Ramiro Polla c9977acbc6 swscale/ops: clear range values SWS_OP_{MIN,MAX} ... This gives partial range values on conversions with floats, after the values have been clamped. gbrpf32be -> rgb8: [f32 XXXX -> zzzX] SWS_OP_READ : 3 elem(s) planar >> 0 [f32 ...X -> ...X] SWS_OP_SWAP_BYTES [f32 ...X -> ...X] SWS_OP_SWIZZLE : 2013 [f32 ...X -> ...X] SWS_OP_LINEAR : diag3 [[7 0 0 0 0] [0 7 0 0 0] [0 0 3 0 0] [0 0 0 1 0]] [f32 ...X -> ...X] SWS_OP_DITHER : 16x16 matrix + {0 3 2 5} [f32 ...X -> ...X] SWS_OP_MAX : {0 0 0 0} <= x + min: {0, 0, 0, _}, max: {nan, nan, nan, _} [f32 ...X -> ...X] SWS_OP_MIN : x <= {7 7 3 _} + min: {0, 0, 0, _}, max: {7, 7, 3, _} [f32 ...X -> +++X] SWS_OP_CONVERT : f32 -> u8 + min: {0, 0, 0, _}, max: {7, 7, 3, _} [ u8 ...X -> +XXX] SWS_OP_PACK : {3 3 2 0} - min: {0, _, _, _}, max: {0, _, _, _} + min: {0, _, _, _}, max: {255, _, _, _} [ u8 .XXX -> +XXX] SWS_OP_WRITE : 1 elem(s) packed >> 0 - min: {0, _, _, _}, max: {0, _, _, _} + min: {0, _, _, _}, max: {255, _, _, _} (X = unused, z = byteswapped, + = exact, 0 = zero) Sponsored-by: Sovereign Tech Fund Signed-off-by: Ramiro Polla <ramiro.polla@gmail.com>		2026-02-24 20:25:59 +01:00
..
acodec	avcodec/adpcm: improve decooding output for 4-bit ADPCM IMA WAV	2025-09-22 23:46:29 +00:00
fate	swscale/ops: clear range values SWS_OP_{MIN,MAX}	2026-02-24 20:25:59 +01:00
lavf	avformat/matroskadec: take into account priming samples in audio tracks	2026-02-05 23:21:49 -03:00
lavf-fate	lavc/av1_parser: Extract SAR from render_size	2026-01-14 23:56:39 +00:00
pixfmt	swscale/range_convert: fix truncation bias in range conversion	2025-11-06 20:36:08 +00:00
seek	avformat/pcmdec: Mark raw PCM stream as having no timestamps	2026-01-05 01:23:17 +00:00
vsynth	fate/vcodec: Test median prediction in ffvhuff,huffyuv	2025-12-14 10:13:29 +01:00