Fraser Cormack [Thu, 26 May 2022 14:11:02 +0000 (15:11 +0100)]
[RISCV][NFC] Unify compatibility checks under one function
Split off from D125021.
We were duplicating logic across different phases. Since we want to
ensure a consistency of logic across phases for correctness, this patch
combines our multiple compatibility checks into one function to better
convey this.
Several methods were made const too.
Reviewed By: reames
Differential Revision: https://reviews.llvm.org/D126472
Groverkss [Fri, 27 May 2022 10:21:40 +0000 (15:51 +0530)]
[MLIR][Presburger] Add intersectDomain/Range to IntegerRelation
This patch adds support for intersection a set with a relation.
Reviewed By: arjunp
Differential Revision: https://reviews.llvm.org/D126328
Anastasia Stulova [Fri, 27 May 2022 10:12:44 +0000 (11:12 +0100)]
[Doc][OpenCL] Misc wording improvements for SPIR-V
Ben Dunbobbin [Fri, 27 May 2022 10:02:04 +0000 (11:02 +0100)]
[llvm-ar][test] add special case of replace converting a regular to a thin archive
Add a regression test for:
https://github.com/llvm/llvm-project/issues/55527
Differential Revision: https://reviews.llvm.org/D125785
Andrew Ng [Thu, 26 May 2022 15:03:09 +0000 (16:03 +0100)]
[LLD][ELF] Drop the string null terminator from the hash in splitStrings
Differential Revision: https://reviews.llvm.org/D126484
River Riddle [Tue, 10 May 2022 17:23:57 +0000 (10:23 -0700)]
[mlir][Tablegen-LSP] Add support for a tracking definitions and references
This essentially builds an index for the parsed records and record values (fields).
This covers quite a few cases, but is limited by the currently lackluster location
tracking in tablegen. A followup will work on plumbing more locations through
tablegen, which should greatly improve what we can do here.
Differential Revision: https://reviews.llvm.org/D125443
River Riddle [Mon, 9 May 2022 22:22:12 +0000 (15:22 -0700)]
[mlir][Tablegen-LSP] Add support for include file link and hover
This allows for following links to include files. This support is effectively
identical to the logic in the PDLL language server, and code is shared as
much as possible.
Differential Revision: https://reviews.llvm.org/D125442
River Riddle [Mon, 9 May 2022 22:05:03 +0000 (15:05 -0700)]
[mlir][Tablegen-LSP] Add support for a compilation database
This provides a format for externally specifying the include directories
for a source file. The format of the tablegen database is exactly the
same as that for PDLL, namely it includes the absolute source file name and
the set of include directories. The database format is shared to simplify
the infra, and also because the format itself is general enough to share. Even
if we desire to expand in the future to contain the actual compilation command,
nothing there is specific enough that we would need two different formats.
As with PDLL, support for generating the database is added to our mlir_tablegen
cmake command.
Differential Revision: https://reviews.llvm.org/D125441
Florian Hahn [Fri, 27 May 2022 09:37:02 +0000 (10:37 +0100)]
[AArch64] Add support for FMA intrinsics to shouldSinkOperands.
If the fma operates on a legal vector type, the indexed variants can be
used, if the second operand is a splat of a valid index.
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D126234
Andrew Ng [Wed, 25 May 2022 18:18:18 +0000 (19:18 +0100)]
[Support] Reduce allocations in parallelForEach with move
Differential Revision: https://reviews.llvm.org/D126458
Balazs Benics [Fri, 27 May 2022 09:18:05 +0000 (11:18 +0200)]
Revert "[llvm][clang][bolt][NFC] Use llvm::less_first() when applicable"
This reverts commit
3988bd13988aad72ec979beb2361e8738584926b.
Did not build on this bot:
https://lab.llvm.org/buildbot#builders/215/builds/6372
/usr/include/c++/9/bits/predefined_ops.h:177:11: error: no match for call to
‘(llvm::less_first) (std::pair<long unsigned int, llvm::bolt::BinaryBasicBlock*>&, const std::pair<long unsigned int, std::nullptr_t>&)’
177 | { return bool(_M_comp(*__it, __val)); }
Balazs Benics [Fri, 27 May 2022 09:15:23 +0000 (11:15 +0200)]
[llvm][clang][bolt][NFC] Use llvm::less_first() when applicable
One could reuse this functor instead of rolling out your own version.
There were a couple other cases where the code was similar, but not
quite the same, such as it might have an assertion in the lambda or other
constructs. Thus, I've not touched any of those, as it might change the
behavior in some way.
As per https://discourse.llvm.org/t/submitting-simple-nfc-patches/62640/3?u=steakhal
Chris Lattner
> LLVM intentionally has a “yes, you can apply common sense judgement to
> things” policy when it comes to code review. If you are doing mechanical
> patches (e.g. adopting less_first) that apply to the entire monorepo,
> then you don’t need everyone in the monorepo to sign off on it. Having
> some +1 validation from someone is useful, but you don’t need everyone
> whose code you touch to weigh in.
Differential Revision: https://reviews.llvm.org/D126068
Balazs Benics [Fri, 27 May 2022 09:05:50 +0000 (11:05 +0200)]
[analyzer][NFCi] Annotate major nonnull returning functions
This patch annotates the most important analyzer function APIs.
Also adds a couple of assertions for uncovering any potential issues
earlier in the constructor; in those cases, the member functions were
already dereferencing the members unconditionally anyway.
Measurements showed no performance impact, nor crashes.
Reviewed By: martong
Differential Revision: https://reviews.llvm.org/D126198
Gabor Marton [Fri, 27 May 2022 09:02:11 +0000 (11:02 +0200)]
[analyzer][NFC] Rename GREngine->CoreEngine, GRExprEngine->ExprEngine in comments and txt files
fixes #115
Balazs Benics [Fri, 27 May 2022 08:54:25 +0000 (10:54 +0200)]
[analyzer][NFC] Remove unused default SVal constructors
Reviewed By: martong, xazax.hun
Differential Revision: https://reviews.llvm.org/D125708
Balazs Benics [Fri, 27 May 2022 08:54:25 +0000 (10:54 +0200)]
[analyzer][NFC] Remove unused friend SVal declarations
Reviewed By: martong, xazax.hun
Differential Revision: https://reviews.llvm.org/D125707
Balazs Benics [Fri, 27 May 2022 08:54:25 +0000 (10:54 +0200)]
[analyzer][NFC] Use idiomatic classof instead of isKind
- Rename `isKind()` to `classof()` to follow the llvm style RTTI.
- Take SVal by-value instead of reference.
- Mark `classof` public.
Reviewed By: martong, xazax.hun
Differential Revision: https://reviews.llvm.org/D125706
Nikolas Klauser [Thu, 26 May 2022 15:10:53 +0000 (17:10 +0200)]
[libc++] Enable ranges_robust_against* and niebloid tests for implemented ranges algorithms
Spies: libcxx-commits
Differential Revision: https://reviews.llvm.org/D126477
Luo, Yuanke [Fri, 27 May 2022 02:54:37 +0000 (10:54 +0800)]
[X86][AMX] Replace LDTILECFG with PLDTILECFGV on auto-config.
There is intrinsic `@llvm.x86.ldtilecfg` which is lowered to LDTILECFG.
This intrinsic is open for user to configure tile registers by
themselves. There is a chance that `@llvm.x86.ldtilecfg` would be mixed
with the new AMX intrinsics which depend on compiler to configure tile
registers. Separate pusedo instruction PLDTILECFGV would avoid
unexpected behavious when `@llvm.x86.ldtilecfg` is mixed with new AMX
intrinsics. Though user should not mix the two programming model,
compiler should avoid crash or UB when they are mixed.
Differential Revision: https://reviews.llvm.org/D126519
LLVM GN Syncbot [Fri, 27 May 2022 08:29:03 +0000 (08:29 +0000)]
[gn build] Port
11e3ad299fee
Nikolas Klauser [Thu, 26 May 2022 14:08:55 +0000 (16:08 +0200)]
[libc++] Implement ranges::is_sorted{, _until}
Reviewed By: Mordante, var-const, #libc
Spies: libcxx-commits, mgorny
Differential Revision: https://reviews.llvm.org/D125608
Balazs Benics [Fri, 27 May 2022 08:07:06 +0000 (10:07 +0200)]
[analyzer][docs] Fix typo in checker name
Fixes #55720
Balazs Benics [Fri, 27 May 2022 08:07:06 +0000 (10:07 +0200)]
[analyzer][NFC] Use MemRegion::getRegion()'s return value unconditionally
Reviewed By: martong
Differential Revision: https://reviews.llvm.org/D126123
Balazs Benics [Fri, 27 May 2022 08:07:06 +0000 (10:07 +0200)]
[analyzer][NFC] Remove unused SVal::hasConjuredSymbol
Reviewed By: martong
Differential Revision: https://reviews.llvm.org/D126130
Balazs Benics [Fri, 27 May 2022 08:07:06 +0000 (10:07 +0200)]
[analyzer][NFC] Remove unused nonloc::ConcreteInt::evalBinOp
Reviewed By: martong
Differential Revision: https://reviews.llvm.org/D126129
Balazs Benics [Fri, 27 May 2022 08:07:06 +0000 (10:07 +0200)]
[analyzer][NFC] Inline loc::ConcreteInt::evalBinOp
This patch also refactored some of the enclosing parts.
Reviewed By: martong
Differential Revision: https://reviews.llvm.org/D126128
Balazs Benics [Fri, 27 May 2022 08:07:06 +0000 (10:07 +0200)]
[analyzer][NFC] Inline ExprEngine::evalMinus
Reviewed By: martong
Differential Revision: https://reviews.llvm.org/D126125
Balazs Benics [Fri, 27 May 2022 08:07:06 +0000 (10:07 +0200)]
[analyzer][NFC] Inline ExprEngine::evalComplement
Reviewed By: martong
Differential Revision: https://reviews.llvm.org/D126124
River Riddle [Mon, 16 May 2022 23:52:49 +0000 (16:52 -0700)]
[mlir-vscode] Add support for highlighting pdll and tablegen markdown code blocks
This essentially just piggy backs off of the existing mlir support.
Differential Revision: https://reviews.llvm.org/D125734
Serge Pavlov [Fri, 11 Mar 2022 07:40:44 +0000 (14:40 +0700)]
[GlobalISel] Add G_IS_FPCLASS
Add a generic opcode to represent `llvm.is_fpclass` intrinsic.
Differential Revision: https://reviews.llvm.org/D121454
Alexander Batashev [Fri, 27 May 2022 06:23:27 +0000 (09:23 +0300)]
[mlir][LLVM] Add support for Calling Convention in LLVMFuncOp
This patch adds support for Calling Convention attribute in LLVM
dialect, including enums, custom syntax and import from LLVM IR.
Additionally fix import of dso_local attribute.
Reviewed By: ftynse
Differential Revision: https://reviews.llvm.org/D126161
Liqin.Weng [Fri, 27 May 2022 06:27:26 +0000 (06:27 +0000)]
[RISCV] Add test for experimental.vector.reverse
```
void vector_reverse_i64(int *A, int *B, int n) {
#pragma clang loop vectorize_width(4, scalable)
for (int i = n-1; i >= 0; i--)
A[i] = B[i] + 1;
}
```
When option: scalable-vectorization is on (or set #pragma clang loop vectorize_width(elements, scalable)), Reverse Iterators can't loop vectorization as <vscale x elements x elementType>
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D125866
Pavel Samolysov [Thu, 26 May 2022 14:03:08 +0000 (17:03 +0300)]
[libunwind][CMake] Fix name of LIBUNWIND_SUPPORTS_NODEFAULTLIBS_FLAG
The CMake variable LIBUNWIND_SUPPORTS_NODEFAULTLIBS_FLAG has been renamed
into C_SUPPORTS_NODEFAULTLIBS_FLAG because the last one is used in the
confix-ix.cmake file while the variable with the original name is not used
at al.
Differential Revision: https://reviews.llvm.org/D126466
Piggy NL [Thu, 26 May 2022 14:59:49 +0000 (22:59 +0800)]
[demangler][RISCV] Fix for long double
Summary:
The size of long double in RISCV (both RV32 and RV64) is 16 bytes, thus
the mangled_size shouble be 32.
This patch will fix test case
"_ZN5test01hIfEEvRAcvjplstT_Le4001a000000000000000E_c"
in test_demangle.pass.cpp, which is expected to be invalid but demangler
returned "void test0::h<float>(char (&) [(unsigned int)((sizeof (float))
+ (0x0.000000004001ap-16382L))])" in RISCV environment without this patch.
Reviewed By: urnathan
Differential Revision: https://reviews.llvm.org/D126480
Ping Deng [Fri, 27 May 2022 05:45:17 +0000 (05:45 +0000)]
[SelectionDAG][NFC] Simplify integer promotion in setcc/vp.setcc
Reviewed By: craig.topper
Differential Revision: https://reviews.llvm.org/D126516
Siva Chandra Reddy [Tue, 24 May 2022 08:13:11 +0000 (08:13 +0000)]
[libc] Add global stdout and stderr objects.
They are added as entrypoint object targets. The header-gen
infrastructure has been extended to enable handling standard required
global objects. The libc-api-test has also been extended to verify the
global object declarations.
Reviewed By: lntue
Differential Revision: https://reviews.llvm.org/D126329
LLVM GN Syncbot [Fri, 27 May 2022 02:54:55 +0000 (02:54 +0000)]
[gn build] Port
08cc05851872
Rahman Lavaee [Fri, 27 May 2022 02:27:05 +0000 (19:27 -0700)]
Reland "[Propeller] Promote functions with propeller profiles to .text.hot."
This relands commit
4d8d2580c53e130c3c3dd3877384301e3c495554.
The major change here is using 'addUsedIfAvailable<BasicBlockSectionsProfileReader>()` to make sure we don't change the pipeline tests.
Differential Revision: https://reviews.llvm.org/D126518
Chenbing Zheng [Fri, 27 May 2022 02:50:30 +0000 (10:50 +0800)]
[InstCombine] [NFC] precommit tests for bitcast-extelt
Luo, Yuanke [Fri, 27 May 2022 02:35:05 +0000 (10:35 +0800)]
[X86][AMX] Update test case with automation tool.
Hanhan Wang [Fri, 27 May 2022 02:20:36 +0000 (19:20 -0700)]
[mlir][Linalg] Relax vectorization condition to allow transposed output.
Reviewed By: ThomasRaoux, dcaballe
Differential Revision: https://reviews.llvm.org/D126454
Enna1 [Fri, 27 May 2022 01:39:04 +0000 (09:39 +0800)]
Add !nosanitize to FixedMetadataKinds
This patch adds !nosanitize metadata to FixedMetadataKinds.def, !nosanitize indicates that LLVM should not insert any sanitizer instrumentation.
Reviewed By: vitalybuka
Differential Revision: https://reviews.llvm.org/D126294
Rahman Lavaee [Fri, 27 May 2022 01:43:58 +0000 (18:43 -0700)]
Revert "[Propeller] Promote functions with propeller profiles to .text.hot."
This reverts commit
4d8d2580c53e130c3c3dd3877384301e3c495554.
Fangrui Song [Fri, 27 May 2022 01:13:38 +0000 (18:13 -0700)]
[RISCV] Simplify code after D125905
wren romano [Thu, 26 May 2022 19:38:58 +0000 (12:38 -0700)]
[mlir][sparse] Moving some functions around
This is a followup to D126105 to move functions in SparseTensorUtils.cpp to match their locations in SparseTensorUtils.h
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D126106
wren romano [Thu, 26 May 2022 19:31:58 +0000 (12:31 -0700)]
[mlir][sparse] Improving ExecutionEngine/SparseTensorUtils.h
This change makes the public API of SparseTensorUtils.cpp explicit, whereas before the publicity of these functions was only implicit. Implicit publicity is sufficient for mlir-opt to generate calls to these functions, but it's not enough to enable C/C++ code to call them directly in the usual way (i.e., without going through codegen). Thus, leaving the publicity implicit prevents development of other tools (e.g., microbenchmarks).
In addition this change also marks the functions MLIR_CRUNNERUTILS_EXPORT, which is required by the JIT under certain configurations (albeit not for anything in our test suite).
Reviewed By: aartbik
Differential Revision: https://reviews.llvm.org/D126105
Argyrios Kyrtzidis [Fri, 27 May 2022 00:18:32 +0000 (17:18 -0700)]
[test/ClangScanDeps] Add a target triple for `macro-expansions.cpp`
This should fix the `clang-ppc64-aix` builder.
Chelsea Cassanova [Thu, 26 May 2022 23:57:57 +0000 (16:57 -0700)]
[lldb/fuzzer] Moving target fuzzer into separate subdirectory
Moving lldb-target-fuzzer into its own subdirectory for better
organization and modularity.
Differential revision: https://reviews.llvm.org/D126507
Damian Rouson [Thu, 31 Mar 2022 20:55:43 +0000 (13:55 -0700)]
[flang] expand the this_image test coverage
Add a test with a range of this_image() intrinsic function
invocations, including a comprehensive set of standard-conforming
keyword and non-keyword arguments with and without optional
arguments present and with argument positions covering all
possible orderings. Also test that several non-conforming
this_image() invocations generate the correct error messages.
Differential Revision: https://reviews.llvm.org/D123331
Rahman Lavaee [Tue, 24 May 2022 17:54:36 +0000 (10:54 -0700)]
[Propeller] Promote functions with propeller profiles to .text.hot.
Today, text section prefixes (none, .unlikely, .hot, and .unkown) are determined based on PGO profile. However, Propeller may deem a function hot when PGO doesn't. Besides, when `-Wl,-keep-text-section-prefix=true` Propeller cannot enforce a global section ordering as the linker can only reorder sections within each output section (.text, .text.hot, .text.unlikely).
This patch promotes all functions with Propeller profiles (functions listed in the basic-block-sections profile) to .text.hot. The feature is hidden behind the flag `--bbsections-guided-section-prefix` which defaults to `true`.
The new implementation refactors the parsing of basic block sections profile into a new `BasicBlockSectionsProfileReader` analysis pass. This allows us to use the information earlier in `CodeGenPrepare` in order to set the functions text prefix. `BasicBlockSectionsProfileReader` will be used both by `BasicBlockSections` pass and `CodeGenPrepare`.
Differential Revision: https://reviews.llvm.org/D122930
Arthur Eubanks [Thu, 26 May 2022 21:18:58 +0000 (14:18 -0700)]
[NFC][JumpThreading] Remove InsertFreezeWhenUnfoldingSelect pass parameter
All callers pass true.
select-unfold-freeze.ll is now a subset of select.ll so delete it.
Reviewed By: nikic
Differential Revision: https://reviews.llvm.org/D126501
Philip Reames [Thu, 26 May 2022 22:24:41 +0000 (15:24 -0700)]
[RISCV] Add a subtarget feature to enable unaligned scalar loads and stores
A RISCV implementation can choose to implement unaligned load/store support. We currently don't have a way for such a processor to indicate a preference for unaligned load/stores, so add a subtarget feature.
There doesn't appear to be a formal extension for unaligned support. The RISCV Profiles (https://github.com/riscv/riscv-profiles/blob/main/profiles.adoc#rva20u64-profile) docs use the name Zicclsm, but a) that doesn't appear to actually been standardized, and b) isn't quite what we want here anyway due to the perf comment.
Instead, we can follow precedent from other backends and have a feature flag for the existence of misaligned load/stores with sufficient performance that user code should actually use them.
Differential Revision: https://reviews.llvm.org/D126085
Rafael Auler [Sat, 21 May 2022 00:42:58 +0000 (17:42 -0700)]
[BOLT] Fix AND evaluation bug in shrink wrapping
Fix a bug where shrink-wrapping would use wrong stack offsets
because the stack was being aligned with an AND instruction, hence,
making its true offsets only available during runtime (we can't
statically determine where are the stack elements and we must give up
on this case).
Reviewed By: Amir
Differential Revision: https://reviews.llvm.org/D126110
Sam Clegg [Thu, 26 May 2022 21:56:36 +0000 (14:56 -0700)]
[lld][WebAssembly] Update test after
87628f5804e2
Jim Ingham [Wed, 18 May 2022 21:39:03 +0000 (14:39 -0700)]
Store a by name list of signals with their actions in the Target
so that they can be used to prime new Process runs. "process handle"
was also changed to populate the dummy target if there's no selected
target, so that the settings will get copied into new targets.
Differential Revision: https://reviews.llvm.org/D126259
Matthias Braun [Thu, 26 May 2022 21:47:49 +0000 (14:47 -0700)]
Use cmake Python3_EXECUTABLE variable instead of hardcoding
Sam Clegg [Thu, 26 May 2022 21:05:05 +0000 (14:05 -0700)]
[lld][WebAssembly] Require double dash for modern linker flags
This matches the behaviour of the ELF backend (in fact this change
is mostly just copying directly from ELF/Options.td).
Differential Revision: https://reviews.llvm.org/D126500
Fangrui Song [Thu, 26 May 2022 21:41:03 +0000 (14:41 -0700)]
[Driver][test] Change -target i386-unknown-linux-gnu to --target=i386 and remove unused -o %t.o
Use a generic ELF triple to demonstrate that these apply to all ELF OSes.
Also migrate away from the legacy `-target ` spelling.
Sebastian Peryt [Thu, 26 May 2022 21:27:25 +0000 (14:27 -0700)]
[DOC] Improve LangRef description of declare
This patch fixes formatting inside Functions section of declare
by making it consistent with the way how define is written.
Fixes #39844
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D125581
Sebastian Peryt [Thu, 26 May 2022 21:19:13 +0000 (14:19 -0700)]
[DOC] Refactor Functions section in LangRef
This change is a small refactor of Functions section
to update placement of define syntax.
Reviewed By: RKSimon
Differential revision: https://reviews.llvm.org/D125831
Mehdi Amini [Tue, 24 May 2022 00:20:55 +0000 (00:20 +0000)]
Apply clang-tidy fixes for llvm-else-after-return in OpenMPDialect.cpp (NFC)
Mehdi Amini [Tue, 24 May 2022 00:02:05 +0000 (00:02 +0000)]
Apply clang-tidy fixes for performance-unnecessary-value-param in Bufferize.cpp (NFC)
Michael Kruse [Thu, 26 May 2022 20:38:04 +0000 (15:38 -0500)]
[Polly][Test] Fix race condition while printing dot files.
The tests dot-scops.ll and dot-scops-npm.ll both wrote to the same file
scops.func.dot. If they are executed in parallel they will race for
the file. Fix by renaming func to func_npm in dot-scops-npm.ll so this
test writes dot scops.func_npm.dot.
Long-term, we will probably pass a file name (prefix) to the
printer pass such that we can use the guaranteed-unique LIT %t
placeholder in tests.
owenca [Wed, 25 May 2022 07:55:03 +0000 (00:55 -0700)]
[clang-format][NFC] Refactor UnwrappedLineParser::parseBlock()
Differential Revision: https://reviews.llvm.org/D126358
Louis Dionne [Thu, 26 May 2022 20:52:16 +0000 (16:52 -0400)]
[libc++][NFC] Fix whitespace
Sam Clegg [Thu, 26 May 2022 17:32:04 +0000 (10:32 -0700)]
[lld][WebAssembly] Avoid importing/exporting hidden symbols in shared libraries
We have some special handling for weakly defined symbols where we both
import and export them, but this is not needed for hidden symbols which
should never be imported or exported.
See https://github.com/emscripten-core/emscripten/pull/16972
This should also help with:
https://github.com/emscripten-core/emscripten/issues/15487
Differential Revision: https://reviews.llvm.org/D126491
owenca [Thu, 26 May 2022 02:27:54 +0000 (19:27 -0700)]
[clang-format] Fix an invalid code generation in RemoveBracesLLVM
Fixes #55706.
Differential Revision: https://reviews.llvm.org/D126438
Sam Clegg [Thu, 26 May 2022 20:11:20 +0000 (13:11 -0700)]
[lld][WebAssemlby] Check for command line flags with missing arguments
I'm really not sure how this was overlooked when we first ported lld
to Wasm. The upstream code in the ELF backend has these two lines but
for some reason they never make it into the Wasm version.
Differential Revision: https://reviews.llvm.org/D126497
William Huang [Thu, 26 May 2022 20:15:00 +0000 (20:15 +0000)]
[ValueTracking] Added support to deduce PHI Nodes values being a power of 2
Add Value Tracking support to deduce induction variable being a power of 2, allowing urem optimizations
Reviewed By: davidxl
Differential Revision: https://reviews.llvm.org/D126018
Sanjay Patel [Thu, 26 May 2022 18:28:05 +0000 (14:28 -0400)]
[InstCombine] fold mul of signbit directly to X < 0 ? Y : 0
This is effectively NFC (intentionally no test diffs)
because we already have the related fold that converts
the 'and' pattern to select. So this is just an efficiency
improvement.
Sanjay Patel [Thu, 26 May 2022 18:03:51 +0000 (14:03 -0400)]
[InstCombine] add tests for mul with sign-splat operand; NFC
Sanjay Patel [Thu, 26 May 2022 16:21:48 +0000 (12:21 -0400)]
[InstCombine] fold icmp equality with sdiv and SMIN
This extends the fold from D126410 /
3952c905ef08
to allow for the only case where it works with signed
division:
https://alive2.llvm.org/ce/z/k7_ypu
(X s/ Y) == SMIN --> (X == SMIN) && (Y == 1)
(X s/ Y) != SMIN --> (X != SMIN) || (Y != 1)
This is another improvement based on #55695.
Sanjay Patel [Thu, 26 May 2022 16:02:30 +0000 (12:02 -0400)]
[InstCombine] reduce code duplication in icmp+div folds; NFC
Sanjay Patel [Thu, 26 May 2022 15:29:59 +0000 (11:29 -0400)]
[InstCombine] add tests for icmp with sdiv operand; NFC
Adapted from unsigned division tests from:
ea6171c108c47c1ee4863
Craig Topper [Thu, 26 May 2022 19:15:04 +0000 (12:15 -0700)]
[LegalizeTypes] Fix bug in expensive checks verification
With a fix for an expensive checks build failure exposed by new RISC-V tests.
Something about expanding two rotates in type legalization caused a change
in the remapping tables that the expensive checks verifying wasn't expecting.
See comment in the code for how it was fixed.
Tests came from this commit that exposed the bug
[RISCV] Add test cases showing failure to remove mask on rotate amounts.
If the masking AND has multiple users we fail to remove it.
Reviewed By: RKSimon
Differential Revision: https://reviews.llvm.org/D126036
Craig Topper [Thu, 26 May 2022 19:13:37 +0000 (12:13 -0700)]
[RISCV] Simplfy creation of IndexVT in lowerMaskedGather/lowerMaskedScatter. NFC
The scalar element width is not a factor in how ContainerVT is
determined. We don't need to check the relative size of VT and
IndexVT.
LLVM GN Syncbot [Thu, 26 May 2022 20:01:22 +0000 (20:01 +0000)]
[gn build] Port
b58a420ff4f9
Argyrios Kyrtzidis [Thu, 12 May 2022 04:56:29 +0000 (21:56 -0700)]
[Tooling/DependencyScanning & Preprocessor] Refactor dependency scanning to produce pre-lexed preprocessor directive tokens, instead of minimized sources
This is a commit with the following changes:
* Remove `ExcludedPreprocessorDirectiveSkipMapping` and related functionality
Removes `ExcludedPreprocessorDirectiveSkipMapping`; its intended benefit for fast skipping of excluded directived blocks
will be superseded by a follow-up patch in the series that will use dependency scanning lexing for the same purpose.
* Refactor dependency scanning to produce pre-lexed preprocessor directive tokens, instead of minimized sources
Replaces the "source minimization" mechanism with a mechanism that produces lexed dependency directives tokens.
* Make the special lexing for dependency scanning a first-class feature of the `Preprocessor` and `Lexer`
This is bringing the following benefits:
* Full access to the preprocessor state during dependency scanning. E.g. a component can see what includes were taken and where they were located in the actual sources.
* Improved performance for dependency scanning. Measurements with a release+thin-LTO build shows ~ -11% reduction in wall time.
* Opportunity to use dependency scanning lexing to speed-up skipping of excluded conditional blocks during normal preprocessing (as follow-up, not part of this patch).
For normal preprocessing measurements show differences are below the noise level.
Since, after this change, we don't minimize sources and pass them in place of the real sources, `DependencyScanningFilesystem` is not technically necessary, but it has valuable performance benefits for caching file `stat`s along with the results of scanning the sources. So the setup of using the `DependencyScanningFilesystem` during a dependency scan remains.
Differential Revision: https://reviews.llvm.org/D125486
Differential Revision: https://reviews.llvm.org/D125487
Differential Revision: https://reviews.llvm.org/D125488
Argyrios Kyrtzidis [Thu, 12 May 2022 04:21:17 +0000 (21:21 -0700)]
[Tooling/DependencyScanning] Rename refactorings towards transitioning dependency scanning to use pre-lexed preprocessor directive tokens
This is first of a series of patches for making the special lexing for dependency scanning a first-class feature of the `Preprocessor` and `Lexer`.
This patch only includes NFC renaming changes to make reviewing of the functionality changing parts easier.
Differential Revision: https://reviews.llvm.org/D125484
python3kgae [Tue, 24 May 2022 18:04:51 +0000 (11:04 -0700)]
[DirectX] Update test for dxil-dis.
Update metadata index which changed for dx.valver.
Reviewed By: beanz
Differential Revision: https://reviews.llvm.org/D126311
Louis Dionne [Thu, 26 May 2022 14:19:25 +0000 (10:19 -0400)]
[libunwind] Tidy-up the testing configuration for libunwind
Start testing Apple backdeployment with older libunwinds, and stop
explicitly specifying the libunwind testing config, since it is
already selected correctly by default.
Differential Revision: https://reviews.llvm.org/D126470
Stella Stamenova [Thu, 26 May 2022 19:16:24 +0000 (12:16 -0700)]
[lldb] Fix enums-layout test on Windows
The test was broken by: https://reviews.llvm.org/D125604
Will Hawkins [Thu, 26 May 2022 19:14:22 +0000 (12:14 -0700)]
[lldb] Fix broken bad-address-breakpoint test
After changing the "fallback" behavior when a user sets a breakpoint
without specifying a module the bad-address-breakpoint test case failed
incorrectly. This patch updates that test case in order to more
thoroughly discover an illegal address and use that as the means for
testing whether a breakpoint set at an illegal address fails to resolve.
Differential revision: https://reviews.llvm.org/D126109
Joel E. Denny [Thu, 26 May 2022 18:59:16 +0000 (14:59 -0400)]
[OpenMP] Extend omp teams to permit nested omp atomic
OpenMP 5.2, sec. 10.2 "teams Construct", p. 232, L9-12 restricts what
regions can be strictly nested within a `teams` construct. This patch
relaxes Clang's enforcement of this restriction in the case of nested
`atomic` constructs unless `-fno-openmp-extensions` is specified.
Cases like the following then seem to work fine with no additional
implementation changes:
```
#pragma omp target teams map(tofrom:x)
#pragma omp atomic update
x++;
```
Reviewed By: ABataev
Differential Revision: https://reviews.llvm.org/D126323
Louis Dionne [Thu, 26 May 2022 18:58:24 +0000 (14:58 -0400)]
[libc++] Time tests during CI
Mircea Trofin [Thu, 26 May 2022 18:48:29 +0000 (11:48 -0700)]
Fix break introduced by D124306
argparse.BooleanOptionalAction is not supported until python 3.9.
Aaron Ballman [Thu, 26 May 2022 18:43:28 +0000 (14:43 -0400)]
Roll back use of #warning for header deprecations
e5ccd668019888de2704ae670da88a7be8cf7e0f and
5029dce492b3cf3ac191eda0b5bf268c3acac2e0 added deprecation warnings to
the <stdbool.h> and <stdnoreturn.h> headers, respectively, because the
headers are deprecated in C2x.
However, there are system headers that include these headers
unconditionally, and #warning diagnostics within system headers are
shown to users instead of suppressed, which means these deprecation
warnings are being triggered in circumstances that users have no
control over except to disable all the warnings through the
_CLANG_DISABLE_CRT_DEPRECATION_WARNINGS macro or other means.
This removes the problematic #warning uses until we find a more
palatable solution.
Adrian Tong [Wed, 4 May 2022 23:00:42 +0000 (23:00 +0000)]
Give option to use isCopyInstr to determine which MI is
treated as Copy instruction in MCP.
This is then used in AArch64 to remove copy instructions after taildup
ran in machine block placement
Differential Revision: https://reviews.llvm.org/D125335
Jason Molenda [Thu, 26 May 2022 18:24:13 +0000 (11:24 -0700)]
One further tweak for realpathing filepath to match dyld
I missed one place I need to realpath the build artifact path,
to make it match the path we get back from dyld.
Zongwei Lan [Thu, 26 May 2022 18:22:41 +0000 (11:22 -0700)]
[Target] use getSubtarget<> instead of static_cast<>(getSubtarget())
Differential Revision: https://reviews.llvm.org/D125391
Jason Molenda [Thu, 26 May 2022 18:11:43 +0000 (11:11 -0700)]
Check that a FileSpec has a Directory component before using
A follow on to my patch for https://reviews.llvm.org/D126435
hit by an x86_64 linux bot; I assumed that a FileSpec had a
directory component and checked if the first character was a
'~'. This was not a valid assumption.
Jason Molenda [Thu, 26 May 2022 07:30:12 +0000 (00:30 -0700)]
Defer source path remap tilde expansion until source file use
When reading source path remappings out of a dSYM, lldb currently
does tilde expansion -- expanding the tilde-username and checking
that the destination pathname exists, for each dSYM with the path
remappings. This cost happens during lldb's initial process launch
/ load, an especially perf-sensitive time. Inside Apple, we have
dSYMs with source path remappings pointing to NFS directories where
these extra stats for every dSYM can be very expensive if the network
is slow.
This patch instead keeps the source path mapping in the original
tilde-username terms and does the tilde expansion when we need
to read a specific source file from one of the modules. We'll
be stat'ing all of those inodes to load the source file anyway,
so the fact that we do the tilde expansion on every source file
we load, it doesn't cost us significantly.
Differential Revision: https://reviews.llvm.org/D126435
rdar://
77091379
(cherry picked from commit
c274b6e5830ea88d3f55d6dc1d2b99e38cf6595e)
Bruno Cardoso Lopes [Tue, 24 May 2022 23:56:22 +0000 (16:56 -0700)]
[Clang][CoverageMapping] Fix switch counter codegen compile time explosion
C++ generated code with huge amount of switch cases chokes badly while emitting
coverage mapping, in our specific testcase (~72k cases), it won't stop after hours.
After this change, the frontend job now finishes in 4.5s and shrinks down `@__covrec_`
by 288k when compared to disabling simplification altogether.
There's probably no good way to create a testcase for this, but it's easy to
reproduce, just add thousands of cases in the below switch, and build with
`-fprofile-instr-generate -fcoverage-mapping`.
```
enum type : int {
FEATURE_INVALID = 0,
FEATURE_A = 1,
...
};
const char *to_string(type e) {
switch (e) {
case type::FEATURE_INVALID: return "FEATURE_INVALID";
case type::FEATURE_A: return "FEATURE_A";}
...
}
```
Differential Revision: https://reviews.llvm.org/D126345
Mircea Trofin [Fri, 22 Apr 2022 19:49:15 +0000 (12:49 -0700)]
[UpdateTestChecks] Auto-generate stub bodies for unused prefixes
This is scoped to autogenerated tests.
The goal is to support having each RUN line specify a list of
check-prefixes where one can specify potentially redundant prefixes. For example,
for X86, if one specified prefixes for both AVX1 and AVX2, and the codegen happened to
match today, one of the prefixes would be used and the onther one not.
If the unused prefix were dropped, and later, codegen differences were
introduced, one would have to go figure out where to add what prefix
(paraphrasing
https://lists.llvm.org/pipermail/llvm-dev/2021-February/148326.html)
To avoid getting errors due to unused prefixes, whole directories can be
opted out (as discussed on that thread), but that means that tests that
aren't autogenerated in such directories could have undetected unused
prefix bugs.
This patch proposes an alternative that both avoids the above, dir-level
optout, and supports the main autogen scenario discussed first. The autogen
tool appends at the end of the test file the list of unused prefixes,
together with a note explaining that is the case. Each prefix is set up
to always pass.
This way, unexpected unused prefixes are easily discoverable, and
expected cases "just work".
Differential Revision: https://reviews.llvm.org/D124306
Florian Hahn [Thu, 26 May 2022 17:16:37 +0000 (18:16 +0100)]
[SCEV] Collect conditions from assumes same way as for branches.
Also collect conditions from assume up-front in applyLoopGuards.
This allows re-using the logic to handle logical ANDs as assume
conditions.
It should should pave the road for a fix for #55645.
David Penry [Mon, 28 Mar 2022 19:18:37 +0000 (12:18 -0700)]
[ARM] Recognize t2LoopEnd for software pipelining
- Add t2LoopEnd to TargetInstrInfo::analyzeBranch and
related functions. As there are many side effects of
analyzing a branch, only do so if software pipelining
is enabled to maintain previous behavior when pipelining
is not desired.
- Make sure that t2LoopEndDec is immediately followed by
a t2B when it is synthesized from a t2LoopEnd. This is
done because the t2LoopEnd might have acquired a
fall-through path, but IfConversion assumes that
fall-through are only possible on analyzable branches.
Reviewed By: dmgreen
Differential Revision: https://reviews.llvm.org/D126322
Mike Rice [Wed, 25 May 2022 22:02:26 +0000 (15:02 -0700)]
[OpenMP] Use the align clause value from 'omp allocate' for globals
Refactor the code that handles the align clause of 'omp allocate' so
it can be used with globals as well as local variables.
Differential Revision: https://reviews.llvm.org/D126426
Owen Anderson [Thu, 26 May 2022 16:50:36 +0000 (09:50 -0700)]
Revert "Replace the custom linked list in LeaderTableEntry with TinyPtrVector."
This reverts commit
1e9114984490b83d4665f12a11f84c83f50ca8f0.
Pending further discussion.
Philip Reames [Thu, 26 May 2022 15:55:47 +0000 (08:55 -0700)]
[RISCV] reorganize getFrameIndexReference to reduce code duplication [nfc]
This change reorganizes the majority of frame index resolution into a two strep process.
Step 1 - Select which base register we're going to use.
Step 2 - Compute the offset from that base register.
The key point is that this allows us to share the step 2 logic for the SP case. This reduces the code duplication, and (I think) makes the code much easier to follow.
I also went ahead and added assertions into phase 2 to catch errors where we select an illegal base pointer. In general, we can't index from a base register to a stack location if that requires crossing a variable and unknown region. In practice, we have two such cases: dynamic stack realign and var sized objects. Note that crossing the scalable region is fine since while variable, it's a known variability which can be expressed in the offset.
Differential Revision: https://reviews.llvm.org/D126403
Shoaib Meenai [Thu, 26 May 2022 16:34:18 +0000 (09:34 -0700)]
Revert "[runtimes] Detect changes to Tests.cmake"
This reverts commit
ec10ac750a8ad96983d85263323635f3dabe92fd.
See https://discourse.llvm.org/t/cmake-regeneration-is-broken/62788.
This change caused Ninja's CMake regeneration to depend on the build,
which prevented CMake regeneration from functioning properly and caused
spurious build failures on incremental builds when a CMake change
occurred.