(see Change I9b2ccc88: Makes all mode token tables const)
Further remove runtime table initialization and use
precalculated const data. Data footprint reduced
by 4112 bytes.
Change-Id: Ia3ae9fc19f77316b045cabff01f6e5f0876a86ab
Ensure that RTCD function pointers are set at most once, to silence
some data race warnings. Implementation provided for POSIX threads and
Win32, with the prior unsynchronized behavior left in place for other
platforms.
Change-Id: I65c5856df43ef67043b3d5f26ddafddd8fcb2f7e
We can get rid of all remaining global initializers now:
vp8_scale_machine_specific_config()
vp8_initialize()
vp8dx_initialize()
Change-Id: I2825cea5d1c01ad9f6c45df49a0f86d803bfeb69
Mode token tabels precalculated in entropymode.c.
Removes vp8_initialize_common()as all common global data
is precalculated const now.
Change-Id: I9b2ccc883e4f618069e1bc180dad3a823394eb73
Removes all runtime initialization of global data in entropy.c.
Precalculated values are used for initializing all entropy related
tabels.
First patch in a series to make sure code is reentrant.
Change-Id: I9aac91a2a26f96d73c6470d772a343df63bfe633
Local variable offsets are now consistent for the functions,
removed unused parameters, reworked the assembly to eliminate
stalls/instructions.
Change-Id: Iaa37668f8a9bb8754df435f6a51c3a08d547f879
This patch includes:
1. fixes to disable block based termporal mixing when motion
is detected (because this version of mfqe only handles zero motion).
2. The criterion used for determining whether to mix or
not are changed to use squared differences rather than
absolute differences.
3. Additional checks on color mismatch and excessive block
flatness added. If the block as decoded has very low activity
it is unlikely to yield benefits for mixing.
Change-Id: I07331e5ab5ba64844c56e84b1a4b7de823eac6cb
In cases where you have a flat background occluded by a moving object
of similar luminosity in the foreground, it was likely that the
foreground blocks would persist for a few frames after the background
is uncovered. This is particularly noticable when the object has a
different color than the background, so add the chroma planes in as an
additional check.
In addition, for block sizes of 8 and 16, the luma threshold is
applied on four subblocks independently, which helps when only part of
the background in the block has been uncovered.
This fixes issue #392, which includes a test clip to reproduce the
issue.
BUG=392
Change-Id: I2bd7b2b0e25e912dcac342e5ad6e8914f5afd302
Reduced the size of the struct by 8 bytes, which would be
a memory savings of 64800 bytes for 1080 resolutions. Had
an extra byte, so created an is_4x4 for B_PRED or SPLITMV
modes. This simplified the mode checks in
vp8_reset_mb_tokens_context and vp8_decode_mb_tokens.
Change-Id: Ibec27784139abdc34d4d01f73c09f43e9e10e0f5
Adds logic to disable mfqe for the first frame after a configuration
change such as change in resolution. Also adds some missing
if CONFIG_POSTPROC macro checks.
Change-Id: If29053dad50b676bd29189ab7f9fe250eb5d30b3
Break MFQE code into it's own file.
It is currently only valid for 16x16 and 8x8 Y blocks. It also filters
4x4 U/V blocks.
Refactor filtering and add associated assembly. Limited test cases show
--mfqe introduces a penalty of ~20% with HD content. The assembly
reduces the penalty to ~15%
Change-Id: I4b8de6b5cdff5413037de5b6c42f437033ee55bf
Reworked the code to use vp8_build_intra_predictors_mby_s,
vp8_intra_prediction_down_copy, and vp8_intra4x4_predict_d_c
functions instead. vp8_intra4x4_predict_d_c is a decoder-only
version of vp8_intra4x4_predict. Future commits will fix this
code duplication.
Change-Id: Ifb4507103b7c83f8b94a872345191c49240154f5
Reworked the code to use vp8_build_intra_predictors_mbuv_s
instead. This is WIP with the goal of eliminating all
functions in reconintra_mt.h
Change-Id: I61c4a132684544b24a38c4a90044597c6ec0dd52
On Android NDK, rand() is inlined function. But, on our SSE optimization,
we need symbol for rand()
Change-Id: I42ab00e3255208ba95d7f9b9a8a3605ff58da8e1
As an optimization some architectures use the max_sad argument to break
out early from the SAD. Pass in INT_MAX instead of 0 to prevent this.
Change-Id: I653c476834b97771578d63f231233d445388629d
MFQE postproc crashed with stream dimensions not a multiple of 16.
The buffer was memset unconditionally, so if the buffer allocation
fails we end up trying to write to NULL.
This patch traps an allocation failure with vpx_internal_error(),
and aligns the buffer dimensions to what vp8_yv12_alloc_frame_buffer()
expects.
Change-Id: I3915d597cd66886a24f4ef39752751ebe6425066
This is the final commit in the series converting to the new RTCD
system. It removes the encoder csystemdependent files and the remaining
global function pointers that didn't conform to the old RTCD system.
Change-Id: I9649706f1bb89f0cbf431ab0e3e7552d37be4d8e
This commit continues the process of converting to the new RTCD
system. It removes the last of the VP8_ENCODER_RTCD struct references.
Change-Id: I2a44f52d7cccf5177e1ca98a028ead570d045395
This is a proof of concept RTCD implementation to replace the current
system of nested includes, prototypes, INVOKE macros, etc. Currently
only the decoder specific functions are implemented in the new system.
Additional functions will be added in subsequent commits.
Overview:
RTCD "functions" are implemented as either a global function pointer
or a macro (when only one eligible specialization available).
Functions which have RTCD specializations are listed using a simple
DSL identifying the function's base name, its prototype, and the
architecture extensions that specializations are available for.
Advantages over the old system:
- No INVOKE macros. A call to an RTCD function looks like an ordinary
function call.
- No need to pass vtables around.
- If there is only one eligible function to call, the function is
called directly, rather than indirecting through a function pointer.
- Supports the notion of "required" extensions, so in combination with
the above, on x86_64 if the best function available is sse2 or lower
it will be called directly, since all x86_64 platforms implement
sse2.
- Elides all references to functions which will never be called, which
could reduce binary size. For example if sse2 is required and there
are both mmx and sse2 implementations of a certain function, the
code will have no link time references to the mmx code.
- Significantly easier to add a new function, just one file to edit.
Disadvantages:
- Requires global writable data (though this is not a new requirement)
- 1 new generated source file.
Change-Id: Iae6edab65315f79c168485c96872641c5aa09d55
Commit 892e23a5b introduced support for the VP8D_GET_LAST_REF_USED,
but missed the mapping of the control id to the underlying function,
so it was unavailable to applications.
In addition, the underlying function vp8_references_buffer() is
moved from common/postproc.c to decoder/onyxd_if.c as postproc.c is
not built in all configurations.
Change-Id: I426dd254e7e6c4c061b70d729b69a6c384ebbe44
Commit e06c242ba introduced a change to call vp8_find_near_mvs() only
once instead of once per reference frame by observing that the only
effect that the frame had was on the bias applied to the motion
vector. By keeping track of the sign_bias value, the mv to use could
be flip-flopped by multiplying its components by -1.
This behavior was subtley wrong in the case when clamping was applied
to the motion vectors found by vp8_find_near_mvs(). A motion vector
could be in-bounds with one sign bias, but out of bounds after
inverting the sign, or vice versa. The clamping must match that done
by the decoder.
This change modifies vp8_find_near_mvs() to remove the clamping from
that function. The vp8_pick_inter_mode() and vp8_rd_pick_inter_mode()
functions instead track the correctly clamped values for both bias
values, switching between them by simple assignment. The common
clamping and inversion code is in vp8_find_near_mvs_bias()
Change-Id: I17e1a348d1643497eca0be232e2fbe2acf8478e1
A processor with ARMv7 instructions does not
necessarily have NEON dsp extensions. This CL
has the added side effect of allowing the ability
to enable/disable the dsp extensions cleanly.
Change-Id: Ie1e879b8fe131885bc3d4138a0acc9ffe73a36df
Makes the thresholds for the multiframe quality enhancement module
depend on the difference between the base quantizers. Also modifies
the mixing function to weigh the current low quality frame less if
the difference in quantizer is large. With the above modifications
mfqe works well for both scalable patterns as well as low quality
key frames.
Change-Id: If24e94f63f3c292f939eea94f627e7ebfb27cb75
filter in a way such that when there is a single bad frame, the
post-processing is applied not only to just that frame but a
few subsequent frames as well.
Change-Id: Iba5d9896eed77244eb76b4a74692a93f8ecff634
When running multi-layer (ML) encodes and dynamically
changing coding parameters on the fly (e.g. frame
duration/rate, bandwidths allocated to each layer)
the encoder would not produce sensible output.
In certain cases the rate targeting would be
hideously inaccurate.
These fixes make it possible to change these coding
parameters correctly and to maintain accurate control
of the rate targeting.
I also added the specification of the input timebase
into the test program, vp8_scalable_patterns.c.
Patch 2: Moved declaration to appease MS compiler)
Change-Id: Ic8bb5a16daa924bb64974e740696e040d07ae363
with deblock or demacroblock filters. When --mfqe is used together
with --demacroblock or --deblock, mfqe is applied first and then
demacroblock/deblock is applied to the mfqe result.
Change-Id: Id83ee01f1b4a33a116f071dcf26d59c7f3497c32
precision used in initialization of roundoff is not a constant
updated to use #define MFQE_PRECISION 4
Change-Id: If2fc3d3d633d58a7f4ab34d258c232ec1e5f0a79
This patch removes the local copies of the dequantize
constants and implements John's idea as described
in "Make a local copy of the dequantized data" commit.
Change-Id: Ic6b7d681f00bf63263f71ff1e39ab2f80729e8b2
Adds a multiframe postprocessing module to enhance the quality of
certain frames that are coded at lower quality than preceding frames.
The module can be invoked from the commandline by use of the --mfqe
option, and will be most beneficial for enhancing the quality of
frames decoded using scalable patterns.
Uses the vp8_variance_var16x16 and vp8_variance_sad16x16 function
pointers to compute SAD and Variance of blocks.
Change-Id: Id73d2a6e3572d07f9f8e36bbce00a4fc5ffd8961
The code had a number of constructs like (condition)?1:0,
which is redundant with C's semantics. In the cases where a boolean
operator was used in the condition, simply remove the ternary part.
Otherwise adjust the surrounding expression to remove the condition
(eg, for rounding up. See pickinter.c and rdopt.c)
Change-Id: Icb2372defa3783cf31857d90b2630d06b2c7e1be
Multithreaded encoding was breaking at low bitrates
Please review/comment. Not sure if this is the best fix.
Change-Id: I87468c765372593fd865bc82e25121ebb8ca6af2
These functions are now used by the encoder.
This is WIP with the goal of creating a common idct/add for
the encoder and decoder. A boost of 1.8% was seen for
the HD rt test clip used.
[Tero] Added needed changes to ARM side.
Change-Id: Ibbb8000be09034203d7adffc457d3c3f8b06a5bf
The example encoder down-samples the input video frames a number of
times with a down-sampling factor, and then encodes and outputs
bitstreams with different resolutions.
Support arbitrary down-sampling factor, and down-sampling factor
can be different for each encoding level.
For example, the encoder can be tested as follows.
1. Configure with multi-resolution encoding enabled:
../libvpx/configure --target=x86-linux-gcc --disable-codecs
--enable-vp8 --enable-runtime_cpu_detect --enable-debug
--disable-install-docs --enable-error-concealment
--enable-multi-res-encoding
2. Run make
3. Encode:
If input video is 1280x720, run:
./vp8_multi_resolution_encoder 1280 720 input.yuv 1.ivf 2.ivf 3.ivf 1
(output: 1.ivf(1280x720); 2.ivf(640x360); 3.ivf(320x180).
The last parameter is set to 1/0 to show/not show PSNR.)
4. Decode:
./simple_decoder 1.ivf 1.yuv
./simple_decoder 2.ivf 2.yuv
./simple_decoder 3.ivf 3.yuv
5. View video:
mplayer 1.yuv -demuxer rawvideo -rawvideo w=1280:h=720 -loop 0 -fps 30
mplayer 2.yuv -demuxer rawvideo -rawvideo w=640:h=360 -loop 0 -fps 30
mplayer 3.yuv -demuxer rawvideo -rawvideo w=320:h=180 -loop 0 -fps 30
The encoding parameters can be modified in vp8_multi_resolution_encoder.c,
for example, target bitrate, frame rate...
Modified API. John helped a lot with that. Thanks!
Change-Id: I03be9a51167eddf94399f92d269599fb3f3d54f5
to the dqcoeff or qcoeff buffer. The encoder would
populate the dc coeffs of the y blocks as a separate
stage (recon_dcblock) and the decoder would use a special
version of the idct. This change eliminates the extra copy
and reduces the code footprint.
[Tero] Added needed changes to armv6 and NEON assembly.
Change-Id: I83202ffdbaf83f6e5dd69f4ba2519fcf0b13b3ba
Storing vp8_bilinear_filters_mmx in an mmx file and using it in an sse2
file is bad
Moving towards allowing --disable-mmx
Change-Id: I20493b35bdedcdcfc0915e6f05fdbe6c81a4a742
The calculated frame_rate is a state variable in the codec, and
shouldn't be maintained in the configuration struct. Move it to the
main part of cpi so that it isn't clobbered when the configuration
struct is updated. The initial framerate estimate is moved from the
vp8_cx_iface.c wrapper into the body of init_config() in onyx_if.c, so
that it is only called once and not reset on every call to
vp8_change_config().
Change-Id: I8d9a3d1283330d1ee297d07e9d78d1f2875f2465
Ronald recently sent me this patch that he did in April.
> From: Ronald S. Bultje <rbultje@google.com>
> Date: Thu, 28 Apr 2011 17:30:15 -0700
> Subject: [PATCH] SSE2 optimizations for
> vp8_build_intra_predictors_mby{,_s}().
HD decode tests have shown a performance boost up to 1.5%,
depending on material.
Patch set 3: Fixed encoder crash.
Change-Id: Ie1fd1fa3dc750eec1a7a20bfa2decc079dcf48c8
Call the idct/add after the tokenize. This is WIP with
the goal of creating a common idct/add for the encoder and
decoder. This move is necessary because the decoder's version
of the idct clobbers qcoeff, which is used by the tokenize.
Change-Id: I6b08d8e8397cd873647fa4fb9469884e3c876756
Added ARM optimized intra 4x4 prediction
- 2x faster on Profiler compared to C-code compiled with -O3
- Function interface changed a little to improve BLOCKD structure
access
Change-Id: I9bc2b723155943fe0cf03dd9ca5f1760f7a81f54
The referenced function (SignalObjectAndWait) isn't used. Reduces the
warnings with mingw32-w64 which defines this.
Change-Id: I4ce592879ec9372bf196dac640204c4d370bd210
Changed 'int eob' to 'char *eob' in BLOCKD so that both encoder and
decoder will use eobs[25] array from MACROBLOCKD structure. In future,
this will enable use of the decoder side IDCT in the encoder.
Change-Id: I6e1c011628cb8864fd4a0b80f0279ce16a5ca978
Adding support for several partitions within one input fragment.
This is necessary to fully support all possible packetization
combinations in the VP8 RTP profile. Several partitions can
be transmitted in the same packet, and they can only be split
by reading the partition lengths from the bitstream.
Change-Id: If7d7ea331cc78cb7efd74c4a976b720c9a655463
Tests showed ~1.2% performance boost on the HD clip used.
Performance will vary based on material.
Change-Id: Icbcf1a828750d5b4ae5252bf596b3ef594042e8a
The partial frame copy function used to copy an extra 8 lines above
and below. The partial frame filtering can only modify 3 pixel rows
above the partial frame. Reduce copy to bare minimum needed, which is
4 lines, so that partial filtering on copied frame is possible.
Define the "magic" fraction number for partial filtering in
loopfilter.h .
Change-Id: I4791ffc541b6884b12759a0d0714a8faf16147ec
check to make sure that cx_data buffer has enough room before
writting to it, prior behavior did not which could result in a crash.
Change-Id: I3fab6f2bc4a96d7c675ea81acd39ece121738b28
Instead of using the predict buffer, the decoder now writes
the predictor into the recon buffer. For blocks with eob=0,
unnecessary idcts can be eliminated. This gave a performance
boost of ~1.8% for the HD clips used.
Tero: Added needed changes to ARM side and scheduled some
assembly code to prevent interlocks.
Patch Set 6: Merged (I1bcdca7a95aacc3a181b9faa6b10e3a71ee24df3)
into this commit because of similarities in the idct
functions.
Patch Set 7: EC bug fix.
Change-Id: Ie31d90b5d3522e1108163f2ac491e455e3f955e6
Added the ability to create rate-targeted, temporally
scalable, VP8 compatible bitstreams.
The application vp8_scalable_patterns.c demonstrates how
to use this capability. Users can create output bitstreams
containing upto 5 temporally separable streams encoded
as a single VP8 bitstream.
(previously abandoned as:
I92d1483e887adb274d07ce9e567e4d0314881b0a)
Change-Id: I156250a3fe930be57c069d508c41b6a7a4ea8d6a
buffer_level in VP8_COMP and starting_buffer_level, optimal_buffer_level
and maximum_buffer_size in VP8_CONFIG changed from int to int64_t
to avoid potential crash issues for larger target bit rates.
Change-Id: I0d5ab6c8a44c2fef51f30cd8df4bb4b739c5df26
The data processed by the loopfilter overlaps. At the block level, this
results in some redundant transforms. Grouping the filtering allows for
a single 16x16 transpose (and inversion) instead of three 16x8 transposes
(and three more inversions).
This implementation is x86_64 only. We retain the previous
implementation for x86.
Improvements are obviously material dependant, but it seems to be ~%1 in
tests here.
Change-Id: I467b7ec3655be98fb5f1a94b5d145e5e5a660007
In the "Removed bmi copy to/from BLOCKD" commit, the copy
to the bmi in BLOCKD was eliminated. The clamp_mvs() used
the bmi in BLOCKD, which now contains incorrect values. This
patch fixes this problem.
Change-Id: I8eca1eaf4015052b0b63e90876f7ad321aba7cff
for SPLITMV and B_PRED modes. Modified code to use the bmi
found in mode_info_context instead of BLOCKD. On the decode
side, the uvmvs are calculated only when required, instead of
every macroblock. This is WIP. (bmi should eventually be
removed from BLOCKD)
Small performance gains noticed for RT encodes and decodes.(VGA)
Change-Id: I2ed7f0fd5ca733655df684aa82da575c77a973e7
Prepend idct function names with vp8_
so that under profiling they show up
associated with libvpx.
Change-Id: I4fe357b50236cb7730a4cc00164c0a3487a1d8b4
The data that the simple horizontal loopfilter reads is aligned, treat
it accordingly.
For the vertical, we only use the bottom 4 bytes, so don't read in 16
(and incur the penalty for unaligned access).
This shows a small improvement on older processors which have a
significant penalty for unaligned reads.
postproc_mmx.c is unused
Change-Id: I87b29bbc0c3b19ee1ca1de3c4f47332a53087b3d
Prepend . to local labels in assembly code. This
allows non unique labels within a file. Also
makes profiling information more informative
by keeping the function name with the loop name.
Change-Id: I7a983cb3a5ba2413d5dafd0a37936b268fb9e37f
When active map is specified and the current frame is not a key frame,
golden frame nor a altref frame then copy only those active regions.
This significantly reduces encoding time by as much as 19% on the test
system where realtime encoding is used. This is particularly useful
when the frame size is large (e.g. 2560x1600) and there's only a few
action macroblocks.
Change-Id: If394a813ec2df5a0201745d1348dbde4278f7ad4
This reverts commit b5ea2fbc2c. Further
testing showed noticable keyframe popping in some cases, reverting this
for now to give time for a proper fix.
Conflicts:
vp8/encoder/onyx_if.c
vp8/encoder/ratectrl.c
Change-Id: I159f53d1bf0e24c035754ab3ded8ccfd58fd04af
the neon code made several assumptions which were broken by a recent
change: https://review.webmproject.org/2676
update the code with new assumptions and guard them with a compile time
assert
Change-Id: I32a8378030759966068f34618d7b4b1b02e101a0
sharpness was not recalculated in vp8cx_pick_filter_level_fast
remove last_filter_type. all values are calculated, don't need to update
the lfi data when it changes.
always use cm->sharpness_level. the extra indirection was annoying.
don't track last frame_type or sharpness_level manually. frame type
only matters for motion search and sharpness_level is taken care of in
frame_init
move function declarations to their proper header
Change-Id: I7ef037bd4bf8cf5e37d2d36bd03b5e22a2ad91db
In sub-pixel motion search, the search range is small(+/- 3 pixels).
Preload whole search area from reference buffer into a 32-byte
aligned buffer. Then in search, load reference data from this buffer
instead. This keeps data in cache, and reduces the crossing cache-
line penalty. For tulip clip, tests on Intel Core2 Quad machine(linux)
showed encoder speed improvement:
3.4% at --rt --cpu-used =-4
2.8% at --rt --cpu-used =-3
2.3% at --rt --cpu-used =-2
2.2% at --rt --cpu-used =-1
Test on Atom notebook showed only 1.1% speed improvement(speed=-4).
Test on Xeon machine also showed less improvement, since unaligned
data access latency is greatly reduced in newer cores.
Next, I will apply similar idea to other 2 sub-pixel search functions
for encoding speed > 4.
Make this change exclusively for x86 platforms.
Change-Id: Ia7bb9f56169eac0f01009fe2b2f2ab5b61d2eb2f
