hashirama/uvg266
1
0
Fork 0
mirror of https://github.com/ultravideo/uvg266.git synced 2024-11-24 10:34:05 +00:00
Code Issues Actions Projects Releases Wiki Activity
3317 commits 3 branches 6 tags 9.2 MiB
Commit graph

2812 commits

Author SHA1 Message Date
Pauli Oikkonen c18adc5ee0 Redo sao_band_ddistortion_avx2 
Avoid branching and do the entire thing on 32 pixels at once in YMMs.
Also make the sao_bands function parameter const.
 2019-08-07 16:35:24 +03:00
Pauli Oikkonen 2827c3e3ab Make calc_sao_bands less opaque 2019-08-07 16:35:24 +03:00
Pauli Oikkonen 1bb9a079a8 Fix indentation 2019-08-07 16:35:24 +03:00
Reima Hyvönen 7bc959c7c5 3 sao functions are now working 2019-08-07 16:35:24 +03:00
Reima Hyvönen 0e0f2d3490 made to clear sum vector after it has been set to memory 2019-08-07 16:35:24 +03:00
Reima Hyvönen f146de7acb removed some variables to prevent memory losses 2019-08-07 16:35:24 +03:00
Reima Hyvönen 247c3a7a71 conversed gined to unsigned int 2019-08-07 16:35:24 +03:00
Reima Hyvönen ac5c216974 Some more memory error preventing to sao_edge_ddistortion_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen 3fb1cbca35 more editing sao_edge_ddistortion_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen afbb6fb960 some more modifications to sao_edge_ddistortion_avx2 to prevent memory failures 2019-08-07 16:35:24 +03:00
Reima Hyvönen 3496a57f7a Edited sao_edge_ddistortion_avx2 to avoid memory overflow 2019-08-07 16:35:24 +03:00
Reima Hyvönen 267ba1d6ce Modified sao_band_ddistortion_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen e70663b245 added some sub commands to avoid memory read errors 2019-08-07 16:35:24 +03:00
Reima Hyvönen 59dfb4570c Converted some loads to load int8_t instead ints 2019-08-07 16:35:24 +03:00
Reima Hyvönen 8b253209a8 Found false address load from calc_sao_edge_dir. Should now work like generic 2019-08-07 16:35:24 +03:00
Reima Hyvönen 50e0a47b7a Took away __restrict 2019-08-07 16:35:24 +03:00
Reima Hyvönen 8a39eb674e Removed c-variable from calc_sao_edge_dir_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen bc0a36830d Clerified some 6 pixel loads 2019-08-07 16:35:24 +03:00
Reima Hyvönen 1a8b211e05 Added break to line 170 2019-08-07 16:35:24 +03:00
Reima Hyvönen d05e750ebe Added some switches to prevent segmentation fault from reading 2019-08-07 16:35:24 +03:00
Reima Hyvönen 203580047d Defined some AVX functions 2019-08-07 16:35:24 +03:00
Reima Hyvönen c884c738b1 Updated some commands to match the standard 2019-08-07 16:35:24 +03:00
Reima Hyvönen b412ed2f59 Removed some setr and used loads calc_sao_edge_dir_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen c6cc063534 converted some hadd operations at calc_sao_edge_dir_avx2 to cast and extract 2019-08-07 16:35:24 +03:00
Reima Hyvönen 47ac109b10 optimated some sao_reconstruct_color_avx2 when sao->type == SAO_TYPE_BAND 2019-08-07 16:35:24 +03:00
Reima Hyvönen 96dc60a1ed first working optimation 2019-08-07 16:35:24 +03:00
Reima Hyvönen c148aff9fb Some optimation done to function sao_reconstruct_color_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen bf16ba6cc4 Remade sao_edge_ddistortion_avx2 and calc_sao_edge_dir_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen 79dc39a676 Some editing for sao_edge_ddistortion_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen 06ee52924e some reconst done to calc_sao_edge_dir_avx2 2019-08-07 16:35:24 +03:00
Reima Hyvönen 5fbc65d823 reconst optimation doesn't work yet 2019-08-07 16:35:24 +03:00
Reima Hyvönen d29f834a69 Remove useless function 2019-08-07 16:35:24 +03:00
Reima Hyvönen a232a12160 calc_sao_edge_dir_avx2 updated 2019-08-07 16:35:24 +03:00
Reima Hyvönen b1febc02a5 sao_edge_ddistortion_avx2 now working proberly 2019-08-07 16:35:24 +03:00
Reima Hyvönen cd6092a1ec Still too much bits, looking for where they appear 2019-08-07 16:35:24 +03:00
Reima Hyvönen 7853be8eeb Incomple optimation 2019-08-07 16:35:24 +03:00
Ari Lemmetti 40609aa865 Add missing headers to Makefile.am 2019-07-12 19:15:51 +03:00
Ari Lemmetti 5db3a78499 Bump versions for release 1.3 2019-07-09 22:09:32 +03:00
Ari Lemmetti d513ab1999 Add missing newline 2019-07-09 21:06:05 +03:00
Ari Lemmetti 4967072625 Do not bypass search on skip cu if early_skip is not enabled 2019-07-09 20:20:12 +03:00
Ari Lemmetti b20992a9f3 Rename functions more descriptive 2019-07-09 20:20:11 +03:00
Ari Lemmetti a348a0ec23 Fix transform depth in early skip 2019-07-09 20:05:48 +03:00
Pauli Oikkonen 8d48bee180 Tidy fast coeff cost code 2019-07-09 18:01:54 +03:00
Pauli Oikkonen 201a43b08e Clean up the RD-estimation code 2019-07-09 18:01:54 +03:00
Pauli Oikkonen b111df5073 Create preliminary version of improved cost estimator 2019-07-09 18:01:54 +03:00
Ari Lemmetti be08a87d94 Add missing parameter max-merge to the help message 2019-07-09 16:28:46 +03:00
Ari Lemmetti d0bb9b4a6d Add parameter max-merge to presets 2019-07-09 16:26:03 +03:00
Ari Lemmetti 4097331fd6 Early skip 2019-07-09 15:59:31 +03:00
Joose Sainio 977e885ea2 Fix issue with gop=0 introduced in 1c36f68d0c 2019-07-05 12:57:27 +03:00
Mikko Pitkänen a7f09c8114 Merge branch 'threadwrapper' 2019-06-24 16:54:59 +03:00
Mikko Pitkänen 3dd606ce2e Add new threadwrapper 2019-06-18 18:45:45 +03:00
Joose Sainio c94077d15e remove hardcoded value 2019-06-12 14:37:41 +03:00
Joose Sainio ac68c8444d remove negation that wasn't supposed to be there 2019-06-12 14:35:24 +03:00
Joose Sainio 5851dcc3be missing negation 2019-06-12 14:08:18 +03:00
Joose Sainio 1c36f68d0c Fix owf>=9 gop=8 and add test to catch such problem in future 2019-06-12 14:04:41 +03:00
Ari Lemmetti 933ff6ed55 Merge branch 'set-qp-in-cu-fix' 2019-06-07 09:01:03 +03:00
Ari Lemmetti c6da839002 Set lcu sqrt lambda according to lcu lambda instead of frame lambda when ROI is used 2019-05-29 18:32:10 +03:00
Ari Lemmetti 9339845e8b Set QP completely at CU level as the name '--set-qp-in-cu' implies 
-Move slice delta QP to CU level when using --set-qp-in-cu
-Separate functionality from roi
 2019-05-24 20:38:39 +03:00
Pauli Oikkonen 081d16fc33 Fix intrinsics that may be missing on some systems 
Create a header to collect all the workarounds for missing intrinsics
in one place
 2019-05-23 19:59:40 +03:00
Pauli Oikkonen 87a9208db8 Eliminate cvtsi64_si128 intrinsic 
Apparently it'll cause Win32 builds to break because it emits the movq
instruction or something..
 2019-04-17 16:30:40 +03:00
Pauli Oikkonen 7175d20bb2 Still include stdint.h for non-vector builds 2019-04-15 19:36:01 +03:00
Pauli Oikkonen 1315c7e2b0 Do not compile any vector code for non-SSE4/AVX2 builds 2019-04-15 19:10:48 +03:00
Pauli Oikkonen f5f70e7bc5 Merge branch 'sad-optimization' 2019-04-15 19:02:01 +03:00
Jan Beich 85f46e17a9 Detect AltiVec via elf_aux_info() on FreeBSD 12+ 2019-04-01 13:08:04 +00:00
Jan Beich 82486255da Simplify AltiVec detection on Linux 2019-04-01 13:08:04 +00:00
Pauli Oikkonen 6d43759604 Create a border-respecting 32-wide AVX hor_sad 2019-03-07 18:01:22 +02:00
Pauli Oikkonen f218cecb38 Remove offending hor_sad_avx2_w32 function 
Consider possibly creating a non-offending AVX2 version instead, the
way hor_sad_sse41_w32 works. Or maybe there's more essential work to
do.
 2019-03-05 22:51:41 +02:00
Pauli Oikkonen df2e6c54fd 4-unroll hor_sad_sse41_arbitrary 
This may not increase perf though because it's so rarely used
function, so keeping icache footprint may be more essential...
 2019-03-05 22:45:23 +02:00
Pauli Oikkonen 448eacba7b Avoid overreading block borders in hor_sad_sse41_arbitrary 2019-03-05 22:34:50 +02:00
Eemeli Kallio c159e275b7 Merge branch 'max_merge' 2019-03-05 14:39:03 +02:00
Pauli Oikkonen 41f51c08c4 Avoid overrunning buffer in hor_sad_sse41_w32 2019-03-01 15:37:38 +02:00
Pauli Oikkonen bcd9879359 Include quant coeff range check in non-scaling list execution path too 2019-02-27 17:26:44 +02:00
Pauli Oikkonen 24e6363f64 Remove the kvz_quant_avx2 wrapper function 2019-02-27 16:32:58 +02:00
Pauli Oikkonen 748820f3c5 Eliminate unnecessary loading of coeffs if scaling lists are off 2019-02-27 16:26:35 +02:00
Pauli Oikkonen 5994350f40 Allow quant_flat_avx2 to be used with scaling lists on 2019-02-27 16:25:59 +02:00
Eemeli Kallio 7f4e0acf41 Added check if max-merge is out of bounds 2019-02-19 13:53:42 +02:00
Pauli Oikkonen 9b0e079262 Use SSE instructions for 64-bit SADs instead of MMX 
VC++ seems to choke on MMX instructions
 2019-02-18 20:13:33 +02:00
Pauli Oikkonen d8b8923028 Add LGPL notices to reg_sad headers 2019-02-18 17:52:47 +02:00
Eemeli Kallio 2a40560888 some variables to const 2019-02-12 11:24:10 +02:00
Eemeli Kallio 8f8e7bb53c Added possibility to reduce number of maximum number of merge candidates. 2019-02-12 09:21:03 +02:00
Pauli Oikkonen 770db825b9 Create hor_sad_w8 and w4 epol mask the way w16 works 2019-02-06 19:34:26 +02:00
Pauli Oikkonen aa19bcac8a Avoid branching in creating shuffle mask in hor_sad_w16 2019-02-06 18:58:46 +02:00
Pauli Oikkonen 2d05ca8520 Remove width from constant-width hor_sad func params 
They should kinda know it already
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 57db234d95 Move 32-wide SSE4.1 hor_sad to picture-sse41.c 
It's not used by picture-avx2.c that also includes the header, so
it should not be in the header
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen dd7d989a39 Implement 32-wide hor_sad on AVX2 2019-02-04 20:41:40 +02:00
Pauli Oikkonen ff70c8a5ec Utilize horizontal SAD functions for SSE4.1 as well 2019-02-04 20:41:40 +02:00
Pauli Oikkonen f5ff4db01f 4-wide hor_sad border agnostic 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 35e7f9a700 Fix hor_sad w8 to work with both borders 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 836783dd6e Use hor_sad_w32 for both left and right borders 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 69687c8d24 Modify hor_sad_sse41_w16 to work over left and right borders 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 51c2abe99a Modify image_interpolated_sad to use kvz_hor_sad 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 1e0eb1af30 Add generic strategy for hor_sad'ing an non-split width block 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 686fb2c957 Unroll arbitrary-width SSE4.1 hor_sad by 4 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 768203a2de First version of arbitrary-width SSE4.1 hor_sad 2019-02-04 20:41:40 +02:00
Pauli Oikkonen ccf683b9b6 Start work on left and right border aware hor_sad 
Comes with 4, 8, 16 and 32 pixel wide implementations now, at some point
investigate if this can start to thrash icache
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 760bd0397d Pad the image buffer by 64 bytes from both ends 
This will be necessary for an efficient and straightforward
implementation of hor_sad for blocks over 16 pixels wide, because they
cannot use the shuffle trick because inter-lane shuffling is so hard to
do
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen c36482a11a Fix bug in 24-wide SAD 
*facepalm*
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen f781dc31f0 Create strategy for ver_sad 
Easy to vectorize
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen ca94ae9529 Handle extrapolated blocks with unmodified width using optimized_sad pointer 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 91b30c7064 Tidy up kvz_image_calc_sad 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 9db0a1bcda Create get_optimized_sad func for SSE4.1 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 91380729b1 Add generic get_optimized_sad implementation 
NOTE: To force generic SAD implementation on devices supporting
vectorized variants, you now have to override both get_optimized_sad
and reg_sad to generic (only overriding get_optimized_sad on AVX2
hardware would just run all SAD blocks through reg_sad_avx2). Let's
see if there's a more sensible way to do it, but it's not trivial.
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 45f36645a6 Move choosing of tailored SAD function higher up the calling chain 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 91cb0fbd45 Create strategy for directly obtaining pointer to constant-width SAD function 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 94035be342 Unify unrolling naming conventions 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 517a4338f6 Unroll SSE SAD for 8-wide blocks to process 4 lines at once 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 0f665b28f6 Unroll arbitrary width SSE4.1 SAD by 4 2019-02-04 20:41:40 +02:00
Pauli Oikkonen cbca3347b5 Unroll 64-wide AVX2 SAD by 2 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 84cf771dea Unroll 32 and 16 wide SAD vector implementations by 4 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 5df5c5f8a4 Cast all pointers to const types in vector SAD funcs 
Also tidy up the pointer arithmetic
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen a711ce3df5 Inline fixed width vectorized SAD functions 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 6504145cce Remove 16-pixel wide AVX2 SAD implementation 
At least on Skylake, it's noticeably slower than the very simple
version using SSE4.1
 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 4cb371184b Add SSE4.1 strategy for 24px wide SAD and an AVX2 strategy for 16 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 796568d9cc Add SSE4.1 strategies for SAD on widths 4 and 12 and AVX2 strategies for 32 and 64 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 4d45d828fa Use constant-width SSE4.1 SAD funcs for AVX2 2019-02-04 20:41:40 +02:00
Pauli Oikkonen 2eaa7bc9d2 Move SSE4.1 SAD functions to separate header 2019-02-04 20:41:40 +02:00
Pauli Oikkonen d2db0086e1 Create constant width SAD versions for 8 and 16 pixels 2019-02-04 20:41:40 +02:00
Pauli Oikkonen a13fc51003 Include a blank AVX2 strategy registration function even in non-AVX2 builds 2019-02-04 19:52:24 +02:00
Pauli Oikkonen d55414db66 Only build AVX2 coeff encoding when supported 
..whoops
 2019-02-04 19:34:30 +02:00
Pauli Oikkonen 3fe2f29456 Merge branch 'encode-coeffs-avx2' 2019-02-04 18:52:31 +02:00
Pauli Oikkonen 722b738888 Fix more naming issues 2019-02-04 16:05:43 +02:00
Pauli Oikkonen e26d98fb75 Rename a couple variables and add crucial comments 2019-02-04 15:57:07 +02:00
Pauli Oikkonen f186455619 Move encode_last_significant_xy out of strategy modules 
It's the exact same in both AVX2 and generic, and does not seem to
be worth even trying to vectorize
 2019-02-04 14:55:41 +02:00
Pauli Oikkonen 3f7340c932 Fine-tune pack_16x16b_to_16x2b 
Avoid mm_set1 operation when it's possible to create the constant with
one bit-shift operation from another instead. Thanks Intel for
3-operand instruction encoding!
 2019-02-04 14:44:47 +02:00
Pauli Oikkonen 314f5b0e1f Rename 16x2b cmpgt function, comment it better, optimize it slightly 
Eliminate an unnecessary bit masking to make it even more messy
 2019-02-04 14:44:32 +02:00
Pauli Oikkonen d8ff6a6459 Fix _andn_u32 to work on old Visual Studio 2019-02-01 15:34:42 +02:00
Pauli Oikkonen 26e1b2c783 Use (u)int32_t instead of (unsigned) int in reg_sad_sse41 2019-01-10 14:37:04 +02:00
Pauli Oikkonen 3a1f2eb752 Prefer SSE4.1 implementation of SAD over AVX2 
It seems that the 128-bit wide version consistently outperforms the
256-bit one
 2019-01-10 13:48:55 +02:00
Pauli Oikkonen 9b24d81c6a Use SSE instead of AVX for small widths 
Highly dubious if this will help performance at all
 2019-01-07 20:12:13 +02:00
Pauli Oikkonen b2176bf72a Optimize SSE4.1 version of SAD 
Make it use the same vblend trick as AVX2. Interestingly, on my test
setup this seems to be faster than the same code using 256-bit AVX
vectors.
 2019-01-07 19:40:57 +02:00
Pauli Oikkonen 887d7700a8 Modify AVX2 SAD to mask data by byte granularity in AVX registers 
Avoids using any SAD calculations narrower than 256 bits, and
simplifies the code. Also improves execution speed
 2019-01-07 18:53:15 +02:00
Pauli Oikkonen 7585f79a71 AVX2-ize SAD calculation 
Performance is no better than SSE though
 2019-01-07 16:26:24 +02:00
Pauli Oikkonen ab3dc58df6 Copy SAD SSE4.1 impl to AVX2 2019-01-03 18:31:57 +02:00
Pauli Oikkonen 45ac6e6d03 Tidy pack_16x16b_to_16x2b comments 2019-01-03 16:37:05 +02:00
Ari Lemmetti cd818db724 Add missing quantization and residual in cost calculation (inter rd=2). 2018-12-21 15:55:29 +02:00
Pauli Oikkonen 016eb014ad Move packing 16x16b -> 16x2b into separate function 2018-12-20 10:51:44 +02:00
Ari Lemmetti b234897e8a Fix smp and amp blocks in fme and revert previous change. 
Filter 8x8 (sub)blocks even with 8x4, 4x8, 16x4, 4x16 etc.
Calculate SATD on the 8x4, ... part
 2018-12-19 21:30:53 +02:00
Pauli Oikkonen 9aaa6f260d Fixes to enable portability 2018-12-18 20:42:09 +02:00
Pauli Oikkonen 2fdbbe9730 Move CG reordering code from quant-avx2 to shared header 2018-12-18 19:42:18 +02:00
Pauli Oikkonen d02207306d Create a header file for shared AVX2 code 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 361bf0c7db Precompute >=2 coeff encoding loop with 2-bit arithmetic 
Who needs 16x16b vectors when you can do practically the same with
16x2b pseudovectors in 32-bit general purpose registers!
 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 940b0e9e6a Require BMI2 for AVX2 build 
Any processor implementing AVX2 should also implement BMI2
 2018-12-18 19:41:09 +02:00
Pauli Oikkonen f66cb23d5b Optimize greater1 encoding loop 
Calculating the c1 variable need not be a serial operation!
 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 8c8b791c35 Vectorize kvz_context_get_sig_ctx_inc 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 033261eb74 Eliminate two branches using bit magic 2018-12-18 19:41:09 +02:00
Pauli Oikkonen c4434e8d04 Scan CG's in forward order to simplify finding last significant 2018-12-18 19:41:09 +02:00
Pauli Oikkonen efd097f5a5 Vectorize the coeff group loop to some extent 2018-12-18 19:41:09 +02:00
Pauli Oikkonen a01362e638 use the efficient method of reordering raster->scan 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 50a888e789 Use the efficient method to find first and last nz coeffs in block 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 7e9203f566 Scan coeff groups in scan order to help find last significant one 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 9a5a6fdbc7 Simplify two ifs in encode_coeff_nxn-avx2 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 37a2a8bac8 See if loop can be optimized by rearranging 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 584f2f74b6 Vectorize significant coeff group scanning loop 2018-12-18 19:41:09 +02:00
Pauli Oikkonen 1bfed73221 Add AVX2 strategy for encode_coding_tree 2018-12-18 19:41:09 +02:00
Pauli Oikkonen c3a6f3112a Add generic strategy group for encode_coding_tree 2018-12-18 19:41:09 +02:00
Marko Viitanen 1ef851ab4b Disable FME on amp/smp blocks with width or height not divisible by 8 2018-12-18 10:28:21 +02:00
Joose Sainio b71c5573f0 Merge branch 'rate_control_fix' 2018-12-17 12:39:27 +02:00
Sergei Trofimovich 68a70e45a1 x86 asm: mark stack as non-executable 
Gentoo's `scanelf` QA tool detects writable/executable stack
of assembly-writtent files as:

```
$ scanelf -qRa  .
 0644 LE !WX --- ---     ./src/strategies/x86_asm/.libs/picture-x86-asm-sad.o
 0644 LE !WX --- ---     ./src/strategies/x86_asm/.libs/picture-x86-asm-satd.o
 0644 LE !WX --- ---     ./src/strategies/x86_asm/picture-x86-asm-sad.o
 0644 LE !WX --- ---     ./src/strategies/x86_asm/picture-x86-asm-satd.o
```

Normally C compiler emits non-executable stack marking (or GNU assembler
via `-Wa,--noexecstack`).

The change adds non-executable stack marking for yasm-based assmbly files.

https://wiki.gentoo.org/wiki/Hardened/GNU_stack_quickstart has more details.

Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org>
 2018-12-16 11:31:56 +00:00
Reima Hyvönen 1fcc5c6a8d Merge branch 'bipred_recon' 2018-12-11 09:59:35 +02:00
Reima Hyvönen e4a10880f3 Added case 12 to bipred_recon no mov 2018-12-11 09:52:17 +02:00
Marko Viitanen a4f3968e52 Fix Visual Studio errors by initializing some variables used in AVX2 signhiding 2018-12-11 09:33:26 +02:00
Ari Lemmetti ac943147e3 Calculate satd cost for whole non-square blocks as well. 2018-12-10 17:04:29 +02:00
Pauli Oikkonen c465578048 Add a descriptive comment to coefficient reordering 2018-12-03 15:36:32 +02:00
Pauli Oikkonen f78bf2ebcb Optimize q_coefs usage for indexed fetch 2018-12-03 15:36:32 +02:00
Pauli Oikkonen d9591f1b49 Eliminate midway buffering of reordered coefs 
TODO: For some mysterious reason seems slightly slower than the
buffered one
 2018-12-03 15:36:32 +02:00
Pauli Oikkonen 7fe454c51f Optimize get_cheapest_alternative() 2018-12-03 15:36:32 +02:00
Pauli Oikkonen 6bbd3e5a44 Optimize rearrange_512 function 2018-12-03 15:36:32 +02:00
Pauli Oikkonen cb8209d1b3 Vectorize transform coefficient reordering loop 2018-12-03 15:36:32 +02:00
Pauli Oikkonen 7cf4c7ae5f Rename "reduce" functions to hsum 
That's what the functions fundamendally do anyway
 2018-12-03 15:36:32 +02:00
Pauli Oikkonen 316cd8a846 Fix ALIGNED keyword and grow alignment to 64B 2018-12-03 15:36:32 +02:00
Pauli Oikkonen 1befc69a4c Implement sign bit hiding in AVX2 2018-12-03 15:36:32 +02:00
Pauli Oikkonen c5cd03497e Require BMI and ABM instruction sets for AVX2 build 
AVX2 support on a processor should always imply BMI and ABM support.
The lzcnt and tzcnt instructions have more suitable semantics in the
corner case that source word is 0, and allow us to even handle that
scenario without a branch. Apparently Visual Studio will already
include this support when building with AVX2 enabled, so only the
automake files need to be tweaked.
 2018-12-03 15:36:32 +02:00
Reima Hyvönen f8696b54a4 Updated bipred_recon_avx2 in avx2/picture-avx2.c. Now it detects blocks that can be not equal to 8 (ie. width = 12) 2018-11-20 17:09:19 +02:00
Marko Viitanen a5a10a33c3 Enable --scaling-list parameter and add to the documentation 2018-11-19 10:47:30 +02:00
Reima Hyvönen 710ba288db Chroma has some problems 2018-11-15 16:42:48 +02:00
Sami Ahovainio 8f98d4aac7 Added square search 2018-11-14 14:50:31 +02:00
Marko Viitanen 6871490dd5 Simplify get_mvd_coding_cost(), only include golomb coding 2018-11-14 14:33:31 +02:00
Ari Lemmetti a832206bb6 Replace 32-bit incompatible instrinsics 2018-11-12 18:54:33 +02:00
Ari Lemmetti 5c774c4105 Rewrite most of FME and interpolation filters 
Changes had to break a lot of stuff and were just squashed into this horrible code dump
 2018-11-08 20:21:16 +02:00
Joose Sainio 1c8a1f24e2 Don't assume anything about bits spent 2018-11-07 16:03:38 +02:00
Joose Sainio 3471e2470d Fix using uninitialized value for the first frame 2018-11-07 08:17:39 +02:00
Joose Sainio d95ac11a3b Fix rate_control for other LP-GOPS 2018-11-06 14:20:44 +02:00
Joose Sainio 67a6ba667e Fix rate control for flat lp-gop 2018-11-06 09:38:17 +02:00
Reima Hyvönen 7406c33a42 Some more cleaning 2018-10-26 12:25:18 +03:00
Reima Hyvönen 4c71546b2e Cleaned some coding 2018-10-26 12:19:44 +03:00
Reima Hyvönen 4fe3909e48 Switched luma to use 32bits size ints intstead of 16bit size 2018-10-24 18:24:46 +03:00
Eemeli Kallio 284e73839e Calculating zero cost moved to its own function 2018-10-16 11:02:01 +03:00
Reima Hyvönen 381e786e10 Trying to find the bug in luma 2018-10-11 18:08:41 +03:00
Marko Viitanen c589e5ed36 Fix closed-gop frame feed, the ordering was incorrect after the first GOP 2018-10-10 11:12:03 +03:00
Reima Hyvönen 2f5f81bac3 removed the non-optimated bipred function 2018-10-09 11:19:23 +03:00
Marko Viitanen 75dce4f3ce Fix low-delay-gop usage with --no-open-gop 2018-10-04 15:16:02 +03:00
Marko Viitanen de71b58f76 Change closed GOP structure to include an additional IDR between GOPs 2018-10-04 11:17:03 +03:00
Reima Hyvönen 212a8e68fa Modified to avoid memory overflow, still some bug inside luma 2018-10-02 20:23:32 +03:00
Marko Viitanen 954f07e3d7 Add --(no-)open-gop option 2018-10-02 10:05:32 +03:00
Marko Viitanen 8bef85e056 Merge branch 'set-qp-in-cu' 2018-09-03 08:33:33 +03:00
Ari Lemmetti 2fdcc2b79d Add option --set-qp-in-cu 2018-09-03 08:32:45 +03:00
Reima Hyvönen 896034b7cf Some renamed functions back 2018-08-28 15:31:10 +03:00
Reima Hyvönen e8b5e6db4c Did some merging 2018-08-28 15:26:27 +03:00
Reima Hyvönen 7de5c74434 Updated bipred_recon to work faster 2018-08-28 15:12:31 +03:00
Reima Hyvönen 47b357cca2 Comment one test 2018-08-27 18:52:14 +03:00
Reima Hyvönen 2ca99a44e8 Updated shuffle operation to be in right order 2018-08-27 18:16:38 +03:00
Marko Viitanen b85ae3688e Signal QP in slice header if tiles and slices=tiles are enabled 
Keeps the PPS constant for various purposes
 2018-08-16 08:44:39 +03:00
Reima Hyvönen 508b218a12 some modifications made to prevent reading too much 2018-08-14 10:50:39 +03:00
Reima Hyvönen 1d935ee888 some useless stuff removed 2018-08-13 16:47:11 +03:00
Reima Hyvönen ce3ac4c05e some modifications to no_mov 2018-08-13 16:41:02 +03:00
Reima Hyvönen 15a613ae94 test if no_mov breaks testing 2018-08-13 16:02:56 +03:00
Reima Hyvönen 97a2049e58 removed pointer declaration out from switch 2018-08-10 16:42:26 +03:00
Reima Hyvönen aa94bcedbc Stream is now pointer 2018-08-10 16:38:49 +03:00
Reima Hyvönen fa5b227ece 256 to 32 doesn't work, made them by hand 2018-08-10 16:01:20 +03:00
Reima Hyvönen 408dedbcc8 removed _mm256_extract_epi8 and replaced with _mm_stream 2018-08-10 15:53:26 +03:00
Reima Hyvönen 31c35091c6 _mm256_cvtsi256_si32 removed 2018-08-10 10:06:40 +03:00
Reima Hyvönen 99dc43074f _mm256_cvtsi256_si32 breaks system, too much bits. back to extract 2018-08-10 09:59:33 +03:00
Reima Hyvönen 4f1f80b2cb Transformed convert from 256 to cast 256 -> 128 and then convert from 128 2018-08-09 15:35:54 +03:00
Reima Hyvönen 4957555eb3 Removed leftover from 939 2018-08-09 15:25:03 +03:00
Reima Hyvönen 28b165c971 Clearified some sections, added _MM_SHUFFLE macro 2018-08-09 15:23:01 +03:00
Reima Hyvönen dd04df8667 testing if error in both avx2 functions 2018-08-03 11:49:00 +03:00
Reima Hyvönen ed50d71fde Switched some variables to different location, altered inter_recon_bipred_avx2 function 2018-08-02 16:08:59 +03:00
Reima Hyvönen f5739a0028 Renaming and removing useless prints 2018-08-02 14:47:17 +03:00
Reima Hyvönen bc09f59bb6 Edited some definitions 2018-08-02 11:54:53 +03:00
Arttu Ylä-Outinen 83555c3d6d Enable --fast-residual-cost with fastest presets 2018-07-16 12:31:20 +03:00
Arttu Ylä-Outinen c438bb4a19 Add an option to skip CABAC for residual costs 
Adds command line option --fast-residual-cost=<limit>. When QP is below
the limit, estimates the cost of coding the residual coefficients from
the sum of absolute coefficients. Skipping CABAC is not worth it with
high QPs because there are fewer coefficients so CABAC is not as slow.
 2018-07-16 12:31:20 +03:00
Reima Hyvönen a4bf77f208 Tested some extract functions 2018-07-12 09:29:32 +03:00
Reima Hyvönen c05033a893 Even more useless vectors removed 2018-07-11 15:09:14 +03:00
Reima Hyvönen 884cb77238 Removed some not used vectors 2018-07-11 15:06:11 +03:00
Reima Hyvönen 792689a5ff Removed for-loops, added extract instead 2018-07-11 14:56:41 +03:00
Reima Hyvönen f9c7f6ee66 Added some break-operations for avx2 optimation 2018-07-11 14:15:38 +03:00
Reima Hyvönen cc064da143 some more optimation for bipred 2018-07-11 11:27:54 +03:00
Reima Hyvönen 9a339eef89 Merge branch 'bipred_recon' of https://gitlab.tut.fi/TIE/ultravideo/kvazaar into HEAD 
# Conflicts:
#	build/kvazaar_lib/kvazaar_lib.vcxproj
 2018-07-10 16:21:04 +03:00
Reima Hyvönen a22cf03ddb Updated to have no movement function to avx2 strategies 2018-07-10 16:07:15 +03:00
Arttu Ylä-Outinen b7474eb532 Fix SAO buffer sizes 
Increases sizes of buffers used for SAO reconstruction to avoid stack
buffer overflow in AVX2 SAO reconstruction.
 2018-07-05 15:56:30 +03:00
Arttu Ylä-Outinen b37470e80f
Merge pull request #207 from jbeich/maltivec 
Unbreak build on PowerPC if AltiVec isn't supported
 2018-07-04 11:06:41 +03:00
Reima Hyvönen ea83ae45f0 Toimiva ratkaisu 2018-07-03 11:18:51 +03:00
Jan Beich 4f4bea7496 Check -maltivec is supported before using 
PowerPC target may lack or have non-standard FPU:

$ cc -dumpmachine
powerpcspe-undermydesk-freebsd
$ cc -c -maltivec -Isrc src/strategies/altivec/picture-altivec.c
src/strategies/altivec/picture-altivec.c:1: error: AltiVec and E500 instructions cannot coexist
 2018-07-02 23:25:23 +00:00
Jan Beich b892d820f8 Clean up macOS includes on powerpc* after 93e1c9f1c3 
strategyselector.c:426:25: machine/cpu.h: No such file or directory
 2018-07-02 21:52:45 +00:00
Reima Hyvönen 17babfffa4 25.6 working optimation, ~50% faster than original 2018-06-25 17:06:16 +03:00
Arttu Ylä-Outinen 2f995f4325
Merge pull request #205 from jbeich/powerpc 
Unbreak build on non-Linux powerpc*
 2018-06-19 13:28:00 +03:00
Arttu Ylä-Outinen c1398ef818 Permit --period=1 with any GOP structure 
All intra coding is a special case so it can be permitted even though
Kvazaar normally only supports intra periods that are divisible by the
GOP length.
 2018-06-18 12:26:11 +03:00
Arttu Ylä-Outinen abdebe0bf9 Fix --owf help message 
The number of parallel frames is --owf plus one, not --owf minus one.

Fixes #204.
 2018-06-18 09:33:36 +03:00
Jan Beich 93e1c9f1c3 Add AltiVec detection for BSDs 
strategyselector.c:377:26: linux/auxvec.h: No such file or directory
 2018-06-17 15:38:24 +00:00
Miika Metsoila 98972d26c2 Document that the high tier requires level 4 or higher 2018-06-14 12:41:03 +03:00
Miika Metsoila 62b44efaa4 Write the encoding tier (main/high) into the bitstream 2018-06-14 12:41:03 +03:00
Arttu Ylä-Outinen a343f6d587 Prepare for delta QPs at CU-level 
- Replaces lcu_dqp_enabled with max_qp_delta_depth in encoder_control_t.
- Fixes set_cu_qps so that it can handle quantization groups of
  arbitrary size.
- Fixes computation of QP predictors so that it works for quantization
  groups of arbitrary size.
 2018-06-13 15:36:19 +03:00
Arttu Ylä-Outinen dc6b2024ea Modify reference count asserts to fix data races 
Changes asserts on the reference count of objects to assert the value
after KVZ_ATOMIC_INC instead of directly checking the value. Fixes some
data races detected by TSan.
 2018-06-12 09:35:07 +03:00
Ari Lemmetti 4fb1c16c61 Add early termination for intra rdo when a zero coefficient block is found. 2018-06-08 21:03:07 +03:00
Ari Lemmetti 492529fb7a Add the same comment to help message as well... 2018-05-30 14:13:15 +03:00
Ari Lemmetti 0d5972bf03 Add missing sort to intra transform split search so mode at 0 is the best 2018-05-21 13:10:38 +03:00
Sebastien Alaiwan 954bca7d6e Fix memset parameter 2018-05-17 11:24:49 +02:00
Jaakko Laitinen f9466efcbb Close file on error 2018-05-15 11:50:16 +03:00
Reima Hyvönen 9fed29f950 optimation for inter_recon_bipred 2018-04-18 15:25:44 +03:00
Arttu Ylä-Outinen 5c585c4fbc Update help message 
Updates the default option values to match the medium preset.
 2018-04-03 10:40:37 +03:00
Arttu Ylä-Outinen 2b4e22111a Update presets 
The new presets are slower but have better coding efficiency.
 2018-04-03 10:37:30 +03:00
Arttu Ylä-Outinen 7185519a1b Update command line help 
- Adds missing default values.
- Adds help for --crypto and --key.
- Adds help for --rd=3.
- Adds help for --sao options.
- Some changes to help wording.
 2018-03-23 14:33:04 +02:00
Arttu Ylä-Outinen 3606860504 Add --no-cpuid option 
Equivalent to --cpuid=0.
 2018-03-23 12:32:27 +02:00
Arttu Ylä-Outinen fb462b25ef Fix transform skip for inter 
The transform skip flag in cu_info_t was stored under the intra
substruct even though transform skip can be used for inter as well. This
caused bitstream errors. Fixed by moving the flag out of the substruct.
 2018-03-20 11:01:33 +02:00
Arttu Ylä-Outinen b64e46707d Skip raster scan step in TZ search 
Raster scan is very slow and the BD-rate improvement is marginal.
 2018-03-01 14:04:03 +02:00
Arttu Ylä-Outinen 6877064230 Add zero neighborhood check to TZ search 
Adds an additional grid search step that starts from the zero motion
vector after the normal grid search. The search range for this step is
half of the normal range.
 2018-03-01 14:02:13 +02:00
Arttu Ylä-Outinen 74a413c46a Switch to star refinement in TZ search 2018-03-01 13:06:14 +02:00
Arttu Ylä-Outinen ebee428ee1 Add loop termination to TZ grid search 
Terminates the grid search if no better motion vector was found in the
last three iterations.
 2018-03-01 13:06:06 +02:00
Arttu Ylä-Outinen 4c175621dd Fix TZ grid search and star refinement 
- Changes TZ grid search and star refinement to keep the origin constant
  instead of moving to the best position after each iteration.
- Changes star refinement to loop until there is no more improvement,
  instead of running the step only once.
 2018-03-01 12:56:57 +02:00
Arttu Ylä-Outinen 9c2d0074a2 Add rounding of motion vectors in inter search 
When the starting point for integer motion estimation was selected among
the merge candidates, the candidate motion vectors were always rounded
down. This commit changes the rounding so that they are rounded to the
nearest integer MV instead.
 2018-03-01 09:39:21 +02:00
Ari Lemmetti 662430d441 Select CU type based on SSD, transform unit tree and mode cost of luma and chroma on --rd=2 2018-02-22 19:26:48 +02:00
Arttu Ylä-Outinen cb06cfeadb Drop temporary arrays in bipred search 
Changes bipred search to use the original source and reconstruction
arrays directly instead of copying them.
 2018-02-14 11:20:51 +02:00
Arttu Ylä-Outinen 0ea516ba30 Move bipred search to a separate function 2018-02-14 09:56:53 +02:00
Arttu Ylä-Outinen 6f506be12d Drop dynamic allocation from bipred search 
Moves the temporary LCU struct used in bipred search from the heap to
the stack. The single malloc call was a huge bottleneck in bipred.
 2018-02-14 09:55:02 +02:00
Arttu Ylä-Outinen 7155dd0db7 Add negative references to L1 list 
Changes reference index list creation so that the negative references
are added to L1 in addition to L0 when biprediction is enabled and no
reordering of pictures is done. Biprediction can now be used with the
low-delay GOP structure.
 2018-02-07 14:54:52 +02:00
Arttu Ylä-Outinen 4b24cd03a2 Update for crypto++ 6.0.0 compatibility 
Changes the crypto module to use unsigned char instead of byte. The byte
typedef is no longer included in the global namespace in crypto++ 6.0.0.
See https://github.com/weidai11/cryptopp/issues/442.

Fixes #184.
 2018-02-05 13:35:03 +02:00
Arttu Ylä-Outinen 8c53417006 Check zero coefficient cost for inter 
Checks the cost of flushing all coefficients of an inter block to zero.
This is much faster than doing full RDOQ but can still reduce bitrate
significantly. Encoding speed is increased since fewer coefficient bits
have to be coded with CABAC.
 2018-01-29 12:41:56 +02:00
Arttu Ylä-Outinen 018b5ffa64 Move inter CU reconstruction to a new function 
Moves code for reconstructing all PUs in an inter CU to a new function
kvz_inter_recon_cu in inter.c.
 2018-01-24 15:05:39 +02:00
Arttu Ylä-Outinen 405b8c1069 Refactor inter MVD cost functions 
Moves duplicate code for writing the MVD of a single motion vector from
kvz_get_mvd_coding_cost_cabac and encoder_inter_prediction_unit to a new
function.
 2018-01-19 08:29:17 +02:00
Arttu Ylä-Outinen c1cca1ad7f Refactor inter MV candidate selection 
Moves duplicate code for checking the best MV candidate from functions
calc_mvd_cost, search_pu_inter_ref and search_pu_inter to a new
function.
 2018-01-19 08:29:17 +02:00
Arttu Ylä-Outinen 9067aa4535 Remove an unnecessary copy in SMP/AMP search 
SMP/AMP search is performed using a lower work tree level than the
normal inter search so the prediction info must be copied up if an
SMP/AMP mode is chosen. Previously pixels and coefficient were copied as
well. Changed to only copy prediction info.
 2018-01-18 10:36:26 +02:00
Arttu Ylä-Outinen 89a930d6dd Add part mode bitcost when using SMP/AMP blocks 2018-01-18 10:36:26 +02:00
Arttu Ylä-Outinen fc43643ba5 Use a transform split for SMP and AMP blocks 2018-01-18 10:36:25 +02:00
Arttu Ylä-Outinen c74ede148b Fix CBF flags for 4x4 luma blocks 
CBF flags were not being propagated to the upper level from blocks of
size 4x4.
 2018-01-18 10:36:25 +02:00
Arttu Ylä-Outinen 0a69e6d18f Fix selection of transform function for 4x4 blocks 
DST function was returned for inter luma transform blocks of size 4x4
even though they must use DCT. Fixed by checking the prediction mode of
the block in addition to whether it is chroma or luma.
 2018-01-18 10:36:25 +02:00
Miika Metsoila bcedfd6669 Remove the usage of errno in me-steps argument parsing 2018-01-16 14:38:43 +02:00
Miika Metsoila 39ed36830e Merge branch 'me_steps' 2018-01-16 14:22:59 +02:00
Miika Metsoila 61213e3ad9 Improve step parameter parsing and usage 2018-01-10 15:16:52 +02:00
Arttu Ylä-Outinen 649113a821 Fix inter search being used for 4x4 blocks 
When 4x4 intra blocks are enabled and inter search is limited to 16x16
and larger blocks, it is possible that inter search is accidentally done
for 4x4 blocks. Fixed by checking that block size is at least 8x8 before
doing inter search.
 2018-01-10 14:21:48 +02:00
Miika Metsoila e8e0e7596a Add a step-cutoff parameter for motion estimation search 2017-12-22 14:04:25 +02:00
Miika Metsoila 4e13608b01 Merge branch 'diamond_search' 2017-12-18 14:11:53 +02:00
Miika Metsoila 2cde0d1a18 Document diamond search option 2017-12-12 14:45:01 +02:00
Miika Metsoila b923b63b42 Add diamond search 2017-12-12 14:40:14 +02:00
Ari Lemmetti 14892fda00 Replace simple coefficient cost estimation with CABAC. Substantial improvement. 
Approximation proved to be too inaccurate while not giving actually that much speedup.
 2017-12-10 01:23:48 +02:00
Miika Metsoila ea79069dc8 Fix a type warning in encmain.c 2017-12-08 16:22:40 +02:00
Miika Metsoila 6aa4cd7528 Fix type warnings 2017-12-08 16:16:36 +02:00
Miika Metsoila b3486b5114 Fix gcc/clang warnings and errors in cfg.c 2017-12-08 16:09:00 +02:00
Miika Metsoila bac07457ea Merge branch 'hevc_level' 2017-12-08 15:57:38 +02:00
Miika Metsoila c67a24e6ec Update readme and --help text 2017-12-07 12:32:46 +02:00
Ari Lemmetti 713e694d82 Define HAVE_STRUCT_TIMESPEC on Visual Studio 2015 and later 
Fixes redefinition of timespec that Pthreads-Win32 does even if it has been already defined.
 2017-12-05 18:26:12 +02:00
Miika Metsoila f64d42169f Improve bitrate checking to accommodate non-integer and less than 1 framerates 2017-12-01 17:20:12 +02:00
Miika Metsoila 57cf92d35f Implement level's bitrate limit checking during encoding 2017-11-28 16:19:44 +02:00
Miika Metsoila 021fb27787 Add high-tier flag 2017-11-20 16:05:28 +02:00
Miika Metsoila d249059d61 Minor refactoring of level checking 2017-11-20 13:25:26 +02:00
Arttu Ylä-Outinen cf85d52b9d Kvazaar version 1.2.0 2017-11-17 15:23:33 +02:00
Miika Metsoila 4c1512e8c5 Add a check for maximum picture width and height for the given level 2017-11-15 16:39:59 +02:00
Arttu Ylä-Outinen 4cb054295a Fix linkers 
Overrides the linkers used for kvazaar, libkvazaar.la and kvazaar_tests.
When crypto++ is enabled, the C++ linker is used and when it is
disabled, the C linker is used.

This removes the need to explicitly specify -lstdc++ in configure when
crypto++ is used and fixes the build with crypto++ when libstd++ is not
installed.
 2017-11-13 15:09:38 +02:00
Miika Metsoila f9a4aba867 Update documentation, fix input fps default value, remove 0 as default level 2017-11-09 16:53:31 +02:00
Miika Metsoila ebba0a4f01 Test if input conforms to it's level's limits (excluding bitrate) 2017-11-08 16:15:41 +02:00
Miika Metsoila fb4d0c3cf2 Move level argument parsing to the correct place and give it initial values 2017-11-03 15:47:35 +02:00
Miika Metsoila 61a31054e1 Add level command-line parameter 2017-11-03 13:04:05 +02:00
Arttu Ylä-Outinen 9974380cdd Fix bipred and temporal MVP 
- Fixes two errors in calculating the POC for the reference frame for
  temporal candidate MV scaling.
- Fixes using the MV for the wrong direction when the temporal MV
  predictor block uses bi-prediction.

Fixes #160.
 2017-10-25 12:26:41 +03:00
Arttu Ylä-Outinen 841597e123 Fix picture and slice types 
Changes handling of intra pictures for --gop=8 so that every picture
with POC divisible by the intra period is intra. The first picture is
IDR and the rest of the intra pictures are CRA. POC is not reset at CRA
pictures. The leading pictures that follow the CRA picture are changed
to RASL so they are allowed to refer to pictures before the CRA picture.

Changes inter slice types to P when the L1 reference list is empty and
to B otherwise.

In all-intra, all pictures are now IDR pictures with POC zero.
 2017-10-20 13:35:26 +03:00
Jaakko Laitinen 957b6850c3 Change ref list printout to match hm decoded printout 2017-09-25 13:48:56 +03:00
Arttu Ylä-Outinen 20aea8df63 Fix POCs when using --gop=8 
When using --gop=8 with an intra period greater than one, a single POC
would be skipped before every intra frame. This commit fixes the problem
by turning the intra frames into BLA frames with leading pictures when
using --gop=8.
 2017-09-19 09:31:58 +03:00
Miika Metsoila 6e00f63469 Remove unused variables from search_pu_inter_ref function 2017-09-18 15:36:37 +03:00
Miika Metsoila 7b0101ce3d Merge branch 'reflist_changes' 
# Conflicts:
#	src/encoderstate.c
#	src/search_inter.c
 2017-09-18 14:59:37 +03:00
Miika Metsoila 769b17768d Change max function to MAX macro for clang/gcc compatibility. 
Remove couple of unnecessary comments
 2017-09-15 14:21:51 +03:00
Miika Metsoila 5f7c5443a3 Remove inter.poc 2017-09-12 14:23:19 +03:00
Miika Metsoila 6bd78a3da7 Reverse L0 list sort direction 2017-09-12 14:23:18 +03:00
Miika Metsoila 83dc7e7f50 Made L0 to sort and fixed mv_ref_coded in search_pu_inter 2017-09-12 14:23:18 +03:00
Timothe FRIGNAC d3362a238e changed strtod to strtol 2017-08-31 15:14:31 +02:00
Timothe FRIGNAC 3a1ab54ff0 Fixed memory leaks 2017-08-31 11:51:41 +02:00
Timothe FRIGNAC 466297fd77 Fixed build error 2017-08-29 17:01:18 +02:00
Timothe FRIGNAC 2e130912cb Add --key opt 2017-08-28 17:15:13 +02:00
Miika Metsoila a5f4cf09b5 Switched from storing POCs in inter.poc to state->frame->refLXs array 2017-08-21 16:34:57 +03:00
Arttu Ylä-Outinen 409d2114f0 Fix motion vector constraints 
Fixes integer motion vectors being constrained more than what was
necessary when using --mv-constraint or --wpp.
 2017-08-11 14:41:36 +03:00
Arttu Ylä-Outinen 7144a00beb Rewrite thread queue 
Changes thread queue so that only the jobs that are ready to run are
stored in the queue. Other jobs are kept track of by pointers in the
reverse dependency lists of other jobs. When a job is ready to run it is
appended to the queue. The job queue is stored as a linked list.

The definitions of threadqueue_queue_t and threadqueue_job_t are moved
to the .c file, turning them into opaque structs.

Makes thread queue code simpler. Fixes some TSan errors.
 2017-08-11 14:18:12 +03:00
Arttu Ylä-Outinen bc47fe94af Drop thread queue debug code 2017-08-11 14:18:12 +03:00
Eemeli Kallio e5cbc7a205 --sao now enables full sao 2017-08-11 13:26:55 +03:00
Eemeli Kallio 4c3453d26f Fixed issue with no-sao argument 2017-08-11 13:12:22 +03:00
Eemeli Kallio 8674c0f5ee Added paremeter for band and edge sao. 2017-08-11 11:57:09 +03:00
Eemeli Kallio d9b93ea368 Added possibility to skip edge or band sao. 2017-08-11 11:51:49 +03:00
Arttu Ylä-Outinen 4b73bdd9aa Skip checked motion vectors in early termination 
Changes the second iteration of early termination to skip the motion
vectors that were already checked in the first iteration.
 2017-08-09 14:29:09 +03:00
Arttu Ylä-Outinen 606d441362 Skip computing MV cost twice in hexagon search 
Changes the first step of hexagon search to skip the zero offset since
the cost of the motion vector has already been computed.
 2017-08-09 14:29:09 +03:00
Arttu Ylä-Outinen fa4648061d Add mv, cost and bitcost to inter_search_info_t 2017-08-09 14:29:08 +03:00
Arttu Ylä-Outinen 328f051d7f Put inter search parameters in a single struct 
Adds struct inter_search_info_t for holding the parameters that are used
by most function related to inter search. Passing the parameters in
a single struct greatly reduces the number of parameters for many
functions.
 2017-08-09 14:27:53 +03:00
Miika Metsoila 0dd069f8af Fixed using wrong POC in add_temporal_candidate 2017-08-09 13:50:21 +03:00
Miika Metsoila 25e0a954c7 Fixed 2 bugs causing incorrect video output 2017-08-09 13:50:21 +03:00
Arttu Ylä-Outinen 24ecddd2a5 Fix wrong strides in SAO reconstruction 
Functions kvz_sao_reconstruct and encoder_sao_reconstruct used
frame->width as the stride instead of frame->rec->stride when accessing
frame->rec->data. This caused errors when using tiles and SAO.
 2017-08-01 15:40:49 +03:00
Arttu Ylä-Outinen f0bf959d17 Fix alignment errors in 32-bit build with MSVC 
Changes the work_tree parameter in search.c functions from an array to
a pointer. Fixes "formal parameter with requested alignment of 8 won't
be aligned" errors.
 2017-07-28 09:27:02 +03:00
Arttu Ylä-Outinen 9694bd2fae Fix build on 32-bit systems 
Function coeff_abs_sum_avx2 that was added in e950c9b was outside the
AVX2 #if directive.
 2017-07-28 09:19:29 +03:00
Arttu Ylä-Outinen ecb0275cdd Store CU arrays as pointers to the main array 
Changes field state->tile->frame->cu_array->data to point to the CU
array in the main encoder state. Removes the need to copy the CU array
to the main CU array after search.
 2017-07-28 08:36:45 +03:00
Arttu Ylä-Outinen e950c9b101 Add AVX2 implementation for coefficient sum 2017-07-28 07:39:36 +03:00
Arttu Ylä-Outinen d50ae6990c Add sum of absolute coefficients to strategies 2017-07-28 07:39:15 +03:00
Arttu Ylä-Outinen 59faca0646 Skip CABAC coefficient cost for --rd=0 2017-07-28 07:33:03 +03:00
Arttu Ylä-Outinen 19e051ea40 Reduce intra threshold 
Reduces intra threshold for --rd=0 from 20 to 8. Threshold of 20
increased BD-Rate too much.
 2017-07-25 13:26:38 +03:00
Arttu Ylä-Outinen e9cf15465e Fix inter cost in bipred 
The cost of coding MV ref indices and MV direction was added to bitcost
but not inter cost. Fixed by adding the extra bits to inter as well.
 2017-07-24 15:24:04 +03:00
Arttu Ylä-Outinen edbe00763e Drop extra parameter in kvz_image_calc_sad 
Drops the parameter max_lcu_below which was always set to -1.
 2017-07-24 15:21:19 +03:00
Arttu Ylä-Outinen ffac29061f Fix extrapolated inter SATD 2017-07-24 15:11:05 +03:00
Arttu Ylä-Outinen 631ef53d2a Fix inter cost calculations 
Inter costs are computed using SAD except when fractional motion
estimation or bi-prediction is enabled. This commit changes
search_pu_inter_ref to recalculate the cost with SATD. Fixes inter/intra
cost comparisons since intra costs are always SATD costs.
 2017-07-24 15:11:05 +03:00
Arttu Ylä-Outinen 6ce2fb1238 Add pixel offsets to encoder_state_config_tile_t 
Adds fields offset_x and offset_y to encoder_state_config_tile_t.
 2017-07-24 15:11:05 +03:00
Arttu Ylä-Outinen 2380ba0d41 Reduce copying in kvz_get_coeff_cost 
Changes function kvz_get_coeff_cost to only copy the CABAC contexts and
not the whole encoder state.

Other threads could be simultaneously using the other parts of the
encoder state. Only copying the CABAC fixes a TSan data race warning.
 2017-07-24 12:38:41 +03:00
Arttu Ylä-Outinen 24b462f801 Align coefficients to 8 bytes 
Adds alignment attribute to lcu_coeff_t. The coefficients are sometimes
handled as 64-bit integers containing four coefficients so the arrays
should be aligned to 8 bytes.

Fixes a UBSan error about misaligned reads.
 2017-07-24 12:37:37 +03:00
Arttu Ylä-Outinen 5ddb43c6fe Fix undefined left shifts in rdo 
Replaces left shifts by multiplications when the operand may be
a negative value. Left shift of a negative value is undefined behavior.
 2017-07-24 12:35:10 +03:00
Arttu Ylä-Outinen d1e64ad62b Fix undefined left shifts 
Replaces left shifts by multiplications when the operand may be
a negative value. Left shift of a negative value is undefined behavior.
 2017-07-20 11:15:30 +03:00
Arttu Ylä-Outinen 07b5fb9caf Fix out-of-bounds read in encoderstate 
When calling encoder_state_encode_leaf with POC 0, index -1 of the GOP
array would be accessed. Fixed by skipping the code for I-frames.
 2017-07-20 11:15:30 +03:00
Arttu Ylä-Outinen 8c4a3473a8 Change --owf=auto and --threads=auto selection 
Changes OWF selection so that it is chosen based on the maximum number
of parallel CTUs. Number of threads is limited to prevent overhead from
extra threads.
 2017-07-20 09:42:28 +03:00
Arttu Ylä-Outinen 4fc9b743c1 Drop an unnecessary pthread_cond_broadcast 
Drop pthread_cond_broadcast on threadqueue->cond in function
kvz_threadqueue_waitfor. The broadcast caused threads to be woken up
more often than necessary.
 2017-07-19 11:09:30 +03:00
Arttu Ylä-Outinen 14003c6a30 Disable printing PSNR with --no-psnr 2017-07-19 10:38:37 +03:00
Arttu Ylä-Outinen e90bde5c62 Clarify PSNR output 
Adds letters Y, U and V to the PSNR output to make it clearer that the
printed values are the luma and chroma PSNR.
 2017-07-19 10:33:43 +03:00
Arttu Ylä-Outinen fdb3480b54 Enable strategies for SAO reconstruction 
Re-enables strategies for SAO reconstruction. They were disabled in
commit ec9ff42.
 2017-07-11 10:35:18 +03:00
Arttu Ylä-Outinen 333dba3884 Add static to SAO strategies 2017-07-11 10:02:01 +03:00
Miika Metsoila e8cc2d8f6a Small fixes 2017-07-07 13:58:19 +03:00
Arttu Ylä-Outinen 67a60a35e3 Fix invalid calls to normalize_lcu_weights 
Changes encoder_state_init_new_frame to only call normalize_lcu_weights
when the weights have been written to the array and rate control is
enabled. When rate control is disabled, the weights are not used.
 2017-07-07 11:05:31 +03:00
Arttu Ylä-Outinen 563bc26e71 Fix out-of-bounds read in AVX2 SAO 
AVX2 version of SAO loaded offsets with a 256 bit read even though there
are only five 32 bit integers.
 2017-07-06 13:04:52 +03:00
Arttu Ylä-Outinen 0850b17f96 Drop get_wpp_limit in search_inter 
WPP limit for motion vectors is now computed inside fracmv_within_tile.
 2017-07-05 13:22:53 +03:00
Arttu Ylä-Outinen 2a85f0f5a4 Move hard-coded MV limits to encoder_control_t 
Adds field max_inter_ref_lcu to encoder_control_t. It is used to set up
inter-LCU dependencies in encoder_state_encode_leaf and restrict motion
vectors in fracmv_within_tile.
 2017-07-05 13:22:53 +03:00
Arttu Ylä-Outinen bb5354f7e2 Relax inter-CTU dependencies when SAO is off 
When using WPP and OWF, the first CTU of a row depends on the last CTU
of the row below in the reference frame. This is necessary when SAO is
enabled since we currently do SAO for a whole CTU row at a time. When
SAO is disabled, however, it is unnecessary to wait for the whole row.

Changes CTUs to depend only on the CTU below in the reference frame
instead of the whole row when WPP and OWF are enabled and SAO disabled.
Gives a significant speedup when running on a machine with many CPU
cores.
 2017-07-05 13:21:06 +03:00
Arttu Ylä-Outinen 1efa2708b2 Do SAO reconstruction for a single CTU at a time 
Moves SAO reconstruction into encoder_state_worker_encode_lcu instead of
doing it in a separate step for the whole CTU row. Reconstruction of the
rightmost 10 pixels and bottommost 10 pixels of a CTU is delayed until
the neighboring CTU has been deblocked.

Doing SAO for the whole CTU row at a time caused unnecessary inter-CTU
dependencies when using WPP and OWF. The first CTU of a row would need
to wait until SAO was done for the row below in the previous frame.
Moving SAO reconstruction to immediately after deblocking each CTU fixes
this problem.
 2017-07-04 15:14:31 +03:00
Arttu Ylä-Outinen ec9ff42077 Rewrite SAO recon to handle arbitrary sized blocks 
Adds width and height parameters to function kvz_sao_reconstruct and
changes it to take coordinates in units of pixels. This will be useful
for doing SAO for areas smaller than a whole CTU.
 2017-06-30 16:09:18 +03:00
Miika Metsoila dcd7acf4fd Fixed crash and incorrect info output 2017-06-27 16:05:15 +03:00
Miika Metsoila f8b6234fdb Changes to refence lists to behave more like L0/L1 lists from the specification 2017-06-27 16:05:15 +03:00
Arttu Ylä-Outinen 2c66e0bbd2 Fix warnings about invalid reads in AVX2 ipol 
AVX2 filter functions read pixels in chunks of 8 or 16 bytes. At the end
of the block, the read goes out of the bounds of the pixels array. The
extra pixels do not affect the result.

Fixes valgrind complaining about the invalid reads by allocating 5 extra
pixels in kvz_get_extended_block_avx2
 2017-06-22 09:37:55 +03:00
Arttu Ylä-Outinen 4d20e156db Fix handling intra period not multiple of GOP length 
With low delay GOP structure, it is possible to use an intra period that
is not a multiple of the GOP structure length. Commit 00c9f52 changed
encoder_state_init_new_frame to reset POC on intra frames. GOP offset,
however, was not reset, resulting in invalid POCs and references for the
following frames.

This commit changes function kvz_encoder_feed_frame so that GOP offset
is correctly reset on intra frames.
 2017-06-22 09:29:00 +03:00
Arttu Ylä-Outinen 00c9f52bd4 Fix setting picture type when using GOP 
Changes encoder_state_init_new_frame to set intra frame pictype to
KVZ_NAL_IDR_W_RADL even when using GOP.
 2017-06-21 13:21:47 +03:00
Arttu Ylä-Outinen f54a25f112 Fix crash when immediately closing encoder 
When closing the encoder, the pictures stored in the input frame buffer
are freed by repeatedly calling kvz_encoder_feed_frame. If the encoder
was closed immediately after opening it, kvz_encoder_feed_frame would be
called with an unprepared encoder state. This would trigger an assert.

Fixed by changing kvz_encoder_feed_frame so that it does not require the
encoder state to be prepared.
 2017-06-15 11:57:46 +03:00
Arttu Ylä-Outinen b74e0458fd Set inter transform depth to zero 
Sets max_transform_hierarchy_depth_inter to 0 in SPS. This saves some
bits because split_transform_flag does not need to be coded for inter
blocks.

When SMP and AMP blocks are enabled the depth is set to 1 instead.
Otherwise inter split flag would default to 1 for SMP and AMP blocks,
resulting in an unnecessary transform split.
 2017-06-08 10:08:20 +03:00
Arttu Ylä-Outinen 8dd01ba5a9 Refactor helper functions in search 
Combines functions lcu_set_intra_mode and lcu_set_inter_pu to a single
function. Removes some duplicated code.
 2017-06-06 10:32:09 +03:00
Arttu Ylä-Outinen 1bbecf7584 Refactor work tree copy functions 
Extracts common code shared by work_tree_copy_up and work_tree_copy_down
to a separate function.
 2017-06-06 10:32:00 +03:00
Arttu Ylä-Outinen 2b169d5d63 Fix crash in kvazaar_close 
Changes kvazaar_close to stop all threads before freeing encoder states.
Fixes a crash when the encoder is closed before all pictures have been
encoded.
 2017-06-02 10:05:33 +03:00
Arttu Ylä-Outinen eb9a05b7ef Fix memory leak 
Changes kvazaar_close to free the remaining pictures in the the input
frame buffer. Fixes a memory leak when the encoder is closed while there
are pictures left in the buffer.
 2017-06-01 15:39:35 +03:00
Arttu Ylä-Outinen 8b2483ca1c Combine intra reconstruction functions 
Replaces function kvz_intra_recon_lcu_luma and
kvz_intra_recon_lcu_chroma in intra.c with function kvz_intra_recon_cu.
The new function can handle reconstruction for both luma and chroma.
Removes some duplicated code.
 2017-05-24 12:07:31 +03:00
Arttu Ylä-Outinen e67fdb853d Move intra leaf TB recon to a separate function 
Moves code for intra leaf transform block reconstruction from functions
kvz_intra_recon_lcu_luma and kvz_intra_recon_lcu_chroma to a new
function intra_recon_tb_leaf. Removes some duplicated code.
 2017-05-24 12:07:31 +03:00
Arttu Ylä-Outinen 13d2fdbd21 Drop unused kvz_videoframe_get_cu functions 2017-05-24 11:15:31 +03:00
Arttu Ylä-Outinen f5eef7f33c Use luma pixel coordinates in encode_coding_tree 
Changes functions encode_intra_coding_unit and encode_coding_tree to
take coordinate arguments in units of luma pixels instead of 8 px
blocks. This should make the code easier to understand.
 2017-05-24 11:15:31 +03:00
Arttu Ylä-Outinen 525a5180ff Combine intra CU encoding functions 
Merges functions encode_intra_coding_unit and
encode_intra_coding_unit_encry. Removes a lot of duplicated code.
 2017-05-24 11:12:40 +03:00
Arttu Ylä-Outinen 610c91b0c5 Use luma pixel coordinates in TU coding functions 
Changes functions encode_transform_unit and encode_transform_coeff to
take coordinate arguments in units of luma pixels instead of 4 px
blocks. This should make the code easier to understand.
 2017-05-23 15:36:16 +03:00
Arttu Ylä-Outinen 2e8838de6e Fix crash when crypto compiled in but disabled 
When kvazaar was built with crypto++ but running without using
encryption features, kvazaar attempted to delete an uninitialized crypto
handle. Fixed by setting the handle to NULL in kvz_encoder_state_init.
 2017-05-23 14:01:48 +03:00
Arttu Ylä-Outinen 2f2c281e8e Fix a memory leak in crypto 
A CryptoPP::CFB_Mode<CryptoPP::AES>::Encryption was allocated at the
beginning of encoder_state_encode_leaf and was never freed. This commit
changes encoder_state_worker_encode_lcu to delete the CFB_Mode. Also
moves crypto handle from encoder_state_config_tile_t to encoder_state_t
so that it can be safely deleted without affecting other threads in the
same tile.
 2017-05-23 11:51:25 +03:00
Arttu Ylä-Outinen 22155950c1 Rewrite crypto to conform to kvazaar code style 2017-05-23 11:51:25 +03:00
Arttu Ylä-Outinen 6829865190 Fix inline declaration in intra_mode_encryption 
Moves the inline declaration of intra_mode_encryption before the type
and changes it to use the INLINE macro. Inline declaration after type
triggered a warning on GCC.
 2017-05-23 11:50:32 +03:00
Arttu Ylä-Outinen 5f8e17d4ba Eliminate a race condition in threadqueue 
Fixes the order of acquiring locks for the job and its dependency in
kvz_threadqueue_job_dep_add. The dependency is locked before the job
that depends on it. This is the same order as in threadqueue_worker.

Acquiring the locks in different order in kvz_threadqueue_job_dep_add
and threadqueue_worker would sometimes result in a deadlock.
 2017-05-18 12:25:53 +03:00
Arttu Ylä-Outinen 4b213477f0 Return best MV from inter early terminate 
When using --me-early-termination=sensitive, early termination of inter
search used to always return the starting point if no tested motion
vector was good enough to continue the search. This commit changes
early_termination to always return the best motion vector and cost
found.
 2017-05-18 09:05:14 +03:00
Arttu Ylä-Outinen 382636de55 Fix handling too large QPs 
Changes kvz_config_validate to output an error if the given QP is out of
range and changes kvz_set_picture_lambda_and_qp to clip the QP to the
valid range if is too large after applying QP offset from GOP structure.
 2017-05-17 12:41:51 +03:00
Arttu Ylä-Outinen de8b59c681 Drop unused function kvz_coefficients_blit 2017-05-12 16:48:30 +03:00
Arttu Ylä-Outinen bcfa5a3cd9 Add a comment explaining the coefficient order 2017-05-12 16:46:57 +03:00
Arttu Ylä-Outinen 95775a1645 Change coefficient storage order 
Changes coefficient storage order to a zig-zag order. Reduces
unnecessary copying of coefficients to temporary arrays.
 2017-05-12 16:46:57 +03:00
Arttu Ylä-Outinen 9395867a9a Quantize all colors in a single traversal 
Changes kvz_quantize_lcu_residual to process all three colors in
a single traversal of the TU tree.
 2017-05-12 16:42:41 +03:00
Arttu Ylä-Outinen 1e58fd6b16 Split kvz_quantize_lcu_residual 
Splits kvz_quantize_lcu_residual to two functions that handle the TU
tree recursion and quantization of a single TU.
 2017-05-12 16:42:41 +03:00
Arttu Ylä-Outinen cc87e0dcc7 Combine luma and chroma quantization functions 
Replaces functions kvz_quantize_lcu_luma_residual and
kvz_quantize_lcu_chroma_residual in transform.c with function
kvz_quantize_lcu_residual. The new function can handle any of the YUV
colors. Removes some duplicated code.
 2017-05-12 16:42:41 +03:00
Arttu Ylä-Outinen 1357dd0599 Pass coeffs through encoder state 
Changes the way coefficients are passed from kvz_search_lcu to
kvz_encode_coding_tree. Drops fields coeff_y, coeff_u and coeff_v in
videoframe_t and instead passes them through field coeff in
endoder_state_t.
 2017-05-12 16:42:41 +03:00
Eemeli Kallio 2cad3173ec Reduced amount of modes for search_intra_rdo 2017-05-12 15:56:07 +03:00
Arttu Ylä-Outinen 26adef4492 Merge branch 'erp-aqp' 2017-05-12 15:05:24 +03:00
Eemeli Kallio 55e0e65733 Added INLINE to kvz_get_ic_rate and kvz_get_coded_level in rdo.c 2017-05-12 15:03:30 +03:00
Arttu Ylä-Outinen ee3d4d0e78 Add adaptive QP for 360 degree video 
Adds option --erp-aqp for enabling adaptive QP for 360 degree video with
equirectangular projection. When projected into a spherical surface,
the middle part of the video covers relatively larger area than the top
and bottom parts. Enabling --erp-aqp sets up a ROI delta QP array which
uses higher QPs for the top and bottom of the video and lower QPs for
the middle part.
 2017-05-11 12:31:53 +03:00
Arttu Ylä-Outinen 79cb3a2fd3 Permit negative QP deltas in ROI 
Delta QPs should not be arbitrarily restricted to positive values.
 2017-05-11 12:13:47 +03:00
Arttu Ylä-Outinen edfbd6f122 Add field lcu_dqp_enabled to encoder_control_t 
Delta QPs for LCUs are enabled when either ROI coding or rate control is
enabled. Having a single field is simpler than always checking whether
ROI or rate control is enabled.
 2017-05-11 12:13:47 +03:00
Arttu Ylä-Outinen 2f2405dfe6 Fix crash when PU depth is limited 
When video width or height was not a multiple of the smallest CU size,
no prediction would be performed at the border CUs. Kvazaar would later
crash at an assertion failure when attempting to write the bitstream for
the CU.

Fixed by permitting inter and intra prediction when the CU split is
forced, even if CUs of that size would otherwise be disabled.
 2017-04-27 10:35:48 +03:00
Arttu Ylä-Outinen 9130b5107c Change handling of infinite PSNR in encmain 
Changes encmain to print 999.99 as PSNR when SSE is zero. This behavior
is in line with HM. Previously SSE was set to 99 when it was zero.
 2017-04-27 10:35:13 +03:00