Sireesha Sanga cc3069fb5e Performance Improvement for ztrsm small sizes ...

Details:
- Optimization of ztrsm for Non-unit Diag Variants.
- Handled Overflow and Underflow Vulnerabilites in
  ztrsm small implementations.
- Fixed failures observed in libflame testing.
- Fine-tuned ztrsm small implementations for specific
  sizes 64<= m,n <= 256, by keeping the number of
  threads to the optimum value, under AOCL_DYNAMIC flag.
- For small sizes, ztrsm small implementation is
  used for all variants.

AMD-Internal: [SWLCSG-1194]
Change-Id: I066491bb03e5cda390cb699182af4350ae60be2d

2022-05-17 18:10:39 +05:30

..

sup

BLIS : Compiler warning fixes

2021-11-12 08:58:52 +05:30

bli_dgemm_ref_k1.c

DGEMM kernel implementation for case k = 1.

2021-11-12 08:58:53 +05:30

bli_gemm_small.c

Changes to enable zgemm small from BLAS Layer

2022-05-17 18:10:38 +05:30

bli_gemm_sqp_kernels.c

Induced method turned off, fix for beta=0 & C = NAN

2021-11-12 08:58:50 +05:30

bli_gemm_sqp_kernels.h

3m_sqp conjugate support added

2021-07-05 18:44:55 +05:30

bli_gemm_sqp.c

3m_sqp conjugate support added

2021-07-05 18:44:55 +05:30

bli_trsm_small.c

Performance Improvement for ztrsm small sizes

2022-05-17 18:10:39 +05:30

CMakeLists.txt

Removed syrk_small code

2021-11-12 08:58:49 +05:30