platform/upstream/llvm.git
3 years ago[NFC][Doc] Mention SystemZ supports StackMap generation
Ulrich Weigand [Thu, 21 Jan 2021 17:29:46 +0000 (18:29 +0100)]
[NFC][Doc] Mention SystemZ supports StackMap generation

Support available as of commit 5eb64110d241cf2506f54ade3c2693beed42dd8f.

Differential Revision: https://reviews.llvm.org/D95040

3 years ago[RISCV] Update V instructions constraints to conform to v1.0
Hsiangkai Wang [Mon, 28 Dec 2020 03:57:41 +0000 (11:57 +0800)]
[RISCV] Update V instructions constraints to conform to v1.0

Upgrade RISC-V V extension to v1.0-08a0b46.
Update instruction constraints to conform to v1.0.

Differential Revision: https://reviews.llvm.org/D93612

3 years ago[OpenMP] Add time profiling support in libomp
Giorgis Georgakoudis [Sat, 16 Jan 2021 02:50:06 +0000 (18:50 -0800)]
[OpenMP] Add time profiling support in libomp

Profiling has been recently implemented in libomptarget (D93055). This patch enables time profiling support for libomptarget in libomp, to support profiling of multi-threaded execution of offloaded regions.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D94855

3 years agoRevert "[AMDGPU] Implement mir parseCustomPseudoSourceValue"
Sebastian Neubauer [Thu, 21 Jan 2021 17:11:48 +0000 (18:11 +0100)]
Revert "[AMDGPU] Implement mir parseCustomPseudoSourceValue"

This reverts commit ba7dcd8542abfc784255efcb0767701dec42fe83.

(caused memory leaks)

3 years ago[libc++] Use ioctl when available to get random_device entropy.
Marek Kurdej [Thu, 21 Jan 2021 16:55:19 +0000 (17:55 +0100)]
[libc++] Use ioctl when available to get random_device entropy.

Implemented the idea from D94571 to improve entropy on Linux.

Reviewed By: ldionne, #libc

Differential Revision: https://reviews.llvm.org/D94953

3 years ago[RISCV] Add new V instructions in v1.0-08a0b46.
Hsiangkai Wang [Mon, 21 Dec 2020 04:14:06 +0000 (12:14 +0800)]
[RISCV] Add new V instructions in v1.0-08a0b46.

Add new V instructions.
vfrsqrte7.v
vfrece7.v
vrgatherei16.vv
vneg.v
vncvt.x.x.w
vfneg.v

3 years ago[flang][driver] Move fixed/free from detection out of FrontendAction API
Andrzej Warzynski [Thu, 21 Jan 2021 15:42:56 +0000 (15:42 +0000)]
[flang][driver] Move fixed/free from detection out of FrontendAction API

All Fortran options should be set in `CompilerInstance` (via its
`CompilerInvocation`) before any of `FrontendAction` is entered -
that's one of the tasks of the driver. However, this is a bit tricky
with fixed and free from detection introduced in
https://reviews.llvm.org/D94228.

Fixed-free form detection needs to happen:
  * before any frontend action (we need to specify `isFixedForm` in
    `Fortran::parser::Options` before running any actions)
  * separately for every input file (we might be compiling multiple
    Fortran files, some in free form, some in fixed form)
In other words, we need this to happen early (before any
`FrontendAction`), but not too early (we need to know what the current
input file is). In practice, `isFixedForm` can only be set later
than other options (other options are inferred from compiler flags). So
we can't really set all of them in one place, which is not ideal.

All changes in this patch are NFCs (hence no new tests). Quick summary:
  * move fixed/free form detection from `FrontendAction::ExecuteAction` to
    `CompilerInstance::ExecuteAction`
  * add a bool flag in `FrontendInputFile` to mark a file as fixed/free
    form
  * updated a few comments

Differential Revision: https://reviews.llvm.org/D95042

3 years ago[RISCV] Make LMUL field in VTYPE continuous.
Hsiangkai Wang [Mon, 21 Dec 2020 04:08:09 +0000 (12:08 +0800)]
[RISCV] Make LMUL field in VTYPE continuous.

Upgrade RISC-V V extension to v1.0-08a0b46.
Update the VTYPE encoding. Make LMUL encoding in a continuous field.

3 years ago[mlir]][SPIRV] Define OrderedOp and UnorderedOp and add lowerings from Standard.
MaheshRavishankar [Wed, 20 Jan 2021 22:47:14 +0000 (14:47 -0800)]
[mlir]][SPIRV] Define OrderedOp and UnorderedOp and add lowerings from Standard.

Define OrderedOp and UnorderedOp instructions in SPIR-V and convert
cmpf operations with `ord` and `uno` tag to these instructions
respectively.

Differential Revision: https://reviews.llvm.org/D95098

3 years ago[mlir][SPIRV] Rename OpSpecConstantOperation -> OpSpecConstantOp
MaheshRavishankar [Wed, 20 Jan 2021 22:46:36 +0000 (14:46 -0800)]
[mlir][SPIRV] Rename OpSpecConstantOperation -> OpSpecConstantOp

The SPIR-V spec uses OpSpecConstantOp. Using an inconsistent name
makes the dialect generation scripts fail. Update to use the right
operation name, and fix the auto generation scripts as well.

Differential Revision: https://reviews.llvm.org/D95097

3 years ago[AMDGPU][GlobalISel] Run SIAddImgInit
Jay Foad [Thu, 21 Jan 2021 14:31:43 +0000 (14:31 +0000)]
[AMDGPU][GlobalISel] Run SIAddImgInit

This pass is required to get correct codegen for image instructions with
the tfe or lwe bits set.

Differential Revision: https://reviews.llvm.org/D95132

3 years agoAMDGPU: Remove v_rsq_f64 patterns
Matt Arsenault [Wed, 20 Jan 2021 22:07:41 +0000 (17:07 -0500)]
AMDGPU: Remove v_rsq_f64 patterns

This isn't accurate enough without correction

3 years agoAMDGPU: Use more accurate fast f64 fdiv
Matt Arsenault [Wed, 20 Jan 2021 18:55:55 +0000 (13:55 -0500)]
AMDGPU: Use more accurate fast f64 fdiv

A raw v_rcp_f64 isn't accurate enough, so start applying correction.

3 years ago[OpenMP][NVPTX] Added forward declaration for atomic operations
Shilei Tian [Thu, 21 Jan 2021 15:37:05 +0000 (10:37 -0500)]
[OpenMP][NVPTX] Added forward declaration for atomic operations

Pretty similar to D95058, this patch added forward declaration for
CUDA atomic functions. We already have definitions with right mangled names in
internal CUDA headers so the forward declaration here can work properly.

Reviewed By: jdoerfert, JonChesterfield

Differential Revision: https://reviews.llvm.org/D95085

3 years agoAArch64/GlobalISel: Factor out parametersInCSRMatch
Matt Arsenault [Tue, 12 Jan 2021 23:58:30 +0000 (18:58 -0500)]
AArch64/GlobalISel: Factor out parametersInCSRMatch

Make this look more like the DAG handling and move to common code.

I also noticed AArch64 seems to not be properly adding the
physreg:virtreg mapping to the function live ins.

3 years ago[AMDGPU] Implement mir parseCustomPseudoSourceValue
Sebastian Neubauer [Fri, 15 Jan 2021 12:57:24 +0000 (13:57 +0100)]
[AMDGPU] Implement mir parseCustomPseudoSourceValue

Allow parsing generated mir with custom pseudo source value tokens.
Also rename pseudo source values to have more meaningful names.

Differential Revision: https://reviews.llvm.org/D94768

3 years ago[ARM] Fix vector saddsat costs.
David Green [Thu, 21 Jan 2021 15:30:39 +0000 (15:30 +0000)]
[ARM] Fix vector saddsat costs.

It turns out the vectorizer calls the getIntrinsicInstrCost functions
with a scalar return type and vector VF. This updates the costmodel to
handle that, still producing the correct vector costs.

A vectorizer test is added to show it vectorizing at the correct factor
again.

3 years ago[flang][driver] Make the driver report diagnostics from the prescanner
Andrzej Warzynski [Thu, 21 Jan 2021 15:02:22 +0000 (15:02 +0000)]
[flang][driver] Make the driver report diagnostics from the prescanner

This patch makes sure that diagnostics from the prescanner are reported
when running `flang-new -E` (i.e. only the preprocessor phase is
requested). More specifically, the `PrintPreprocessedAction` action is
updated.

With this patch we make sure that the `f18` and `flang-new` provide
identical output when running the preprocessor and the prescanner
generates diagnostics.

Differential Revision: https://reviews.llvm.org/D94782

3 years ago[OpenMP] Add support for mapping names in mapper API
Joseph Huber [Fri, 15 Jan 2021 18:09:35 +0000 (13:09 -0500)]
[OpenMP] Add support for mapping names in mapper API

Summary:
The custom mapper API did not previously support the mapping names added previously. This means they were not present if a user requested debugging information while using the mapper functions. This adds basic support for passing the mapped names to the runtime library.

Reviewers: jdoerfert

Differential Revision: https://reviews.llvm.org/D94806

3 years agoAMDGPU: Add occupancy to serialized MachineFunctionInfo
Matt Arsenault [Fri, 15 Jan 2021 21:07:37 +0000 (16:07 -0500)]
AMDGPU: Add occupancy to serialized MachineFunctionInfo

Not sure about the default value handling, but also not sure
defaulting to a theoretically subtarget dependent value.

3 years ago[lldb][NFC] Fix build with GCC<6
Raphael Isemann [Thu, 21 Jan 2021 14:04:40 +0000 (15:04 +0100)]
[lldb][NFC] Fix build with GCC<6

GCC/libstdc++ before 6.1 can't handle scoped enums as unordered_map keys. LLVM
(and some build) bots officially support some GCC 5.x versions, so this patch
just makes the enum unscoped until we can require GCC 6.x.

3 years ago[clang][AST] Add get functions for CXXFoldExpr paren locations.
Balázs Kéri [Thu, 21 Jan 2021 09:17:57 +0000 (10:17 +0100)]
[clang][AST] Add get functions for CXXFoldExpr paren locations.

Reviewed By: hokein

Differential Revision: https://reviews.llvm.org/D94787

3 years ago[InstCombine] avoid crashing on attribute propagation
Sanjay Patel [Thu, 21 Jan 2021 12:54:41 +0000 (07:54 -0500)]
[InstCombine] avoid crashing on attribute propagation

In https://llvm.org/PR48810 , we are crashing while trying to
propagate attributes from mempcpy (returns void*) to memcpy
(returns nothing - void).

We can avoid the crash by removing known incompatible
attributes for the void return type.

I'm not sure if this goes far enough (should we just drop all
attributes since this isn't the same function?). We also need
to audit other transforms in LibCallSimplifier to make sure
there are no other cases that have the same problem.

Differential Revision: https://reviews.llvm.org/D95088

3 years ago[MC] Use std::make_tuple to make some toolchains happy again
Mikael Holmen [Thu, 21 Jan 2021 13:04:44 +0000 (14:04 +0100)]
[MC] Use std::make_tuple to make some toolchains happy again

My toolchain (LLVM 8.0, libstdc++ 5.4.0) complained with:

12:27:43 ../lib/MC/MCDwarf.cpp:814:10: error: chosen constructor is explicit in copy-initialization
12:27:43   return {Offset, Size, SetDelta};
12:27:43          ^~~~~~~~~~~~~~~~~~~~~~~~
12:27:43 /proj/flexasic/app/llvm/8.0/bin/../lib/gcc/x86_64-unknown-linux-gnu/5.4.0/../../../../include/c++/5.4.0/tuple:479:19: note: explicit constructor declared here
12:27:43         constexpr tuple(_UElements&&... __elements)
12:27:43                   ^
12:27:43 1 error generated.

This commit adds explicit calls to std::make_tuple to work around
the problem.

3 years agoAdd log1p lowering from standard to ROCDL intrinsics
Frederik Gossen [Thu, 21 Jan 2021 12:54:48 +0000 (13:54 +0100)]
Add log1p lowering from standard to ROCDL intrinsics

Differential Revision: https://reviews.llvm.org/D95129

3 years ago[DAGCombiner] Enable SimplifyDemandedBits vector support for TRUNCATE (REAPPLIED).
Simon Pilgrim [Thu, 21 Jan 2021 12:58:16 +0000 (12:58 +0000)]
[DAGCombiner] Enable SimplifyDemandedBits vector support for TRUNCATE (REAPPLIED).

Add DemandedElts support inside the TRUNCATE analysis.

REAPPLIED - this was reverted by @hans at rGa51226057fc3 due to an issue with vector shift amount types, which was fixed in rG935bacd3a724 and an additional test case added at rG0ca81b90d19d

Differential Revision: https://reviews.llvm.org/D56387

3 years agoAdd log1p lowering from standard to NVVM intrinsics
Frederik Gossen [Thu, 21 Jan 2021 12:56:25 +0000 (13:56 +0100)]
Add log1p lowering from standard to NVVM intrinsics

Differential Revision: https://reviews.llvm.org/D95130

3 years ago[X86][SSE] Add uitofp(trunc(and(lshr(x,c)))) vector test
Simon Pilgrim [Thu, 21 Jan 2021 12:29:50 +0000 (12:29 +0000)]
[X86][SSE] Add uitofp(trunc(and(lshr(x,c)))) vector test

Reduced from regression reported by @hans on D56387

3 years ago[DAG] SimplifyDemandedBits - correctly adjust truncated shift amount type
Simon Pilgrim [Thu, 21 Jan 2021 12:24:35 +0000 (12:24 +0000)]
[DAG] SimplifyDemandedBits - correctly adjust truncated shift amount type

As noticed on D56387, for vectors we must always correctly adjust the shift amount type during truncation (not just after legalization). We were getting away with it as we currently only accepted scalars via the dyn_cast<ConstantSDNode>.

3 years agoReland [lldb] Fix TestThreadStepOut.py after "Flush local value map on every instruction"
Raphael Isemann [Thu, 26 Nov 2020 08:40:02 +0000 (09:40 +0100)]
Reland [lldb] Fix TestThreadStepOut.py after "Flush local value map on every instruction"

The original patch got reverted as a dependency of cf1c774d6ace59c5adc9ab71b31e .
That patch got relanded so it's also necessary to reland this patch.

Original summary:

After cf1c774d6ace59c5adc9ab71b31e762c1be695b1, Clang seems to generate code
that is more similar to icc/Clang, so we can use the same line numbers for
all compilers in this test.

3 years ago[lldb] Make TestBSDArchives a no-debug-info-test
Raphael Isemann [Thu, 21 Jan 2021 12:06:47 +0000 (13:06 +0100)]
[lldb] Make TestBSDArchives a no-debug-info-test

The DSYM variant of this test is failing since D94890. But as we explicitly
try to disable the DSYM generation in the makefile and build the archive on
our own, I don't see why we even need to run the DSYM version of the test.

This patch disables the generated derived versions of this test for the
different debug information containers (which includes the failing DSYM one).

3 years ago[lldb][import-std-module] Do some basic file checks before trying to import a module
Raphael Isemann [Thu, 21 Jan 2021 11:05:59 +0000 (12:05 +0100)]
[lldb][import-std-module] Do some basic file checks before trying to import a module

Currently when LLDB has enough data in the debug information to import the `std` module,
it will just try to import it. However when debugging libraries where the sources aren't
available anymore, importing the module will generate a confusing diagnostic that
the module couldn't be built.

For the fallback mode (where we retry failed expressions with the loaded module), this
will cause the second expression to fail with a module built error instead of the
actual parsing issue in the user expression.

This patch adds checks that ensures that we at least have any source files in the found
include paths before we try to import the module. This prevents the module from being
loaded in the situation described above which means we don't emit the bogus 'can't
import module' diagnostic and also don't waste any time retrying the expression in the
fallback mode.

For the unit tests I did some refactoring as they now require a VFS with the files in it
and not just the paths. The Python test just builds a binary with a fake C++ module,
then deletes the module before debugging.

Fixes rdar://73264458

Reviewed By: JDevlieghere

Differential Revision: https://reviews.llvm.org/D95096

3 years agoMC: AArch64: Add support for gotpage_lo15
Adhemerval Zanella [Wed, 13 Jan 2021 17:27:42 +0000 (17:27 +0000)]
MC: AArch64: Add support for gotpage_lo15

It is not used bt LLVM itself, but it would be used on lld tests
to implement R_AARCH64_LD64_GOTPAGE_LO15 support.

3 years ago[DAG] CombineToPreIndexedLoadStore - use const APInt& for getAPIntValue(). NFCI.
Simon Pilgrim [Thu, 21 Jan 2021 11:01:02 +0000 (11:01 +0000)]
[DAG] CombineToPreIndexedLoadStore - use const APInt& for getAPIntValue(). NFCI.

Cleanup some code to use auto* properly from cast, and use const APInt& for getAPIntValue() to avoid an unnecessary copy.

3 years ago[X86] Avoid a std::string copy by replacing auto with const auto&. NFC.
Simon Pilgrim [Thu, 21 Jan 2021 10:43:07 +0000 (10:43 +0000)]
[X86] Avoid a std::string copy by replacing auto with const auto&. NFC.

Fixes msvc analyzer warning.

3 years agoRevert "[X86][AMX] Fix tile config register spill issue."
Luo, Yuanke [Thu, 21 Jan 2021 10:00:43 +0000 (18:00 +0800)]
Revert "[X86][AMX] Fix tile config register spill issue."

This reverts commit 20013d02f3352a88d0838eed349abc9a2b0e9cc0.

3 years ago[clangd] Fix a missing override keyword, NFC.
Haojian Wu [Thu, 21 Jan 2021 10:06:43 +0000 (11:06 +0100)]
[clangd] Fix a missing override keyword, NFC.

3 years ago[LoopUnswitch] Implement first version of partial unswitching.
Florian Hahn [Thu, 21 Jan 2021 09:32:04 +0000 (09:32 +0000)]
[LoopUnswitch] Implement first version of partial unswitching.

This patch applies the idea from D93734 to LoopUnswitch.

It adds support for unswitching on conditions that are only
invariant along certain paths through a loop.

In particular, it targets conditions in the loop header that
depend on values loaded from memory. If either path from
the true or false successor through the loop does not modify
memory, perform partial loop unswitching.

That is, duplicate the instructions feeding the condition in the pre-header.
Then unswitch on the duplicated condition. The condition is now known
in the unswitched version for the 'invariant' path through the original loop.

On caveat of this approach is that one of the loops created can be partially
unswitched again. To avoid this behavior, `llvm.loop.unswitch.partial.disable`
metadata is added to the unswitched loops, to avoid subsequent partial
unswitching.

If that's the approach to go, I can move the code handling the metadata kind
into separate functions.

This increases the cases we unswitch quite a bit in SPEC2006/SPEC2000 &
MultiSource. It also allows us to eliminate a dead loop in SPEC2017's omnetpp

```
Tests: 236
Same hash: 170 (filtered out)
Remaining: 66
Metric: loop-unswitch.NumBranches

Program                                        base   patch  diff
 test-suite...000/255.vortex/255.vortex.test     2.00  23.00 1050.0%
 test-suite...T2006/401.bzip2/401.bzip2.test     7.00  55.00 685.7%
 test-suite :: External/Nurbs/nurbs.test         5.00  26.00 420.0%
 test-suite...s-C/unix-smail/unix-smail.test     1.00   3.00 200.0%
 test-suite.../Prolangs-C++/ocean/ocean.test     1.00   3.00 200.0%
 test-suite...tions/lambda-0.1.3/lambda.test     1.00   3.00 200.0%
 test-suite...yApps-C++/PENNANT/PENNANT.test     2.00   5.00 150.0%
 test-suite...marks/Ptrdist/yacr2/yacr2.test     1.00   2.00 100.0%
 test-suite...lications/viterbi/viterbi.test     1.00   2.00 100.0%
 test-suite...plications/d/make_dparser.test    12.00  24.00 100.0%
 test-suite...CFP2006/433.milc/433.milc.test    14.00  27.00 92.9%
 test-suite.../Applications/lemon/lemon.test     7.00  12.00 71.4%
 test-suite...ce/Applications/Burg/burg.test     6.00  10.00 66.7%
 test-suite...T2006/473.astar/473.astar.test    16.00  26.00 62.5%
 test-suite...marks/7zip/7zip-benchmark.test    78.00 121.00 55.1%
```

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D93764

3 years ago[mlir] Remove complex ops from Standard dialect.
Alexander Belyaev [Wed, 20 Jan 2021 20:26:54 +0000 (21:26 +0100)]
[mlir] Remove complex ops from Standard dialect.

`complex` dialect should be used instead.
https://llvm.discourse.group/t/rfc-split-the-complex-dialect-from-std/2496/2

Differential Revision: https://reviews.llvm.org/D95077

3 years agoMCDwarf: Delete uneeded parameter
Fangrui Song [Thu, 21 Jan 2021 08:55:07 +0000 (00:55 -0800)]
MCDwarf: Delete uneeded parameter

And change signature

3 years ago[llvm-nm][ELF] - Make -D display symbol versions.
Georgii Rymar [Thu, 14 Jan 2021 13:20:35 +0000 (16:20 +0300)]
[llvm-nm][ELF] - Make -D display symbol versions.

This fixes https://bugs.llvm.org/show_bug.cgi?id=48670.

Since binutils 2.35, nm -D displays symbol versions by default.
This patch teaches llvm-nm to do the same.

Differential revision: https://reviews.llvm.org/D94907

3 years ago[X86][AMX] Fix tile config register spill issue.
Luo, Yuanke [Tue, 5 Jan 2021 13:41:51 +0000 (21:41 +0800)]
[X86][AMX] Fix tile config register spill issue.

Previous code build the model that tile config register is the user of
each AMX instruction. There is a problem for the tile config register
spill. When across function, the ldtilecfg instruction may be inserted
on each AMX instruction which use tile config register. This cause all
tile data register clobber.
To fix this issue, we remove the model of tile config register. We
analyze the regmask of call instruction and insert ldtilecfg if there is
any tile data register live across the call. Inserting the sttilecfg
before the call is unneccessary, because the tile config doesn't change
and we can just reload the config.
Besides we also need check tile config register interference. Since we
don't model the config register we should check interference from the
ldtilecfg to each tile data register def.
             ldtilecfg
             /       \
            BB1      BB2
            /         \
           call       BB3
           /           \
       %1=tileload   %2=tilezero
We can start from the instruction of each tile def, and backward to
ldtilecfg. If there is any call instruction, and tile data register is
not preserved, we should insert ldtilecfg after the call instruction.

Differential Revision: https://reviews.llvm.org/D94155

3 years ago[yaml2obj/obj2yaml] - Improve dumping/creating of ELF versioning sections.
Georgii Rymar [Mon, 18 Jan 2021 14:42:49 +0000 (17:42 +0300)]
[yaml2obj/obj2yaml] - Improve dumping/creating of ELF versioning sections.

This makes the following improvements.

For `SHT_GNU_versym`:
 * yaml2obj: set `sh_link` to index of `.dynsym` section automatically.
For `SHT_GNU_verdef`:
 * yaml2obj: set `sh_link` to index of `.dynstr` section automatically.
 * yaml2obj: set `sh_info` field automatically.
 * obj2yaml: don't dump the `Info` field when its value matches the number of version definitions.
For `SHT_GNU_verneed`:
 * yaml2obj: set `sh_link` to index of `.dynstr` section automatically.
 * yaml2obj: set `sh_info` field automatically.
 * obj2yaml: don't dump the `Info` field when its value matches the number of version dependencies.

Also, simplifies few test cases.

Differential revision: https://reviews.llvm.org/D94956

3 years ago[IndirectFunctions] Skip propagating attributes to address taken functions
madhur13490 [Wed, 13 Jan 2021 09:32:29 +0000 (09:32 +0000)]
[IndirectFunctions] Skip propagating attributes to address taken functions

In case of indirect calls or address taken functions,
skip propagating any attributes to them. We just
propagate features to such functions.

Reviewed By: rampitec

Differential Revision: https://reviews.llvm.org/D94585

3 years ago[llvm] Use hasSingleElement (NFC)
Kazu Hirata [Thu, 21 Jan 2021 05:35:55 +0000 (21:35 -0800)]
[llvm] Use hasSingleElement (NFC)

3 years ago[Transforms] Use llvm::append_range (NFC)
Kazu Hirata [Thu, 21 Jan 2021 05:35:53 +0000 (21:35 -0800)]
[Transforms] Use llvm::append_range (NFC)

3 years ago[llvm] Construct SmallVector with iterator ranges (NFC)
Kazu Hirata [Thu, 21 Jan 2021 05:35:51 +0000 (21:35 -0800)]
[llvm] Construct SmallVector with iterator ranges (NFC)

3 years ago[X86] Add experimental option to separately tune alignment of innermost loops
Max Kazantsev [Thu, 21 Jan 2021 04:15:16 +0000 (11:15 +0700)]
[X86] Add experimental option to separately tune alignment of innermost loops

We already have an experimental option to tune loop alignment. Its impact
is very wide (and there is a suspicion that it's not always profitable). We want
to have something more narrow to play with. This patch adds similar option that
overrides preferred alignment for innermost loops. This is for experimental
purposes, default values do not change the existing behavior.

Differential Revision: https://reviews.llvm.org/D94895
Reviewed By: pengfei

3 years ago[RISCV] Implement vssseg intrinsics.
Hsiangkai Wang [Sat, 16 Jan 2021 13:40:41 +0000 (21:40 +0800)]
[RISCV] Implement vssseg intrinsics.

Define vlsseg intrinsics and pseudo instructions. Lower vlsseg
intrinsics to pseudo instructions in RISCVDAGToDAGISel.

Differential Revision: https://reviews.llvm.org/D94863

3 years ago[RISCV] Implement vlsseg intrinsics.
Hsiangkai Wang [Fri, 15 Jan 2021 11:29:51 +0000 (19:29 +0800)]
[RISCV] Implement vlsseg intrinsics.

Define vlsseg intrinsics and pseudo instructions. Lower vlsseg intrinsics
to pseudo instructions in RISCVDAGToDAGISel.

Differential Revision: https://reviews.llvm.org/D94763

3 years ago[RISCV] Implement vsseg intrinsics.
Hsiangkai Wang [Thu, 14 Jan 2021 09:07:18 +0000 (17:07 +0800)]
[RISCV] Implement vsseg intrinsics.

Define vsseg intrinsics and pseudo instructions. Lower vsseg intrinsics
to pseudo instructions in RISCVDAGToDAGISel.

Differential Revision: https://reviews.llvm.org/D94688

3 years ago[lldb] Upstream eCore_arm_arm64e enum value in ArchSpec
Jonas Devlieghere [Thu, 21 Jan 2021 02:49:19 +0000 (18:49 -0800)]
[lldb] Upstream eCore_arm_arm64e enum value in ArchSpec

Upstream the eCore_arm_arm64e enum value in ArchSpec. All the other
arm64e triple changes already landed in LLVM.

Differential revision: https://reviews.llvm.org/D95110

3 years ago[RISCV] Use update_llc_test_checks.py to regenerate check lines in vleff-rv32.ll...
Craig Topper [Thu, 21 Jan 2021 02:35:31 +0000 (18:35 -0800)]
[RISCV] Use update_llc_test_checks.py to regenerate check lines in vleff-rv32.ll and vleff-rv64.ll.

This should minimize change in a future patch.

3 years ago[dsymutil] Compare object modification times using second precision
Jonas Devlieghere [Thu, 21 Jan 2021 02:38:56 +0000 (18:38 -0800)]
[dsymutil] Compare object modification times using second precision

The modification time in the debug map is expressed using second
precision, while the modification time returned by the filesystem could
be more precise. Avoid spurious warnings about timestamp mismatches by
truncating the modification time reported by the system to seconds.

3 years agoUse CXX_SOURCES and point to the right source file.
Jim Ingham [Thu, 21 Jan 2021 02:38:07 +0000 (18:38 -0800)]
Use CXX_SOURCES and point to the right source file.

Copy paste error, but the test still built on macOS.  Weird.
It failed on debian linux with an error about -fno-limit-debug-info
not being a supported flag???  Not sure how this goof would cause
that error, but let's see if it did...

3 years ago[MSan] Move origins for overlapped memory transfer
Jianzhou Zhao [Wed, 13 Jan 2021 01:37:16 +0000 (01:37 +0000)]
[MSan] Move origins for overlapped memory transfer

Reviewed-by: eugenis
Differential Revision: https://reviews.llvm.org/D94572

3 years agoFix a bug with setting breakpoints on C++11 inline initialization statements.
Jim Ingham [Thu, 21 Jan 2021 01:58:34 +0000 (17:58 -0800)]
Fix a bug with setting breakpoints on C++11 inline initialization statements.

If they occurred before the constructor that used them, we would refuse to
set the breakpoint because we thought they were crossing function boundaries.

Differential Revision: https://reviews.llvm.org/D94846

3 years ago[lld-macho] Add dependency on ObjCARC to fix shared build
Jez Ng [Thu, 21 Jan 2021 01:41:24 +0000 (20:41 -0500)]
[lld-macho] Add dependency on ObjCARC to fix shared build

3 years ago[Clang][OpenMP] Use `clang_cc1` test for `declare_target_device_only_compilation...
Shilei Tian [Thu, 21 Jan 2021 01:34:03 +0000 (20:34 -0500)]
[Clang][OpenMP] Use `clang_cc1` test for `declare_target_device_only_compilation.cpp`

Use `clang_cc1` test for `declare_target_device_only_compilation.cpp`

Reviewed By: echristo

Differential Revision: https://reviews.llvm.org/D95089

3 years ago[DAGCombiner] Precommit test case for D95086
Guozhi Wei [Thu, 21 Jan 2021 01:15:47 +0000 (17:15 -0800)]
[DAGCombiner] Precommit test case for D95086

This is the test case for D95086 with worse result.

Differential Revision: https://reviews.llvm.org/D95103

3 years ago[mlir][OpFormatGen] Fix incorrect kind used for RegionsDirective
River Riddle [Thu, 21 Jan 2021 00:47:00 +0000 (16:47 -0800)]
[mlir][OpFormatGen] Fix incorrect kind used for RegionsDirective

I attempted to write a test case for this, but the situations in which the kind is used for RegionDirective and ResultsDirective have zero overlap; meaning that there isn't a situation in which sharing the kind creates a conflict.

Differential Revision: https://reviews.llvm.org/D94988

3 years ago[mlir] Make MLIRContext::getOrLoadDialect(StringRef, TypeID, ...) public
mfehr [Thu, 21 Jan 2021 00:29:51 +0000 (00:29 +0000)]
[mlir] Make MLIRContext::getOrLoadDialect(StringRef, TypeID, ...) public

Having this function in a public scope is helpful to register dialects that are
defined at runtime, and thus that need a runtime-defined TypeID.

Also, a similar function in DialectRegistry, insert(TypeID, StringRef, ...), has
a public scope.

Reviewed By: mehdi_amini

Differential Revision: https://reviews.llvm.org/D95091

3 years ago[mlir] Add a new builtin `unrealized_conversion_cast` operation
River Riddle [Thu, 21 Jan 2021 00:17:26 +0000 (16:17 -0800)]
[mlir] Add a new builtin `unrealized_conversion_cast` operation

An `unrealized_conversion_cast` operation represents an unrealized conversion
from one set of types to another, that is used to enable the inter-mixing of
different type systems. This operation should not be attributed any special
representational or execution semantics, and is generally only intended to be
used to satisfy the temporary intermixing of type systems during the conversion
of one type system to another.

This operation was discussed in the following RFC(and ODM):

https://llvm.discourse.group/t/open-meeting-1-14-dialect-conversion-and-type-conversion-the-question-of-cast-operations/

Differential Revision: https://reviews.llvm.org/D94832

3 years ago[mlir] Add an interface for Cast-Like operations
River Riddle [Thu, 21 Jan 2021 00:17:13 +0000 (16:17 -0800)]
[mlir] Add an interface for Cast-Like operations

A cast-like operation is one that converts from a set of input types to a set of output types. The arity of the inputs may be from 0-N, whereas the arity of the outputs may be anything from 1-N. Cast-like operations are removable in cases where they produce a "no-op", i.e when the input types and output types match 1-1.

Differential Revision: https://reviews.llvm.org/D94831

3 years ago[NFC] Minor cleanup for ValueHandle code.
Varun Gandhi [Thu, 7 Jan 2021 16:50:31 +0000 (08:50 -0800)]
[NFC] Minor cleanup for ValueHandle code.

Based on feedback in https://reviews.llvm.org/D93433.

Reviewed By: dblaikie

Differential Revision: https://reviews.llvm.org/D94238

3 years ago[libc][NFC][obvious] fix the names of MPFR tests
Michael Jones [Wed, 20 Jan 2021 23:42:01 +0000 (23:42 +0000)]
[libc][NFC][obvious] fix the names of MPFR tests

I missed the MPFR tests in my previous commit. They have now been fixed
to not fail the prefix check in the test macro.

3 years ago[libc][NFC] add "LlvmLibc" as a prefix to all test names
Michael Jones [Fri, 15 Jan 2021 21:54:23 +0000 (21:54 +0000)]
[libc][NFC] add "LlvmLibc" as a prefix to all test names

Summary:
Having a consistent prefix makes selecting all of the llvm libc tests
easier on any platform that is also using the gtest framework.
This also modifies the TEST and TEST_F macros to enforce this change
moving forward.

Reviewers: sivachandra

Subscribers:

3 years ago[BuildLibcalls, Attrs] Support more variants of C++'s new, add attributes for C+...
Dávid Bolvanský [Wed, 20 Jan 2021 23:12:28 +0000 (00:12 +0100)]
[BuildLibcalls, Attrs] Support more variants of C++'s new, add attributes for C++'s delete

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D95095

3 years ago[RISCV] Add another isel pattern for slliu.w.
Craig Topper [Wed, 20 Jan 2021 22:52:03 +0000 (14:52 -0800)]
[RISCV] Add another isel pattern for slliu.w.

Previously we only matched (and (shl X, C1), 0xffffffff << C1)
which matches the InstCombine canonicalization order. But its
possible to see (shl (and X, 0xffffffff), C1) if the pattern
is introduced in SelectionDAG. For example, through expansion of
a GEP.

3 years ago[RISCV] Add addu.w and slliu.w test that uses getelementptr with zero extended indices.
Craig Topper [Wed, 20 Jan 2021 22:32:20 +0000 (14:32 -0800)]
[RISCV] Add addu.w and slliu.w test that uses getelementptr with zero extended indices.

This is closer to the kind of code that these intrinsics are
targeted at. Note we fail to match slliu.w here because our pattern
looks for (and (shl X, C1), 0xffffffff << C1) rather than
(shl (and X, 0xffffffff), C1). I'll fix this in a follow up
commit.

3 years agoRevert "[mlir][Affine] Add support for multi-store producer fusion"
Diego Caballero [Wed, 20 Jan 2021 22:36:37 +0000 (00:36 +0200)]
Revert "[mlir][Affine] Add support for multi-store producer fusion"

This reverts commit 7dd198852b4db52ae22242dfeda4eccda83aa8b2.

ASAN issue.

3 years ago[mlir][sparse] add asserts on reading in tensor data
Aart Bik [Wed, 20 Jan 2021 18:37:22 +0000 (10:37 -0800)]
[mlir][sparse] add asserts on reading in tensor data

Rationale:
Since I made the argument that metadata helps with extra
verification checks, I better actually do that ;-)

Reviewed By: penpornk

Differential Revision: https://reviews.llvm.org/D95072

3 years agoD94954: Fixes Snapdragon Kryo CPU core detection
Ryan Houdek [Wed, 20 Jan 2021 22:23:43 +0000 (22:23 +0000)]
D94954: Fixes Snapdragon Kryo CPU core detection

All of these families were claiming to be a73 based, which was causing
-mcpu/mtune=native to never use the newer features available to these
cores.

Goes through each and bumps the individual cores to their respective Big
counterparts. Since this code path doesn't support big.little detection,
there was already a precedent set with the Qualcomm line to choose the
big cores only.

Adds a comment on each line for the product's name that the part number
refers to. Confirmed on-device and through Linux header naming
convections.

Additionally newer SoCs mix CPU implementer parts from multiple
implementers. Both 0x41 (ARM) and 0x51 (Qualcomm) in the Snapdragon case

This was causing a desync in information where the scan at the start to
find the implementer would mismatch the part scan later on.
Now scan for both implementer and part at the start so these stay in
sync.

Differential Revision: https://reviews.llvm.org/D94954

3 years agoMakefile.rules: Avoid redundant .d generation (make restart) and inline archive rule...
Fangrui Song [Wed, 20 Jan 2021 22:22:33 +0000 (14:22 -0800)]
Makefile.rules: Avoid redundant .d generation (make restart) and inline archive rule to the only test

Take an example when `CXX_SOURCES` is main.cpp.

main.d is an included file. make will rebuild main.d, re-executes itself [1] to read
in the new main.d file, then rebuild main.o, finally link main.o into a.out.
main.cpp is parsed twice in this process.

This patch merges .d generation into .o generation [2], writes explicit rules
for .c/.m and deletes suffix rules for %.m and %.o. Since a target can be
satisfied by either of .c/.cpp/.m/.mm, we use multiple pattern rules. The
rule with the prerequisite (with VPATH considered) satisfied is used [3].

Since suffix rules are disabled, the implicit rule for archive member targets is
no long available [4]. Rewrite, simplify the archive rule and inline it into the
only test `test/API/functionalities/archives/Makefile`.

[1]: https://www.gnu.org/software/make/manual/html_node/Remaking-Makefiles.html
[2]: http://make.mad-scientist.net/papers/advanced-auto-dependency-generation/
[3]: https://www.gnu.org/software/make/manual/html_node/Pattern-Match.html
[4]: https://www.gnu.org/software/make/manual/html_node/Archive-Update.html

ObjC/ObjCXX tests only run on macOS. I don't have testing environment.  Hope
someone can do it for me.

Reviewed By: JDevlieghere

Differential Revision: https://reviews.llvm.org/D94890

3 years ago[mlir] NFC - Fix unused variable in non-debug mode
Nicolas Vasilache [Wed, 20 Jan 2021 22:19:59 +0000 (22:19 +0000)]
[mlir] NFC - Fix unused variable in non-debug mode

3 years ago[NFC][AMDGPU] Document target ID syntax for code object V2 to V3
Tony Tye [Wed, 20 Jan 2021 03:38:55 +0000 (03:38 +0000)]
[NFC][AMDGPU] Document target ID syntax for code object V2 to V3

Differential Revision: https://reviews.llvm.org/D95018

3 years ago[hip] Fix `<complex>` compilation on Windows with VS2019.
Michael Liao [Wed, 20 Jan 2021 19:55:06 +0000 (14:55 -0500)]
[hip] Fix `<complex>` compilation on Windows with VS2019.

Differential Revision: https://reviews.llvm.org/D95075

3 years agoReland "[NPM][Inliner] Factor ImportedFunctionStats in the InlineAdvisor"
Mircea Trofin [Wed, 20 Jan 2021 19:25:43 +0000 (11:25 -0800)]
Reland "[NPM][Inliner] Factor ImportedFunctionStats in the InlineAdvisor"

This reverts commit d97f776be5f8cd3cd446fe73827cd355f6bab4e1.

The original problem was due to build failures in shared lib builds. D95079
moved ImportedFunctionsInliningStatistics under Analysis, unblocking
this.

3 years ago[mlir:async] Fix data races in AsyncRuntime
Eugene Zhulenev [Wed, 20 Jan 2021 13:17:12 +0000 (05:17 -0800)]
[mlir:async] Fix data races in AsyncRuntime

Resumed coroutine potentially can deallocate the token/value/group and destroy the mutex before the std::unique_ptr destructor.

Reviewed By: mehdi_amini

Differential Revision: https://reviews.llvm.org/D95037

3 years ago[gn build] Port 95ce32c7878d
LLVM GN Syncbot [Wed, 20 Jan 2021 21:18:20 +0000 (21:18 +0000)]
[gn build] Port 95ce32c7878d

3 years ago[NFC] Move ImportedFunctionsInliningStatistics to Analysis
Mircea Trofin [Wed, 20 Jan 2021 20:39:55 +0000 (12:39 -0800)]
[NFC] Move ImportedFunctionsInliningStatistics to Analysis

This is related to D94982. We want to call these APIs from the Analysis
component, so we can't leave them under Transforms.

Differential Revision: https://reviews.llvm.org/D95079

3 years ago[flang] Fix creation of deferred shape arrays by POINTER statement
Peter Steinfeld [Wed, 20 Jan 2021 20:34:08 +0000 (12:34 -0800)]
[flang] Fix creation of deferred shape arrays by POINTER statement

It's possible to  declare deferred shape array using the POINTER
statement, for example:

  POINTER :: var(:)

When analyzing POINTER declarations, we were not capturing the array
specification information, if present.  I fixed this by changing the
"Post" function for "parser::PointerDecl" to check to see if the
declaration contained a "DeferredShapeSpecList".  In such cases, I
analyzed the shape and used to information to declare an "ObjectEntity"
that contains the shape information rather than an "UnknownEntity".

I also added a couple of small tests that fail to compile without these
changes.

Differential Revision: https://reviews.llvm.org/D95080

3 years ago[OpenMP][NVPTX] Added forward declaration to pave the way for building deviceRTLs...
Shilei Tian [Wed, 20 Jan 2021 20:55:52 +0000 (15:55 -0500)]
[OpenMP][NVPTX] Added forward declaration to pave the way for building deviceRTLs with OpenMP

Once we switch to build deviceRTLs with OpenMP, primitives and CUDA
intrinsics cannot be used directly anymore because `__device__` is not recognized
by OpenMP compiler. To avoid involving all CUDA internal headers we had in `clang`,
we forward declared these functions. Eventually they will be transformed into
right LLVM instrinsics.

Reviewed By: JonChesterfield

Differential Revision: https://reviews.llvm.org/D95058

3 years ago[CodeView] Emit function types in -gline-tables-only.
Amy Huang [Tue, 12 Jan 2021 00:37:29 +0000 (16:37 -0800)]
[CodeView] Emit function types in -gline-tables-only.

This change adds function types to further differentiate between
FUNC_IDs in -gline-tables-only.

Size increase of object files in clang are
Before: 917990 kb
After:  999312 kb

Bug: https://bugs.llvm.org/show_bug.cgi?id=48432

Differential Revision: https://reviews.llvm.org/D95001

3 years ago[flang] Infrastructure improvements in utility routines
peter klausler [Wed, 20 Jan 2021 01:14:41 +0000 (17:14 -0800)]
[flang] Infrastructure improvements in utility routines

* IsArrayElement() needs another option to control whether it
  should ignore trailing component references.
* Add IsObjectPointer().
* Add const Scope& variants of IsFunction() and IsProcedure().
* Make TypeAndShape::Characterize() work with procedure bindings.
* Handle CHARACTER length in MeasureSizeInBytes().
* Fine-tune FindExternallyVisibleObject()'s handling of dummy arguments
  to conform with Fortran 2018: only INTENT(IN) and dummy pointers
  in pure functions signify; update two tests accordingly.

Also: resolve some stylistic inconsistencies and add a missing
"const" in the expression traversal template framework.

Differential Revision: https://reviews.llvm.org/D95011

3 years ago[mlir][Linalg] NFC - Fully compose map and operands when creating AffineMin in tiling.
Nicolas Vasilache [Wed, 20 Jan 2021 20:29:34 +0000 (20:29 +0000)]
[mlir][Linalg] NFC - Fully compose map and operands when creating AffineMin in tiling.

This may simplify the composition of patterns but is otherwise NFC.

3 years ago[mlir] Add ComplexDialect to SCF->GPU pass.
Alexander Belyaev [Wed, 20 Jan 2021 20:11:49 +0000 (21:11 +0100)]
[mlir] Add ComplexDialect to SCF->GPU pass.

3 years ago[mlir] Fix SubTensorInsertOp semantics
Nicolas Vasilache [Wed, 20 Jan 2021 19:54:02 +0000 (19:54 +0000)]
[mlir] Fix SubTensorInsertOp semantics

Like SubView, SubTensor/SubTensorInsertOp are allowed to have rank-reducing/expanding semantics. In the case of SubTensorInsertOp , the rank of offsets/sizes/strides should be the rank of the destination tensor.

Also, add a builder flavor for SubTensorOp to return a rank-reduced tensor.

Differential Revision: https://reviews.llvm.org/D95076

3 years ago[PredicateInfo] Handle logical and/or
Nikita Popov [Wed, 20 Jan 2021 19:52:23 +0000 (20:52 +0100)]
[PredicateInfo] Handle logical and/or

Teach PredicateInfo to handle logical and/or the same way as
bitwise and/or. This allows handling logical and/or inside IPSCCP
and NewGVN.

3 years ago[lldb/Commands] Align process launch --plugin with process attach (NFC)
Med Ismail Bennani [Wed, 20 Jan 2021 19:29:13 +0000 (20:29 +0100)]
[lldb/Commands] Align process launch --plugin with process attach (NFC)

Following `7169d3a315f4cdc19c4ab6b8f20c6f91b46ba9b8`, this patch updates
the short option for the plugin command option to (`-p` to `-P`) to
align with the `process attach` command options.

The long option remains the same since there are already the same for both
commands.

Signed-off-by: Med Ismail Bennani <medismail.bennani@gmail.com>
3 years ago[PredicateInfo][SCCP][NewGVN] Add tests for logical and/or (NFC)
Nikita Popov [Mon, 11 Jan 2021 19:17:10 +0000 (20:17 +0100)]
[PredicateInfo][SCCP][NewGVN] Add tests for logical and/or (NFC)

Duplicate some existing and/or tests using logical form.

3 years agoReland "[PDB] Defer relocating .debug$S until commit time and parallelize it"
Reid Kleckner [Wed, 20 Jan 2021 19:46:05 +0000 (11:46 -0800)]
Reland "[PDB] Defer relocating .debug$S until commit time and parallelize it"

This reverts commit 5b7aef6eb4b2930971029b984cb2360f7682e5a5 and relands
6529d7c5a45b1b9588e512013b02f891d71bc134.

The ASan error was debugged and determined to be the fault of an invalid
object file input in our test suite, which was fixed by my last change.
LLD's project policy is that it assumes input objects are valid, so I
have added a comment about this assumption to the relocation bounds
check.

3 years ago[mlir][Linalg] NFC - Expose getSmallestBoundingIndex as an utility function
Nicolas Vasilache [Wed, 20 Jan 2021 19:24:49 +0000 (19:24 +0000)]
[mlir][Linalg] NFC - Expose getSmallestBoundingIndex as an utility function

3 years ago[libomptarget][devicertl][nfc] Simplify target_atomic abstraction
Jon Chesterfield [Wed, 20 Jan 2021 19:50:50 +0000 (19:50 +0000)]
[libomptarget][devicertl][nfc] Simplify target_atomic abstraction

[libomptarget][devicertl][nfc] Simplify target_atomic abstraction

Atomic functions were implemented as a shim around cuda's atomics, with
amdgcn implementing those symbols as a shim around gcc style intrinsics.

This patch folds target_atomic.h into target_impl.h and folds amdgcn.

Further work is likely to be useful here, either changing to openmp's atomic
interface or instantiating the templates on the few used types in order to
move them into a cuda/c++ implementation file. This change is mostly to
group the remaining uses of the cuda api under nvptx' target_impl abstraction.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D95062

3 years agoRevert "[lldb] Re-enable TestPlatformProcessConnect on macos"
Pavel Labath [Wed, 20 Jan 2021 19:49:03 +0000 (20:49 +0100)]
Revert "[lldb] Re-enable TestPlatformProcessConnect on macos"

This reverts commit 079e664661770a78e30c0d27a12d50047f1b1ea8. It needs
more work.

3 years ago[COFF] Fix relocation offsets in pdb-file-statics test input
Reid Kleckner [Wed, 20 Jan 2021 19:16:01 +0000 (11:16 -0800)]
[COFF] Fix relocation offsets in pdb-file-statics test input

The relocation offsets were incorrect. I fixed them with llvm-readobj
-codeview -codeview-subsection-bytes, which has a helpful printout of
the relocations that apply to a given symbol record with their offsets.
With this, I was able to update the relocation offsets in the yaml to
fix the line table and the S_DEFRANGE_REGISTER records.

There is still some remaining inconsistency in yaml2obj and obj2yaml
when round tripping MSVC objects, but that isn't a blocker for relanding
D94267.

3 years ago[libomptarget][devicertl][nfc] Remove some cuda intrinsics, simplify
Jon Chesterfield [Wed, 20 Jan 2021 19:45:05 +0000 (19:45 +0000)]
[libomptarget][devicertl][nfc] Remove some cuda intrinsics, simplify

[libomptarget][devicertl][nfc] Remove some cuda intrinsics, simplify

Replace __popc, __ffs with clang intrinsics. Move kmpc_impl_min to only file
that uses it and replace template with explictly typed.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D95060

3 years ago[PredicateInfo] Generalize processing of conditions
Nikita Popov [Mon, 11 Jan 2021 21:18:12 +0000 (22:18 +0100)]
[PredicateInfo] Generalize processing of conditions

Branch/assume conditions in PredicateInfo are currently handled in
a rather ad-hoc manner, with some arbitrary limitations. For example,
an `and` of two `icmp`s will be handled, but an `and` of an `icmp`
and some other condition will not. That also includes the case where
more than two conditions and and'ed together.

This patch makes the handling more general by looking through and/ors
up to a limit and considering all kinds of conditions (though operands
will only be taken for cmps of course).

Differential Revision: https://reviews.llvm.org/D94447

3 years ago[flang][driver] Refactor one unit-test case to use fixtures (nfc)
Andrzej Warzynski [Wed, 20 Jan 2021 18:27:04 +0000 (18:27 +0000)]
[flang][driver] Refactor one unit-test case to use fixtures (nfc)

Move the unit test from InputOutputTest.cpp to FrontendActionTest.cpp
and re-implement it in terms of the FrontendActionTest fixture. This is
just a small code clean-up and a continuation of:
  * https://reviews.llvm.org/D93544

Moving forward, we should try be implementing all unit-test cases for
Flang's frontend actions in terms of FrontendActionTest.

Reviewed By: sameeranjoshi

Differential Revision: https://reviews.llvm.org/D94922

3 years ago[EXTINT][OMP] Fix _ExtInt type checking in device code
Erich Keane [Wed, 20 Jan 2021 19:33:22 +0000 (11:33 -0800)]
[EXTINT][OMP] Fix _ExtInt type checking in device code

_ExtInt gets stuck in the device-type-checking for __int128 if it is
between 65 and 128 bits inclusive.  Anything larger or smaller was
permitted despite this, so this is simply enabling 65-128 bit _ExtInts.
_ExtInt is supported on all our current ABIs, but we stil use the
hasExtIntType in the target info to differentiate here so that it can be
disabled.