apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-06 09:07:10 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

supports dynamic Cfp8 (#3767)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
Details
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
Details
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
Details
CE Compile Job / Show Code Archive Output (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
Details
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Details
Deploy GitHub Pages / deploy (push) Has been cancelled
Details

Browse Source 
* supports dynamic Cfp8

* add unittest
This commit is contained in:
lzy
2025-09-08 11:41:29 +08:00
committed by GitHub
parent b5e20e3015
commit af49b81ffd
20 changed files with 1417 additions and 225 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
custom_ops/gpu_ops/append_attn/append_attention_kernel.h
View File

@@ -103,6 +103,7 @@ void CascadeAppendAttentionC8Kernel(
const bool causal,
const bool is_decoder,
const bool enable_prefill,
const std::string& cache_quant_type_str,
cudaStream_t& stream,
paddle::Tensor* out);
@@ -264,9 +265,10 @@ void CascadeAppendAttentionKernel(
causal,
is_decoder,
enable_prefill,
cache_quant_type_str,
stream,
out);
} else if (cache_quant_type_str == "cache_fp8") {
} else if (cache_quant_type_str == "cache_fp8" or cache_quant_type_str == "block_wise_fp8") {
CascadeAppendAttentionC8Kernel<T, OutT, true>(meta_data,
qkv,
cache_k,
@@ -299,6 +301,7 @@ void CascadeAppendAttentionKernel(
causal,
is_decoder,
enable_prefill,
cache_quant_type_str,
stream,
out);
} else if (cache_quant_type_str == "cache_int4_zp") {