[XPU][TritonGPUToLLVM] Use reqd_work_group_size
#2845
Draft
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
Use `llvm.func`'s `reqd_work_group_size` instead of `max_work_group_size` to express known work-group size information. Signed-off-by: victor-eds <victor.perez@codeplay.com>
victor-eds
force-pushed
the
use-reqd-work-group-size
branch
from
109de74 to
3d95a81
Compare
|
I found why setting reqd_work_group_size fails. Basically, IGC expects a non-standard format for that metadata in the LLVM module, so it has to preprocess it. Preprocessing only takes place if the kernel is listed under opencl.kernels metadata. As preprocessing is not taking place, IGC sees the attribute is present as the function is marked, so it tries to retrieve that info, but it just returns 0 as the metadata hasn't been preprocessed. In the end, the static work-group size is 0 always. Once #2869 is completed, this PR can be continued. |
|
Blocked. Related to #2923 (comment) |
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Outdated suggestions cannot be applied.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.
Use
llvm.func'sreqd_work_group_sizeinstead ofmax_work_group_sizeto express known work-group size information.