]>
granicus.if.org Git - libvpx/log
Ronald S. Bultje [Thu, 1 Oct 2015 01:37:20 +0000 (21:37 -0400)]
vp10: skip unreachable cat6 token extrabits.
We have historically added new bits to cat6 whenever we added a new
transform size (or bitdepth, for that matter). However, we have
always coded these new bits regardless of the actual transform size,
which means that for smaller transforms, we code bits that cannot
possibly be set. The coding (quality) impact of this is negligible,
but the bigger issue is that this allows creating bitstreams with
coefficient values that are nonsensible and can cause int overflows,
which then de facto become part of the bitstream spec. By not coding
these bits, we remove this possibility.
See issue 1065.
Change-Id: Ib3186eca2df6a7a15ddc60c8b55af182aadd964d
Ronald S. Bultje [Wed, 30 Sep 2015 21:16:28 +0000 (17:16 -0400)]
vp10: remove superframe size field for last frame in superframe.
This is identical to what the tile size does for the last tile. See
issue 1042 (which covers generalizing the superframe/tile concepts).
Change-Id: I1f187d2e3b984e424e3b6d79201b8723069e1a50
Ronald S. Bultje [Wed, 30 Sep 2015 20:42:59 +0000 (16:42 -0400)]
vp10: use superframe marker index/size mechanism for tile size.
See issue 1042. Should provide slight bitstream savings in most cases
where tiles are being used.
Change-Id: Ie2808cf8ef30b3efe50804396900c4d63a3fa026
Ronald S. Bultje [Thu, 1 Oct 2015 02:01:41 +0000 (22:01 -0400)]
vp10: extend range for delta Q values.
See issue 1051. 6 bits is fairly arbitrary but at least allows writing
delta Q values that are fairly normal in other codecs. I can extend to
8 if people want full range, although I personally don't have any need
for that.
Change-Id: I0a5a7c3d9b8eb3de4418430ab0e925d4a08cd7a0
Ronald S. Bultje [Thu, 1 Oct 2015 01:57:26 +0000 (21:57 -0400)]
vp10: make render_width/height referenceable through ref frames.
See issue 1079.
Change-Id: I754a61ec011c3508bbb26826cf8e11dbdfdd8379
Ronald S. Bultje [Fri, 2 Oct 2015 17:15:59 +0000 (17:15 +0000)]
Merge "vp10: reimplement d45/4x4 to match vp8 instead of vp9."
Debargha Mukherjee [Fri, 2 Oct 2015 06:55:55 +0000 (06:55 +0000)]
Merge "Accelerated transform in high bit depth"
Marco Paniconi [Fri, 2 Oct 2015 03:20:22 +0000 (03:20 +0000)]
Merge "Two-steps scaling in VP9 encoder dynamic resizing."
jackychen [Thu, 1 Oct 2015 21:03:22 +0000 (14:03 -0700)]
Two-steps scaling in VP9 encoder dynamic resizing.
Dynamic resizing now support two-steps scaling: first go down to
3/4 and then 1/2. This feature is under a flag which controls the
switch between two-steps scaling and one-step scaling (1/2 only).
Change-Id: I3a6c1d3d5668cf8e016a0a02aeca737565604a0f
hui su [Thu, 1 Oct 2015 18:19:13 +0000 (11:19 -0700)]
Small cleanup
Change-Id: I5aeaa94b743f84738d288f8b027fec4c164f2ec3
Scott LaVarnway [Thu, 1 Oct 2015 17:31:10 +0000 (17:31 +0000)]
Merge "VP9: remove plane_type from macroblockd_plane"
Ronald S. Bultje [Wed, 30 Sep 2015 22:45:08 +0000 (18:45 -0400)]
vp10: reimplement d45/4x4 to match vp8 instead of vp9.
This is more a proof of concept than anything else. The problem here
isn't so much how to code it, but rather where to place the resulting
code. All intrapred DSP code lives in vpx_dsp, so do we want the vp10
specific intra pred functions to live there, or in vp10/?
See issue 1015.
Change-Id: I675f7badcc8e18fd99a9553910ecf3ddf81f0a05
Ronald S. Bultje [Thu, 1 Oct 2015 12:30:49 +0000 (08:30 -0400)]
vp8: align left pixel array by 16 bytes.
The x86 simd expects this. Identical alignment can be found in vp9
and vp10 also. Fixes crashes on 32bit x86 systems.
Change-Id: I229c88d8f696acbef5337c8fa9503528df4e1c40
James Zern [Thu, 1 Oct 2015 02:10:22 +0000 (02:10 +0000)]
Merge "sixtap_predict_test: enable NEON tests"
Ronald S. Bultje [Thu, 1 Oct 2015 01:01:57 +0000 (01:01 +0000)]
Merge "vp8: change build_intra4x4_predictors() to use vpx_dsp."
Ronald S. Bultje [Thu, 1 Oct 2015 01:01:45 +0000 (01:01 +0000)]
Merge "vp8: change build_intra_predictors_mbuv_s to use vpx_dsp."
Ronald S. Bultje [Thu, 1 Oct 2015 00:57:37 +0000 (00:57 +0000)]
Merge "vp8: change build_intra_predictors_mby_s to use vpx_dsp."
Ronald S. Bultje [Wed, 30 Sep 2015 22:44:37 +0000 (18:44 -0400)]
vp8: change build_intra4x4_predictors() to use vpx_dsp.
I've added a few new functions (d45e, d63e, he, ve) to cover the
filtered h/v 4x4 predictors that are vp8-specific, the "correct"
d45 with the correctly filtered bottom-right pixel (as opposed to
the unfiltered version in vp9), and the "broken" d63 with weirdly
filtered bottom-right pixels (which is correctly filtered in vp9).
There may be a minor performance impact on all systems because we
have to do an extra copy of the Above pixel array to incorporate
the topleft pixel in the same array (thus fitting the vpx_dsp API).
In addition, armv6 will have a more serious performance impact b/c
I removed the armv6/vp8-specific assembly. I'm not sure anyone
cares...
Change-Id: I7f9e5ebee11d8e21aca2cd517a69eefc181b2e86
Ronald S. Bultje [Wed, 30 Sep 2015 15:46:35 +0000 (11:46 -0400)]
vp8: change build_intra_predictors_mbuv_s to use vpx_dsp.
Change-Id: I936c2430c3c5b1e0ab5dec0a20110525e925b5e4
Ronald S. Bultje [Wed, 30 Sep 2015 15:07:35 +0000 (11:07 -0400)]
vp8: change build_intra_predictors_mby_s to use vpx_dsp.
Change-Id: I2000820e0c04de2c975d370a0cf7145330289bb2
Scott LaVarnway [Tue, 29 Sep 2015 17:40:27 +0000 (10:40 -0700)]
VP9: remove plane_type from macroblockd_plane
Change-Id: Ia5072a3a92212d8565f33359f6c146469bdfbbec
Scott LaVarnway [Wed, 30 Sep 2015 22:11:21 +0000 (22:11 +0000)]
Merge "VP9: remove plane_type checks in loopfilter functions"
James Zern [Wed, 30 Sep 2015 21:52:38 +0000 (21:52 +0000)]
Merge changes I264e75bf,Ifb0f41fb
* changes:
vp9_loopfilter: remove unnecessary masks
vp9_reset_lfm: harmonize function signature
James Zern [Wed, 30 Sep 2015 21:51:51 +0000 (21:51 +0000)]
Merge changes I68c4f189,Ia5a752db
* changes:
vp9_thread_test: clarify test case names
vp9_thread_test: add non-frame-parallel files
James Zern [Wed, 30 Sep 2015 19:55:36 +0000 (19:55 +0000)]
Merge "test/*.h: (windows) fix min/max conflict"
James Zern [Wed, 30 Sep 2015 03:55:14 +0000 (20:55 -0700)]
vp9_loopfilter: remove unnecessary masks
Change-Id: I264e75bf3ddd083ee5311c50a37fb18fe634ddc3
James Zern [Wed, 30 Sep 2015 03:28:47 +0000 (20:28 -0700)]
test/*.h: (windows) fix min/max conflict
define NOMINMAX to allow the std:: versions to be used; min/max will be
defined transitively via windows.h otherwise
Change-Id: I692b03fa3e70b7a53962d3fd209498f70f712fed
James Zern [Wed, 30 Sep 2015 03:46:37 +0000 (20:46 -0700)]
vp9_reset_lfm: harmonize function signature
Change-Id: Ifb0f41fb43564a777be29b4c66443b366fa146a3
Alex Converse [Wed, 30 Sep 2015 01:49:21 +0000 (18:49 -0700)]
Change dynamic_cast to static_cast to fix no-rtti build
Change-Id: Iad73b490b171cdda5c368ada69fb8eab2a86c156
Alex Converse [Wed, 30 Sep 2015 01:20:30 +0000 (01:20 +0000)]
Merge "Add a test for the interaction between active map and cyclic referesh."
Scott LaVarnway [Tue, 29 Sep 2015 15:34:26 +0000 (08:34 -0700)]
VP9: remove plane_type checks in loopfilter functions
vp9_filter_block_plane_ss11() and vp9_filter_block_plane_non420()
are only called for the uv planes.
Change-Id: Iacd3b3242c8ce581edd37c8f06d95efc8a0f88a3
Scott LaVarnway [Tue, 29 Sep 2015 21:40:48 +0000 (21:40 +0000)]
Merge "VP9: move loopfilter build masks to decode loop"
Tom Finegan [Tue, 29 Sep 2015 18:56:21 +0000 (18:56 +0000)]
Merge "vpxenc: Allow non i420 input for VP10."
Marco Paniconi [Tue, 29 Sep 2015 17:47:39 +0000 (17:47 +0000)]
Merge "aq-mode for SVC: Add consec_zero_mv to layer context."
Tom Finegan [Tue, 29 Sep 2015 17:45:00 +0000 (10:45 -0700)]
vpxenc: Allow non i420 input for VP10.
BUG=https://code.google.com/p/webm/issues/detail?id=1066
Change-Id: I3bd26a516ef3d2742c523af570f639f9312df6df
Yaowu Xu [Tue, 29 Sep 2015 17:22:49 +0000 (17:22 +0000)]
Merge "Fix a macro definition"
Tom Finegan [Tue, 29 Sep 2015 17:08:55 +0000 (17:08 +0000)]
Merge "build/make/iosbuild.sh: Remove jobs argument."
Marco [Tue, 29 Sep 2015 17:00:16 +0000 (10:00 -0700)]
aq-mode for SVC: Add consec_zero_mv to layer context.
Change-Id: I63fadf1c7240d4b2893384f75c519311e9659d47
Yaowu Xu [Tue, 29 Sep 2015 16:34:42 +0000 (09:34 -0700)]
Fix a macro definition
to be consistent with the head file name.
Change-Id: I9634332a2b3fac7e7f3b7ef58821ea7c81c5c813
Scott LaVarnway [Tue, 29 Sep 2015 12:20:49 +0000 (05:20 -0700)]
VP9: move loopfilter build masks to decode loop
The loopfilter masks are now built in the decode loop.
This is done so we can eventually reduce the number of
MODE_INFO structs required by the decoder.
The encoder builds the masks for the entire frame prior
to calling the loopfilter.
Change-Id: Ia2146b07e0acb8c50203e586dfae0c4c5b316f11
Julia Robson [Mon, 28 Sep 2015 15:50:39 +0000 (16:50 +0100)]
Accelerated transform in high bit depth
When configured with high bitdepth enabled, the 8bit transform
stopped using optimised code. This made 8bit content decode slowly.
Change-Id: I67d91f9b212921d5320f949fc0a0d3f32f90c0ea
Marco Paniconi [Mon, 28 Sep 2015 19:53:10 +0000 (19:53 +0000)]
Merge "VP8: Update rate correction factor for drop_overshoot feature."
Marco [Mon, 28 Sep 2015 15:31:06 +0000 (08:31 -0700)]
VP8: Update rate correction factor for drop_overshoot feature.
Update rate correction factor when we drop the frame due to overshoot.
Only affects when the drop_overshoot feature is on: screen_content_mode = 2.
Change-Id: I67e24de979b4c74744151d2ceb3cd75fec2a1e7a
Angie Chiang [Mon, 28 Sep 2015 17:26:22 +0000 (17:26 +0000)]
Merge "comment out fdct32"
Ronald S. Bultje [Mon, 28 Sep 2015 16:25:28 +0000 (16:25 +0000)]
Merge "vp9/10: improve support for render_width/height."
Ronald S. Bultje [Mon, 28 Sep 2015 16:25:20 +0000 (16:25 +0000)]
Merge "Rename display_{size,width,height} to render_*."
Ronald S. Bultje [Mon, 28 Sep 2015 16:23:11 +0000 (16:23 +0000)]
Merge "vp10: code reference_mode in uncompressed header."
Ronald S. Bultje [Mon, 28 Sep 2015 16:23:00 +0000 (16:23 +0000)]
Merge "vp10: split UV int4x4 loopfilter flag in one for each covered edge."
Ronald S. Bultje [Sat, 26 Sep 2015 01:51:55 +0000 (21:51 -0400)]
vp9/10: improve support for render_width/height.
In the decoder, map this to the output variable vpx_image_t.r_w/h.
This is intended as an improved version of VP9D_GET_DISPLAY_SIZE,
which doesn't work with parallel frame decoding. In the encoder,
map this to a codec control func (VP9E_SET_RENDER_SIZE) that takes
a w/h pair argument in a int[2] (identical to VP9D_GET_DISPLAY_SIZE).
Also add render_size to the encoder_param_get_to_decoder unit test.
See issue 1030.
Change-Id: I12124c13602d832bf4c44090db08c1009c94c7e8
James Zern [Sat, 26 Sep 2015 01:52:52 +0000 (01:52 +0000)]
Merge "vp9/10 encoder: prevent NULL access on failure"
Ronald S. Bultje [Sat, 26 Sep 2015 01:34:29 +0000 (21:34 -0400)]
Rename display_{size,width,height} to render_*.
The name "display_*" (or "d_*") is used for non-compatible information
(that is, the cropped frame dimensions in pixels, as opposed to the
intended screen rendering surface size). Therefore, continuing to use
display_* would be confusing to end users. Instead, rename the field
to render_*, so that struct vpx_image can include it.
Change-Id: Iab8d2eae96492b71c4ea60c4bce8121cb2a1fe2d
Ronald S. Bultje [Sat, 26 Sep 2015 01:20:14 +0000 (01:20 +0000)]
Merge "vp10: remove MACROBLOCK.{highbd_,}itxfm_add function pointer."
Ronald S. Bultje [Sat, 26 Sep 2015 01:19:49 +0000 (01:19 +0000)]
Merge "vp10: remove MACROBLOCK.fwd_txm4x4 function pointer."
Angie Chiang [Sat, 26 Sep 2015 01:18:27 +0000 (18:18 -0700)]
comment out fdct32
comment out fdct32
remove fdct32 test
Change-Id: I31c47fb435377465cd3265e39621ca50d3aae656
Ronald S. Bultje [Sat, 26 Sep 2015 00:32:14 +0000 (20:32 -0400)]
vp10: code reference_mode in uncompressed header.
See issue 1041 point 2.
Change-Id: I6fc6427b1a0edff828e39d43428e3271491f8ac5
Ronald S. Bultje [Wed, 9 Sep 2015 20:50:07 +0000 (16:50 -0400)]
vp10: split UV int4x4 loopfilter flag in one for each covered edge.
In practice, this fixes the issue that if you have an odd number of
mi_cols, on the full right of the image, the UV int4x4 loopfilter
will be skipped over odd cols as well as odd rows (because it holds a
single variable for both edges).
See issue 1016.
Change-Id: Id53b501cbff9323a8239ed4775ae01fe91874b7e
James Zern [Sat, 26 Sep 2015 00:12:43 +0000 (00:12 +0000)]
Merge "Revert "remove static from fdct4/8/16/32""
Ronald S. Bultje [Fri, 11 Sep 2015 19:45:23 +0000 (15:45 -0400)]
vp10: remove MACROBLOCK.{highbd_,}itxfm_add function pointer.
This is preparatory work for allowing per-segment lossless coding.
See issue 1035.
Change-Id: I9487d02717ee3e766aee61a487780056bb35d2d3
Ronald S. Bultje [Tue, 22 Sep 2015 18:01:48 +0000 (14:01 -0400)]
vp10: remove MACROBLOCK.fwd_txm4x4 function pointer.
This is preparatory work for allowing per-segment lossless coding.
See issue 1035.
Change-Id: Idd72e2a42d90fa7319c10122032d1a7c7a54dc05
Tom Finegan [Fri, 25 Sep 2015 22:18:17 +0000 (15:18 -0700)]
build/make/iosbuild.sh: Remove jobs argument.
This can be handled via MAKEFLAGS.
Change-Id: I3a58a8a41f6570cb3b80c7c97e51735b82bf4ec9
Tom Finegan [Fri, 25 Sep 2015 22:14:38 +0000 (22:14 +0000)]
Merge "build/make/configure.sh: Embed bitcode in arm darwin targets."
Alex Converse [Wed, 16 Sep 2015 04:18:32 +0000 (21:18 -0700)]
Add a test for the interaction between active map and cyclic referesh.
Fails with Icac63051bf37c7355e661837b57c257d58c764fc reverted.
Change-Id: I460d7a5a74faa4daace25f911f8dc5f68e16c951
James Zern [Fri, 25 Sep 2015 19:31:07 +0000 (19:31 +0000)]
Merge "vp9/10 decoder_remove: check pbi pointer"
Marco Paniconi [Fri, 25 Sep 2015 18:59:58 +0000 (18:59 +0000)]
Merge "VP8: Adjust rate correction factor for drop due to overshoot."
Marco [Tue, 23 Jun 2015 15:34:06 +0000 (08:34 -0700)]
Limit cyclic refresh on steady background blocks.
Use the existing QP condition on limiting cyclic refresh, and add
addiitonal condition that block has been encoded with zero/small motion
x frames in row (where x is at least several times the refresh period).
Additional condition only affect non-screen content mode.
This helps to improve visual stability for noisy input, where on steady
background areas the application of delta_qp may lead to encoding the noise.
Also added a change to use the true skip (after encoding) to update the
last QP.
Change-Id: I234a1128d017d284cf767fdb58ef6c59d809f679
James Zern [Fri, 25 Sep 2015 03:22:15 +0000 (03:22 +0000)]
Merge "configure: reference the README for missing yasm"
James Zern [Fri, 25 Sep 2015 02:50:42 +0000 (19:50 -0700)]
vp9/10 decoder_remove: check pbi pointer
fixes crash on error
Change-Id: Ibb1ef5565fb833cdee1a49335473d98f1187ef43
Jacky Chen [Fri, 25 Sep 2015 01:04:07 +0000 (01:04 +0000)]
Merge "Change size on first frame and change config cause crash."
Marco [Thu, 24 Sep 2015 22:15:29 +0000 (15:15 -0700)]
VP8: Adjust rate correction factor for drop due to overshoot.
Change-Id: Id70ca2e18a46247720eb631ae13a8430bd8b0954
Tom Finegan [Thu, 24 Sep 2015 21:55:51 +0000 (14:55 -0700)]
build/make/configure.sh: Embed bitcode in arm darwin targets.
When the iOS SDK major version is 9 or higher:
- Pass -fembed-bitcode to compiler, assembler, and linker.
- Add a warning for simulator targets since yasm doesn't know
what -fembed-bitcode means, and exits with an error.
BUG=https://code.google.com/p/webm/issues/detail?id=1075
Change-Id: I38c997a0225e53c5dd1b4ddf7935d21362953f76
Tom Finegan [Thu, 24 Sep 2015 18:44:32 +0000 (18:44 +0000)]
Merge "build/make/configure.sh: Silence arm target Xcode7 link warnings."
Tom Finegan [Thu, 24 Sep 2015 18:44:23 +0000 (18:44 +0000)]
Merge "build/make/configure.sh: Fix armv7 builds in Xcode7."
Tom Finegan [Fri, 18 Sep 2015 23:01:32 +0000 (16:01 -0700)]
build/make/configure.sh: Silence arm target Xcode7 link warnings.
Always add IOS_VERSION_MIN to darwin arm cflags. The warning occured
because the default (9.0) does not match the value set by configure
(6.0).
BUG=https://code.google.com/p/webm/issues/detail?id=1075
Change-Id: Ia9085ceeca10e057f9eb781c14f07581bb6280a5
Tom Finegan [Fri, 18 Sep 2015 22:58:41 +0000 (15:58 -0700)]
build/make/configure.sh: Fix armv7 builds in Xcode7.
- Use the iphoneos SDK path (instead of macosx).
- Detect iOS SDK major version and disable media (armv6) when using
iOS SDK version 9 or higher.
BUG=https://code.google.com/p/webm/issues/detail?id=1075
Change-Id: I12f77dbeee4c0084e8322f6841813da8b5e91c16
Tom Finegan [Wed, 23 Sep 2015 18:42:37 +0000 (11:42 -0700)]
build/make/configure.sh: docs for soft_{dis|en}able.
Add function comments explaining what the functions do and do not do.
Change-Id: I23dea09f93bc5cdbea6a0077f90683a1df2f74dc
James Zern [Fri, 11 Sep 2015 04:08:46 +0000 (21:08 -0700)]
vp9_thread_test: clarify test case names
rename Decode[2-4] to something more precise
Change-Id: I68c4f189796eb11ac1a5b7b682f24efb71708187
James Zern [Fri, 11 Sep 2015 04:04:39 +0000 (21:04 -0700)]
vp9_thread_test: add non-frame-parallel files
these have been supported in tile-threaded decoding since:
b3b7645 vp9_dthread: remove frame_parallel_decoding_mode requirement
Change-Id: Ia5a752db9be937153cf4830d9258752136356d1b
James Zern [Thu, 24 Sep 2015 00:55:51 +0000 (17:55 -0700)]
vp9/10 encoder: prevent NULL access on failure
Change-Id: I1fc8e0b3d48675cd5428b7b36f7cc28ab32cbf71
James Zern [Thu, 24 Sep 2015 00:51:42 +0000 (17:51 -0700)]
configure: reference the README for missing yasm
Change-Id: I2ad799901385011764affadeaddcc271df21509f
James Zern [Thu, 24 Sep 2015 00:45:27 +0000 (17:45 -0700)]
Revert "remove static from fdct4/8/16/32"
This reverts commit
8903b9fa8345726efbe9b92a759c98cc21c4c14b .
there is no reason for these to be global
Change-Id: I66a31c06f8426aeca348ef12d9b9ab59d6d5e55d
James Zern [Thu, 24 Sep 2015 00:33:16 +0000 (00:33 +0000)]
Merge "VP9: Remove frame_parallel_decoding_mode from macroblockd"
Marco Paniconi [Thu, 24 Sep 2015 00:12:02 +0000 (00:12 +0000)]
Merge "Non-rd mode: Limit transform size for intra to 16x16."
Scott LaVarnway [Wed, 23 Sep 2015 23:06:46 +0000 (16:06 -0700)]
VP9: Remove frame_parallel_decoding_mode from macroblockd
Not used.
Change-Id: I71527d0ee43a5730f1a2527e7ab687a77a137db4
Hui Su [Wed, 23 Sep 2015 21:39:43 +0000 (21:39 +0000)]
Merge "Adjust rd calculation in choose_tx_size_from_rd"
Marco [Wed, 23 Sep 2015 18:47:20 +0000 (11:47 -0700)]
Adjust rate-boost threshold in cyclic refresh for seg#2.
Small gain in metrics (average ~0.2dB), small
reduction in rate fluctuation.
Change-Id: Id75bd89c168486f075308fb474ebd26e3bdfb85b
Marco [Tue, 22 Sep 2015 17:42:55 +0000 (10:42 -0700)]
Non-rd mode: Limit transform size for intra to 16x16.
Limit transform size for intra to 16x16, for non-screen content mode.
Little/no change in speed or metrics.
32x32 intra block is rarley selected in RTC (non-screen content) case,
but some visual improvement can be seen in some example,
e.g., captured_video_dark_whd.yuv.
Change-Id: I68e2db87875343b3fb9bb407a7709f0088f84072
hui su [Tue, 22 Sep 2015 21:08:28 +0000 (14:08 -0700)]
Adjust rd calculation in choose_tx_size_from_rd
Coding gain:
derflr 0.142%
hevclr 0.153%
hevcmr 0.124%
Change-Id: I63b56ae3a9002c3a266e10e2964135ed43b0ba53
Johann [Mon, 21 Sep 2015 23:55:28 +0000 (16:55 -0700)]
Restrict get_msb inputs
Add a warning and assert that inputs for get_msb must not be zero.
Change-Id: I8c6f289ff13248f6e3a8bc24aab3712ed33022a6
Angie Chiang [Mon, 21 Sep 2015 23:25:26 +0000 (23:25 +0000)]
Merge "remove static from fdct4/8/16/32"
Johann Koenig [Mon, 21 Sep 2015 19:17:18 +0000 (19:17 +0000)]
Merge "Remove vpx_filter_block1d16_v8_intrin_ssse3"
Angie Chiang [Sat, 19 Sep 2015 01:46:22 +0000 (18:46 -0700)]
remove static from fdct4/8/16/32
remove static from fdct4/8/16/32 in vp10/encoder/dct.c
add prefix vp10_ to fdct4/8/16/32
add vp10/encoder/dct.h
Change-Id: I644827a191c1a7761850ec0b1da705638b618c66
Marco Paniconi [Mon, 21 Sep 2015 18:26:28 +0000 (18:26 +0000)]
Merge "Non-rd pickmode: Don't skip checking zeromv-last mode."
jackychen [Mon, 21 Sep 2015 16:37:46 +0000 (09:37 -0700)]
Change size on first frame and change config cause crash.
Reallocation of mi buffer fails if change size on the first frame and
change config in subsequent frames. Add a condition for resolution
check to avoid assertion failure.
BUG=1074
Change-Id: Ie26ed816a57fa871ba27a72db9805baaaeaba9f3
Marco [Mon, 21 Sep 2015 17:20:14 +0000 (10:20 -0700)]
Non-rd pickmode: Don't skip checking zeromv-last mode.
Reference frame masking logic may skip checking zeromv-last mode.
Fix to avoid this and make sure zero-last is always checked.
No noticeable change in speed, and PSNR/SSIM metrics on RTC set overall
neutral (very small gain ~0.02).
Small visual improvement on few RTC clips.
Change-Id: I26eacdc449126424001a4a64e5ac31949f064417
Jingning Han [Mon, 21 Sep 2015 16:13:37 +0000 (16:13 +0000)]
Merge "Create sub8x8 block inter prediction function"
James Zern [Sat, 19 Sep 2015 16:14:18 +0000 (09:14 -0700)]
vp9_end_to_end_test: disable vp10 w/high bitdepth
the range check in dct.c (abs(input[i]) < (1 << bit)) will fail in many
cases. this was broken at the time this check was added
BUG=1076
Change-Id: I3df8c7a555e95567d73ac16acda997096ab8d6e2
Jingning Han [Fri, 18 Sep 2015 23:27:52 +0000 (16:27 -0700)]
Create sub8x8 block inter prediction function
Change-Id: Ib161e6fb3eb081f7176a1d969fed16a7d1ffc320
James Zern [Fri, 18 Sep 2015 23:18:10 +0000 (23:18 +0000)]
Merge "configure: add --extra-cxxflags option"
Johann [Fri, 18 Sep 2015 22:59:12 +0000 (15:59 -0700)]
Remove vpx_filter_block1d16_v8_intrin_ssse3
This was rewritten and moved to vpx_dsp/x86/vpx_subpixel_8t_ssse3.asm
in
195883023bb39b5ee5c6811a316ab96d9225034d
Change-Id: I117ce983dae12006e302679ba7f175573dd9e874
Tom Finegan [Fri, 18 Sep 2015 19:38:55 +0000 (19:38 +0000)]
Merge "iosbuild: Enable PIC for x86 targets."