[fix][cpu] fix prefill attention in CPU attention backend (#27035)

Signed-off-by: Fadi Arafeh <fadi.arafeh@arm.com>
This commit is contained in:
Fadi Arafeh
2025-10-18 13:30:21 +00:00
committed by GitHub
parent 245e4f2c01
commit ab4be40fc5
2 changed files with 16 additions and 4 deletions
+9 -1
View File
@@ -1293,7 +1293,8 @@ class EngineArgs:
# Set default arguments for V1 Engine.
self._set_default_args(usage_context, model_config)
# Disable chunked prefill for POWER (ppc64le)/ARM/s390x/RISCV CPUs in V1
# Disable chunked prefill and prefix caching for:
# POWER (ppc64le)/ARM/s390x/RISCV CPUs in V1
if current_platform.is_cpu() and current_platform.get_cpu_architecture() in (
CpuArchEnum.POWERPC,
CpuArchEnum.S390X,
@@ -1306,6 +1307,13 @@ class EngineArgs:
"disabling it for V1 backend."
)
self.enable_chunked_prefill = False
logger.info(
"Prefix caching is not supported for ARM and POWER, "
"S390X and RISC-V CPUs; "
"disabling it for V1 backend."
)
self.enable_prefix_caching = False
assert self.enable_chunked_prefill is not None
sliding_window: int | None = None