[HIP] [JIT] [gfx950] Optimize Minimax M3 scoring & top-k kernels - #4787
Merged
valarLip merged 12 commits intoSep 1, 2026
Merged
Conversation
Contributor
🏷️ CI GuideRuns automatically on every PR:
Extended tests (opt-in via labels):
|
ukannika
marked this pull request as ready for review
August 23, 2026 02:44
github-actions
Bot
changed the title
andyluo7
added a commit
to andyluo7/vllm
that referenced
this pull request
Aug 26, 2026
Integrate the AITER MiniMax M3 score and top-k kernels on the sparse paged-attention stack from vllm-project#52849, including the emitted sparse page table used by TP2/TP4. Gate selection to the exact shapes compiled by ROCm/aiter#4787, preserve actionable FP8 fallback errors, and use the current keyword-only page-table API. Keep the AITER path ROCm-local so CUDA and other platforms retain their existing MSA/Triton selection. Add focused selector, device-isolation, metadata, speculative-decode, mixed-batch, padding, and page-table parity coverage. Depends-on: vllm-project#52849 Depends-on: ROCm/aiter#4787 Co-authored-by: Yamini Preethi Kamisetty <YaminiPreethi.Kamisetty@amd.com> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: OpenAI Codex <codex@openai.com> Signed-off-by: Andy Luo <andy.luo@amd.com>
Contributor
|
The current head Could an AITER maintainer please review #4787 for merge? This is the kernel dependency rather than a duplicate vLLM implementation. |
andyluo7
added a commit
to andyluo7/vllm
that referenced
this pull request
Aug 26, 2026
Integrate the AITER MiniMax M3 score and top-k kernels on the sparse paged-attention stack from vllm-project#52849, including the emitted sparse page table used by TP2/TP4. Gate selection to the exact shapes compiled by ROCm/aiter#4787, preserve actionable FP8 fallback errors, and use the current keyword-only page-table API. Keep the AITER path ROCm-local so CUDA and other platforms retain their existing MSA/Triton selection. Add focused selector, device-isolation, metadata, speculative-decode, mixed-batch, padding, and page-table parity coverage. Depends-on: vllm-project#52849 Depends-on: ROCm/aiter#4787 Co-authored-by: Yamini Preethi Kamisetty <YaminiPreethi.Kamisetty@amd.com> Co-authored-by: Cursor <cursoragent@cursor.com> Co-authored-by: OpenAI Codex <codex@openai.com> Signed-off-by: Andy Luo <andy.luo@amd.com>
valarLip
requested changes
Aug 30, 2026
valarLip
approved these changes
Aug 31, 2026
1 task
weitliao
pushed a commit
to weitliao/vllm
that referenced
this pull request
Aug 31, 2026
Rebase onto vLLM vllm-project#52849 and apply final integration aligned with the 08_24 Docker image: AITER msa_attention indexer (ROCm/aiter#4787), fused sparse page-table emit, and TP>1 support via emits_sparse_block_table. Co-authored-by: Cursor <cursoragent@cursor.com>
4 tasks
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
3 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.
Add fp8 MFMA sparse block scoring and top-k with fused page-table needed for pa gluon kernel.