DragonFly BSD
DragonFly commits List (threaded) for 2010-12
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

Re: git: kernel - Add MONITOR/MWAIT support to the LWKT scheduler


From: Matthew Dillon <dillon@xxxxxxxxxxxxxxxxxxxx>
Date: Sun, 26 Dec 2010 11:53:41 -0800 (PST)

:This commit removes the RQF_IDLECHECK_WK_MASK check *before*
:the CLI instruction when MONITOR/MWAIT is unavailable, because
:of how C evaluates expressions.  Some strange panic in vm_map_lookup()
:started annoying me after upgrading the kernel recently, and
:bisecting showed that this is the first commit with the problem.
:
:I'm running the kernel with the following change to see if it
:fixes the panic, which usually takes several hours to be triggered.
:
:Best Regards,
:YONETANI Tomokazu

    Theoretically it shouldn't matter.  All that happens is that the
    splz() gets run while in CLI and then the WK_MASK check after that
    prevents the idle thread from halting.

    What's the strange panic in vm_map_lookup()?  The changes might be
    revealing other bugs which have been present for a longer period
    of time rather than being bugs themselves.

    In anycase, this week I will try to work through all the reported
    bugs.  I was able to reproduce the 'lockmgr: locking against myself'
    bug using your grep --mmap test.  I'll try to track that down in
    what little time I have today.  I will have a lot more time to
    work on things this coming week.

						-Matt




[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]