John Koleszar [Tue, 21 Sep 2010 14:35:52 +0000 (10:35 -0400)]
Add getter functions for the interface data symbols
Having these symbols be available as functions rather than data is
occasionally more convenient. Implemented this way rather than a
get-codec-by-id style to avoid creating a link-time dependency
between the encoder and the decoder.
John Koleszar [Tue, 21 Sep 2010 15:54:36 +0000 (11:54 -0400)]
Don't reset mb clamping state during splitmv decoding
The MV decoding changes in c5fb0eb introduced a bug where the
macroblock clamping state was reset for each partition, so if an
earlier partition needed clamping but a subsequent one didn't,
the MB wouldn't receive clamping. Instead, the state is only
set during splitmv decoding, never cleared.
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.
Johann [Thu, 9 Sep 2010 19:55:19 +0000 (15:55 -0400)]
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
Yunqing Wang [Thu, 16 Sep 2010 18:08:52 +0000 (14:08 -0400)]
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.
Fritz Koenig [Tue, 14 Sep 2010 22:46:37 +0000 (15:46 -0700)]
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.
Scott LaVarnway [Thu, 9 Sep 2010 18:42:48 +0000 (14:42 -0400)]
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.
Johann [Tue, 7 Sep 2010 18:21:27 +0000 (14:21 -0400)]
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
John Koleszar [Thu, 9 Sep 2010 16:57:23 +0000 (12:57 -0400)]
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.
Jim Bankoski [Thu, 22 Jul 2010 20:07:13 +0000 (16:07 -0400)]
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.
Jim Bankoski [Thu, 22 Jul 2010 20:07:13 +0000 (16:07 -0400)]
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.
Scott LaVarnway [Thu, 2 Sep 2010 20:17:52 +0000 (16:17 -0400)]
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.
James Zern [Fri, 20 Aug 2010 20:06:56 +0000 (16:06 -0400)]
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.
Frank Galligan [Wed, 1 Sep 2010 20:40:18 +0000 (16:40 -0400)]
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.
The memory being zeroed in vp8_update_mode_info_border() was just
allocated with calloc, and so the entire function is actually
redundant, but it should be made correct in case someone expects
it to actually work in the future.
The timebase was being set to the value in the Y4M file on each
pass, but only doubled to account for the altref placement on
the first past.
This avoids reseting it on the second pass.
Fritz Koenig [Tue, 24 Aug 2010 23:27:49 +0000 (16:27 -0700)]
Allow --cpu= to work for x86.
--cpu was already implemented for most of our embedded
platforms, this just extends it to x86. Corner case for
Atom processor as it doesn't respond to the --march=
option under icc.
Johann [Tue, 24 Aug 2010 22:23:16 +0000 (18:23 -0400)]
clean up compiler warnings
did a test compile with clang and got rid of some warnings that have
been annoying me for a while:
vp8/decoder/detokenize.c: In function 'vp8_init_detokenizer':
vp8/decoder/detokenize.c:121: warning: assignment discards qualifiers from pointer target type
vp8/decoder/detokenize.c:122: warning: assignment discards qualifiers from pointer target type
vp8/decoder/detokenize.c:123: warning: assignment from incompatible pointer type
vp8/decoder/detokenize.c:124: warning: assignment discards qualifiers from pointer target type
vp8/decoder/detokenize.c:125: warning: assignment discards qualifiers from pointer target type
vp8/decoder/detokenize.c:128: warning: assignment discards qualifiers from pointer target type
vp8/decoder/detokenize.c:129: warning: assignment discards qualifiers from pointer target type
vp8/decoder/detokenize.c:130: warning: assignment discards qualifiers from pointer target type
vp8/decoder/detokenize.c:131: warning: assignment discards qualifiers from pointer target type
Fritz Koenig [Fri, 20 Aug 2010 17:58:19 +0000 (10:58 -0700)]
Rework idct calling structure.
Moving the eob structure allows for a non-struct based
function to handle decoding an entire mb of
idct/dequant/recon data. This allows for SIMD functions
to idct/dequant/recon multiple blocks at once.
John Koleszar [Fri, 20 Aug 2010 15:04:10 +0000 (11:04 -0400)]
increase rate control buffer level precision
The external API exposes the RC initial/optimal/full buffer level in
milliseconds, but this value was truncated internally to seconds. This
patch allows the use of the full precision during the conversion from
time to bits.
John Koleszar [Mon, 16 Aug 2010 13:34:30 +0000 (09:34 -0400)]
arm: fix missing dependency with --enable-shared
The C version of the dequant/idct/add function depends on the C
version of the IDCT, but this isn't compiled in on ARM. Since this
code has asm version, we can just remove this file to eliminate the
link error.
Johann [Thu, 12 Aug 2010 13:05:37 +0000 (09:05 -0400)]
framework for assembly version of the detokenizer
adds a compile time option: --enable-arm-asm-detok which pulls in
vp8/decoder/arm/detokenize.asm
currently about break even speed wise, but changes are pending to
the fill code (branch and load 3 bytes versus conditionally always
load one) and the error handling. Currently it doesn't handle zero
runs or overrunning the buffer.
this is really just so i don't have to rebase my changes all the
time to run benchmarks - now just need to replace one file!
Scott LaVarnway [Thu, 12 Aug 2010 20:25:43 +0000 (16:25 -0400)]
Removed unnecessary MB_MODE_INFO copies
These copies occurred for each macroblock in the encoder and decoder.
Thetemp MB_MODE_INFO mbmi was removed from MACROBLOCKD. As a result,
a large number compile errors had to be fixed.
John Koleszar [Mon, 9 Aug 2010 17:27:26 +0000 (13:27 -0400)]
avoid negative array subscript warnings
The mv_ref and sub_mv_ref token encodings are indexed from NEARESTMV
and LEFT4X4, respectively, rather than being zero-based like the
other token encodings.
Yaowu Xu [Wed, 11 Aug 2010 04:12:04 +0000 (21:12 -0700)]
Normalize quantizer's zero bin and rounding factors
This patch changes a few numbers in the two constant arrays
for quantizer's zerobin and rounding factors, in general to
make the sum of the two factors for any Q to be 128. While
it might be beneficial to calibrate the two arrays for best
quantizer performance, it is not the purpose of this patch.
Normalizing the two arrays will enable quick optimization
of the current faster quantizer, i.e .zerobin check can be
removed.
Replace the exponential search for optimal rounding during
quantization with a linear Viterbi trellis and enable it
by default when using --best.
Right now this operates on top of the output of the adaptive
zero-bin quantizer in vp8_regular_quantize_b() and gives a small
gain.
It can be tested as a replacement for that quantizer by
enabling the call to vp8_strict_quantize_b(), which uses
normal rounding and no zero bin offset.
Ultimately, the quantizer will have to become a function of lambda
in order to take advantage of activity masking, since there is
limited ability to change the quantization factor itself.
However, currently vp8_strict_quantize_b() plus the trellis
quantizer (which is lambda-dependent) loses to
vp8_regular_quantize_b() alone (which is not) on my test clip.
Patch Set 3:
Fix an issue related to the cost evaluation of successor
states when a coefficient is reduced to zero. With this
issue fixed, now the trellis search almost exactly matches
the exponential search.
Patch Set 2:
Overall, the goal of this patch set is to make "trellis"
search to produce encodings that match the exponential
search version. There are three main differences between
Patch Set 2 and 1:
a. Patch set 1 did not properly account for the scale of
2nd order error, so patch set 2 disable it all together
for 2nd blocks.
b. Patch set 1 was not consistent on when to enable the
the quantization optimization. Patch set 2 restore the
condition to be consistent.
c. Patch set 1 checks quantized level L-1, and L for any
input coefficient was quantized to L. Patch set 2 limits
the candidate coefficient to those that were rounded up
to L. It is worth noting here that a strategy to check
L and L+1 for coefficients that were truncated down to L
might work.
(a and b get trellis quant to basically match the exponential
search on all mid/low rate encodings on cif set, without
a, b, trellis quant can hurt the psnr by 0.2 to .3db at
200kbps for some cif clips)
(c gets trellis quant to match the exponential search
to match at Q0 encoding, without c, trellis quant can be
1.5 to 2db lower for encodings with fixed Q at 0 on most
derf cif clips)
Yunqing Wang [Thu, 29 Jul 2010 20:24:26 +0000 (16:24 -0400)]
First modification of multi-thread decoder
This is the first modification of VP8 multi-thread decoder, which uses
same threads to decode macroblocks and then do loopfiltering for each
frame.
Inspired by Rob Clark, synchronization was done on every 8 macroblocks
instead of every macroblock to reduce lock contention.
Comparing with the original code, this implementation gave about 15%-
20% performance gain while decoding my test clips on a Core2 Quad
platform (Linux).
John Koleszar [Mon, 9 Aug 2010 13:33:00 +0000 (09:33 -0400)]
Mark loopfilter C functions as static
Clang defaults to C99 mode, and inline works differently in C99.
(gcc, on the other hand, defaults to a special gnu-style inlining,
which uses different syntax.) Making the functions static makes sure
clang doesn't decide to discard a function because it's too large to
inline.
The regex which postprocesses the gcc make-deps (-M) output was too
greedy and matching in the dependencies part of the rule rather than
the target only. The patch provided with the issue was not correct, as
it tried to match the .o at the end of the line, which isn't correct
at least for my GCC version. This patch matches word characters
instead of .*
Thanks to raimue and the MacPorts community for isolating this issue.
Jan Kratochvil [Sat, 31 Jul 2010 15:12:31 +0000 (17:12 +0200)]
nasm: avoid space before the :data symbol type.
global label:data
^^
Provide nasm compatibility. No binary change by this patch with yasm
on {x86_64,i686}-fedora13-linux-gnu. Few longer opcodes with nasm on
{x86_64,i686}-fedora13-linux-gnu have been checked as safe.
Jan Kratochvil [Sat, 31 Jul 2010 15:12:31 +0000 (17:12 +0200)]
nasm: end labels with colon (':')
Labels should end by colon (':'), nasm requires it.
Provide nasm compatibility. No binary change by this patch with yasm
on {x86_64,i686}-fedora13-linux-gnu. Few longer opcodes with nasm on
{x86_64,i686}-fedora13-linux-gnu have been checked as safe.
Jan Kratochvil [Sat, 31 Jul 2010 15:12:31 +0000 (17:12 +0200)]
nasm: use OWORD vs DQWORD
nasm knows only OWORD. yasm knows both OWORD and DQWORD.
Provide nasm compatibility. No binary change by this patch with yasm on
{x86_64,i686}-fedora13-linux-gnu. Few longer opcodes with nasm on
{x86_64,i686}-fedora13-linux-gnu have been checked as safe.
projects / libvpx / log