On Tuesday, September 22, 2015 06:18:12 PM Jason Ekstrand wrote: > We don't use any of the code after the switch and, as far as I can tell, it > wouldn't work anyway. The only reason this wasn't causing us problems is > that it's all dead and DCE cleans it up.
Huh, really? Right after that switch statement I see:
if (instr->dest.dest.ssa.num_components == 1)
return;
and reductions should always produce a single component, so this change
seems like it should do nothing. It seems like a reasonable change
regardless...but...
> ---
> src/glsl/nir/nir_lower_alu_to_scalar.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/glsl/nir/nir_lower_alu_to_scalar.c
> b/src/glsl/nir/nir_lower_alu_to_scalar.c
> index 5ef5ec2..84d4943 100644
> --- a/src/glsl/nir/nir_lower_alu_to_scalar.c
> +++ b/src/glsl/nir/nir_lower_alu_to_scalar.c
> @@ -86,7 +86,7 @@ lower_alu_instr_scalar(nir_alu_instr *instr, nir_builder *b)
> case name##3: \
> case name##4: \
> lower_reduction(instr, chan, merge, b); \
> - break;
> + return;
>
> switch (instr->op) {
> case nir_op_vec4:
>
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ mesa-dev mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/mesa-dev
