minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[Fix] Allow flashinfer_sparse_mla DSA backend for HiSparse on SM120 FP8 KV (#33075)

Browse Source
This commit is contained in:
gongwei1027
2026-08-11 15:05:04 -07:00
committed by GitHub
parent 59450c4f18
commit 2c07ca5e8d
3 changed files with 23 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
python/sglang/srt/arg_groups/hisparse_hook.py
+4 -5
View File
@@ -10,7 +10,7 @@ logger = logging.getLogger(__name__)
HISPARSE_CUDA_DSA_BACKENDS_BY_DTYPE = {
"bfloat16": {"flashmla_sparse"},
"fp8_e4m3": {"flashmla_kv"},
"fp8_e4m3": {"flashmla_kv", "flashinfer_sparse_mla"},
}
HISPARSE_ROCM_DSA_BACKENDS = {"tilelang", "aiter"}
HISPARSE_KV_CACHE_DTYPES = ("bfloat16", "fp8_e4m3")
@@ -32,7 +32,7 @@ def _hisparse_allowed_backends(kv_cache_dtype: str) -> set[str]:
if _is_hip():
return HISPARSE_ROCM_DSA_BACKENDS
return HISPARSE_CUDA_DSA_BACKENDS_BY_DTYPE.get(
kv_cache_dtype, {"flashmla_sparse", "flashmla_kv"}
kv_cache_dtype, {"flashmla_sparse", "flashmla_kv", "flashinfer_sparse_mla"}
)
@@ -56,9 +56,8 @@ def validate_hisparse_dsa_backend(
f"HiSparse supports DSA {label} backend(s) {sorted(allowed_backends)} "
f"on this platform with --kv-cache-dtype={kv_cache_dtype}, "
f"but got --dsa-{label}-backend={backend}. "
f"Please use --dsa-{label}-backend="
f"{_hisparse_default_backend(kv_cache_dtype)} "
"or omit it."
f"Please use one of {sorted(allowed_backends)}, or omit the option "
"to let SGLang pick a backend for this platform."
)