Results 1 to 5 of 5

Thread: [News] AMD Ryzen-optimized C and C++ Compilers Improve Performance

  1. #1
    Join XS BOINC Team StyM's Avatar
    Join Date
    Mar 2006
    Location
    Tropics
    Posts
    9,468

    [News] AMD Ryzen-optimized C and C++ Compilers Improve Performance

    https://www.techpowerup.com/233548/a...ve-performance

    AMD followed up its Ryzen processor launch with support for the software development ecosystem by releasing special C and C++ compilers that let you make software that can fully take advantage of the "Zen" micro-architecture. The new AOCC 1.0 C/C++ compilers by AMD are based on LLVM Clang, with "Zen" specific patches. AMD claims AOCC offers improved vectorization and better code generation for "Zen" based CPUs. It also includes a "Zen" optimized linker.

    Phoronix benchmarked AOCC against other more common compilers such as GCC 6.3, GCC 7.1, GCC 8, LLVM Clang 4.0, and LLVM Clang 5.0 using a Ryzen 7-1700 eight-core processor powered machine, running on Ubuntu 17.04 Linux, and found that AOCC offers higher performance than GCC in most cases, LLVM Clang in some cases, and marginally higher performance than LLVM Clang in some cases. Find more results in the link below.

  2. #2
    Xtreme Guru
    Join Date
    May 2007
    Location
    Ace Deuce, Michigan
    Posts
    3,955
    Wait.

    I was about to be impressed, until I saw it was based off LLVM CLang. Performance increase appears to be negligible?
    Quote Originally Posted by Hans de Vries View Post

    JF-AMD posting: IPC increases!!!!!!! How many times did I tell you!!!

    terrace215 post: IPC decreases, The more I post the more it decreases.
    terrace215 post: IPC decreases, The more I post the more it decreases.
    terrace215 post: IPC decreases, The more I post the more it decreases.
    .....}
    until (interrupt by Movieman)


    Regards, Hans

  3. #3
    Xtreme Enthusiast
    Join Date
    Feb 2010
    Posts
    578
    I'm surprised LLVM Clang performs better than GCC, period.

  4. #4
    Xtreme Addict
    Join Date
    Sep 2010
    Location
    US, MI
    Posts
    1,680
    Here's the real articles:

    http://www.phoronix.com/scan.php?pag...zen-aocc&num=1
    http://www.phoronix.com/scan.php?pag...izations&num=1

    There's nothing to really see, it's like a said before, using opts doesn't usually net that much perf, though it's usually worth it, any little bit helps.
    It looks like whatever amd is trying to do though isn't helping and is just increasing compile time.

    Besides, I don't use clang...
    I use gcc or ms's compiler...
    When you use a 3rd party compiler you have to ensure there's a project file, a working build path, etc.
    It just doesn't usually work out of the box.
    So I don't know what they're trying todo, who's gonna redo there code to compile with a compiler that has a short lifetime expectancy... ^^
    I get they're trying to get patches to clang I guess, upstreamed, in that sense ok, sure, but who actually uses clang?
    I know it's supposed to be better when it does work, and etc, but I just want my code to compile lol, ie gcc...
    Or maybe that was codebox or whatever it was called, or codeblocks, I don't remember, shows you how much I care about 3rd party compilers...

  5. #5
    Xtreme Enthusiast
    Join Date
    Feb 2010
    Posts
    578
    I had always thought Clang produced superior compile times but binaries with inferior performance/optimization. Been too long since I checked, I guess.

    And not having ICC in the mix = teh fail

Bookmarks

Bookmarks

Posting Permissions

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