virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2025-01-24 13:56:33 +02:00
Code Issues Releases Activity

lavc/aarch64: Fix ff_pred8x8_plane_neon_10

Browse Source 
Fix test failure on aarch64:
./tests/checkasm/checkasm --test=h264pred 479612

The mismatch between neon and C functions can also be reproduced using the following bitstream and command line.

wget https://streams.videolan.org/ffmpeg/incoming/intra8x8pred_10bit.264
 ./ffmpeg -cpuflags 0  -threads 1 -i intra8x8pred_10bit.264  -f framemd5 -y md5_ref
 ./ffmpeg              -threads 1 -i intra8x8pred_10bit.264  -f framemd5 -y md5_neon

Signed-off-by: Bin Peng <pengbin@visionular.com>
Signed-off-by: Martin Storsjö <martin@martin.st>
(cherry picked from commit decc9e643cc3ac5537f42b465e2637fbefbf41cc)
This commit is contained in:
Bin Peng 2024-12-16 10:31:23 +08:00 committed by Martin Storsjö
parent 2a29fe87c4
commit d0c887017a
1 changed files with 2 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
libavcodec/aarch64/h264pred_neon.S
View File

@ -595,12 +595,11 @@ function ff_pred8x8_plane_neon_10, export=1
ssubl v2.4s, v2.4h, v3.4h
ext v0.16b, v0.16b, v0.16b, #14
mov v0.h[0], wzr
mul v0.8h, v0.8h, v5.h[0]
dup v1.4s, v2.s[0]
dup v2.4s, v2.s[0]
dup v3.8h, v5.h[1]
saddw v1.4s, v1.4s, v0.4h
saddw2 v2.4s, v2.4s, v0.8h
smlal v1.4s, v0.4h, v5.h[0]
smlal2 v2.4s, v0.8h, v5.h[0]
mov w3, #8
mvni v4.8h, #0xFC, lsl #8 // 1023 for clipping
1: