From: Jeff Cody <jcody@redhat.com>
To: Fam Zheng <famz@redhat.com>
Cc: kwolf@redhat.com, pbonzini@redhat.com, qemu-devel@nongnu.org,
stefanha@redhat.com
Subject: Re: [Qemu-devel] [PATCH 8/8] nbd: use BlockDriverState refcnt
Date: Thu, 25 Jul 2013 21:29:39 -0400 [thread overview]
Message-ID: <20130726012939.GA5445@localhost.localdomain> (raw)
In-Reply-To: <20130725130141.GC19682@localhost.localdomain>
On Thu, Jul 25, 2013 at 09:01:41AM -0400, Jeff Cody wrote:
> On Thu, Jul 25, 2013 at 05:01:46PM +0800, Fam Zheng wrote:
> > Previously, nbd calls drive_get_ref() on the drive of bs. A BDS doesn't
> > always have associated dinfo, which nbd doesn't care either. We already
> > have BDS ref count, so use it to make it safe for a BDS w/o blockdev.
> >
> > Signed-off-by: Fam Zheng <famz@redhat.com>
> > ---
> > blockdev-nbd.c | 10 +---------
> > nbd.c | 5 +++++
> > 2 files changed, 6 insertions(+), 9 deletions(-)
> >
> > diff --git a/blockdev-nbd.c b/blockdev-nbd.c
> > index 95f10c8..922cf56 100644
> > --- a/blockdev-nbd.c
> > +++ b/blockdev-nbd.c
> > @@ -69,12 +69,6 @@ static void nbd_close_notifier(Notifier *n, void *data)
> > g_free(cn);
> > }
> >
> > -static void nbd_server_put_ref(NBDExport *exp)
> > -{
> > - BlockDriverState *bs = nbd_export_get_blockdev(exp);
> > - drive_put_ref(drive_get_by_blockdev(bs));
> > -}
> > -
> > void qmp_nbd_server_add(const char *device, bool has_writable, bool writable,
> > Error **errp)
> > {
> > @@ -105,11 +99,9 @@ void qmp_nbd_server_add(const char *device, bool has_writable, bool writable,
> > writable = false;
> > }
> >
> > - exp = nbd_export_new(bs, 0, -1, writable ? 0 : NBD_FLAG_READ_ONLY,
> > - nbd_server_put_ref);
> > + exp = nbd_export_new(bs, 0, -1, writable ? 0 : NBD_FLAG_READ_ONLY, NULL);
> >
> > nbd_export_set_name(exp, device);
> > - drive_get_ref(drive_get_by_blockdev(bs));
> >
> > n = g_malloc0(sizeof(NBDCloseNotifier));
> > n->n.notify = nbd_close_notifier;
> > diff --git a/nbd.c b/nbd.c
> > index 2606403..f258cdd 100644
> > --- a/nbd.c
> > +++ b/nbd.c
> > @@ -881,6 +881,7 @@ NBDExport *nbd_export_new(BlockDriverState *bs, off_t dev_offset,
> > exp->nbdflags = nbdflags;
> > exp->size = size == -1 ? bdrv_getlength(bs) : size;
> > exp->close = close;
> > + bdrv_ref(bs);
> > return exp;
> > }
> >
> > @@ -927,6 +928,10 @@ void nbd_export_close(NBDExport *exp)
> > }
> > nbd_export_set_name(exp, NULL);
> > nbd_export_put(exp);
> > + if (exp->bs) {
> > + bdrv_unref(exp->bs);
> > + exp->bs = NULL;
> > + }
>
> Same thing here - if the refcnt is ever > 1, then this leaks exp->bs.
>
Nevermind - since bs continues to exist elsewhere (and would
presumably be unrefed there), then this would not leak exp->bs.
> > }
> >
> > void nbd_export_get(NBDExport *exp)
> > --
> > 1.8.3.2
> >
> >
prev parent reply other threads:[~2013-07-26 1:30 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-25 9:01 [Qemu-devel] [PATCH 0/8] Implement reference count for BlockDriverState Fam Zheng
2013-07-25 9:01 ` [Qemu-devel] [PATCH 1/8] vvfat: use bdrv_new() to allocate BlockDriverState Fam Zheng
2013-07-25 9:01 ` [Qemu-devel] [PATCH 2/8] iscsi: use bdrv_new() instead of stack structure Fam Zheng
2013-07-25 9:01 ` [Qemu-devel] [PATCH 3/8] block: implement reference count for BlockDriverState Fam Zheng
2013-07-25 13:15 ` Jeff Cody
2013-07-26 1:13 ` Fam Zheng
2013-07-26 1:50 ` Jeff Cody
2013-07-26 1:56 ` Fam Zheng
2013-07-25 9:01 ` [Qemu-devel] [PATCH 4/8] block: make bdrv_delete() static Fam Zheng
2013-07-25 9:01 ` [Qemu-devel] [PATCH 5/8] block: use BlockDriverState refcnt for device attach/detach Fam Zheng
2013-07-25 12:49 ` Jeff Cody
2013-07-25 9:01 ` [Qemu-devel] [PATCH 6/8] migration: omit drive ref as we have bdrv_ref now Fam Zheng
2013-07-25 9:01 ` [Qemu-devel] [PATCH 7/8] xen_disk: simplify blk_disconnect with refcnt Fam Zheng
2013-07-25 12:56 ` Jeff Cody
2013-07-26 1:30 ` Jeff Cody
2013-07-25 9:01 ` [Qemu-devel] [PATCH 8/8] nbd: use BlockDriverState refcnt Fam Zheng
2013-07-25 13:01 ` Jeff Cody
2013-07-26 1:29 ` Jeff Cody [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20130726012939.GA5445@localhost.localdomain \
--to=jcody@redhat.com \
--cc=famz@redhat.com \
--cc=kwolf@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).