Skip to content

[None][fix] Remove MHC fused hidden-size guard #13611

Merged
lfr-0531 merged 1 commit into
NVIDIA:feat/deepseek_v4from
mingyangHao:mingyangh/remove-mhc-fused-hidden-size-guard
Apr 29, 2026
Merged

[None][fix] Remove MHC fused hidden-size guard #13611
lfr-0531 merged 1 commit into
NVIDIA:feat/deepseek_v4from
mingyangHao:mingyangh/remove-mhc-fused-hidden-size-guard

Conversation

@mingyangHao
Copy link
Copy Markdown
Collaborator

@mingyangHao mingyangHao commented Apr 29, 2026

@coderabbitai summary

Description

  • Drop the hidden_size == FHC_HIDDEN runtime guard in mhcFusedHcLaunch and mhcFusedHcAllInOneLaunch; the templated kernel already specializes on FHC_HIDDEN so the check is redundant for the configurations
    we ship and blocks valid call sites.
  • The companion hc_mult == FHC_HC_MULT guard is kept.

Test Coverage

PR Checklist

Please review the following before submitting your PR:

  • PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

  • PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

  • Test cases are provided for new code paths (see test instructions)

  • Any new dependencies have been scanned for license and vulnerabilities

  • CODEOWNERS updated if ownership changes

  • Documentation updated as needed

  • Update tava architecture diagram if there is a significant design change in PR.

  • The reviewers assigned automatically/manually are appropriate for the PR.

  • Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

Signed-off-by: Mingyang Hao <mingyangHao@users.noreply.github.com>
@mingyangHao mingyangHao changed the title fix: Remove MHC fused hidden-size guard [None][fix] Remove MHC fused hidden-size guard Apr 29, 2026
@mingyangHao mingyangHao marked this pull request as draft April 29, 2026 11:14
@mingyangHao mingyangHao marked this pull request as ready for review April 29, 2026 11:14
@lfr-0531 lfr-0531 changed the title [None][fix] Remove MHC fused hidden-size guard [None][fix] Remove MHC fused hidden-size guard Apr 29, 2026
@mingyangHao mingyangHao changed the title [None][fix] Remove MHC fused hidden-size guard [None][fix] Remove MHC fused hidden-size guard Apr 29, 2026
@lfr-0531 lfr-0531 merged commit 7aa1771 into NVIDIA:feat/deepseek_v4 Apr 29, 2026
6 of 13 checks passed
lfr-0531 pushed a commit that referenced this pull request May 7, 2026
Signed-off-by: Mingyang Hao <mingyangHao@users.noreply.github.com>
Co-authored-by: Mingyang Hao <mingyangHao@users.noreply.github.com>
lfr-0531 pushed a commit that referenced this pull request May 14, 2026
Signed-off-by: Mingyang Hao <mingyangHao@users.noreply.github.com>
Co-authored-by: Mingyang Hao <mingyangHao@users.noreply.github.com>
Signed-off-by: Fanrong Li <lfr-0531@users.noreply.github.com>
lfr-0531 pushed a commit to lfr-0531/TensorRT-LLM that referenced this pull request May 29, 2026
Signed-off-by: Mingyang Hao <mingyangHao@users.noreply.github.com>
Co-authored-by: Mingyang Hao <mingyangHao@users.noreply.github.com>
Signed-off-by: Fanrong Li <lfr-0531@users.noreply.github.com>
(cherry picked from commit b3f45bb)
Signed-off-by: Fanrong Li <lfr-0531@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants