summaryrefslogtreecommitdiffstats
path: root/src/gallium/drivers/nouveau/nouveau_heap.c
diff options
context:
space:
mode:
authorIlia Mirkin <[email protected]>2014-01-10 17:58:37 -0500
committerMaarten Lankhorst <[email protected]>2014-01-27 16:40:42 +0100
commit18d97a8df776863c89c52294055160a17fc0f9e6 (patch)
treef79284462588e22fd4e5eda65381be71c7958919 /src/gallium/drivers/nouveau/nouveau_heap.c
parent45b64e52f4df49ac01ac100fba2c01633d492a6d (diff)
nouveau/codegen: set dType to S32 for OP_NEG U32
It doesn't make sense to do an OP_NEG from U32 to U32. This was manifested on nv50 in glsl-fs-atan-3 which was generating a UMAD TEMP[0].x, TEMP[0].xxxx, -TEMP[5].xxxx, TEMP[0].xxxx instruction. (For some reason, nvc0 causes a different shader to be generated.) This led to a cvt neg u32 $r1 u32 $r1 Which did not yield the desired result. This changes the final output to cvt neg s32 $r1 u32 $r1 which produces the desired output and the piglit tests passes. My assumption is that this is also what we want on nvc0, but could not test as there was no suitable shader that generated the problem instruction. Signed-off-by: Ilia Mirkin <[email protected]>
Diffstat (limited to 'src/gallium/drivers/nouveau/nouveau_heap.c')
0 files changed, 0 insertions, 0 deletions