Recherche avancée

Sur d’autres sites (5986)

• lavu : add new D3D11 pixfmt and hwcontext

  6 juin 2017, par wm4

  lavu : add new D3D11 pixfmt and hwcontext
  To be used with the new d3d11 hwaccel decode API.
  With the new hwaccel API, we don't want surfaces to depend on the
  
  decoder (other than the required dimension and format). The old D3D11VA
  
  pixfmt uses ID3D11VideoDecoderOutputView pointers, which include the
  
  decoder configuration, and thus is incompatible with the new hwaccel
  
  API. This patch introduces AV_PIX_FMT_D3D11, which uses ID3D11Texture2D
  
  and an index. It's simpler and compatible with the new hwaccel API.
  The introduced hwcontext supports only the new pixfmt.
  Frame upload code untested.
  Significantly based on work by Steve Lhomme <robux4@gmail.com>, but with
  
  heavy changes/rewrites.
  Merges Libav commit fff90422d181744cd75dbf011687ee7095f02875.
  Signed-off-by : Diego Biurrun <diego@biurrun.de>
  

  • [DH] doc/APIchanges
  • [DH] libavutil/Makefile
  • [DH] libavutil/hwcontext.c
  • [DH] libavutil/hwcontext.h
  • [DH] libavutil/hwcontext_d3d11va.c
  • [DH] libavutil/hwcontext_d3d11va.h
  • [DH] libavutil/hwcontext_internal.h
  • [DH] libavutil/pixdesc.c
  • [DH] libavutil/pixfmt.h
  • [DH] libavutil/version.h

• Missing functions in new version of FFmpeg

  25 avril 2013, par Asik

  I am upgrading a wrapper around ffmpeg to the most recent version of the library. Some functions have been renamed or replaced. For the most part I've been able to found equivalents easily, but the following two are giving me trouble :

  av_abort_set_callback
  
  av_abort_default_callback

  I could not find these functions by searching online, so what happened to them and what should they be replaced with now ?

  Here is a copy of the header file "abort.h" where they were located :

  /* 
  
   * This file is part of FFmpeg. 
  
   * 
  
   * FFmpeg is free software; you can redistribute it and/or 
  
   * modify it under the terms of the GNU Lesser General Public 
  
   * License as published by the Free Software Foundation; either 
  
   * version 2.1 of the License, or (at your option) any later version. 
  
   * 
  
   * FFmpeg is distributed in the hope that it will be useful, 
  
   * but WITHOUT ANY WARRANTY; without even the implied warranty of 
  
   * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU 
  
   * Lesser General Public License for more details. 
  
   * 
  
   * You should have received a copy of the GNU Lesser General Public 
  
   * License along with FFmpeg; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA 
  
   */
  
  
  
  #ifndef AVUTIL_ABORT_H
  
  #define AVUTIL_ABORT_H
  
  
  
  /** 
  
   * Causes abnormal program termination. By default, av_abort_callback calls 
  
   * abort() from the stdlib. This behavior can be altered by setting a 
  
   * different av_abort callback function. 
  
   */
  
  
  
  void av_abort_internal(void);
  
  void av_abort_set_callback(void (*)(void));
  
  void av_abort_default_callback(void);
  
  
  
  #define av_abort()    do { av_log(NULL, AV_LOG_ERROR, "Abort at %s:%d\n", __FILE__, __LINE__); av_abort_internal(); } while(0)
  
  
  
  #endif /* AVUTIL_ABORT_H */

• Problems with get_buffer2 using ffmpeg / dxva2

  1er mars 2018, par Nick Doran

  I am trying to use hardware acceleration to decode h264 using ffmpeg and dxva2. I am using dx9. I have confirmed that my build of ffmpeg supports dxva, though I didn’t build it myself.

  My problem is related to this post :

  FFMPEG : While decoding video, is possible to generate result to user’s provided buffer ?

  I suspect that the problem is in my get_buffer2 method. After that call, I get these errors / messages from ffmpeg :

  Increasing reorder buffer to 1
  
  Discarding mismatching reference
  
  Discarding mismatching reference
  
  Missing reference picture, default is 0
  
  decode_slice_header error
  
  hardware accelerator failed to decode picture
  
  Failed to send packet: Operation not permitted

  Based on the "Discarding mismatching reference" output, I believe the problem is the mismatch between the height of the AVFrame versus the height in the AVCodecContext.

  The AVFrame coming into myGetBuffer is 1920x1088, but the AVCodecContext has the resolution at 1920x1080 (the actual resolution of the stream). I understand that the buffers must be 32-bit aligned, and I get this output from ffmpeg prior to myGetBuffer being called :

  Reinit context to 1920x1088, pix_fmt: yuv420p

  myGetBuffer populates the frame data, linesize, and buffer using dx9 surfaces and av_buffer_create, pretty much the same way as in the linked question above. As I understand it, buf[0] is for Y, buf[1] is for U, and buf[2] is for V, and frame will be decoded into these buffers.

  I believe my hwaccel is set up correctly. These calls are all successful :

  hr = DXVA2CreateDirect3DDeviceManager9
  
  hr = pD3DManager->ResetDevice
  
  hr = pD3DManager->OpenDeviceHandle
  
  hr = DXVA2CreateVideoService
  
  hr = pDecoderService->GetDecoderRenderTargets
  
  hr = pDecoderService->GetDecoderDeviceGuids
  
  hr = pDecoderService->GetDecoderConfigurations
  
  hr = pDecoderService->CreateSurface
  
  hr = pDecoderService->CreateVideoDecoder

  I am using the "newer" paradigm of avcodec_send_packet and avcodec_receive_frame, not the deprecated avcodec_decode_video2. Everything works fine with software decoding.

  Source is h264 at 1920x1080 from a live camera.

  Is this just a byte alignment problem ? If so, how do I get the AvCodecContext in myGetBuffer to be 1088 ?

  I have done a lot of googling on this, and haven’t really found any good samples. Any suggestions of something I missed are greatly appreciated !

  NOTE -
  When I call avcodec_align_dimensions2, it wants to set the height to 1090, which is not 32-bit aligned. Which seems very odd. Using that value does not change the results that I am getting.