[Mesa-dev] [PATCH 11/22 v2] nir: Recognize some more open-coded fmin / fmax
Jason Ekstrand
jason at jlekstrand.net
Tue Mar 6 02:41:38 UTC 2018
On Mon, Mar 5, 2018 at 2:25 PM, Ian Romanick <idr at freedesktop.org> wrote:
> On 03/05/2018 02:07 PM, Jason Ekstrand wrote:
> > On Mon, Mar 5, 2018 at 11:11 AM, Ian Romanick <idr at freedesktop.org
> > <mailto:idr at freedesktop.org>> wrote:
> >
> > On 02/28/2018 12:48 PM, Jason Ekstrand wrote:
> > > Rb
> >
> > For both or just this one?
> >
> >
> > If you put the ~ on there, both.
>
> I'm not sure what you mean. That's all the other patch ([PATCH 11.1/22]
> nir: Mark bcsel-to-fmin (or fmax) transformations as inexact) does.
>
Sorry, I got confused. Yes, RB to both.
> > > On Wed, Feb 28, 2018 at 12:18 PM, Ian Romanick <
> idr at freedesktop.org <mailto:idr at freedesktop.org>
> > > <mailto:idr at freedesktop.org <mailto:idr at freedesktop.org>>> wrote:
> > >
> > > From: Ian Romanick <ian.d.romanick at intel.com <mailto:
> ian.d.romanick at intel.com>
> > > <mailto:ian.d.romanick at intel.com
> > <mailto:ian.d.romanick at intel.com>>>
> > >
> > > This transformation is inexact because section 4.7.1 (Range and
> > > Precision) says:
> > >
> > > Operations and built-in functions that operate on a NaN
> > are not
> > > required to return a NaN as the result.
> > >
> > > The fmin or fmax might not return NaN in cases where the
> original
> > > expression would be required to return NaN.
> > >
> > > v2: Reorder operands and mark as inexact. The latter
> suggested by
> > > Jason.
> > >
> > > shader-db results:
> > >
> > > Haswell, Broadwell, and Skylake had similar results. (Skylake
> > shown)
> > > total instructions in shared programs: 14514817 -> 14514808
> > (<.01%)
> > > instructions in affected programs: 229 -> 220 (-3.93%)
> > > helped: 3
> > > HURT: 0
> > > helped stats (abs) min: 1 max: 4 x̄: 3.00 x̃: 4
> > > helped stats (rel) min: 2.86% max: 4.12% x̄: 3.70% x̃: 4.12%
> > >
> > > total cycles in shared programs: 533145211 -> 533144939 (<.01%)
> > > cycles in affected programs: 37268 -> 36996 (-0.73%)
> > > helped: 8
> > > HURT: 0
> > > helped stats (abs) min: 2 max: 134 x̄: 34.00 x̃: 2
> > > helped stats (rel) min: 0.02% max: 14.22% x̄: 3.53% x̃: 0.05%
> > >
> > > Sandy Bridge and Ivy Bridge had similar results. (Ivy Bridge
> > shown)
> > > total cycles in shared programs: 257618409 -> 257618403 (<.01%)
> > > cycles in affected programs: 12582 -> 12576 (-0.05%)
> > > helped: 3
> > > HURT: 0
> > > helped stats (abs) min: 2 max: 2 x̄: 2.00 x̃: 2
> > > helped stats (rel) min: 0.05% max: 0.05% x̄: 0.05% x̃: 0.05%
> > >
> > > No changes on Iron Lake or GM45.
> > >
> > > Signed-off-by: Ian Romanick <ian.d.romanick at intel.com
> > <mailto:ian.d.romanick at intel.com>
> > > <mailto:ian.d.romanick at intel.com
> > <mailto:ian.d.romanick at intel.com>>>
> > > ---
> > > src/compiler/nir/nir_opt_algebraic.py | 2 ++
> > > 1 file changed, 2 insertions(+)
> > >
> > > diff --git a/src/compiler/nir/nir_opt_algebraic.py
> > > b/src/compiler/nir/nir_opt_algebraic.py
> > > index d40d59b..17f4d9c 100644
> > > --- a/src/compiler/nir/nir_opt_algebraic.py
> > > +++ b/src/compiler/nir/nir_opt_algebraic.py
> > > @@ -170,6 +170,8 @@ optimizations = [
> > > (('fge', ('fneg', ('fabs', a)), 0.0), ('feq', a, 0.0)),
> > > (('bcsel', ('flt', b, a), b, a), ('fmin', a, b)),
> > > (('bcsel', ('flt', a, b), b, a), ('fmax', a, b)),
> > > + (('~bcsel', ('fge', a, b), b, a), ('fmin', a, b)),
> > > + (('~bcsel', ('fge', b, a), b, a), ('fmax', a, b)),
> > > (('bcsel', ('inot', a), b, c), ('bcsel', a, c, b)),
> > > (('bcsel', a, ('bcsel', a, b, c), d), ('bcsel', a, b, d)),
> > > (('bcsel', a, True, 'b at bool'), ('ior', a, b)),
> > > --
> > > 2.9.5
> > >
> > > _______________________________________________
> > > mesa-dev mailing list
> > > mesa-dev at lists.freedesktop.org
> > <mailto:mesa-dev at lists.freedesktop.org>
> > <mailto:mesa-dev at lists.freedesktop.org
> > <mailto:mesa-dev at lists.freedesktop.org>>
> > > https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> > <https://lists.freedesktop.org/mailman/listinfo/mesa-dev>
> > > <https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> > <https://lists.freedesktop.org/mailman/listinfo/mesa-dev>>
> > >
> > >
> >
> >
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20180305/20ec238a/attachment-0001.html>
More information about the mesa-dev
mailing list