Subject: Re: LFS stability: kern/36608
To: Blair Sadewitz <email@example.com>
From: Sverre Froyen <firstname.lastname@example.org>
Date: 09/22/2007 09:47:47
On Thursday 13 September 2007, you wrote:
> I suspect it masks the problem. Try this out:
> set vfs.lfs.pagetrip to something sane, such as ssize/4096/4. Each
> time you decrease it (by powers of two), try untarring pkgsrc (this is
> on a multiprocessor machine, since we're talking about locking). The
> last decrement in pagetrip should be ssize/PAGE_SIZE. Notice how, if
> you get deadlocks, etc., they seem to happen more often as you go
> lower. Now, try setting pagetrip to something one or two powers of
> two lower than ssize/PAGE_SIZE, such as 64. Now notice how quickly it
OK, dumpfs_lfs reports that ssize = 1048576 and sysctl says hw.pagesize =
4096. Thus, ssize/4096/4 = 64 and ssize/PAGE_SIZE = 256, so I set
vfs.lfs.pagetrip=64 (it was 0).
Running my bogofilter test case (see kern/36608) no longer triggers the
LOCKDEBUG assertion. I will resume regular use of bogofilter in my email
client in order to test more thoroughly.
BTW, my system is a ThinkPad T42, i.e., a uniprocessor system.
> I suspected this has to do with the lfs writer daemon ltsleeping,
> given the time it sleeps is fixed. Haven't gotten a chance to look at
> it more, though.
Do you think it is the lfs_writer daemon that makes the call to lfs_segunlock,
around line 2290 in lfs_vnops.c (see the PR). If so, your suspicion sounds
reasonable. How can I verify this?
> Any thoughts?
I still think the LFS code (see the PR) looks questionable but I've not gotten
any feedback on my comments.
Thanks for your help!