[libre-riscv-dev] porting AMDVLK to the Libre RISC-V 3D GPU: NLNet EUR 50, 000 Grant application

[Michael Pham]

On Wed, Sep 25, 2019 at 2:32 AM Luke Kenneth Casson Leighton
<lkcl at lkcl.net> wrote:
>
> On Wed, Sep 25, 2019 at 5:15 AM Michael Pham <pham.michael.98 at gmail.com> wrote:
> >
> > > > Is there a specific reason you are basing it on the AMDVLK vulkan
> > > > driver instead of Mesa's RADV vulkan driver for amdgpu instead?
>
> honestly: i'd trust what AMD did to be feature-complete over something
> created through reverse-engineering.
>

Actually, I just did some additional research. It looks like AMD
actually pays their employees to contribute to Mesa's RADV as well.
Also, it's not completely reverse-engineering because AMD's stuff is
much more open than Nvidia. RADV takes some code from AMDVLK but
AMDVLK can't take in RADV's optimizations because of corporate
policies and the fact that AMD has to keep their linux vulkan driver
in sync with their windows driver.
Several benchmarks have shown that RADV and AMDVLK are matched
performance wise, however, AMDVLK actually has less features than RADV
(an example is that RADV supports AMD's freesync technology but AMDVLK
doesn't even support it yet).
In general, RADV is more stable and more advanced than AMDVLK, the
general consensus is that RADV is the better overall driver at the
moment.

https://www.reddit.com/r/Amd/comments/au220w/linux_50_3_games_in_amdvlk_future_amd_vulkan/eh5696a/
https://www.reddit.com/r/Amd/comments/au220w/linux_50_3_games_in_amdvlk_future_amd_vulkan/eh5w30d/
https://www.reddit.com/r/Amd/comments/au220w/linux_50_3_games_in_amdvlk_future_amd_vulkan/eh766g2/

In the end though, since you are turning the AMD driver into a Libre
RISC-V version, it might not even matter. But the point of me bringing
this up is that instead of porting AMDVLK, RADV might be a better base
to start from. Since Jacob is a graphics expert though, he'll have
better comments than mine (once he gets back).

-- Michael