[CPU] Fix FP8 attention scratchpad sizing (#50194)
Signed-off-by: Li, Tianmu <tianmu.li@intel.com> Co-authored-by: Codex <noreply@openai.com> Co-authored-by: Li, Jiang <jiang1.li@intel.com>
This commit is contained in:
co-authored by
Codex
Li, Jiang <jiang1.li@intel.com>
parent
100d655a23
commit
65a1a16594
@@ -3814,6 +3814,7 @@ def cpu_attn_get_scheduler_metadata(
|
||||
isa: str,
|
||||
enable_kv_split: bool,
|
||||
dynamic_causal: torch.Tensor | None = None,
|
||||
kv_cache_dtype: str = "auto",
|
||||
) -> torch.Tensor:
|
||||
scheduler_metadata = torch.ops._C.get_scheduler_metadata(
|
||||
num_reqs,
|
||||
@@ -3828,6 +3829,7 @@ def cpu_attn_get_scheduler_metadata(
|
||||
isa,
|
||||
enable_kv_split,
|
||||
dynamic_causal,
|
||||
kv_cache_dtype,
|
||||
)
|
||||
return scheduler_metadata
|
||||
|
||||
|
||||
Reference in New Issue
Block a user