Page 1 of 2 12 LastLast
Results 1 to 10 of 18

Thread: AMD R600g LLVM Back-End Is Working A Bit Better

  1. #1
    Join Date
    Jan 2007
    Posts
    14,787

    Default AMD R600g LLVM Back-End Is Working A Bit Better

    Phoronix: AMD R600g LLVM Back-End Is Working A Bit Better

    The R600g LLVM shader compiler back-end that's primarily intended for the Radeon Gallium3D compute support is now working a bit better for graphics support compared to when it was first committed...

    http://www.phoronix.com/vr.php?view=MTExNzc

  2. #2
    Join Date
    Aug 2009
    Location
    Russe, Bulgaria
    Posts
    511

    Default

    I believe when VLIW packetizer is finished, the performance will be greatly enhanced.

  3. #3
    Join Date
    Jan 2007
    Posts
    459

    Default

    "
    Granted, the tests were just measuring the in-game frame-rates and not the shader compiler time"
    it seems a little pointless to not have/provide a benchmark result for shader compiler time/speed in this case or how is anyone expected to estimate how many CPU cycles have now been saved for use elsewhere in the system! etc.

  4. #4
    Join Date
    Feb 2008
    Location
    Linuxland
    Posts
    5,102

    Default

    The shader compile time is uninteresting, the speed of the outputted shader is not. Michael benched the right thing.

    It really does not matter if a game takes 0.3s more or less to load.

  5. #5
    Join Date
    Nov 2008
    Location
    Madison, WI, USA
    Posts
    874

    Default

    Quote Originally Posted by curaga View Post
    The shader compile time is uninteresting, the speed of the outputted shader is not. Michael benched the right thing.

    It really does not matter if a game takes 0.3s more or less to load.
    Wasn't Valve reportedly having issues with recompile times of dynamic/changing shaders during game run-time? /nitpick

  6. #6
    Join Date
    Feb 2008
    Location
    Linuxland
    Posts
    5,102

    Default

    That should be a) threaded b) cached and c) the complaint was about fglrx, IIRC, not mesa.

  7. #7
    Join Date
    Jan 2007
    Posts
    459

    Default

    Quote Originally Posted by curaga View Post
    The shader compile time is uninteresting, the speed of the outputted shader is not. Michael benched the right thing.

    It really does not matter if a game takes 0.3s more or less to load.
    OC i agree that it makes no difference at all that a game takes a few micro seconds more to load or not
    but in this case the code is primarily for Gallium3D compute support and all those saved microseconds per routine cycle do add up and so output speed and cycles saved is interesting overall (at least to me) as an initial estimate.

  8. #8
    Join Date
    Nov 2008
    Location
    Madison, WI, USA
    Posts
    874

    Default

    Quote Originally Posted by curaga View Post
    That should be a) threaded b) cached and c) the complaint was about fglrx, IIRC, not mesa.
    N/M then. I thought their complaints were centered around Mesa + shader recompile times. *Goes and grabs the first cup of coffee of the day*

  9. #9
    Join Date
    Nov 2008
    Location
    Germany
    Posts
    5,411

    Default

    last time i checked the compiler it cashes HON i hope they do some stability fixes to.

  10. #10
    Join Date
    Jun 2010
    Location
    ฿ 16LDJ6Hrd1oN3nCoFL7BypHSEYL84ca1JR
    Posts
    1,052

    Default

    Can I check which shader compiler is really used?

    I (HD 6550M) have built mesa with --enable-r600-llvm-compiler and R600_LLVM set to 1. kwin with "opengl2 shaders" and xonotic work absolutely fine. By the way, xonotic on the "ultra" preset runs pretty good, so the radeon guys are doing something right.

    edit: I guess I am using it. Portal 2 in wine with 32 bit crashes:
    Code:
    Backtrace:
    =>0 0xf59deca4 in r600_dri.so (+0x14bca4) (0x0000027e)
      1 0xf5f95e10 in r600_dri.so (+0x702e0f) (0x7d6456ec)
      2 0xf5ee80ce _ZN4llvm19MachineFunctionPass13runOnFunctionERNS_8FunctionE+0x7d() in r600_dri.so (0xf696f0e8)
      3 0xf63a62b3 _ZN4llvm13FPPassManager13runOnFunctionERNS_8FunctionE+0x272() in r600_dri.so (0x7d62e538)
      4 0xf63a630c _ZN4llvm13FPPassManager11runOnModuleERNS_6ModuleE+0x4b() in r600_dri.so (0x7d560068)
      5 0xf63a5f04 _ZN4llvm13MPPassManager11runOnModuleERNS_6ModuleE+0x1f3() in r600_dri.so (0x7d5c4038)
      6 0xf63a5fe9 _ZN4llvm15PassManagerImpl3runERNS_6ModuleE+0x78() in r600_dri.so (0x00000000)
      7 0xf63a6036 _ZN4llvm11PassManager3runERNS_6ModuleE+0x25() in r600_dri.so (0x00339a18)
      8 0xf59d21ef in r600_dri.so (+0x13f1ee) (0x00339a18)
      9 0xf59cfdfe in r600_dri.so (+0x13cdfd) (0x00339a90)
    <snip>
    Last edited by ChrisXY; 06-11-2012 at 10:38 AM.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •