mirror of
https://chromium.googlesource.com/libyuv/libyuv
synced 2025-12-09 18:26:49 +08:00
432d186116
We can use the dot product instructions to apply the coefficients directly without the need for LD4 de-interleaving load instructions, since these are known to be slow on some micro-architectures. ST4 is also known to be slow on more modern micro-architectures, however avoiding this is left for a future SVE implementation where we can make use of interleaving-narrowing instructions. Reduction in cycle counts observed compared to existing Neon code: Cortex-A55: -5.8% Cortex-A510: -18.9% Cortex-A76: -21.8% Cortex-A720: -30.2% Cortex-X1: -28.6% Cortex-X2: -23.4% Bug: b/42280946 Change-Id: I5887559649cc805a810d867b652c85d48285657d Reviewed-on: https://chromium-review.googlesource.com/c/libyuv/libyuv/+/5790970 Reviewed-by: Justin Green <greenjustin@google.com> Reviewed-by: Frank Barchard <fbarchard@chromium.org> |
||
|---|---|---|
| .. | ||
| basic_types.h | ||
| compare_row.h | ||
| compare.h | ||
| convert_argb.h | ||
| convert_from_argb.h | ||
| convert_from.h | ||
| convert.h | ||
| cpu_id.h | ||
| loongson_intrinsics.h | ||
| macros_msa.h | ||
| mjpeg_decoder.h | ||
| planar_functions.h | ||
| rotate_argb.h | ||
| rotate_row.h | ||
| rotate.h | ||
| row.h | ||
| scale_argb.h | ||
| scale_rgb.h | ||
| scale_row.h | ||
| scale_uv.h | ||
| scale.h | ||
| version.h | ||
| video_common.h | ||