Meghana Vankadari d5b4d3aa5e Fixing control flow in aocl_gemm_bf16s4f32of32|bf16 ...

- Fixed framework of bf16s4f32of32 API to correct
  pointer updations.
- Modified pre_op structure to exclude pre-op-offset.
  Now offset is passed as a separate parameter to the
  scale-pack functions.
- Fixed work-distribution among threads in MT scenario.
- Added Blocksizes and kernel-pointers and verified
  functionality for the new API.

AMD-Internal: [SWLCSG-2943]
Change-Id: I58fece240d62c798c880a2b2b7fa64e560cc753d

2024-07-29 05:12:09 -04:00

..

config

Fixing control flow in aocl_gemm_bf16s4f32of32|bf16

2024-07-29 05:12:09 -04:00

frame

Fixing control flow in aocl_gemm_bf16s4f32of32|bf16

2024-07-29 05:12:09 -04:00

JIT

Fix for offset issue while reading constants from JIT code

2024-06-24 07:14:15 -04:00

kernels

Fixing control flow in aocl_gemm_bf16s4f32of32|bf16

2024-07-29 05:12:09 -04:00

aocl_bf16_type.h

Added low precision GEMM - bf16bf16f32of32

2022-08-24 03:27:00 -04:00

aocl_gemm_bf16_utils.c

Fixing control flow in aocl_gemm_bf16s4f32of32|bf16

2024-07-29 05:12:09 -04:00

aocl_gemm_bf16bf16f32obf16.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_bf16bf16f32of32.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_bf16s4f32of32.c

Fixing control flow in aocl_gemm_bf16s4f32of32|bf16

2024-07-29 05:12:09 -04:00

aocl_gemm_check.h

Added parameter checking for LPGEMM APIs

2023-11-09 21:50:55 -05:00

aocl_gemm_f32f32f32of32_utils.c

Implemented optimal AVX512-variant of f32 LPGEMV

2024-03-04 23:53:23 +05:30

aocl_gemm_f32f32f32of32.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_interface_apis.h

Added new API in aocl_gemm to support A bf16 data type and B s4 data type

2024-07-25 11:59:03 +00:00

aocl_gemm_post_ops.h

Added new API in aocl_gemm to support A bf16 data type and B s4 data type

2024-07-25 11:59:03 +00:00

aocl_gemm_s8s8s16os8.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_s8s8s16os16_utils.c

Implemented LPGEMV(n=1) for AVX2-INT8 variants

2024-06-18 12:09:18 +05:30

aocl_gemm_s8s8s16os16.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_s8s8s32os8.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_s8s8s32os32_utils.c

Added transB in all AVX512 based int8 API's

2024-05-23 03:46:13 +05:30

aocl_gemm_s8s8s32os32.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_u8s4s32os32_utils.c

Int4 B matrix reordering support in LPGEMM.

2024-06-24 07:55:34 -04:00

aocl_gemm_u8s8s16os8.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_u8s8s16os16_utils.c

Implemented LPGEMV(n=1) for AVX2-INT8 variants

2024-06-18 12:09:18 +05:30

aocl_gemm_u8s8s16os16.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_u8s8s16ou8.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_u8s8s32os8.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm_u8s8s32os32_utils.c

Added transB in all AVX512 based int8 API's

2024-05-23 03:46:13 +05:30

aocl_gemm_u8s8s32os32.c

Quantization (scale + zero point) support for BF16 LPGEMM api.

2024-07-18 05:32:51 -04:00

aocl_gemm.h

Implemented JIT-based microkernel for bf16 datatype

2024-03-13 05:55:18 +05:30

aocl_util_interface_apis.h

aocl_softmax_f32 api's for softmax computation as part of lpgemm.

2023-04-21 05:26:08 -04:00

aocl_util_l1_ops.c

aocl_softmax_f32 api's for softmax computation as part of lpgemm.

2023-04-21 05:26:08 -04:00