On 9/11/05, Matthias Scheler <tron@zhadum.de> wrote:
> On Sun, Sep 11, 2005 at 07:06:24PM +0200, Julio M. Merino Vidal wrote:
> > Yes, sure it's a bug, although it looks certainly strange.  I have to
> > review the parts that control file system and node usage, because
> > they are not efficient nor correct (as seems from this bug).
>=20
> Is it possible that your memory allocate fails to grab a new page if
> all the free pages are used for the buffer cache? That machine is
> configured to use a lot of memorfy for the buffer cache.

The thing is that there are checks all around to ensure that memory
allocations went correctly (i.e., they didn't return NULL); this includes
calls to malloc as well as to pool_get.  If the allocation failed, it shoul=
d
be trapped by one of these assertions...

... unless malloc/pool_get can return an invalid non-null pointer.  Is
this the case?  (I don't think so but...)

--=20
Julio M. Merino Vidal <jmmv84@gmail.com>
http://www.livejournal.com/users/jmmv/
The NetBSD Project - http://www.NetBSD.org/