[FLYDSL] mfma16_hip GDR K5 prefill chunk_gdn_fwd_h for MI308#4353
Open
huizzhan wants to merge 2 commits into
Open
[FLYDSL] mfma16_hip GDR K5 prefill chunk_gdn_fwd_h for MI308#4353huizzhan wants to merge 2 commits into
huizzhan wants to merge 2 commits into
Conversation
Trim the FlyDSL GDN prefill down to the mfma16_hip fork for the MI308/gfx942 target: drop the other forks (kv/mfma32_vk/2wave/3wave/naive/naive_opt/hipport) and the flydsl csv/heuristic BV logic; BV selection now uniformly reuses the hip K5 _select_bv_for_varlen/_dense. Test kept to the mfma16_hip correctness suite; run script runs it. Co-authored-by: Cursor <cursoragent@cursor.com>
Contributor
🏷️ CI GuideRuns automatically on every PR:
Extended tests (opt-in via labels):
|
…BV select The mfma16_hip-only dispatch now uses the hip K5 BV selector, so the flydsl csv/heuristic BV constants (_BV_CANDIDATES, _TUNED_BV_*, _GFX942/950 CTA targets, _IS_GFX950) and their comments are unreferenced dead code -- drop them (keep _IS_GFX942, used by the sched_gfx942 gate). Co-authored-by: Cursor <cursoragent@cursor.com>
huizzhan
changed the title
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Trim the FlyDSL GDN prefill down to the mfma16_hip fork for the MI308/gfx942 target: drop the other forks (kv/mfma32_vk/2wave/3wave/naive/naive_opt/hipport) and the flydsl csv/heuristic BV logic; BV selection now uniformly reuses the hip K5 _select_bv_for_varlen/_dense. Test kept to the mfma16_hip correctness suite; run script runs it.
Motivation
Technical Details
Test Plan
Test Result
Submission Checklist