Recherche avancée

Sur d’autres sites (11828)

• Beware the builtins

  14 janvier 2010, par Mans — Compilers

  GCC includes a large number of builtin functions allegedly providing optimised code for common operations not easily expressed directly in C. Rather than taking such claims at face value (this is GCC after all), I decided to conduct a small investigation to see how well a few of these functions are actually implemented for various targets.

  For my test, I selected the following functions :

  • __builtin_bswap32 : Byte-swap a 32-bit word.
  • __builtin_bswap64 : Byte-swap a 64-bit word.
  • __builtin_clz : Count leading zeros in a word.
  • __builtin_ctz : Count trailing zeros in a word.
  • __builtin_prefetch : Prefetch data into cache.

  To test the quality of these builtins, I wrapped each in a normal function, then compiled the code for these targets :

  • ARMv7
  • AVR32
  • MIPS
  • MIPS64
  • PowerPC
  • PowerPC64
  • x86
  • x86_64

  In all cases I used compiler flags were -O3 -fomit-frame-pointer plus any flags required to select a modern CPU model.
  

  ARM

  Both __builtin_clz and __builtin_prefetch generate the expected CLZ and PLD instructions respectively. The code for __builtin_ctz is reasonable for ARMv6 and earlier :

  rsb     r3, r0, #0
  and     r0, r3, r0
  clz     r0, r0
  rsb     r0, r0, #31
  

  For ARMv7 (in fact v6T2), however, using the new bit-reversal instruction would have been better :

  rbit    r0, r0
  clz     r0, r0
  

  I suspect this is simply a matter of the function not yet having been updated for ARMv7, which is perhaps even excusable given the relatively rare use cases for it.

  The byte-reversal functions are where it gets shocking. Rather than use the REV instruction found from ARMv6 on, both of them generate external calls to __bswapsi2 and __bswapdi2 in libgcc, which is plain C code :

  SItype
  __bswapsi2 (SItype u)
  
    return ((((u) & 0xff000000) >> 24)
            | (((u) & 0x00ff0000) >>  8)
            | (((u) & 0x0000ff00) <<  8)
            | (((u) & 0x000000ff) << 24)) ;
  
  DItype
  
  __bswapdi2 (DItype u)
  
  
  
     return ((((u) & 0xff00000000000000ull) >> 56)
  
            | (((u) & 0x00ff000000000000ull) >> 40)
  
            | (((u) & 0x0000ff0000000000ull) >> 24)
  
            | (((u) & 0x000000ff00000000ull) >>  8)
  
            | (((u) & 0x00000000ff000000ull) <<  8)
  
            | (((u) & 0x0000000000ff0000ull) << 24)
  
            | (((u) & 0x000000000000ff00ull) << 40)
  
            | (((u) & 0x00000000000000ffull) << 56)) ;
  
  
  

  While the 32-bit version compiles to a reasonable-looking shift/mask/or job, the 64-bit one is a real WTF. Brace yourselves :

  push    r4, r5, r6, r7, r8, r9, sl, fp
  mov     r5, #0
  mov     r6, #65280 ; 0xff00
  sub     sp, sp, #40 ; 0x28
  and     r7, r0, r5
  and     r8, r1, r6
  str     r7, [sp, #8]
  str     r8, [sp, #12]
  mov     r9, #0
  mov     r4, r1
  and     r5, r0, r9
  mov     sl, #255 ; 0xff
  ldr     r9, [sp, #8]
  and     r6, r4, sl
  mov     ip, #16711680 ; 0xff0000
  str     r5, [sp, #16]
  str     r6, [sp, #20]
  lsl     r2, r0, #24
  and     ip, ip, r1
  lsr     r7, r4, #24
  mov     r1, #0
  lsr     r5, r9, #24
  mov     sl, #0
  mov     r9, #-16777216 ; 0xff000000
  and     fp, r0, r9
  lsr     r6, ip, #8
  orr     r9, r7, r1
  and     ip, r4, sl
  orr     sl, r1, r2
  str     r6, [sp]
  str     r9, [sp, #32]
  str     sl, [sp, #36] ; 0x24
  add     r8, sp, #32
  ldm     r8, r7, r8
  str     r1, [sp, #4]
  ldm     sp, r9, sl
  orr     r7, r7, r9
  orr     r8, r8, sl
  str     r7, [sp, #32]
  str     r8, [sp, #36] ; 0x24
  mov     r3, r0
  mov     r7, #16711680 ; 0xff0000
  mov     r8, #0
  and     r9, r3, r7
  and     sl, r4, r8
  ldr     r0, [sp, #16]
  str     fp, [sp, #24]
  str     ip, [sp, #28]
  stm     sp, r9, sl
  ldr     r7, [sp, #20]
  ldr     sl, [sp, #12]
  ldr     fp, [sp, #12]
  ldr     r8, [sp, #28]
  lsr     r0, r0, #8
  orr     r7, r0, r7, lsl #24
  lsr     r6, sl, #24
  orr     r5, r5, fp, lsl #8
  lsl     sl, r8, #8
  mov     fp, r7
  add     r8, sp, #32
  ldm     r8, r7, r8
  orr     r6, r6, r8
  ldr     r8, [sp, #20]
  ldr     r0, [sp, #24]
  orr     r5, r5, r7
  lsr     r8, r8, #8
  orr     sl, sl, r0, lsr #24
  mov     ip, r8
  ldr     r0, [sp, #4]
  orr     fp, fp, r5
  ldr     r5, [sp, #24]
  orr     ip, ip, r6
  ldr     r6, [sp]
  lsl     r9, r5, #8
  lsl     r8, r0, #24
  orr     fp, fp, r9
  lsl     r3, r3, #8
  orr     r8, r8, r6, lsr #8
  orr     ip, ip, sl
  lsl     r7, r6, #24
  and     r5, r3, #16711680 ; 0xff0000
  orr     r7, r7, fp
  orr     r8, r8, ip
  orr     r4, r1, r7
  orr     r5, r5, r8
  mov     r9, r6
  mov     r1, r5
  mov     r0, r4
  add     sp, sp, #40 ; 0x28
  pop     r4, r5, r6, r7, r8, r9, sl, fp
  bx      lr
  

  That’s right, 91 instructions to move 8 bytes around a bit. GCC definitely has a problem with 64-bit numbers. It is perhaps worth noting that the bswap_64 macro in glibc splits the 64-bit value into 32-bit halves which are then reversed independently, thus side-stepping this weakness of gcc.

  As a side note, ARM RVCT (armcc) compiles those functions perfectly into one and two REV instructions, respectively.

  AVR32

  There is not much to report here. The latest gcc version available is 4.2.4, which doesn’t appear to have the bswap functions. The other three are handled nicely, even using a bit-reverse for __builtin_ctz.

  MIPS / MIPS64

  The situation MIPS is similar to ARM. Both bswap builtins result in external libgcc calls, the rest giving sensible code.

  PowerPC

  I scarcely believe my eyes, but this one is actually not bad. The PowerPC has no byte-reversal instructions, yet someone seems to have taken the time to teach gcc a good instruction sequence for this operation. The PowerPC does have some powerful rotate-and-mask instructions which come in handy here. First the 32-bit version :

  rotlwi  r0,r3,8
  rlwimi  r0,r3,24,0,7
  rlwimi  r0,r3,24,16,23
  mr      r3,r0
  blr
  

  The 64-bit byte-reversal simply applies the above code on each half of the value :

  rotlwi  r0,r3,8
  rlwimi  r0,r3,24,0,7
  rlwimi  r0,r3,24,16,23
  rotlwi  r3,r4,8
  rlwimi  r3,r4,24,0,7
  rlwimi  r3,r4,24,16,23
  mr      r4,r0
  blr
  

  Although I haven’t analysed that code carefully, it looks pretty good.

  PowerPC64

  Doing 64-bit operations is easier on a 64-bit CPU, right ? For you and me perhaps, but not for gcc. Here __builtin_bswap64 gives us the now familiar __bswapdi2 call, and while not as bad as the ARM version, it is not pretty :

  rldicr  r0,r3,8,55
  rldicr  r10,r3,56,7
  rldicr  r0,r0,56,15
  rldicl  r11,r3,8,56
  rldicr  r9,r3,16,47
  or      r11,r10,r11
  rldicr  r9,r9,48,23
  rldicl  r10,r0,24,40
  rldicr  r0,r3,24,39
  or      r11,r11,r10
  rldicl  r9,r9,40,24
  rldicr  r0,r0,40,31
  or      r9,r11,r9
  rlwinm  r10,r3,0,0,7
  rldicl  r0,r0,56,8
  or      r0,r9,r0
  rldicr  r10,r10,8,55
  rlwinm  r11,r3,0,8,15
  or      r0,r0,r10
  rldicr  r11,r11,24,39
  rlwinm  r3,r3,0,16,23
  or      r0,r0,r11
  rldicr  r3,r3,40,23
  or      r3,r0,r3
  blr
  

  That is 6 times longer than the (presumably) hand-written 32-bit version.

  x86 / x86_64

  As one might expect, results on x86 are good. All the tested functions use the available special instructions. One word of caution though : the bit-counting instructions are very slow on some implementations, specifically the Atom, AMD chips, and the notoriously slow Pentium4E.

  Conclusion

  In conclusion, I would say gcc builtins can be useful to avoid fragile inline assembler. Before using them, however, one should make sure they are not in fact harmful on the required targets. Not even those builtins mapping directly to CPU instructions can be trusted.

• FFmpeg, how to skip late input ?

  14 novembre 2017, par user3343357

  I’m running ffmpeg to display incoming stream on a Decklink BlackMagic card with the following command line :

  ffmpeg -y -f ourFmt -probesize 32 -i - -f decklink -preset ultrafast 
  
     -pix_fmt uyvy422 -s 1920x1080 -r 30 -af volume=0.1 -max_delay 10000 
  
       DeckLink Mini Monitor

  Basically I get the video over the internet by UDP and stream it to ffmpeg stdin. Both audio and video streams have pts and dts and are fully in sync, if the connection is good there is no problems.

  However if there are issues with the connection i start getting errors, sometimes the video delay grows significantly, and audio stops working.
  The errors i get are :

  ffmpeg : [decklink @ 0x26cc600] There are not enough buffered video
  frames. Video may misbehave ! ffmpeg : [decklink @ 0x26cc600] There’s no
  buffered audio. Audio will misbehave ! ffmpeg : Last message
  repeated 4 times ffmpeg : [decklink @ 0x26cc600] There are not enough
  buffered video frames. Video may misbehave ! ffmpeg : [decklink @
  0x26cc600] There’s no buffered audio. Audio will misbehave ! ffmpeg :
  Last message repeated 3 times ffmpeg : frame= 5204 fps= 30 q=-0.0
  size=N/A time=00:02:53.76 bitrate=N/A dup=385 drop=5 speed=0.993x
  ffmpeg : [decklink @ 0x26cc600] There’s no buffered audio. Audio will
  misbehave ! ffmpeg : Last message repeated 18 times ffmpeg :
  [decklink @ 0x26cc600] There are not enough buffered video frames.
  Video may misbehave ! ffmpeg : [decklink @ 0x26cc600] There’s no
  buffered audio. Audio will misbehave !

  The problem is when the connection is back to normal, the video keeps misbehaving until I restart the stream. What I want to do is for FFmpeg to skip to the content of the last second and play synchronized video from there, drop all the late data in between, is it possible ?

• FFMPEG Seeking with concat demuxer causes video & audio to be out of sync

  20 février 2023, par Garuuk

  I have a very simple use case that's driving me bananas.

  


  My problem and question :

  


  I'm using ffmpeg version 5.1.2 on a MacOS and i'm using ffmpeg seeking and concat demuxer to cut many 1 minute videos into 15 seconds chopped up over 12 clips where every clip is just 2 seconds from the same video (kind of like a mini teasers for the video). I would really like to not have to re-encode to make the video processing as fast as possible.

  


  First, I take each 1 minute video and cut it up into 12 clips (I do all this programmatically in python fwiw)

  


  ffmpeg -ss 0 -i input.mp4 -t 2 -c copy -y cut_1.mp4
ffmpeg -ss 4 -i input.mp4 -t 2 -c copy -y cut_2.mp4
ffmpeg -ss 8 -i input.mp4 -t 2 -c copy -y cut_3.mp4
...
...


  


  I then write all the output file names to my concat_manifest.txt

  


  file cut_1.mp4
file cut_2.mp4
...
...


  


  Then I run my concat command :

  


  ffmpeg -f concat -i concat_manifest.txt -c copy -y concat_video.mp4


  


  This works really fast but the audio and video at the stitch point get out of sync and sometimes the video just chokes & lags. It's mostly not a smooth experience.

  


  What I have tried :

  


  

  1. using the concat protocol with intermediate profiles : ffmpeg.org/wiki/Concatenate#demuxer

  


  2. Putting the -ss when I seek after the -i. This makes everything worse

  


  3. Playing around with different -ss values. This has some noticeable affects but it's not obvious why yet.

  


  4. I've also read from the ffmpeg resource regarding seeking and copying :

  


  


  Which leads me to believe that maybe because ffmpeg is using timestamps instead of frames, seeking isn't accurate using -ss when using the concat demuxer

  


  Is there a way to get concat demuxer cutting and concatenating the video where the audio is somewhat in sync with the video ?

  


  Thanks

  


  EDIT : I found an answer and i'll be posting the solution in the coming few days.