generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
13,127 Commits 55 Branches 16 Tags 63 MiB
14ef4aeafb
Commit Graph

125 Commits

Author SHA1 Message Date
Johann
eabb793f3b Use correct buffer size in vp8 subpixel variance 
In vp8_sub_pixel_variance8x8_neon the temp2 buffer is only initialized
to kHeight8 * kWidth8. However, in the case that xoffset != 0 and
yoffset == 0, var_filter_block2d_bil_w8 is called with output_width
kHeight8PlusOne.

Thanks to cmugurel for diagnosing and yulius for the patch.

Change-Id: Ib71ffd96ffad963c92b8b7ca23f303942785b8e0
https://code.google.com/p/webrtc/issues/detail?id=4190
 2015-02-03 09:11:05 -08:00
Johann
f6be2f3c87 Clarify GCC version check 
The version check was incorrectly matching some versions of clang
which reported as gcc 4.2

Change-Id: I686d3576e71883fe1463206b56ab5e2aa9bb68a8
 2014-09-25 11:53:45 -07:00
Jia Jia
0ae866bd19 vp8/vp9: neon: msvc: move the 'ifdef _MSC_VER' bit to vpx_ports/mem.h. 
fix compiling warning.

Change-Id: If8706a9046436f704c597e4275a6810c76ba7daa
 2014-09-14 01:43:54 +08:00
Jia Jia
c97f5e8b86 vp8 common: change 'HAVE_NEON_ASM' to 'HAVE_NEON' for compiling functions of NEON intrinsics. 
Change-Id: I975e5eac16f8b623ff589f0ec072cdaff2183b04
 2014-09-05 12:24:05 +00:00
James Zern
35fadf1d25 bilinearpredict_neon: fix type conversion warnings 
make bifilter4_coeff[][] uint8_t, no values exceed this range and
they're loaded with vdup_n_u8().

Change-Id: I921983e9edd828d29820e40ac30a7801dbe0fb4f
 2014-09-04 20:50:42 -07:00
James Zern
f61e00c79d Merge "arm: Fix building vp8_subpixelvariance_neon.c with MSVC" 2014-09-04 11:00:53 -07:00
Scott LaVarnway
ec94967ffe Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 10"" 
This reverts commit 677fb5123e

Compiles with 4.6.

Change-Id: I7f87048911b6bc28a61741d95501fa45ee97b819
 2014-09-04 08:51:20 -07:00
Martin Storsjo
0002da32e6 arm: Fix building vp8_subpixelvariance_neon.c with MSVC 
Use the right return values - vget_low_s64 returns int64x1_t, not
a normal int64_t.

Also make __builtin_prefetch a no-op on MSVC for this file.

Change-Id: I4d2fce01d0ba106b98d3d53b137803119c2c2c08
 2014-09-04 09:49:30 +03:00
Scott LaVarnway
dcbfacbb98 Neon version of vp8_build_intra_predictors_mby_s() and 
vp8_build_intra_predictors_mbuv_s().

This patch replaces the assembly version with an intrinsic
version.

On a Nexus 7, vpxenc (in realtime mode, speed -12)
reported a performance improvement of ~2.6%.

Change-Id: I9ef65bad929450c0215253fdae1c16c8b4a8f26f
 2014-09-03 13:41:27 -07:00
Scott LaVarnway
9293d267d2 VP8 for ARMv8 by using NEON intrinsics 17 
Add vp8_subpixelvariance_neon.c
- vp8_sub_pixel_variance16x16_neon_func
- vp8_variance_halfpixvar16x16_h_neon
- vp8_variance_halfpixvar16x16_v_neon
- vp8_variance_halfpixvar16x16_hv_neon
- vp8_sub_pixel_variance8x8_neon

Change-Id: I3e5d85b2eafc26be0eef6a777789b80e4579257b
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-09-03 13:33:44 -07:00
Johann
5b788c0cbe Merge "Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 06" This reverts commit 81ad047ee5. Revert "VP8 for ARMv8 by using NEON intrinsics 15" This reverts commit 727af7cebe3698b8493ba6c1360b0a6606c310fb."" 2014-09-03 13:27:11 -07:00
Scott LaVarnway
652ef29d09 Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 08"" 
This reverts commit 928ff03889

Compiles with 4.6 now.

Change-Id: Ib455da1098bb0e0623248be07579882a425fcbd1
 2014-08-29 13:29:36 -07:00
Johann
911e96a4eb Revert "Revert "VP8 for ARMv8 by using NEON intrinsics 06" This reverts commit 81ad047ee5. Revert "VP8 for ARMv8 by using NEON intrinsics 15" This reverts commit 727af7cebe3698b8493ba6c1360b0a6606c310fb." 
This reverts commit 920f803f2e

Change-Id: I410d9036214a1b18427cca70b4bc6d8239740737
 2014-08-20 09:41:50 -07:00
James Zern
74ed33cf6e vp8_bilinear_predict4x4_neon: init src vectors 
quiets uninitialized warnings on the first load.

Change-Id: I58a5af337087d96b4eaea8991a0f85c4ba58aebe
 2014-07-11 00:05:25 -07:00
James Zern
5e30127c7a vp8_sixtap_predict4x4_neon: init src vectors 
quiets uninitialized warnings on the first load.

Change-Id: Ied9b03928537a9ed2cd414b9e8a0be00191b0f32
 2014-07-10 23:48:47 -07:00
Johann
f625b2ac93 Correct HAVE_NEON_ASM define 
These optimizations are currently disabled.

Change-Id: I19c58c9cb82d017638b86196641b9e001dfa798b
 2014-05-16 08:20:13 -07:00
Johann
2f6f955a17 Remove intermediate step in vp8_dequantize_b 
With the intrinsics it is no longer necessary to have a stub/helper
function.

Change-Id: I3695961c3c94f1bb750d3b7b29716e509ebba482
 2014-05-14 12:24:18 -07:00
Johann
4dcc6d9707 Build armv7a-only code 
Allow disabling the more generic NEON code.
Use filtered option to disable rtcd code.

Change-Id: Icb4500c1a2bac16eed3c5e3ec0c35e92e6bbbb9f
 2014-05-14 12:23:33 -07:00
Johann
920f803f2e Revert "VP8 for ARMv8 by using NEON intrinsics 06" 
This reverts commit 81ad047ee5.
Revert "VP8 for ARMv8 by using NEON intrinsics 15"
This reverts commit 727af7cebe.

This exposes a bug in gcc 4.9 regarding register allocation. Will reland
when 4.9 is fixed.

Change-Id: I2d8a04e4edde93719280e41550f4c0765608ec4d
 2014-05-13 13:21:17 -07:00
Johann
ce23931a3f Only build neon assembly for armv7 targets 
Allow selectively building just the intrinsics for armv8

Change-Id: I2f29b2e4508b8b8e5649c2906b3159ad1d4ec477
 2014-05-12 08:52:02 -07:00
Johann
4bffb75ba3 Merge "Revert "VP8 for ARMv8 by using NEON intrinsics 10"" 2014-05-07 06:47:48 -07:00
Johann
3a695015ad Merge "arm: Use a correct neon vector type for 64 bit integers" 2014-05-07 06:34:25 -07:00
Martin Storsjo
d5d82a5e1a arm: Add a no-op define of __builtin_prefetch for MSVC 
Both GCC and RVCT/ARMCC support __builtin_prefetch, but MSVC
doesn't.

Change-Id: I44e1eecead61bc88d8fdfd3fef03d76d4f5afe08
 2014-05-07 10:43:24 +03:00
Martin Storsjo
82a83c4fe0 arm: Use a correct neon vector type for 64 bit integers 
This fixes building with MSVC.

Change-Id: I763ba8855c8083d82c8b477d3a297e310e93a335
 2014-05-07 10:22:40 +03:00
Johann
677fb5123e Revert "VP8 for ARMv8 by using NEON intrinsics 10" 
This reverts commit c500fc22c1

There is an issue with gcc 4.6 in the Android NDK:
loopfiltersimpleverticaledge_neon.c: In function 'vp8_loop_filter_bvs_neon':
loopfiltersimpleverticaledge_neon.c:176:1: error: insn does not satisfy its constraints:

Change-Id: I95b6509d12f075890308914cc691b813d2e5cd9f
 2014-05-06 14:28:00 -07:00
Johann
928ff03889 Revert "VP8 for ARMv8 by using NEON intrinsics 08" 
This reverts commit a5d79f43b9

There is an issue with gcc 4.6 in the Android NDK:
loopfilter_neon.c: In function 'vp8_loop_filter_vertical_edge_y_neon':
loopfilter_neon.c:394:1: error: insn does not satisfy its constraints:

Change-Id: I2b8c6ee3fa595c152ac3a5c08dd79bd9770c7b52
 2014-05-06 13:20:24 -07:00
Johann
34843e9784 Merge "VP8 for ARMv8 by using NEON intrinsics 16" 2014-05-05 13:09:43 -07:00
Johann
ad42b09f2f Merge "VP8 for ARMv8 by using NEON intrinsics 15" 2014-05-05 13:09:12 -07:00
Johann
1b7291d52c Merge "VP8 for ARMv8 by using NEON intrinsics 14" 2014-05-05 07:08:08 -07:00
Johann
a7355f3bbb Merge changes Iaf7d6b0a,Iece0bf56 
* changes:
  Use INLINE and include vpx_config.h instead of plain 'inline'
  Use vreinterpret instead of casting neon vector types
 2014-05-05 05:36:54 -07:00
Martin Storsjo
7afed9a1b6 Use INLINE and include vpx_config.h instead of plain 'inline' 
This fixes compilation with MSVC.

Change-Id: Iaf7d6b0a0134968a6addf315fde6d852f298db8c
 2014-05-04 22:42:13 +03:00
Martin Storsjo
dfb8fc917a Use vreinterpret instead of casting neon vector types 
MSVC doesn't support casting neon vector types but requires using
vreinterpret.

Change-Id: Iece0bf5632567efd7f37f527abea38afeab4926d
 2014-05-04 22:40:57 +03:00
James Yu
4ea9cf3e2d VP8 for ARMv8 by using NEON intrinsics 16 
Add variance_neon.c
- vp8_variance16x16_neon
- vp8_variance16x8_neon
- vp8_variance8x16_neon
- vp8_variance8x8_neon

Change-Id: Idfb9c96134a1c6a696a98ce68b4f7ed593a00660
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 19:07:40 -07:00
James Yu
727af7cebe VP8 for ARMv8 by using NEON intrinsics 15 
Add idct_dequant_0_2x_neon.c
- idct_dequant_0_2x_neon

Change-Id: I8e129172ef1b2517cf72ff267788921f1a792586
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 19:07:33 -07:00
James Yu
08e38f06db VP8 for ARMv8 by using NEON intrinsics 14 
Add sixtappredict_neon.c
- vp8_sixtap_predict16x16_neon
- vp8_sixtap_predict8x8_neon
- vp8_sixtap_predict8x4_neon
- vp8_sixtap_predict4x4_neon

Change-Id: I3b02fce48ae2e6c6099041ba5ddd7b090f1463b9
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 19:07:12 -07:00
James Yu
18e9caad47 VP8 for ARMv8 by using NEON intrinsics 13 
Add shortidct4x4llm_neon.c
- vp8_short_idct4x4llm_neon

Change-Id: I5a734bbffca8dacf8633c2b0ff07b98aa2f438ba
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 19:07:05 -07:00
James Yu
feaf766bd0 VP8 for ARMv8 by using NEON intrinsics 12 
Add sad_neon.c
- vp8_sad16x16_neon
- vp8_sad16x8_neon
- vp8_sad8x8_neon
- vp8_sad8x16_neon
- vp8_sad4x4_neon

Change-Id: I08eaae49ec03fb91b394354660a5df0367cea311
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 04:54:39 -07:00
James Yu
4a8336fa9d VP8 for ARMv8 by using NEON intrinsics 11 
Add mbloopfilter_neon.c
- vp8_mbloop_filter_horizontal_edge_y_neon
- vp8_mbloop_filter_horizontal_edge_uv_neon
- vp8_mbloop_filter_vertical_edge_y_neon
- vp8_mbloop_filter_vertical_edge_uv_neon

Change-Id: Ia9084e0892d4d49412d9cf2b165a0f719f2382d7
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 04:54:33 -07:00
James Yu
c500fc22c1 VP8 for ARMv8 by using NEON intrinsics 10 
Add loopfiltersimpleverticaledge_neon.c
- vp8_loop_filter_bvs_neon
- vp8_loop_filter_mbvs_neon

Change-Id: I7cf0a161ad4ae37c881b94cc0122f895d3baae79
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 04:11:00 -07:00
James Yu
55c95f2d2c VP8 for ARMv8 by using NEON intrinsics 09 
Add loopfiltersimplehorizontaledge_neon.c
- vp8_loop_filter_bhs_neon
- vp8_loop_filter_mbhs_neon

Change-Id: I77f9721b20585da8bf3869a3850ff0ae4b4bfeea
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-03 04:10:45 -07:00
James Yu
a5d79f43b9 VP8 for ARMv8 by using NEON intrinsics 08 
Add loopfilter_neon.c
- vp8_loop_filter_horizontal_edge_y_neon
- vp8_loop_filter_horizontal_edge_uv_neon
- vp8_loop_filter_vertical_edge_y_neon
- vp8_loop_filter_vertical_edge_uv_neon

Change-Id: I50b57dedabd42d2a3c183c1738cc5346f0e71ed8
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-02 09:32:11 -07:00
James Yu
930557be10 VP8 for ARMv8 by using NEON intrinsics 07 
Add iwalsh_neon.c
- vp8_short_inv_walsh4x4_neon

Change-Id: I8beda6ce11ad8ce9e80cc0a38d40161938359162
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-02 09:24:54 -07:00
James Yu
81ad047ee5 VP8 for ARMv8 by using NEON intrinsics 06 
Add idct_dequant_full_2x_neon.c
- idct_dequant_full_2x_neon

==== Summary of apply VP8 decode patch series ====
Benchmark on Samsung Chromebook, Cortex-A15, 1.7GHz, Dual core
Toolchain: linaro-1.13.1-4.8-2014.01
Compile argument: CROSS=arm-linux-gnueabihf- ../libvpx/configure
                     --target=armv7-linux-gcc --prefix=$HOME/out
                     --enable-shared --cpu=cortex-a7
Test argument: vpxdec --summary --noblit ./tears_of_steel_1080p.webm

NEON assembly   46.68 (fps)
Apply patch 06  46.65, -0.03
Apply patch 07  46.86, +0.21
Apply patch 08  46.58, -0.28
Apply patch 09  46.57, -0.01
Apply patch 10  46.51, -0.06
Apply patch 11  46.13, -0.38
Apply patch 12  45.42, -0.71
Apply patch 13  46.06, +0.64
Apply patch 14  45.19, -0.87
Apply patch 15  45.93, +0.74
Apply patch 16  45.48, -0.45
Apply patch 17  45.84, +0.36
Apply patch 18  45.91, +0.07  <= With all NEON intrinsics patches
                 Total -0.77 fps, 1.65% performance regression

Change-Id: I77bfc9eaccfb97b8d401e949ceff8795e26ca6b7
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-05-02 11:57:47 +08:00
Yunqing Wang
096eaba728 Remove VP8 save_reg_neon function 
This patch did a cleanup following the commit "Save NEON registers
in VP8 NEON functions". The pushing/poping of callee-saved NEON
registers was moved into individual NEON functions. Therefore,
we don't need to save those registers at the beginning of codec.
The related code was removed.

Change-Id: I5648166514fc9beffb780aa138495597731f49ea
 2014-04-29 16:13:24 -07:00
Yunqing Wang
33df6d1fc1 Save NEON registers in VP8 NEON functions 
The recent compiler can generate optimized code that uses NEON registers
for various operations besides floating-point operations. Therefore,
only saving callee-saved registers d8 - d15 at the beginning of the
encoder/decoder is not enough anymore. This patch added register saving
code in VP8 NEON functions that use those registers.

Change-Id: Ie9e44f5188cf410990c8aaaac68faceee9dffd31
 2014-04-28 14:51:53 -07:00
James Yu
fb5d281bb6 VP8 for ARMv8 by using NEON intrinsics 05 
Add dequantizeb_neon.c
- vp8_dequantize_b_loop_neon

vpxdec  --summary --noblit ../videos/tears_of_steel_1080p.webm
Before => After, 13.25 => 13.23 (fps)

Change-Id: Iebe3b0c6ed2359c778b0570763c5681ae25fef0c
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-02-26 10:16:00 +08:00
James Yu
28b2f82f97 VP8 for ARMv8 by using NEON intrinsics 04 
Add dequant_idct_neon.c
- vp8_dequant_idct_add_neon

vpxdec  --summary --noblit ../videos/tears_of_steel_1080p.webm
Before => After, 13.25 => 13.22 (fps)

Change-Id: Id48f39e1da58dd3d8d37658e94989411997f4f7c
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-02-26 09:59:23 +08:00
James Yu
d749ab6221 VP8 for ARMv8 by using NEON intrinsics 03 
Add dc_only_idct_add_neon.c
- vp8_dc_only_idct_add_neon

vpxdec  --summary --noblit ../videos/tears_of_steel_1080p.webm
Before => After, 13.25 => 13.24 (fps)

Change-Id: I5e9e277ec3a3ca67e13c8cc4c324a6fbe8a897fc
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-02-26 09:28:29 +08:00
James Yu
300a3bfc73 VP8 for ARMv8 by using NEON intrinsics 02 
Add copymem_neon.c
- vp8_copy_mem16x16_neon
- vp8_copy_mem8x8_neon
- vp8_copy_mem8x4_neon

vpxdec  --summary --noblit ../videos/tears_of_steel_1080p.webm
Before => After, 13.25 => 13.25 (fps)

Change-Id: Ib956b5a20522ff57dc8a580bf0aef7b252bddba6
Signed-off-by: James Yu <james.yu@linaro.org>
 2014-02-23 22:56:53 +08:00
Andrew Russell
549c31f8ae minor spelling cleanup in comments 
Change-Id: Ia91c6c406273345b08505097ffe1af3896980f06
 2014-02-12 16:32:51 -08:00