[CIR][Dialect] Add convergent attribute to functions for SIMT languages
#840
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
Hugobros3
pushed a commit
to shady-gang/clangir
that referenced
this pull request
Oct 2, 2024
…ages (llvm#840) Fix llvm#805. This PR includes end-to-end implementation. The `convergent` attribute is set depending on languages, which is wrapped as `langOpts.assumeFunctionsAreConvergent()`. Therefore, in ClangIR, every `cir.func` under `#cir.lang<opencl_c>` is set to be convergent. After lowering to LLVM IR, `PostOrderFunctionAttrs` pass will remove unnecessary `convergent` then. In other words, we will still see `convergent` on every function with `-O0`, but not with default optimization level. The test taken from `CodeGenOpenCL/convergent.cl` is a bit complicated. However, the core of it is that `convergent` is set properly for `convfun()` `non_convfun()` `f()` and `g()`. Merge of two `if` is more or less a result of generating the same LLVM IR as OG.
smeenai
pushed a commit
to smeenai/clangir
that referenced
this pull request
Oct 9, 2024
…ages (llvm#840) Fix llvm#805. This PR includes end-to-end implementation. The `convergent` attribute is set depending on languages, which is wrapped as `langOpts.assumeFunctionsAreConvergent()`. Therefore, in ClangIR, every `cir.func` under `#cir.lang<opencl_c>` is set to be convergent. After lowering to LLVM IR, `PostOrderFunctionAttrs` pass will remove unnecessary `convergent` then. In other words, we will still see `convergent` on every function with `-O0`, but not with default optimization level. The test taken from `CodeGenOpenCL/convergent.cl` is a bit complicated. However, the core of it is that `convergent` is set properly for `convfun()` `non_convfun()` `f()` and `g()`. Merge of two `if` is more or less a result of generating the same LLVM IR as OG.
smeenai
pushed a commit
to smeenai/clangir
that referenced
this pull request
Oct 9, 2024
…ages (llvm#840) Fix llvm#805. This PR includes end-to-end implementation. The `convergent` attribute is set depending on languages, which is wrapped as `langOpts.assumeFunctionsAreConvergent()`. Therefore, in ClangIR, every `cir.func` under `#cir.lang<opencl_c>` is set to be convergent. After lowering to LLVM IR, `PostOrderFunctionAttrs` pass will remove unnecessary `convergent` then. In other words, we will still see `convergent` on every function with `-O0`, but not with default optimization level. The test taken from `CodeGenOpenCL/convergent.cl` is a bit complicated. However, the core of it is that `convergent` is set properly for `convfun()` `non_convfun()` `f()` and `g()`. Merge of two `if` is more or less a result of generating the same LLVM IR as OG.
keryell
pushed a commit
to keryell/clangir
that referenced
this pull request
Oct 19, 2024
…ages (llvm#840) Fix llvm#805. This PR includes end-to-end implementation. The `convergent` attribute is set depending on languages, which is wrapped as `langOpts.assumeFunctionsAreConvergent()`. Therefore, in ClangIR, every `cir.func` under `#cir.lang<opencl_c>` is set to be convergent. After lowering to LLVM IR, `PostOrderFunctionAttrs` pass will remove unnecessary `convergent` then. In other words, we will still see `convergent` on every function with `-O0`, but not with default optimization level. The test taken from `CodeGenOpenCL/convergent.cl` is a bit complicated. However, the core of it is that `convergent` is set properly for `convfun()` `non_convfun()` `f()` and `g()`. Merge of two `if` is more or less a result of generating the same LLVM IR as OG.
lanza
pushed a commit
that referenced
this pull request
Nov 5, 2024
…ages (#840) Fix #805. This PR includes end-to-end implementation. The `convergent` attribute is set depending on languages, which is wrapped as `langOpts.assumeFunctionsAreConvergent()`. Therefore, in ClangIR, every `cir.func` under `#cir.lang<opencl_c>` is set to be convergent. After lowering to LLVM IR, `PostOrderFunctionAttrs` pass will remove unnecessary `convergent` then. In other words, we will still see `convergent` on every function with `-O0`, but not with default optimization level. The test taken from `CodeGenOpenCL/convergent.cl` is a bit complicated. However, the core of it is that `convergent` is set properly for `convfun()` `non_convfun()` `f()` and `g()`. Merge of two `if` is more or less a result of generating the same LLVM IR as OG.
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.
Fix #805. This PR includes end-to-end implementation.
The
convergentattribute is set depending on languages, which is wrapped aslangOpts.assumeFunctionsAreConvergent().Therefore, in ClangIR, every
cir.funcunder#cir.lang<opencl_c>is set to be convergent.After lowering to LLVM IR,
PostOrderFunctionAttrspass will remove unnecessaryconvergentthen. In other words, we will still seeconvergenton every function with-O0, but not with default optimization level.The test taken from
CodeGenOpenCL/convergent.clis a bit complicated. However, the core of it is thatconvergentis set properly forconvfun()non_convfun()f()andg(). Merge of twoifis more or less a result of generating the same LLVM IR as OG.