Ramiro Polla bf743bb9e0 swscale/ops_memcpy: also use bulk memset/memcpy for negative strides ...

The operations themselves have already been generalized to support
this, so all we need to do is modify the base input/output pointers.

2026-11-05 19:43:24 +00:00

..

aarch64

swscale/aarch64: fix signed load in yuv2planeX_10

2026-11-03 21:19:50 +00:00

arm

swscale/arm: align unscaled NEON YUV to RGB slices to source chroma

2026-10-20 13:39:20 +00:00

loongarch

swscale/loongarch: add LSX fast bilinear horizontal scale

2026-10-09 10:24:34 +00:00

ppc

swscale/ppc: build the AltiVec and VSX files only when enabled

2026-11-02 07:24:04 +00:00

riscv

swscale/riscv: fix uyvy/yuyv to yuv422p on odd widths

2026-09-27 12:06:07 +00:00

tests

swscale/tests: cover chroma range conversion with FATE references

2026-10-29 03:10:43 +00:00

vulkan

swscale/uops: factor out ff_sws_uop_list_print()

2026-10-29 15:47:35 +00:00

x86

swscale/uops: check every product of an input before fusing it

2026-11-02 18:19:15 +00:00

.gitignore

Add missing .gitignore.

2026-08-17 22:21:44 +00:00

alphablend.c

…

bayer_template.c

…

cms.c

avutil/slicethread: add return value to slicethread callbacks

2026-08-14 10:41:40 +02:00

cms.h

…

csputils.c

swscale/csputils: derive the RGB to XYZ matrix without dividing by y

2026-10-28 19:22:28 +00:00

csputils.h

swscale/uops: add SWS_UOP_LUT_3D reference implementation

2026-08-03 09:32:30 +00:00

filters.c

swscale/filters: add ability to set a virtual output size

2026-06-20 03:08:21 +02:00

filters.h

swscale/filters: add ability to set a virtual output size

2026-06-20 03:08:21 +02:00

format.c

swscale: compile ops code unconditionally

2026-10-30 10:55:13 +00:00

format.h

swscale: build the X'Y'Z' tables from the transfer functions of lavu

2026-10-28 19:22:28 +00:00

framepool.c

avutil/framepool: provide a single copy for static builds

2026-10-26 16:53:24 +00:00

gamma.c

…

graph.c

swscale: compile ops code unconditionally

2026-10-30 10:55:13 +00:00

graph.h

swscale/graph: add plane copy metadata

2026-08-09 21:03:33 +02:00

half2float.c

…

hscale_fast_bilinear.c

swscale: avoid overflow in fast bilinear edge handling

2026-08-05 19:34:30 +00:00

hscale.c

…

input.c

swscale/input: fix rgbf32 input (add half input functions)

2026-06-05 22:33:58 +02:00

jit.c

swscale/jit: Fix getting MAP_ANONYMOUS defined on older glibc

2026-09-08 12:03:06 +00:00

jit.h

swscale/jit: move jit-related code to its own file

2026-07-16 08:32:50 +00:00

libswscale.v

…

log2_tab.c

…

lut3d.c

swscale/lut3d: simplify and hard-code 3DLUT format

2026-08-03 09:32:30 +00:00

lut3d.h

swscale/lut3d: simplify and hard-code 3DLUT format

2026-08-03 09:32:30 +00:00

Makefile

swscale: build and run ops tests unconditionally

2026-10-30 10:55:13 +00:00

op_list_gen_template.c

swscale/ops: use slice threading for ff_sws_enum_op_lists()

2026-08-14 10:41:40 +02:00

ops_chain.c

swscale/ops_chain: remove unused SwsOp setup helpers

2026-10-02 14:11:50 +02:00

ops_chain.h

swscale/ops_chain: remove unused SwsOp setup helpers

2026-10-02 14:11:50 +02:00

ops_dispatch.c

swscale/ops: rename ff_sws_op_list_update_comps()

2026-10-20 21:33:01 +00:00

ops_dispatch.h

swscale/ops_dispatch: add SwsOpBackend.compile_uops()

2026-07-14 12:24:01 +00:00

ops_internal.h

swscale/ops_internal: make pack decoding helper take the pattern directly

2026-10-24 19:10:48 +00:00

ops_memcpy.c

swscale/ops_memcpy: also use bulk memset/memcpy for negative strides

2026-11-05 19:43:24 +00:00

ops_optimizer.c

swscale/ops_optimizer: mark split scaling intermediates progressive

2026-10-25 10:11:09 +00:00

ops.c

swscale/ops: clamp lev_extra >= lev

2026-10-29 15:47:35 +00:00

ops.h

swscale/ops: rename ff_sws_op_list_update_comps()

2026-10-20 21:33:01 +00:00

options.c

swscale: remove GLSL backend

2026-07-13 18:57:37 +09:00

output.c

swscale/output: write the last partial byte in the mono error-diffusion paths

2026-10-19 00:51:53 +00:00

rational64.c

swscale/rational64: use 64-bit math whenever possible

2026-09-30 11:51:24 +00:00

rational64.h

swscale/rational64: replace av_ prefix with ff_

2026-09-24 11:18:49 -03:00

rgb2rgb_template.c

libswscale: handle any width in the planar to packed YUV converters

2026-10-18 22:45:57 +00:00

rgb2rgb.c

swscale/rgb2rgb: Remove set-but-unused functions

2026-03-01 23:45:11 +00:00

rgb2rgb.h

libswscale: handle any width in the planar to packed YUV converters

2026-10-18 22:45:57 +00:00

slice.c

…

swscale_internal.h

swscale/ppc: check the CPU flags outside the AltiVec files

2026-11-02 07:24:04 +00:00

swscale_unscaled.c

swscale/loongarch: add unscaled NV12/NV21 to 32-bit RGB conversions

2026-10-09 10:24:34 +00:00

swscale.c

configure: make swscale_vulkan depend on unstable

2026-10-30 10:55:13 +00:00

swscale.h

swscale: remove GLSL backend

2026-07-13 18:57:37 +09:00

swscaleres.rc

…

uops_backend.c

swscale/uops: factor out ff_sws_uop_list_print()

2026-10-29 15:47:35 +00:00

uops_list.h

swscale/uops: add SWS_UOP_LUT_3D reference implementation

2026-08-03 09:32:30 +00:00

uops_macros_gen.c

swscale/uops: add SWS_UOP_FLAG_PUNNING

2026-10-27 18:53:16 +00:00

uops_macros.h

swscale/uops: correctly mark masked linear rows as identity rows

2026-10-27 18:53:16 +00:00

uops_tmpl.c

swscale/uops: round every float operation of the reference backend

2026-11-02 18:19:15 +00:00

uops_tmpl.h

swscale/uops: do not request loop vectorization under UBSan

2026-09-24 18:08:16 +00:00

uops.c

swscale/uops: check every product of an input before fusing it

2026-11-02 18:19:15 +00:00

uops.h

swscale/uops: round every float operation of the reference backend

2026-11-02 18:19:15 +00:00

utils.c

swscale: build the X'Y'Z' tables from the transfer functions of lavu

2026-10-28 19:22:28 +00:00

version_major.h

libs: Bump major version of all libraries

2026-06-23 19:15:57 +02:00

version.c

…

version.h

Bump minor after branching release/9.1

2026-10-24 17:05:20 -03:00

vscale.c

…

yuv2rgb.c

swscale/yuv2rgb: add explicit BE/LE 565/555 cases

2026-06-10 17:54:20 +00:00