Identified with bugprone-unused-local-non-trivial-variable.
See https://discourse.llvm.org/t/psa-opty-create-now-with-100-more-tab-complete/87339. I plan to mark these as deprecated in https://github.com/llvm/llvm-project/pull/164649.
This PR shifts from using the LLVM OpenMP enumerator bit flags to an OpenMP dialect specific enumerator. This allows us to better represent map types that wouldn't be of interest to the LLVM backend and runtime in the dialect. Primarily things like ref_ptr/ref_ptee/ref_ptr_ptee/atach_none/attach_always/attach_auto which are of interest to the compiler for certrain transformations (primarily in the FIR transformation passes dealing with mapping), but the runtime has no need to know about them. It also means if another OpenMP implementation comes along they won't need to stick to the same bit flag system LLVM chose/do leg work to address it.
Extends `do concurrent` to OpenMP device mapping by adding support for mapping `reduce` specifiers to omp `reduction` clauses. The changes attach 2 `reduction` clauses to the mapped OpenMP construct: one on the `teams` part of the construct and one on the `wloop` part. - https://github.com/llvm/llvm-project/pull/155754 - https://github.com/llvm/llvm-project/pull/155987 - https://github.com/llvm/llvm-project/pull/155992 - https://github.com/llvm/llvm-project/pull/155993 - https://github.com/llvm/llvm-project/pull/157638 - https://github.com/llvm/llvm-project/pull/156610 ◀️ - https://github.com/llvm/llvm-project/pull/156837
Extends support for mapping `do concurrent` on the device by adding support for `local` specifiers. The changes in this PR map the local variable to the `omp.target` op and uses the mapped value as the `private` clause operand in the nested `omp.parallel` op. - https://github.com/llvm/llvm-project/pull/155754 - https://github.com/llvm/llvm-project/pull/155987 - https://github.com/llvm/llvm-project/pull/155992 - https://github.com/llvm/llvm-project/pull/155993 - https://github.com/llvm/llvm-project/pull/157638 ◀️ - https://github.com/llvm/llvm-project/pull/156610 - https://github.com/llvm/llvm-project/pull/156837
Upstreams further parts of `do concurrent` to OpenMP conversion pass from AMD's fork. This PR extends the pass by adding support for mapping to the device. PR stack: - https://github.com/llvm/llvm-project/pull/155754 - https://github.com/llvm/llvm-project/pull/155987 ◀️ - https://github.com/llvm/llvm-project/pull/155992 - https://github.com/llvm/llvm-project/pull/155993 - https://github.com/llvm/llvm-project/pull/157638 - https://github.com/llvm/llvm-project/pull/156610 - https://github.com/llvm/llvm-project/pull/156837
This commit generalizes `replaceUsesOfBlockArgument` to `replaceAllUsesWith`. In rollback mode, the same restrictions keep applying: a value cannot be replaced multiple times and a call to `replaceAllUsesWith` will replace all current and future uses of the `from` value. `replaceAllUsesWith` is now fully supported and its behavior is consistent with the remaining dialect conversion API. Before this commit, `replaceAllUsesWith` was immediately reflected in the IR when running in rollback mode. After this commit, `replaceAllUsesWith` changes are materialized in a delayed fashion, at the end of the dialect conversion. This is consistent with the `replaceUsesOfBlockArgument` and `replaceOp` APIs. `replaceAllUsesExcept` etc. are still not supported and will be deactivated on the `ConversionPatternRewriter` (when running in rollback mode) in a follow-up commit. Note for LLVM integration: Replace `replaceUsesOfBlockArgument` with `replaceAllUsesWith`. If you are seeing failures, you may have patterns that use `replaceAllUsesWith` incorrectly (e.g., being called multiple times on the same value) or bypass the rewriter API entirely. E.g., such failures were mitigated in Flang by switching to the walk-patterns driver (#156171). You can temporarily reactivate the old behavior by calling `RewriterBase::replaceAllUsesWith`. However, note that that behavior is faulty in a dialect conversion. E.g., the base `RewriterBase::replaceAllUsesWith` implementation does not see uses of the `from` value that have not materialized yet and will, therefore, not replace them.
Fixes #155273 This PR introduces 2 changes: 1. The `do concurrent` to OpenMP pass is now a module pass rather than a function pass. 2. Reduction ops are looked up in the parent module before being created. The benefit of using a module pass is that the same reduction operation can be used across multiple functions if the reduction type matches.
See https://github.com/llvm/llvm-project/pull/147168 for more info.
Now that we have changes introduced by #145837, mapping reductions from `do concurrent` to OpenMP is almost trivial. This PR adds such mapping. PR stack: - https://github.com/llvm/llvm-project/pull/145837 - https://github.com/llvm/llvm-project/pull/146025 - https://github.com/llvm/llvm-project/pull/146028 - https://github.com/llvm/llvm-project/pull/146033 (this one)