mirror of
https://github.com/amd/blis.git
synced 2026-05-12 10:05:38 +00:00
c8f14edcf5
- Added support for 2 new APIs: 1. sgemm_compute() 2. dgemm_compute() These are dependent on the ?gemm_pack_get_size() and ?gemm_pack() APIs. - ?gemm_compute() takes the packed matrix buffer (represented by the packed matrix identifier) and performs the GEMM operation: C := A * B + beta * C. - Whenever the kernel storage preference and the matrix storage scheme isn't matching, and the respective matrix being loaded isn't packed either, on-the-go packing has been enabled for such cases to pack that matrix. - Note: If both the matrices are packed using the ?gemm_pack() API, it is the responsibility of the user to pack only one matrix with alpha scalar and the other with a unit scalar. - Note: Support is presently limited to Single Thread only. Both, pack and compute APIs are forced to take n_threads=1. AMD-Internal: [CPUPL-3560] Change-Id: I825d98a0a5038d31668d2a4b84b3ccc204e6c158
92 lines
3.8 KiB
Plaintext
92 lines
3.8 KiB
Plaintext
sgemm_ S N N P U 1 1 1 1 0 1 1 1 0 1
|
|
sgemm_ S N N P U 2 2 2 1 0 2 2 1 0 2
|
|
sgemm_ S N N P U 3 3 3 1 0 3 3 1 0 3
|
|
sgemm_ S N N P U 4 4 4 1 0 4 4 1 0 4
|
|
sgemm_ S N N P U 5 5 5 1 0 5 5 1 0 5
|
|
sgemm_ S N N P U 6 6 6 1 0 6 6 1 0 6
|
|
sgemm_ S N N P U 7 7 7 1 0 7 7 1 0 7
|
|
sgemm_ S N N P U 8 8 8 1 0 8 8 1 0 8
|
|
sgemm_ S N N P U 9 9 9 1 0 9 9 1 0 9
|
|
sgemm_ S N N P U 10 10 10 1 0 10 10 1 0 10
|
|
sgemm_ S N N P U 20 20 20 1 0 20 20 1 0 20
|
|
sgemm_ S N N P U 30 30 30 1 0 30 30 1 0 30
|
|
sgemm_ S N N P U 40 40 40 1 0 40 40 1 0 40
|
|
sgemm_ S N N P U 50 50 50 1 0 50 50 1 0 50
|
|
sgemm_ S N N P U 60 60 60 1 0 60 60 1 0 60
|
|
sgemm_ S N N P U 70 70 70 1 0 70 70 1 0 70
|
|
sgemm_ S N N P U 80 80 80 1 0 80 80 1 0 80
|
|
sgemm_ S N N P U 90 90 90 1 0 90 90 1 0 90
|
|
sgemm_ S N N P U 100 100 100 1 0 100 100 1 0 100
|
|
sgemm_ S N N P U 200 200 200 1 0 200 200 1 0 200
|
|
sgemm_ S N N P U 300 300 300 1 0 300 300 1 0 300
|
|
sgemm_ S N N P U 400 400 400 1 0 400 400 1 0 400
|
|
sgemm_ S N N P U 500 500 500 1 0 500 500 1 0 500
|
|
dgemm_ D N N P U 1 1 1 1 0 1 1 1 0 1
|
|
dgemm_ D N N P U 2 2 2 1 0 2 2 1 0 2
|
|
dgemm_ D N N P U 3 3 3 1 0 3 3 1 0 3
|
|
dgemm_ D N N P U 4 4 4 1 0 4 4 1 0 4
|
|
dgemm_ D N N P U 5 5 5 1 0 5 5 1 0 5
|
|
dgemm_ D N N P U 6 6 6 1 0 6 6 1 0 6
|
|
dgemm_ D N N P U 7 7 7 1 0 7 7 1 0 7
|
|
dgemm_ D N N P U 8 8 8 1 0 8 8 1 0 8
|
|
dgemm_ D N N P U 9 9 9 1 0 9 9 1 0 9
|
|
dgemm_ D N N P U 10 10 10 1 0 10 10 1 0 10
|
|
dgemm_ D N N P U 20 20 20 1 0 20 20 1 0 20
|
|
dgemm_ D N N P U 30 30 30 1 0 30 30 1 0 30
|
|
dgemm_ D N N P U 40 40 40 1 0 40 40 1 0 40
|
|
dgemm_ D N N P U 50 50 50 1 0 50 50 1 0 50
|
|
dgemm_ D N N P U 60 60 60 1 0 60 60 1 0 60
|
|
dgemm_ D N N P U 70 70 70 1 0 70 70 1 0 70
|
|
dgemm_ D N N P U 80 80 80 1 0 80 80 1 0 80
|
|
dgemm_ D N N P U 90 90 90 1 0 90 90 1 0 90
|
|
dgemm_ D N N P U 100 100 100 1 0 100 100 1 0 100
|
|
dgemm_ D N N P U 200 200 200 1 0 200 200 1 0 200
|
|
dgemm_ D N N P U 300 300 300 1 0 300 300 1 0 300
|
|
dgemm_ D N N P U 400 400 400 1 0 400 400 1 0 400
|
|
dgemm_ D N N P U 500 500 500 1 0 500 500 1 0 500
|
|
sgemm_ S N N U P 1 1 1 1 0 1 1 1 0 1
|
|
sgemm_ S N N U P 2 2 2 1 0 2 2 1 0 2
|
|
sgemm_ S N N U P 3 3 3 1 0 3 3 1 0 3
|
|
sgemm_ S N N U P 4 4 4 1 0 4 4 1 0 4
|
|
sgemm_ S N N U P 5 5 5 1 0 5 5 1 0 5
|
|
sgemm_ S N N U P 6 6 6 1 0 6 6 1 0 6
|
|
sgemm_ S N N U P 7 7 7 1 0 7 7 1 0 7
|
|
sgemm_ S N N U P 8 8 8 1 0 8 8 1 0 8
|
|
sgemm_ S N N U P 9 9 9 1 0 9 9 1 0 9
|
|
sgemm_ S N N U P 10 10 10 1 0 10 10 1 0 10
|
|
sgemm_ S N N U P 20 20 20 1 0 20 20 1 0 20
|
|
sgemm_ S N N U P 30 30 30 1 0 30 30 1 0 30
|
|
sgemm_ S N N U P 40 40 40 1 0 40 40 1 0 40
|
|
sgemm_ S N N U P 50 50 50 1 0 50 50 1 0 50
|
|
sgemm_ S N N U P 60 60 60 1 0 60 60 1 0 60
|
|
sgemm_ S N N U P 70 70 70 1 0 70 70 1 0 70
|
|
sgemm_ S N N U P 80 80 80 1 0 80 80 1 0 80
|
|
sgemm_ S N N U P 90 90 90 1 0 90 90 1 0 90
|
|
sgemm_ S N N U P 100 100 100 1 0 100 100 1 0 100
|
|
sgemm_ S N N U P 200 200 200 1 0 200 200 1 0 200
|
|
sgemm_ S N N U P 300 300 300 1 0 300 300 1 0 300
|
|
sgemm_ S N N U P 400 400 400 1 0 400 400 1 0 400
|
|
sgemm_ S N N U P 500 500 500 1 0 500 500 1 0 500
|
|
dgemm_ D N N U P 1 1 1 1 0 1 1 1 0 1
|
|
dgemm_ D N N U P 2 2 2 1 0 2 2 1 0 2
|
|
dgemm_ D N N U P 3 3 3 1 0 3 3 1 0 3
|
|
dgemm_ D N N U P 4 4 4 1 0 4 4 1 0 4
|
|
dgemm_ D N N U P 5 5 5 1 0 5 5 1 0 5
|
|
dgemm_ D N N U P 6 6 6 1 0 6 6 1 0 6
|
|
dgemm_ D N N U P 7 7 7 1 0 7 7 1 0 7
|
|
dgemm_ D N N U P 8 8 8 1 0 8 8 1 0 8
|
|
dgemm_ D N N U P 9 9 9 1 0 9 9 1 0 9
|
|
dgemm_ D N N U P 10 10 10 1 0 10 10 1 0 10
|
|
dgemm_ D N N U P 20 20 20 1 0 20 20 1 0 20
|
|
dgemm_ D N N U P 30 30 30 1 0 30 30 1 0 30
|
|
dgemm_ D N N U P 40 40 40 1 0 40 40 1 0 40
|
|
dgemm_ D N N U P 50 50 50 1 0 50 50 1 0 50
|
|
dgemm_ D N N U P 60 60 60 1 0 60 60 1 0 60
|
|
dgemm_ D N N U P 70 70 70 1 0 70 70 1 0 70
|
|
dgemm_ D N N U P 80 80 80 1 0 80 80 1 0 80
|
|
dgemm_ D N N U P 90 90 90 1 0 90 90 1 0 90
|
|
dgemm_ D N N U P 100 100 100 1 0 100 100 1 0 100
|
|
dgemm_ D N N U P 200 200 200 1 0 200 200 1 0 200
|
|
dgemm_ D N N U P 300 300 300 1 0 300 300 1 0 300
|
|
dgemm_ D N N U P 400 400 400 1 0 400 400 1 0 400
|
|
dgemm_ D N N U P 500 500 500 1 0 500 500 1 0 500 |