apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

fix split_rope_cache_kv_encoder in mix mtp (#5384)

Browse Source
This commit is contained in:
cmcamdy
2025-12-05 14:33:17 +08:00
committed by GitHub
parent b5a7abe624
commit 86b6430582
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
custom_ops/xpu_ops/src/ops/block_attn.cc
View File

@@ -585,7 +585,8 @@ std::vector<paddle::Tensor> BlockAttnKernel(
int,
E_Scale>(
xpu_ctx->x_context(),
reinterpret_cast<const XPU_XType*>(qkv.data<data_t>()), // qkv
reinterpret_cast<const XPU_XType*>(qkv.data<data_t>()) +
total_enc_len * qkv_shape[qkv_shape.size() - 1], // qkv
reinterpret_cast<const float*>(
rotary_embs.data<float>()), // rotary_pos_emb
reinterpret_cast<const int*>(