Commit Graph

479 Commits

Author SHA1 Message Date
Yunqing Wang
a23ccf8f8c Merge "Restructure multi-threaded decoder" 2010-09-21 05:00:30 -07:00
John Koleszar
99c611fea6 Merge remote branch 'internal/upstream' into HEAD 2010-09-21 00:05:03 -04:00
Fritz Koenig
b7dc9398f2 Use movq instead of movdqu.
Movdqu is more expensive (throughput, uops) than movq.  Minimal
impact for newer big cores, but ~2.25% gain on Atom.

Change-Id: I62c80bb1cc01d8a91c350c4c7719462809a4ef7f
2010-09-20 11:34:26 -07:00
Fritz Koenig
1c906448cc Merge "Better choice of instruction filter mask comparision." 2010-09-20 11:01:51 -07:00
Johann
6cf2b4aa0e Merge "reorder data to use wider instructions" 2010-09-20 10:47:33 -07:00
Johann
9c9afbab85 Merge "Update NEON wide idcts" 2010-09-20 10:47:22 -07:00
Fritz Koenig
8eae7fe7e8 Better choice of instruction filter mask comparision.
Use pmaxub instead of a combination of psubusb/por to
determine if any comparisons go over the limit.

Change-Id: I3f0bd7d2aabe5fee9ba6620508e2b60605abcb82
2010-09-20 10:20:38 -07:00
Guillermo Ballester Valor
236906863a Add high limit check for unsigned parameters
The patch related with issue #55 (5a72620) fixed some warnings, but the
fix was not optimal. It actually was a trick to confuse compiler rather
than a fix.

This patch fixes it by creating a new macro used when needed just a high
limit check for an unsigned.

Change-Id: I94b322e0f7fb07604b3b1df1f9321185f48cfcb5
2010-09-20 10:03:05 -04:00
Johann
022323bf85 reorder data to use wider instructions
the previous commit laid the groundwork by doing two sets of idcts
together. this moved that further by grouping the interesting data
(q[0], q+16[0]) together to allow using wider instructions. also
managed to drop a few instructions by recognizing that the constant
for sinpi8sqrt2 could be downshifted all the time which avoided a
dowshift as well as workarounds for a function which only accepted
signed data

looks like a modest gain for performance: at qcif, went from ~180
fps to ~183
Change-Id: I842673f3080b8239e026cc9b50346dbccbab4adf
2010-09-17 16:47:39 -04:00
Yunqing Wang
f857a85088 Restructure multi-threaded decoder
On each MB, loopfiltering is done right after MB decoding. This
combines two loops in multi-threaded code into one, which reduces
number of synchronizations to half.

The above-row/left-col data are saved in temp buffers for
next-row/next MB decoding.

Tests on 4-core gLucid machine showed 10% decoder performance
gain with threads=4 (tulip clip). Testing on other platforms
isn't done yet.

Change-Id: Id18ea7c1e84965dabea65d4c01ca5bc056ddeac9
2010-09-17 09:56:05 -04:00
John Koleszar
14b322e466 Merge remote branch 'internal/upstream' into HEAD 2010-09-17 00:05:04 -04:00
John Koleszar
9100073e8d cleanup: remove unused xprintf
These files aren't currently used, and we can get them back if we
need them.

Change-Id: I62aa3bff828e491a80c80eeb84a7c44903df29b5
2010-09-16 13:14:12 -04:00
John Koleszar
147b125b15 Reduce size of tokenizer tables
This patch reduces the size of the global tables maintained by the
tokenizer to 16k from 80k-96k. See issue #177.

Change-Id: If0275d5f28389af11ac83c5d929d1157cde90fbe
2010-09-16 10:00:04 -04:00
John Koleszar
4e31cbbe58 Merge remote branch 'internal/upstream' into HEAD 2010-09-16 00:05:08 -04:00
Fritz Koenig
746439ef6c Modify GET_GOT macro for performance.
GET_GOT was producing a zero length call.  This resulted in
pipeline flushes occuring when returing from the assembly
functions.  Masked on out of order cores, but evident on
Atom cores.

Change-Id: I8c375af313e8a169c77adbaf956693c0cfeb5ccd
2010-09-15 12:41:15 -07:00
John Koleszar
b1879d9754 Merge remote branch 'internal/upstream' into HEAD 2010-09-15 00:05:04 -04:00
John Koleszar
fe46476e98 Merge remote branch 'internal/upstream' into HEAD 2010-09-14 00:05:04 -04:00
Fritz Koenig
769f2424cc Removed unnecessary pxor.
There is no need to make sure that the lower byte of the
register is 0 because the downshift by 11 overwrites that byte.

Change-Id: I89cbf004b2ff532a2c68e0dc399c45a49cdad5a1
2010-09-13 18:34:34 -07:00
Fritz Koenig
71a1c19754 Merge "Make block access to frame buffer sequential" 2010-09-13 11:04:22 -07:00
John Koleszar
eeca6b786a Remove legacy release.sh script
This script is part of a legacy release process and is unsupported. Most
of this functionality has been moved into 'make dist.'

Change-Id: Id67936302083352b628869e2988876cf56558ca5
2010-09-13 09:46:51 -04:00
John Koleszar
887d6ef49a configure: support for ppc32-linux-gcc
Fixes issue 89. Thanks to josejx for the patch.

Change-Id: I7e664fed703b49f2fb3af4c5e6ce1173742000c2
2010-09-13 09:04:55 -04:00
John Koleszar
7f1a908b97 cosmetics: expand tabs in configure
Change-Id: I88ddb0afb56ef2be8184b56fe125ad938ead7a84
2010-09-13 09:02:18 -04:00
Fritz Koenig
a65cd3def0 Make block access to frame buffer sequential
Sequentially accessing memory from a low address to a high
address should make it easier for the processor to predict
the cache.

Change-Id: I1921ce996bdd547144fe864fea6435f527f5842d
2010-09-10 16:27:28 -07:00
John Koleszar
42cb1d84f2 Merge remote branch 'internal/upstream' into HEAD 2010-09-10 00:05:04 -04:00
Scott LaVarnway
a32ded1d5f Merge "Improved subset block search" 2010-09-09 11:51:29 -07:00
Scott LaVarnway
c5fb0eb8d9 Improved subset block search
Improved the subset block search and fill.  (about 3% improvement for
32 bit)  Modified/merged the code in order to create
vp8_read_mb_modes_mv which can decode the modes/mvs on a macroblock
level. This will allow the decode loop (in the future) to decode
modes/mvs on a frame, row, or mb level.

Change-Id: If637d994b508792f846d39b5d44a7bf9aa5cddf3
2010-09-09 14:42:48 -04:00
Johann
14ba764219 Update NEON wide idcts
Expand 93c32a55 which used SSE2 instructions to do two
idct/dequant/recons at a time to NEON. Initial working
commit. More work needs to be put into rearranging and
interlacing the data to take advantage of quadword
operations, which is when we'll hopefully see a much
better boost

Change-Id: I86d59d96f15e0d0f9710253e2c098ac2ff2865d1
2010-09-09 14:08:12 -04:00
John Koleszar
edcbb1c199 Fix GF interval for non-lagged ARFs
When ARFs are enabled in non-lagged compress modes, the GF interval
was being reset to zero. Non-lagged ARF updates were enabled in commit
63ccfbd, but this incorrect GF interval caused a quality regression.

Change-Id: I615c3b493f4ce2127044f4e68d0bcb07d6b730c3
2010-09-09 13:18:54 -04:00
Fritz Koenig
6d90f867e4 Merge branch 'master' of git://review.webmproject.org/libvpx 2010-09-09 08:54:21 -07:00
John Koleszar
c2140b8af1 Use WebM in copyright notice for consistency
Changes 'The VP8 project' to 'The WebM project', for consistency
with other webmproject.org repositories.

Fixes issue #97.

Change-Id: I37c13ed5fbdb9d334ceef71c6350e9febed9bbba
2010-09-09 10:01:21 -04:00
John Koleszar
5d012fc26b Merge remote branch 'internal/upstream' into HEAD 2010-09-09 00:05:04 -04:00
Jim Bankoski
69ae8f475d Skip unnecessary search of identical frames
vp8_get_compressed_data() was defeating logic in
encode_frame_to_datarate() that determined the reference buffers to
search and forcing all frames to be eligible to search. In cases
where buffers have identical contents, this is unnecessary extra
work.

Change-Id: I9e667ac39128ae32dc455a3db4c62e3efce6f114
2010-09-08 11:31:34 -04:00
Jim Bankoski
63ccfbd545 Enable ARFs for non-lagged compress
ARFs were explicitly disabled except in lagged compress mode. New
ARF logic allows for the ARF buffer to hold an older golden frame,
which does not require lagged compress.

Change-Id: I1dff82b6f53e8311f1e0514b1794ae05919d5f79
2010-09-08 11:26:13 -04:00
Fritz Koenig
3fb37162a8 Bilinear subpixel optimizations for ssse3.
Used pmaddubsw for multiply and add of two filter taps
at once for 16x16 and 8x8 blocks.

Change-Id: Idccf2d6e094561624407b109fa7e80ba799355ea
2010-09-07 17:19:40 -07:00
John Koleszar
b6a41d66df Merge remote branch 'internal/upstream' into HEAD 2010-09-04 00:05:04 -04:00
Scott LaVarnway
0de458f6b9 Reduced the size of MB_MODE_INFO
Moved partition_bmi and partition_count out of MB_MODE_INFO and
placed into MACROBLOCK.  Also reduced the size of other members
of the MB_MODE_INFO struct.  For 1080p, the memory was reduced
by 1,209,516 bytes.  The decoder performance appeared to improve
by 3% for the clip used.
Note:  The main goal for this change is to improve the decoder
performance.  The encoder will be revisited at a later date for
further structure cleanup.

Change-Id: I4733621292ee9cc3fffa4046cb3fd4d99bd14613
2010-09-03 16:43:23 -04:00
John Koleszar
773eff6267 Merge remote branch 'internal/upstream' into HEAD 2010-09-03 00:05:04 -04:00
John Koleszar
b0519a26b1 Update CHANGELOG for v0.9.2 release
Change-Id: I184e927987544e9f34f890249b589ea13a93a330
2010-09-02 14:56:47 -04:00
John Koleszar
e4b5002490 Update AUTHORS
Change-Id: I0395ffa107651a773fd11d12682ab9372f76a90b
2010-09-02 13:41:03 -04:00
John Koleszar
4496db45e3 Whitespace: nuke CRLFs
Change-Id: I8b9fdf9875a8fcff4cb49a3357ce44f18108c2e7
2010-09-02 13:33:01 -04:00
John Koleszar
daab4bcba6 Use native win32 timers on mingw
Changed to use QueryPerformanceCounter on Windows rather than only
when building with MSVC, so that MSVC can link libs built with
MinGW.

Fixes issue #149.

Change-Id: Ie2dc7edc8f4d096cf95ec5ffb1ab00f2d67b3e7d
2010-09-02 12:03:51 -04:00
John Koleszar
d6ee72a7bf Fix target detection on mingw32
gcc -dumpmachine returns only 'mingw32'

Change-Id: I774d05a97c5131fc12009e436712c319e54490a5
2010-09-02 11:52:39 -04:00
John Koleszar
21039ce16e Use -fno-common for mingw
Fixes http://code.google.com/p/webm/issues/detail?id=112

Thanks to Ramiro Polla for the issue/fix.

Change-Id: I7f7b547a4ea3270e183f59280510066cc29a619e
2010-09-02 11:52:38 -04:00
James Zern
76640f85da encoder: remove postproc dependency
Remove the dependency on postproc.c for the encoder in general, the only
unchecked need for it is when CONFIG_PSNR is enabled. All other cases
are already wrapped in CONFIG_POSTPROC. In the CONFIG_PSNR case the file
will still be included.

Additionally, when VP8_SET_POSTPROC is used with the encoder when post
processing has been disabled an error will be returned.

This addresses issue #153.

Change-Id: Ia6dfe20167f7077734a6058cbd1d794550346089
2010-09-02 11:52:37 -04:00
John Koleszar
7a3e0a1d93 Merge "added separate rounding/zbin constants for 2nd order" 2010-09-02 08:42:29 -07:00
John Koleszar
9398be0f46 Merge "Disable frame dropping by default" 2010-09-02 08:41:46 -07:00
Yaowu Xu
fca129203a added separate rounding/zbin constants for 2nd order
This allows experiments of using different rounding and
zerobin constants for 2nd order blocks.

Change-Id: Idd829adba3edd1f713c66151a8d29bb245e33a71
2010-09-02 10:27:03 -04:00
John Koleszar
23216211bc Disable frame dropping by default
This is not the behavior that most users expect.

Change-Id: I226126ea400c22cf1f7918e80ea7fe0771c569cb
2010-09-02 09:32:03 -04:00
John Koleszar
2675df2d05 Merge remote branch 'internal/upstream' into HEAD 2010-09-02 00:05:04 -04:00
Frank Galligan
d45e55015e Fix rare deadlock before loop filter
There was an extremely rare deadlock that happened when one thread
was waiting to start the loop filter on frame n while the other
threads were starting to work on frame n+1.

Change-Id: Icc94f728b3b6663405435640d9a2996735ba19ef
2010-09-01 22:01:21 -04:00