[SVE][InstCombine] Don't convert calls to fp binop intrinsics to instructions when...
authorPaul Walker <paul.walker@arm.com>
Tue, 4 Apr 2023 12:51:25 +0000 (12:51 +0000)
committerPaul Walker <paul.walker@arm.com>
Thu, 6 Apr 2023 10:23:44 +0000 (10:23 +0000)
commit65031c189bc818ee449294b2f02abc55a5f88691
tree17463ae8f99d7f1265e5da44dcb5075da0ccec17
parent5f2145adf9e343e950935cac1c9030f8ca597912
[SVE][InstCombine] Don't convert calls to fp binop intrinsics to instructions when strictfp is required.

There's no support to lower scalable vector constrained operations for SVE
and thus without this change we'll crash during code generation.

Differential Revision: https://reviews.llvm.org/D147600
llvm/lib/Target/AArch64/AArch64TargetTransformInfo.cpp
llvm/test/Transforms/InstCombine/AArch64/sve-intrinsic-strictfp.ll