From: Jason Ekstrand <jason.ekstrand@intel.com>
Date: Fri, 12 May 2017 18:55:51 +0000 (-0700)
Subject: intel/blorp: Don't use ffma directly
X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=302c0488cf37c145a538b7560c2aeea1b75e6890;p=mesa.git

intel/blorp: Don't use ffma directly

It isn't supported prior to gen6 and, on gen6+, NIR will fuse the fmul
and fadd into an ffma automatically for us anyway.

Reviewed-by: Topi Pohjolainen <topi.pohjolainen@intel.com>
---

diff --git a/src/intel/blorp/blorp_blit.c b/src/intel/blorp/blorp_blit.c
index 23e33defa7b..75fc3f5080b 100644
--- a/src/intel/blorp/blorp_blit.c
+++ b/src/intel/blorp/blorp_blit.c
@@ -135,7 +135,7 @@ blorp_blit_apply_transform(nir_builder *b, nir_ssa_def *src_pos,
    nir_ssa_def *mul = nir_vec2(b, nir_channel(b, coord_transform, 0),
                                   nir_channel(b, coord_transform, 2));
 
-   return nir_ffma(b, src_pos, mul, offset);
+   return nir_fadd(b, nir_fmul(b, src_pos, mul), offset);
 }
 
 static inline void