mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-14 10:09:41 +00:00
* fix compile error
* fix typo of padding
* Add smoothquant op
* Add smoothquant instance library
* refine type
* add test script
* Re-generate smoothquant.hpp
* Always use 'current year' in copyright
* use Generic2dBlockShape instead
* Add vector = 8 instance back
* Find exe path automatically
* Simplify the api condition
* Remove debugging code
* update year
* Add blank line between function declaration
* explicitly cast return value to dim3
* refine return value
* Fix default warmup and repeat value
* Add comment
* refactor sommthquant cmake
* Add README
* Fix typo
---------
Co-authored-by: Po Yen, Chen <PoYen.Chen@amd.com>
[ROCm/composable_kernel commit: fbd654545a]
37 lines
1.9 KiB
Bash
Executable File
37 lines
1.9 KiB
Bash
Executable File
|
|
EXE="$(find . -name tile_smoothquant -type f | head -n 1)"
|
|
|
|
$EXE -m=1 -n=1 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=80 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=128 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=144 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=168 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=184 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=256 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=288 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=344 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=376 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=448 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=512 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=924 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=1024 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=1078 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=1996 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
$EXE -m=700 -n=4080 -e=1e-12 -v=1 -prec=bf16 -repeat=1000
|
|
|
|
$EXE -m=700 -n=80 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=128 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=144 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=168 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=184 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=256 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=288 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=344 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=376 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=448 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=512 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=924 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=1024 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=1078 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=1996 -e=1e-12 -v=1 -prec=fp16 -repeat=1000
|
|
$EXE -m=700 -n=4080 -e=1e-12 -v=1 -prec=fp16 -repeat=1000 |