Hi Jeremy, thanks for taking care. It seems the problem has been fixed in the meantime.
Firstly, shortly after opening this bug I found a nice workaround: Using the BFS scheduler (2.6.31-13bfsbfq1-generic from the PPA https://launchpad.net/~darxus/+archive/bfsbfq) everything was as fast as expected. I have since compiled my own kernels with the CK patchset to get at newer kernel versions. In between, out of curiosity I tested some of the Karmic kernels that the automatic upgrades entailed and found 2.6.31.14 still behaving badly. I seem to remember that 2.6.31.20 was still bad, too. Later I found 2.6.32-020632rc7 (from mainline) to be OK. Today, triggered by your mail, I have tested 2.6.33-020633-generic #020633 SMP and 2.6.34-999-generic #201004031003 SMP and both don't show the issue. (Everything except the kernel is still Karmic.) So, it seems from 2.6.32 on the issue has gone away. Thanks again and regards, Lutz ** Tags removed: needs-kernel-logs needs-upstream-testing -- Severe disk performance regression in 2.6.31-11.38 https://bugs.launchpad.net/bugs/442232 You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
