harsdave cd83fc38b5 Add packing support M edge cases in DGEMM 24xk pack kernel ...

Previously, the DGEMM implementation used `dscalv` for cases
where the M dimension of matrix A is not in multiple of 24,
resulting in a ~40% performance drop.

This commit introduces a specialized edge cases in pack kernel
to optimize performance for these cases.

The new packing support significantly improves the performance.

- Removed reliance on `dscalv` for edge cases, addressing the
  performance bottleneck.

AMD-Internal: [CPUPL-6677]

Change-Id: I150d13eb536d84f8eb439d7f4a77a04a0d0e6d60

2025-05-06 09:22:49 +05:30

..

sup

Additional optimizations to ZGEMM SUP and Tiny codepaths(ZEN4 and ZEN5)

2025-04-23 00:56:58 -04:00

bli_cgemm_avx512_k1.c

AVX512 optimizations for CGEMM(rank-1 kernel)

2025-03-06 20:14:05 +05:30

bli_cgemm_zen4_asm_4x24.c

AVX512 optimizations for CGEMM(Native)

2025-02-28 03:18:24 -05:00

bli_cgemm_zen4_asm_24x4.c

AVX512 optimizations for CGEMM(Native)

2025-02-28 03:18:24 -05:00

bli_cgemm_zen4_asm_macros.h

AVX512 optimizations for CGEMM(Native)

2025-02-28 03:18:24 -05:00

bli_dgemm_avx512_asm_8x24.c

Fixed C prefetch in 8x24 DGEMM kernel

2025-02-27 12:17:29 -05:00

bli_dgemm_avx512_k1.c

Added k=1 avx512 dgemm kernel.

2023-11-07 01:10:09 -05:00

bli_dgemm_zen4_asm_32x6.c

Code cleanup: Copyright notices

2024-08-05 15:35:08 -04:00

bli_gemm_tiny_adapter_zen4.c

Add packing support M edge cases in DGEMM 24xk pack kernel

2025-05-06 09:22:49 +05:30

bli_gemm_tiny_avx512.c

Code cleanup: Copyright notices (2)

2025-02-07 05:41:44 -05:00

bli_gemm_tiny_avx512.h

Support for Tiny-GEMM interface(ZGEMM)

2025-01-24 12:59:26 -05:00

bli_gemmtrsm_l_zen4_8x24.c

Code cleanup: Copyright notices

2024-08-05 15:35:08 -04:00

bli_gemmtrsm_l_zen_16x14.c

Code cleanup: AMD copyright notice

2023-11-23 08:54:31 -05:00

bli_gemmtrsm_u_zen4_8x24.c

Code cleanup: Copyright notices

2024-08-05 15:35:08 -04:00

bli_gemmtrsm_u_zen_16x14.c

Code cleanup: AMD copyright notice

2023-11-23 08:54:31 -05:00

bli_trsm_small_AVX512.c

Fixed compiler warnings due to prefetch in AVX2 and AVX512 kernels

2024-09-12 11:37:06 +05:30

bli_zero_zmm.c

BLIS: Missing clobbers (batch 8)

2025-02-07 10:39:24 -05:00

bli_zgemm_avx512_k1.c

AVX512 optimizations for ZGEMM to handle k = 1 cases

2024-07-09 07:07:24 -04:00

bli_zgemm_zen4_asm_4x12.c

Code cleanup: Copyright notices

2024-08-05 15:35:08 -04:00

bli_zgemm_zen4_asm_12x4.c

BLIS: Missing clobbers (batch 1)

2023-05-23 11:51:18 -04:00

bli_zgemmtrsm_l_4x12.c

Code cleanup: Copyright notices

2024-08-05 15:35:08 -04:00

bli_zgemmtrsm_u_4x12.c

Code cleanup: Copyright notices

2024-08-05 15:35:08 -04:00

bli_ztrsm_small_AVX512.c

Code cleanup: spelling corrections

2024-08-05 16:18:51 -04:00