On RADV, we already support fp16 with the LLVM backend (not saying it's
always optimized though) and with ACO it should be mostly working but
not yet enabled because I think we would like fast packed math support
first and I'm not sure if fp16 I/O are implemented.
There is also some missing fp16 optimizations in the ACO backend which
makes some games (eg. Youngblood) a bit slower if fp16
features/extensions are exposed but the gap should be filled soon hopefully.
I think the most important thing missing is a SLP vectorizer in NIR for us.
On 5/4/20 8:43 PM, Marek Olšák wrote:
Hi,
This is the status of mediump support in Mesa. What I listed is what
AMD GPUs can do. "Yes" means what Mesa supports.
*Feature* *FP16 support* *Int16 support*
ALU Yes No
Uniforms No No
VS in No No
VS out / FS in No No
FS out No No
TCS, TES, GS out / in No No
Sampler coordinates (only coord, derivs, lod, bias; not offset and
compare) No ---
Image coordinates --- No
Return value from samplers (incl. sampler buffers) Yes
No
Return value from image loads (incl. image buffers) No No
Data source for image stores (incl. image buffers) No No
If 16-bit sampler/image instructions are surrounded by conversions,
promote them to 32 bits No No
Please let me know if you don't see the table correctly.
I'd like to know if I can enable some of them using the existing FP16
CAP. The only drivers supporting FP16 are currently Freedreno and
Panfrost.
Thanks,
Marek
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev