F18 Clause 19.4p9 says:
The associate names of an ASSOCIATE construct have the scope of the
block.
Clause 11.3.1p1 says the ASSOCIATE statement is not itself in the block:
R1102 associate-construct is: associate-stmt block end-associate-stmt
Associate statement associations are currently fully processed from left
to right, incorrectly interposing associating entities earlier in the
list on same-named entities in the host scope.
1 program p
2 logical :: a = .false.
3 real :: b = 9.73
4 associate (b => a, a => b)
5 print*, a, b
6 end associate
7 print*, a, b
8 end
Associating names 'a' and 'b' at line 4 in this code are now both
aliased to logical host entity 'a' at line 2. This happens because the
reference to 'b' in the second association incorrectly resolves 'b' to
the entity in line 4 (already associated to 'a' at line 2), rather than
the 'b' at line 3. With bridge code to process these associations,
f18 output is:
F F
F 9.73
It should be:
9.73 F
F 9.73
To fix this, names in right-hand side selector variables/expressions
must all be resolved before any left-hand side entities are resolved.
This is done by maintaining a stack of lists of associations, rather
than a stack of associations. Each ASSOCIATE statement's list of
assocations is then visited once for right-hand side processing, and
once for left-hand side processing.
Note that other construct associations do not have this problem.
SELECT RANK and SELECT TYPE each have a single assocation, not a list.
Constraint C1113 prohibits the right-hand side of a CHANGE TEAM
association from referencing any left-hand side entity.
Differential Revision: https://reviews.llvm.org/D95010
When using 2 InlinePass instances in the same CGSCC - one for other
mandatory inlinings, the other for the heuristic-driven ones - the order
in which the ImportedFunctionStats would be output-ed would depend on
the destruction order of the inline passes, which is not deterministic.
This patch moves the ImportedFunctionStats responsibility to the
InlineAdvisor to address this problem.
Differential Revision: https://reviews.llvm.org/D94982
It caused "Vector shift amounts must be in the same as their first arg"
asserts in Chromium builds. See the code review for repro instructions.
> Add DemandedElts support inside the TRUNCATE analysis.
>
> Differential Revision: https://reviews.llvm.org/D56387
This reverts commit cad4275d69.
This reverts commit 275f30df8a.
As noted on the code review (https://reviews.llvm.org/D92892), this
change causes us to reject valid code in a few cases. Reverting so we
have more time to figure out what the right fix{es are, is} here.
We already handle "vperm2x128 (ins ?, X, C1), (ins ?, X, C1), 0x31" for shuffling of the upper subvectors, but we weren't dealing with the case when we were splatting the upper subvector from a single source.
Exploits the instruction xxsplti32dx.
It can be used to materialize any 64 bit scalar/vector splat by using two instances, one for the upper 32 bits and the other for the lower 32 bits. It should not materialize the cases which can be materialized by using the instruction xxspltidp.
Differential Revision: https://https://reviews.llvm.org/D90173
This patch refactors the current implementation of
`ProcessLaunchCommandOptions` to be generated by TableGen.
The patch also renames the class to `CommandOptionsProcessLaunch` to
align better with the rest of the codebase style and moves it to
separate files.
Differential Review: https://reviews.llvm.org/D95059
Signed-off-by: Med Ismail Bennani <medismail.bennani@gmail.com>
The patch adapts the rocm runtime wrapper due to subtle differences between the cuda and the rocm/hip runtime api.
Reviewed By: csigg
Differential Revision: https://reviews.llvm.org/D95027
There can be muliple patterns that map to the same compressed
instruction. Reversing those leads to multiple ways to uncompress
an instruction, but its not easily controllable which one will
be chosen by the tablegen backend.
This patch adds a flag to mark patterns that should only be used
for compressing. This allows us to leave one canonical pattern
for uncompressing.
The obvious benefit of this is getting c.mv to uncompress to
the addi patern that is aliased to the mv pseudoinstruction. For
the add/and/or/xor/li patterns it just removes some unreachable
code from the generated code.
Reviewed By: frasercrmck
Differential Revision: https://reviews.llvm.org/D94894
This patch adds support for producer-consumer fusion scenarios with
multiple producer stores to the AffineLoopFusion pass. The patch
introduces some changes to the producer-consumer algorithm, including:
* For a given consumer loop, producer-consumer fusion iterates over its
producer candidates until a fixed point is reached.
* Producer candidates are gathered beforehand for each iteration of the
consumer loop and visited in reverse program order (not strictly guaranteed)
to maximize the number of loops fused per iteration.
In general, these changes were needed to simplify the multi-store producer
support and remove some of the workarounds that were introduced in the past
to support more fusion cases under the single-store producer limitation.
This patch also preserves the existing functionality of AffineLoopFusion with
one minor change in behavior. Producer-consumer fusion didn't fuse scenarios
with escaping memrefs and multiple outgoing edges (from a single store).
Multi-store producer scenarios will usually (always?) have multiple outgoing
edges so we couldn't fuse any with escaping memrefs, which would greatly limit
the applicability of this new feature. Therefore, the patch enables fusion for
these scenarios. Please, see modified tests for specific details.
Reviewed By: andydavis1, bondhugula
Differential Revision: https://reviews.llvm.org/D92876
Replaced CUDA builtin vars with LLVM intrinsics such that we don't need
definitions of those intrinsics.
Reviewed By: JonChesterfield
Differential Revision: https://reviews.llvm.org/D95013
The loop-unswitch transform should not be performed on a loop whose
condition is divergent. For this to happen correctly, divergence
analysis must be available. The existing divergence analysis has not
been ported to the new pass manager yet. As a result, loop unswitching
on the new pass manager is currently unsafe on targets that care about
divergence.
This test is temporarily disabled to unblock work on the new pass
manager. The issue is now tracked in bug 48819.
Reviewed By: foad
Differential Revision: https://reviews.llvm.org/D95051
We were able to remove almost all of the state from
OperationData, so these don't make sense as members
of that class - just pass the RecurKind in as a param.
More streamlining is possible, but I'm trying to avoid
logic/typo bugs while fixing this. Eventually, we should
not need the `OperationData` class.
We were able to remove almost all of the state from
OperationData, so these don't make sense as members
of that class - just pass the RecurKind in as a param.
The documentation for contributing to LLVM currently links to the section
explaining how to submit a Phabricator review using the web interface.
I believe it would be better to link to the general page for using
Phabricator instead, which explains how to sign up with Phabricator,
and also how to submit patches using either the web interface or the
command-line.
I think this is worth changing because what currently *appears* to be our
preferred way of submitting a patch (through the web interface) isn't
actually what we prefer. Indeed, patches submitted from the command-line
have more meta-data available (such as which repository the patch targets),
and also can't suffer from missing context.
Differential Revision: https://reviews.llvm.org/D94929
Loop peeling assumes that the loop's latch is a conditional branch. Add
a check to canPeel that explicitly checks for this, and testcases that
otherwise fail an assertion when trying to peel a loop whose back-edge
is a switch case or the non-unwind edge of an invoke.
Reviewed By: skatkov, fhahn
Differential Revision: https://reviews.llvm.org/D94995
This reduces template bloat, but more importantly, makes it possible to
construct one from clang-query without template types.
Differential Revision: https://reviews.llvm.org/D94879
[libomptarget][devicertl] Wrap source in declare target pragmas
Factored out of D93135 / D94745. C++ and cuda ignore unknown pragmas
so this is a NFC for the current implementation language. Removes noise
from patches for building deviceRTL as openmp.
Reviewed By: tianshilei1992
Differential Revision: https://reviews.llvm.org/D95048
Currently all these tests are XFAILED on Linux even though the problem
only seems to be with the few checks that look at collation. To retain
test coverage this splits the locale-dependent tests into a separate
.pass.cpp that is XFAILed as before.
This commit also XFAILs the locale-dependent tests on FreeBSD since the
[=M=] and [.ch.] behaviour for cs_CZ also doesn't seem to match the
behaviour that is expected by these tests.
Reviewed By: #libc, ldionne
Differential Revision: https://reviews.llvm.org/D94969
Combined with 'da98651 - Revert "DR2064:
decltype(E) is only a dependent', this change (5a391d3) caused verifier
errors when building Chromium. See https://crbug.com/1168494#c1 for a
reproducer.
Additionally it reverts changes that were dependent on this one, see
below.
> Following up on PR48517, fix handling of template arguments that refer
> to dependent declarations.
>
> Treat an id-expression that names a local variable in a templated
> function as being instantiation-dependent.
>
> This addresses a language defect whereby a reference to a dependent
> declaration can be formed without any construct being value-dependent.
> Fixing that through value-dependence turns out to be problematic, so
> instead this patch takes the approach (proposed on the core reflector)
> of allowing the use of pointers or references to (but not values of)
> dependent declarations inside value-dependent expressions, and instead
> treating template arguments as dependent if they evaluate to a constant
> involving such dependent declarations.
>
> This ends up affecting a bunch of OpenMP tests, due to OpenMP
> imprecisely handling instantiation-dependent constructs, bailing out
> early instead of processing dependent constructs to the extent possible
> when handling the template.
>
> Previously committed as 8c1f2d15b8, and
> reverted because a dependency commit was reverted.
This reverts commit 5a391d38ac.
It also restores clang/test/SemaCXX/coroutines.cpp to its state before
da986511fb.
Revert "[c++20] P1907R1: Support for generalized non-type template arguments of scalar type."
> Previously committed as 9e08e51a20, and
> reverted because a dependency commit was reverted. This incorporates the
> following follow-on commits that were also reverted:
>
> 7e84aa1b81 by Simon Pilgrim
> ed13d8c667 by me
> 95c7b6cadb by Sam McCall
> 430d5d8429 by Dave Zarzycki
This reverts commit 4b574008ae.
Revert "[msabi] Mangle a template argument referring to array-to-pointer decay"
> [msabi] Mangle a template argument referring to array-to-pointer decay
> applied to an array the same as the array itself.
>
> This follows MS ABI, and corrects a regression from the implementation
> of generalized non-type template parameters, where we "forgot" how to
> mangle this case.
This reverts commit 18e093faf7.
As discussed on D56387, if we're shifting to extract the upper/lower half of a vXi64 vector then we're actually better off performing this at the subvector level as its very likely to fold into something.
combineConcatVectorOps can perform this in reverse if necessary.
Add the aarch64[_be]-*-gnu_ilp32 targets to support the GNU ILP32 ABI for AArch64.
The needed codegen changes were mostly already implemented in D61259, which added support for the watchOS ILP32 ABI. The main changes are:
- Wiring up the new target to enable ILP32 codegen and MC.
- ILP32 va_list support.
- ILP32 TLSDESC relocation support.
There was existing MC support for ELF ILP32 relocations from D25159 which could be enabled by passing "-target-abi ilp32" to llvm-mc. This was changed to check for "gnu_ilp32" in the target triple instead. This shouldn't cause any issues since the existing support was slightly broken: it was generating ELF64 objects instead of the ELF32 object files expected by the GNU ILP32 toolchain.
This target has been tested by running the full rustc testsuite on a big-endian ILP32 system based on the GCC ILP32 toolchain.
Reviewed By: kristof.beyls
Differential Revision: https://reviews.llvm.org/D94143
This patch pre-commits a test case with wrong exit count
analysis for D92367.
Reviewed by: mkazantsev
Differential Revision: https://reviews.llvm.org/D94657
The pass analysis uses "sets" implemented using a SmallVector type
to keep track of Used, Preserved, Required and RequiredTransitive
passes. When having nested analyses we could end up with duplicates
in those sets, as there was no checks to see if a pass already
existed in the "set" before pushing to the vectors. This idea with
this patch is to avoid such duplicates by avoiding pushing elements
that already is contained when adding elements to those sets.
To align with the above PMDataManager::collectRequiredAndUsedAnalyses
is changed to skip adding both the Required and RequiredTransitive
passes to its result vectors (since RequiredTransitive always is
a subset of Required we ended up with duplicates when traversing
both sets).
Main goal with this is to avoid spending time verifying the same
analysis mulitple times in PMDataManager::verifyPreservedAnalysis
when iterating over the Preserved "set". It is assumed that removing
duplicates from a "set" shouldn't have any other negative impact
(I have not seen any problems so far). If this ends up causing
problems one could do some uniqueness filtering of the vector being
traversed in verifyPreservedAnalysis instead.
Reviewed By: foad
Differential Revision: https://reviews.llvm.org/D94416
NameMatch could be a float close to zero, in such cases we were
dividing by zero and moreover propogating a "NaN" to clients, which is invalid
per JSON.
This fixes the issue by only using Quality scores whenever the NameMatch is low,
as we do in CodeCompletion ranking.
Fixes https://github.com/clangd/clangd/issues/648.
Differential Revision: https://reviews.llvm.org/D94755
This puts it in alphabetical order, matching the rest of the list.
Reviewed by: MaskRay, saugustine
Differential Revision: https://reviews.llvm.org/D94481
In 8031785f4a the temporary object being built was moved to %t/main.o,
but not all run lines were updated to reflect this. Observe the failure
on this buildbot:
http://lab.llvm.org:8011/#/builders/5/builds/3646/steps/9/logs/stdio
It might pass locally for some people due to a stale %t.o hanging around
the build directory.
Add a check if regions do not implement the RegionBranchOpInterface. This is not
allowed in the current deallocation steps. Furthermore, we handle edge-cases,
where a single region is attached and the parent operation has no results.
This fixes: https://bugs.llvm.org/show_bug.cgi?id=48575
Differential Revision: https://reviews.llvm.org/D94586
The runtime-wrappers depend on LLVMSupport, pulling in static initialization code (e.g. command line arguments). Dynamically loading multiple such libraries results in ODR violoations.
So far this has not been an issue, but in D94421, I would like to load both the async-runtime and the cuda-runtime-wrappers as part of a cuda-runner integration test. When doing this, code that asserts that an option category is only registered once fails (note that I've only experienced this in Google's bazel where the async-runtime depends on LLVMSupport, but a similar issue would happen in cmake if more than one runtime-wrapper starts to depend on LLVMSupport).
The underlying issue is that we have a mix of static and dynamic linking. If all dependencies were loaded as shared objects (i.e. if LLVMSupport was linked dynamically to the runtime wrappers), each dependency would only get loaded once. However, linking dependencies dynamically would require special attention to paths (one could dynamically load the dependencies first given explicit paths). The simpler approach seems to be to link all dependencies statically into a single shared object.
This change basically applies the same logic that we have in the c_runner_utils: we have a shared object target that can be loaded dynamically, and we have a static library target that can be linked to other runtime-wrapper shared object targets.
Reviewed By: herhut
Differential Revision: https://reviews.llvm.org/D94399
Summary: This is to address bug48712.
The solution in this patch is that when we want to merge two variable a
into the storage frame of variable b only if the alignment of a is
multiple of b.
There may be other strategies. But now I think they are hard to handle
and benefit little. Or we can implement them in the future.
Test-plan: check-llvm
Reviewers: jmorse, lxfind, junparser
Differential Revision: https://reviews.llvm.org/D94891
If constants are hidden behind G_ANYEXT we can treat them same way as G_SEXT.
For that purpose we extend getConstantVRegValWithLookThrough with option
to handle G_ANYEXT same way as G_SEXT.
Differential Revision: https://reviews.llvm.org/D92219
Jez Ng