Commit graph

1801 commits

Author SHA1 Message Date
Ari Koivula 86219aa0fc Fix non-determinism with tiles
Earlier fix that fixed the supply side of the cu_array to take tile
coordinates into account should have been accompanied with this one
that does the same thing to demand side.
2016-03-03 17:39:20 +02:00
Arttu Ylä-Outinen 626b53ce85 Move sao search from encoderstate to sao.
Moves sao search from function encoder_state_worker_encode_lcu in
encoderstate.c to function kvz_sao_search_lcu in sao.c. Makes functions
kvz_init_sao_info, kvz_sao_search_chroma and kvz_sao_search_luma static
since they are no longer used outside sao.c.
2016-03-01 14:56:16 +02:00
Ari Koivula cfa722e448 Reduce parallelism for tiles
There is still some race-condition with encoding tiles from multiple
frames, so disable this to keep the bitstream deterministic.
2016-02-29 20:20:21 +02:00
Ari Koivula 3dcc0957f8 Deal with impossible mv constraints
If 0,0 vector is illegal, it's possible that no legal movement vector,
is found, in which case a large cost is returned instead. The cost
overflowed and there is all sorts of silliness with converting from
double to int, but I'm not going to fix all of it because when we
remove the doubles it will all get fixed.
2016-02-29 19:18:14 +02:00
Ari Koivula b1adf1576a Add --mv-constraint=frametilemargin
Add an even stricter motion vector constraint to prevent motion vectors
to fractional pixel positions that would need pixels outside the tile.
2016-02-29 19:18:14 +02:00
Ari Koivula f808cbf608 Allow increased parallelism for tiles
When movement vectors are constrained to tiles, only the same tile in
previous frame needs to be depended upon.
2016-02-29 14:33:06 +02:00
Ari Koivula f4ebff12b0 Combine tile mv constraint with OWF mv constraint
This also fixes movement vectors in tiles when OWF is on. The OWF mv
constraint assumed WPP, so it didn't work with tiles.
2016-02-29 14:33:06 +02:00
Ari Koivula 7981609cd0 Add --mv-constraint=frametile 2016-02-29 14:33:06 +02:00
Ari Koivula 9dbbb7fdbc Add --mv-constraint argument 2016-02-29 14:33:06 +02:00
Ari Koivula 1be877faf9 Fix chroma reconstruction with tiles
An incorrect frame boundary check caused a checksum error, because the
chroma reconstruction of the encoder was wrong. The encoder treated
horizontal tile boundaries as frame boundaries when the vertical
component of the movement vector was a multiple of 8.
2016-02-29 14:32:51 +02:00
Ari Koivula c0dc490dd1 Fix inter non-determinism with tiles
CU data was being copied to the wrong place in the reference frames
cu_array, which led to uninitialized data being used as a starting
point for motion vector search.

Fixes #99.
2016-02-26 17:05:04 +02:00
Ari Koivula 719d72925b Add loop-input option
This option is useful for testing long encodes, as you don't have to
find an actual infinite input.
2016-02-18 20:00:55 +02:00
Ari Koivula d23a5a15f1 Fix overflow in rate control
A 32 bit int overflowed after 2^31 bits (2Gb). It will still overflow
eventually, after 500 years of outputting 1Gb/s, but by that time,
I recon we will have fixed this properly and it's time to upgrade.
2016-02-18 16:48:21 +02:00
Ari Koivula eeafe14946 Clean up search initialization
Copy lcu explicitly instead of initializing with the same parameters.
2016-02-17 14:57:31 +02:00
Arttu Ylä-Outinen e5c84c361c Eliminate a race condition with input thread.
Changes communication between the input thread and main thread in
encmain.c so that only one of them uses img_in and retval at a time.
Fixes a race condition which would sometimes result in a deadlock.
2016-02-17 12:09:19 +02:00
Ari Koivula c40ede56ad Allow more frame parallelism in LP-gop
Add dependency to the reference frame instead of the previous frame,
in order to allow more frames to be encoded in parallel when temporal
stepping >1 in LP-gop (such as --gop=lp-g8d4r1t2).
2016-02-05 17:08:24 +02:00
Arttu Ylä-Outinen 40c7198f7d Add a script for updating README
Adds script tools/update_readme.sh for regenerating the "Using Kvazaar"
section of README.md from the output of "kvazaar --help".
2016-02-05 16:21:39 +02:00
Arttu Ylä-Outinen aac5373095 Fix typos in documentation
Fixes a few typos in README and command line help.
2016-02-05 16:21:27 +02:00
Ari Koivula a4915dc547 Update man and README 2016-02-04 14:16:58 +02:00
Ari Koivula e941e21cd6 Enable errors about non-existing CLI options
Set opterr and optind to their normal default values.
2016-02-04 13:48:58 +02:00
Ari Koivula 7a4bf94a52 Add --version and --help
Also don't print help by default, because it's too long. Print a
shorter usage message instead.
2016-02-04 13:48:48 +02:00
Ari Lemmetti 99e37ec235 Update old pixel type to the current one 2016-01-30 19:33:09 +02:00
Ari Koivula c76a0951cf Change version to 0.8.3 2016-01-28 21:21:02 +02:00
Ari Koivula cb2121b1aa Double time scale when field coding is used 2016-01-28 21:04:52 +02:00
Ari Koivula 8ad7d2a714 Move interlacing stuff to libkvazaaar API
This moves the interlacing from CLI code to api->encoder_encode, in
order to make it possible to use field coding through the lib API.

The field order is now determined per frame, as FFmpeg gives it per
frame and it's signaled per frame.

As a side effect, the CLI also now prints info from frames instead of
fields. While we might want to extend the API in the future to allow
printing of more detailed information about fields, for now it's
more important that the CLI uses the real lib API.

PSNR calculation for interlaced frames disabled until we have a way to
avoid deinterlacing the frame when it's not necessary.
2016-01-27 15:29:45 +02:00
Ari Koivula 6952f0fcc6 Refactor interlaced reading
Doesn't change the way it works. Just rearranges things so it's easier
to see what is going on.
2016-01-26 13:42:41 +02:00
Ari Koivula a46351efe1 Fix out of bounds error in interlacing
When field height was padded to a multiple of 8, yuv_io_extract_field
would read outside the buffer.
2016-01-26 13:41:52 +02:00
Arttu Ylä-Outinen 49677810b5 Rename config module to cfg.
Prevents a conflict with config.h and src/config.h so that the config.h
generated by configure is included in global.h. Fixes problems with
large input files on 32-bit systems.
2016-01-25 12:26:46 +02:00
Marko Viitanen 8e6c12b859 Merge branch 'input_reading_thread' 2016-01-25 12:00:03 +02:00
Marko Viitanen b4a4ce848c Use field parity for extracting correct fields from the interlaced picture 2016-01-25 10:58:12 +02:00
Marko Viitanen 441ce7728f Fix for input_read_thread() in the case when interlaced source-scan-type is used 2016-01-25 10:57:51 +02:00
Marko Viitanen 198204a20a Fix when using --source-scan-type=bff, offset was used for output lines 2016-01-25 10:13:51 +02:00
Ari Koivula 22b8ed43dc Remove global.h include from kvazaar.h
It shouldn't have been put there as it's the lib interface.
2016-01-22 15:23:34 +02:00
Ari Koivula 249c88011e Fix problem with >2GB input files on 32bit 2016-01-22 15:15:02 +02:00
Ari Koivula fa1af14637 Fix includes to include global.h first everywhere 2016-01-22 15:07:49 +02:00
Ari Koivula 3bf278529c Fix interlacing when using lib interface
Some flags used for interlacing were set in CLI interface, which
meant that interlacing didn't work correctly when used through
libkvazaar.
2016-01-22 14:35:20 +02:00
Marko Viitanen 0128ee26e7 Clear img_in pointer after reading it 2016-01-22 14:29:35 +02:00
Marko Viitanen b5459c1f23 Fixed performance monitoring by adding KVZ_ prefix to GET_TIME 2016-01-22 11:27:25 +02:00
Marko Viitanen e36237335e Fixed memory leaks caused by the input handler thread and cleaned up the code 2016-01-22 11:27:25 +02:00
Marko Viitanen ad9a1f6539 Input thread implementation
- Handle input processing in a separate thread to allow main thread more time with thread handling etc
  - Significant speedup can be seen when run on ultrafast settings and on a system with great number of cores
2016-01-22 11:27:25 +02:00
Ari Koivula 5e734593c0 Add psnr argument to CLI
To disable calculation of PSNR for frames, printing 0.0dB instead.
2016-01-21 15:08:34 +02:00
Ari Koivula 9eba3a83cc Add compiler flag checking to configure 2016-01-20 16:32:34 +00:00
Arttu Ylä-Outinen d452709795 Fix compiling AVX2 strategies.
Option -mavx2 was omitted when compiling AVX2 strategies. This commit
moves strategies to convenience libraries so that their compilation
flags can be easily set and adds -mavx2 to CFLAGS of the AVX2 library.
2016-01-20 11:04:12 +02:00
Ari Koivula 8060e2f6ec Delete kvazaar_version.h
It's not used anymore.
2016-01-19 20:40:35 +02:00
Ari Lemmetti 44656aeb19 Remove useless calculation 2016-01-19 16:35:16 +02:00
Marko Viitanen e822c16659 Removed unneeded cpu flags causing compiling to fail on powerpc, closes #121 2016-01-18 08:55:32 +02:00
Ari Koivula c8c0b4e8e8 Change version number for v0.8.2 2016-01-15 19:42:07 +02:00
Ari Koivula e2402c0000 Remove kva_api_get versioning.
We have soname versioning now, so we should focus on getting that right
instead. This also serves as an example of correctly incrementing the
lib-version.
2016-01-15 19:39:24 +02:00
Ari Koivula caf809f26d Remove scons build scripts
Because we are not going to maintain them.
2016-01-15 17:35:35 +02:00
Ari Koivula 15e1110997 Remove reference to Makefile-old
Makefile-old was deleted and this reference breaks make dist.
2016-01-15 17:32:54 +02:00
Ari Lemmetti a9decd2f40 Bump for yet another release 2016-01-14 23:23:11 +02:00
Ari Koivula 7718ac378f Add fractional FPS support.
Now that we put the timing info into the bitstream, the time base must
be precisely known. Represent framerate as a fraction and add timing
info only if the old floating point framerate was not used.

Deprecate cfg->framerate so it can be removed once we get patches to
FFmpeg and libav.

Add support for (num)/(denom) format to --input-fps.
2016-01-14 22:16:53 +02:00
Ari Lemmetti a9bd7b9e63 Bump version numbers for release v0.8.0 2016-01-14 20:38:28 +02:00
Ari Lemmetti b605e3866e Bye bye Makefile 2016-01-14 20:38:01 +02:00
Marko Viitanen 242edf98ad Added calculation and writing of VUI num_units_in_tick and time_scale 2016-01-14 15:32:33 +02:00
Ari Lemmetti daf39e348f Add dedicated handling for blitting NxN coeffs when N is 4, 8 or 16 2016-01-13 19:27:45 +02:00
Ari Lemmetti a2fc9920e6 Merge branch 'alternative-satd' 2016-01-13 15:00:43 +02:00
Ari Lemmetti 1ed34f2df8 Add some planar pred optimization for blocks larger than 8x8 2016-01-13 14:50:17 +02:00
Ari Lemmetti 0df88697ff Copy generic function to AVX2 strategy 2016-01-12 23:51:18 +02:00
Ari Lemmetti 62799a9fc3 Create generic strategy of planar prediction 2016-01-12 23:50:47 +02:00
Ari Lemmetti 3cb1cebfe5 Add missing inlines 2016-01-12 23:03:31 +02:00
Ari Lemmetti 6a0b13b8b6 Remove unused functions 2016-01-12 22:55:37 +02:00
Ari Lemmetti 61155f0edd Add 128-bit version of the functions as well 2016-01-12 22:52:00 +02:00
Ari Lemmetti a6afb8a8f4 Small refactoring 2016-01-12 22:29:33 +02:00
Ari Lemmetti a756f6133a Manually unroll vertical Hadamard transform 2016-01-12 21:45:02 +02:00
Ari Lemmetti 66350aa20e Experiment with alternative implementation of FWHT 2016-01-11 16:25:56 +02:00
Arttu Ylä-Outinen e14858f41a Fix build and tests.
- Remove non-existent file interface_main.c from library sources.
- Add file mv_cand_tests.c to test sources.
2015-12-21 16:03:55 +02:00
Arttu Ylä-Outinen 9abdee7cc3 Merge branch 'autotools' 2015-12-21 15:54:30 +02:00
Arttu Ylä-Outinen eb6fa3d980 Fix exporting functions in library.
Rewrites definition of macro KVZ_PUBLIC in kvazaar.h so that
KVZ_STATIC_LIB need not be defined when building a static library.
2015-12-21 14:38:59 +02:00
darealshinji 8427a85d36 Add tests 2015-12-19 08:24:35 +01:00
Ari Koivula 1270da3626 Move files under their modules in Visual Studio
Also moves CLI stuff under CLI project, so they are compiled as their
own lib just like when the Makefile is used.

The file interface_main.c was an artifact from a bygone era and should have
been deleted long ago.
2015-12-17 15:39:45 +02:00
Ari Koivula 947bae24f9 Update Doxygen documentation
Add module information to all header files.

Update all header file documentations to briefly say what they are, and
to use the javadoc format so the brief actually gets included into the
doxygen documentation.

Remove \file from implementation files, in order to not repeat the info
from the header files.

Add files under strategies and tools to Doxygen and update the Doxygen
settings to be just plain better.

Make README be the main page of Doxygen documentation.
2015-12-17 14:05:50 +02:00
Ari Koivula a6ea705e19 Add missing lambda to some bit costs
Bits were being added to rate distortion without being multiplied by
lambda in a few places. Fixing this bug also finally allows us to remove
the magic bits from the Coding Unit split decision.

I tried to find new optimum value for CU_COST and it turned out to be 2
for veryslow and 0 for superfast. The difference between 0 and 2 on
veryslow was only 0.1% however, so I don't think this parameter is
needed any longer. Before this fix the effect of removing CU_COST would
have been 0.8%.
2015-12-15 16:32:38 +02:00
Arttu Ylä-Outinen 0e33049d9e Enable full mv search once again.
- Updates function search_mv_full so that it compiles and handles
  non-square blocks.
- Enables compilation of search_mv_full.
- Sets full search radius to 32.
- Enables selecting full mv search with "--me full".
2015-12-15 12:26:26 +02:00
Arttu Ylä-Outinen dbb9b0df85 Enable search for AMP blocks. 2015-12-15 11:21:46 +02:00
Arttu Ylä-Outinen 7e4f4538a4 Implement encoding AMP part modes.
Also adds parameter --amp for enabling AMP blocks.
2015-12-15 11:21:45 +02:00
Arttu Ylä-Outinen c3716f7803 Add --smp option for enabling SMP blocks. 2015-12-15 11:21:45 +02:00
Arttu Ylä-Outinen 38b881c36f Implement search_frac for rectangular blocks.
Replaces parameter depth of function search_frac with parameters width
and height.
2015-12-15 11:21:45 +02:00
Arttu Ylä-Outinen 864c77f6eb Use kvz_satd_any_size in inter search.
Changes search_frac and kvz_search_cu_iter to use kvz_satd_any_size for
computing the SATDs instead of getting the SATD function with
kvz_pixels_get_satd_func.
2015-12-15 11:21:45 +02:00
Arttu Ylä-Outinen 056fa09ba5 Add arbitrary-sized SATD functions.
Adds strategy satd_any_size for generic and AVX2. The satd_any_size
functions are implemented with macro SATD_ANY_SIZE defined in
strategies-picture.h.
2015-12-15 11:21:45 +02:00
Arttu Ylä-Outinen 6bdc08b6eb Drop unused function declaration.
Removes declaration of non-existent function satd_8bit_8x8_generic in
strategyselector.h.
2015-12-15 11:21:44 +02:00
Arttu Ylä-Outinen 728a6abecc Extract macro SATD_NxN.
Combines definitions of macros SATD_NXN and SATD_NXN_AVX2 to macro
SATD_NxN and moves it to strategies-picture.h.
2015-12-15 11:21:44 +02:00
Arttu Ylä-Outinen 1eebfde0c5 Make tz search work with non-square blocks.
Replaces parameter depth with parameters width and height.
2015-12-15 11:21:44 +02:00
Arttu Ylä-Outinen e203883f3d Refactor kvz_filter_deblock_lcu.
Moves code for filtering the rightmost 4 pixels of an LCU to a separate
function filter_deblock_lcu_rightmost.
2015-12-15 11:21:44 +02:00
Arttu Ylä-Outinen 21ca74fe86 Replace deblock filter with a simple loop.
- Adds function is_pu_boundary.
- Moves code for filtering an edge of a single PU or TU to a new
  function filter_deblock_unit.
- Replaces recursive CU tree traversal in filter_deblock_cu with
  a simple loop and renames it to filter_deblock_lcu_inside.
2015-12-15 11:21:43 +02:00
Arttu Ylä-Outinen 7516fda970 Make fractional recon work with non-square blocks.
Adds parameter block_height to functions inter_recon_frac_luma,
inter_recon_14bit_frac_luma and inter_recon_14bit_frac_chroma so that
they can handle SMP blocks.
2015-12-15 11:21:43 +02:00
Arttu Ylä-Outinen 591a1ce6db Turn some inter recon functions static.
Makes the following functions static since they are not used outside
inter.c:
- kvz_inter_recon_frac_luma
- kvz_inter_recon_14bit_frac_luma
- kvz_inter_recon_frac_chroma
- kvz_inter_recon_14bit_frac_chroma
2015-12-15 11:21:43 +02:00
Arttu Ylä-Outinen 0f531362bf Enable Nx2N partitions. 2015-12-15 11:21:43 +02:00
Arttu Ylä-Outinen 4402e251ae Fix kvz_get_extended_block functions.
The buffers allocated in functions kvz_get_extended_block_avx2 and
kvz_get_extended_block_generic were too small when the width of the
block was less than its height. Fixed to allocate correctly sized
buffers.
2015-12-15 11:21:43 +02:00
Arttu Ylä-Outinen bdd8b1c0aa Implement 2NxN partitions in inter search.
- Try using 2NxN partitions after the usual 2Nx2N.
- Adds function kvz_search_cu_smp to search_inter module.
2015-12-15 11:21:42 +02:00
Arttu Ylä-Outinen 410064e880 Split lcu_set_inter into two functions.
Moves code for setting the inter modes for a single PU to a new function
lcu_set_inter_pu.
2015-12-15 11:21:42 +02:00
Arttu Ylä-Outinen 3236428e4d Make hexbs search work with non-square blocks.
Replaces parameter depth with parameters width and height.
2015-12-15 11:21:42 +02:00
Arttu Ylä-Outinen 31ba8d61c3 Implement fractional chroma recon for SMP blocks.
Adds parameter block_height to function kvz_inter_recon_frac_chroma.
2015-12-15 11:21:42 +02:00
Arttu Ylä-Outinen 0b6cef7be5 Remove unused function kvz_inter_set_block. 2015-12-15 11:21:42 +02:00
Arttu Ylä-Outinen e63486b23f Make lcu_set_inter work with SMP blocks. 2015-12-15 11:21:41 +02:00
Arttu Ylä-Outinen 7b99eb2970 Call recon functions correctly for SMP blocks.
Makes calls to kvz_inter_recon_lcu and kvz_inter_recon_lcu_bipred in
function search_cu work correctly when using SMP blocks.
2015-12-15 11:21:41 +02:00
Arttu Ylä-Outinen dc4525c0e3 Implement inter recon for non-square blocks.
Adds parameter height to functions kvz_inter_recon_lcu and
kvz_inter_recon_lcu_bipred and makes them work on non-square sizes.
Fractional reconstruction functions do not handle non-square blocks yet.
2015-12-15 11:21:41 +02:00
Arttu Ylä-Outinen f874c8614e Add part_mode binarization table comment. 2015-12-15 11:21:41 +02:00
Arttu Ylä-Outinen c77074a7ff Implement encoding SMP blocks. 2015-12-15 11:21:41 +02:00
Arttu Ylä-Outinen 98707a1288 Move encoding intra CU to a separate function.
Moves code for encoding a single intra coding unit from function
kvz_encode_coding_tree to a new function encode_intra_coding_unit.
2015-12-15 11:21:40 +02:00