Li, Jiang 05339a7b20 [Bugfix][CPU] Fix llama4 inference on CPU (#34321) ...

Signed-off-by: jiang1.li <jiang1.li@intel.com>

2026-02-11 19:07:23 +08:00

..

micro_gemm

…

sgl-kernels

[Bugfix]: Fix display errors in TORCH_CHECK messages (#32942)

2026-01-31 09:48:48 -08:00

activation.cpp

…

cpu_arch_macros.h

…

cpu_attn_amx.hpp

[CPU] Split attention dispatch by head_dim alignment (#32161)

2026-02-03 19:37:15 -08:00

cpu_attn_impl.hpp

[CPU] Enable FP16 (Half dtype) support for s390x (#34116)

2026-02-11 14:41:42 +08:00

cpu_attn_neon_bfmmla.hpp

[cpu][performance] CPU Paged Attention NEON BFMMLA BF16 Implementation (#32263)

2026-02-06 15:01:48 +08:00

cpu_attn_neon.hpp

[cpu][performance] CPU Paged Attention NEON BFMMLA BF16 Implementation (#32263)

2026-02-06 15:01:48 +08:00

cpu_attn_vec16.hpp

…

cpu_attn_vec.hpp

…

cpu_attn.cpp

[CPU] Split attention dispatch by head_dim alignment (#32161)

2026-02-03 19:37:15 -08:00

cpu_fused_moe.cpp

[Bugfix][CPU] Fix llama4 inference on CPU (#34321)

2026-02-11 19:07:23 +08:00

cpu_types_arm.hpp

[Feature][CPU Backend]: Optimize ARM vectorization backend (#30329)

2026-02-02 20:17:56 -08:00

cpu_types_scalar.hpp

…

cpu_types_vsx.hpp

…

cpu_types_vxe.hpp

[CPU] Enable FP16 (Half dtype) support for s390x (#34116)

2026-02-11 14:41:42 +08:00

cpu_types_x86.hpp

…

cpu_types.hpp

…

cpu_wna16.cpp

[Doc]: fixing multiple typos in diverse files (#33256)

2026-01-29 16:52:03 +08:00

dnnl_helper.cpp

[Fix] [CPU Backend] : Prepack weights for w8a8 oneDNN matmul (#33901)

2026-02-09 18:04:41 +08:00

dnnl_helper.h

…

dnnl_kernels.cpp

[Feature][CPU Backend]: Optimize ARM vectorization backend (#30329)

2026-02-02 20:17:56 -08:00

float_convert.hpp

…

generate_cpu_attn_dispatch.py

[CPU] Split attention dispatch by head_dim alignment (#32161)

2026-02-03 19:37:15 -08:00

layernorm.cpp

…

mla_decode.cpp

[CPU] Enable FP16 (Half dtype) support for s390x (#34116)

2026-02-11 14:41:42 +08:00

pos_encoding.cpp

…

shm.cpp

[Bugfix][CPU] Fix thread num for shared memory communication (#33317)

2026-01-29 03:26:58 -08:00

torch_bindings.cpp

[Bugfix][CPU] Fix llama4 inference on CPU (#34321)

2026-02-11 19:07:23 +08:00

utils.cpp

…

utils.hpp

[Feature][CPU Backend]: Optimize ARM vectorization backend (#30329)

2026-02-02 20:17:56 -08:00