Recherche avancée

Sur d’autres sites (9422)

• lavu/x86 : add FFT assembly

  10 avril 2021, par Lynne

  lavu/x86 : add FFT assembly
  This commit adds a pure x86 assembly SIMD version of the FFT in libavutil/tx.
  
  The design of this pure assembly FFT is pretty unconventional.
  On the lowest level, instead of splitting the complex numbers into
  
  real and imaginary parts, we keep complex numbers together but split
  
  them in terms of parity. This saves a number of shuffles in each transform,
  
  but more importantly, it splits each transform into two independent
  
  paths, which we process using separate registers in parallel.
  
  This allows us to keep all units saturated and lets us use all available
  
  registers to avoid dependencies.
  
  Moreover, it allows us to double the granularity of our per-load permutation,
  
  skipping many expensive lookups and allowing us to use just 4 loads per register,
  
  rather than 8, or in case FMA3 (and by extension, AVX2), use the vgatherdpd
  
  instruction, which is at least as fast as 4 separate loads on old hardware,
  
  and quite a bit faster on modern CPUs).
  Higher up, we go for a bottom-up construction of large transforms, foregoing
  
  the traditional per-transform call-return recursion chains. Instead, we always
  
  start at the bottom-most basis transform (in this case, a 32-point transform),
  
  and continue constructing larger and larger transforms until we return to the
  
  top-most transform.
  
  This way, we only touch the stack 3 times per a complete target transform :
  
  once for the 1/2 length transform and two times for the 1/4 length transform.
  The combination algorithm we use is a standard Split-Radix algorithm,
  
  as used in our C code. Although a version with less operations exists
  
  (Steven G. Johnson and Matteo Frigo's "A modified split-radix FFT with fewer
  
  arithmetic operations", IEEE Trans. Signal Process. 55 (1), 111–119 (2007),
  
  which is the one FFTW uses), it only has 2% less operations and requires at least 4x
  
  the binary code (due to it needing 4 different paths to do a single transform).
  
  That version also has other issues which prevent it from being implemented
  
  with SIMD code as efficiently, which makes it lose the marginal gains it offered,
  
  and cannot be performed bottom-up, requiring many recursive call-return chains,
  
  whose overhead adds up.
  We go through a lot of effort to minimize load/stores by keeping as much in
  
  registers in between construcring transforms. This saves us around 32 cycles,
  
  on paper, but in reality a lot more due to load/store aliasing (a load from a
  
  memory location cannot be issued while there's a store pending, and there are
  
  only so many (2 for Zen 3) load/store units in a CPU).
  
  Also, we interleave coefficients during the last stage to save on a store+load
  
  per register.
  Each of the smallest, basis transforms (4, 8 and 16-point in our case)
  
  has been extremely optimized. Our 8-point transform is barely 20 instructions
  
  in total, beating our old implementation 8-point transform by 1 instruction.
  
  Our 2x8-point transform is 23 instructions, beating our old implementation by
  
  6 instruction and needing 50% less cycles. Our 16-point transform's combination
  
  code takes slightly more instructions than our old implementation, but makes up
  
  for it by requiring a lot less arithmetic operations.
  Overall, the transform was optimized for the timings of Zen 3, which at the
  
  time of writing has the most IPC from all documented CPUs. Shuffles were
  
  preferred over arithmetic operations due to their 1/0.5 latency/throughput.
  On average, this code is 30% faster than our old libavcodec implementation.
  
  It's able to trade blows with the previously-untouchable FFTW on small transforms,
  
  and due to its tiny size and better prediction, outdoes FFTW on larger transforms
  
  by 11% on the largest currently supported size.
  

  • [DH] libavutil/tx.c
  • [DH] libavutil/tx_priv.h
  • [DH] libavutil/x86/Makefile
  • [DH] libavutil/x86/tx_float.asm
  • [DH] libavutil/x86/tx_float_init.c

• Merge commit ’26ec75aec3576daea691dee53a78ec67c0dc4040’

  19 janvier 2016, par Hendrik Leppkes

  Merge commit ’26ec75aec3576daea691dee53a78ec67c0dc4040’
  * commit ’26ec75aec3576daea691dee53a78ec67c0dc4040’ :
  
    checkasm : Check register clobbering on arm
  Merged-by : Hendrik Leppkes <h.leppkes@gmail.com>
  

  • [DH] tests/checkasm/arm/Makefile
  • [DH] tests/checkasm/arm/checkasm.S
  • [DH] tests/checkasm/checkasm.c
  • [DH] tests/checkasm/checkasm.h

• FFmpeg avutil.lib unresolved external symbol

  2 août 2022, par Sere

  i'm trying to use FFmpeg with visual sudio 2022 .NET 6 through an MSVC project.
I followed this tutorial : https://www.youtube.com/watch?v=qMNr1Su-nR8&ab_channel=HakanSoyalp.
I mean I have configureg Include (.h) file, library (.lib) files and dynamic (.dll) file copied into bin folder.
If I call for example the avformat_alloc_context() method inside the avformat.lib all work rightly, if I call for example av_file_map(...) inside the avutil.lib the compiler give me an error :
LNK2019 unresolved external symbol "int __cdecl av_file_map ...
But the avutil.lib is linked !!!
The same happen if I call other methods inside avutil.lib module.

  


  Thanks for help...

  


  Code :

  


  extern "C"&#xA;{&#xA;    #include <libavformat></libavformat>avformat.h>&#xA;    #include <libavutil></libavutil>file.h> // av_file_map&#xA;}&#xA;&#xA;int ConvertJPEGtoNALs(int argc, char* argv[])&#xA;{&#xA;    AVFormatContext* fmt_ctx = NULL;&#xA;    AVIOContext* avio_ctx = NULL;&#xA;    uint8_t* buffer = NULL, * avio_ctx_buffer = NULL;&#xA;    size_t buffer_size, avio_ctx_buffer_size = 4096;&#xA;    char* input_filename = NULL;&#xA;    int ret = 0;&#xA;    struct buffer_data bd = { 0 };&#xA;    if (argc != 2) {&#xA;        fprintf(stderr, "usage: %s input_file\n"&#xA;            "API example program to show how to read from a custom buffer "&#xA;            "accessed through AVIOContext.\n", argv[0]);&#xA;        return 1;&#xA;    }&#xA;    input_filename = argv[1];&#xA;    /* register codecs and formats and other lavf/lavc components*/&#xA;    //av_register_all();    //!deprecated&#xA;    /* slurp file content into buffer */&#xA;    ret = av_file_map(input_filename, &amp;buffer, &amp;buffer_size, 0, NULL);&#xA;    /* fill opaque structure used by the AVIOContext read callback */&#xA;    bd.ptr = buffer;&#xA;    bd.size = buffer_size; ???&#xA;    if (!(fmt_ctx = avformat_alloc_context())) {&#xA;        ret = AVERROR(ENOMEM);&#xA;        goto end;&#xA;    }&#xA;    //... to be continue ...&#xA;}&#xA;