[PATCH] D22898: AMDGPU: Fix ffloor for SI
Matt Arsenault via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Aug 30 12:03:14 PDT 2017
arsenm added inline comments.
================
Comment at: lib/Target/AMDGPU/SIInstructions.td:3545
$x,
(V_CMP_CLASS_F64_e64 SRCMODS.NONE, $x, 3/*NaN*/)),
DSTCLAMP.NONE, DSTOMOD.NONE)
----------------
This might be the problem. This is using SRCMODS.NONE rather than preserving it like the other uses. It might be less error prone to do this as a custom expansion of floor rather than expanding the fract here
https://reviews.llvm.org/D22898
More information about the llvm-commits
mailing list