shithub: dav1d

Info  •  Files  •  Log  •  Branches

36d615d1 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2019/10/22 09:57

x86: adapt SSSE3 wiener filter to SSE2

4866abab – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2019/10/21 07:12

x86: adapt SSSE3 warp_affine_8x8{,t} to SSE2

0526e1ea – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/22 13:10

arm: looprestoration: Fix register names in a comment

06ca5744 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/10 11:33

arm64: looprestoration: Minimal scheduling improvements

8b3985fd – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/22 08:24

arm64: looprestoration: Fix a typo

cf9146c3 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/21 19:24

arm64: looprestoration: Fix register references in comments

a3641268 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/08 05:40

arm64: looprestoration: Use ld2r instead of ld1+dup+dup

2eaabafc – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/07 19:21

arm64: looprestoration: Pass a correct height parameter to sgr_box3_h_neon for the top slice

14d4edcd – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/07 19:26

arm: looprestoration: Port the ARM64 SGR NEON assembly to 32 bit arm

3b33c52d – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/10/24 10:37

x86: Add minor ipred_z AVX2 optimizations

6c81623e – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/10/22 19:44

Shrink some stack buffers in the C versions of ipred_z

55951027 – Luc Trudeau <ltrudeau@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/22 15:50

Don't backup pixel if next block not "CDEFed"

47790541 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored and Henrik Gramner <gramner@twoorioles.com> committed on 2019/10/17 14:41

x86inc: fix LOAD_MM_PERMUTATION for AVX512

3e9f9676 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2019/10/12 21:30

x86: adapt SSSE3 cdef_filter_{4x4,4x8,8x8} to SSE2

11b72506 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2019/10/14 13:49

tools: fix SSE2 cpu masking

62fcd0cb – Martin Storsjö <martin@martin.st> authored on 2019/10/11 10:06

ci: Try switching two GCC based arm/aarch64 build configurations to debugoptimized

a6228f47 – Martin Storsjö <martin@martin.st> authored on 2019/10/11 09:17

arm64: ipred: Make sure all symbols are aligned

5f86e719 – Jean-Baptiste Kempf <jb@videolan.org> authored on 2019/10/11 06:57

Update news for 0.5.0: z2-avx2, ipred-neon and wiener-vsx

5d014b41 – Martin Storsjö <martin@martin.st> authored on 2019/10/10 19:16

arm: util: Split movrel into movrel and movrel_local

b7d7c8ce – Luc Trudeau <ltrudeau@twoorioles.com> authored on 2019/10/10 19:07

Check loopfilter levels prior to calling lf_mask

57dd0aae – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2019/10/01 05:52

arm64: ipred: NEON implementation of the cfl_ac functions

c7693386 – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2019/09/30 10:32

arm64: ipred: NEON implementation of the cfl_pred functions

d322d451 – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2019/09/26 05:46

arm64: ipred: NEON implementation of the filter function

4f14573c – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2019/09/25 17:50

arm64: ipred: NEON implementation of palette prediction

4318600e – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2019/09/24 11:40

arm64: ipred: NEON implementation of smooth prediction

8ab69afb – Martin Storsjö <martin@martin.st> authored and Janne Grunau <janne-vlc@jannau.net> committed on 2019/09/19 06:34

arm64: ipred: NEON implementation of paeth prediction

ea9fc9d9 – Henrik Gramner <gramner@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/09 19:05

x86: Add ipred_z2 AVX2 asm

afe901a6 – Henrik Gramner <gramner@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/09 19:14

Simplify ipred_z C code

dfadb6df – Henrik Gramner <gramner@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/09 19:05

checkasm: Improve ipred_z tests

a7c024ce – James Almer <jamrial@gmail.com> authored on 2019/10/09 19:09

x86: fix generate_grain_uv checkasm crashes on Windows x64

c688d5b2 – Jean-Baptiste Kempf <jb@videolan.org> authored on 2019/10/09 04:55

Update NEWS for 0.5.0

be60b142 – Michail Alvanos <malvanos@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/09 04:42

Add VSX wiener filter implementation

3e0f1508 – Jean-Baptiste Kempf <jb@videolan.org> authored on 2019/10/08 04:37

Move snap to package/ subfolder

61442bee – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/07 09:29

arm: mc: Port the ARM64 warp filter to arm32

5647a57e – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/07 08:24

arm64: mc: Use addp instead of addv+trn1 in warp

3489a9c1 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/02 09:41

arm: cdef: Port the ARM64 CDEF NEON assembly to 32 bit arm

32ae5dd0 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/07 06:27

arm: Support PIC loading of non-global symbols in the movrel macro on apple platforms

7bbc5e3d – Luc Trudeau <ltrudeau@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/08 03:52

Remove branch when changing bit in LR edges mask

dfaa2a10 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/05 04:29

arm64: cdef: Improve find_dir

fa6a0924 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/04 09:53

arm64: cdef: Calculate two initial parameters in the same vector

1f835750 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/03 07:31

arm64: cdef: Use loads with postincrement in more places in the padding function

bc26e300 – Martin Storsjö <martin@martin.st> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/01 19:05

arm64: cdef: Rewrite an expression slightly

d2c94ee1 – Luc Trudeau <ltrudeau@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/07 15:53

Don't backup pixels if next restoration unit is NONE

4e22ef3a – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/10/07 05:21

Add AVX2 version of generate_grain_uv (4:2:0)

ff41197b – Martin Storsjö <martin@martin.st> authored on 2019/10/04 19:07

arm64: mc: Schedule instructions better in the warp8x8 functions

e570088d – Luc Trudeau <ltrudeau@twoorioles.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/10/03 19:22

Check for RESTORATION_NONE once per frame

a4ceff6f – Martin Storsjö <martin@martin.st> authored on 2019/10/02 19:32

arm64: mc: Use sbfx instead of ubfx+sxth in the warp function

d4dfa85c – Henrik Gramner <gramner@twoorioles.com> authored on 2019/10/01 14:13

x86: Increase precision of SSSE3 IDCT intermediates

de561b3b – Henrik Gramner <gramner@twoorioles.com> authored on 2019/09/27 19:44

x86: Increase precision of AVX2 IDCT intermediates

f404c722 – Henrik Gramner <gramner@twoorioles.com> authored on 2019/10/01 19:34

checkasm: Add a function listing feature

16e0741a – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/10/01 14:43

Simplify README build instructions

f6a8cc0c – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/09/30 11:30

Minor cleanup

f7743da1 – Martin Storsjö <martin@martin.st> authored on 2019/09/17 20:07

arm64: ipred: NEON implementation of dc/h/v prediction modes

a91a03b0 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2019/09/18 05:25

x86: add warp_affine SSE4 and SSSE3 asm

713aa34c – Martin Storsjö <martin@martin.st> authored on 2019/09/28 19:37

arm64: itx: Fix overflows in idct

0ed3ad19 – Martin Storsjö <martin@martin.st> authored on 2019/09/29 13:34

arm64: itx: Consistently use the factor 2896 in adst

a4950bce – Martin Storsjö <martin@martin.st> authored on 2019/09/28 20:43

arm64: itx: Use smull+smlal instead of addl+mul

490a1420 – Niklas Haas <git@haasn.xyz> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/09/15 00:38

dav1dplay: initial support for --zerocopy

3f35ef1f – Niklas Haas <git@haasn.xyz> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/09/14 22:55

dav1dplay: add --untimed for benchmarking purposes

f6ae8c9c – Niklas Haas <git@haasn.xyz> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/09/14 22:40

dav1dplay: add --highquality to toggle render quality

c0865f35 – Victorien Le Couviour--Tuffet <victorien@videolan.org> authored on 2019/07/19 07:53

x86: add 32-bit support to SSSE3 deblock lpf

1e4e6c7a – Ronald S. Bultje <rsbultje@gmail.com> authored and Victorien Le Couviour--Tuffet <victorien@videolan.org> committed on 2019/04/17 06:11

x86: add deblocking loopfilters SSSE3 asm (64-bit)

556890be – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/09/10 07:00

AVX2 for chroma 4:2:0 film grain reconstruction

6d363223 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/09/05 13:03

Remove luma width check in fguv_32x32xn

99307bf3 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/09/05 10:42

Y grain AVX2 implementations

04ca7112 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/09/01 08:00

Add film grain checkasm tests

b9d4630c – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/09/01 07:18

Split out film grain block functions into a DSPContext

79c4aa95 – James Almer <jamrial@gmail.com> authored on 2019/09/05 20:50

obu: fix deriving render_width and render_height from reference frames

acad1a99 – Henrik Gramner <gramner@twoorioles.com> authored on 2019/09/04 21:55

Silence some clang-cl warnings

69dae683 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/09/05 18:50

x86: Fix buffer overead in mc put

a9315f5f – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/09/04 18:06

x86: Increase precision of the final inverse ADST transform stages

e2702eaf – Martin Storsjö <martin@martin.st> authored on 2019/09/02 19:13

arm64: itx: Do the final calculation of adst4/adst8/adst16 in 32 bit to avoid too narrow clipping

c0e1988b – Henrik Gramner <gramner@twoorioles.com> authored on 2019/09/04 10:31

Prefer __builtin_unreachable() over __assume() on clang-cl

666c71a0 – Henrik Gramner <gramner@twoorioles.com> authored on 2019/09/03 22:17

Fix clang-cl assertion warning

e65abadf – Janne Grunau <janne-vlc@jannau.net> authored and Martin Storsjö <martin@martin.st> committed on 2019/09/04 11:08

arm: Fix assembling with older binutils

863c3731 – Janne Grunau <janne-vlc@jannau.net> authored on 2019/09/01 20:46

TileContext: reorder scratch buffer to avoid conflicts

bfc9f72a – Janne Grunau <janne-vlc@jannau.net> authored on 2019/09/01 15:02

CI: use "needs:" to break the static build, test stage dependency

91b0af2f – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/08/30 03:38

Apply high-bitdepth adjustment of pixel index after delta calculation

1ffbeda0 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/08/29 17:26

Use linear interpolation for high bit-depth pixel values

c09f1072 – Ronald S. Bultje <rsbultje@gmail.com> authored and Jean-Baptiste Kempf <jb@videolan.org> committed on 2019/08/29 15:35

Fix bugs in film grain generation

cfd6fe6d – B Krishnan Iyer <krishnaniyer97@gmail.com> authored on 2019/08/29 16:07

arm: mc: Making code style consistent

f01bbbdd – Martin Storsjö <martin@martin.st> authored on 2019/08/29 10:17

arm: mc: Push fewer registers in w_mask

c3e5ad04 – Martin Storsjö <martin@martin.st> authored on 2019/08/29 10:08

arm: mc: Remove an unused instruction in w_mask

37a03fc7 – Ronald S. Bultje <rsbultje@gmail.com> authored on 2019/08/28 15:17

Check absolute tile positions in sb-to-tile_idx table generation

2c1467b4 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/28 13:07

Use 64-bit integers for warp_affine mvx/mvy calculations

eeca6f25 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/28 11:30

x86: Fix inverse ADST transform overflows

70b66ff1 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/20 14:59

Optimize coef ctx calculations

a62c445d – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/23 13:01

Consolidate horizontal scan tables

6757cab9 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/21 14:35

Change scan tables from int16_t to uint16_t

6751c980 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/09 18:41

Utilize the constraints in assertions to improve code generation

3d94fb9a – B Krishnan Iyer <krishnaniyer97@gmail.com> authored on 2019/08/01 18:45

arm64: mc: NEON implementation of w_mask_444/422/420 function

1dc2dc7d – B Krishnan Iyer <krishnaniyer97@gmail.com> authored on 2019/07/22 19:20

arm64: mc: NEON implementation of blend, blend_h and blend_v function

d20d70e8 – Michael Bradshaw <mjbshaw@google.com> authored and Henrik Gramner <gramner@twoorioles.com> committed on 2019/08/13 22:12

Prefer `do {} while (0);` over `while (0);`

0f4edbff – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/07 10:47

Cosmetics: CDF tables

61dcd11b – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/10 10:34

x86: Add an msac function for coefficient hi_tok decoding

e29fd5c0 – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/06 11:17

Add msac optimizations

a819653e – Henrik Gramner <gramner@twoorioles.com> authored and Henrik Gramner <henrik@gramner.com> committed on 2019/08/08 21:45

Remove unused CDF:s

dff0a08c – James Almer <jamrial@gmail.com> authored on 2019/08/10 11:50

dav1dplay: abort if no input filename is provided

3a77c57b – James Almer <jamrial@gmail.com> authored on 2019/08/09 12:56

meson: move dav1dplay to a new examples section

42ea146f – Luc Trudeau <ltrudeau@twoorioles.com> authored and Henrik Gramner <gramner@twoorioles.com> committed on 2019/07/29 08:12

decode_coefs reuse lossless variable