Announcement

Collapse
No announcement yet.

Linux+Mesa Git Remains Problematic For Some Regressed R9 290 GPUs

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

  • Linux+Mesa Git Remains Problematic For Some Regressed R9 290 GPUs

    Phoronix: Linux+Mesa Git Remains Problematic For Some Regressed R9 290 GPUs

    Last week a Mesa fix landed to target the Radeon R9 290/390 performance regression that's been covered a few time on Phoronix since the issue was originally noticed. While the Mesa fix is working for some users, it didn't fix all problems, including with my Hawaii test card...

    Phoronix, Linux Hardware Reviews, Linux hardware benchmarks, Linux server benchmarks, Linux benchmarking, Desktop Linux, Linux performance, Open Source graphics, Linux How To, Ubuntu benchmarks, Ubuntu hardware, Phoronix Test Suite

  • #2
    Can anyone who can reproduce this bug please bisect the kernel?

    Comment


    • #3
      Originally posted by agd5f View Post
      Can anyone who can reproduce this bug please bisect the kernel?
      I thought that was already done in the bug..

      Comment


      • #4
        Originally posted by Azpegath View Post

        I thought that was already done in the bug..
        I think they fixed that particular one. Or at least fixed it to where most 290s are performing where they should. If it fixed most it's likely an issue somewhere else, but you can't know for sure without a bisect.

        Comment


        • #5
          Originally posted by phoronix View Post
          with the newest Padoka PPA, updated on Monday
          Padoka has just had a slow cycle, packages being a week old that time around. Then an update was just around the end of Monday or beginning of Tuesday. And of course it also took some time for them to get published. Is it certain that the patches needed for the test were actually in the test?

          Comment


          • #6
            Originally posted by eydee View Post

            Padoka has just had a slow cycle, packages being a week old that time around. Then an update was just around the end of Monday or beginning of Tuesday. And of course it also took some time for them to get published. Is it certain that the patches needed for the test were actually in the test?
            Yes, specifically waited.
            Michael Larabel
            https://www.michaellarabel.com/

            Comment


            • #7
              I opened the bug report, https://bugs.freedesktop.org/show_bug.cgi?id=97260 tested the last patch for it Monday night, still have the bug. But this morning (sniff), after 3 years (sniff), my beloved Sapphire Triple X R9 290 is not among us anymore... One minute of silence please:
              .
              .
              .

              Okay, I am cool now.

              So I cannot test it anymore.

              agd5f, in the bug report it was already bisected, but the fault was not in the Kernel. It only exposed a new feature that mesa has trouble making it work for the R9 290.

              Comment


              • #8
                Michael, I got a message stuck in moderation.

                Comment


                • #9
                  Originally posted by M@GOid View Post
                  ...But this morning (sniff), after 3 years (sniff), my beloved Sapphire Triple X R9 290 is not among us anymore...
                  Funny enough, shortly after this patch was announced, my 390X died. Thankfully it was still under warranty and I was just contacted yesterday that they are sending a replacement. I plan on testing it out as soon as it comes back.

                  Comment


                  • #10
                    Originally posted by Azpegath View Post

                    I thought that was already done in the bug..
                    Maybe this bug is due to a different regression introduced around the same time, that affects slightly different hardware.

                    Comment

                    Working...
                    X