Messages in this thread |  | | | Date | Fri, 2 Jul 2004 13:00:19 -0700 (PDT) | | Subject | Re: inode_unused list corruption in 2.4.26 - spin_lock problem? | | From | Chris Caputo <> |
| |
On Fri, 25 Jun 2004, Marcelo Tosatti wrote: > On Wed, Jun 23, 2004 at 06:50:48PM -0700, Chris Caputo wrote: > > Is it safe to assume that the x86 version of atomic_dec_and_lock(), which > > iput() uses, is well trusted? I figure it's got to be, but doesn't hurt > > to ask. > > Pretty sure it is, used all over. You can try to use non-optimize version > at lib/dec_and_lock.c for a test.
Hi Marcelo. Just an update...
I normally run irqbalance on these systems which are experiencing the inode_unused list corruption problem.
I disabled irqbalance on two servers which have experienced the inode_unused list corruption problem and they have now been running without crashing for considerably longer than they had been before. The length of run time without problem is not yet conclusive, but I wanted to give you a heads-up on my progress.
My current theory is that occasionally when irqbalance changes CPU affinities that the resulting set_ioapic_affinity() calls somehow cause either inter-CPU locking or cache coherency or ??? to fail.
I CC'ed Arjan because this involves irqbalance, although I'd expect the fix will be needed in set_ioapic_affinity() or related kernel code.
Any insights welcome. I am not sure what to do next except wait and see if these servers with irqbalance disabled continue to stay up.
Chris
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |