diff options
author | Jason Ekstrand <[email protected]> | 2019-05-06 10:23:26 -0500 |
---|---|---|
committer | Jason Ekstrand <[email protected]> | 2019-05-24 08:38:11 -0500 |
commit | ddd08e1888cfdf7a643b8fa1ce61f77bb66aa3eb (patch) | |
tree | d7a679f59482db5f44756bbb482f55e61861d057 /src/intel/compiler/brw_nir_attribute_workarounds.c | |
parent | 6c2ca2a5d33cdfc4202ebb0df54567b587ecc578 (diff) |
nir/builder: Remove the use_fmov parameter from nir_swizzle
This flag has caused more confusion than good in most cases. You can
validly use imov for floats or fmov for integers because, without source
modifiers, neither modify their input in any way. Using imov for floats
is more reliable so we go that direction.
Reviewed-by: Kristian H. Kristensen <[email protected]>
Acked-by: Alyssa Rosenzweig <[email protected]>
Diffstat (limited to 'src/intel/compiler/brw_nir_attribute_workarounds.c')
-rw-r--r-- | src/intel/compiler/brw_nir_attribute_workarounds.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/intel/compiler/brw_nir_attribute_workarounds.c b/src/intel/compiler/brw_nir_attribute_workarounds.c index 55a3b3e4546..ecd0f0752b0 100644 --- a/src/intel/compiler/brw_nir_attribute_workarounds.c +++ b/src/intel/compiler/brw_nir_attribute_workarounds.c @@ -80,7 +80,7 @@ apply_attr_wa_block(nir_block *block, struct attr_wa_state *state) /* Apply BGRA swizzle if required. */ if (wa_flags & BRW_ATTRIB_WA_BGRA) { - val = nir_swizzle(b, val, (unsigned[4]){2,1,0,3}, 4, true); + val = nir_swizzle(b, val, (unsigned[4]){2,1,0,3}, 4); } if (wa_flags & BRW_ATTRIB_WA_NORMALIZE) { |