> > @@ -136,17 +136,27 @@ static bool
> > cf_node_has_side_effects(nir_cf_node *node)
> > {
> > nir_foreach_block_in_cf_node(block, node) {
> > + bool inside_loop = node->type == nir_cf_node_loop;
> > + for (nir_cf_node *n = &block->cf_node; !inside_loop && n != node; n
> > = n->parent) {
> >
>
> Is there some reason why you added !inside_loop to the condition? Just to
> avoid looping if we detect it early?
Yes.
> The only part of this comment change that looks relevant is the first bit
> where you make it also talk about ifs. That said, I think your updates are
> reasonable, they just probably belong in their own patch.
Will split the comment changes to its own patch (and fix the spurious
whitespace).
Thanks,
Caio
_______________________________________________
mesa-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
