public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* balance_dirty_pages broken
@ 2002-10-17  4:36 Doug Ledford
  2002-10-17  5:09 ` Andrew Morton
  0 siblings, 1 reply; 7+ messages in thread
From: Doug Ledford @ 2002-10-17  4:36 UTC (permalink / raw)
  To: linux-kernel

Actually, I don't know if it's balance_dirty_pages fault or some other
part of the kernel's fault, but it is broken here.  Failure mode is that
balance_dirty_pages would loop forever.  Reason it would loop forever is
because the ps struct had a negative entry for nr_dirty.  Breaking out of
the loop when ps.nr_dirty < 0 allows my machine to live.  Now, *why*
ps.nr_dirty is < 0 is another issue.  I have no clue.  Well, I have one 
clue, but no idea if it's valid or another red herring ;-)  Machine only 
has 256MB RAM, but has 4GB Highmem support enabled.  Don't know if that 
confuses the page counter stuff, having no actual highmem available.

If you need details about the machine, just ask.  This was easy for me to
trigger by running mke2fs on a scsi disk.  Would hit this every time.


-- 
  Doug Ledford <dledford@redhat.com>     919-754-3700 x44233
         Red Hat, Inc. 
         1801 Varsity Dr.
         Raleigh, NC 27606
  

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2002-10-17  6:03 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2002-10-17  4:36 balance_dirty_pages broken Doug Ledford
2002-10-17  5:09 ` Andrew Morton
2002-10-17  5:22   ` Doug Ledford
2002-10-17  5:41     ` Doug Ledford
2002-10-17  5:54       ` Andrew Morton
2002-10-17  6:01         ` Doug Ledford
2002-10-17  6:09           ` Andrew Morton

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox