Dragonfly under KVM
Matthew Dillon
dillon at apollo.backplane.com
Sun Jun 15 19:02:13 PDT 2008
:Hello,
:
:I've applied the patch and left the system running a "make -j8
:buildworld" in a loop, cleaning out /usr/obj between each run.
:
:It has been running for over 10 hours with no lock-ups with each build
:taking ~45mins. I have also tested -j32 while taxing the disk and CPU of
:the Linux host OS.
:
:All now appears to be Ok.
:
:Thanks
:
:Gary
The patch also appears to fix the system lockups I was experiencing
testing HAMMER. My worst-case stress test ran over the weekend without
any problems.
The two things in common here are that both KVM and HAMMER (under stress)
could impose very long latencies on particular cpus, creating a
deadlock between the IPI sending code and the thread cpu-migration
code.
-Matt
Matthew Dillon
<dillon at backplane.com>
More information about the Bugs
mailing list