Re: dogfooding over in clusteradm land
--Kg9bXG3rqf9MxIZW
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Tue, Jan 03, 2012 at 01:45:26AM -0800, Don Lewis wrote:
> On 3 Jan, Kostik Belousov wrote:
>=20
> > This sounds very plausible. I think that there is no sense in restarting
> > the scan if it is requested in async mode at all. See below.
> >=20
> > Would be thrilled if this finally solves the svn2cvs issues.
> >=20
> > commit 41aaafe5e3be5387949f303b8766da64ee4a521f
> > Author: Kostik Belousov <kostik@sirion>
> > Date: Tue Jan 3 11:16:30 2012 +0200
> >=20
> > Do not restart the scan in vm_object_page_clean() if requested
> > mode is async.
> > =20
> > Proposed by: truckman
> >=20
> > diff --git a/sys/vm/vm_object.c b/sys/vm/vm_object.c
> > index 716916f..52fc08b 100644
> > --- a/sys/vm/vm_object.c
> > +++ b/sys/vm/vm_object.c
> > @@ -841,7 +841,8 @@ rescan:
> > if (p->valid =3D=3D 0)
> > continue;
> > if (vm_page_sleep_if_busy(p, TRUE, "vpcwai")) {
> > - if (object->generation !=3D curgeneration)
> > + if ((flags & OBJPC_SYNC) !=3D 0 &&
> > + object->generation !=3D curgeneration)
> > goto rescan;
> > np =3D vm_page_find_least(object, pi);
> > continue;
>=20
> I wonder if it would make more sense to just skip the busy pages in
> async mode instead of sleeping ...
>=20
It would be too much weakening the guarantee of the vfs_msync(MNT_NOWAIT)
to not write such pages, IMO. Busy state indeed means that the page most
likely undergoing the i/o, but in case it is not, we would not write it
at all.
Lets see whether the change alone helps. Do you agree ?
--Kg9bXG3rqf9MxIZW
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.18 (FreeBSD)
iEYEARECAAYFAk8C178ACgkQC3+MBN1Mb4hHgACdE/rxPa1jc/HtDf0e5lllCt1F
5oMAn0w8LJtV/gHa1ogbWvGypHTkxiAF
=f+AK
-----END PGP SIGNATURE-----
--Kg9bXG3rqf9MxIZW--
討論串 (同標題文章)
完整討論串 (本文為第 15 之 20 篇)：