[CPU][Perf] Accelerate Attention head for s390x using vector intrinsics (#34434)

Signed-off-by: Rehan Khan <Rehan.Khan7@ibm.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
This commit is contained in:
R3hankhan
2026-02-24 07:25:39 -08:00
committed by GitHub
co-authored by Li, Jiang <jiang1.li@intel.com>
parent 0de5333989
commit 34ce0ffd1f
6 changed files with 424 additions and 8 deletions
+3 -4
View File
@@ -2017,21 +2017,20 @@ class EngineArgs:
)
# Disable chunked prefill and prefix caching for:
# POWER (ppc64le)/s390x/RISCV CPUs in V1
# POWER (ppc64le)/RISCV CPUs in V1
if current_platform.is_cpu() and current_platform.get_cpu_architecture() in (
CpuArchEnum.POWERPC,
CpuArchEnum.S390X,
CpuArchEnum.RISCV,
):
logger.info(
"Chunked prefill is not supported for POWER, "
"S390X and RISC-V CPUs; "
"and RISC-V CPUs; "
"disabling it for V1 backend."
)
self.enable_chunked_prefill = False
logger.info(
"Prefix caching is not supported for POWER, "
"S390X and RISC-V CPUs; "
"and RISC-V CPUs; "
"disabling it for V1 backend."
)
self.enable_prefix_caching = False