Bump to f355cd6f6c51 [Nr 5] #242
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This adds the `CExpression` trait to additional ops to allow to use these ops within the expression operation. Furthermore, the operator precedence is defined for those ops.
Summary: Some dependencies on the standard C extensions are added transitively. This patch adds the new values.
…lvm#80378) This patch is stacked on llvm#80372, llvm#80307, and llvm#80306. ShuffleVector on scalable vector types gets IRTranslate'd to G_SPLAT_VECTOR since a ShuffleVector that has operates on scalable vectors is a splat vector where the value of the splat vector is the 0th element of the first operand, because the index mask operand is the zeroinitializer (undef and poison are treated as zeroinitializer here). This is analogous to what happens in SelectionDAG for ShuffleVector. `buildSplatVector` is renamed to`buildBuildVectorSplatVector`. I did not make this a separate patch because it would cause problems to revert that change without reverting this change too.
…4270) This switches the default of `LLDB_TEST_USE_VENDOR_PACKAGES` from `ON` to `OFF` in preparation for eventually deleting it. All known LLDB buildbots have this package installed, so flipping the default will uncover any other users. If this breaks anything, the preferred fix is to install `pexpect` on the host system. The second fix is to build with cmake option `-DLLDB_TEST_USE_VENDOR_PACKAGES=ON` as a temporary measure until `pexpect` can be installed. If neither of those work, reverting this patch is OK.
Details in the linked issue. Might fail on other architectures but I can't confirm, they can add to this if it does.
…r types" (llvm#84330) Reverts llvm#80378 causing Buildbot failures that did not show up with check-llvm or CI.
This improves overall analysis for minbitwidth in SLP. It allows to
analyze the trees with store/insertelement root nodes. Also, instead of
using single minbitwidth, detected from the very first analysis stage,
it tries to detect the best one for each trunc/ext subtree in the graph
and use it for the subtree.
Results in better code and less vector register pressure.
Metric: size..text
Program size..text
results results0 diff
test-suite :: SingleSource/Benchmarks/Adobe-C++/simple_types_loop_invariant.test 92549.00 92609.00 0.1%
test-suite :: External/SPEC/CINT2017speed/625.x264_s/625.x264_s.test 663381.00 663493.00 0.0%
test-suite :: External/SPEC/CINT2017rate/525.x264_r/525.x264_r.test 663381.00 663493.00 0.0%
test-suite :: MultiSource/Benchmarks/Bullet/bullet.test 307182.00 307214.00 0.0%
test-suite :: External/SPEC/CFP2017speed/638.imagick_s/638.imagick_s.test 1394420.00 1394484.00 0.0%
test-suite :: External/SPEC/CFP2017rate/538.imagick_r/538.imagick_r.test 1394420.00 1394484.00 0.0%
test-suite :: External/SPEC/CFP2017rate/510.parest_r/510.parest_r.test 2040257.00 2040273.00 0.0%
test-suite :: External/SPEC/CFP2017rate/526.blender_r/526.blender_r.test 12396098.00 1239585.00 -0.0%
test-suite :: External/SPEC/CINT2006/445.gobmk/445.gobmk.test 909944.00 909768.00 -0.0%
SingleSource/Benchmarks/Adobe-C++/simple_types_loop_invariant - 4 scalar
instructions remain scalar (good).
Spec2017/x264 - the whole function idct4x4dc is vectorized using <16
x i16> instead of <16 x i32>, also zext/trunc are removed. In other
places last vector zext/sext removed and replaced by
extractelement + scalar zext/sext pair.
MultiSource/Benchmarks/Bullet/bullet - reduce or <4 x i32> replaced by
reduce or <4 x i8>
Spec2017/imagick - Removed extra zext from 2 packs of the operations.
Spec2017/parest - Removed extra zext, replaced by extractelement+scalar
zext
Spec2017/blender - the whole bunch of vector zext/sext replaced by
extractelement+scalar zext/sext, some extra code vectorized in smaller
types.
Spec2006/gobmk - fixed cost estimation, some small code remains scalar.
Reviewers: RKSimon
Pull Request: llvm#84334
Summary: The clang-offload-bundler uses an empty file to control the bundles made for embedding. Previously this still used `/dev/null` by mistake even on Windows.
…llvm#84223) This is in preparation for an upcoming commit that will add "funclet" OpBundle to the inserted runtime calls where the function's EH personality requires it. See PR llvm#82533
…lvm#84142) Arithmetic constants for vector types can be constructed from objects implementing Python buffer protocol such as `array.array`. Note that until Python 3.12, there is no typing support for buffer protocol implementers, so the annotations use array explicitly. Reverts llvm#84103
…vm#83643) As per the OpenMP standard, "If a variable appears in a link clause on a declare target directive that does not have a device_type clause with the nohost device-type-description then it is treated as if it had appeared in a map clause with a map-type of tofrom" is an implicit mapping rule. Before this change, such variables were mapped as to by default.
Instead of the longer ArrayElemPtr + Load.
…#83458) Clean-up InputSignal checks.
llvm#82985) This adds a rewrite that converts illegal 2D unit-dim `shape_casts` into `vector.transpose` ops. E.g. ```mlir // Case 1: %a = vector.shape_cast %0 : vector<[4]x1xf32> to vector<1x[4]xf32> // Case 2: %b = vector.shape_cast %1 : vector<[4]x1xf32> to vector<[4]xf32> ``` Becomes: ```mlir // Case 1: %a = vector.transpose %0 : [1, 0] vector<[4]x1xf32> to vector<1x[4]xf32> // Case 2: %t = vector.transpose %1 : [1, 0] vector<[4]x1xf32> to vector<1x[4]xf32> %b = vector.shape_cast %t : vector<1x[4]xf32> to vector<[4]xf32> ``` Various lowerings and drop unit-dims patterns add such shape_casts, however, if they do not cancel out (which they likely won't if we've reached the vector-legalization pass) they will prevent lowering the IR. Rewriting them as a transpose gives `LiftIllegalVectorTransposeToMemory` a chance to eliminate the illegal types.
…4332) Summary: The new driver does not need this hash and it can lead to redefined symbol errors when the CUID hash isn't set.
This would have helped identify problems with llvm#83905 which only showed up in an LLVM_ENABLE_EXPENSIVE_CHECKS build.
…lvm#80378) Recommits llvm#80378 which was reverted in llvm#84330. The problem was that the change in llvm/test/CodeGen/AArch64/GlobalISel/legalizer-info-validation.mir used 217 as an opcode instead of a regex.
This PR adds the following to the mlir c api: - The disctinct mlir builtin attribute. - LLVM attributes (mostly debug related ones)
Targets the dynamic realignment pattern of `(Ptr + Align - 1) & -Align;` as implemented by gep then ptrmask. Specifically, when the pointer already has alignment information, dynamically realigning it to less than is already known should be a no-op. Discovered while writing test cases for another patch. For the zero low bits of a known aligned pointer, adding the gep index then removing it with a mask is a no-op. Folding the ptrmask effect entirely into the gep is the ideal result as that unblocks other optimisations that are not aware of ptrmask. In some other cases the gep is known to be dead and is removed without changing the ptrmask. In the least effective case, this transform creates a new gep with a rounded-down index and still leaves the ptrmask unchanged. That simplified gep is still a minor improvement, geps are cheap and ptrmask occurs in address calculation contexts so I don't think it's worth special casing to avoid the extra instruction.
The buildbot seems to complain about `strcmp` function not available in the vfork patch (llvm#81564): https://lab.llvm.org/buildbot/#/builders/68/builds/70093/steps/6/logs/stdio Unfortunately, I can't reproduce the failure on my linux machine so this is a guessing fix. If anyone has a way to reproduce and very this fix, please feel free to merge this change. Co-authored-by: jeffreytan81 <jeffreytan@fb.com>
…2->s64 anyext load." This reverts commit 7524ad9. Broke sanitizer build bots, e.g. https://lab.llvm.org/buildbot/#/builders/5/builds/41588/steps/9/logs/stdio
Currently, progress events reported by the ProgressManager and broadcast to eBroadcastBitProgressCategory always specify they're complete. The problem is that the ProgressManager reports kNonDeterministicTotal for both the total and the completed number of (sub)events. Because the values are the same, the event reports itself as complete. This patch fixes the issue by reporting 0 as the completed value for the start event and kNonDeterministicTotal for the end event.
We already import Record.h.
Combined constructs are decomposed into separate operations. However, this does not adhere to `acc` dialect's goal to be able to regenerate semantically equivalent clauses as user's intent. Thus, add an attribute to keep track of the combined constructs.
Removes an unused field. Retypes unshared smart pointers to `unique_ptr`.
Allows linalg structured operations to be handled during spmdization and sharding propagation. There is only support for projected permutation indexing maps.
…m#83922) During block signature conversion, a new block is inserted and ops are moved from the old block to the new block. This commit changes the implementation such that ops are moved in bulk (`splice`) instead of one-by-one; that's what `splitBlock` is doing. This also makes it possible to pass the new block argument types directly to `createBlock` instead of using `addArgument` (which bypasses the rewriter). This doesn't change anything from a technical point of view (there is no rewriter API for adding arguments at the moment), but the implementation reads a bit nicer.
…3425) This commit adds listener support to the dialect conversion. Similarly to the greedy pattern rewrite driver, an optional listener can be specified in the configuration object. Listeners are notified only if the dialect conversion succeeds. In case of a failure, where some IR changes are first performed and then rolled back, no notifications are sent. Due to the fact that some kinds of rewrite are reflected in the IR immediately and some in a delayed fashion, there are certain limitations when attaching a listener; these are documented in `ConversionConfig`. To summarize, users are always notified about all rewrites that happened, but the notifications are sent all at once at the very end, and not interleaved with the actual IR changes. This change is in preparation improvements to `transform.apply_conversion_patterns`, which currently invalidates all handles. In the future, it can use a listener to update handles accordingly, similar to `transform.apply_patterns`.
llvm-project/mlir/lib/Dialect/Linalg/Transforms/MeshShardingInterfaceImpl.cpp:96:8:
error: unused variable 'resultElementType' [-Werror,-Wunused-variable]
Type resultElementType =
^
llvm-project/mlir/lib/Dialect/Linalg/Transforms/MeshShardingInterfaceImpl.cpp:122:1:
error: non-void function does not return a value in all control paths [-Werror,-Wreturn-type]
}
^
2 errors generated.
Close llvm#71034 See https://discourse.llvm.org/t/rfc-c-20-modules-introduce-thin-bmi-and-decls-hash/74755 This patch introduces reduced BMI, which doesn't contain the definitions of functions and variables if its definitions won't contribute to the ABI. Testing is a big part of the patch. We want to make sure the reduced BMI contains the same behavior with the existing and relatively stable fatBMI. This is pretty helpful for further reduction. The user interfaces part it left to following patches to ease the reviewing.
…umer (NFC)
llvm-project/clang/lib/Frontend/FrontendActions.cpp:213:10:
error: moving a local object in a return statement prevents copy elision [-Werror,-Wpessimizing-move]
213 | return std::move(Consumers);
| ^
/Users/jiefu/llvm-project/clang/lib/Frontend/FrontendActions.cpp:213:10: note: remove std::move call here
213 | return std::move(Consumers);
| ^~~~~~~~~~ ~
1 error generated.
The test case demonstrates how meta variables are needlessly renamed, making diffs harder to read.
The match argument is unused.
The match argument isn't used.
Prior to this change, running UTC on larger tests, especially tests with unnamed IR values, often resulted in a spuriously large diff because e.g. TMPnn variables in the CHECK lines were renumbered. This change attempts to reduce the diff by keeping those variable names the same. There are cases in which this "drift" of variable names can end up being more confusing. The old behavior can be re-enabled with the --reset-variable-names command line argument. The improvement may not be immediately apparent in the diff of this change. The point is that the diff of stable_ir_values.ll against stable_ir_values.ll.expected after this change is smaller. Ideally, we'd also keep meta variables for "global" objects stable, e.g. for attributes (#nn) and metadata (!nn). However, that would require a much more substantial refactoring of how we generate check lines, so I left it for future work.
PR llvm#75982 had been created before these tests were added, therefore some test were not updated.
This reverts commit fb02f9a. Looks like some Python version incompatibility, will investigate.
Resubmitting this after previous revert with the following changes: - Split table into table_rhs_idx and table_candidate_idx so that bisect.bisect_left can be used without the `key` argument, which was introduced in Python 3.10 - Remove a re.Pattern type annotation Original commit message: Prior to this change, running UTC on larger tests, especially tests with unnamed IR values, often resulted in a spuriously large diff because e.g. TMPnn variables in the CHECK lines were renumbered. This change attempts to reduce the diff by keeping those variable names the same. There are cases in which this "drift" of variable names can end up being more confusing. The old behavior can be re-enabled with the --reset-variable-names command line argument. The improvement may not be immediately apparent in the diff of this change. The point is that the diff of stable_ir_values.ll against stable_ir_values.ll.expected after this change is smaller. Ideally, we'd also keep meta variables for "global" objects stable, e.g. for attributes (#nn) and metadata (!nn). However, that would require a much more substantial refactoring of how we generate check lines, so I left it for future work.
Older versions of clang do not have __builtin_complex, but they may define `__GNUC__`.
…s. (llvm#84187) With the many pseudos used in SVE codegen it can be too easy to miss instructions. This enables the existing test we have for checking the scheduling info of the pseudos matches the real instructions, and adjusts the scheduling info in the NeoverseV1 model to make sure all are handled. In the cases I could I opted to use the same info as in the NeoverseV2 model, to keep the differences smaller.
…this` with values. (llvm#84164) This is the constructor's job, and we want to be able to test that it does this.
…y value. (llvm#84317) I'm making some changes to `Environment::getResultObjectLocation()`, with the ultimate goal of eliminating `RecordValue` entirely, and I'd like to make sure I don't break this behavior (and I've realized we don't have a test for it yet).
These checks have been broken since 6afe972. The check_cxx_compiler_flag macro only takes two arguments and passing three to it ends up calling `cmake_check_compiler_flag(CXX "${_FLAG}" ${_RESULT})` with ${_FLAG} equal to `-Werror` and the result variable being the actually tested compiler flag. I noticed this because some of the flags that I know should be supported were being flagged as not supported. `--debug-trycompile` shows the following surprising line in the generated CMakeLists.txt: `add_definitions([==[-D-Wempty-body]==] [==[-Werror]==])` which then results in the following error while running the check: ``` FAILED: CMakeFiles/cmTC_72736.dir/src.cxx.o tmp/upstream-llvm-readonly/bin/clang++ -nodefaultlibs -std=c++17 -fcolor-diagnostics -D-Wempty-body -Werror -MD -MT CMakeFiles/cmTC_72736.dir/src.cxx.o -MF CMakeFiles/cmTC_72736.dir/src.cxx.o.d -o CMakeFiles/cmTC_72736.dir/src.cxx.o -c .../cmake-build-all-sanitizers/CMakeFiles/CMakeScratch/TryCompile-nyh3QR/src.cxx In file included from <built-in>:450: <command line>:1:9: error: macro name must be an identifier 1 | #define -Wempty-body 1 | ^ 1 error generated. ``` It would be great if CMake could be a bit more helpful here so I've filed https://gitlab.kitware.com/cmake/cmake/-/issues/25735. See also https://reviews.llvm.org/D146920. Reviewed By: nikic Pull Request: llvm#83779
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Small conflict resolution