diff options
| author | Roland Scheidegger <sroland@vmware.com> | 2021-07-26 16:55:10 +0200 |
|---|---|---|
| committer | Dylan Baker <dylan.c.baker@intel.com> | 2021-07-27 11:43:31 -0700 |
| commit | fdc5858caf7bb5efd765463f96a4447bb561767c (patch) | |
| tree | b86fd5b67aa4dc376afdb382ce260f725c676454 /.pick_status.json | |
| parent | d23a2a0128a2a457453e1f7349130c23ee6ae025 (diff) | |
llvmpipe: fix nir dot products (fsum op)
When the dot product uses a source which can be optimized to a scalar,
after a bunch of nir optimization steps the source to fsum will be a scalar
with a x replicate swizzle. Hence nir_src_num_components is just 1 and the
fsum was just a no-op which is not correct. Arguably this could be optimized
a bit better, but just determine the number of addends by using nir_op_infos
instead (the operand fetch was fixed already by 39a938ecf41b doing the same).
Fixes: 4eb0475b5a00 ("gallivm/nir: add fsum support")
Reviewed-by: Erik Faye-Lund <erik.faye-lund@collabora.com>
Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Dave Airlie <airlied@redhat.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/12066>
(cherry picked from commit cac5711d433fa1519be101ad3dae1e1d53c9a55a)
Diffstat (limited to '.pick_status.json')
| -rw-r--r-- | .pick_status.json | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/.pick_status.json b/.pick_status.json index b6a4f8bc98b..e83f24d486e 100644 --- a/.pick_status.json +++ b/.pick_status.json @@ -301,7 +301,7 @@ "description": "llvmpipe: fix nir dot products (fsum op)", "nominated": true, "nomination_type": 1, - "resolution": 0, + "resolution": 1, "main_sha": null, "because_sha": "4eb0475b5a0047e368129550e4c07a450a233a74" }, |