shithub: dav1d

Info  •  Files  •  Log  •  Branches

041f6535 – James Almer <[email protected]> authored on 2018/11/25 14:14

Put layout back in Dav1dSequenceHeader

513dfa99 – Martin Storsjö <[email protected]> authored on 2018/11/15 11:15

arm64: looprestoration: NEON optimized wiener filter

b6bb8536 – Ronald S. Bultje <[email protected]> authored on 2018/11/24 06:23

Store literal bitstream values in Dav1dSequence/FrameHeader

3e95d8ed – Ronald S. Bultje <[email protected]> authored on 2018/11/23 15:57

Move Dav1dPictureParameters-related entries to top of Dav1dSeqHdr

9c4bf181 – Ronald S. Bultje <[email protected]> authored on 2018/11/23 15:45

Add some more constants from section 3

14406338 – Ronald S. Bultje <[email protected]> authored on 2018/11/18 04:33

Fix some crashes (ninja test) from the previous patches

c19fb21d – Ronald S. Bultje <[email protected]> authored on 2018/11/18 04:11

Remove entries from c->refs[] that can be retrieved from frame_hdr

5a4fea2c – Ronald S. Bultje <[email protected]> authored on 2018/11/17 16:11

Add frame_hdr/seq_hdr to Dav1dPicture

7cc54b91 – Ronald S. Bultje <[email protected]> authored on 2018/11/17 15:55

Move Av1FrameHeader and Av1SequenceHeader into public headers

03d4ede0 – Ronald S. Bultje <[email protected]> authored on 2018/11/18 11:47

Make frame_hdr a referenced object

c371907f – Ronald S. Bultje <[email protected]> authored on 2018/11/17 14:13

Make frame_hdr a pointer

e071d257 – Ronald S. Bultje <[email protected]> authored on 2018/11/17 13:31

Make sequence header a referenced object

41e64ac1 – Raphaël Zumer <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/11/20 18:09

Match aomdec Y4M output for colocated chroma

3f410fd9 – Ronald S. Bultje <[email protected]> authored on 2018/11/24 13:20

Rewrite flushing logic

c1b0808c – Janne Grunau <[email protected]> authored on 2018/11/24 07:44

backup_lpf: do not store 4 pixels rows at the bottom edge of the picture

81b10e8c – James Almer <[email protected]> authored on 2018/11/23 17:26

obu: fix temporal/spatial layer checks when decoder model is enabled

70dc7bd5 – Janne Grunau <[email protected]> authored on 2018/11/23 18:58

align tl_edge on stack for interintra prediction

951b0c98 – Janne Grunau <[email protected]> authored on 2018/11/23 16:40

scalable: output only frames of the highest selected spatial layer

3a4445bd – Janne Grunau <[email protected]> authored on 2018/11/23 16:50

API/scalable: add all_layers Dav1dSettings

0eafb6f5 – Janne Grunau <[email protected]> authored on 2018/11/23 16:07

API/scalable: add operating point Dav1dSetting

0e3c2197 – David Michael Barr <[email protected]> authored on 2018/11/21 18:20

checkasm: Add unit tests for chroma-from-luma subsampling

6af37afa – Ronald S. Bultje <[email protected]> authored on 2018/11/22 03:51

Add API to match input images to specific output pictures

f8e1a621 – Janne Grunau <[email protected]> authored on 2018/11/22 14:23

frame-mt: check delayed pics for decoding errors

3fa2f04f – Niklas Haas <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/22 15:52

film_grain: don't attempt copying chroma planes for I400

b28788c0 – Michael Bradshaw <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/21 14:27

Modify release_picture_callback to take a picture instead of a data pointer

27860ee5 – Janne Grunau <[email protected]> authored on 2018/11/22 16:12

CI: move test-debian-msan to amd64 to avoid timeouts

516333ca – Konstantin Pavlov <[email protected]> authored on 2018/11/22 14:21

CI: use aarch64 images from VideoLAN registry.

82404efb – James Almer <[email protected]> authored on 2018/11/22 06:43

obu: set have_frame_hdr to true only when there were no errors

2b39cfb0 – Janne Grunau <[email protected]> authored on 2018/11/22 08:26

film_grain: round subsampled width/height up

790dd353 – Michael Bradshaw <[email protected]> authored on 2018/11/21 14:04

Fix some minor typos

12337860 – Ronald S. Bultje <[email protected]> authored on 2018/11/21 14:29

Set correct width for post-scaling picture if super-res is enabled

9abc8747 – Janne Grunau <[email protected]> authored on 2018/11/21 17:25

film_grain: fix build when bit depths are disabled

12e0c905 – Martin Storsjö <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/05 06:45

arm64: Work around a clang/LLVM misoptimization

241dafa0 – Ronald S. Bultje <[email protected]> authored and Jean-Baptiste Kempf <[email protected]> committed on 2018/11/21 03:11

It is not an error to signal a temporal update while prev_segmap=NULL

76f56606 – Nathan E. Egge <[email protected]> authored on 2018/11/21 01:23

Rewrite the y4m header to match aomdec.

c5622d89 – Janne Grunau <[email protected]> authored on 2018/11/20 18:29

CI: use meson's '--werror' instead of CFLAGS

32accaa4 – James Almer <[email protected]> authored on 2018/11/20 14:16

dav1d_fuzzer: include config.h before other headers

c4be071d – Ronald S. Bultje <[email protected]> authored on 2018/11/20 11:32

Merge dav1d_picture_alloc() and dav1d_picture_alloc_copy()

b716083c – Janne Grunau <[email protected]> authored on 2018/11/19 15:13

fuzzer: mask all CPU flags under memory sanitizer

1b536ad4 – Janne Grunau <[email protected]> authored on 2018/11/20 15:05

mark dav1d_set_cpu_flags_mask visible

d0fa8ce9 – Janne Grunau <[email protected]> authored on 2018/11/19 14:55

common.h: prefix multiple inclusion guard with DAV1D_

ce258fa4 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/17 06:05

Normatively allow invalid global motion parameters

bb1a9352 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/20 02:56

Fix use of uninitialized memory in resize_c()

e8f5a93e – Janne Grunau <[email protected]> authored on 2018/11/20 03:53

film_grain: replace left shift of a signed value with a multiplication

5098b23a – Janne Grunau <[email protected]> authored on 2018/11/12 18:59

itx: clip according to spec, fixes #103, #158

c627f16f – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/20 03:28

Don't set LR coefficient defaults at image edges

86fd0b6d – Martin Storsjö <[email protected]> authored and Jean-Baptiste Kempf <[email protected]> committed on 2018/11/20 09:22

film_grain: Fix compilation with MSVC

acee4345 – Janne Grunau <[email protected]> authored on 2018/11/19 17:52

frame mt: mark frame as failed in dav1d_close()

b3c522d5 – Marvin Scholz <[email protected]> authored and Jean-Baptiste Kempf <[email protected]> committed on 2018/11/19 12:10

build: Only add libdav1d_nasm_objs if needed

9f77d9c3 – James Almer <[email protected]> authored on 2018/11/19 13:03

film_grain: include config.h before other headers

a194d478 – Ronald S. Bultje <[email protected]> authored and Luc Trudeau <[email protected]> committed on 2018/11/16 10:59

Slightly simplify the loop in lr_sbrow()

cfa986fe – Niklas Haas <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/11/13 11:53

film_grain: implement film grain synthesis

20e9f4df – Niklas Haas <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/11/13 11:52

picture: make the film grain metadata public

df5230ef – Niklas Haas <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/10/27 20:45

obu: parse uv_mult etc. as signed integers

5698bc91 – Nathan E. Egge <[email protected]> authored on 2018/11/18 13:32

Call msac_decode_bool() for 2 element CDFs.

9f812914 – Nathan E. Egge <[email protected]> authored and Nathan E. Egge <[email protected]> committed on 2018/11/14 05:09

Remove confusing math and redundant computations.

ecf72597 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/18 11:58

Clip resize height to image size

92020899 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/18 11:42

Don't initialize the LR values if LR is disabled for a plane

c496fab4 – Janne Grunau <[email protected]> authored on 2018/11/18 17:02

decode: check once per tile sb row for symbol decoder overreads

4aa0363a – Martin Storsjö <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/10/26 20:54

arm64: mc: Implement 8tap and bilin functions

842b2074 – James Almer <[email protected]> authored on 2018/11/17 07:03

obu: support frame_refs_short_signaling

58bcccc9 – Janne Grunau <[email protected]> authored on 2018/11/18 12:12

arm: define PIC based on __PIC__ or __pic__ if not defined

1d16d094 – Ronald S. Bultje <[email protected]> authored on 2018/11/17 04:26

Revert "Disable warped motion for invalid warped global motion parameters"

8bef2dc2 – Janne Grunau <[email protected]> authored on 2018/11/17 05:17

set chroma width/height correctly for 422/444 in dav1d_read_coef_blocks

f1d6eb93 – Janne Grunau <[email protected]> authored on 2018/11/14 09:38

CI: add asan/msan/ubsan test jobs for oss-fuzz regression testing

623132ac – Janne Grunau <[email protected]> authored on 2018/11/14 13:42

checkasm: fix undefined signed int left shift by 31 in loopfilter test

014dbe01 – Janne Grunau <[email protected]> authored on 2018/11/15 17:37

CI: change build-debian-werror job to debug build

31bf9aa1 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/16 10:56

Fix calculation of unit_idx to identify Av1RestorationUnit

242c35f3 – Ronald S. Bultje <[email protected]> authored on 2018/11/16 11:32

If Y/AC is 0 (but lossless=0), only allow coding DCT_DCT

0fdee4da – Ronald S. Bultje <[email protected]> authored on 2018/11/15 03:11

Add support for super-res

d27598e4 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/15 10:24

Clip unscaled output pixels in put_8tap_scaled_c()

14388c3c – Janne Grunau <[email protected]> authored on 2018/11/15 15:35

call dav1d_get_picture() until it returns -EAGAIN for draining

e890a66f – Janne Grunau <[email protected]> authored on 2018/11/15 16:40

frame mt: fix memleak caused by race between dav1d_close and dav1d_decode_frame

597a6eb9 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/15 03:38

Disable warped motion for invalid warped global motion parameters

ae1eb5f2 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/15 03:12

Use seg->globalmv/skip in determining is_comp

e5510796 – Ronald S. Bultje <[email protected]> authored on 2018/11/15 07:47

Don't clip in prep_8tap_scaled_c

4149c72b – James Almer <[email protected]> authored on 2018/11/15 06:58

obu: add support for OBUs with no size field

6c27b662 – Boyuan Xiao <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/11/09 08:43

Only do intra edge filtering/upsampling if enable_intra_edge set

2c3eaffd – Boyuan Xiao <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/11/13 12:06

Correct the condition for freeing references in dav1d_parse_obus

a1e945ca – Janne Grunau <[email protected]> authored on 2018/11/15 05:10

obu: tile_group and frame OBUs do not have trailing bits

a6b94ca9 – Janne Grunau <[email protected]> authored and Marvin Scholz <[email protected]> committed on 2018/11/14 10:52

meson: fix disabling asm for arm/arm64

949853f2 – Konstantin Pavlov <[email protected]> authored on 2018/11/14 18:12

CI: bump the dav1d-debian-unstable image version.

ca33a9b7 – Rupert Swarbrick <[email protected]> authored on 2018/11/14 12:15

Fix operator order in obu.c

a79cebf2 – Janne Grunau <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/11/14 05:27

dav1d_fuzzer: skip empty frames to avoid assertions after 026069693ef

c59f1940 – Rupert Swarbrick <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/10/17 13:49

Correctly flush at the end of OBUs

2532642b – Rupert Swarbrick <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/10/17 14:16

Fix how we read the UV quantization level

066b02c2 – Boyuan Xiao <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/10/15 12:37

Segmentation map reference logic

2f7eb1e9 – Boyuan Xiao <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/10/15 12:37

Fix parsing segmentation data in parse_frame_hdr

0bf59f09 – Ronald S. Bultje <[email protected]> authored and Janne Grunau <[email protected]> committed on 2018/11/13 15:09

Fix segmentation map size check

cf9ec49a – Janne Grunau <[email protected]> authored on 2018/11/14 08:58

mc: use width/height of reference frame in warp_affine

faa09008 – Janne Grunau <[email protected]> authored and Ronald S. Bultje <[email protected]> committed on 2018/11/14 07:21

mc: ensure order of evaluation of macro arguments in FILTER_BILIN

3fdb6cc9 – Ronald S. Bultje <[email protected]> authored on 2018/11/13 15:01

Support skip/globalmv/ref segmentation features

66adb457 – Janne Grunau <[email protected]> authored on 2018/11/14 06:49

mc: use PXSTRIDE consistently in {put,prep}_8tap_scaled_c

02606969 – James Almer <[email protected]> authored on 2018/11/08 09:32

implement a decoupled decode API

ec53ec6d – Ronald S. Bultje <[email protected]> authored on 2018/11/13 09:24

Add support for scaled references

e10b855c – David Michael Barr <[email protected]> authored and Henrik Gramner <[email protected]> committed on 2018/11/09 14:41

Pass dimensions to cfl_ac and derive log2sz

d401106b – Ronald S. Bultje <[email protected]> authored on 2018/11/13 06:51

Fix parsing of render size

7ef74ca1 – Ronald S. Bultje <[email protected]> authored on 2018/11/12 16:24

Align block size to bw/bh for z2 filter extension

4a646a0c – Janne Grunau <[email protected]> authored on 2018/11/12 19:33

intra_pred: fix ubsan issues with max_width/height in 2f251bd11528

8948e5af – Ronald S. Bultje <[email protected]> authored on 2018/11/12 11:58

Add missing .data suffix so we don't overwrite in memset()