search

llvm / llvm-project

The LLVM Project is a collection of modular and reusable compiler and toolchain technologies.
http://llvm.org
Other
28.31k stars 11.69k forks source link

[AArch64] NEON intrinsic compilation error occurs when using -fno-lax-vector-conversions #71362

Open yyctw opened 11 months ago

yyctw commented 11 months ago

This appears to be the same issue as the previous one https://github.com/llvm/llvm-project/issues/42686.

I discovered that certain NEON intrinsics failed to compile with the -fno-lax-vector-conversions flag, a problem that did not occur in GCC.

The NEON intrinsics that have this issue are listed below:

Thank you for your reading.

llvmbot commented 11 months ago

@llvm/issue-subscribers-backend-aarch64

Author: Yi-Yen Chung (yyctw)

This appears to be the same issue as the previous one https://github.com/llvm/llvm-project/issues/42686. I discovered that certain NEON intrinsics failed to compile with the `-fno-lax-vector-conversions` flag, a problem that did not occur in GCC. - Reproduce error: https://godbolt.org/z/Mhxqj9151 - Compare with GCC: https://godbolt.org/z/qc31v3no8 The NEON intrinsics that have this issue are listed below: - `vdupb_lane_p8`, `vdupb_laneq_p8`, `vdupbh_lane_p16`, `vduph_laneq_p16` - `vset_lane_p8`, `vset_lane_p16`, `vset_lane_p64`, `vsetq_lane_p8`, `vsetq_lane_p16`, `vsetq_lane_p64` - `vget_lane_p8`, `vget_lane_p16`, `vget_lane_p64`, `vgetq_lane_p8`, `vgetq_lane_p16`, `vgetq_lane_p64` - `vcopy_lane_p8`, `vcopy_lane_p16`, `vcopy_lane_p64`, `vcopyq_lane_p8`, `vcopyq_lane_p16`, `vcopyq_lane_p64`, - `vcopy_laneq_p8`, `vcopy_laneq_p16`, `vcopy_laneq_p64`, `vcopyq_laneq_p8`, `vcopyq_laneq_p16`, `vcopyq_laneq_p64` Thank you for your reading.