Add packing support M edge cases in ZGEMM 12xk pack kernel (#89)

Browse Source

Previously, the ZGEMM implementation used `zscalv` 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 `zscalv` for edge cases, addressing the
      performance bottleneck.

    AMD-Internal: [CPUPL-6677]

Co-authored-by: harsh dave <harsdave@amd.com>

...

This commit is contained in:

Dave, Harsh

2025-08-14 14:29:03 +05:30

committed by GitHub

parent 76c4872718

commit 1b1b19486b

1 changed files with 2241 additions and 167 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download Patch File Download Diff File Expand all files Collapse all files

2408

kernels/zen4/1m/bli_packm_zen4_asm_z12xk.c

File diff suppressed because it is too large Load Diff