diff options
| author | Faith Ekstrand <faith.ekstrand@collabora.com> | 2023-10-26 08:01:15 -0500 |
|---|---|---|
| committer | Eric Engestrom <eric@engestrom.ch> | 2023-10-30 15:47:22 +0000 |
| commit | 8a7498e13f081c054ebe7f14994ccca89677557e (patch) | |
| tree | 68173eb78e20c69ff2df8d83513a0a6a237b016d | |
| parent | 0144e4d0b3f069d883e4771847f1bb990a2ea675 (diff) | |
nir/lower_bit_size: Fix subgroup lowering for floats
Using u2u is always correct for integers, including signed integers,
because we're doing a down-cast. It's wrong for floats, though.
Fixes: f95665cfebbd ("nir/lower_bit_size: Add support for lowering subgroup ops")
Reviewed-by: Daniel Schürmann <daniel@schuermann.dev>
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25894>
(cherry picked from commit 5979e74177eb3013ca7e78867295e9a015e6c8ed)
| -rw-r--r-- | .pick_status.json | 2 | ||||
| -rw-r--r-- | src/compiler/nir/nir_lower_bit_size.c | 2 |
2 files changed, 2 insertions, 2 deletions
diff --git a/.pick_status.json b/.pick_status.json index d2d18bc7421..8247aa12bca 100644 --- a/.pick_status.json +++ b/.pick_status.json @@ -984,7 +984,7 @@ "description": "nir/lower_bit_size: Fix subgroup lowering for floats", "nominated": true, "nomination_type": 1, - "resolution": 0, + "resolution": 1, "main_sha": null, "because_sha": "f95665cfebbd943ca3795c026e3dabf4af3f262c", "notes": null diff --git a/src/compiler/nir/nir_lower_bit_size.c b/src/compiler/nir/nir_lower_bit_size.c index 37889a39faf..35097b85d83 100644 --- a/src/compiler/nir/nir_lower_bit_size.c +++ b/src/compiler/nir/nir_lower_bit_size.c @@ -203,7 +203,7 @@ lower_intrinsic_instr(nir_builder *b, nir_intrinsic_instr *intrin, if (intrin->intrinsic != nir_intrinsic_vote_feq && intrin->intrinsic != nir_intrinsic_vote_ieq) - res = nir_u2uN(b, res, old_bit_size); + res = nir_convert_to_bit_size(b, res, type, old_bit_size); nir_def_rewrite_uses(&intrin->def, res); break; |