Recherche avancée

Sur d’autres sites (8984)

• vulkan_decode : remove informative queries

  24 décembre 2024, par Lynne

  vulkan_decode : remove informative queries
  We queried the decoder whether it was able to decode sucessfully, but
  
  since we operated asynchronously, we weren't able to do anything with
  
  this information but let the user know decoding failed for the previous
  
  frame(s).
  Since we parse the slice headers ourselves and we're reasonably sure we
  
  can decode before actually starting to decode, this was rarely triggered
  
  on corrupt data, and hardware's understanding of whether there was an error
  
  or not is vague.
  There's also a semantic problem with our use of the queries - if there's
  
  a seek, we flush, but what happens to the queries is vague according to
  
  the spec. Most hardware dealt fine, since queries are nothing more than
  
  GPU memory with integers stored. But with Intel, they seem to be more of
  
  a register to which a driver must keep track of, leading to issues if there's
  
  been a reset (seek) and we query the previous submission before the seek.
  Just get rid of them. The query code is still used in encoding.
  This fixes seeking with HEVC and AV1 on Intel.
  

  • [DH] libavcodec/vulkan_decode.c

• avcodec/aarch64/opusdsp_neon : Simplify opus_postfilter_neon

  7 février, par Krzysztof Pyrkosz

  avcodec/aarch64/opusdsp_neon : Simplify opus_postfilter_neon
  This change removes one extra floating point operation and simplifies
  
  load operations at the beginning of the loop by using dedicated register
  
  for each of the 5 pointers and interleaving it with calculations. The
  
  first case seems to be a bit slower, but the performance increase is
  
  substantial in the other two.
  A78 before :
  
  postfilter_15_neon :                                   1684.8 ( 4.23x)
  
  postfilter_512_neon :                                  1395.5 ( 5.10x)
  
  postfilter_1022_neon :                                 1357.0 ( 5.25x)
  After :
  
  postfilter_15_neon :                                   1742.2 ( 4.09x)
  
  postfilter_512_neon :                                  1169.8 ( 6.09x)
  
  postfilter_1022_neon :                                 1160.0 ( 6.12x)
  A72 before :
  
  postfilter_15_neon :                                   3144.8 ( 2.39x)
  
  postfilter_512_neon :                                  3141.2 ( 2.39x)
  
  postfilter_1022_neon :                                 3230.0 ( 2.33x)
  After :
  
  postfilter_15_neon :                                   2847.8 ( 2.64x)
  
  postfilter_512_neon :                                  2877.8 ( 2.61x)
  
  postfilter_1022_neon :                                 2837.2 ( 2.65x)
  x13s before :
  
  postfilter_15_neon :                                   1615.4 ( 2.61x)
  
  postfilter_512_neon :                                   963.1 ( 4.39x)
  
  postfilter_1022_neon :                                  963.6 ( 4.39x)
  After :
  
  postfilter_15_neon :                                   1749.6 ( 2.41x)
  
  postfilter_512_neon :                                   707.1 ( 5.97x)
  
  postfilter_1022_neon :                                  706.1 ( 5.99x)
  Signed-off-by : Martin Storsjö <martin@martin.st>
  

  • [DH] libavcodec/aarch64/opusdsp_neon.S

• x86/hevc_res_add : refactor ff_hevc_transform_add{16,32}_8

  21 août 2014, par James Almer

  x86/hevc_res_add : refactor ff_hevc_transform_add16,32_8
  * Reduced xmm register count to 7 (As such they are now enabled for x86_32).
  
  * Removed four movdqa (affects the sse2 version only).
  
  * pxor is now used to clear m0 only once.
   5% faster.
  Reviewed-by : Christophe Gisquet <christophe.gisquet@gmail.com>
  
  Signed-off-by : James Almer <jamrial@gmail.com>
  

  • [DH] libavcodec/x86/hevc_res_add.asm
  • [DH] libavcodec/x86/hevcdsp_init.c