Date: Tue, 22 Feb 2005 08:23:09 -0800 From: Eric Anholt <eta@lclark.edu> To: Dag-Erling =?ISO-8859-1?Q?Sm=F8rgrav?= <des@des.no> Cc: amd64@freebsd.org Subject: Re: VM panics Message-ID: <1109089389.4267.7.camel@leguin> In-Reply-To: <86y8dje3p2.fsf@xps.des.no> References: <86y8dje3p2.fsf@xps.des.no>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 2005-02-20 at 14:55 +0100, Dag-Erling Sm=F8rgrav wrote:
> With recent -CURRENT, I get recurrent VM panics on amd64:
>=20
> panic: mutex vm object not owned at /usr/src/sys/vm/vm_page.c:608
>=20
> which corresponds to the following code at the beginning of in
> vm_page_remove():
>=20
> 605: mtx_assert(&vm_page_queue_mtx, MA_OWNED);
> 606: if ((object =3D m->object) =3D=3D NULL)
> 607: return;
> 608: VM_OBJECT_LOCK_ASSERT(object, MA_OWNED);
> 609: if (m->flags & PG_BUSY) {
> 610: vm_page_flag_clear(m, PG_BUSY);
> 611: vm_page_flash(m);
>=20
> This only happens under load (e.g. buildworld).
>=20
> Serial console does not work properly ("sio0: configured irq 3 not in
> bitmap of probed irqs 0"), and I don't currently have a firewire
> cable.
Backing vm/uma_core.c to r1.114 has fixed the issue for me, sufficient
to survive two savecore -f (which no kernel with 1.115 had yet), a night
of being a kaffe tinderbox, and me on the desktop sending this email.
--=20
Eric Anholt eta@lclark.edu =20
http://people.freebsd.org/~anholt/ anholt@FreeBSD.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1109089389.4267.7.camel>