Arnav Sharma f0416cff08 SGEMM SUP Panel Stride Bug Fix ...

- The AVX512 SGEMM SUP rv m and n kernels did not accomodate for the
  use of panel strides in case of packed matrices, thus resulting in
  incorrect matrix strides when packing was explicitly enabled using
  BLIS_PACK_A=1, BLIS_PACK_B=1 or both.
- The kernels are updated to use panel strides for traversing both A
  and B matrix buffers accurately.

[AMD-Internal]: CPUPL-3673
Change-Id: I4341ed7e1e1419cc3e2063b06f278edcb9145adb

2023-09-27 03:02:24 -04:00

..

1

Fixes for GEMV Functionality Issues

2023-08-14 13:54:48 +05:30

1m

Windows CMake bugfix in object libraries for shared library option

2023-05-24 17:30:16 +05:30

3

SGEMM SUP Panel Stride Bug Fix

2023-09-27 03:02:24 -04:00

aocl_smart

DGEMM: Added decision logic to choose between sup vs native for zen4 architecture

2023-04-17 13:08:34 -04:00

lpgemm

Fix for lack of BF16 instruction when compiled with GCC-11

2023-05-19 10:15:08 +00:00

bli_kernels_zen4.h

Added AVX512 ZDSCALV kernel

2023-08-06 01:51:47 -04:00

CMakeLists.txt

Update CMake system to use object libraries for haswell, skx and zen4.

2023-05-12 10:04:16 -04:00