zink: lower byte/word extract ops in nir

we don't implement these, and pre-optimizing them breaks things in ntv->vtn

Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5562>

diff --git a/src/gallium/drivers/zink/zink_compiler.c b/src/gallium/drivers/zink/zink_compiler.c
index 8ffdcc4607cdea9352d8d446c882d7d84cf63db4..a4e79d1ed26c795001a337e882ab359d4fad1b8f 100644 (file)
--- a/src/gallium/drivers/zink/zink_compiler.c
+++ b/src/gallium/drivers/zink/zink_compiler.c
@@ -90,6 +90,8 @@ static const struct nir_shader_compiler_options nir_options = {
    .lower_flrp32 = true,
    .lower_fpow = true,
    .lower_fsat = true,
+   .lower_extract_byte = true,
+   .lower_extract_word = true,
 };
 
 const void *