aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/c93.c
diff options
context:
space:
mode:
authorRémi Denis-Courmont <remi@remlab.net>2024-06-28 21:56:12 +0300
committerRémi Denis-Courmont <remi@remlab.net>2024-07-04 19:57:42 +0300
commit5a6e333fc7ac514255bcd6b424924d92ef558bf0 (patch)
tree94d07565f764cc0a52d30acaa44baa8af06253b5 /libavcodec/c93.c
parent03175b587cdc445d6d0de87ada0af0e6692e3dea (diff)
downloadffmpeg-5a6e333fc7ac514255bcd6b424924d92ef558bf0.tar.gz
lavc/h264dsp: R-V V 8-bit luma loop filter
T-Head C908 (cycles): h264_h_loop_filter_luma_8bpp_c: 297.5 h264_h_loop_filter_luma_8bpp_rvv_i32: 369.2 h264_v_loop_filter_luma_8bpp_c: 862.7 h264_v_loop_filter_luma_8bpp_rvv_i32: 199.7 Performance in the horizontal scenario seems worse than scalar. x86 SSE2 and AVX optimisations are similarly affected. This is presumably caused by unlucky inputs from checkasm, such that the C code short-circuits almost all filter calculations.
Diffstat (limited to 'libavcodec/c93.c')
0 files changed, 0 insertions, 0 deletions