Per SPIR-V validation rules, explict layout decorations are only
needed for StorageBuffer, PhysicalStorageBuffer, Uniform, and
PushConstant storage classes. (And even that is for Shader
capabilities). So we don't need such decorations on the rest.
Reviewed By: hanchung
Differential Revision: https://reviews.llvm.org/D124543
ptxas is a proprietary compiler from Nvidia that can compile PTX to
machine code (SASS). It has a lot of diagnostics to catch errors
in PTX, which can be used to verify PTX output from llc.
Set -DPXTAS_EXECUTABLE=/path/to/ptxas CMake option to enable it.
If this option is not set, then ptxas is substituted to true which
effectively disables all ptxas RUN lines.
LLVM_PTXAS_EXECUTABLE environment variable takes precedence over
the CMake option, and allows to override ptxas executable that is used for LIT
without complete re-configuration.
Differential Revision: https://reviews.llvm.org/D121727
Support for loading shared objects as plugins into clang-tidy was added
in http://reviews.llvm.org/D111100. Unfortunately, the utility scripts
`clang-tidy-diff.py` and `run-clang-tidy.py` did not receive
corresponding arguments to forward such plugins to clang-tidy.
This diff adds a `-load=plugin` option to both scripts.
Differential Revision: http://reviews.llvm.org/D12306
Reviewed By: aaron.ballman
Previously MASK= elements were accessed in assumption that mask is an array of
input argument rank (and in combination with explicit DIM= argument we had
out-of-bounds access), but for MAXLOC/MINLOC/FINDLOC mask should be be
conformable and could be scalar.
Add new regression tests with scalar mask for verification.
Reviewed By: klausler
Differential Revision: https://reviews.llvm.org/D124408
This getLValue function was declared in 98db1f990f ([Analyzer] [NFC]
Parameter Regions, 2020-05-11) but was never implemented.
Reviewed By: NoQ
Differential Revision: https://reviews.llvm.org/D124461
- Do not traverse concept decl inside `AutoType`. We only traverse
declaration and definitions, not references to a declaration.
- Do not visit implicit AST node the relevant traversal mode.
- Add traversal extension points for concept requirements.
- Renamed `TraverseConceptReference` to mark as helper to share
the code. Having an extension point there seems confusing given that
there are many concept refences in the AST that do not call the
helper. Those are `AutoType`, `AutoTypeLoc` and constraint requirements.
Only clangd code requires an update.
There are no use-cases for concept requirement traversals yet, but
I added them in the earlier version of the patch and decided to keep
them for completeness.
Reviewed By: sammccall
Differential Revision: https://reviews.llvm.org/D124532
This reverts commit d46fa023ca.
Regressed include order in some cases with trailing comments, see the
comments on https://reviews.llvm.org/D121370. Will add a regression test
in a follow-up commit.
Current index value generation uses fixed-length vector ops, this patch
adds an alterantive codegen path compatible with scalable vectors by
using `LLVM::StepVectorOp`.
Differential Revision: https://reviews.llvm.org/D124454
When using opaque pointers, convert GEPs into offset representation
of the form P + V1 * Scale1 + V2 * Scale2 + ... + ConstantOffset.
This allows us to recognize equivalent address calculations even if
the GEPs don't use the same source element type.
This fixes an opaque pointer codegen regression seen in rustc.
Differential Revision: https://reviews.llvm.org/D124527
The `llvm.x86.cast.tile.to.vector` intrinsic is lowered to
`llvm.x86.tilestored64.internal` and `load <256 x i32>`. The
`llvm.x86.cast.vector.to.tile` is lowered to `store <256 x i32>` and
`llvm.x86.tileloadd64.internal`. When `llvm.x86.cast.tile.to.vector` is
used by `store <256 x i32>` or `load <256 x i32>` is used by
`llvm.x86.cast.vector.to.tile`, they can be combined by
`llvm.x86.tilestored64.internal` and `llvm.x86.tileloadd64.internal`.
Differential Revision: https://reviews.llvm.org/D124378
They can already be available, and even if not, DT/LI can be available.
We should not recompute them. Old PM is unchanged because it would
require changing dependencies, and we don't care enough about it.
Differential Revision: https://reviews.llvm.org/D124439
Reviewed By: nikic, aeubanks
isNoopAddrSpaceCast is expecting SrcAS is different from DestAS.
If the two AS are the same, consider ptrtoint/inttoptr as noop cast.
Reviewed By: arsenm
Differential Revision: https://reviews.llvm.org/D123573
Previously all entries in global_ctors had to have the void()* type and
we'd skip evaluating bitcasted functions. With opaque pointers we may
see the function directly.
Fixes#55147.
Reviewed By: #opaque-pointers, nikic
Differential Revision: https://reviews.llvm.org/D124553
This patch provides the basic infrastructure for lowering declarative
constructs for OpenMP and OpenACC.
This is part of the upstreaming effort from the fir-dev branch in [1].
[1] https://github.com/flang-compiler/f18-llvm-project
Reviewed By: kiranchandramohan, shraiysh, clementval
Differential Revision: https://reviews.llvm.org/D124225
Git wants to check in 'text' files with LF endings, so this changes them
in the repository but not in the checkout, where they keep CRLF endings.
Differential Revision: https://reviews.llvm.org/D124563
Fix escaping when launching in terminal with AppleScript. The invocation
we're building up is wrapped in single quotes when passed to bash and
wrapped in double quotes for AppleScript.
Here's an example invocation with the new escaping:
tell application "Terminal"
activate
do script "/bin/bash -c 'arch -arch arm64 'darwin-debug'
--unix-socket=/tmp/dL2jSh --arch=arm64 --working-dir
\"/private/tmp/with spaces\" --disable-aslr -- \"foo\"
\"bar\" \"baz\" ; echo Process exited with status $?';exit"
end tell
Previously we were using unescaped single quotes which resulted in the
whole bash invocation being passed in pieces. That works most of the
time but breaks when you have a space in your current working directory
for example.
rdar://91870763
Differential revision: https://reviews.llvm.org/D124568
The BinaryHolder has two caches for object and archive entries. These
are implemented as StringMaps of ObjectEntry and ArchiveEntry
respectively. The fact that they're stored by value is problematic
because the BinaryHolder hands out references that become invalidate
when the data structure grows. This patch wraps those object instances
in unique pointers and changes the interface to hand out pointers. This
resulted in transient failures.
rdar://90412671
Differential revision: https://reviews.llvm.org/D124567
Using the legacy PM for the optimization pipeline was deprecated in 13.0.0.
Following recent changes to remove non-core features of the legacy
PM/optimization pipeline, remove ThreadSanitizerLegacyPass.
Reviewed By: #sanitizers, vitalybuka
Differential Revision: https://reviews.llvm.org/D124209
Emitting metadata for the same ivar multiple times can lead to
miscompilations. Objective-C runtime adds offsets to calculate ivar
position in memory and presence of duplicate offsets causes wrong final
position thus overwriting unrelated memory.
Such a situation is impossible with modules disabled as clang diagnoses
ivar redeclarations during sema checks after parsing
(`Sema::ActOnFields`). Fix the case with modules enabled by checking
during deserialization if ivar is already declared. We also support
a use case where the same category ends up in multiple modules. We
don't want to treat this case as ivar redeclaration and instead merge
corresponding ivars.
rdar://83468070
Differential Revision: https://reviews.llvm.org/D121177
There were two problems with directly copying the MMOs from the old
function. The MMOs are owned by the function's Allocator, so need to
be reallocated anyways (surprisingly I didn't notice breakage on
this). Second, the PseudoSourceValues are also allocated per function
and need to be reallocated.
The current testcase I'm trying to reduce only reproduces with IPRA
enabled and requires handling multiple functions.
The only real difference vs. the IR is the extra indirect to look for
the underlying MachineFunction, so treat the ReduceWorkItem as the
module instead of the function.
The ugliest piece of this is really the ugliness of
MachineModuleInfo. It not only tracks actual module state, but has a
number of transient fields used for isel and/or the asm printer. These
shouldn't do any harm for the use here, though they should be
separated out.
When passing a scalar .FALSE. as the MASK argument to MAXLOC, we were getting
bad memory references. We were falling into the code intended when the MASK
argument was missing.
I fixed this by checking for a scalar MASK with a .FALSE. value and
setting the result to all zeroes in that case. I also added tests for
MAXLOC and MINLOC with scalar values of .TRUE. and .FALSE. for the MASK
argument.
I also special cased situations where the MASK argument is a scalar with
a .TRUE. value and passed along a nullptr in such cases.
Along the way, I eliminated the unused "chars" argument from the constructor
for ExtremumLocAccumulator.
Differential Revision: https://reviews.llvm.org/D124484
Lei Zhang