[Ffmpeg-devel-irc] ffmpeg-devel.log.20130729

[burek]

[00:00] <durandal_1707> what I should do instead?
[00:01] <cehoyos> You wrote above that you don't want to revert but instead fix the issue properly.
[00:02] <cehoyos> (I would of course have been able to revert but you said you are against reverting.)
[00:07] <durandal_1707> i thought it was possible to fix bug without reverting, but i found it revert is better
[00:07] <durandal_1707> i will rewrite this code anyway
[00:07] <cone-858> ffmpeg.git 03Carl Eugen Hoyos 07master:2c046c718aef: Revert "Avoid overflows when reading pgm files with maxval != 255 and != 65535."
[00:07] <cone-858> ffmpeg.git 03Paul B Mahol 07release/1.1:ef81f55ec7a9: Revert "pnm: remove nonsense code"
[00:07] <cone-858> ffmpeg.git 03Paul B Mahol 07release/1.2:9f2d73168f54: Revert "pnm: remove nonsense code"
[00:07] <cone-858> ffmpeg.git 03Paul B Mahol 07release/2.0:0047a310905c: Revert "pnm: remove nonsense code"
[00:25] <durandal_1707> if filter_ctx->input/output_count is deprecated what is used instead?
[00:28] <cone-858> ffmpeg.git 03Paul B Mahol 07master:329221eee7d6: pnm: fix 255 < maxval < 65535 for pam with depth 1
[00:28] <cone-858> ffmpeg.git 03Paul B Mahol 07master:a5155294e571: pnm: do not set avctx->coded_frame
[00:43] <cone-858> ffmpeg.git 03Paul B Mahol 07master:e999f2339ab0: truemotion2: make code independent of sizeof(AVFrame)
[00:43] <cone-858> ffmpeg.git 03Paul B Mahol 07master:19da86f3ff43: targaenc: remove unused TargaContext
[01:04] <durandal_1707> hey http://ffmpeg.gusari.org/viewtopic.php?f=8&t=1014
[02:07] <michaelni> is jon severinsson on irc ?
[02:18] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:f3c300d0a4d3: update all trac links to use the trac subdomain
[02:44] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:73746237a188: Prepare for 0.8.7 Release
[02:44] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:a563e4af9f56: oma: Validate sample rates
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:2eaf8698a3bb: avfiltergraph: check for sws opts being non-NULL before using them.
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:c65fb5b41b8e: xmv: do not leak memory in the error paths in xmv_read_header()
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:0f6364b62bb5: bmv: check for len being valid in bmv_decode_frame().
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:9aa2eee31389: xmv: check audio track parameters validity.
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:4c7f40c6df83: dfa: check for invalid access in decode_wdlt().
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:881526744eea: lavf: make sure stream probe data gets freed.
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:74753cf1a99b: indeo3: fix data size check
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:c5084a176549: rv10: check that extradata is large enough
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:9b2af4d080c7: indeo3: check motion vectors.
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:c579d4283edb: indeo3: switch parsing the header to bytestream2
[02:44] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:43c0a87279e7: qdm2: check that the FFT size is a power of 2
[02:44] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:be90f0279d07: Merge commit '43c0a87279e717c1384314c6da7155c306ee7c60' into release/0.10
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:d2d2ddf9a34e: oggdec: fix faulty cleanup prototype
[03:26] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:7bf6a86f2e49: update Changelog
[03:26] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:a590979988b1: aac: check the maximum number of channels
[03:26] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:6742f0408dba: indeo3: fix off by one in MV validity check
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:f4bb72d33db2: id3v2: check for end of file while unescaping tags
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:31ed79af7f47: wav: Always seek to an even offset
[03:26] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:2f61e4090820: update Changelog
[03:26] <cone-858> ffmpeg.git 03Michael Smith 07release/0.10:4941dfb4f6de: proresdec: support mixed interlaced/non-interlaced content
[03:26] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:e6617580e3f2: update Changelog
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:a03dcec36474: mjpegdec: properly report unsupported disabled features
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:2ebabfff4804: jpegls: return meaningful errors
[03:26] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:ca4a25acf89b: jpegls: factorize return paths
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:d26bc6c6b69f: jpegls: check the scan offset
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:6711d410dc13: wavpack: validate samples size parsed in wavpack_decode_block
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:da5cf7e45263: ljpeg: use the correct number of components in yuv
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:33492ad81000: mjpeg: Validate sampling factors
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:7ca8d8223db2: mjpegdec: validate parameters in mjpeg_decode_scan_progressive_ac
[03:26] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:42fed7f433e6: wavpack: check packet size early
[03:26] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:7118358a950e: Merge commit '42fed7f433e6d2167ffd4aae31905b583a53b988' into release/0.10
[04:02] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:5a6af4fd74a4: wavpack: return meaningful errors
[04:02] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:ea7ba1d8717d: apetag: use int64_t for filesize
[04:02] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:96de1c5ed90b: tiff: do not overread the source buffer
[04:02] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:e98f95670bf1: Prepare for 0.8.8 Release
[04:02] <cone-858> ffmpeg.git 03Kostya Shishkov 07release/0.10:e96aaa5622ed: smacker: fix an off by one in huff.length computation
[04:02] <cone-858> ffmpeg.git 03Kostya Shishkov 07release/0.10:d7b7b10518cc: smacker: check the return value of smacker_decode_tree
[04:02] <cone-858> ffmpeg.git 03Kostya Shishkov 07release/0.10:db0c8061fe54: smacker: pad the extradata allocation
[04:02] <cone-858> ffmpeg.git 03Kostya Shishkov 07release/0.10:b40870e63640: smacker: check frame size validity
[04:02] <cone-858> ffmpeg.git 03Alexandra Khirnova 07release/0.10:9248f789d1fa: vmdav: convert to bytestream2
[04:02] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:078e68d2617f: 4xm: don't rely on get_buffer() initializing the frame.
[04:02] <cone-858> ffmpeg.git 03Anton Khirnov 07release/0.10:e797b7787b25: 4xm: check the return value of read_huffman_tables().
[04:02] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:284ac9191b75: 4xm: use the correct logging context
[04:02] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:e5679444fd60: 4xm: reject frames not compatible with the declared version
[04:02] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:6a4f1e784e39: 4xm: check bitstream_size boundary before using it
[04:02] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:0c943d1cdd18: 4xm: do not overread the source buffer in decode_p_block
[04:02] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:b20004b2e616: lavc: move put_bits_left in put_bits.h
[04:02] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:4ff5167ee7fd: wmapro: make sure there is room to store the current packet
[04:02] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:a3539d26ecee: Merge commit '4ff5167ee7fdee6d35c1bb2558172329ae6ec770' into release/0.10
[04:40] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:4e1999ebcb67: wmapro: check num_vec_coeffs against the actual available buffer
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:46e09894acb2: wmapro: return early on unsupported condition
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:bd5ff335ec55: wmapro: check the min_samples_per_subframe
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:f16aa5843ff9: wmapro: error out on impossible scale factor offsets
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:7658333c1755: indeo: Properly forward the error codes
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:7eff48029f99: indeo: use proper error code
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:0a1d02ca77c9: indeo: check for reference when inheriting mvs
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:e6a365b5d2fc: lavc: check for overflow in init_get_bits
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:c9d8424395fc: indeo: use a typedef for the mc function pointer
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:e22a5d490de1: kmvc: use fixed sized arrays in the context
[04:41] <cone-858> ffmpeg.git 03Luca Barbato 07release/0.10:79edb9adf619: kmvc: Clip pixel position to valid range
[04:41] <cone-858> ffmpeg.git 03Reinhard Tartler 07release/0.10:8c62082b5127: Changelog for 0.8.8
[04:41] <cone-858> ffmpeg.git 03Justin Ruggles 07release/0.10:e786cc333120: swfdec: do better validation of tag length
[04:41] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:62d9d4d9d4be: Merge remote-tracking branch 'qatar/release/0.8' into release/0.10
[05:14] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:4e17e9f8afc4: update for 0.10.8 Signed-off-by: Michael Niedermayer <michaelni at gmx.at>
[05:14] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:5bb347a3b4a8: dct-test: fix order of evaluation bug
[05:14] <cone-858> ffmpeg.git 03Michael Niedermayer 07release/0.10:e3e25777944e: ffserver: fix order of evaluation bugs
[11:14] <cone-858> ffmpeg.git 03Luca Barbato 07master:e13a929314e4: xl: K&R formatting cosmetics
[11:15] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:688331c51ce6: Merge commit 'e13a929314e4a9a09d3146648f876f162f46cc32'
[11:21] <saste> durandal_1707, you should not consider user ffmpeg-related incompetence as a personal attack
[11:22] <saste> about workarounds, most people prefer that way, just because they can fix it quickly
[11:22] <saste> it will broke apart the next time they upgrade, but in the short time workarounds represent the shortest path solution
[11:23] <saste> the weird thing is that the guy found the time to document all that stuff, for no obvious reason
[11:32] <cone-858> ffmpeg.git 03Luca Barbato 07master:f1cb490d6d73: xl: Fix the buffer size check
[11:32] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:03853b1091e4: Merge commit 'f1cb490d6d7391ff7e28cc376908cc98a652228d'
[11:38] <cone-858> ffmpeg.git 03Luca Barbato 07master:0e78ef0f94cf: flac: use meaningful return values
[11:38] <cone-858> ffmpeg.git 03Luca Barbato 07master:45ee556d51ef: qdm2: Whitespace cosmetics
[11:38] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:5f312139dff4: Merge commit '45ee556d51ef04d79d52bf6b0b7f28a4d231cb0c'
[11:42] <saste> log level bikeshed
[11:50] <cone-858> ffmpeg.git 03Luca Barbato 07master:62b1e3b1031e: aasc: Check minimum buffer size
[11:50] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:0c6f382e8557: Merge commit '62b1e3b1031e901105d78e831120de8e4c3e0013'
[11:52] <durandal_1707> michaelni: if i prefer to merge every single commit stop doing merges
[12:06] <cone-858> ffmpeg.git 03Paul B Mahol 07master:b09d86c6366f: utvideoenc: use av_image_copy_plane()
[12:06] <cone-858> ffmpeg.git 03Zhang Rui 07master:0f51c398beac: http: Support reading gzip/deflate compressed data
[12:06] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:39a69d9dfbc7: Merge commit '0f51c398beac87682b2249662b97e30512f7868c'
[12:20] <cone-858> ffmpeg.git 03Luca Barbato 07master:3ca1dd2502a8: xl: Make sure the width is valid
[12:20] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:6561aae8a955: Merge commit '3ca1dd2502a860989ee5289316af6b9816cee78f'
[12:33] <cone-858> ffmpeg.git 03Luca Barbato 07master:4f03a77e5259: oma: refactor seek function
[12:33] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:bc672a016fa1: Merge commit '4f03a77e52596cbe9ec179666ddb3e0345a8133a'
[12:46] <michaelni> mergig comits one at a time makes later bisecting and debuging much easier
[12:48] <michaelni> git can bisect of course over arbitrary complex merges but that doesnt help if one ancestor branch is 2+ years outdated and missing the feature one wants to debug
[12:49] <michaelni> also it avoid having to rebase huge merges when others push changes while confilcts are being resolved
[12:51] <cone-858> ffmpeg.git 03Luca Barbato 07master:9d0b45ade864: oma: check geob tag boundary
[12:51] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:9d8fb2374723: Merge commit '9d0b45ade864f3d2ccd8610149fe1fff53c4e937'
[12:51] <ubitux> BBB: i really don't know what is the timeline and workload required for each thing; as long as i learn about video codecs i'm happy with anything
[12:59] <durandal_1707> michaelni: cherry-pick, actually pushing real commits and not merges actually show what is changed
[12:59] <durandal_1707> how i'm supposed to see what really changed?
[12:59] <durandal_1707> by using merges you helping noone
[13:01] <cone-858> ffmpeg.git 03Luca Barbato 07master:2219e27b5b17: oma: correctly mark and decrypt partial packets
[13:01] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:5fb652dddb79: Merge commit '2219e27b5b17d146e4ab71a3ed86dfc013fb7a93'
[13:02] <michaelni> the gitlog script could be changed to show the changes to the first parent on merges
[13:03] <michaelni> also somethig like "git log --first-parent -p -m" might work
[13:04] <durandal_1707> this is far from useful
[13:20] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:c5b46a064d99: lavf: Don't interpret just slightly broken timestamps as wraparound
[13:20] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:f7bc03bcaff5: Merge commit 'c5b46a064d9991f2cd045c90179fccf35ecffc34'
[13:28] <michaelni> if you think cherry picks work better (against everything linus says about rebasing others history), noone stops you from using them to "merge" changes from libav into your private git repo. If they (against my expectations) do work better nothing would be in the way to reconsider things (given there are volunteers, lack of technical issues ...)
[13:30] <durandal11707> michaelni: i read what linus said, and what you are doing is definitely not following him
[13:32] <michaelni> what libav is doing is not following him really, and thats where all the problems stem from
[13:32] <durandal11707> this is like tallking with wall
[13:35] <michaelni> i cant change libav, nor can i change your oppinon on merges but cherry picks dont work as replacement for merges, you are welcome to show me wrong on any point i said
[13:36] <durandal11707> i'm not looking for replacement of merges, i do not want merges at all
[13:36] <durandal11707> just commit normal way
[13:38] <michaelni> normal way is to use merges (see linux or any other large project)
[13:38] <nevcairiel> merges allow the history to be used for conflict resolving, ie. with a 3-way merge, only if the merge history is known it can use the full power of this merge and resolve more cases automatically
[13:38] <durandal11707> but merges: hides actuall changes, introduces new changes, introduces subtle bugs
[13:40] <durandal11707> nevcairiel: you cleary do not know what you are talking about, should ffmpeg merge itself to 'speed' development and help conflict resolving?
[13:40] <nevcairiel> i know what i talk about, but you clearly dont know the full power of git
[13:41] <durandal11707> if the full power of git is to create subtle bugs than its really nice feature
[13:42] <nevcairiel> this wouldnt change if you apply the changes any other way
[13:43] <durandal11707> not at all, who check that merges are actually doing the right thing?
[13:43] <nevcairiel> who checks if a cherry-pick does the right thing?
[13:44] <nevcairiel> a cherry-pick is just a rebased fast-forward merge, it uses the same mechanism internally
[13:53] <durandal11707> with merges, person that do merges usually just let git do it and do not care single bit if result is optimal
[14:03] <cone-858> ffmpeg.git 03Martin Storsjö 07master:eb33ba04e03d: hls: Return all packets from a previous variant before moving on to the next one
[14:03] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:07440c938073: Merge commit 'eb33ba04e03d9f36e23fffd442510c824be709c3'
[14:08] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:82bf8c878345: hls: Copy the time base from the chained demuxer
[14:08] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:cdd2d73d315e: hls: Don't check discard flags until the parent demuxer's streams actually exist
[14:08] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:d31c0fcfa55a: Merge commit 'cdd2d73d315ecaf19ff49e64c91923275f1bda68'
[14:15] <cone-858> ffmpeg.git 03Martin Storsjö 07master:c11e33a3d966: hls: Set stream offset before opening a chained demuxer
[14:15] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:2f9e97b32a7a: Merge commit 'c11e33a3d9665dd1fc5dbdecdd03a4860ac6a622'
[14:29] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:9d64f236292b: hls: Respect the different stream time bases when comparing dts
[14:29] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:65dcb5449344: Merge commit '9d64f236292ba28018dd9afd2d57f8f944b33f81'
[14:36] <cone-858> ffmpeg.git 03LYF 07master:c110cbf6b59d: hls: Create an AVProgram for each variant
[14:36] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:994e09345ef9: Merge remote-tracking branch 'qatar/master'
[16:38] <durandal_1707> michaelni: why no merges have bunch of See: .... i'm not interested what commits are being duplicated in commit history
[16:38] <durandal_1707> *now
[16:46] <cone-858> ffmpeg.git 03Paul B Mahol 07master:2786ac777d2c: zmbv: use av_image_copy_plane()
[17:00] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:a9903f7ec123: Revert "bitstream: make vlc init of static tables thread safe."
[17:00] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:6af8326354ed: avcodec/ff_init_vlc_sparse: use a spinlock for thread sync
[17:00] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:f80ccf074a3b: avcodec/ff_init_vlc_sparse: use a local array for static vlcs avoiding malloc()
[17:07] <durandal_1707> ubitux: why your site is dead?
[19:41] <cone-858> ffmpeg.git 03Carl Eugen Hoyos 07master:f2118d44c9f1: Fix compilation with --disable-everything --enable-shared.
[20:42] <cxo> On Android if I use stlport_static, at runtime I get a Zdlpv symbol could not be found. If I use gnustl_port, at compile I get undefined reference to _M_allocate/deallocate. What's going on?
[20:44] <cxo> (Trying to use build_stagefright)
[20:58] <cone-858> ffmpeg.git 03Michael Niedermayer 07master:3cd8aaa2b2e7: avcodec/kmvc: fix MV checks
[21:13] <cone-858> ffmpeg.git 03Paul B Mahol 07master:6b68e2a43b34: lavfi: add compand filter
[21:24] <llogan> durandal_1707: you missed the me/be typo
[21:37] <cxo> I think I got disconnected. Any body know if build_stagefright works?
[22:06] <durandal_1707> cxo: give exact build failure
[22:08] <cone-858> ffmpeg.git 03Paul B Mahol 07master:c96b3ae8db35: doc/filters: fix typo
[22:21] <cxo> durandal_1707, sure
[22:28] <cxo> I'm just getting the output now.... By the way, is ffmpeg not compatible with google's stlport_static?
[22:30] <durandal_1707> i dunno what is that
[22:33] <cxo> https://raw.github.com/flyskywhy/android-ndk-r7b/master/docs/CPLUSPLUS-SUPPORT.html
[22:36] <cxo> I just did a git pull and remake and it compiles now with gnustl_port (it didnt yesterday). I get a different runtime error (compared to when using stlport)  http://pastebin.com/Ljk3rnt1
[22:38] <durandal_1707> cxo: i see you disabled bunch of stuff, it happens with normal setup too?
[22:39] <cxo> What is a normal setup?
[22:40] <durandal_1707> the one you do not disable everything
[22:41] <cxo> But I am not using --disable-everything
[22:42] <durandal_1707> but you use --disable-decoders
[22:42] <cxo> Do you want me to remove that and rebuild?
[22:43] <durandal_1707> note this just to find out if the source of problem is bug in Makefile
[22:43] <durandal_1707> cxo: yes, if noone give some better solution
[22:44] <cxo> I dont understand c++ symbols, is OMXCodec::create the same as the missing  "_ZN7android8OMXCodec6CreateERKNS_2spINS_4IOMXEEERKNS1_INS_8MetaDataEEEbRKNS1_INS_11MediaSourceEEEPKcj"
[22:44] <cxo> If its the same, then that symbol is not part of FFmpeg
[22:46] <durandal_1707> i can't dechiper that error at all, thus way i proposed ....
[22:50] <cxo> No difference, same error
[22:52] <durandal_1707> perhaps shared is broken, what if you try static?
[22:54] <cxo> where do i go for static stagefright libs?
[22:54] <cxo> And back to my original question; has anyone know if tools/build_stagefright works?
[22:55] <durandal_1707> well it worked for someone
[22:55] <cxo> who?
[22:56] <cxo> maybe they can help me
[22:56] <durandal_1707> you should also open bug report
[22:58] <cxo> I will submit a bug report
[00:00] --- Tue Jul 30 2013