diff options
| author | Jonathan Wright <jonathan.wright@arm.com> | 2021-02-02 01:31:32 +0000 |
|---|---|---|
| committer | Jonathan Wright <jonathan.wright@arm.com> | 2021-02-03 14:01:15 +0000 |
| commit | 9a00ff96fad209ebde56b227d313cad5d769dc55 (patch) | |
| tree | 308dbfa2da68f7caa25034c945720d990bc43827 /libcpp | |
| parent | b2c4cf7b19d2441307132727dde0fb63f27d1530 (diff) | |
| download | gcc-9a00ff96fad209ebde56b227d313cad5d769dc55.zip gcc-9a00ff96fad209ebde56b227d313cad5d769dc55.tar.gz gcc-9a00ff96fad209ebde56b227d313cad5d769dc55.tar.bz2 | |
aarch64: Use RTL builtins for [su]mlal_high_lane[q] intrinsics
Rewrite [su]mlal_high_lane[q] Neon intrinsics to use RTL builtins
rather than inline assembly code, allowing for better scheduling and
optimization.
gcc/ChangeLog:
2021-02-02 Jonathan Wright <jonathan.wright@arm.com>
* config/aarch64/aarch64-simd-builtins.def: Add
[su]mlal_hi_lane[q] builtin generator macros.
* config/aarch64/aarch64-simd.md
(aarch64_<su>mlal_hi_lane<mode>_insn): Define.
(aarch64_<su>mlal_hi_lane<mode>): Define.
(aarch64_<su>mlal_hi_laneq<mode>_insn): Define.
(aarch64_<su>mlal_hi_laneq<mode>): Define.
* config/aarch64/arm_neon.h (vmlal_high_lane_s16): Use RTL
builtin instead of inline asm.
(vmlal_high_lane_s32): Likewise.
(vmlal_high_lane_u16): Likewise.
(vmlal_high_lane_u32): Likewise.
(vmlal_high_laneq_s16): Likewise.
(vmlal_high_laneq_s32): Likewise.
(vmlal_high_laneq_u16): Likewise.
(vmlal_high_laneq_u32): Likewise.
Diffstat (limited to 'libcpp')
0 files changed, 0 insertions, 0 deletions