000)]
[InstCombine] Merge debug locations when folding through a phi node
If all the operands to a phi node are of the same operation, instcombine
will try to pull them through the phi node, combining them into a single
operation. When it does this, the debug location of the operation should
be the merged debug locations of the phi node arguments.
Patch 2 of 8 for D26256. Folding of a binary operation.
Differential Revision: https://reviews.llvm.org/D26256
llvm-svn: 289679
Reid Kleckner [Wed, 14 Dec 2016 17:44:11 +0000 (17:44 +0000)]
Improve our handling of tag decls in function prototypes
r289225 broke AST invariants by reparenting enumerators into function
decl contexts. This improves things by only reparenting TagDecls while
also attempting to preserve the lexical declcontext chain. The
interesting example here is:
int f(struct S { enum E { a = 1 } b; } c);
The semantic contexts of E and S should be f, and the lexical context of
S should be f and the lexical context of E should be S. We didn't do
that with r289225, but now we should.
This change should also improve our behavior on this example:
void f() {
extern void ext(struct S { } o);
// S injected here
}
Before r289225 we would only remove 'S' from the surrounding tag
injection context if it was the TU, but now we properly reparent S from
f to ext.
Fixes PR31366
llvm-svn: 289678
Dehao Chen [Wed, 14 Dec 2016 17:23:16 +0000 (17:23 +0000)]
revert r289669 which breaks bots
llvm-svn: 289676
Dehao Chen [Wed, 14 Dec 2016 17:22:53 +0000 (17:22 +0000)]
revert r289670 which breaks bot.
llvm-svn: 289675
Yaxun Liu [Wed, 14 Dec 2016 17:16:52 +0000 (17:16 +0000)]
AMDGPU: Emit runtime metadata version 2 as YAML
Differential Revision: https://reviews.llvm.org/D25046
llvm-svn: 289674
Derek Schuff [Wed, 14 Dec 2016 17:05:34 +0000 (17:05 +0000)]
lit.cfg: Check value of build config rather than converting to boolean
This is a CMake var which never evaluates to false.
llvm-svn: 289673
Eric Liu [Wed, 14 Dec 2016 17:01:52 +0000 (17:01 +0000)]
[change-namespace] don't crash when type reference is in function type parameter list.
Reviewers: hokein
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D27758
llvm-svn: 289672
Matt Arsenault [Wed, 14 Dec 2016 16:52:06 +0000 (16:52 +0000)]
AMDGPU: Make AllocationPriority of SGPRs higher than VGPRs
Since SGPRs should spill to VGPRs, they should be allocated first.
I don't think this is sufficient for SGPRs to always spill to
VGPRs though.
llvm-svn: 289671
Dehao Chen [Wed, 14 Dec 2016 16:49:34 +0000 (16:49 +0000)]
Create SampleProfileLoader pass in llvm instead of clang
Summary:
We used to create SampleProfileLoader pass in clang. This makes LTO/ThinLTO unable to add this pass in the linker plugin. This patch moves the SampleProfileLoader pass creation from
clang to llvm pass manager builder.
Reviewers: tejohnson, davidxl, dnovillo
Subscribers: mehdi_amini, cfe-commits
Differential Revision: https://reviews.llvm.org/D27744
llvm-svn: 289670
Dehao Chen [Wed, 14 Dec 2016 16:49:28 +0000 (16:49 +0000)]
Create SampleProfileLoader pass in llvm instead of clang
Summary: We used to create SampleProfileLoader pass in clang. This makes LTO/ThinLTO unable to add this pass in the linker plugin. This patch moves the SampleProfileLoader pass creation from clang to llvm pass manager builder.
Reviewers: tejohnson, davidxl, dnovillo
Subscribers: llvm-commits, mehdi_amini
Differential Revision: https://reviews.llvm.org/D27743
llvm-svn: 289669
Petr Hosek [Wed, 14 Dec 2016 16:46:50 +0000 (16:46 +0000)]
[Driver] Allow setting the default linker during build
This change allows setting the default linker used by the Clang
driver when configuring the build.
Differential Revision: https://reviews.llvm.org/D25263
llvm-svn: 289668
Nirav Dave [Wed, 14 Dec 2016 16:43:44 +0000 (16:43 +0000)]
Revert "In visitSTORE, always use FindBetterChain, rather than only when UseAA is enabled."
Reverting due to ARM MCJIT and MIPS LLD error.
This reverts commit r289659.
llvm-svn: 289667
Kostya Kortchinsky [Wed, 14 Dec 2016 16:38:11 +0000 (16:38 +0000)]
[scudo] Relax the memalign test
Summary:
Now that we are not rounding up the sizes passed to the secondary allocator,
the memalign test could run out of aligned addresses to return for larger
alignments. We now reduce the size of the quarantine for that test, and
allocate less chunks for the larger alignments.
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D27760
llvm-svn: 289665
Matt Arsenault [Wed, 14 Dec 2016 16:36:12 +0000 (16:36 +0000)]
AMDGPU: Change vintrp printing
llvm-svn: 289664
Derek Schuff [Wed, 14 Dec 2016 16:20:25 +0000 (16:20 +0000)]
Revert gold part of change, just liblto
llvm-svn: 289663
Derek Schuff [Wed, 14 Dec 2016 16:20:22 +0000 (16:20 +0000)]
Disable libLTO tests when libLTO is not built
Summary:
The current test only checks whether ld64 is available, causing tests
to fail when ld64 is avilable but libLTO is not built.
Reviewers: beanz, mehdi_amini
Subscribers: mehdi_amini, llvm-commits
Differential Revision: https://reviews.llvm.org/D27739
llvm-svn: 289662
Robert Lougher [Wed, 14 Dec 2016 16:14:17 +0000 (16:14 +0000)]
New API for merging debug locations. NFC.
Given two debug locations the function getMergedLocation combines the
locations into a single location (which may be an empty location).
Please see https://reviews.llvm.org/D26256 for the discussion leading
up to this API.
Note the function is currently a stub. This allows optimisations to
use the API although no location will actually be used.
This is patch 1 out of 8 for D26256. As suggested by David Blaikie,
each change in D26256 has been broken out into a separate patch.
llvm-svn: 289661
Sylvestre Ledru [Wed, 14 Dec 2016 16:09:29 +0000 (16:09 +0000)]
Update the default of the Mozilla coding style
Summary:
I also proposed the change in Firefox .clang-format file:
https://bugzilla.mozilla.org/show_bug.cgi?id=1322321
Reviewers: klimek
Subscribers: cfe-commits
Differential Revision: https://reviews.llvm.org/D27557
llvm-svn: 289660
Nirav Dave [Wed, 14 Dec 2016 15:44:26 +0000 (15:44 +0000)]
In visitSTORE, always use FindBetterChain, rather than only when UseAA is enabled.
Retrying after fixing after removing load-store factoring through
token factors in favor of improved token factor operand pruning
Simplify Consecutive Merge Store Candidate Search
Now that address aliasing is much less conservative, push through
simplified store merging search which only checks for parallel stores
through the chain subgraph. This is cleaner as the separation of
non-interfering loads/stores from the store-merging logic.
Whem merging stores, search up the chain through a single load, and
finds all possible stores by looking down from through a load and a
TokenFactor to all stores visited. This improves the quality of the
output SelectionDAG and generally the output CodeGen (with some
exceptions).
Additional Minor Changes:
1. Finishes removing unused AliasLoad code
2. Unifies the the chain aggregation in the merged stores across
code paths
3. Re-add the Store node to the worklist after calling
SimplifyDemandedBits.
4. Increase GatherAllAliasesMaxDepth from 6 to 18. That number is
arbitrary, but seemed sufficient to not cause regressions in
tests.
This finishes the change Matt Arsenault started in r246307 and
jyknight's original patch.
Many tests required some changes as memory operations are now
reorderable. Some tests relying on the order were changed to use
volatile memory operations
Noteworthy tests:
CodeGen/AArch64/argument-blocks.ll -
It's not entirely clear what the test_varargs_stackalign test is
supposed to be asserting, but the new code looks right.
CodeGen/AArch64/arm64-memset-inline.lli -
CodeGen/AArch64/arm64-stur.ll -
CodeGen/ARM/memset-inline.ll -
The backend now generates *worse* code due to store merging
succeeding, as we do do a 16-byte constant-zero store efficiently.
CodeGen/AArch64/merge-store.ll -
Improved, but there still seems to be an extraneous vector insert
from an element to itself?
CodeGen/PowerPC/ppc64-align-long-double.ll -
Worse code emitted in this case, due to the improved store->load
forwarding.
CodeGen/X86/dag-merge-fast-accesses.ll -
CodeGen/X86/MergeConsecutiveStores.ll -
CodeGen/X86/stores-merging.ll -
CodeGen/Mips/load-store-left-right.ll -
Restored correct merging of non-aligned stores
CodeGen/AMDGPU/promote-alloca-stored-pointer-value.ll -
Improved. Correctly merges buffer_store_dword calls
CodeGen/AMDGPU/si-triv-disjoint-mem-access.ll -
Improved. Sidesteps loading a stored value and
merges two stores
CodeGen/X86/pr18023.ll -
This test has been removed, as it was asserting incorrect
behavior. Non-volatile stores *CAN* be moved past volatile loads,
and now are.
CodeGen/X86/vector-idiv.ll -
CodeGen/X86/vector-lzcnt-128.ll -
It's basically impossible to tell what these tests are actually
testing. But, looks like the code got better due to the memory
operations being recognized as non-aliasing.
CodeGen/X86/win32-eh.ll -
Both loads of the securitycookie are now merged.
Reviewers: arsenm, hfinkel, tstellarAMD, jyknight, nhaehnle
Subscribers: wdng, nhaehnle, nemanjai, arsenm, weimingz, niravd, RKSimon, aemerson, qcolombet, dsanders, resistor, tstellarAMD, t.p.northover, spatel
Differential Revision: https://reviews.llvm.org/D14834
llvm-svn: 289659
Piotr Padlewski [Wed, 14 Dec 2016 15:42:23 +0000 (15:42 +0000)]
Deleted unused typedef
llvm-svn: 289658
Kelvin Li [Wed, 14 Dec 2016 15:39:58 +0000 (15:39 +0000)]
Fix assert message. NFC.
llvm-svn: 289657
Piotr Padlewski [Wed, 14 Dec 2016 15:29:23 +0000 (15:29 +0000)]
modernize-use-auto NFC fixes
llvm-svn: 289656
Simon Pilgrim [Wed, 14 Dec 2016 15:14:44 +0000 (15:14 +0000)]
Wdocumentation fix
llvm-svn: 289655
Simon Pilgrim [Wed, 14 Dec 2016 15:08:13 +0000 (15:08 +0000)]
[DAGCombiner] Try to use SelectionDAG::isKnownToBeAPowerOfTwo instead of just APInt::isPowerOf2
Generalize sdiv/udiv/srem/urem combines using APInt::isPowerOf2, which only works for const/splat-const values, to call SelectionDAG::isKnownToBeAPowerOfTwo instead which recognises many more cases.
Added a DAGCombiner::BuildLogBase2 helper since PowerOf2 combines often involve taking the log2 of such a value.
Differential Revision: https://reviews.llvm.org/D27714
llvm-svn: 289654
Michael Zuckerman [Wed, 14 Dec 2016 14:57:10 +0000 (14:57 +0000)]
Fix bug 30945- [AVX512] Failure to flip vector comparison to remove not mask instruction
adding new optimization opportunity by adding new X86ISelLowering pattern. The test case was shown in https://llvm.org/bugs/show_bug.cgi?id=30945.
Test explanation:
Select gets three arguments mask, op and op2. In this case, the Mask is a result of ICMP. The ICMP instruction compares (with equal operand) the zero initializer vector and the result of the first ICMP.
In general, The result of "cmp eq, op1, zero initializers" is "not(op1)" where op1 is a mask. By rearranging of the two arguments inside the Select instruction, we can get the same result. Without the necessary of the middle phase ("cmp eq, op1, zero initializers").
Missed optimization opportunity:
vpcmpled %zmm0, %zmm1, %k0
knotw %k0, %k1
can be combine to
vpcmpgtd %zmm0, %zmm2, %k1
Reviewers:
1. delena
2. igorb
Commited after check all
Differential Revision: https://reviews.llvm.org/D27160
llvm-svn: 289653
Simon Pilgrim [Wed, 14 Dec 2016 14:39:51 +0000 (14:39 +0000)]
[X86][SSE] Add AVX1 tests to sdiv/udiv srem/urem combine tests
As requested on D27714
llvm-svn: 289652
Renato Golin [Wed, 14 Dec 2016 13:24:20 +0000 (13:24 +0000)]
Revert "[AVR] Add the very first on-target test"
This reverts commit r289648, as it's an execution test and relies on the
emulator/dispatcher being available on all builders.
llvm-svn: 289651
Neil Hickey [Wed, 14 Dec 2016 13:18:48 +0000 (13:18 +0000)]
Fixing cast condition for removing casts from builtin FPClassification.
The function SemaBuiltinFPClassification removed superfluous float to double
casts, this was changed to also remove float to float casts but this isn't
valid in all cases, for example when doing an rvaluetolvalue cast. Added a
check to only remove if this was a conventional floating cast.
Added additional tests into SemaOpenCL/extensions to cover these cases
llvm-svn: 289650
Stephan Bergmann [Wed, 14 Dec 2016 12:11:35 +0000 (12:11 +0000)]
Adapt to recent APFloat change
llvm-svn: 289649
Dylan McKay [Wed, 14 Dec 2016 12:03:39 +0000 (12:03 +0000)]
[AVR] Add the very first on-target test
This test runs on actual AVR hardware.
llvm-svn: 289648
Stephan Bergmann [Wed, 14 Dec 2016 11:57:17 +0000 (11:57 +0000)]
Replace APFloatBase static fltSemantics data members with getter functions
At least the plugin used by the LibreOffice build
(<https://wiki.documentfoundation.org/Development/Clang_plugins>) indirectly
uses those members (through inline functions in LLVM/Clang include files in turn
using them), but they are not exported by utils/extract_symbols.py on Windows,
and accessing data across DLL/EXE boundaries on Windows is generally
problematic.
Differential Revision: https://reviews.llvm.org/D26671
llvm-svn: 289647
Artur Pilipenko [Wed, 14 Dec 2016 11:55:47 +0000 (11:55 +0000)]
Add a couple of assertions to the load combine code introduced by r289538
llvm-svn: 289646
Dylan McKay [Wed, 14 Dec 2016 11:47:14 +0000 (11:47 +0000)]
[AVR] Add the integrated testing tool to the .gitignore
We build it as an LLVM tool.
llvm-svn: 289645
Hafiz Abid Qadeer [Wed, 14 Dec 2016 11:13:03 +0000 (11:13 +0000)]
Remove code needed for supporting old VS versions.
Summary: This code was probably needed to support VS2013 and is not needed now. I have built it with VS and mingw. Ok to remove it?
Reviewers: zturner, abidh
Subscribers: lldb-commits
Differential Revision: https://reviews.llvm.org/D27707
llvm-svn: 289644
Oliver Stannard [Wed, 14 Dec 2016 10:43:58 +0000 (10:43 +0000)]
[Assembler] Better error messages for .org directive
Currently, the error messages we emit for the .org directive when the
expression is not absolute or is out of range do not include the line
number of the directive, so it can be hard to track down the problem if
a file contains many .org directives.
This patch stores the source location in the MCOrgFragment, so that it
can be used for diagnostics emitted during layout.
Since layout is an iterative process, and the errors are detected during
each iteration, it would have been possible for errors to be reported
multiple times. To prevent this, I've made the assembler bail out after
each iteration if any errors have been reported. This will still allow
multiple unrelated errors to be reported in the common case where they
are all detected in the first round of layout.
Differential Revision: https://reviews.llvm.org/D27411
llvm-svn: 289643
Peter Smith [Wed, 14 Dec 2016 10:36:12 +0000 (10:36 +0000)]
[ELF] Accept first SHT_ARM_ATTRIBUTES section
The eglibc library, as used by Ubuntu 14.04 requires the presence of an
SHT_ARM_ATTRIBUTES section in for the purposes of checking hard/soft float
compatibility when dlopen() is used. Unfortunately when the section is not
present dlopen() fails with a generic could not find file message.
This change makes lld keep the first .ARM.attributes section that it
encounters and propagates it to the output. This is not a complete
SHT_ARM_ATTRIBUTES implementation, that would involve reading the contents
of the section and joining each individual attribute. It should suffice
for a homogenous build all libraries and executables on the same system
with a compatible set of command line options.
Differential revision: https://reviews.llvm.org/D27718
llvm-svn: 289642
Dylan McKay [Wed, 14 Dec 2016 10:15:00 +0000 (10:15 +0000)]
[AVR] Add a function instrumentation pass
This will be used for an on-chip test suite.
llvm-svn: 289641
Andrey Churbanov [Wed, 14 Dec 2016 08:29:00 +0000 (08:29 +0000)]
Cleanup: debug print fixed and moved inside critical section.
Patch by Victor Campos.
Differential Revision: https://reviews.llvm.org/D27647
llvm-svn: 289640
Craig Topper [Wed, 14 Dec 2016 07:46:12 +0000 (07:46 +0000)]
[X86][InstCombine] Handle demanded elements for operand of AVX-512 scalar floating point to integer conversion intrinsics.
llvm-svn: 289639
Hal Finkel [Wed, 14 Dec 2016 07:24:50 +0000 (07:24 +0000)]
[PowerPC] Fix logic dealing with nop after calls (and tail-call eligibility)
This change aims to unify and correct our logic for when we need to allow for
the possibility of the linker adding a TOC restoration instruction after a
call. This comes up in two contexts:
1. When determining tail-call eligibility. If we make a tail call (i.e.
directly branch to a function) then there is no place for the linker to add
a TOC restoration.
2. When determining when we need to add a nop instruction after a call.
Likewise, if there is a possibility that the linker might need to add a
TOC restoration after a call, then we need to put a nop after the call
(the bl instruction).
First problem: We were using similar, but different, logic to decide (1) and
(2). This is just wrong. Both the resideInSameModule function (used when
determining tail-call eligibility) and the isLocalCall function (used when
deciding if the post-call nop is needed) were supposed to be determining the
same underlying fact (i.e. might a TOC restoration be needed after the call).
The same logic should be used in both places.
Second problem: The logic in both places was wrong. We only know that two
functions will share the same TOC when both functions come from the same
section of the same object. Otherwise the linker might cause the functions to
use different TOC base addresses (unless the multi-TOC linker option is
disabled, in which case only shared-library boundaries are relevant). There are
a number of factors that can cause functions to be placed in different sections
or come from different objects (-ffunction-sections, explicitly-specified
section names, COMDAT, weak linkage, etc.). All of these need to be checked.
The existing logic only checked properties of the callee, but the properties of
the caller must also be checked (for example, calling from a function in a
COMDAT section means calling between sections).
There was a conceptual error in the resideInSameModule function in that it
allowed tail calls to functions with weak linkage and protected/hidden
visibility. While protected/hidden visibility does prevent the function
implementation from being replaced at runtime (via interposition), it does not
prevent the linker from using an alternate implementation at link time (i.e.
using some strong definition to replace the provided weak one during linking).
If this happens, then we're still potentially looking at a required TOC
restoration upon return.
Otherwise, in general, the post-call nop is needed wherever ELF interposition
needs to be supported. We don't currently support ELF interposition at the IR
level (see http://lists.llvm.org/pipermail/llvm-dev/2016-November/107625.html
for more information), and I don't think we should try to make it appear to
work in the backend in spite of that fact. This will yield subtle bugs if
interposition is attempted. As a result, regardless of whether we're in PIC
mode, we don't assume that we need to add the nop to support the possibility of
ELF interposition. However, the necessary check is in place (i.e. calling
GV->isInterposable and TM.shouldAssumeDSOLocal) so when we have functions for
which interposition is allowed at the IR level, we'll add the nop as necessary.
In the mean time, we'll generate more tail calls and fewer nops when compiling
position-independent code.
Differential Revision: https://reviews.llvm.org/D27231
llvm-svn: 289638
Justin Lebar [Wed, 14 Dec 2016 06:52:23 +0000 (06:52 +0000)]
[clang-tidy] Suggest including <cmath> if necessary in type-promotion-in-math-fn-check.
Reviewers: alexfh
Subscribers: JDevlieghere, cfe-commits
Differential Revision: https://reviews.llvm.org/D27748
llvm-svn: 289637
Craig Topper [Wed, 14 Dec 2016 06:06:58 +0000 (06:06 +0000)]
[X86][InstCombine] Teach SimplifyDemandedVectorElts to handle masked scalar add/sub/mul/div/max/min intrinsics better.
Now we can remove these intrinsics if element 0 isn't used. Also fix undef element tracking.
llvm-svn: 289636
Craig Topper [Wed, 14 Dec 2016 05:43:05 +0000 (05:43 +0000)]
[X86][InstCombine] Handle scalar fmadd intrinsics correctly in SimplifyDemandedVectorElts.
Now we pass a modified version of DemandedElts to each operand and we calculate undef elts correctly.
llvm-svn: 289632
Mehdi Amini [Wed, 14 Dec 2016 04:56:42 +0000 (04:56 +0000)]
[ThinLTO] Add an API to trigger file-based API for returning objects to the linker
Summary:
The motivation is to support better the -object_path_lto option on
Darwin. The linker needs to write down the generate object files on
disk for later use by lldb or dsymutil (debug info are not present
in the final binary). We're moving this into libLTO so that we can
be smarter when a cache is enabled and hard-link when possible
instead of duplicating the files.
Reviewers: tejohnson, deadalnix, pcc
Subscribers: dexonsmith, llvm-commits
Differential Revision: https://reviews.llvm.org/D27507
llvm-svn: 289631
Richard Smith [Wed, 14 Dec 2016 03:22:16 +0000 (03:22 +0000)]
[c++1z] P0217R3: Allow by-value structured binding of arrays.
llvm-svn: 289630
Craig Topper [Wed, 14 Dec 2016 03:17:30 +0000 (03:17 +0000)]
[X86][InstCombine] Teach SimplifyDemandedVectorElts to handle scalar round intrinsics more correctly.
Now we only pass bit 0 of the DemandedElts to optimize operand 1 as we recurse since the upper bits are unused. Similarly we clear bit 0 for optimizing operand 0.
Also calculate UndefElts correctly.
Simplify InstCombineCalls for these instrinics to just call SimplifyDemandedVectorElts for the call instrution to reuse this support.
llvm-svn: 289629
Craig Topper [Wed, 14 Dec 2016 03:17:27 +0000 (03:17 +0000)]
[X86][InstCombine] Teach SimplifyDemandedVectorElts to handle scalar min/max/cmp intrinsics more correctly.
Now we only pass bit 0 of the DemandedElts to optimize operand 1 as we recurse since the upper bits are unused.
Also calculate UndefElts correctly.
Simplify InstCombineCalls for these instrinics to just call SimplifyDemandedVectorElts for the call instrution to reuse this support.
llvm-svn: 289628
Justin Lebar [Wed, 14 Dec 2016 03:15:01 +0000 (03:15 +0000)]
[ClangTidy] Add new performance-type-promotion-in-math-fn check.
Summary:
This checks for calls to double-precision math.h with single-precision
arguments. For example, it suggests replacing ::sin(0.f) with
::sinf(0.f).
Subscribers: mgorny, cfe-commits
Differential Revision: https://reviews.llvm.org/D27284
llvm-svn: 289627
Mehdi Amini [Wed, 14 Dec 2016 02:35:32 +0000 (02:35 +0000)]
Don't double-initialize cl::opt for iterating in reverse order to uncover non-determinism in codegen by default
Bots are broken and needs to be fixed before having this on by default.
The feature was committed in r289619.
I tried to disable it in r289624 and failed because it was initialized in two places.
llvm-svn: 289626
Paul Robinson [Wed, 14 Dec 2016 02:06:11 +0000 (02:06 +0000)]
[PS4] Undo dialect tweak for Objective-C.
In r267772, we had set the PS4's default dialect for both C and
Objective-C to gnu99. Make that change only for C; we don't really
support Objective-C/C++ so there's no point fiddling the dialect.
llvm-svn: 289625
Mehdi Amini [Wed, 14 Dec 2016 02:02:28 +0000 (02:02 +0000)]
Disable Iterating SmallPtrSet in reverse order to uncover non-determinism in codegen by default
Bots are broken and needs to be fixed before having this on by default.
The feature was committed in r289619.
llvm-svn: 289624
Richard Smith [Wed, 14 Dec 2016 01:32:13 +0000 (01:32 +0000)]
When emitting a multidimensional array copy, only emit a single flattened
cleanup loop for exception handling.
llvm-svn: 289623
Kostya Serebryany [Wed, 14 Dec 2016 01:31:21 +0000 (01:31 +0000)]
[libFuzzer] document one more desired feature of a fuzz target
llvm-svn: 289622
Peter Collingbourne [Wed, 14 Dec 2016 01:17:59 +0000 (01:17 +0000)]
LTO: Add support for multi-module bitcode files.
Differential Revision: https://reviews.llvm.org/D27313
llvm-svn: 289621
Paul Robinson [Wed, 14 Dec 2016 00:27:35 +0000 (00:27 +0000)]
[DWARF] Preserve column number when emitting 'line 0' record
Follow-up to r289256, address a FIXME to avoid resetting the column
number. This reduced .debug_line by 2.6% in a RelWithDebInfo
self-build of clang.
llvm-svn: 289620