fix test fp8 code

This commit is contained in:
ltqin
2025-08-28 09:58:48 +00:00
parent 595d9b9718
commit 526a77134c
2 changed files with 2 additions and 2 deletions

View File

@@ -129,4 +129,4 @@ TBD
## FP8 experimental support
As described in [this blog](https://blog.hippoml.com/8bit-hippoattention-up-to-3x-faster-compared-to-flashattentionv2-8f9def90b482), we have an experimental support for fp8 fmha kernels, you can evaluate the performance by setting the arg `-prec=fp8` to the `tile_example_fmha_fwd`, on a gfx942 machine and ROCm 6.0+.
Currently we only support `-vlayout=c`( `hdim*seqlen` for V matrix) and `-squant=1`(static quantization) with `hdim=128` for fp8 now. Full feature support will come later.
Currently we support `-vlayout=r`( `hdim*seqlen` for V matrix) and `-squant=1`(static quantization) with `hdim=128` for fp8 now. Full feature support will come later.

View File

@@ -73,7 +73,7 @@ run_fp8_tests() {
for b in 1 2 ; do
for hdim in 64 128 256 ; do
$EXE -prec=fp8 -init=3 -b=$b -h=1 -d=128 -s=128 -bias=$bias -iperm=$perm -operm=$perm -vlayout=c -squant=1 -kname=$KNAME $COMMON_ARGS
$EXE -prec=fp8 -init=3 -b=$b -h=1 -d=128 -s=128 -bias=$bias -iperm=$perm -operm=$perm -vlayout=r -squant=1 -kname=$KNAME $COMMON_ARGS
done ; done ; done ; done
}