4 Commits

Author SHA1 Message Date
Changpeng Fang
24267a7e14
AMDGPU: Add f64 to f32 support for llvm.fptrunc.round (#107481) 2024-09-05 22:57:27 -07:00
Changpeng Fang
e44a67543c
AMDGPU: Add a few unsupported checks for llvm.fptrunc.round intrinsic (#107330)
A check here can be removed when we implement support for the
corresponding types/mode.
2024-09-05 15:05:15 -07:00
Changpeng Fang
26b0bef192
AMDGPU: Use pattern to select instruction for intrinsic llvm.fptrunc.round (#105761)
Use GCNPat instead of Custom Lowering to select instructions for
intrinsic llvm.fptrunc.round. "SupportedRoundMode : TImmLeaf" is used as
a predicate to select only when the rounding mode is supported.
"as_hw_round_mode : SDNodeXForm" is developed to translate the round
modes to the corresponding ones that hardware recognizes.
2024-08-29 11:43:58 -07:00
Changpeng Fang
53d95f3056
AMDGPU: Rename fail.llvm.fptrunc.round.ll to llvm.fptrunc.round.err.ll (#106452)
Also correct the suffix of the intrinsic
2024-08-28 13:52:07 -07:00