In commit "3d50da5397d20abc932d81453b26cde758293a40", the stack
pointer was modified while aligning the stack, and it needed to
be pop out at the end.
Change-Id: I062971e195f1f2ab9d0ab5fb84dcf215a0fcaa67
There are many places in handle_inter_mode that need to restore the
dst buffer pointers, due to buffer pointer swap and early rd search
breakout. This commit wraps these operations into an inline function
for clean-up.
Change-Id: I0462e8c41c8bc3cd8db07395489cac03d8e5be54
This patch fixed issue 661: "Decoder produces mismatched outputs
with ssse3 enabled and disabled." In sub-pixel filters, a pixel
value was multiplied by a filter coefficient, and the results
were added up. The order of adding up these multiplications had to
be arranged carefully to prevent incorrect overflowing.
Change-Id: Id08af4200fea9e1b896fc40157b8651c2c7e80f2
Reversing bit order of partition_context_lookup, and modifying accordingly
update_partition_context() and partition_plane_context().
Change-Id: I64a11f1a94962a3bf217de2f50698cb781db71a5
This rebase is a better implementation of the previous ones.
Modifications are done to reduce the total clock cycle.
Speedup: 1.341
Compiled with -O3
Tested with: park_joy_420_720p50.y4m
Change-Id: I940eaf283f60597ca0d9d2e13d518878d55ff02d
Commit a4a5a210 enabled lossless coding, but the commit incorrectly
disabled the usage of skip in encoder even when skip should be used.
This commit make sure that skip is enabled even in lossless mode.
Change-Id: I276954f952c6ac68f17a316ebc72f09001228a08
Since they used in encoder only. This commit also re-order includes
for the files that include vp9_extend.h
Change-Id: I929fc113f2135d3198cd1fc6a17434e5a2f8a459
Explicitly constrain the upper limit of motion search range (in the
unit of full pixel) to be [-1023, +1023]. It is intended to control
the effective motion search range for 4K sequences.
Change-Id: I645539c70885eec0f155781f439d97d333336e88
This patch followed "Rewrite filter_selectively_horiz for parallel
loopfiltering" commit, and added x86 SSE2 optimization to do
16-pixel filtering in parallel. Also, corrected the declaration
of aligned arrays. For 8-pixel-in-parallel case, improved the
calculation of the masks and filters. Updated the threshold loading
since the thresholds were already duplicated. Updated neon C functions
to call neon loopfilters twice.
Using tulip clip, tests showed it gave a ~1.5% decoder speed gain.
Change-Id: Id02638626ac27a4b0e0b09d71792a24c0499bd35
Separate the rounding and right shift operations of forward transform
from those of inverse transform. Take out the assertion check from
inverse transforms. If the transform coefficients were constructed to
cause intermediate steps of inverse transform overflow, the codec will
just let it overflow without breaking the decoding flow.
Change-Id: I73cfc3706c4e840fc543a77cbc4cdb0b05d07730
on arm until we implenment real vp9_idct32x32_34_add_neon.
This issue is due to commit 47665452f0
Merge "Add 32x32 idct function for eob<=34 case".
Change-Id: I56b5f0abc20e7dd1bba521f78a995e85d65ea296
Simplifies the code by implementing band mapping with static arrays.
A lot of the code complexity introduced in a previous patch
disappears.
Change-Id: Ia3fac36e594fb5ad2d55ae141c58bba4c55c2d28
The switch to the rate-correction damping factor
in https://gerrit.chromium.org/gerrit/#/c/67536/ was not conditioned on CBR mode.
Change-Id: I2326704e8ac030a4f7b592dd3fedb94c7dd0644d
The step that sums three input samples could potentially cause the
intermediate result go beyond 16 bit limit, when operating as the
second 1-D transform. This commit fixes the issue.
Change-Id: Iaf512449ac2d25ddd8a806d760afab362c62a516
Overall change (using dual buffer scheme for superblocks of both inter
and intra modes) reduces speed 2 runtime:
bluesky_1080p at 6000kbps: 263553ms -> 257441ms
riverbed_1080p at 8000kbps: 233230ms -> 225308ms.
Change-Id: Idf8d70f768a4b0d97b2a8506372c57b7b4022119
As Jim suggested, 1D array was used to store filter levels instead
of 2D array. This used shift_y in setup_mask directly, and saved
few cycles.
Change-Id: If61ab298784861f1806b1cd396d4e4e2e0f097b9
Implements scan order to band map with arrays in both the encoder
and decoder to remove conditional statements.
Encoding seems to be about 1% faster at speed 0, tested on football.
Decoding seems to be about 0.5-1% faster on a set of 25 videos.
Change-Id: Idb233ca0b9e0efd790e30880642e8717e1c5c8dd
This commit enables the dual buffer rate-distortion optimization
and encoding scheme. It stacks the original transform coefficients,
quantized levels, and reconstructed coefficients, in the rate-
distortion optimization search process, hence eliminates the need
to re-run residual generation, forward transform, and quantization
in the encoding stage.
Change-Id: I011bfad3a59a380a869ee552e91dae0394ec492e
Added loop filter mask checking, and made the caller function
ready for implementation of parallel loopfiltering in horizontal
direction.
Next, we need to go through the loopfilter functions (both c and
optimized versions), and provide 16-byte wide loopfiltering for
each filter type.
Change-Id: Ifef47e7ef9086ebc2fd6ca7ede8f27c9bbf79e66
Allocate memory space of dual buffer sets that store the coeff, qcoeff,
dqcoeff, and eobs. Connect the pointers of macroblock_plane and
macroblockd_plane to the actual buffer in use accordingly.
Change-Id: I2f0b5f482ca879fae39095013eaf8901db20a5a4
Make the macroblockd_plane contain dynamic buffer pointers instead
static pointers to the memory space allocated therein. The decoder
uses the buffer allocated in pbi, while encoder will use a dual
buffer approach for rate-distortion optimization search.
Change-Id: Ie6f24be2dcda35df7c15b4014e5ccf236fb3f76c
We only used "ib" to call get_scan() function, which in turn calls
get_tx_type_4x4() function. The latter one only needs block index if
bsize < BLOCK_8X8 -- under that condition raster_block == block.
Change-Id: I697306a0c3cf937acdd4f5e623d4367c5acc0b2f
This commit fixes the assignment of mode_info pointer per tile. It
makes recognition of tiles in both row and column formats and properly
arrange the use of mode_info.
The bug was first introduced in
I6226456dd11f275fa991e4a7a930549da6675915
https://gerrit.chromium.org/gerrit/#/c/67492/
Change-Id: Ie12cd209f53241513728c461ee3d7b9599ddb860
Inlining set_contexts_on_border() into set_contexts(). The only difference
is the additional check that "has_eob != 0" in addition to
"xd->mb_to_right_edge < 0" and "xd->mb_to_right_edge < 0". If has_eob == 0
then memset does the right thing and works faster.
Change-Id: I5206f767d729f758b14c667592b7034df4837d0e
This patch continued the work done in "Rewrite loop_filter_info_n
struct"(commit:00dbd369c70270428d56da6d15ea5486fc821c52) to further
improve loopfilter function.
1. Instead of storing pointers to thresholds, store loopfilter
levels within 64x64 SB;
2. Since loopfilter levels are already calculated in setup_mask,
we don't need call build_lfi to look up them again. Just save
loopfilter levels in setup_mask.
3. Reorganized and simplified filter_block_plane().
Tests showed a ~0.8% decoder speedup.
Change-Id: I723c7779738bbc2afcb9afa2c6f78580ee6c3af7
This to make sure that prediction residue always get coded in lossless
mode.
This commit also fixed lossless unit test
Change-Id: I537726ee55328d4e4cf0a0196393a67e12bfcde1
The new expression is much more logical than previous one. Surprisingly
both expressions give exactly the same set of dependent values
-- have_top, have_left, have_right -- in vp9_predict_intra_block.
Change-Id: I63eb1b592b8c37883b3a0dbb1f3daa271e446109
This patch fixed the issue reported in "Issue 655: remove textrel's
from 32-bit vp9 encoder". The set of vp9_subpel_variance functions
that used x86inc.asm ABI didn't build correctly for 32bit PIC. The
fix was carefully done under the situation that there was not
enough registers.
After the change, we got
$ eu-findtextrel libvpx.so
eu-findtextrel: no text relocations reported in 'libvpx.so'
Change-Id: I1b176311dedaf48eaee0a1e777588043c97cea82
The term x represents macroblock pointer across encode_block. Change
the two local variable names to avoid confusion.
Change-Id: Ic732e73023525d673c0a678ed2708ac1edf5a3f9
Now tile decoding consists of two stages:
1. Find tile buffer start and its size, put this info into tile_buffers.
2. Decode each tile based on information from tile_buffers.
It seems that stage 1 can also be reused by multithreaded tile decoder.
Change-Id: If0cdaefdd6d10bb41c63561346c9ae4cfac081dd
It is more logical to use dqcoeff buffer to put there *dequantized*
transform coefficients (inside inverse_transform_block and
decode_coefs functions). Dequantization happens inside WRITE_COEF_CONTINUE
macro.
qcoeff buffer should be only used in the encoder for *quantized*
transform coefficients.
Change-Id: Ifd54bef272bbf5311ced6669c4f1079f998af5d7
SVC multiple layer per frame encoding is invoked with vpx_svc_init and
vpx_svc_encode. These interfaces are designed to be invoked from ffmpeg.
Additional improvements:
- make dummy frame handling a bit more explicit
- fixed bug with single layer encodes
- track individual frame sizes and psnrs instead of averages
- parameterized quantizer, 16th scalefactors, more logging,
- enabled single layer encodes to generate baseline
- include new mode for 3 layer I frame with 5 total layers
Change-Id: I46cfa600d102e208c6af8acd6132e0cc25cda8d4
Removing special case handling from vp9_tree_probs_from_distribution(),
tree_merge_probs(), and vp9_tokens_from_tree_offset() functions. Replacing
inter_mode_offset() function with macro INTER_OFFSET which is used now for
vp9_inter_mode_tree definition.
Change-Id: Iff75a1499d460beb949ece543389c8754deaf178
Removes stack-alocation of token_cache in decode_coefs function
Seems to achieve about 1% decode speed improvement as tested on
25 480p videos.
Change-Id: I8e7eb3361fa09d9654dfad0677a6d606701fdc6e
The compound inter prediction could potentially run with initial
motion vectors of invalid value and check the mv_cost, which triggers
overheap read. This commit resolves this issue by forcing a motion
vector value check for compound inter modes of both superblock and
sub8x8 block sizes.
Change-Id: I4f4fc19ce83c8272782bc382f12c82a3f03212fc
We only update partition_probs for inter frames but they are constant
for key frames. It is not necessary to have constants inside frame
context and copy them every time. This change reduces FRAME_CONTEXT size
by at least 48 bytes.
Change-Id: If70a53be51043f37fe7d113853217937710932a7
Removed:
goldfreq, avg_encode_time, avg_pick_mode_time,
cpu_freq, interquantizer
member variables from VP9_COMP since they are no longer
used in the code.
Change-Id: I010a82c217d0da03c3f53d1858d3462190c12dcf
Removed three members from the VP9_COMP data structure:
inter_zz_count, gf_bad_count, gf_update_recommended.
These were part of the VP8 real-time mode implementation
that was removed from the initial VP9 codecbase.
Change-Id: I866b083b88ef02c74837277d50ce532ca88492f3
This commit fixes the use case of plane_block_idx, which determines
the plane (Y/U/V) index based on block index. When block idx >= 4 in
sub8x8 block loop, it should be of chroma components.
Change-Id: I072705aa7b35445524ac607089ca8ce54b7ba478
We don't have to calculate 'new' probability in convert_distribution()
because it is enough to calculate only 'new' counters which could be used
to calculate probability if necessary. That's why removing a lot of unused
temporary probability arrays and reducing number of get_binary_prob()
calls.
Change-Id: I4e14eb7203d1ace61bbddefd6b9b6326be83ba63
1. Reduced the size memset based on eob for 32x32 transform. The reset
of non-zero coefficient should probably go into where they are read in
inverse transform functions. (TODO)
2. Removed a redundant level of indirection.
vp9_iht4x4_add() checks transform type and call vp9_iht4x4_16_add()
for tranforms other than DCT_DCT. In this case, the DCT_DCT case
has been already handled here.
Change-Id: Iacbc77da761f0b308df5acea0f20c9add9f33d20
The change doesn't affect the bitstream. It changes the order or function
calls and affects how we reconstruct intra- and inter-blocks. Speed up is
about 1...1.5%.
For intra-blocks:
Before:
for each transform block read tokens
for each transform block do prediction
for each transform block do inverse transform
Now:
for each transform block
read tokens
do prediction
do inverse transform
For inter-blocks:
Before:
for each transform block read tokens
for each transform block do inverse transform
Now:
for each transform block
read tokens
do inverse transform
Change-Id: I12a79bf1aa5a18c351b8010369bd3ff1deae1570
This CL contains two AVX2 optimized loop filter functions,
mb_lpf_horizontal_edge_w_avx2_8 and mb_lpf_horizontal_edge_w_avx2_16.
Change-Id: I604e4fe6e99752b7800c2ea98721d97f7e0b931b
-Don't reduce maxQ for gold/alt in CBR mode.
-Fix to min/maxQ for first/initial key frame.
-Add more speeds to datarate test and reduce the starting bitrate for test.
Change-Id: Id2a333d76dd3f6a51b322ca984588e2a22159c58
This commit makes zcoeff_blk cache the case where the entire block
is quantized to be zero (without applying zero-forcing) in the rate-
distortion optimization loop, and skip the forward DCT, quantization,
inverse DCT, and reconstruction process in the encode_block stage.
It now works for all the block sizes, including sub8x8 blocks.
Change-Id: I5ae60a9c436ba3637d11666733554bec4580ef98
Both decode_modes_sb and decode_modes_b had conditions to immediately
return at the beginning. Eliminating these conditions here and calling
these functions only to do a real work. Also unrolling loop for
PARTITION_SPLIT.
Change-Id: I2fc41cb74ac491f045a2f04fe68d30ff4aaa555d
"<< SUBPEL_BITS" needs to be added in the calculation. Call
set_scaled_offsets() to calculate x_offset_q4 and y_offset_q4.
Change-Id: Ied130ea771510e918f51cd1dc3abe57f4c0962b5