Meghana Vankadari 8c9a7c21b4 Optimized axpyf kernel for scomplex datatype ...

Details:
- Implemented axpyf kernel with fuse factor=4 for scomplex datatype.
- Modified BLAS interface call for cgemv to reduce framework overhead.
- Directed gemv to dotv in the case where dimension of y vector is 1.
- when alpha = 0, gemv becomes scalv of Y with beta. Added code to
return early after scaling Y vector with beta.

AMD-Internal: [CPUPL-1402]
Change-Id: Ibaab078008d76953332ba4da3515993578c0e586

2021-05-24 14:40:17 +05:30

..

level0

Declare/define static functions via BLIS_INLINE.

2020-08-03 11:23:40 +05:30

bli_arch_config_pre.h

Removed export macros from all internal prototypes.

2020-08-03 11:47:18 +05:30

bli_arch_config.h

BLIS:merge:

2021-04-27 11:09:48 +05:30

bli_blas_macro_defs.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_builtin_macro_defs.h

Rewrote reference kernels to use #pragma omp simd.

2019-01-24 17:23:18 -06:00

bli_complex_macro_defs.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_config_macro_defs.h

Auto-reduce (by default) prime numbers of threads.

2020-11-23 16:54:51 -06:00

bli_error_macro_defs.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_extern_defs.h

Export functions without def file (#303)

2019-03-11 19:05:32 -05:00

bli_f2c.h

Allow lesser Makefiles to reference installed BLIS.

2018-08-25 20:12:36 -05:00

bli_genarray_macro_defs.h

Added sup kernels and code path for gemmt similar to GEMM.GEMMT now also supports complex data types.

2020-07-13 16:26:32 +05:30

bli_gentdef_macro_defs.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_gentfunc_macro_defs.h

Optimized axpyf kernel for scomplex datatype

2021-05-24 14:40:17 +05:30

bli_gentprot_macro_defs.h

Blis: DOTC Additional argument for Complex types when using FLANG

2020-12-16 14:03:10 +05:30

bli_kernel_macro_defs.h

BLIS:merge:

2021-04-27 11:09:48 +05:30

bli_macro_defs.h

AOCL Windows: 3.1 BLIS changes

2021-03-09 17:35:57 +05:30

bli_misc_macro_defs.h

Replaced use of bool_t type with C99 bool.

2020-08-03 11:27:13 +05:30

bli_oapi_ba.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_oapi_ex.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_oapi_macro_defs.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_obj_macro_defs.h

Replaced use of bool_t type with C99 bool.

2020-08-03 11:27:13 +05:30

bli_param_macro_defs.h

BLIS:merge:

2021-04-27 11:09:48 +05:30

bli_pragma_macro_defs.h

Generalized ref kernels' pragma omp simd usage.

2019-02-12 16:01:28 -06:00

bli_sbox.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_scalar_macro_defs.h

Added support for pre-broadcast when packing B.

2019-09-17 17:42:10 -05:00

bli_system.h

BLIS:merge:

2021-04-27 11:09:48 +05:30

bli_tapi_ba.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_tapi_ex.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_tapi_macro_defs.h

Remove UT-Austin from copyright headers' clause 3.

2018-12-04 14:31:06 -06:00

bli_type_defs.h

BLIS:merge:

2021-04-27 11:09:48 +05:30

bli_x86_asm_macros.h

Optionally disable trsm diagonal pre-inversion.

2020-12-04 16:08:15 -06:00

blis.h

"Merge Selective Packing code from amd branch flame/blis"

2020-08-06 10:09:28 +05:30