]> granicus.if.org Git - libvpx/commit
projects / libvpx / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: d478d2d) | patch
vp8 quantization -> intrinsics
authorJohann <johannkoenig@google.com>
Mon, 20 Oct 2014 23:03:28 +0000 (16:03 -0700)
committerJohann <johannkoenig@google.com>
Fri, 31 Oct 2014 20:42:13 +0000 (13:42 -0700)
commit7ae75c3d525d79b9b28652fb34082cf81a5de9ab
tree68c01317e641e59a3936479995833e8893be56e1tree | snapshot
parentd478d2df3728da02bd7424184a7c1eb05ee23f5ccommit | diff
vp8 quantization -> intrinsics

Use intrinsics for neon quantization. Slight loss (<5%) of performance
compared to the assembly. Roughly 10x faster on arm64 because that was
running C code before.

Change-Id: I7cf5242d8f29b7eab5bca6a1c20c89c9fc9ca66d
vp8/common/rtcd_defs.pl diff | blob | history
vp8/encoder/arm/neon/fastquantizeb_neon.asm [deleted file] blob | history
vp8/encoder/arm/neon/fastquantizeb_neon.c [new file with mode: 0644] blob
vp8/vp8cx_arm.mk diff | blob | history
Unnamed repository; edit this file 'description' to name the repository.