On Mon, Oct 15, 2018, at 5:18 PM, Eric Anholt wrote: > This will be used on V3D to cut down the size of the VS inputs in the VPM > (memory area for sharing data between shader stages). > --- > src/compiler/nir/nir_lower_io_to_scalar.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/src/compiler/nir/nir_lower_io_to_scalar.c b/src/compiler/ > nir/nir_lower_io_to_scalar.c > index f0c2a6a95d6d..c64f641a0ae4 100644 > --- a/src/compiler/nir/nir_lower_io_to_scalar.c > +++ b/src/compiler/nir/nir_lower_io_to_scalar.c > @@ -320,7 +320,9 @@ nir_lower_io_to_scalar_early(nir_shader *shader, > nir_variable_mode mask) > if (glsl_type_is_64bit(glsl_without_array(var->type))) > continue; > > - if (var->data.location < VARYING_SLOT_VAR0 && > + if (!(shader->info.stage == MESA_SHADER_VERTEX && > + mode == nir_var_shader_in) && > + var->data.location < VARYING_SLOT_VAR0 &&
As I understand it all current use of this pass skips vetex shader inputs via the mask param so this change shouldn't impact any other drivers so: Reviewed-by: Timothy Arceri <tarc...@itsqueeze.com> > var->data.location >= 0) > continue; > > -- > 2.18.0 > > _______________________________________________ > 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
