Skip to content

[Graph Optimization] Support CUDAGraph Padding + MTP #4545

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Oct 23, 2025
Merged

[Graph Optimization] Support CUDAGraph Padding + MTP #4545

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d873614
Support CUDAGraph Padding + MTP
gongshaotian Oct 22, 2025
567228e
Merge branch 'develop' of https://github.com/PaddlePaddle/FastDeploy …
gongshaotian Oct 23, 2025
0661ebe
support orther write cache kernel
gongshaotian Oct 23, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions custom_ops/gpu_ops/append_attn/speculate_write_cache_with_rope_impl.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -627,6 +627,7 @@ __global__ void append_speculate_cache_fp8_rope_qk_norm_dynamic_kernel(
const int token_id = blockIdx.x;

const int bid = batch_id_per_token[token_id];
if (bid == -1) return; // NOTE(gongshaotian): For CUDAGraph padding

const int start_token_idx = cu_seqlens_q[bid];
const int head_idx = blockIdx.y * NUM_WARPS + wid;
Expand Down
Loading