[Mesa-dev] [RFC PATCH] Add GL_MESA_ieee_fp_alu_mode specification draft

Ilia Mirkin imirkin at alum.mit.edu
Tue Feb 25 00:54:33 UTC 2020

On Mon, Feb 24, 2020 at 6:56 PM Ilia Mirkin <imirkin at alum.mit.edu> wrote:
> On Sun, Feb 23, 2020 at 8:57 PM Ilia Mirkin <imirkin at alum.mit.edu> wrote:
> >
> > ---
> >
> > We talked about something like this a while back, but the end result
> > was inconclusive. I added a TGSI MUL_ZERO_WINS shader property for nine.
> > But it'd be nice for wine to be able to control this too.
> >
> > I couldn't actually find any evidence of the discussion from 2017 or so,
> > so ... let's have another one.
> Ian, Matteo, thanks for your feedback.
> Based on IRC discussion today, it looks like radeonsi is not
> interested in gaining such a feature, and Ian's comments all point
> away from exposing the actual hardware bits, which I think means this
> ext is DOA. I don't think support in just nouveau + r600 justifies the
> effort of getting this going -- too small of a user base between them,
> and they can use nine if they really want better perf. I guess that's
> why this ext died back in 2017 too, but at least now there will
> hopefully be an easier-to-find record of it.
> Cheers,
>   -ilia

Oh, and for posterity, strfllw in #winehackers tracked down the
original thread from 2017:


Not sure how I missed it, I even looked in Jan 2017, but there we are.
Apparently part of the difficulty for radeonsi is LLVM, but perhaps
ACO could soften some of that blow.


More information about the mesa-dev mailing list