https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125279
Bug ID: 125279
Summary: Wrong lowering of `vmaxh_f16` and `vminh_f16`
Product: gcc
Version: 17.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: sayantn05 at gmail dot com
Target Milestone: ---
Currently `vmaxh_f16` and `vminh_f16` are lowered to `fmaxnm` and `fminnm`,
respectively. But according to the ARM intrinsics reference, they should lower
to
[fmax](https://developer.arm.com/architectures/instruction-sets/intrinsics/vmaxh_f16)
and
[fmin](https://developer.arm.com/architectures/instruction-sets/intrinsics/vminh_f16)
respectively.
An example can be seen in [Godbolt](https://godbolt.org/z/b8zaaaT9n), Clang
produces the correct assembly.
GCC version: aarch64-unknown-linux-gnu-gcc (GCC) 17.0.0 20260511
