Changes

[AMDGPU] Control num waves per EU for implicit work-group size

Summary:
If amdgpu-flat-work-group-size is not specified in LLVM IR, the backend
uses default value of 1024. For this, minimum waves per EU should be 4.
However, backend is still setting minimum value to 1 instead of calculated
value. This is not observed normally as frontend always provide
amdgpu-flat-work-group-size attribute.

Reviewers: rampitec, b-sumner, sameerds, msearles

Reviewed By: rampitec

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

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D81991

[lldb/ObjCRuntime] Implement support for small method lists

On macOS 11 (and other aligned Apple OSs), the Objective-C runtime
has a new optimization which saves memory by making the method
lists smaller.
This patch adds support for this new method list encoding (while
also keeping backward compatibility). This is implicitely covered
by some existing Objective-C tests.

[lldb/ObjC] Add support for direct selector references

On macOS 11 (and other aligned OSs), the shared cache method
lists get an additional optimization which removes one level
of indirection to get to the selector.
This patch supports this new optimization. Both codepaths are
covered byt the existing Objective-C tests.

[lldb/build.py] Always pass an SDK to the compiler on Darwin

On macOS 11, system libraries which are part of the shared cache
are not present on the filesystem anymore. This causes issues
with build.py, because it fails to link binaries with libSystem
or libc++.

The real issue is that build.py was not passing an SDK to the
compiler. The script accepts an argument for the SDK, but it
is currently unused. This patch just threads the SDK through
to the compile and link steps and this fixes a bunch of Shell
test failures on very recent macOS builds.

lld/MachO: Remove a useless temporary

ld64.lld: Add janky support for tbd files

With this, a simple hello world links against libSystem.tbd and the
old ld64.lld linker kind of works again with newer SDKs.

The motivation here is to have an arm64 cross linker that's good
enough to be able to run simple configure link checks on non-mac
systems for generating config.h files. Once -flavor darwinnew can
link arm64, we'll switch to that.

[mlir][OpFormatGen] Add support for resolving variadic types from non-variadic

This enables better support for traits such as SameOperandsAndResultType, and other situations in which a variadic operand may be resolved from a non-variadic.

Differential Revision: https://reviews.llvm.org/D83011

Add parenthesized expression to SyntaxTree

Subscribers: cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D82960

This patch adds basic debug info support with basic block sections.

This patch uses ranges for debug information when a function contains basic block sections rather than using [lowpc, highpc]. This is also the first in a series of patches for debug info and does not contain the support for linker relaxation. That will be done as a follow up patch.

Differential Revision: https://reviews.llvm.org/D78851

[clang] Re-add deleted forward declaration.

[X86] Enable multibyte NOPs in 64-bit mode for padding/alignment.

The default CPU used by llvm-mc doesn't have the NOPL feature, but
if we know we're compiling in 64-bit mode we should be able to
use nopl.

[lldb][NFC] Fix a variable name in ClangDiagnosticManagerAdapter

Revert "[clang-tidy] For `run-clang-tidy.py` escape the paths that are used for analysis."

This reverts commit 068fa35746637fde29355a43d17d554a92b32cdf.

Based on a regression reported in https://bugs.llvm.org/show_bug.cgi?id=46536

[CodeGen] Fix warnings in getCopyToPartsVector

Whilst trying to assemble the following test:

  clang/test/CodeGen/aarch64-sve-intrinsics/acle_sve_set2.c

I discovered we were hitting some warnings about possible invalid
calls to getVectorNumElements() in getCopyToPartsVector(). I've
tried to fix these by using ElementCount types where possible and
I've made the assumption that we don't support using a fixed width
vector to copy parts of a scalable vector, and vice versa. Looking
at how the copy is implemented I think that's the right thing for
now.

Differential Revision: https://reviews.llvm.org/D82744

Fix missing build dependency on omp_gen.

Summary:
`include/llvm/Frontend/OpenMP/CMakeLists.txt` creates a new target
called `omp_gen` which builds the generated include file `OMP.h.inc`.
This target must therefore be a dependency of every compilation step
whose transitive #include dependencies contain `OMP.h.inc`, or else
it's possible for builds to fail if Ninja (or make or whatever)
schedules that compilation step before building `OMP.h.inc` at all.

A few of those dependencies are currently missing, which leads to
intermittent build failures, depending on the order that Ninja (or
whatever) happens to schedule its commands. As far as I can see,
compiles in `clang/lib/CodeGen`, `clang/lib/Frontend`, and
`clang/examples` all depend transitivily on `OMP.h.inc` (usually via
`clang/AST/AST.h`), but don't have the formal dependency in the ninja
graph.

Adding `omp_gen` to the dependencies of `clang-tablegen-targets` seems
to be the way to get the missing dependency into the `clang/examples`
subdirectory. This also fixes the other two clang subdirectories, as
far as I can see.

Reviewers: clementval, thakis, chandlerc, jdoerfert

Reviewed By: clementval

Subscribers: cfe-commits, jdenny, mgorny, sstefan1, llvm-commits

Tags: #llvm, #clang

Differential Revision: https://reviews.llvm.org/D82659

[ARM] Rearrange SizeReduction when using -Oz

Move the Thumb2SizeReduce pass to before IfConversion when optimising
for minimal code size.

Running the Thumb2SizeReduction pass before IfConversionallows T1
instructions to propagate to the final output, rather than the
ifConverter modifying T2 instructions and preventing them from being
reduced later.

This change does introduce a regression regarding execution time, so
it's only applied when optimising for size.

Running the LLVM Test Suite with this change produces a geomean
difference of -0.1% for the size..text metric.

Differential Revision: https://reviews.llvm.org/D82439

[NFC] Fix typo in triples from unkown to unknown

[lldb][NFC] Don't pass around passthrough from ClangDiagnosticManagerAdapter

The passthrough DiagnosticConsumer is an implementation detail of
ClangDiagnosticManagerAdapter and we can just hide it behind the normal
DiagnosticConsumer interface that ClangDiagnosticManagerAdapter is supposed
to implement.

[SVE] ACLE: Fix builtins for svdup_lane_bf16 and svcvtnt_bf16_f32_x

bfloat16 variants of svdup_lane were missing, and svcvtnt_bf16_x
was implemented incorrectly (it takes an operand for the inactive
lanes)

Reviewers: fpetrogalli, efriedma

Reviewed By: fpetrogalli

Tags: #clang

Differential Revision: https://reviews.llvm.org/D82908

[AArch64][SVE] Add unpred load/store patterns for bf16 types

Reviewers: kmclaughlin, c-rhodes, efriedma

Reviewed By: efriedma

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82909

[CodeGen][SVE] Don't drop scalable flag in DAGCombiner::visitEXTRACT_SUBVECTOR

There was a rogue 'assert' in AArch64ISelLowering for the tuple.get intrinsics,
that shouldn't really have been there (I suspect this was a remnant from when
we expected the wider vector always to have come from a vector CONCAT).

When I tried to create a more minimal reproducer, I found a bug in
DAGCombiner where it drops the scalable flag when trying to fold:

      extract_subv (bitcast X), Index --> bitcast (extract_subv X, Index')

This patch fixes both issues.

Reviewers: david-arm, efriedma, spatel

Reviewed By: efriedma

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82910

[AMDGPU] Fix formatting in MIR tests

DSE: fix builtin function recognition to take decl into account

[Support][Windows] Prevent 2s delay when renaming a file that does not exist

Differential Revision: https://reviews.llvm.org/D82542

[SVE] Add warnings checks in four more LLVM SVE tests

I have added CHECK lines to the following tests:

  llvm/test/CodeGen/AArch64/sve-breakdown-scalable-vectortype.ll
  llvm/test/CodeGen/AArch64/sve-calling-convention-tuple-types.ll
  llvm/test/CodeGen/AArch64/sve-intrinsics-create-tuple.ll
  llvm/test/CodeGen/AArch64/sve-intrinsics-loads.ll

since they are now free of warnings related to invalid use of
EVT::getVectorNumElements() and VectorType::getNumElements().

Differential Revision: https://reviews.llvm.org/D82957

[VE] Rename VE toolchain source files

Summary:
Rename VE.cpp and VE.h to VEToolchain.cpp and VEToolchain.h respectively
in order to avoid link warning message.  Linker warns that VE.cpp.o and
Arch/VE.cpp.o have the same name.

Reviewers: simoll, k-ishizaka

Reviewed By: simoll

Subscribers: mgorny, cfe-commits

Tags: #llvm, #ve, #clang

Differential Revision: https://reviews.llvm.org/D82968

[gn build] Port 804d9687443

[BasicAA] New basic-aa-recphi test. NFC

[ScalarEvolution] createSCEV(): recognize `udiv`/`urem` disguised as an `sdiv`/`srem`

Summary:
While InstCombine trivially converts that `srem` into a `urem`,
it might happen later than wanted, in particular i'd like
for that to happen on  https://godbolt.org/z/bwuEmJ test case
early in pipeline, before first instcombine run, just before `-mem2reg`.

SCEV should recognize this case natively.

Reviewers: mkazantsev, efriedma, nikic, reames

Reviewed By: efriedma

Subscribers: clementval, hiraditya, javed.absar, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82721

Regenerate llvm/test/CodeGen/X86/optimize-max-0.ll

It surprizingly appears to be affected by the last SCEV patch

[LV] Enable the LoopVectorizer to create pointer inductions

This patch enables the LoopVectorizer to build a phi of pointer
type and provide the vector loads and stores with vector type
getelementptrs built from the pointer induction variable, which
produces much less instructions than the previous approach of
creating scalar getelementpointers and glue them together to a
vector.

Differential Revision: https://reviews.llvm.org/D81267

Fix some typos (unkown -> unknown); NFC

Regenerate apint-shift tests and replace %tmp variable names to silence update_test_checks warnings

[InstCombine] Add some (vXi1 trunc(lshr(x,c))) -> icmp_eq(and(x,c')) tests for non-uniform vectors

As noticed on PR46531

[AArch64][SVE] Add reg+imm addressing mode for unpredicated stores

Reviewers: sdesmalen, efriedma, david-arm

Reviewed By: efriedma

Subscribers: tschuett, kristof.beyls, hiraditya, rkruppe, psnobl, danielkiss, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82985

[Alignment][NFC] VectorLayout now uses Align internally

By rewritting `ScalarizerVisitor::getVectorLayout` in such a way it returns `VectorLayout` (or `None`) it becomes obvious that `VectorLayout::VecAlign` cannot be `0`.

This patch is part of a series to introduce an Alignment type.
See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html
See this patch for the introduction of the type: https://reviews.llvm.org/D64790

Differential Revision: https://reviews.llvm.org/D82981

[Alignment][NFC] Transition and simplify calls to DL::getABITypeAlignment

This patch is part of a series to introduce an Alignment type.
See this thread for context: http://lists.llvm.org/pipermail/llvm-dev/2019-July/133851.html
See this patch for the introduction of the type: https://reviews.llvm.org/D64790

Differential Revision: https://reviews.llvm.org/D82977

[NFC][clang] Add missing VALIDATE_DIAG_SIZE()

Originally when libCrossTU was introduced in commit
e350b0a19629c4cce87f28913b3137f4c7015de3, the macro which thus had all
diagnostic kinds covered was not added.

[clangd] Switch FindSymbolsTests to use TestTU

Summary:
This gets rid of the dependency on ClangdServer and a bunch of extra
infrastructure coming with it. Also enables us to clear SyncAPI, as it was the
sole user of runWorkspace/DocumentSymbols.

Reviewers: sammccall

Subscribers: ilya-biryukov, MaskRay, jkorous, arphaman, usaxena95, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D82944

[analyzer][StdLibraryFunctionsChecker] Add POSIX file handling functions

Adding file handling functions from the POSIX standard (2017).
A new checker option is introduced to enable them.
In follow-up patches I am going to upstream networking, pthread, and other
groups of POSIX functions.

Differential Revision: https://reviews.llvm.org/D82288

[lldb] Skip TestLimitDebugInfo on windows

The test does not work on windows, because clang will emit full type
information for __declspec(dllexport) types even under
-flimit-debug-info. __declspec(dllexport) is needed to be able to use
the type across shared library boundaries on windows, which makes this a
pretty good heuristic, but defeats the purpose of this test.

I am going to create (in another patch) an basic assembly test, so that
the relevant code gets at least some coverage on windows hosts.

This also reverts commit 1276855f2b4485ec312b379c1b8eaf5510d9b157, which
added the __declspec annotations -- they are not necessary anymore, and
they needlessly complicate the test.

[clang] Fix a null-NSS-access crash in DependentNameType.

The DependentNameType must have a non-null NSS. This property could be
violated during typo correction.

Differential Revision: https://reviews.llvm.org/D82738

[BasicAA] Fix recursive phi MustAlias calculations

With the option -basic-aa-recphi we can detect recursive phis that loop
through constant geps, which allows us to detect more no-alias case for
pointer IV's. If the other phi operand and the other alias value are
MustAlias though, we cannot presume that every element in the loop is
also MustAlias. We need to instead be conservative and return MayAlias.

Differential Revision: https://reviews.llvm.org/D82987

[flang] Add changes to codegen to convert it to tablegen passes.

This upstreams changes to codegen to convert the passes to the new
tablegen pass support from MLIR.

Differential revision: https://reviews.llvm.org/D83018

[flang] Add inliner pass.

This adds a minimalist inliner implementation. Along with the inliner, a
minimum number of support files are also included. These will pave the
way for future diffs to add more transformation passes to flang. A
future diff will add the inline test, which cannot be run successfully
quite yet as some components have not yet been upstreamed.

Differential revision:

[clang][Serialization] Don't duplicate the body of LambdaExpr during deserialization

05843dc6ab97a00cbde7aa4f08bf3692eb83109d changed the serialization of the body
of LambdaExpr to avoid a mutation in LambdaExpr::getBody and to avoid a missing
body in LambdaExpr::children.

Unfortunately this replaced one bug by another: we are now duplicating the body
during deserialization; that is after deserialization the identity:

E->getBody() == E->getCallOperator()->getBody() does not hold.

Fix that by instead lazily loading the body from the call operator when needed.

Differential Revision: https://reviews.llvm.org/D83009

Reviewed By: martong, aaron.ballman, vabridgers

[AArch64][SVE] Put zeroing pseudos and patterns under flag.

This patch puts the _ZERO pseudos and corresponding patterns
under the predicate 'UseExperimentalZeroingPseudos', so that they
can be enabled/disabled through compile flags.

This is done because the zeroing pseudos use MOVPRFX to do merging of
the inactive lanes, but it depends on the uarch whether this operation
is actually merged with the destructive operation. If not, it may be
more profitable to use a SELECT and to give the compiler the freedom to
schedule these instructions as normal, rather than keeping them bundled
together. Additionally, this feature is not yet fully implemented and
there are still known bugs (see D80410) that need to be resolved before
the 'experimental' can be dropped from the name.

Reviewers: paulwalker-arm, cameron.mcinally, efriedma

Reviewed By: paulwalker-arm

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82780

Regenerate sext/trunc tests and replace %tmp variable names to silence update_test_checks warnings

call ::pthread_detach on llvm_execute_on_thread_impl

Fixes all TSAN bugs in clangd

Reviewed By: sammccall

Differential Revision: https://reviews.llvm.org/D83039

[lldb] Add a host-independent test for handling -flimit-debug-info

This complements the existing TestLimitDebugInfo.py, which tests this
scenario more comprehensively, but is not able to run on all hosts.
Specifically, it's hard to trigger this code from windows because clang
tries hard to ensure that debug info for types marked with
__declspec(dllexport) is emitted even under -flimit-debug-info (and
dllexport is needed to use a type across shared libraries).

This assembly-based test serves two purposes:
- it tests that -flimit-debug-info code path works for windows binaries
  (even though the aforementioned feature means its less likely to be
  used there)
- it gives basic test coverage for the -flimit-debug-info handling code
  when running the test suite on windows hosts.

[ASTMatchers] Enhanced support for matchers taking Regex arguments

Added new Macros `AST(_POLYMORPHIC)_MATCHER_REGEX(_OVERLOAD)` that define a matchers that take a regular expression string and optionally regular expression flags. This lets users match against nodes while ignoring the case without having to manually use `[Aa]` or `[A-Fa-f]` in their regex. The other point this addresses is in the current state, matchers that use regular expressions have to compile them for each node they try to match on, Now the regular expression is compiled once when you define the matcher and used for every node that it tries to match against. If there is an error while compiling the regular expression an error will be logged to stderr showing the bad regex string and the reason it couldn't be compiled. The old behaviour of this was down to the Matcher implementation and some would assert, whereas others just would never match. Support for this has been added to the documentation script as well. Support for this has been added to dynamic matchers ensuring functionality is the same between the 2 use cases.

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D82706

Fix missing build dependencies on omp_gen

Differential Revision: https://reviews.llvm.org/D83003

[AMDGPU][CODEGEN] Added support of new inline assembler constraints

Added support for constraints 'I', 'J', 'B', 'C', 'DA', 'DB'.

See https://gcc.gnu.org/onlinedocs/gcc/Machine-Constraints.html#Machine-Constraints.

Reviewers: arsenm, rampitec

Differential Revision: https://reviews.llvm.org/D81651

[MLIR] Exact integer emptiness checks for FlatAffineConstraints

This patch adds the capability to perform exact integer emptiness checks for FlatAffineConstraints using the General Basis Reduction algorithm (GBR). Previously, only a heuristic was available for emptiness checks, which was not guaranteed to always give a conclusive result.

This patch adds a `Simplex` class, which can be constructed using a `FlatAffineConstraints`, and can find an integer sample point (if one exists) using the GBR algorithm. Additionally, it adds two classes `Matrix` and `Fraction`, which are used by `Simplex`.

The integer emptiness check functionality can be accessed through the new `FlatAffineConstraints::isIntegerEmpty()` function, which runs the existing heuristic first and, if that proves to be inconclusive, runs the GBR algorithm to produce a conclusive result.

Differential Revision: https://reviews.llvm.org/D80860

[ASTImporter] Add unittest case for friend decl import

Summary:
This change adds a matching test case for the recent bug fix to
VisitFriendDecl in ASTImporterLookup.cpp.

See https://reviews.llvm.org/D82882 for details.

Reviewers: martong, a.sidorin, shafik

Reviewed By: martong

Subscribers: rnkovacs, teemperor, cfe-commits, dkrupp

Tags: #clang

Differential Revision: https://reviews.llvm.org/D83006

Preserve GlobalsAA analysis result in LowerConstantIntrinsics

LowerConstantIntrinsics fails to preserve the analysis result of
GlobalsAA. Not preserving the analysis might affect benchmark
performance. This change fixes this issue.

Patch by Ryan Santhiraraja <rsanthir@quicinc.com>

Reviewers: fpetrogalli, joerg, fhahn

Reviewed By: fhahn

Differential Revision: https://reviews.llvm.org/D82342

[ELF][test] Add some additional .eh_frame/.eh_frame_hdr testing

This patch adds a few extra cases to the existing testing for eh_frame
and eh_frame_hdr behaviour in LLD. They all come from a private
testsuite we are trying to migrate to lit.

Reviewed by: grimar, MaskRay

Differential Revision: https://reviews.llvm.org/D82852

[InstCombine] Add some (vXi1 trunc(lshr(x,c))) -> icmp_eq(and(x,c')) tests for vectors with undef elements

Suggested on D83035

[InstCombine] Add (vXi1 trunc(lshr(x,c))) -> icmp_eq(and(x,c')) support for non-uniform vectors

As noted on PR46531, we were only performing this transform on uniform vectors as we were using the m_APInt pattern matcher to extract the shift amount.

Differential Revision: https://reviews.llvm.org/D83035

[AArch64][SVE] NFC: Rename isOrig -> isReverseInstr

This is a non-functional to clarify some of the terminology in the
AArch64SVEInstrInfo/SVEInstrFormats.td files around the tables
for mapping an instruction to it's reverse instruction counter part,
and vice versa. e.g. DIV -> DIVR and DIVR -> DIV.

Reviewers: paulwalker-arm, cameron.mcinally, rengolin, efriedma

Reviewed By: paulwalker-arm, efriedma

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82979

[lldb] Fix type conversion in the Scalar getters

Summary:
The Scalar class claims to follow the C type conversion rules. This is
true for the Promote function, but it is not true for the implicit
conversions done in the getter methods.

These functions had a subtle bug: when extending the type, they used the
signedness of the *target* type in order to determine whether to do
sign-extension or zero-extension. This is not how things work in C,
which uses the signedness of the *source* type. I.e., C does
(sign-)extension before it does signed->unsigned conversion, and not the
other way around.

This means that: (unsigned long)(int)-1
      is equal to (unsigned long)0xffffffffffffffff
      and not (unsigned long)0x00000000ffffffff

Unsurprisingly, we have accumulated code which depended on this
inconsistent behavior. It mainly manifested itself as code calling
"ULongLong/SLongLong" as a way to get the value of the Scalar object in
a primitive type that is "large enough". Previously, the ULongLong
conversion did not do sign-extension, but now it does.

This patch makes the Scalar getters consistent with the declared
semantics, and fixes the couple of call sites that were using it
incorrectly.

Reviewers: teemperor, JDevlieghere

Subscribers: lldb-commits

Tags: #lldb

Differential Revision: https://reviews.llvm.org/D82772

[InstCombine] Add some sext/trunc tests to show missing support for non-uniform vectors

[MLIR][SPIRVToLLVM] Implementation of spv.BitFieldInsert pattern

This patch introduces conversion pattern for `spv.BitFiledInsert` op,
as well as some utility functions to facilitate code reading.
Since `spv.BitFiledInsert` may take both vector and integer operands,
this case was specifically handled by broadcasting values (`count`
and `offset` here) to vectors. Moreover, the types had to be converted
to same bitwidth in order to conform with LLVM dialect rules.
This was done with `zext` when extending (Note that `count` and
`offset` are treated as unsigned) and `trunc` in the opposite case.
For the latter one, truncation is safe since the op is defined only when
`count`/`offset`/their sum is less than the bitwidth of the result.
This introduces a natural bound of the value of 64, which can be
expressed as `i8`.

Reviewed By: antiagainst, ftynse

Differential Revision: https://reviews.llvm.org/D82639

[MLIR][SPIRVToLLVM] SPIR-V function call conversion pattern

Added conversion pattern for SPIR-V `FunctionCallOp`. Based on
specification, it returns no results or a single result, so
can be mapped directly to LLVM dialect's `llvm.call`.

Reviewed By: antiagainst, ftynse

Differential Revision: https://reviews.llvm.org/D83030

[MLIR][SPIRV] Support two memory access attributes in OpCopyMemory.

This commit augments spv.CopyMemory's implementation to support 2 memory
access operands. Hence, more closely following the spec. The following
changes are introduces:

- Customize logic for spv.CopyMemory serialization and deserialization.
- Add 2 additional attributes for source memory access operand.

Reviewed By: antiagainst

Differential Revision: https://reviews.llvm.org/D82710

[PowerPC] Remove undefs from splat input when changing shuffle mask

As of 1fed131660b2c5d3ea7007e273a7a5da80699445, we have code that
changes shuffle masks so that we can put the shuffle in a canonical
form that can be matched to a single instruction. However, it
does not properly account for undef elements in the BUILD_VECTOR
that is the RHS splat so we can end up with undefs where they
shouldn't be. This patch converts the splat input with undefs to
one without.

[LLD] Add required dependency after shared libs break due to ba5087f13025

The dependency on TextAPI was not added and is required for shared
libs builds.

[OpenMP][CMake] Fix version detection of testing compiler

When configuring in-tree, the correct names are LLVM_VERSION_MAJOR
and LLVM_VERSION_MINOR. This has been wrong since the code was added
in commits fc473dee98 and 821649229e.

[OpenMPOpt][Fix] Remove double initialization of omp::types.

[NewPM][LSR] Rename strength-reduce -> loop-reduce

The legacy pass was called "loop-reduce".

This lowers the number of check-llvm failures under NPM by 83.

Reviewed By: ychen

Differential Revision: https://reviews.llvm.org/D82925

[gn build] make building on an arm mac work

Currently requires `llvm_targets_to_build = [ "X86", "AArch64" ]`:
building just the host arch (i.e. aarch64) causes some linker errors.

[MLIR][SPIRVToLLVM] Convert spv.constant scalars and vectors

This patch introduces conversion pattern for `spv.constant` with scalar
and vector types. There is a special case when the constant value is a
signed/unsigned integer (vector of integers). Since LLVM dialect does not
have signedness semantics, the types had to be converted to signless ints.

Reviewed By: antiagainst

Differential Revision: https://reviews.llvm.org/D82936

[NewPM] Add -basic-aa to pr33196.ll

The legacy pass manager implicitly adds BasicAA, but the new PM does
not. This causes pr33196.ll to fail under NPM.

There are almost certainly lots of other failures like this, wanted to
get some input on if adding -basic-aa to tests makes sense at scale.

Reviewed By: fhahn

Differential Revision: https://reviews.llvm.org/D82915

[clang][NFC] Store a pointer to the ASTContext in ASTDumper and TextNodeDumper

In general there is no way to get to the ASTContext from most AST nodes
(Decls are one of the exception). This will be a problem when implementing
the rest of APValue::dump since we need the ASTContext to dump some kinds of
APValues.

The ASTContext* in ASTDumper and TextNodeDumper is not always
non-null. This is because we still want to be able to use the various
dump() functions in a debugger.

No functional changes intended.

[DebugInfo] Fix LineTest byteswap for cross-targeting builds

Summary:
The byte swap fix for big endian hosts in 9782c922cb21 (for D81570)
swaps based on the host endianess,  but for cross-targeting builds (i.e.
big endian host targeting little endian) the host-endianess won't
necessarily match the generated DWARF. This change updates the test
to use symmetrical constants so the results aren't endian dependent.

Reviewers: jhenderson, hubert.reinterpretcast, stevewan, ikudrin

Reviewed By: ikudrin

Subscribers: ikudrin, aprantl, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D82827

Revert "[clang][NFC] Store a pointer to the ASTContext in ASTDumper and TextNodeDumper"

This reverts commit aa7fd905e4e1bc510448431da9310e8cf5197523.

I missed some dump() functions.

[flang] Fix bug determining alternate return

The arguments have been moved out of the analyzer so we can't get the
expected number there. Instead use the argument count from the newly
built callee.

Differential Revision: https://reviews.llvm.org/D83063

[flang] Clean up binary dependences of runtime libraries

There were dependences upon LLVM libraries in the Fortran
runtime support library binaries due to some indirect #includes
of llvm/Support/raw_ostream.h, which caused some kind of internal
ABI version consistency checking to get pulled in.  Fixed by
cleaning up some includes.

Reviewed By: tskeith, PeteSteinfeld, sscalpone

Differential Revision: https://reviews.llvm.org/D83060

ld64.lld: Make janky support for tbd files actually work sometimes

Also fix a bug in the test input that made the test miss this issue.

[PowerPC]Implement Vector Permute Extended Builtin

Implements vector permute builtin: vec_permx()

Differential Revision: https://reviews.llvm.org/D82869

Revert "[MLIR][SPIRV] Support two memory access attributes in OpCopyMemory."

This reverts commit ef2f46e1f6a63040734c48ed53893298df14b6fa, which
likely triggers a compiler internal error for MSVC.

Differential Revision: https://reviews.llvm.org/D83075

[X86] Add test cases for v32i8 rotate with min-legal-vector-width=256

We currently don't mark ROTL as custom when avx512bw is enabled
under the assumption we'll be able to promote the shifts in the
rotate idiom. But if we don't have 512-bit registers enabled we
can't promote.

[X86] Modify the conditions for when we stop making v16i8/v32i8 rotate Custom based on having avx512 features.

The comments here indicate that we prefer to promote the shifts
instead of allowing rotate to be pattern matched. But we weren't
taking into account whether 512-bit registers are enabled or
whethever we have vpsllvw/vpsrlvw instructions.

splatvar_rotate_v32i8 is a slight regrssion, but the other cases
are neutral or improved.

[x86] add tests for vector select with bit-test condition; NFC

[mlir] [VectorOps] Add choice between dot and axpy lowering of vector.contract

Default vector.contract lowering essentially yields a series of sdot/ddot
operations. However, for some layouts a series of saxpy/daxpy operations,
chained through fma are more efficient. This CL introduces a choice between
the two lowering paths. A default heuristic is to follow.

Some preliminary avx2 performance numbers for matrix-times-vector.
Here, dot performs best for 64x64 A x b and saxpy for 64x64 A^T x b.

```
------------------------------------------------------------
            A x b                          A^T x b
------------------------------------------------------------
GFLOPS    sdot (reassoc)    saxpy    sdot (reassoc)    saxpy
------------------------------------------------------------
1x1        0.6               0.9       0.6             0.9
2x2        2.5               3.2       2.4             3.5
4x4        6.4               8.4       4.9             11.8
8x8       11.7               6.1       5.0             29.6
16x16     20.7              10.8       7.3             43.3
32x32     29.3               7.9       6.4             51.8
64x64     38.9                                         79.3
128x128   32.4                                         40.7
------------------------------------------------------------
```

Reviewed By: nicolasvasilache, ftynse

Differential Revision: https://reviews.llvm.org/D83012

[InstSimplify] Add test for sext/zext comparisons (NFC)

[test] Deflake test/profile/ContinuousSyncMode/online-merging.c

This test spawns 32 child processes which race to update counters on
shared memory pages. On some Apple-internal machines, two processes race
to perform an update in approximately 0.5% of the test runs, leading to
dropped counter updates. Deflake the test by using atomic increments.

Tested with:

```
$ for I in $(seq 1 1000); do echo ":: Test run $I..."; ./bin/llvm-lit projects/compiler-rt/test/profile/Profile-x86_64h/ContinuousSyncMode/online-merging.c -av || break; done
```

rdar://64956774

[X86] Add test case for unfolding broadcast load from vpternlog.

[X86] Add vpternlog to the broadcast unfolding table.

[PowerPC]Add Vector Insert Instruction Definitions and MC Test

Adds td definitions and asm/disasm tests for the following instructions:

  VINSBVLX
  VINSBVRX
  VINSHVLX
  VINSHVRX
  VINSWVLX
  VINSWVRX
  VINSBLX
  VINSBRX
  VINSHLX
  VINSHRX
  VINSWLX
  VINSWRX
  VINSDLX
  VINSDRX
  VINSW
  VINSD

Differential Revision: https://reviews.llvm.org/D83052

[SelectionDAG] don't split branch on logic-of-vector-compares

SelectionDAGBuilder converts logic-of-compares into multiple branches based
on a boolean TLI setting in isJumpExpensive(). But that probably never
considered the pattern of extracted bools from a vector compare - it seems
unlikely that we would want to turn vector logic into control-flow.

The motivating x86 reduction case is shown in PR44565:
https://bugs.llvm.org/show_bug.cgi?id=44565
...and that test shows the expected improvement from using pmovmsk codegen.

For AArch64, I modified the test to include an extra op because the simpler
test gets transformed by a codegen invocation of SimplifyCFG.

Differential Revision: https://reviews.llvm.org/D82602

[x86] remove redundant tests with no check lines; NFC

These were accidentally included with:
rGb93e6650c8ac

Fix typo and check commit access.

[PowerPC] Implement Vector Blend Builtins in LLVM/Clang

Implements vec_blendv()

Differential Revision: https://reviews.llvm.org/D82774

[flang] External I/O runtime work, repackaged (part 1)

Add a isFixedRecordLength flag member to Connection to
disambiguate the state of "record has known variable length"
from "record has fixed length".  Code that sets and tests this
flag will appear in later patches.  Rearrange data members to
reduce storage requirements, since Connection might indirectly
end up on a program stack frame.  Add a utility member function
BeginRecord(); use it in internal I/O processing.

Reviewed By: tskeith, sscalpone

Differential Revision: https://reviews.llvm.org/D83098

[X86] Teach lower512BitShuffle to try bitmask and bitblend before splitting v32i16/v64i8 on av512f only targets.

We consider v32i16/v64i8 to be legal types on avx512f, but we
don't have most operations until avx512bw. But we can use
and/or/xor operations. So try those before splitting.

This is especially helpful since we turn some ands with constant
masks into shuffles in early DAG combines. So we should make sure
we recover those back to AND.

[gn build] get everything to build when llvm_targets_to_build is just AArch64

[NFC][Scalarizer] Add some insertelement/extractelement tests

See D82961/D82970/D83101/D83102.

[NFC][Scalarizer] Also scalarize loads in newly-added tests

Should help better showcase improvements

[flang] External I/O runtime work, repackaged (part 2)

Clean up the input editing path so external input works better
when combined with further changes.  List-directed input needed
to allow for advancement to following records.

Reviewed By: tskeith, sscalpone

Differential Revision: https://reviews.llvm.org/D83104

[AMDGPU] Unify early PS termination blocks

Generate a single early exit block out-of-line and branch to this
if all lanes are killed. This avoids branching if lanes are active.

Reviewed By: nhaehnle

Differential Revision: https://reviews.llvm.org/D82641

[AMDGPU] Insert PS early exit at end of control flow

Exit early if the exec mask is zero at the end of control flow.
Mark the ends of control flow during control flow lowering and
convert these to exits during the insert skips pass.

Reviewed By: nhaehnle

Differential Revision: https://reviews.llvm.org/D82737

[PowerPC][NFC] Refactor lowerDynamicAlloc

When performing dynamic stack allocation, calculation of frame pointer
and actual negsize can be separated. This patch refactors
`lowerDynamicAlloc` in preparation of supporting
`-fstack-clash-protection` which also has to calculate actual frame
pointer and negsize.

Differential Revision: https://reviews.llvm.org/D81354

Revert "[AMDGPU] Insert PS early exit at end of control flow"

This reverts commit 2bfcacf0ad362956277a1c2c9ba00ddc453a42ce.

There appears to be an issue to analysis preservation.

[lld-macho] Support binding dysyms to any section

Previously, we only supported binding dysyms to the GOT. This
diff adds support for binding them to any arbitrary section. C++
programs appear to use this, I believe for vtables and type_info.

This diff also makes our bind opcode encoding a bit smarter -- we now
encode just the differences between bindings, which will make things
more compact.

I was initially concerned about the performance overhead of iterating
over these relocations, but it turns out that the number of such
relocations is small. A quick analysis of my llvm-project build
directory showed that < 1.3% out of ~7M relocations are RELOC_UNSIGNED
bindings to symbols (including both dynamic and static symbols).

Reviewed By: #lld-macho, smeenai

Differential Revision: https://reviews.llvm.org/D83103

[PowerPC][NFC] Prevent unused error when assertion is disabled.

[AMDGPU] Insert PS early exit at end of control flow

Exit early if the exec mask is zero at the end of control flow.
Mark the ends of control flow during control flow lowering and
convert these to exits during the insert skips pass.

Reviewed By: nhaehnle

Differential Revision: https://reviews.llvm.org/D82737

[X86] Add back support for matching VPTERNLOG from back to back logic ops.

I think this mostly looks ok. The only weird thing I noticed was
a couple rotate vXi8 tests picked up an extra logic op where we have

(and (or (and), (andn)), X). Previously we matched the (or (and), (andn))
to vpternlog, but now we match the (and (or), X) and leave the and/andn
unmatched.

[PowerPC] Implement probing for dynamic stack allocation

This patch is part of supporting `-fstack-clash-protection`. Mainly do
such things compared to existing `lowerDynamicAlloc`

- Added a new pseudo instruction PPC::PREPARE_PROBED_ALLOC to get
  actual frame pointer and final stack pointer.
- Synthesize a loop to probe by blocks.
- Use DYNAREAOFFSET to get MaxCallFrameSize which is calculated in
  prologepilog.

Differential Revision: https://reviews.llvm.org/D81358

[CostModel] Fix cast crash

Don't presume instruction operands while matching reductions.

Bugzilla: https://bugs.llvm.org/show_bug.cgi?id=46430

Differential Revision: https://reviews.llvm.org/D82453

[clang] Check ValueDependent instead of InstantiationDependent before executing the align expr for builtin align functions.

in general, value dependent is a subset of instnatiation dependent. This
would allows us to produce diagnostics for the align expression (which
is instantiation dependent but not value dependent).

Differential Revision: https://reviews.llvm.org/D83074

[X86] Remove MODRM_SPLITREGM from the disassembler tables.

This offers a very minor table size reduction due to only being
used for one AMX opcode.

[NFC] Use ADT/Bitfields in Instructions

This is an example patch for D81580.

Differential Revision: https://reviews.llvm.org/D81662

Fix stack-clash probing for large static alloca

Differential Revision: https://reviews.llvm.org/D82867

[Attributor] Create getName() method for abstract attribute

Summary: The `getName()` method returns the name of the abstract attribute

Reviewers: jdoerfert, sstefan1, uenoku, homerdin, baziotis

Reviewed By: sstefan1

Subscribers: uenoku, kuter, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D83109