Announcement

Collapse
No announcement yet.

OpenGL 3.3 Support Lands For RadeonSI Gallium3D

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Guest
    Guest replied
    Originally posted by Pontostroy View Post
    for mesa,xorg,etc
    Code:
    zypper ar http://download.opensuse.org/repositories/home:/pontostroy:/X11/openSUSE_13.1/  pont_x11
    zypper ref
    zypper dup --from pont_x11
    kernel
    Code:
    zypper ar  http://download.opensuse.org/repositories/home:/pontostroy:/drm-next/openSUSE_13.1/ kern
    zypper ref
    zypper up
    Finally got around to doing this. Alternatively, could I just add those two repos, up their priority over the openSUSE repos (from 99 to 98), and then just do zypper dup?

    Leave a comment:


  • Rakot
    replied
    Originally posted by Pontostroy View Post
    No, intel and radeonsi very stable for my 7790 and hd2500, but DRI_PRIME don't work. I had gpu hangs and other kernel related problem, didn't see X segfaults in a long time.

    Can you provide some logs or backtrace?
    [ 6082.060] (EE)
    [ 6082.060] (EE) Backtrace:
    [ 6082.060] (EE) 0: /usr/bin/Xorg (xorg_backtrace+0x48) [0x58be18]
    [ 6082.060] (EE) 1: /usr/bin/Xorg (0x400000+0x18fb09) [0x58fb09]
    [ 6082.060] (EE) 2: /lib64/libpthread.so.0 (0x7fbab8cb3000+0xf9f0)
    [0x7fbab8cc29f0]
    [ 6082.060] (EE) 3: /usr/lib64/xorg/modules/drivers/intel_drv.so
    (0x7fbab32ea000+0x2a28a) [0x7fbab331428a]
    [ 6082.060] (EE) 4: /usr/lib64/xorg/modules/drivers/intel_drv.so
    (0x7fbab32ea000+0xd4bf1) [0x7fbab33bebf1]
    [ 6082.060] (EE) 5: /usr/lib64/xorg/modules/drivers/intel_drv.so
    (0x7fbab32ea000+0xd5695) [0x7fbab33bf695]
    [ 6082.060] (EE) 6: /usr/lib64/xorg/modules/drivers/intel_drv.so
    (0x7fbab32ea000+0xdb9d9) [0x7fbab33c59d9]
    [ 6082.060] (EE) 7: /usr/lib64/xorg/modules/drivers/intel_drv.so
    (0x7fbab32ea000+0x4fab1) [0x7fbab3339ab1]
    [ 6082.060] (EE) 8: /usr/lib64/xorg/modules/drivers/intel_drv.so
    (0x7fbab32ea000+0x31554) [0x7fbab331b554]
    [ 6082.060] (EE) 9: /usr/lib64/xorg/modules/drivers/intel_drv.so
    (0x7fbab32ea000+0x3184b) [0x7fbab331b84b]
    [ 6082.061] (EE) 10: /usr/bin/Xorg (0x400000+0x1195fa) [0x5195fa]
    [ 6082.061] (EE) 11: /usr/bin/Xorg (0x400000+0x39791) [0x439791]
    [ 6082.061] (EE) 12: /usr/bin/Xorg (0x400000+0x3d23e) [0x43d23e]
    [ 6082.061] (EE) 13: /usr/bin/Xorg (0x400000+0x4106a) [0x44106a]
    [ 6082.061] (EE) 14: /lib64/libc.so.6 (__libc_start_main+0xf5)
    [0x7fbab7920be5]
    [ 6082.061] (EE) 15: /usr/bin/Xorg (0x400000+0x2c601) [0x42c601]
    [ 6082.061] (EE)
    [ 6082.061] (EE) Segmentation fault at address 0x18
    [ 6082.061] (EE)
    Fatal server error:
    [ 6082.061] (EE) Caught signal 11 (Segmentation fault). Server aborting
    [ 6082.061] (EE)
    [ 6082.061] (EE)
    Please consult the The X.Org Foundation support
    at http://wiki.x.org
    for help.
    [ 6082.061] (EE) Please also check the log file at "/var/log/Xorg.0.log" for
    additional information.
    [ 6082.061] (EE)
    [ 6082.061] (II) AIGLX: Suspending AIGLX clients for VT switch
    [ 6082.183] (EE) Server terminated with error (1). Closing log file.

    Something like this was yesterday. Also this crash happened today. Once I updated x-server from your repo everything looks fine so far. Hopefully the issue is resolved.
    As for DRI_PRIME it works fine on my muxless laptop. There is an issue with intel driver in 3.13 (bug report was filled), radeonsi shows an error (another bug report exists) and runpm doesn't work properly (again there is a bug report). Once all this stuff is fixed I can say that open source drivers are in much-much better state than catalyst.

    Leave a comment:


  • FutureSuture
    replied
    Originally posted by Tyler_K View Post
    Well all you who pay much more attention to the nitty gritty details of the devices would know better then I (about the position the device takes in the performance/capabilities echelon), but see Pontostroy's results here with a 7790:



    Also note Alex's suggestion just further down in the thread in regards to turning on tiling.

    Given those pts, I'd guess that 24fps isn't rubbish at all .... (again, I'm not an expert as to which cards are uber fast and which ones are not)
    I see what you mean. Maybe I was underestimating Uniengine Valley.

    Leave a comment:


  • Tyler_K
    replied
    Originally posted by FutureSuture View Post
    24 frames per second with a Sapphire AMD Radeon HD 7870 GHz Edition card? Sorry, but that sounds completely rubbish to me. What am I misunderstanding?
    Well all you who pay much more attention to the nitty gritty details of the devices would know better then I (about the position the device takes in the performance/capabilities echelon), but see Pontostroy's results here with a 7790:



    Also note Alex's suggestion just further down in the thread in regards to turning on tiling.

    Given those pts, I'd guess that 24fps isn't rubbish at all .... (again, I'm not an expert as to which cards are uber fast and which ones are not)

    Leave a comment:


  • FutureSuture
    replied
    Originally posted by lordheavy View Post
    Unigine Heaven with phoronix test suite http://openbenchmarking.org/result/1...SO-HOME1575701

    Really stable, now i cn play oilrush
    24 frames per second with a Sapphire AMD Radeon HD 7870 GHz Edition card? Sorry, but that sounds completely rubbish to me. What am I misunderstanding?

    Leave a comment:


  • V10lator
    replied
    Originally posted by Tom_G View Post
    Hi

    MLL works out of the box, but there are rendering issues. Some people [1] discussed about and asked the publisher of the game for a patch. It seems, MLL has obviously some bugs in the shaders. For details (in german) read this [2].

    Tom

    [1] http://steamcommunity.com/app/43160/...6998172578/#p5
    [2] http://pastie.org/8623348
    Just wanted to point out that V10lator is me and tell you guys that sadly I didn't get any more replys after the mails showed on pastie.

    Originally posted by Pontostroy View Post
    Metro LL has text render problem, i can't find and start new game or change options.

    https://bugs.freedesktop.org/show_bug.cgi?id=71870
    The bug report is for intel. Even when the bugs are the same there's a separate report for radeon:


    But I guess these bug reports will be closed when the mesa devs are 100% sure it's the fault of the game (I'm 99% sure about that). So let's hope the Metro devs are listening and working on it.

    Leave a comment:


  • Tom_G
    replied
    Hi

    MLL works out of the box, but there are rendering issues. Some people [1] discussed about and asked the publisher of the game for a patch. It seems, MLL has obviously some bugs in the shaders. For details (in german) read this [2].

    Tom

    [1] http://steamcommunity.com/app/43160/...6998172578/#p5
    [2] http://pastie.org/8623348

    Leave a comment:


  • Pontostroy
    replied
    Originally posted by dietrdan View Post
    I just want to say thanks to every developer involved to make that happen

    Has anybody already tried Metro Last Light now? I haven't purchased it yet, but would like to know if its playable now with radeonsi.
    Metro LL has text render problem, i can't find and start new game or change options.

    Leave a comment:


  • dietrdan
    replied
    I just want to say thanks to every developer involved to make that happen

    Has anybody already tried Metro Last Light now? I haven't purchased it yet, but would like to know if its playable now with radeonsi.

    Leave a comment:


  • lordheavy
    replied
    Unigine Heaven with phoronix test suite http://openbenchmarking.org/result/1...SO-HOME1575701

    Really stable, now i cn play oilrush

    Leave a comment:

Working...
X