[Ffmpeg-devel-irc] ffmpeg-devel.log.20171026

[burek]

[04:36:04 CEST] <cone-773> ffmpeg 03Zhong Li 07master:984b882b769c: fate: fix mpeg2-ticket6677 faillures on some platforms
[05:33:59 CEST] <rcombs> huh, I think https://trac.ffmpeg.org/ticket/6778#comment:1 is the same as an issue I'm investigating where dts/pts only advances by 960 after encoding a 1024-sample packet (with time_base==sample_rate)
[05:39:02 CEST] <rcombs> https://gist.github.com/4be2a14b70a2eb4252ac73fa984f7d6d thanks matroska timestamps
[05:40:00 CEST] <rcombs> packet durations indicate 480-sample packets; they're actually 512
[05:41:47 CEST] <rcombs> and wow this interleave is bad
[05:51:41 CEST] <cone-773> ffmpeg 03James Almer 07master:6bd665b7c579: avcodec/tak: remove GetBitContext usage from avpriv_tak_parse_streaminfo()
[05:51:42 CEST] <cone-773> ffmpeg 03James Almer 07master:ae100046ca32: avcodec/exif: remove GetByteContext usage from avpriv_exif_decode_ifd()
[08:31:41 CEST] <cone-773> ffmpeg 03Tobias Rapp 07master:a07ac9302335: avformat/wavenc: skip writing incorrect peak-of-peaks position value
[11:10:34 CEST] <arpu> [Parsed_amovie_0 @ 0x2044320] EOF timestamp not reliable 
[11:10:38 CEST] <arpu> what does this means? 
[11:10:54 CEST] <arpu> streams breaks after this message
[11:12:44 CEST] <arpu> ok end of file :>
[11:12:59 CEST] <arpu> i use a amovie with loop 
[11:13:14 CEST] <arpu> looks like the loop is ended ...
[13:33:31 CEST] <J_Darnley> Gramner: will you give me permission adapt the AVX-512 cpu detection code from x264 into FFmpeg?  Noting the change from GPL to LGPL?
[17:08:19 CEST] <atomnuker> J_Darnley: x86inc is released under the ISC license, I don't think there's a need to ask
[17:14:21 CEST] <J_Darnley> Not the abstraction layer
[17:14:21 CEST] <J_Darnley> cpuid detection
[17:23:46 CEST] <Gramner> J_Darnley: yes, I'm ok with LGPL:ing the avx-512 cpu detection code from x264
[17:25:35 CEST] <J_Darnley> Okay, thank you very much
[17:25:38 CEST] <thardin> how is animated webp different from webm?
[17:25:47 CEST] <thardin> what are even webdevs doing
[17:25:57 CEST] <J_Darnley> No audio permitted?
[17:26:15 CEST] <J_Darnley> Max of 15fps?
[17:26:56 CEST] <jamrial> thardin: same as png vs animated png :p
[17:27:52 CEST] <thardin> maybe there's less expectation to retain sync
[17:30:11 CEST] <thardin> I can only guess it's OK to pause gifs and such when out of view
[17:32:16 CEST] <thardin> webp is 4:2:0 only? yikes
[17:33:27 CEST] <jamrial> thardin: webp is vp8, so yeah
[17:33:47 CEST] <jamrial> there's also a lossless mode, which is rgb
[17:34:13 CEST] <thardin> so jpeg is still better if you have high chroma contrast
[17:34:15 CEST] <jamrial> and it can have alpha channel, so also rgba and yuva420
[17:34:20 CEST] <atomnuker> J_Darnley: are you going to write some avx512 asm?
[17:34:36 CEST] <thardin> so close but still managing to mess it up
[17:35:46 CEST] <jamrial> maybe they will update the spec to use vp9 or av1 eventually
[17:36:07 CEST] <J_Darnley> atomnuker: yes
[17:36:20 CEST] <thardin> I'm a bit amazed at the flexibility of the jpeg spec. I wonder if any decoder supports all its features
[17:36:53 CEST] <J_Darnley> It will probably just be a length extension of existing avx2 first then maybe I will checkout newer instructions.
[17:39:16 CEST] <atomnuker> J_Darnley: cool, what for?
[17:45:24 CEST] <kierank> atomnuker: v210 and friends of course
[17:48:01 CEST] <atomnuker> michaelni: what do *bits, *codes and *symbols mean in init_vlc?
[17:48:43 CEST] <BBB> thardin: I believe webp is intra-only, no?
[17:48:58 CEST] <BBB> thardin: so animated webp would be intra-coded images, it wouldnt actually have any inter coding
[17:49:17 CEST] <michaelni> atomnuker, length of codewords, codeword length and int value associted with each codeword IIRC
[17:49:49 CEST] <michaelni> s/codeword length/codeword bits/
[17:50:13 CEST] <thardin> BBB: sounds like that would be worse than gif even
[17:50:25 CEST] <thardin> for the kind of graphics gif is intended for
[17:50:30 CEST] <BBB> I dont think the goal of animated webp is to compress well
[17:50:37 CEST] <BBB> if you wanted compression, youd use webm
[17:51:04 CEST] <thardin> I guess
[17:54:22 CEST] <thardin> I wonder if any RGBA / YUVA formats use alpha prediction
[17:55:10 CEST] <atomnuker> michaelni: what's the difference between nb_bits and bits_size?
[17:57:33 CEST] <atomnuker> oh is bits/codes_size the number of bits per each entry of *bits and *codes?
[17:58:06 CEST] <atomnuker> (so the stride basically)
[18:00:04 CEST] <michaelni> atomnuker, see ff_init_vlc_sparse() in libavcodec/bitstream.c that documents all the parameters
[18:28:04 CEST] <atomnuker> btw michaelni, could you upload this file to samples? I'd like to add a fate test for it
[18:28:05 CEST] <atomnuker> https://pars.ee/temp/zoneplate_1920x1080_legall_32x8x3_medium_1frames_1threads.vc2
[18:28:16 CEST] <atomnuker> (should go in the dirac directory)
[18:28:49 CEST] <atomnuker> actually nevermind
[18:28:55 CEST] <atomnuker> we already have a test for it
[18:29:07 CEST] <atomnuker> (in fact a whole bunch in the vsynth stuff)
[18:30:40 CEST] <cone-366> ffmpeg 03Carl Eugen Hoyos 07master:f2c867051c8f: lavc/avcodec: Constify the return value of av_bitstream_filter_next().
[18:33:01 CEST] <cone-366> ffmpeg 03Mateusz 07master:a4743d257425: avformat/yuv4mpeg: add gray9/10/12 support
[18:39:37 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:bf6cd808bed4: avcodec/pngdec: Clean up on av_frame_ref() failure
[18:39:38 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:4e341bd90411: avcodec/svq3: Fix overflow in svq3_add_idct_c()
[18:39:39 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:6d1ebb9def18: avcodec/ffv1dec: Fix integer overflow in read_quant_table()
[18:39:40 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:940659036f47: avcodec/dirac_dwt: Fix integer overflow in COMPOSE_FIDELITYi*()
[18:39:41 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:b5b52c0ca726: avcodec/takdec: Fix integer overflows in decode_subframe()
[18:39:42 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:4bc16930efdb: avcodec/proresdec2: Check bits in DECODE_CODEWORD(), fixes invalid shift
[18:39:43 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:ab845587d135: avcodec/takdec: Fix integer overflow in decode_lpc()
[18:39:44 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:c54317a17e57: avcodec/jpeg2000: Check that codsty->log2_prec_widths/heights has been initialized
[18:39:45 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:520daf8c0ec4: avcodec/hevcdsp_template: Fix undefined shift
[18:39:46 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:6b9ffcdb2bac: avcodec/proresdec2: SKIP_BITS() does not work with len=32
[18:39:47 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:4768b30b5bdf: avcodec/aacdec_template: Clear tns present flag on error
[18:39:48 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:ab43bc50c017: avcodec/truemotion2: Fix integer overflows in tm2_high_chroma()
[18:39:49 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:0a231e7dd32b: avcodec/x86/lossless_videoencdsp: Fix handling of small widths
[18:39:50 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:7cc854ce1531: avcodec/x86/lossless_videoencdsp: Fix warning: signed dword value exceeds bounds
[18:39:51 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:0009272f943a: avcodec/mpeg4videodec: Use 64 bit intermediates for sprite delta
[18:39:52 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:aa6c44c33399: avcodec/mpeg_er: Clear mcsel in mpeg_er_decode_mb()
[18:39:53 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:d13d3feba2d0: avcodec/dirac_dwt: Fix integer overflow in COMPOSE_53iL0()
[18:39:54 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:48933f28c2a8: avcodec/ffv1dec: Fix out of array read in slice counting
[18:39:55 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:20fd9217d8ae: avcodec/pafvideo: Check for bitstream end in decode_0()
[18:39:56 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:e8fd32b69ffd: avcodec/snowdec: Check mv_scale
[18:39:57 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:0ccb27e09427: ffserver: Fix off by 1 error in path
[18:39:58 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:b33434ec62a4: tests/ffserver.regression.ref: update checksums to what ffserver currently produces
[18:39:59 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.3:60b385a5bf9f: Update for 3.3.5
[18:45:14 CEST] <jamrial> michaelni: the id3v2 patch by Fredrik Hubinette needs to be applied to 3.3 before you tag the point release. i'll push it in a bit
[18:45:46 CEST] <michaelni> jamrial, ok
[18:49:26 CEST] <Gramner> J_Darnley: be aware that using zmm registers induces significant frequency drops which reduces performance of everything else, so if you want to use 512-bit vectors you better go all in on it to make up for it. you probably want to spend at least 20-30% of overall runtime in avx-512 code
[18:50:00 CEST] <Gramner> the alternative is to stay in 256-bit mode and just leverage new instructions and opmasks
[18:53:30 CEST] <cone-366> ffmpeg 03Fredrik Hubinette 07release/3.3:670d3189e9ef: avformat/id3v2: fix leak in chapter parsing
[19:08:10 CEST] <jya> peloverde: ping
[19:09:18 CEST] <cone-366> ffmpeg 03Martin Storsjö 07master:10f4511f14a4: libavutil: Make LOCAL_ALIGNED(xx be equal to LOCAL_ALIGNED_xx(
[19:09:19 CEST] <cone-366> ffmpeg 03James Almer 07master:5de85c102988: Merge commit '10f4511f14a4e830c0ed471df4cd1cc2a18a481a'
[19:13:15 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:c2bebfc80128: pthread_frame: Propagate sw_pix_fmt across threads
[19:13:16 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:9aa251c98ce6: vf_deinterlace_vaapi: Add support for field rate output
[19:13:17 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:f64d1100a54d: avconv: Flush output BSFs when encode reaches EOF
[19:13:18 CEST] <cone-366> ffmpeg 03James Almer 07master:b848eb834791: Merge commit 'f64d1100a54d12c78ce436181bb64229c56da6b3'
[19:14:13 CEST] <jya> anyone familiar with the aac parser/decoder ?
[19:14:28 CEST] <kierank> atomnuker probably
[19:15:48 CEST] <atomnuker> I'm the worst possible person to ask, I know nothing about aac and aac dts packets
[19:15:51 CEST] <cone-366> ffmpeg 03Vodyannikov Aleksandr 07release/3.0:8642322b9f23: avcodec/cfhd: Fix decoding regression due to height check
[19:15:52 CEST] <cone-366> ffmpeg 03Brice Waegeneire 07release/3.0:d57345e8d076: doc/filters: typo in frei0r
[19:15:53 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:d4bc7fc412ba: avcodec/aacdec_fixed: fix: left shift of negative value -1
[19:15:54 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:3fd54e444022: avcodec/aacps: Fix multiple integer overflow in map_val_34_to_20()
[19:15:55 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:8c05ac89d353: avformat/oggparsecelt: Do not re-allocate os->private
[19:15:56 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:665311ab1f96: avcodec/hevc_ps: fix integer overflow in log2_parallel_merge_level_minus2
[19:15:57 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:f31fc4755f69: avcodec/dnxhddec: Move mb height check out of non hr branch
[19:15:57 CEST] <atomnuker> jya: but I think rcombs might know
[19:15:58 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:654e157d2161: avcodec/diracdec: Fix integer overflow in signed multiplication in UNPACK_ARITH()
[19:15:59 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:2954ce9dea00: avformat/rtmppkt: Convert ff_amf_tag_size() to bytestream2
[19:16:00 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:81c940b15101: avformat/rtmppkt: Convert ff_amf_get_field_value() to bytestream2
[19:16:01 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:43d7b1e42fd7: avcodec/takdec: Fix integer overflow in decode_subframe()
[19:16:02 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:ac0fbaf8acd8: avcodec/diracdec: Fix integer overflow in divide3()
[19:16:03 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:693db350dd0b: avcodec/dirac_dwt: Fix multiple integer overflows in COMPOSE_DD97iH0()
[19:16:04 CEST] <cone-366> ffmpeg 03Steven Siloti 07release/3.0:a371850d58e0: avformat/utils: fix memory leak in avformat_free_context
[19:16:05 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:b59d6183c4ff: avcodec/h264_slice: Fix overflow in slice offset
[19:16:06 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:410f709bec2b: avcodec/aacdec_fixed: fix invalid shift in predict()
[19:16:07 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:f91733e1415b: avcodec/dirac_dwt: Fixes integer overflows in COMPOSE_DAUB97*
[19:16:08 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:e154826a2f02: avcodec/mpeg4videodec: Clear mcsel before decoding an image
[19:16:09 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:9bfa8b692ebd: avcodec/diracdec: Check perspective_exp and zrs_exp.
[19:16:10 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:f7509e9fce7a: avcodec/snowdec: Fix off by 1 error
[19:16:11 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:689e97fcada5: avcodec/fic: Fixes signed integer overflow
[19:16:12 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:2f49580e0329: avcodec/dirac_dwt_template: Fix integer overflow in vertical_compose53iL0()
[19:16:13 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:c543ff526bba: avcodec/me_cmp: Fix crashes on ARM due to misalignment
[19:16:14 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:2820ffe39236: avcodec/aacdec_template: Fix running cleanup in decode_ics_info()
[19:16:15 CEST] <cone-366> ffmpeg 03Vitaly Buka 07release/3.0:616154a6a535: avcodec/utils: Fix signed integer overflow in rc_initial_buffer_occupancy initialization
[19:16:16 CEST] <cone-366> ffmpeg 03Vitaly Buka 07release/3.0:64af458bb862: avformat/mov: Fix signed integer overflows with total_size
[19:16:17 CEST] <cone-366> ffmpeg 03Vitaly Buka 07release/3.0:aadd7fbc14a3: avformat/aviobuf: Fix signed integer overflow in avio_seek()
[19:16:18 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:9e98eee39dd3: avcodec/hevc_ps: Check delta_pocs in ff_hevc_decode_short_term_rps()
[19:16:19 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:cf838b8fd263: ffprobe: Fix null pointer dereference with color primaries
[19:16:20 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:ffdc430c4aeb: ffprobe: Fix NULL pointer handling in color parameter printing
[19:16:21 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:b9fa2a86e6dd: avformat/hls: Fix DoS due to infinite loop
[19:16:22 CEST] <cone-366> ffmpeg 03Yi and  *®() 07release/3.0:39ddbd204aad: avformat/asfdec: Fix DoS due to lack of eof check
[19:16:23 CEST] <cone-366> ffmpeg 03Yi and  *®() 07release/3.0:adca94d65e5c: avformat/cinedec: Fix DoS due to lack of eof check
[19:16:24 CEST] <cone-366> ffmpeg 03Yi and  *®() 07release/3.0:81e6a95e54ff: avformat/rmdec: Fix DoS due to lack of eof check
[19:16:25 CEST] <cone-366> ffmpeg 03Yi and  *®() 07release/3.0:4c6bed6e3be1: avformat/rl2: Fix DoS due to lack of eof check
[19:16:26 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:9a73a776816b: avformat/mvdec: Fix DoS due to lack of eof check
[19:16:27 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:a5018026af1d: avcodec/sbrdsp_fixed: Fix undefined overflows in autocorrelate()
[19:16:28 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:51ee15df5813: avcodec/hevc_ps: Fix undefined shift in pcm code
[19:16:29 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:e89125faba1a: avcodec/snowdec: Fix integer overflow in decode_subband_slice_buffered()
[19:16:30 CEST] <cone-366> ffmpeg 03Yi(SÑ) 07release/3.0:c6d3640cf71c: avformat/nsvdec: Fix DoS due to lack of eof check in nsvs_file_offset loop.
[19:16:31 CEST] <cone-366> ffmpeg 03Yi(SÑ) 07release/3.0:74c067e95572: avformat/mxfdec: Fix DoS issues in mxf_read_index_entry_array()
[19:16:32 CEST] <cone-366> ffmpeg 03Yi(SÑ) 07release/3.0:b2aa633d663d: avformat/mxfdec: Fix Sign error in mxf_read_primer_pack()
[19:16:33 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:92f4341ed113: avcodec/diracdec: Fix integer overflow in INTRA_DC_PRED()
[19:16:34 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:fd4500df5c87: avcodec/dirac_dwt: Fix multiple overflows in 9/7 lifting
[19:16:35 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:d08abbd0bd16: avformat/mov: Fix DoS in read_tfra()
[19:16:36 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:e4a9790bac92: avformat/asfdec: Fix DoS in asf_build_simple_index()
[19:16:37 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:bed626f845cd: avcodec/diracdec: Fix overflow in DC computation
[19:16:38 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:03d8e9fec5c1: avcodec/hevcdsp_template: Fix undefined shift in put_hevc_pel_bi_w_pixels
[19:16:39 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:9bc5df5ec8b7: avcodec/jpeg2000dsp: Fix multiple integer overflows in ict_int()
[19:16:40 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:791d06da1a00: avcodec/hevc_ps: Fix c?_qp_offset_list size
[19:16:41 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:d54cc921a4b2: avcodec/pngdec: Clean up on av_frame_ref() failure
[19:16:42 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:145fce8c4b2d: avcodec/svq3: Fix overflow in svq3_add_idct_c()
[19:16:43 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:3386a57b086f: avcodec/ffv1dec: Fix integer overflow in read_quant_table()
[19:16:44 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:977d6d8bffe3: avcodec/dirac_dwt: Fix integer overflow in COMPOSE_FIDELITYi*()
[19:16:45 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:a89a340e4f1b: avcodec/takdec: Fix integer overflows in decode_subframe()
[19:16:46 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:09913c5ec4cc: avcodec/proresdec2: Check bits in DECODE_CODEWORD(), fixes invalid shift
[19:16:47 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:2c84969cb4ae: avcodec/takdec: Fix integer overflow in decode_lpc()
[19:16:48 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:d164c49af76b: avcodec/jpeg2000: Check that codsty->log2_prec_widths/heights has been initialized
[19:16:49 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:9f2beced0a52: avcodec/hevcdsp_template: Fix undefined shift
[19:16:50 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:fc1acb103dde: avcodec/proresdec2: SKIP_BITS() does not work with len=32
[19:16:51 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:d0e4c3410c90: avcodec/aacdec_template: Clear tns present flag on error
[19:16:52 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:9825fcea99e4: avcodec/truemotion2: Fix integer overflows in tm2_high_chroma()
[19:16:53 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:6fa58eabb18c: avcodec/x86/lossless_videoencdsp: Fix handling of small widths
[19:16:54 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:789ad4d361e1: avcodec/mpeg4videodec: Use 64 bit intermediates for sprite delta
[19:16:55 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:ad3b198f479a: avcodec/mpeg_er: Clear mcsel in mpeg_er_decode_mb()
[19:16:56 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:87a8a4a507f0: avcodec/dirac_dwt: Fix integer overflow in COMPOSE_53iL0()
[19:16:57 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:df441441c552: avcodec/ffv1dec: Fix out of array read in slice counting
[19:16:58 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:5422cdca42e7: avcodec/pafvideo: Check for bitstream end in decode_0()
[19:16:59 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:abb3ec84bb7d: avcodec/snowdec: Check mv_scale
[19:17:00 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.0:479e65ba47a1: Update for 3.0.10
[19:18:36 CEST] <peloverde> jya: pong
[19:19:08 CEST] <jya> peloverde: I have a question about the aac_parser, what type of input is it expecting? raw aac or something encapsulated in adts?
[19:19:33 CEST] <peloverde> It's expecting adts
[19:19:58 CEST] <jya> ah that explains why it gives me rubbish answer when fed raw aac
[19:20:10 CEST] <jya> thank you, that is what I had been expecting
[19:21:26 CEST] <jya> peloverde: are you familiar with the flac parser too?
[19:21:35 CEST] <peloverde> not at all, sorry
[19:22:32 CEST] <jamrial> BBB: ping
[19:22:38 CEST] <BBB> y0
[19:23:02 CEST] <jamrial> BBB: https://pastebin.com/NjxYqKxV
[19:23:19 CEST] <jamrial> i'm merging the superframe bsf from libav. those are the changes they added on top of it
[19:23:51 CEST] <jamrial> the do{} while(0) is good so i'm going to apply it, but i want your opinion on the rest
[19:25:23 CEST] <peloverde> As long as we are talking about the superframe filter, is it possible to get it to properly set packet positions for the subframes?
[19:25:39 CEST] <BBB> the functional change is wrong
[19:25:51 CEST] <BBB> it allows recusrively packing superframes in superframes
[19:26:31 CEST] <atomnuker> jamrial: the do { part should be on a new line IMO
[19:26:39 CEST] <BBB> so I would ignore the remainder
[19:27:10 CEST] <BBB> peloverde: what do you mean about packet positions of subframes?
[19:27:20 CEST] <BBB> peloverde: you mean the splitter? or the merger? superframe=merger
[19:28:32 CEST] <peloverde> I meant split, my bad
[19:28:51 CEST] <BBB> we can probably do that for the splitter yes
[19:35:30 CEST] <J_Darnley> :)  avutil/tests/cpu now reports avx512 detected
[19:38:46 CEST] <jamrial> BBB: ok, will skip the functional change and only apply the do while
[19:41:53 CEST] <cone-366> ffmpeg 03Ronald S. Bultje 07master:0cf949a01193: vp9: Add bsf to merge superframes
[19:41:54 CEST] <cone-366> ffmpeg 03James Almer 07master:4c0426c478dc: Merge commit '0cf949a01193dcf6f83fd95d46792dd94479b4e4'
[19:47:54 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:34e051d16850: vp9: Add bsf to fix reordering in raw streams
[19:47:55 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:0fd91e4bfc00: vaapi_encode: Add VP9 support
[19:47:56 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:41dda860870f: doc: Add VAAPI encoders
[19:47:57 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:4936a48b1e6f: qsv: Add ability to create a session from a device
[19:47:58 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:8848ba0bd6b0: qsvdec: Allow use of hw_device_ctx to make the internal session
[19:47:59 CEST] <cone-366> ffmpeg 03Mark Thompson 07master:3d197514e613: qsvenc: Allow use of hw_device_ctx to make the internal session
[19:48:00 CEST] <cone-366> ffmpeg 03James Almer 07master:4576ee100b9e: Merge commit '3d197514e613ccd9eab43180c0a7c8b09a307606'
[19:48:03 CEST] <TD-Linux> thardin, that's why in practice, the spec of jpeg is "what libjpeg v6 implements"
[19:51:29 CEST] <kierank> Gramner: hmmm
[19:54:28 CEST] <Shiz> TD-Linux: unrelatedly https://twitter.com/enginetankard/status/894115533729087488 <-- very good but I've got two (hopefully-soon-networked) PC98s ;p
[19:55:07 CEST] <TD-Linux> Shiz, a pc9821 is on my shopping list but I need a 24khz monitor as well
[19:55:57 CEST] <Shiz> theres some specific LCD monitors that can do 24khz
[19:56:13 CEST] <Shiz> theres a list of monitors that work with old ataris and the like -- if it goes down to 15khz, it can do 24khz as well
[19:56:42 CEST] <Shiz> http://15khz.wikidot.com/
[19:56:44 CEST] <Shiz> here
[19:56:52 CEST] <Shiz> i happen to luckily own a benq g2420hd :p
[19:57:50 CEST] <iive> is that horizontal frequency?
[19:58:23 CEST] <TD-Linux> yeah the monitors attached to the x68000s are of that variety. I also have a nice bvm but it's 15khz only. I wouldn't mind a 24khz crt, either.
[19:58:36 CEST] <Shiz> yeah hsync
[19:59:25 CEST] <Shiz> if you need pc98 shopping advice hmu anyway, it can be annoying to nail down the model you want which has everything
[20:00:59 CEST] <Shiz> https://up.shiz.me/priv/2017-10-17%2021.27.39.mp4 setup
[20:02:24 CEST] <Compn> jya : maybe alex converse ?
[20:06:23 CEST] <TD-Linux> Shiz, thanks, I'll probably take you up on that offer later and maybe run some yaj links past you.
[20:06:34 CEST] <Shiz> #o
[20:06:39 CEST] <Shiz> \o *
[20:21:41 CEST] <BBB> J_Darnley: nice
[20:23:26 CEST] <cone-366> ffmpeg 03Diego Biurrun 07master:604fbb3132e8: mov: Move code shared with CAF to a separate file
[20:23:27 CEST] <cone-366> ffmpeg 03James Almer 07master:4bb09bf5a04a: Merge commit '604fbb3132e88727e496c96c92cfe02748c25a1a'
[20:34:57 CEST] <cone-366> ffmpeg 03James Almer 07master:c64800e976f9: configure: add optional zlib dependency to avformat
[20:56:01 CEST] <atomnuker> michaelni: does the built-in ratecontrol system for snow/mpeg4 take VFR video into account?
[20:59:51 CEST] <michaelni> atomnuker, if not then you have to maybe change a few lines of picture_number related code to use pts
[21:11:41 CEST] <thardin> TD-Linux: so I've suspected
[21:12:10 CEST] <thardin> I'm not even sure I've seen hierarchical jpegs anywhere
[21:15:48 CEST] <cone-366> ffmpeg 03Diego Biurrun 07master:b62ed6873821: configure: Better names for functions that sanitize input
[21:15:49 CEST] <cone-366> ffmpeg 03James Almer 07master:cdba33c45f79: Merge commit 'b62ed6873821c8fce8f7e2c2927ae54f86afeb22'
[21:21:58 CEST] <cone-366> ffmpeg 03Sean McGovern 07master:6ac0e7818399: mpeg4videodec: raise an error if sprite_trajectory.table is NULL
[21:21:59 CEST] <cone-366> ffmpeg 03James Almer 07master:af0505ed9515: Merge commit '6ac0e7818399a57e4684202bac79f35b3561ad1e'
[21:22:36 CEST] <cone-366> ffmpeg 03James Almer 07n3.3.5:HEAD: Merge commit '6ac0e7818399a57e4684202bac79f35b3561ad1e'
[21:35:10 CEST] <cone-366> ffmpeg 03James Almer 07n3.5-dev:HEAD: Merge commit '6ac0e7818399a57e4684202bac79f35b3561ad1e'
[21:42:03 CEST] <cone-366> ffmpeg 03Luca Barbato 07master:d32d59bc977b: matroska: Read only the data written in the scratch buffer
[21:42:04 CEST] <cone-366> ffmpeg 03James Almer 07master:ce9e07e6b675: Merge commit 'd32d59bc977b43031007bb2ab21e232f96d2ebcb'
[21:47:34 CEST] <cone-366> ffmpeg 03Vittorio Giovara 07master:f5950b8fd61e: lavfi: Drop unused and empty header file
[21:47:35 CEST] <cone-366> ffmpeg 03James Almer 07master:99ba85a0e24b: Merge commit 'f5950b8fd61ec85e0ad8790bea56b37ceea19436'
[21:52:23 CEST] <cone-366> ffmpeg 03Vittorio Giovara 07master:839010aca971: filtfmts-test: Mark filter as const
[21:52:24 CEST] <cone-366> ffmpeg 03Vittorio Giovara 07master:efddf2c09aed: decode: Initialize ret before using it
[21:52:25 CEST] <cone-366> ffmpeg 03James Almer 07master:7066a92f87a4: Merge commit 'efddf2c09aed7400c73ecf327f86a4d0452b94b5'
[22:01:49 CEST] <atomnuker> michaelni: it actually already does that, nice
[22:10:55 CEST] <cone-366> ffmpeg 03Andreas Unterweger 07master:b200a2c8da40: examples: Fixed and extended Doxygen documentation
[22:10:56 CEST] <cone-366> ffmpeg 03James Almer 07master:8a3d3b624075: Merge commit 'b200a2c8da403b5a5c8b50f8cb4a75fd4f0131b1'
[22:14:06 CEST] <cone-366> ffmpeg 03Diego Biurrun 07master:193b09189004: thread: Define ff_mutex_* macros as stub functions when threads are disabled
[22:14:07 CEST] <cone-366> ffmpeg 03James Almer 07master:00bfe8509c9b: Merge commit '193b09189004ede4a6998e69192d1a9f63602088'
[22:30:14 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:ab2d991382c2: avcodec/pngdec: Clean up on av_frame_ref() failure
[22:30:15 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:6b66cd8c40df: avcodec/svq3: Fix overflow in svq3_add_idct_c()
[22:30:16 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:5cf5a1034c07: avcodec/ffv1dec: Fix integer overflow in read_quant_table()
[22:30:17 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:720a44f3a327: avcodec/dirac_dwt: Fix integer overflow in COMPOSE_FIDELITYi*()
[22:30:18 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:397674a3fb1c: avcodec/takdec: Fix integer overflows in decode_subframe()
[22:30:19 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:9e8a636551b2: avcodec/proresdec2: Check bits in DECODE_CODEWORD(), fixes invalid shift
[22:30:20 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:04be199f8ea0: avcodec/takdec: Fix integer overflow in decode_lpc()
[22:30:21 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:74b28c54d9af: avcodec/jpeg2000: Check that codsty->log2_prec_widths/heights has been initialized
[22:30:22 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:2293e5a78a6a: avcodec/hevcdsp_template: Fix undefined shift
[22:30:23 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:79bec49315ba: avcodec/proresdec2: SKIP_BITS() does not work with len=32
[22:30:24 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:d9630deca429: avcodec/aacdec_template: Clear tns present flag on error
[22:30:25 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:2db9b3199641: avcodec/truemotion2: Fix integer overflows in tm2_high_chroma()
[22:30:26 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:0eb0b21c7f4f: avcodec/x86/lossless_videoencdsp: Fix handling of small widths
[22:30:27 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:16b07d59de97: avcodec/mpeg4videodec: Use 64 bit intermediates for sprite delta
[22:30:28 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:b2c9771dd435: avcodec/mpeg_er: Clear mcsel in mpeg_er_decode_mb()
[22:30:29 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:11f5a1319646: avcodec/dirac_dwt: Fix integer overflow in COMPOSE_53iL0()
[22:30:30 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:f052d1df3e59: avcodec/ffv1dec: Fix out of array read in slice counting
[22:30:31 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:1959c0f9f921: avcodec/pafvideo: Check for bitstream end in decode_0()
[22:30:32 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:70d0cc822161: avcodec/snowdec: Check mv_scale
[22:30:33 CEST] <cone-366> ffmpeg 03Fredrik Hubinette 07release/3.2:b76dcd07868e: avformat/id3v2: fix leak in chapter parsing
[22:30:34 CEST] <cone-366> ffmpeg 03Michael Niedermayer 07release/3.2:4b1e87dc32b7: Update for 3.2.9
[22:59:03 CEST] <cone-366> ffmpeg 03James Almer 07n3.2.9:HEAD: Merge commit '193b09189004ede4a6998e69192d1a9f63602088'
[23:43:50 CEST] <cone-366> ffmpeg 03Carl Eugen Hoyos 07master:693a11b8a269: lavf: Remove AVFMT_RAWPICTURE.
[00:00:00 CEST] --- Fri Oct 27 2017