commit f00c43f4d6e812b581f64edc53a655f8e2413938
author George Steed <george.steed@arm.com> Tue May 07 13:41:47 2024 +0100
committer Frank Barchard <fbarchard@chromium.org> Wed Oct 30 17:58:29 2024 +0000
tree 7ac0170c9f89bffaf220a7f488e2349c88248893
parent 51d07554a039004ff278009852f9d33c0f76bf91

[AArch64] Unroll HalfFloat{,1}Row_NEON

The existing C implementation compiled with a recent LLVM is
auto-vectorised and unrolled to process four vectors per loop iteration,
making the Neon implementation slower than the C implementation on
little cores. To avoid this, unroll the Neon implementation to also
process four vectors per iteration.

Reduction in cycle counts observed compared to the existing Neon
implementation:

            | HalfFloat1Row_NEON | HalfFloatRow_NEON
Cortex-A510 |             -37.1% |            -40.8%
Cortex-A520 |             -32.3% |            -37.4%
Cortex-A720 |               0.0% |            -10.6%
  Cortex-X2 |               0.0% |             -7.8%
  Cortex-X4 |              +0.3% |             -6.9%

Bug: b/42280945
Change-Id: I12b474c970fc4355d75ed924c4ca6169badda2bc
Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/5872805
Reviewed-by: Frank Barchard <fbarchard@chromium.org>
Reviewed-by: Justin Green <greenjustin@google.com>