<div dir="auto"><div><div class="gmail_extra"><div class="gmail_quote">On Dec 21, 2016 9:17 PM, "Timothy Arceri" <<a href="mailto:timothy.arceri@collabora.com">timothy.arceri@collabora.com</a>> wrote:<br type="attribution"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="quoted-text">On Mon, 2016-12-19 at 20:11 -0800, Jason Ekstrand wrote:<br>
> ---<br>
> src/mesa/drivers/dri/i965/<wbr>brw_nir.c | 2 ++<br>
> 1 file changed, 2 insertions(+)<br>
><br>
> diff --git a/src/mesa/drivers/dri/i965/<wbr>brw_nir.c<br>
> b/src/mesa/drivers/dri/i965/<wbr>brw_nir.c<br>
> index 0c1fb44..a091861 100644<br>
> --- a/src/mesa/drivers/dri/i965/<wbr>brw_nir.c<br>
> +++ b/src/mesa/drivers/dri/i965/<wbr>brw_nir.c<br>
> @@ -429,6 +429,8 @@ nir_optimize(nir_shader *nir, const struct<br>
> brw_compiler *compiler,<br>
> OPT(nir_opt_algebraic);<br>
> OPT(nir_opt_constant_<wbr>folding);<br>
> OPT(nir_opt_dead_cf);<br>
> + OPT(nir_opt_trivial_<wbr>continues);<br>
> + OPT(nir_opt_if);<br>
<br>
</div>I'm seeing regressions in my series for Vulkan (the Vulkan CTS was only<br>
enabled for my branch today). I wonder if we should be applying your<br>
series before mine? Either way this or a similar patch is:<br></blockquote></div></div></div><div dir="auto"><br></div><div dir="auto">If you're seeing regressions, there's probably a real bug there. I assume you applied the else case fix?</div><div dir="auto"><br></div><div dir="auto"><div class="gmail_extra"><div class="gmail_quote"><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Reviewed-by: Timothy Arceri <<a href="mailto:timothy.arceri@collabora.com">timothy.arceri@collabora.com</a>><br>
<div class="elided-text"><br>
> if (nir->options->max_unroll_<wbr>iterations != 0) {<br>
> OPT(nir_opt_loop_<wbr>unroll, indirect_mask);<br>
> }<br>
</div></blockquote></div><br></div></div></div>