shithub: dav1d

Info  •  Files  •  Log  •  Branches

bce8fae9 – Janne Grunau <janne-vlc@jannau.net> authored on 2020/03/04 17:26

CI: run a selection of jobs on a node with avx2

3a6a55d8 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/03/04 18:05

x86: Fix crash in AVX2 cdef_filter with <32-byte stack alignment

fb348f64 – Martin Storsjö <martin@martin.st> authored on 2020/03/03 11:21

arm64: mc: NEON implementation of blend for 16bpc

52e9b435 – Martin Storsjö <martin@martin.st> authored on 2020/03/04 05:51

arm: mc: Optimize blend_v

a7f6fe32 – Martin Storsjö <martin@martin.st> authored on 2020/03/03 10:31

arm64: mc: Treat the stride as a full 64 bit (potential signed) value in blend_8bpc_neon

48ffb05e – Martin Storsjö <martin@martin.st> authored on 2020/03/03 09:51

arm64: mc: Fix indentation

83c62716 – Martin Storsjö <martin@martin.st> authored on 2020/03/03 09:49

arm64: mc: Use more intuitive lane specifications for loads/stores

f4dac1a3 – Jean-Baptiste Kempf <jb@videolan.org> authored on 2020/02/25 05:21

Update NEWS for 0.6.0

abaad816 – Janne Grunau <janne-vlc@jannau.net> authored on 2020/03/02 19:16

CI/armv7: use `linux32 meson ...` to allow running on aarch64

360243c2 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2020/02/17 19:10

arm64: loopfilter: NEON implementation of loopfilter for 16 bpc

ebbf91f4 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2020/02/17 19:13

arm: loopfilter: Prepare for 16 bpc

ac492552 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2020/02/24 19:08

arm: loopfilter: Fix a comment

d398da88 – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2020/02/16 19:38

fuzzing: link the fuzzing binaries as C++

7675eb16 – Janne Grunau <janne-vlc@jannau.net> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2020/02/16 19:30

fuzzing: split the fuzzing targets to their own meson.build file

64f9db55 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/02/20 10:25

x86: Add mc w_mask 4:4:4 AVX-512 (Ice Lake) asm

d4a7c647 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/02/20 10:25

x86: Add mc w_mask 4:2:2 AVX-512 (Ice Lake) asm

50e9a39a – Henrik Gramner <gramner@twoorioles.com> authored on 2020/02/20 10:25

x86: Add mc w_mask 4:2:0 AVX-512 (Ice Lake) asm

d085424c – Henrik Gramner <gramner@twoorioles.com> authored on 2020/02/20 10:25

x86: Add mc avg/w_avg/mask AVX-512 (Ice Lake) asm

22080aa3 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2020/02/14 10:46

x86: optimize cdef_filter_{4x{4,8},8x8}_avx2

1bd078c2 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2020/02/12 09:25

x86: add a seperate fully edged case to cdef_filter_avx2

efbdf7a0 – Henrik Gramner <gramner@twoorioles.com> authored and Victorien Le Couviour--Tuffet <victorien@videolan.org> committed on 2020/02/19 15:49

checkasm: Improve the cdef input randomization algorithm

296d1dc0 – Luc Trudeau <ltrudeau@twoorioles.com> authored on 2020/02/21 05:23

cli: Replace malloc + memset(0) with calloc in input.c

cacc8e35 – Luc Trudeau <ltrudeau@twoorioles.com> authored on 2020/02/21 05:01

cli: remove init_[de]muxers() functions

0c885607 – Luc Trudeau <ltrudeau@twoorioles.com> authored on 2020/02/20 16:01

Replace malloc+memset(0) with calloc

bf56afde – Janne Grunau <janne-vlc@jannau.net> authored on 2020/02/18 16:45

CI: update aarch64 docker image to buster with meson 0.49

b33f46e8 – Martin Storsjö <martin@martin.st> authored on 2020/02/12 19:23

arm: cdef: Do an 8 bit implementation for cases with all edges present

aff9a210 – Martin Storsjö <martin@martin.st> authored on 2020/02/14 18:48

arm32: cdef: Fix a typo for consistency

09d90658 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/02/15 09:52

cli: Implement line buffering in print_stats()

eb7077ed – Martin Storsjö <martin@martin.st> authored on 2020/02/12 18:42

arm: cdef: Remove leftover unused labels and macro parameters

e3dbf926 – Martin Storsjö <martin@martin.st> authored on 2020/02/10 05:03

arm64: looprestoration: NEON implementation of SGR for 10 bpc

7cf5d753 – Martin Storsjö <martin@martin.st> authored on 2020/02/09 18:39

arm64: looprestoration: Prepare for 16 bpc by splitting code to separate files

32e265a8 – Martin Storsjö <martin@martin.st> authored on 2020/02/10 05:02

arm: looprestoration: Add 8bpc to existing function names, add HIGHBD_*_SUFFIX

96da9cc2 – Martin Storsjö <martin@martin.st> authored on 2020/02/04 16:03

looprestoration: Add a bpc parameter to the init func

8fb30657 – Martin Storsjö <martin@martin.st> authored on 2020/02/09 18:14

arm: looprestoration: Improve scheduling in box3/5_h slightly

8e8fb84d – Martin Storsjö <martin@martin.st> authored on 2020/02/05 05:17

arm: Use int16_t for the tmp intermediate buffer

feeaf785 – Martin Storsjö <martin@martin.st> authored on 2020/02/06 18:45

arm: looprestoration: Fix a comment

e4208e85 – Jean-Baptiste Kempf <jb@videolan.org> authored on 2020/02/10 16:19

NEWS: Official naming is AVX2, not AVX-2

d4c5ad49 – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2020/02/01 09:34

arm64: mc: Reduce the width of a register copy

b1167ce1 – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2020/02/01 09:33

arm64: mc: Use two regs for alternating output rows for w4/8 in avg/w_avg/mask

0bad117e – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2020/02/01 09:33

arm64: mc: Simplify avg/w_avg/mask by always using the w16 macro

2e68c1f3 – Jean-Baptiste Kempf <jb@videolan.org> authored on 2020/02/09 09:35

Update NEWS for 0.6.0

8974c155 – Martin Storsjö <martin@martin.st> authored on 2020/02/06 04:36

arm64: mc: NEON implementation of warp for 16 bpc

e6cebeb7 – Martin Storsjö <martin@martin.st> authored on 2020/02/06 04:10

arm64: cdef: Add NEON implementations of CDEF for 16 bpc

1d5ef8df – Martin Storsjö <martin@martin.st> authored on 2020/02/06 04:09

arm: cdef: Prepare for 16bpc

19ce77e0 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/01/28 12:42

x86: Add cdef_filter_4x4 AVX-512 (Ice Lake) asm

58a4ba07 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/02/05 22:43

Reorder the Dav1dFrameHeader struct to fix alignment issues

c89eb564 – Martin Storsjö <martin@martin.st> authored on 2020/02/05 05:43

arm64: looprestoration: NEON implementation of wiener filter for 16 bpc

c2a2e6ee – Martin Storsjö <martin@martin.st> authored on 2020/02/05 05:43

arm: looprestoration: Fix the wiener C wrapper function for high bitdepths

5bc8a500 – Martin Storsjö <martin@martin.st> authored on 2020/02/05 05:43

arm: looprestoration: Prepare for 16bpc wiener filter by adding _8bpc to function names

2653292c – Martin Storsjö <martin@martin.st> authored on 2020/02/05 05:43

arm: looprestoration: Clarify a comment

fe44861b – Martin Storsjö <martin@martin.st> authored on 2020/02/04 06:50

arm64: mc: NEON implementation of put/prep 8tap/bilin for 16 bpc

c851c65c – kossh1 <jonvito16@gmail.com> authored and Ronald S. Bultje <rsbultje@gmail.com> committed on 2020/02/05 14:41

Update README

3c8110a9 – Lynne <dev@lynne.ee> authored on 2020/01/13 06:06

x86/msac: add an avx2 version for msac_decode_symbol_adapt16

35ab85bb – Lynne <dev@lynne.ee> authored on 2020/01/13 06:05

msac: make symbol_adapt16 a function pointer on x86_64

03511f8c – Martin Storsjö <martin@martin.st> authored on 2020/01/31 17:35

arm64: mc: NEON implementation of avg/mask/w_avg for 16 bpc

a285204a – Martin Storsjö <martin@martin.st> authored on 2020/01/31 17:35

arm: mc: Prepare the init file for higher bitdepths

7eaa7c9f – Martin Storsjö <martin@martin.st> authored on 2020/02/01 18:17

arm: Make the existing 8bpc assembly only built if 8bpc is enabled

b8399319 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/02/03 18:56

x86: Avoid cmov instructions that depends on multiple flags

d21dc801 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/02/03 18:56

x86: Add miscellaneous minor scalar optimizations

81a26458 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/02/03 18:55

x86: Use unsigned pointer comparisons

bb178db0 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/01/29 09:17

Rework the CDEF top edge handling

dccc21b7 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/01/31 12:57

checkasm: Fix missing shift in high bit-depth cdef_filter test

fbc1b420 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/02/01 09:17

Avoid masking the lsb in high bit-depth stride calculations

9c29f229 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/01/24 15:34

checkasm: Increase buffer alignment to 64-byte on x86-64

361a3c8e – Martin Storsjö <martin@martin.st> authored on 2020/01/28 06:07

arm: cdef: Add special cased versions for pri_strength/sec_strength being zero

6ad9bd5f – Martin Storsjö <martin@martin.st> authored on 2020/01/28 05:30

arm: cdef: Fix some comment typos

ba23ac8c – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/28 18:49

Fix crash in dav1d_apply_grain() with negative picture strides

eaaf2218 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/27 12:42

Optimize the cdef_filter C implementation

fad6db20 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/27 12:41

checkasm: Improve cdef_filter test

6385cde2 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/24 14:40

Avoid redundant calls to CDEF DSP functions

447b7c63 – Henrik Gramner <gramner@twoorioles.com> authored on 2020/01/21 19:10

x86: Bump nasm version requirement to 2.14

e636a2f4 – Konstantin Pavlov <thresh@videolan.org> authored on 2020/01/21 16:25

CI: Use a newer image to build snap packages

e706fac9 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored and Victorien Le Couviour--Tuffet <psilokos@twoorioles.com> committed on 2019/12/11 19:25

x86: add prep_8tap AVX512 asm

b83cb964 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored and Victorien Le Couviour--Tuffet <psilokos@twoorioles.com> committed on 2020/01/17 08:42

x86: replace "mov hb, Xb" by "movzx hd, Xb" in MC

289ca2ce – Victorien Le Couviour--Tuffet <psilokos@twoorioles.com> authored on 2020/01/20 05:35

x86inc: save xmm_regs_used in spill_xmm for non-win64

010eae8b – Martin Storsjö <martin@martin.st> authored on 2020/01/15 07:41

arm64: itx: Fix overflow/clipping in negation in idct16

ef64567e – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/14 16:53

x86: Fix overflows in SSSE3 idct

3a3af969 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/14 15:52

x86: Fix missing saturations in inverse identity asm

8ff89463 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/12/21 11:06

SSSE3 implementations of film grain

e79e5ceb – Dale Curtis <dalecurtis@chromium.org> authored on 2020/01/10 08:57

Reduce scope of NO_SANITIZE usage

c192e0db – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/09 12:29

Add a workaround for -fsanitize=cfi + dlsym() issue

5462c2a8 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored and Ronald S. Bultje <rsbultje@gmail.com> committed on 2019/12/11 19:24

x86: add prep_bilin AVX512 asm

40891aab – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored and Ronald S. Bultje <rsbultje@gmail.com> committed on 2019/12/18 11:01

x86: add avx512icl cpu flag to x86inc.asm

430967a6 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored and Ronald S. Bultje <rsbultje@gmail.com> committed on 2019/10/29 12:19

checkasm: x86: ensure all SIMD lanes are turned on at all times

a4721225 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/07 19:44

Add misc. inverse transform C optimizations

c37b5ee3 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/07 19:44

Skip clipping in the inverse wht transform C implementation

a7ca7b22 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/07 19:44

x86: Fix SSSE3 inverse identity transform overflow/clipping

f16b43cd – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2020/01/07 19:43

x86: Fix AVX2 inverse identity transform overflow/clipping

50220456 – Xavier Claessens <xavier.claessens@collabora.com> authored and Xavier Claessens <xclaesse@gmail.com> committed on 2019/12/30 14:58

Fix building as a meson subproject

115fe773 – Dale Curtis <dalecurtis@google.com> authored and Dale Curtis <dalecurtis@chromium.org> committed on 2020/01/07 18:56

Fix missing include for limits.h

8d574f70 – Martin Storsjö <martin@martin.st> authored on 2020/01/04 19:41

arm64: msac: Avoid 32 bit intermediates in symbol_adapt

9f084b0d – Martin Storsjö <martin@martin.st> authored on 2020/01/02 10:43

arm64: itx: Use sqrdmulh in the preexisting identity transform functions

e36088e4 – Martin Storsjö <martin@martin.st> authored on 2020/01/01 18:12

arm64: itx: Specialcase transforms with identity in the first pass with downshift

33e65d80 – Martin Storsjö <martin@martin.st> authored on 2020/01/02 02:58

arm64: itx: Adjust .irp in the 4x16/16x4/8x16/16x8 functions

4504ae3f – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/12/31 14:46

Don't interleave the skip mode index finding loops

3d166b97 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2020/01/01 12:42

Prevent shift by >= 32

acadacfa – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/12/31 14:46

Take lossless into account when assigning loopfilter strength

de6e3170 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/12/31 07:15

Deal with chroma coefficients that are exactly 0x100000

f306f969 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/12/30 23:14

x86: Fix inverse (flip)adst 8x4 clipping

81aba493 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/12/30 23:13

Fix C inverse ADST clipping