Update our in-tree libvpx code to upstream master from September 25.
This includes a work-around added to skip neon intrinsics which
android toolchains fail to compile properly. See bug 1087096.
Results of running ./update.py --ndk ~/android/android-ndk-r9 --commit
587ff646f
This updates our in-tree copy of libvpx to the
v1.3.0 git tag (2e88f2f2ec777259bda1714e72f1ecd2519bceb5)
libvpx 1.3.0 adds support for VP9. VP9 support is built
but not yet exposed with this commit.
Our update.sh script is replaced with update.py that can
update the build system to a given git commit.
- checkout out upstream git
- create platform dependend config files
- add/remove changed libvpx files
- update moz.build
- warn about new build categories in libvpx
This updates our in-tree copy of libvpx to match the 1.2.0 git
tag. All but one of the patches we were carrying are in this
upstream version. Our update.sh script should copy the new
files needed but will not remove the old ones for you.
Runtime cpu detection was rewritten upstream. We now generate
per-platform headers for this and include the correct one from
a vpx_rtcd.h wrapper like we were already doing for vpx_config.h
This revision includes improved assembly optimizations and should
be faster on all platforms.
Includes work by Jan Gerber and Ralph Giles.
--HG--
rename : media/libvpx/vp8/common/arm/neon/save_neon_reg.asm => media/libvpx/vp8/common/arm/neon/save_reg_neon.asm
rename : media/libvpx/vp8/common/arm/armv6/vp8_mse16x16_armv6.asm => media/libvpx/vp8/encoder/arm/armv6/vp8_mse16x16_armv6.asm
rename : media/libvpx/vp8/common/arm/neon/vp8_mse16x16_neon.asm => media/libvpx/vp8/encoder/arm/neon/vp8_mse16x16_neon.asm
rename : media/libvpx/vp8/encoder/x86/x86_csystemdependent.c => media/libvpx/vp8/encoder/x86/vp8_enc_stubs_mmx.c
rename : media/libvpx/vpx_config_arm-linux-gcc.c => media/libvpx/vpx_config_armv7-android-gcc.c
rename : media/libvpx/vpx_config_arm-linux-gcc.h => media/libvpx/vpx_config_armv7-android-gcc.h
