DeepSeek/FlashMLA
1
0
Fork 0
mirror of https://github.com/deepseek-ai/FlashMLA synced 2025-06-26 18:15:54 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

enable fp8 compile

Browse Source
This commit is contained in:
chenhongmin.will
2025-02-25 17:48:07 +08:00
parent fed0499301
commit 7409203f44
2 changed files with 12 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
setup.py
View File

@@ -37,7 +37,7 @@ ext_modules.append(
sources=[
"csrc/flash_api.cpp",
"csrc/flash_fwd_mla_bf16_sm90.cu",
#"csrc/flash_fwd_mla_fp8_sm90.cu",
"csrc/flash_fwd_mla_fp8_sm90.cu",
],
extra_compile_args={
"cxx": cxx_args,
@@ -55,7 +55,8 @@ ext_modules.append(
"--expt-relaxed-constexpr",
"--expt-extended-lambda",
"--use_fast_math",
"--ptxas-options=-v,--register-usage-level=10"
"--ptxas-options=-v,--register-usage-level=10",
"--ftemplate-backtrace-limit=0"
]
+ cc_flag
),