[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:
co-authored by
Li, Jiang <jiang1.li@intel.com>
parent
0de5333989
commit
34ce0ffd1f
@@ -12,7 +12,7 @@
|
||||
#include "cpu/utils.hpp"
|
||||
|
||||
namespace cpu_attention {
|
||||
enum class ISA { AMX, VEC, VEC16, NEON };
|
||||
enum class ISA { AMX, VEC, VEC16, NEON, VXE };
|
||||
|
||||
template <ISA isa, typename scalar_t, int64_t head_dim>
|
||||
class AttentionImpl {};
|
||||
|
||||
Reference in New Issue
Block a user