Recherche avancée

Sur d’autres sites (5899)

• Evolution #3071 : Performance boucle DATA sur CSV

  15 octobre 2013, par esj -

  De l’utilité de faire "svn cp" et non une copie dans éditeur perso pour que l’historique d’un code soit facile à retrouver.

  Ce code est apparu en 2007 dans r10948 qui répondait à une difficulté signalée dans Spip-contrib, à laquelle il fut répondu par un code, documenté sur spipnet comme il se doit :
  http://contrib.spip.net/Creer-de-grands-tableaux-dans-SPIP,9#forum401060

  Cette fonction faisait alors 3 lignes, c’était plus rapide à écrire que de regarder si PHP avait l’équivalent en magasin.
  Ensuite, il y a eu quantités de signalements de problèmes, ce format propriétaire à l’origine ayant un RFC tardif et imprécis dont tout le monde se fout,
  d’où un code qui n’a pas cessé de croître avec les années (r11111, r11113, r13859, r14013 surtout). Au vu du commentaire dans la doc PHP de fgetcsv, ce code semble toujours d’actualité. En tout cas il y a intérêt à vérifier que les cas signalés dans les logs des commit ci-dessous sont bien pris en compte par cette fonction avant de mettre le code des autres à la poubelle sans chercher à comprendre, marque de fabrique de SPIP3.

• avformat/matroskadec : Check min_luminance more thoroughly

  16 février 2021, par Andreas Rheinhardt

  avformat/matroskadec : Check min_luminance more thoroughly
  In the absence of an explicitly coded minimal luminance, the current
  
  code inferred it to be -1, an invalid value. Yet it did not check the
  
  value lateron at all, so that if a valid maximum luminance is
  
  encountered, but no minimal luminance, an invalid minimal luminance of
  
   1 is exported. If an minimal luminance element with a negative value is
  
  present, it is exported, too. This can be simply fixed by adding a check
  
  for the value of the element.
  Yet given that a minimal luminance of zero Cd/m² is legal and can be
  
  coded with a length of zero, we must not use a fake default value to
  
  find out whether the element is present or not. Therefore this patch
  
  uses an explicit counter for it.
  While just at it, also check for max_luminance > min_luminance.
  Reviewed-by : Ridley Combs <rcombs@rcombs.me>
  
  Signed-off-by : Andreas Rheinhardt <andreas.rheinhardt@gmail.com>
  

  • [DH] libavformat/matroskadec.c

• Statically built FFMPEG binary segmentation fault

  12 février 2020, par stevendesu

  I want to create a custom build of FFMPEG which rips out everything except for the ability to transmux HLS videos to MP4, and I need this build to be 100% static with no external dependencies

  I tried using the following configuration :

  ./configure \
  
      --extra-cflags='-static -static-libstdc++ -static-libgcc' \
  
      --extra-cxxflags='-static -static-libstdc++ -static-libgcc' \
  
      --extra-ldflags='-static -static-libstdc++ -static-libgcc' \
  
      --pkg-config-flags='--static' \
  
      --enable-static \
  
      --disable-shared \
  
      --disable-runtime-cpudetect \
  
      --disable-autodetect \
  
      --disable-ffplay \
  
      --disable-ffprobe \
  
      --disable-doc \
  
      --disable-avdevice \
  
      --disable-swresample \
  
      --disable-swscale \
  
      --disable-postproc \
  
      --disable-pthreads \
  
      --disable-w32threads \
  
      --disable-os2threads \
  
      --enable-network \
  
      --disable-dct \
  
      --disable-dwt \
  
      --disable-error-resilience \
  
      --disable-lsp \
  
      --disable-lzo \
  
      --disable-mdct \
  
      --disable-rdft \
  
      --disable-fft \
  
      --disable-faan \
  
      --disable-pixelutils \
  
      --disable-encoders \
  
      --disable-decoders \
  
      --disable-hwaccels \
  
      --disable-muxers \
  
      --enable-muxer=mov \
  
      --enable-muxer=mp4 \
  
      --disable-demuxers \
  
      --enable-demuxer=hls \
  
      --enable-demuxer=mpegts \
  
      --enable-demuxer=h264 \
  
      --enable-demuxer=aac \
  
      --disable-parsers \
  
      --enable-parser=h264 \
  
      --enable-parser=aac \
  
      --disable-bsfs \
  
      --disable-protocols \
  
      --enable-protocol=tcp \
  
      --enable-protocol=tls \
  
      --enable-protocol=http \
  
      --enable-protocol=https \
  
      --enable-protocol=hls \
  
      --disable-indevs \
  
      --disable-outdevs \
  
      --disable-devices \
  
      --disable-filters \
  
      --disable-alsa \
  
      --disable-appkit \
  
      --disable-avfoundation \
  
      --disable-bzlib \
  
      --disable-coreimage \
  
      --disable-iconv \
  
      --disable-lzma \
  
      --enable-openssl \
  
      --disable-sndio \
  
      --disable-sdl2 \
  
      --disable-securetransport \
  
      --disable-xlib \
  
      --disable-zlib \
  
      --disable-amf \
  
      --disable-audiotoolbox \
  
      --disable-cuda-llvm \
  
      --disable-cuvid \
  
      --disable-d3d11va \
  
      --disable-dxva2 \
  
      --disable-ffnvcodec \
  
      --disable-nvdec \
  
      --disable-nvenc \
  
      --disable-v4l2-m2m \
  
      --disable-vaapi \
  
      --disable-vdpau \
  
      --disable-videotoolbox \
  
      --disable-debug

  This looked about like what I wanted :

  install prefix            /usr/local
  
  source path               .
  
  C compiler                gcc
  
  C library                 glibc
  
  ARCH                      x86 (generic)
  
  big-endian                no
  
  runtime cpu detection     no
  
  standalone assembly       yes
  
  x86 assembler             nasm
  
  MMX enabled               yes
  
  MMXEXT enabled            yes
  
  3DNow! enabled            yes
  
  3DNow! extended enabled   yes
  
  SSE enabled               yes
  
  SSSE3 enabled             yes
  
  AESNI enabled             yes
  
  AVX enabled               yes
  
  AVX2 enabled              yes
  
  AVX-512 enabled           yes
  
  XOP enabled               yes
  
  FMA3 enabled              yes
  
  FMA4 enabled              yes
  
  i686 features enabled     yes
  
  CMOV is fast              yes
  
  EBX available             yes
  
  EBP available             yes
  
  debug symbols             no
  
  strip symbols             yes
  
  optimize for size         no
  
  optimizations             yes
  
  static                    yes
  
  shared                    no
  
  postprocessing support    no
  
  network support           yes
  
  threading support         no
  
  safe bitstream reader     yes
  
  texi2html enabled         no
  
  perl enabled              yes
  
  pod2man enabled           yes
  
  makeinfo enabled          no
  
  makeinfo supports HTML    no
  
  
  
  External libraries:
  
  openssl
  
  
  
  External libraries providing hardware acceleration:
  
  
  
  Libraries:
  
  avcodec                 avfilter                avformat                avutil
  
  
  
  Programs:
  
  ffmpeg
  
  
  
  Enabled decoders:
  
  
  
  Enabled encoders:
  
  
  
  Enabled hwaccels:
  
  
  
  Enabled parsers:
  
  aac                     h264
  
  
  
  Enabled demuxers:
  
  aac                     h264                    hls                     mpegts
  
  
  
  Enabled muxers:
  
  mov                     mp4
  
  
  
  Enabled protocols:
  
  hls                     http                    https                   tcp                     tls
  
  
  
  Enabled filters:
  
  aformat                 anull                   atrim                   format                  hflip                   null                    transpose               trim                    vflip
  
  
  
  Enabled bsfs:
  
  null
  
  
  
  Enabled indevs:
  
  
  
  Enabled outdevs:
  
  
  
  License: LGPL version 2.1 or later

  It included several filters which I won’t ever need or use, but these filters are pulled in automatically if you don’t specify --disable-avfilter, and specifying --disable-avfilter prevents the ffmpeg binary from being produced. So I’m stuck with those.

  Using these parameters and then running make, I received a binary that was about 5.9 MB in size and looked right :

  $> ldd ffmpeg
  
          not a dynamic executable

  But when I try to run it :

  $> ./ffmpeg -version
  
  Segmentation fault

  Using valgrind to try and inspect the cause of the segmentation fault :

  $> valgrind ./ffmpeg -version
  
  .... lots of stuff ...
  
  ==61362== Jump to the invalid address stated on the next line
  
  ==61362==    at 0x0: ???
  
  ==61362==    by 0x70BB1B: ??? (in /src/FFmpeg/ffmpeg)
  
  ==61362==    by 0x70B2E6: ??? (in /src/FFmpeg/ffmpeg)
  
  ==61362==    by 0x4033F9: ??? (in /src/FFmpeg/ffmpeg)
  
  ==61362==    by 0x1FFF000677: ???
  
  ==61362==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
  
  ==61362==
  
  ==61362==
  
  ==61362== Process terminating with default action of signal 11 (SIGSEGV)
  
  ==61362==  Bad permissions for mapped region at address 0x0
  
  ==61362==    at 0x0: ???
  
  ==61362==    by 0x70BB1B: ??? (in /src/FFmpeg/ffmpeg)
  
  ==61362==    by 0x70B2E6: ??? (in /src/FFmpeg/ffmpeg)
  
  ==61362==    by 0x4033F9: ??? (in /src/FFmpeg/ffmpeg)
  
  ==61362==    by 0x1FFF000677: ???
  
  ==61362==
  
  ==61362== HEAP SUMMARY:
  
  ==61362==     in use at exit: 0 bytes in 0 blocks
  
  ==61362==   total heap usage: 0 allocs, 0 frees, 0 bytes allocated
  
  ==61362==
  
  ==61362== All heap blocks were freed -- no leaks are possible
  
  ==61362==
  
  ==61362== For counts of detected and suppressed errors, rerun with: -v
  
  ==61362== Use --track-origins=yes to see where uninitialised values come from
  
  ==61362== ERROR SUMMARY: 93 errors from 90 contexts (suppressed: 0 from 0)
  
  Segmentation fault

  Attempting to access memory at location 0x0 sounds like trying to follow a null pointer. But I’m not sure how to fix this.

  gdb backtrace

  When I first ran gdb ./ffmpeg gdb immediately gave me a segmentation fault and I wasn’t kicked into the gdb REPL, so I couldn’t investigate

  After rebuilding ffmpeg I was able to get in this time :

  $> gdb ./ffmpeg
  
  
  
  GNU gdb (Ubuntu 8.1-0ubuntu3.2) 8.1.0.20180409-git
  
  Copyright (C) 2018 Free Software Foundation, Inc.
  
  License GPLv3+: GNU GPL version 3 or later /gnu.org/licenses/gpl.html>
  
  This is free software: you are free to change and redistribute it.
  
  There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
  
  and "show warranty" for details.
  
  This GDB was configured as "x86_64-linux-gnu".
  
  Type "show configuration" for configuration details.
  
  For bug reporting instructions, please see:
  
  /www.gnu.org/software/gdb/bugs/>.
  
  Find the GDB manual and other documentation resources online at:
  
  /www.gnu.org/software/gdb/documentation/>.
  
  For help, type "help".
  
  Type "apropos word" to search for commands related to "word"...
  
  Reading symbols from ffmpeg...done.
  
  (gdb) r
  
  Starting program: /src/FFmpeg/ffmpeg
  
  warning: Error disabling address space randomization: Operation not permitted
  
  
  
  Program received signal SIGSEGV, Segmentation fault.
  
  0x0000000000000000 in ?? ()
  
  (gdb) bt
  
  #0  0x0000000000000000 in ?? ()
  
  #1  0x0000000000f9a8d5 in __register_frame_info_bases.part.6 ()
  
  #2  0x00000000004445fd in frame_dummy ()
  
  #3  0x0000000000000001 in ?? ()
  
  #4  0x0000000000ebd20c in __libc_csu_init ()
  
  #5  0x0000000000ebc9d7 in __libc_start_main ()
  
  #6  0x000000000044451a in _start ()
  
  (gdb)

  I tried grep’ing the code base for __register_frame_info_bases and found nothing. So I’m not really sure where to go from here

  A fix, but not an explanation

  By randomly removing configuration parameters and rebuilding I discovered that --disable-pthreads was causing the segmentation fault. When I remove this, ffmpeg runs just fine

  I don’t know why this is the case, though. Why would they make it possible to remove something that you need to run ?