Nikita Popov [Thu, 30 Jul 2020 17:14:02 +0000 (19:14 +0200)]
[NewPM][LVI] Abandon LVI after CVP
As mentioned on D70376, LVI can currently cause performance issues
when running under NewPM. The problem is that, unlike the legacy
pass manager, NewPM will not immediately discard the LVI analysis
if the following pass does not need it. This is a problem, because
LVI has a high memory requirement, and mass invalidation of LVI
values is very inefficient. LVI should only be alive during passes
that actively interact with it.
This patch addresses the issue by explicitly abandoning LVI after CVP,
which gets us back to the LegacyPM behavior.
Differential Revision: https://reviews.llvm.org/D84959
Craig Topper [Sat, 1 Aug 2020 20:25:18 +0000 (13:25 -0700)]
[X86] Add assembler support for {disp8} and {disp32} to control the size of displacement used for memory operands.
These prefixes should override the default behavior and force a larger immediate size. I don't believe gas issues any warning if you use {disp8} when a 32-bit displacement is already required. And this patch doesn't either.
This completes the {disp8} and {disp32} support from PR46650.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D84793
Craig Topper [Sat, 1 Aug 2020 17:01:58 +0000 (10:01 -0700)]
[InstSimplify] Fold abs(abs(x)) -> abs(x)
It's always safe to pick the earlier abs regardless of the nsw flag. We'll just lose it if it is on the outer abs but not the inner abs.
Differential Revision: https://reviews.llvm.org/D85053
Craig Topper [Sat, 1 Aug 2020 16:59:09 +0000 (09:59 -0700)]
[InstCombine] Fold abs(-x) -> abs(x)
Negating the input doesn't matter. I left a FIXME to copy the nsw flag if its present on the neg but not on the abs.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D85055
Florian Hahn [Sat, 1 Aug 2020 19:58:05 +0000 (20:58 +0100)]
[PPC] Adjust run line for hardware-loops-crash.ll
Looks like %s was accidentally dropped.
Florian Hahn [Sat, 1 Aug 2020 19:38:38 +0000 (20:38 +0100)]
[LCSSA] Provide option for caller to clean up unused PHIs.
formLCSSAForInstructions is used by SCEVExpander, which tracks all
inserted instructions including LCSSA phis using asserting value
handles. This means cleanup needs to happen in the caller.
Extend formLCSSAForInstructions to take an optional pointer to a
vector. If this argument is non-nullptr, instead of directly deleting
the phis, add them to the vector, so the caller can process them.
This should address various PPC buildbot failures, including
http://lab.llvm.org:8011/builders/clang-ppc64be-linux-lnt/builds/40567
Simon Pilgrim [Sat, 1 Aug 2020 19:34:39 +0000 (20:34 +0100)]
[X86][AVX512] Fold concat(and(x,y),and(z,w)) -> and(concat(x,z),concat(y,w)) for 512-bit vectors
Helps vpternlog folding on non-AVX512BW targets
zoecarver [Sat, 1 Aug 2020 19:06:31 +0000 (12:06 -0700)]
[libcxx] Add compatible with constraint tests for some shared_ptr constructors.
Add shared_ptr tests where the element type and pointer type aren't 'convertible' but are 'compatible'.
Responding to a comment from D81414.
Differential Revision: https://reviews.llvm.org/D81532
Simon Pilgrim [Sat, 1 Aug 2020 13:54:21 +0000 (14:54 +0100)]
[X86][AVX] Ensure we only combine to PSHUFLW/PSHUFHW on supporting targets
Noticed while investigating combining from concatenated shuffle vectors, we weren't checking that PSHUFLW/PSHUFHW was legal - we were depending on lowering splitting to subvectors.
Florian Hahn [Sat, 1 Aug 2020 16:54:23 +0000 (17:54 +0100)]
[LCSSA] Use IRBuilder for PHI creation.
Use IRBuilder instead PHINode::Create. This should not impact the
generated code, but IRBuilder provides a way to register callbacks for
inserted instructions, which is convenient for some users.
Reviewed By: lebedev.ri
Differential Revision: https://reviews.llvm.org/D85037
Sanjay Patel [Sat, 1 Aug 2020 12:38:28 +0000 (08:38 -0400)]
[VectorCombine] add tests for non-zero gep offsets; NFC
Luofan Chen [Sat, 1 Aug 2020 13:27:16 +0000 (21:27 +0800)]
[Attributor][NFC] Update description for the dependency graph
The word "dependency graph" is a bit misleading. When there is an
edge from node A to B (A -> B), it actually mean that B depends on
A and when the state of A is updated, B should also be updated. So
I update the comment to make the description clearer.
Reviewed By: jdoerfert
Differential Revision: https://reviews.llvm.org/D85065
David Green [Sat, 1 Aug 2020 13:01:18 +0000 (14:01 +0100)]
[ARM] Distribute post-inc for Thumb2 sign/zero extending loads/stores
This adds sign/zero extending scalar loads/stores to the MVE
instructions added in D77813, allowing us to create up more post-inc
instructions. These are comparatively simple, compared to LDR/STR (which
may be better turned into an LDRD/LDM), but still require some additions
over MVE instructions. Because there are i12 and i8 variants of the
offset loads/stores dealing with different signs, we may need to convert
an i12 address to a i8 negative instruction. t2LDRBi12 can also be
shrunk to a tLDRi under the right conditions, so we need to be careful
with codesize too.
Differential Revision: https://reviews.llvm.org/D78625
Benjamin Kramer [Sat, 1 Aug 2020 12:48:42 +0000 (14:48 +0200)]
[mlir][Vector] Simplify code a bit. NFCI.
Sanjay Patel [Sat, 1 Aug 2020 11:46:23 +0000 (07:46 -0400)]
[InstSimplify] simplify abs if operand is known non-negative
abs() should be rare enough that using value tracking is not going
to be a compile-time cost burden, so use it to reduce a variety of
potential patterns. We do this in DAGCombiner too.
Differential Revision: https://reviews.llvm.org/D85043
Sanjay Patel [Sat, 1 Aug 2020 11:44:33 +0000 (07:44 -0400)]
[InstSimplify] add abs test with assume; NFC
Simon Pilgrim [Sat, 1 Aug 2020 11:28:10 +0000 (12:28 +0100)]
[X86][AVX] Extend v2f64 BROADCAST(LOAD) -> BROADCAST_LOAD to v2i64/v4f32/v4i32
Minor precursor fix for D66004, but helps the SSE41 tests as well as they run with -disable-peephole
Nathan James [Sat, 1 Aug 2020 10:04:29 +0000 (11:04 +0100)]
[clang-tidy][NFC] Small refactor
Evgeny Leviant [Sat, 1 Aug 2020 09:58:52 +0000 (12:58 +0300)]
[MachineVerifier] Refactor calcRegsPassed. NFC
Patch improves performance of verify-machineinstrs pass up to 10x.
Differential revision: https://reviews.llvm.org/D84105
Nathan James [Sat, 1 Aug 2020 09:35:13 +0000 (10:35 +0100)]
[clang-tidy] readability-identifier-naming checks configs for included files
When checking for the style of a decl that isn't in the main file, the check will now search for the configuration that the included files uses to gather the style for its decls.
This can be useful to silence warnings in header files that follow a different naming convention without using header-filter to silence all warnings(even from other checks) in the header file.
Reviewed By: aaron.ballman, gribozavr2
Differential Revision: https://reviews.llvm.org/D84814
Craig Topper [Sat, 1 Aug 2020 06:10:47 +0000 (23:10 -0700)]
[X86] Refactor the broadcast and load folding in tryVPTESTM to reduce some code.
Now we try to load and broadcast together for operand 1. Followed
by load and broadcast for operand 1. Previously we tried load
operand 1, load operand 1, broadcast operand 0, broadcast operand 1.
Now we have a single helper that tries load and broadcast for
one operand that we can just call twice.
Chen Zheng [Sat, 1 Aug 2020 05:28:15 +0000 (01:28 -0400)]
[SCEV] don't query getSCEV() for incomplete phis
querying getSCEV() for incomplete phis leads to wrong cache value in `ExprToIVMap`,
because incomplete phis may be simplified to same value before get SCEV expression.
Reviewed By: lebedev.ri, mkazantsev
Differential Revision: https://reviews.llvm.org/D77560
Craig Topper [Sat, 1 Aug 2020 04:38:59 +0000 (21:38 -0700)]
[X86] Use TargetLowering::getRegClassFor to simplify some code in tryVPTESTM. NFCI
Justin Hibbits [Thu, 9 Apr 2020 02:29:40 +0000 (21:29 -0500)]
PowerPC: Don't lower SELECT_CC to PPCISD::FSEL on SPE
SPE doesn't have a fsel instruction, so don't try to lower to it.
This fixes a "Cannot select: tN: f64 = PPCISD::FSEL tX, tY, tZ" error.
Reviewed By: #powerpc, lkail
Differential Revision: https://reviews.llvm.org/D77773
Justin Hibbits [Thu, 16 Apr 2020 14:42:39 +0000 (09:42 -0500)]
PowerPC: Fix SPE extloadf32 handling.
The patterns were incorrect copies from the FPU code, and are
unnecessary, since there's no extended load for SPE. Just let LLVM
itself do the work by marking it expand.
Reviewed By: #powerpc, lkail
Differential Revision: https://reviews.llvm.org/D78670
Kazushi (Jam) Marukawa [Sat, 11 Jul 2020 07:01:13 +0000 (16:01 +0900)]
[VE] Change calling convention to follow ABI
Change to expand all arguments and return values to i64 to follow ABI.
Update regression tests also.
Reviewed By: simoll
Differential Revision: https://reviews.llvm.org/D84581
Nathan James [Sat, 1 Aug 2020 00:45:33 +0000 (01:45 +0100)]
[clang-tidy][NFC] Added convienence methods for getting optional options
These methods abstract away Error handling when trying to read options that can't be parsed by logging the error automatically and returning None.
Reviewed By: gribozavr2
Differential Revision: https://reviews.llvm.org/D84812
Huihui Zhang [Sat, 1 Aug 2020 00:08:17 +0000 (17:08 -0700)]
[AArch64][SVE] Allow vector of pointers as legal type for masked load/store.
Refer to LangRef http://llvm.org/docs/LangRef.html#llvm-masked-load-intrinsics
'llvm.masked.load/store.*’ intrinsics are overloaded intrinsic, which allow the
load/store data to be a vector of any integer, floating-point or pointer data type.
Therefore, allow pointer data type when checking 'isLegalMaskedLoadStore()'.
Reviewed By: paulwalker-arm
Differential Revision: https://reviews.llvm.org/D85045
Richard Smith [Sat, 1 Aug 2020 00:07:36 +0000 (17:07 -0700)]
Don't crash if we deserialize a pack expansion type whose pattern
contains no packs.
Fixes a regression from
740a164dec483225cbd02ab6c82199e2747ffacb.
Craig Topper [Sat, 1 Aug 2020 00:07:20 +0000 (17:07 -0700)]
[X86] Simplify vpternlog immediate selection.
Rather than hardcoding immediate values for 12 different combinations
in a nested pair of switches, we can perform the matched logic
operation on 3 magic constants to calculate the immediate.
Special thanks to this tweet https://twitter.com/rygorous/status/
1187034321992871936
for making me realize I could do this.
cgyurgyik [Sat, 1 Aug 2020 00:14:34 +0000 (20:14 -0400)]
[libc] [obvious] In strrchr, remove cast to unsigned char before
comparison.
cgyurgyik [Fri, 31 Jul 2020 23:59:29 +0000 (19:59 -0400)]
[libc] [obvious] Fix strchr and strrchr tests so that constness is
actually verified.
Julian Lettner [Fri, 31 Jul 2020 23:09:13 +0000 (16:09 -0700)]
[compiler-rt][Darwin] Fix linker errors for check-asan
A recent change broke `ninja check-asan` on Darwin by causing an error
during linking of ASan unit tests [1].
Move the addition of `-ObjC` compiler flag outside of the new
`if(COMPILER_RT_STANDALONE_BUILD)` block. It doesn't add any global
flags (e.g, `${CMAKE_CXX_FLAGS}`) and the decision to add is based
solely on source paths (`${source_rpath}`).
[1]
8b2fcc42b895, https://reviews.llvm.org/D84466
Differential Revision: https://reviews.llvm.org/D85057
Hsiangkai Wang [Thu, 4 Jun 2020 19:34:01 +0000 (03:34 +0800)]
Support experimental v extension v0.9.
Differential revision: https://reviews.llvm.org/D81213
Hsiangkai Wang [Thu, 21 May 2020 09:30:20 +0000 (17:30 +0800)]
Upgrade MC to v0.9.
Differential revision: https://reviews.llvm.org/D80802
Craig Topper [Fri, 31 Jul 2020 22:46:12 +0000 (15:46 -0700)]
[ValueTracking] Improve llvm.abs handling in computeKnownBits.
Add the optimizations we have in the SelectionDAG version.
Known non-negative copies all known bits. Any known one other than
the sign bit makes result non-negative.
Differential Revision: https://reviews.llvm.org/D85000
Shilei Tian [Fri, 31 Jul 2020 22:54:09 +0000 (18:54 -0400)]
[OpenMP] Fixed the issue that target memory deallocation might be called when they're being used
This patch fixed the issue that target memory might be deallocated when
they're still being used or before they're used.
Reviewed By: ye-luo
Differential Revision: https://reviews.llvm.org/D84996
Sanjay Patel [Fri, 31 Jul 2020 20:35:41 +0000 (16:35 -0400)]
[InstSimplify] add tests for abs intrinsic; NFC
Andrei Lebedev [Fri, 31 Jul 2020 22:39:12 +0000 (15:39 -0700)]
Updated the -I option description.
Adrian Prantl [Fri, 31 Jul 2020 20:30:59 +0000 (13:30 -0700)]
Convert to early exit (NFC)
Jacques Pienaar [Fri, 31 Jul 2020 21:46:48 +0000 (14:46 -0700)]
[mlir] Add shape.with_shape op
This is an operation that can returns a new ValueShape with a different shape. Useful for composing shape function calls and reusing existing shape transfer functions.
Just adding the op in this change.
Differential Revision: https://reviews.llvm.org/D84217
River Riddle [Fri, 31 Jul 2020 20:18:13 +0000 (13:18 -0700)]
[mlir][PassIncGen] Refactor how pass registration is generated
The current output is a bit clunky and requires including files+macros everywhere, or manually wrapping the file inclusion in a registration function. This revision refactors the pass backend to automatically generate `registerFooPass`/`registerFooPasses` functions that wrap the pass registration. `gen-pass-decls` now takes a `-name` input that specifies a tag name for the group of passes that are being generated. For each pass, the generator now produces a `registerFooPass` where `Foo` is the name of the definition specified in tablegen. It also generates a `registerGroupPasses`, where `Group` is the tag provided via the `-name` input parameter, that registers all of the passes present.
Differential Revision: https://reviews.llvm.org/D84983
Rahul Joshi [Fri, 31 Jul 2020 18:39:49 +0000 (11:39 -0700)]
[MLIR][NFC] Add FuncOp::getArgumentTypes()
Differential Revision: https://reviews.llvm.org/D85038
Luboš Luňák [Thu, 30 Jul 2020 16:08:13 +0000 (18:08 +0200)]
[lldb] report an error if a CLI option lacks an argument
Differential Revision: https://reviews.llvm.org/D84955
Luboš Luňák [Thu, 30 Jul 2020 19:25:38 +0000 (21:25 +0200)]
[lldb] force full gui redraw on Ctrl+L
As is common with curses apps, this allows to redraw everything
in case something corrupts the screen. Apparently key modifiers
are difficult with curses (curses FAQ it "doesn't do that"),
thankfully Ctrl+key are simply control characters, so it's
(ascii & 037) => 12.
Differential Revision: https://reviews.llvm.org/D84972
Jaydeep Chauhan [Fri, 31 Jul 2020 19:22:22 +0000 (20:22 +0100)]
[clang] Use the location of the void parameters when complaining that only a single void parameter should be present.
Fixes PR46417.
Differential Revision: https://reviews.llvm.org/D84678
Reviewed By: aaron.ballman
Sriraman Tallam [Fri, 31 Jul 2020 19:26:53 +0000 (12:26 -0700)]
Fix a test typo which caused a breakage.
Sriraman Tallam [Fri, 31 Jul 2020 19:00:59 +0000 (12:00 -0700)]
New test for basic block sections options.
This tests lld basic block sections options:
+ --lto-basic-block-sections=
+ --lto-unique-basic-block-section-names
Differential Revision: https://reviews.llvm.org/D84462
Chris Gyurgyik [Fri, 31 Jul 2020 18:57:46 +0000 (14:57 -0400)]
[libc] [obvious] Add rest of strrchr test.
Sriraman Tallam [Fri, 31 Jul 2020 18:14:49 +0000 (11:14 -0700)]
Rename basic block sections options to be consistent.
D68049 created options for basic block sections: -fbasic-block-sections=,
-funique-basic-block-section-names. Rename options in llc and lld (--lto-)
to be consistent. Specifically,
+ Rename basicblock-sections to basic-block-sections
+ Rename unique-bb-section-names to unique-basic-block-section-names
Differential Revision: https://reviews.llvm.org/D84462
Julian Lettner [Fri, 31 Jul 2020 18:38:10 +0000 (11:38 -0700)]
[compiler-rt][Darwin] Fix GetOSMajorKernelOffset() on watchOS
`TARGET_OS_IOS` and `TARGET_OS_WATCH` are not mutually exclusive.
`SANITIZER_IOS` is defined for all embedded platforms. So the branch
for watchOS is never taken. We could fix this by switching the order
of the branches (but the reason for doing so is non-obvious). Instead,
lets use the Darwin-specific `TARGET_OS_*` macros which are mutually
exclusive.
cgyurgyik [Fri, 31 Jul 2020 18:36:23 +0000 (14:36 -0400)]
[libc] Adds strrchr implementation.
Reviewed By: sivachandra
Differential Revision: https://reviews.llvm.org/D84875
LLVM GN Syncbot [Fri, 31 Jul 2020 18:32:54 +0000 (18:32 +0000)]
[gn build] Port
b7cfa6ca928
Sidharth Baveja [Fri, 31 Jul 2020 18:31:58 +0000 (18:31 +0000)]
[Loop Peeling] Separate the Loop Peeling Utilities from the Loop Unrolling Utilities
Summary: This patch separates the Loop Peeling Utilities from Loop Unrolling.
The reason for this change is that Loop Peeling is no longer only being used by
loop unrolling; Patch D82927 introduces loop peeling with fusion, such that
loops can be modified to have to same trip count, making them legal to be
peeled.
Reviewed By: Meinersbur
Differential Revision: https://reviews.llvm.org/D83056
Sourabh Singh Tomar [Thu, 30 Jul 2020 18:17:51 +0000 (23:47 +0530)]
[flang][OpenMP] Added initial support for lowering OpenMP parallel construct
This patch lower `!OMP PARALLEL` construct from PFT to OpenMPDialect operations.
This is first patch in this direction(lowering parallel construct).
OpenMP parallel construct can have multiple clauses and parameters. This patch
only implements lowering of an empty(contains no code in body) parallel construct
without any clauses or parameters.
Patch is carved out of following approved PR:
https://github.com/flang-compiler/f18-llvm-project/pull/322
Reviewed By: kiranchandramohan, DavidTruby
Differential Revision: https://reviews.llvm.org/D84965
sameeran joshi [Fri, 24 Jul 2020 06:55:32 +0000 (12:25 +0530)]
[flang][NFC] Verify C781 from - Clause 7 constraint checks for f18.
Reviewed By: PeteSteinfeld
Differential Revision: https://reviews.llvm.org/D84503
Louis Dionne [Fri, 31 Jul 2020 16:56:36 +0000 (12:56 -0400)]
[libc++] Avoid including <Block.h> from <functional>
Block.h is a pretty common name, which can lead to nasty collisions with
user provided headers. Since we're only getting a few simple declarations
from the header, it's better to declare them manually than to include the
header.
rdar://
66384326
Differential Revision: https://reviews.llvm.org/D85035
Fangrui Song [Fri, 31 Jul 2020 18:20:15 +0000 (11:20 -0700)]
[Support] Fix computeHostNumPhysicalCores() to respect affinity
computeHostNumPhysicalCores() is designed to respect CPU affinity.
D84764 used sysconf(_SC_NPROCESSORS_ONLN) which does not respect
affinity.
SupportTests Threading.PhysicalConcurrency may fail if taskset -c is specified.
sameeran joshi [Fri, 24 Jul 2020 08:19:29 +0000 (13:49 +0530)]
[flang]Verify C7107, C7108, C7109 from - Clause 7 constraint checks for f18.
1. Annotate the sources with constraint numbers.
2. Add tests for
*C7107 (R765) digit shall have one of the values 0 or 1.
*C7108 (R766) digit shall have one of the values 0 through 7.
*C7109 (R764) A boz-literal-constant shall appear only as a data-stmt-constant in a DATA statement, or where explicitly allowed in 16.9 as an actual argument of an intrinsic procedure.
Reviewed By: PeteSteinfeld
Differential Revision: https://reviews.llvm.org/D84504
Tim Keith [Fri, 31 Jul 2020 18:10:44 +0000 (11:10 -0700)]
Revert "[flang] Fix multi-config generator builds."
This reverts commit
332170356e35ea9cdc2c1d612b61a50ec5ea322e.
The change breaks out-of-tree builds.
Discussion in https://reviews.llvm.org/D84022
Sanjay Patel [Fri, 31 Jul 2020 15:50:39 +0000 (11:50 -0400)]
[ConstantFolding] fold abs intrinsic
The handling for minimum value is similar to cttz/ctlz with 0 just above this case.
Differential Revision: https://reviews.llvm.org/D84942
Hans Wennborg [Thu, 23 Jul 2020 09:17:16 +0000 (11:17 +0200)]
RuntimeDyldELF: report_fatal_error instead of asserting for unimplemented relocations (PR46816)
This fixes the ExecutionEngine/MCJIT/stubs-sm-pic.ll test in no-asserts
builds which is set to XFAIL on some platforms like 32-bit x86. More
importantly, we probably don't want to silently error in these cases.
Differential revision: https://reviews.llvm.org/D84390
cgyurgyik [Fri, 31 Jul 2020 18:04:36 +0000 (14:04 -0400)]
[libc] Add islower and isupper implementation.
Reviewed By: sivachandra
Differential Revision: https://reviews.llvm.org/D84960
kristina [Fri, 31 Jul 2020 17:44:02 +0000 (18:44 +0100)]
[libunwind] Add -Wno-suggest-override to CMakeLists.txt.
Set -Wno-suggest-override where such warning is provided
by the compiler when building libunwind, alongside libcxx
and libcxxabi, using recent Clang. This extends behavior
introduced in
77e0e9e17daf0865620abcd41f692ab0642367c4
to libunwind, avoiding a large amount of warnings during
builds. See D84126 for the original patch.
Michael Kruse [Fri, 31 Jul 2020 18:00:06 +0000 (13:00 -0500)]
[OpenMP][docs] Update loop tiling status.
Craig Topper [Fri, 31 Jul 2020 17:43:00 +0000 (10:43 -0700)]
[ValueTracking] Add ComputeNumSignBits support for llvm.abs intrinsic
If absolute value needs turn a negative number into a positive number it reduces the number of sign bits by at most 1.
Differential Revision: https://reviews.llvm.org/D84971
sameeran joshi [Thu, 30 Jul 2020 19:32:04 +0000 (01:02 +0530)]
[Flang] Checks for constraint C7110-C7115.
Added more tests.
Annotate sources and tests.
Improve error message.
Reviewed By: PeteSteinfeld
Differential Revision: https://reviews.llvm.org/D85014
Jonas Devlieghere [Fri, 31 Jul 2020 17:52:49 +0000 (10:52 -0700)]
[lldb/Test] Reduce code duplication by importing subprocess globally
Import the subprocess module once instead of doing it inline which is
error prone and leads to needless code duplication.
Teresa Johnson [Thu, 30 Jul 2020 20:49:49 +0000 (13:49 -0700)]
[ThinLTO] Compile time improvement to propagateAttributes
I found that propagateAttributes was ~23% of a thin link's run time
(almost 4x higher than the second hottest function). The main reason is
that it re-examines a global var each time it is referenced. This
becomes unnecessary once it is marked both non read only and non write
only. I added a set to avoid doing redundant work, which dropped the
runtime of that thin link by almost 15%.
I made a smaller efficiency improvement (no measurable impact) to skip
all summaries for a VI if the first copy is dead. I added an assert to
ensure that all copies are dead if any is. The code in
computeDeadSymbols marks all summaries for a VI as live. There is one
corner case where it was skipping marking an alias as live, that I
fixed. However, since the code earlier marked all copies of a preserved
GUID's VI as live, and each 'visit' marks all copies live, the only case
where this could make a difference is summaries that were marked live
when they were built initially, and that is only a few special compiler
generated symbols and inline assembly symbols, so it likely is never
provoked in practice.
Differential Revision: https://reviews.llvm.org/D84985
Fangrui Song [Fri, 31 Jul 2020 17:46:27 +0000 (10:46 -0700)]
[Support][CommandLine] Delete unused llvm::cl::ParseEnvrironmentOptions
The function was added in 2003. It is not used and can be emulated with ParseCommandLineOptions.
Albion Fung [Fri, 31 Jul 2020 16:57:59 +0000 (11:57 -0500)]
[PowerPC] Add Vector String Isolate instruction definitions and MC Tests
This patch implements the instruction definition and MC tests for the vector
string isolate instructions.
Differential Revision: https://reviews.llvm.org/D84197
Florian Hahn [Fri, 31 Jul 2020 17:16:01 +0000 (18:16 +0100)]
[SCEVExpander] Name temporary instructions for LCSSA insertion (NFC).
Aditya Nandakumar [Fri, 31 Jul 2020 16:41:06 +0000 (09:41 -0700)]
[GISel] Add combiners for G_INTTOPTR and G_PTRTOINT
https://reviews.llvm.org/D84909
Patch adds two new GICombinerRules, one for G_INTTOPTR and one for
G_PTRTOINT. The G_INTTOPTR elides ptr2int(int2ptr(x)) to a copy of x, if
the cast is within the same address space. The G_PTRTOINT elides
int2ptr(ptr2int(x)) to a copy of x. Patch additionally adds new combiner
tests for the AArch64 target to test these new combiner rules.
Patch by mkitzan
Keith Randall [Fri, 31 Jul 2020 17:04:04 +0000 (10:04 -0700)]
Disable getauxval for Go
We want the Go build to not use getauxval, as we must support glibc < 2.16 platforms.
Reviewed By: dvyukov
Differential Revision: https://reviews.llvm.org/D84859
Simon Pilgrim [Fri, 31 Jul 2020 11:51:46 +0000 (12:51 +0100)]
[X86][SSE] Cleanup bitwise reduction check prefixes. NFC
Add AVX512BW/AVX512BWVL prefixes for a future patch
Hiroshi Yamauchi [Thu, 30 Jul 2020 18:22:25 +0000 (11:22 -0700)]
[PGO][test] Add test to check memops changes function hash
Following up D84782.
Differential Revision: https://reviews.llvm.org/D84953
Hongtao Yu [Fri, 31 Jul 2020 01:22:50 +0000 (18:22 -0700)]
[AutoFDO] Avoid merging inlinee samples multiple times
A function call can be replicated by optimizations like loop unroll and jump threading and the replicates end up sharing the sample nested callee profile. Therefore when it comes to merging samples for uninlined callees in the sample profile inliner, a callee profile can be merged multiple times which will cause an assert to fire.
This change avoids merging same callee profile for duplicate callsites by filtering out callee profiles with a non-zero head sample count.
Reviewed By: wenlei, wmi
Differential Revision: https://reviews.llvm.org/D84997
LLVM GN Syncbot [Fri, 31 Jul 2020 16:23:24 +0000 (16:23 +0000)]
[gn build] Port
df69492cdfa
Sameer Arora [Thu, 23 Jul 2020 23:26:42 +0000 (16:26 -0700)]
[llvm-libtool-darwin] Refactor Slice and writeUniversalBinary
Refactoring `Slice` class and function `createUniversalBinary` from
`llvm-lipo` into MachOUniversalWriter. This refactoring is necessary so
as to use the refactored code for creating universal binaries under
llvm-libtool-darwin.
Reviewed by alexshap, smeenai
Differential Revision: https://reviews.llvm.org/D84662
Amy Kwan [Tue, 28 Jul 2020 04:45:54 +0000 (23:45 -0500)]
[PowerPC] Implement low-order Vector Modulus Builtins, and add Vector Multiply/Divide/Modulus Builtins Tests
Power10 introduces new instructions for vector multiply, divide and modulus.
These instructions can be exploited by the builtin functions: vec_mul, vec_div,
and vec_mod, respectively.
This patch aims adds the function prototype, vec_mod, as vec_mul and vec_div
been previously implemented in altivec.h.
This patch also adds the following front end tests:
vec_mul for v2i64
vec_div for v4i32 and v2i64
vec_mod for v4i32 and v2i64
Differential Revision: https://reviews.llvm.org/D82576
Xing GUO [Fri, 31 Jul 2020 15:20:44 +0000 (23:20 +0800)]
[DWARFYAML][debug_aranges] Use yaml::Hex64 rather than uint64_t as length. NFC.
It's better to use yaml::Hex64 as length in the tuples of the address
range table.
Benjamin Kramer [Fri, 31 Jul 2020 15:27:44 +0000 (17:27 +0200)]
Hide some internal symbols. NFC.
Hans Wennborg [Fri, 31 Jul 2020 15:22:49 +0000 (17:22 +0200)]
Add flang to export.sh to it gets source tarballs in releases
Louis Dionne [Fri, 31 Jul 2020 15:18:01 +0000 (11:18 -0400)]
[libc++] Fix eager generator expression in DefineLinkerScript
As explained in https://gitlab.kitware.com/cmake/cmake/-/issues/21045,
both branches of an $<IF> generator expression are evaluated eagerly
by CMake. As a result, if the non-selected branch contains an invalid
generator expression (such as getting the OUTPUT_NAME property of a
non-existent target), a hard error will occur.
This failed builds using the cxxrt ABI library, which doesn't create
a CMake target currently.
Thomas Raoux [Fri, 31 Jul 2020 15:02:21 +0000 (08:02 -0700)]
[mlir][spirv] Relax restriction on pointer type for CooperativeMatrix load/store
This change allow CooperativeMatrix Load/Store operations to use pointer type
that may not match the matrix element type. This allow us to declare buffer
with a larger type size than the matrix element type. This follows SPIR-V spec
and this is needed to be able to use cooperative matrix in combination with
shared local memory efficiently.
Differential Revision: https://reviews.llvm.org/D84993
Richard Barton [Sat, 18 Jul 2020 13:22:18 +0000 (14:22 +0100)]
[flang] Add -h as a synonym for help
As expected by user in http://lists.llvm.org/pipermail/flang-dev/2020-June/000404.html
Depends on D84856
Differential Revision: https://reviews.llvm.org/D84857
Richard Barton [Thu, 25 Jun 2020 15:01:56 +0000 (16:01 +0100)]
[flang] Add details to --help screen on default behaviour
Add a usage string and a defaults section that clarifies:
* If no input files are given, f18 reads from stdin
* If no input files are given, f18 dumps the parse tree.
* The default behaviour is to exec F18_FC.
* The fefault F18_FC setting is 'gfortran'
Adds a simple regression test which tests the top and tail of the help
screen and the exit status.
Depends on D84855
Differential Revision: https://reviews.llvm.org/D84856
Richard Barton [Fri, 17 Jul 2020 08:15:21 +0000 (09:15 +0100)]
[flang] Make interactive behaviour more obvious
When flang is invoked with no files it waits for input on stdin. Make it
print a message saying this to prevent the user being surprised.
Differential Revision: https://reviews.llvm.org/D84855
Matt Arsenault [Fri, 27 May 2016 22:50:12 +0000 (15:50 -0700)]
Support addrspacecast initializers with isNoopAddrSpaceCast
Moves isNoopAddrSpaceCast to the TargetMachine. It logically belongs
with the DataLayout.
Frederik Gossen [Fri, 31 Jul 2020 14:17:31 +0000 (14:17 +0000)]
[MLIR][Shape] Allow unsafe `shape.broadcast`
In a context in which `shape.broadcast` is known not to produce an error value,
we want it to operate solely on extent tensors. The operation's behavior is
then undefined in the error case as the result type cannot hold this value.
Differential Revision: https://reviews.llvm.org/D84933
Nico Weber [Fri, 31 Jul 2020 13:54:39 +0000 (09:54 -0400)]
[gn build] (manually) merge
63d3aeb529
Sourabh Singh Tomar [Fri, 31 Jul 2020 13:22:12 +0000 (18:52 +0530)]
[MLIR,OpenMP][NFCI] Removed loop for accessing regions of ParallelOp
`ParallelOp` has only one region associated with it.
Reviewed By: kiranchandramohan, ftynse
Differential Revision: https://reviews.llvm.org/D85008
Erich Keane [Fri, 17 Jul 2020 16:48:01 +0000 (09:48 -0700)]
DR2303: Prefer 'nearer' base classes during template deduction.
DR2303 fixes the case where the derived-base match for template
deduction is ambiguous if a base-of-base ALSO matches. The canonical
example (as shown in the test) is just like the MSVC implementation of
std::tuple.
This fixes a fairly sizable issue, where if a user inherits from
std::tuple on Windows (with the MS STL), they cannot use that type to
call a function that takes std::tuple.
Differential Revision: https://reviews.llvm.org/D84048
Haojian Wu [Fri, 31 Jul 2020 12:32:18 +0000 (14:32 +0200)]
[clangd] Be more explicit on testing the optional DefLoc in LocatedSymbol.
And also fix a bug where we may return a meaningless location.
Differential Revision: https://reviews.llvm.org/D84919
Bjorn Pettersson [Fri, 31 Jul 2020 12:28:49 +0000 (14:28 +0200)]
[clang-tidy] Fix build problem after commit
45a720a864320bbbeb596a
When building with LLVM8.0 on RHEL7.8 I got failures like this
after commit
45a720a864320bbbe:
/app/llvm/8.0/bin/../lib/gcc/x86_64-unknown-linux-gnu/
5.4.0/../../../../include/c++/5.4.0/ext/new_allocator.h:120:23:
error: no matching constructor for initialization of
'std::pair<std::__cxx11::basic_string<char>,
std::__cxx11::basic_string<char> >'
{ ::new((void *)__p) _Up(std::forward<_Args>(__args)...); }
...
../../clang-tools-extra/clang-tidy/ClangTidyOptions.cpp:73:15:
note: in instantiation of function template specialization
'std::vector<std::pair<std::__cxx11::basic_string<char>,
std::__cxx11::basic_string<char> >,
std::allocator<std::pair<std::__cxx11::basic_string<char>,
std::__cxx11::basic_string<char> > > >::emplace_back<llvm::StringRef,
const std::__cxx11::basic_string<char> &>' requested here
Options.emplace_back(KeyValue.getKey(), KeyValue.getValue().Value);
This is an attempt to avoid such build problems.
Xing GUO [Fri, 31 Jul 2020 04:56:10 +0000 (12:56 +0800)]
[DWARFYAML] Make the debug_aranges entry optional.
This patch makes the 'debug_aranges' entry optional. If the entry is
empty, yaml2obj will only emit the header for it.
Reviewed By: jhenderson
Differential Revision: https://reviews.llvm.org/D84921
Xing GUO [Fri, 31 Jul 2020 12:06:30 +0000 (20:06 +0800)]
[DWARFYAML] Add helper function getDWARFEmitterByName(). NFC.
In this patch, we add a helper function getDWARFEmitterByName(). This
function returns the proper DWARF section emitting method by the name.
Reviewed By: jhenderson
Differential Revision: https://reviews.llvm.org/D84952
Xing GUO [Fri, 31 Jul 2020 12:02:20 +0000 (20:02 +0800)]
[DWARFYAML] Add emitDebug[GNU]Pub[names/types] functions. NFC.
In this patch, emitDebugPubnames(), emitDebugPubtypes(),
emitDebugGNUPubnames(), emitDebugGNUPubtypes() are added.
Reviewed By: jhenderson
Differential Revision: https://reviews.llvm.org/D85003
Kirill Bobyrev [Fri, 31 Jul 2020 12:02:19 +0000 (14:02 +0200)]
[clangd] Fix remote index build on macOS
macOS builds suddenly started failing:
https://github.com/kirillbobyrev/indexing-tools/runs/
925090879
This patch makes use of imported libraries and fixes builds for macOS.
Landing this without a review since the patch is quite straightforward
and I've been testing it on my local macOS machine for a while.
Differential Revision: https://reviews.llvm.org/D84928
Nemanja Ivanovic [Fri, 31 Jul 2020 11:19:47 +0000 (06:19 -0500)]
[Sanitizers] Fix lint failure with Python 3.6
There are some files in compiler-rt that use UTF-8 characters in some of the
comments. This causes lint failures with some versions of Python. This patch
just makes the encoding explicit in the call to open.