Welcome to mirror list, hosted at ThFree Co, Russian Federation.

github.com/FFmpeg/FFmpeg.git - Unnamed repository; edit this file 'description' to name the repository.
summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorReimar Döffinger <Reimar.Doeffinger@gmx.de>2021-01-16 04:03:45 +0300
committerJosh Dekker <josh@itanimul.li>2021-02-18 16:11:53 +0300
commit30f80d855bb0eb3742ac51d8900f636968d0b70b (patch)
tree87df9d0d487ec522bea842503a883a5b8f643c9f /libavcodec/aarch64/Makefile
parent67c8c863c70f2989add027ffccccbe4b87988415 (diff)
lavc/aarch64: port HEVC SIMD idct NEON
Makes SIMD-optimized 8x8 and 16x16 idcts for 8 and 10 bit depth available on aarch64. For a UHD HDR (10 bit) sample video these were consuming the most time and this optimization reduced overall decode time from 19.4s to 16.4s, approximately 15% speedup. Test sample was the first 300 frames of "LG 4K HDR Demo - New York.ts", running on Apple M1. Signed-off-by: Josh Dekker <josh@itanimul.li>
Diffstat (limited to 'libavcodec/aarch64/Makefile')
-rw-r--r--libavcodec/aarch64/Makefile2
1 files changed, 2 insertions, 0 deletions
diff --git a/libavcodec/aarch64/Makefile b/libavcodec/aarch64/Makefile
index f6434e40da..2ea1d74a38 100644
--- a/libavcodec/aarch64/Makefile
+++ b/libavcodec/aarch64/Makefile
@@ -61,3 +61,5 @@ NEON-OBJS-$(CONFIG_VP9_DECODER) += aarch64/vp9itxfm_16bpp_neon.o \
aarch64/vp9lpf_neon.o \
aarch64/vp9mc_16bpp_neon.o \
aarch64/vp9mc_neon.o
+NEON-OBJS-$(CONFIG_HEVC_DECODER) += aarch64/hevcdsp_idct_neon.o \
+ aarch64/hevcdsp_init_aarch64.o