perf(amdgpu): inline offloaded range-for kernel body#12
Closed
kevinjosephamd wants to merge 1 commit intoamd-integrationfrom
Closed
perf(amdgpu): inline offloaded range-for kernel body#12kevinjosephamd wants to merge 1 commit intoamd-integrationfrom
kevinjosephamd wants to merge 1 commit intoamd-integrationfrom
Conversation
Collaborator
|
/run-ci |
jamesETsmith
approved these changes
Apr 24, 2026
Collaborator
jamesETsmith
left a comment
jamesETsmith
left a comment
There was a problem hiding this comment.
This looks good pending CI runs after #3 goes in
Mark the per-iteration loop body emitted by create_offload_range_for as alwaysinline so the AMDGPU backend inlines it into gpu_parallel_range_for instead of leaving it as a separate function call. Measured throughput improvement on an internal end-to-end workload.
gpinkert
force-pushed
the
kejoseph/perf/inline-range-for-body
branch
from
4e5bc24 to
895dfc3
Compare
|
/run-ci |
1 similar comment
Collaborator
|
/run-ci |
Author
|
Closing this PR out since we're unable to replicate the performance improvements originally observed. |
Collaborator
|
/run-ci |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Mark the per-iteration loop body emitted by create_offload_range_for as alwaysinline so the AMDGPU backend inlines it into gpu_parallel_range_for instead of leaving it as a separate function call. Measured throughput improvement on an internal end-to-end workload.