Meghana Vankadari fb6d4b5b8b BLAS compliance for Level-3 routines ...

Details:
- Added checks in top-level BLAS layer for alpha = zero. Scale C with
  beta or set C with 0 appropriately if true and return. This ensures
  that input matrices are not referenced, and thus that any Inf or NaN
  values in them are not propagated to output matrix.
- Scalm internally handles the case where both alpha and beta are zero
  by calling setm to populate the output matrix with zeroes.

AMD-Internal: [CPUPL-3134]
Change-Id: I14f14419049e7fcc55bf7fd30ddbd3d5a12c4427

2023-03-29 04:36:00 -05:00

..

0

AOCL Windows: 3.1 BLIS changes

2021-03-08 19:04:17 +05:30

1

AOCL Windows: 3.1 BLIS changes

2021-03-08 19:04:17 +05:30

1d

AOCL Windows: 3.1 BLIS changes

2021-03-08 19:04:17 +05:30

1f

AOCL Windows: 3.1 BLIS changes

2021-03-08 19:04:17 +05:30

1m

Redirecting scalm to setm when alpha is zero

2023-03-24 00:59:43 -04:00

2

Added vector packing logic to ZGEMV variant 2

2023-03-22 03:19:18 -04:00

3

Added in row storage support for C matrix.

2023-03-24 02:46:21 -04:00

base

Fixed decision logic in choosing between sup vs native GEMM code paths

2023-03-02 02:12:42 -05:00

compat

BLAS compliance for Level-3 routines

2023-03-29 04:36:00 -05:00

include

Fixed Compilation Fails when configured with --disable-blas

2023-03-23 06:11:52 -04:00

ind

3m1 turning OFF

2021-05-26 10:06:54 +05:30

thread

Code cleanup and warnings fixes

2023-01-09 04:34:52 -05:00

util

Remove quick return for zero increments.

2023-03-23 23:35:03 -04:00

CMakeLists.txt

AOCL Windows: 3.1 BLIS changes

2021-03-08 19:04:17 +05:30