<div dir="ltr"><span style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:small;font-style:normal;font-variant-ligatures:normal;font-variant-caps:normal;font-weight:400;letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;background-color:rgb(255,255,255);text-decoration-style:initial;text-decoration-color:initial;float:none;display:inline">Reviewed-by: Bas Nieuwenhuizen <<a href="mailto:bas@basnieuwenhuizen.nl">bas@basnieuwenhuizen.nl</a>></span><br></div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Feb 2, 2018 at 7:04 PM, Samuel Pitoiset <span dir="ltr"><<a href="mailto:samuel.pitoiset@gmail.com" target="_blank">samuel.pitoiset@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">There is no fpow in hardware, so it's always lowered somewhere,<br>
but it appears that lowering at NIR level is better. Figured while<br>
comparing compute shaders between RadeonSI and RADV.<br>
<br>
Polaris10:<br>
Totals from affected shaders:<br>
SGPRS: 18936 -> 18904 (-0.17 %)<br>
VGPRS: 12240 -> 12220 (-0.16 %)<br>
Spilled SGPRs: 2809 -> 2809 (0.00 %)<br>
Code Size: 718116 -> 719848 (0.24 %) bytes<br>
Max Waves: 1409 -> 1410 (0.07 %)<br>
<br>
Vega10:<br>
Totals from affected shaders:<br>
SGPRS: 18392 -> 18392 (0.00 %)<br>
VGPRS: 12008 -> 11920 (-0.73 %)<br>
Spilled SGPRs: 3001 -> 2981 (-0.67 %)<br>
Code Size: 777444 -> 778788 (0.17 %) bytes<br>
Max Waves: 1503 -> 1504 (0.07 %)<br>
<br>
Signed-off-by: Samuel Pitoiset <<a href="mailto:samuel.pitoiset@gmail.com">samuel.pitoiset@gmail.com</a>><br>
---<br>
src/amd/vulkan/radv_shader.c | 1 +<br>
1 file changed, 1 insertion(+)<br>
<br>
diff --git a/src/amd/vulkan/radv_shader.c b/src/amd/vulkan/radv_shader.c<br>
index af094e6220..769e991f93 100644<br>
--- a/src/amd/vulkan/radv_shader.c<br>
+++ b/src/amd/vulkan/radv_shader.c<br>
@@ -67,6 +67,7 @@ static const struct nir_shader_compiler_options nir_options = {<br>
.lower_extract_byte = true,<br>
.lower_extract_word = true,<br>
.lower_ffma = true,<br>
+ .lower_fpow = true,<br>
.vs_inputs_dual_locations = true,<br>
.max_unroll_iterations = 32<br>
};<br>
<span class="HOEnZb"><font color="#888888">--<br>
2.16.1<br>
<br>
______________________________<wbr>_________________<br>
mesa-dev mailing list<br>
<a href="mailto:mesa-dev@lists.freedesktop.org">mesa-dev@lists.freedesktop.org</a><br>
<a href="https://lists.freedesktop.org/mailman/listinfo/mesa-dev" rel="noreferrer" target="_blank">https://lists.freedesktop.org/<wbr>mailman/listinfo/mesa-dev</a><br>
</font></span></blockquote></div><br></div>