shithub: dav1d

39b35e34 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/07 16:26

fix backup of t->l.tx_lpf_uv at tile boundaries for 4:2:2 and 4:4:4

a9a09b70 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/08 08:18

cdef: simplify sec_taps

7775e15a – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/08 09:39

Fix context for comp_type and intrabc

ba5eecc4 – Henrik Gramner <gramner@twoorioles.com> authored on 2018/11/08 12:05

Align blend masks

e4fbbbce – Henrik Gramner <gramner@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/07 16:42

Add blend AVX2 asm

8676fda3 – Henrik Gramner <gramner@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/07 16:41

Add blend checkasm unit test

bbc11c99 – Ronald S. Bultje <rsbultje@gmail.com> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2018/11/07 13:12

cdef: make scalar types non-sized

7fe3b037 – James Almer <jamrial@gmail.com> authored on 2018/11/07 15:08

decode: clear the temporal tile data cache in Dav1dContext after copying to the frame context

1e852dc1 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/06 06:42

AVX2 for emu_edge

ed129818 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/07 04:22

Change stride of emu_edge buffer from 160 to 192

47a581a5 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/06 06:42

Change emu_edge function argument order for easier SIMD

404a8fbf – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/06 02:49

Add emu_edge checkasm unit test

b8b56a17 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/07 09:47

Fix crash if --muxer is not set

fac1a1af – Tristan Laurent <t.laurent@ateme.com> authored and Ronald S. Bultje <rsbultje@gmail.com> committed on 2018/11/07 12:56

Add 'null' muxer in help

55d512c7 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/07 15:20

recon: fix bilinear entry in dav1d_filter_dir table

5945f34f – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/07 11:58

coef/dequant: clip coefs before and after dequantization

48a7486a – James Almer <jamrial@gmail.com> authored on 2018/11/06 17:59

picture: add dav1d_picture_move_ref()

2044016e – Steve Lhomme <robux4@videolan.org> authored on 2018/11/07 03:37

Make the public headers C++ friendly

0ddaaf8e – Henrik Gramner <gramner@twoorioles.com> authored on 2018/11/06 18:18

Fix GCC aliasing warnings

67436a48 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/06 16:45

threading: fix deadlock with multiple tile rows

704a5252 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/04 17:19

frame threading: signal reference frame decoding errors

6f979f81 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/05 15:03

tile threading: pass decoding errors through progress tracking

82cce09a – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/05 18:38

do not use the picture after an allocation error

014b0a06 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/05 16:28

submit_frame: frame threading cleanup on decoding errors

2366b77a – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/05 18:35

submit_frame: always 'goto error' for proper cleanup

e3b5d4d0 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/05 04:27

Use grouped context setting

01386d4c – Ronald S. Bultje <rsbultje@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/06 00:52

Move emu_edge() from recon_tmpl.c to Dav1dMCDSPContext in mc_tmpl.c

17426a6a – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/28 19:33

checkasm: Don't benchmark all the 8tap filters

84b7a861 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/06 09:08

checkasm: Extend mc testing for any combination of 2-32 x 2-32, for OBMC

d9470160 – Tristan Laurent <t.laurent@ateme.com> authored and Ronald S. Bultje <rsbultje@gmail.com> committed on 2018/11/06 08:38

Add "disable_cdf_update" support

bd747b11 – James Almer <jamrial@gmail.com> authored on 2018/11/05 19:14

getbits: prevent undefined shift in dav1d_get_vlc()

44ad79e9 – Henrik Gramner <gramner@twoorioles.com> authored on 2018/11/05 13:48

Fix SGR AVX2 asm on Windows

973d11bf – Henrik Gramner <gramner@twoorioles.com> authored on 2018/11/05 09:13

x86: Add warp8x8 and warp8x8t AVX2 asm

22e144fb – Henrik Gramner <gramner@twoorioles.com> authored on 2018/11/05 09:11

Add checkasm unit tests for warp8x8 and warp8x8t

a0692eb8 – Henrik Gramner <gramner@twoorioles.com> authored on 2018/11/05 09:11

Reorder the mc warp filter array

4a499fd5 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/02 11:39

Add AVX2 implementation for SGR looprestoration

bfdfd1aa – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/01 11:48

Simplify SGR C code

7d1d7d28 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/11/01 11:44

Add unit test for SGR looprestoration

c6fe27cd – David Michael Barr <b@rr-dav.id.au> authored on 2018/11/05 13:30

Simplify rate calculation in update_cdf

aa5f55b2 – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/04 16:33

remove unused function av1_clamp_mv()

d4393288 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/04 17:23

move rl32() to its only user tools/dav1d/input/ivf.c

d5421c30 – Ronald S. Bultje <rsbultje@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/04 02:53

Remove debug and redundant zero-inits from ref_mvs.c

f2f8f57b – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/04 14:50

loopfilter: use width aligned to 4 instead of 8

adaa9ba6 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/04 14:29

loop_filter_h_*_c: removed unnecessary mask to width

22d3b6d9 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/04 12:50

loopfilter: limit filter width to the frame edge

ef677d6a – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/04 14:48

calculate and cache the picture dimension in 4x4 blocks

e0c3186e – Ronald S. Bultje <rsbultje@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/02 11:42

Add simpler version of update_cdf() for boolean symbols

0941aece – Ronald S. Bultje <rsbultje@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/02 15:33

Simplify blend() by making intermediate buffer contiguous

08da01aa – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/02 12:52

CI: update master in the cached test data repository during fetch

4cab7e31 – SmilingWolf <lupo996@gmail.com> authored on 2018/11/03 05:25

Fix typo in dav1d_cli_parse.c

44cd6e25 – Martin Storsjö <martin@martin.st> authored on 2018/11/02 19:03

dav1d: Only return 0/1 from md5_verify

16ce8b1a – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/01 13:24

meson: prevent memory sanitizer builds with asm

d4fcb70d – Martin Storsjö <martin@martin.st> authored on 2018/11/02 06:21

arm: Improve use of getauxval in quirky environments

287ba91b – Dale Curtis <dalecurtis@google.com> authored on 2018/11/01 20:34

Include <stdlib.h> for win32/thread.c

8e5f126f – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/01 18:53

mem: use memalign as fallback for posix_memalign and _aligned_malloc

03d59ea5 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/01 15:52

arm: use defined() for defines which are not guaranteed to be defined

40dc545e – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/01 17:07

obu: max_tile_height_sb is only used for non-uniform tiles

1a8cbc03 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/10/31 16:02

CI: cache dav1d-test-data

9eaebda6 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/10/31 16:08

CI: run testdata tests on all branches and merge requests

ed79e075 – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/11/01 06:00

arm: add /proc/cpuinfo based NEON detection for Android J-B

7b61d4ec – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/31 17:29

arm: add cpu flag detection

2689dfb5 – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/31 17:06

arm: call dav1d_get_cpu_flags_arm only when asm is enabled

9c68e809 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/01 10:09

fix msan use-of-uninitialized-value warning in decode_coefs

24108f83 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/01 09:14

Revert "explicitly clear some Dav1dRef pointers after calling dav1d_ref_dec()"

f55ac48c – Janne Grunau <janne-vlc@jannau.net> authored on 2018/11/01 09:12

dav1d_ref_dec: always set the ref pointer to NULL

9fb690fa – David Michael Barr <b@rr-dav.id.au> authored on 2018/11/01 15:38

Match style for update_cdf and remove nsymbs2speed

aeb5a5ef – James Almer <jamrial@gmail.com> authored on 2018/10/31 14:51

explicitly clear some Dav1dRef pointers after calling dav1d_ref_dec()

b7d2b7d2 – James Almer <jamrial@gmail.com> authored on 2018/10/31 11:06

ref: don't leave dangling pointers around when freeing Dav1dRef

6fd4013a – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/30 19:50

dav1d: add --verify option to verify decoding results

285d1b76 – Luc Trudeau <ltrudeau@twoorioles.com> authored on 2018/10/31 07:46

Remove dav1d_sgr_one_by_x

e0a05e5e – Janne Grunau <janne-vlc@jannau.net> authored on 2018/10/31 05:18

fuzzing: avoid a undefined left shift for lossless coded blocks

33d16d81 – Rostislav Pehlivanov <atomnuker@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/26 16:33

Rewrite msac.c

c0351e1b – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/25 04:55

CI: add testdata based regression tests

12465310 – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/30 16:42

fuzzing: limit maximum frame size to 4kx4k

6e796314 – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/30 17:04

decode: clean up properly after frame allocation error

fa93b8ff – Martin Storsjö <martin@martin.st> authored on 2018/10/30 19:31

meson: Simplify the name of the checkasm test target

8cef1efc – Marvin Scholz <epirat07@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/29 19:40

Readme: Add info how to run tests

4a535b9f – Marvin Scholz <epirat07@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/29 19:27

Do not run other tests parallel with checkasm test

ea707934 – Marvin Scholz <epirat07@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/14 11:10

Add option to include extra testdata tests

8fbd87e5 – James Almer <jamrial@gmail.com> authored on 2018/10/29 16:44

obu: fix parsing film_grain_present and num_operating_points in Sequence Headers

8fac3029 – Ronald S. Bultje <rsbultje@gmail.com> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2018/10/28 08:00

Don't use msac_decode_bool() for edge partition parsing

c52cc17a – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2018/10/28 17:09

API: constify Dav1dData and dav1d_data_wrap()

9f17489c – Janne Grunau <janne-vlc@jannau.net> authored on 2018/10/26 16:48

unref reference pictures after decoding errors

46a3fd20 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/26 13:00

Add a 4x4 cdef_filter AVX2 implementation

e2c6d029 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/29 05:26

Add 8x8 cdef_filter AVX2 implementation

3b02d3a9 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/25 09:55

Add CDEF filter checkasm unit test

8007c79f – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/25 09:22

Simplify/rewrite cdef filtering code

ba08e37c – Marvin Scholz <epirat07@gmail.com> authored on 2018/10/28 22:04

dav1d: Fix theoretical invalid pointer dereference

d8996b18 – Ronald S. Bultje <rsbultje@gmail.com> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2018/10/28 10:26

Clear reference state upon sequence header change

e74daebe – Janne Grunau <janne-vlc@jannau.net> authored on 2018/10/26 06:15

libfuzzer: build oss-fuzz fuzzer binaries with meson

5ff85043 – James Almer <jamrial@gmail.com> authored on 2018/10/28 10:17

picture: make poc the first field in Dav1dPicture

0f4253f2 – Ronald S. Bultje <rsbultje@gmail.com> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2018/10/28 08:03

Backup extra 8 pixels at bottom of frame

172f277b – Ronald S. Bultje <rsbultje@gmail.com> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2018/10/28 04:50

Cast masks to unsigned before upshifting

cacfb8d2 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/10/28 07:25

parse_obu: reset have_{seq,frame}_hdr on new OBU_{SEQ,FRAME}_HDR

ba789ebf – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/27 17:28

Require bits/component to match between picture and references

acde4240 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/27 14:38

Add AVX2 SIMD implementation for cdef_dir

dd797aa2 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/23 04:39

Add CDEF unit test

fdf780f7 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2018/10/23 06:06

Rewrite cdef_dir C code

c6e66595 – Janne Grunau <janne-vlc@jannau.net> authored on 2018/10/27 16:37

fix hang in dav1d_close() with more than 31 tile threads

8d5a8121 – James Almer <jamrial@gmail.com> authored on 2018/10/27 06:50

tools/dav1d: don't call write_header and write_trailer when it's not needed