From: Kevin Wolf <kwolf@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: qemu-block@nongnu.org, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PULL 21/31] block: Avoid bs->blk in bdrv_next()
Date: Fri, 20 May 2016 10:10:17 +0200 [thread overview]
Message-ID: <20160520081017.GC4861@noname.redhat.com> (raw)
In-Reply-To: <20160520080549.GB4861@noname.redhat.com>
Am 20.05.2016 um 10:05 hat Kevin Wolf geschrieben:
> Am 20.05.2016 um 09:54 hat Paolo Bonzini geschrieben:
> > > +/* Iterates over all top-level BlockDriverStates, i.e. BDSs that are owned by
> > > + * the monitor or attached to a BlockBackend */
> > > +BdrvNextIterator *bdrv_next(BdrvNextIterator *it, BlockDriverState **bs)
> > > +{
> > > + if (!it) {
> > > + it = g_new(BdrvNextIterator, 1);
> >
> > Hmm, who frees it? (Especially if the caller exits the loop
> > prematurely, which means you cannot just free it before returning NULL).
> > I think it's better to:
> >
> > - allocate the iterator on the stack and make bdrv_next return a BDS *
> >
> > - and add a bdrv_first function that does this:
> >
> > > + *it = (BdrvNextIterator) {
> > > + .phase = BDRV_NEXT_BACKEND_ROOTS,
> > > + };
> >
> > and then returns bdrv_next(it);
> >
> > - if desirable add a macro that abstracts the calls to bdrv_first and
> > bdrv_next.
>
> Already posted a fix. I chose to keep the interface and free the
> BdrvNextIterator inside bdrv_next(), when we return NULL after the last
> element.
Oops, should have actually read your email... You're right about callers
that prematurely exit the loop, of course.
I still don't really like first/next interfaces, though. Perhaps start
the iteration with bs == NULL instead of it == NULL?
Kevin
next prev parent reply other threads:[~2016-05-20 8:10 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-05-19 15:21 [Qemu-devel] [PULL 00/31] Block layer patches Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 01/31] block: Make sure throttled BDSes always have a BB Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 02/31] block: Introduce BlockBackendPublic Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 03/31] block: throttle-groups: Use BlockBackend pointers internally Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 04/31] block: Convert throttle_group_get_name() to BlockBackend Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 05/31] block: Move throttling fields from BDS to BB Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 06/31] block: Move actual I/O throttling to BlockBackend Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 07/31] block: Move I/O throttling configuration functions " Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 08/31] block: Introduce BdrvChild.opaque Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 09/31] block: Drain throttling queue with BdrvChild callback Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 10/31] block/io: Quiesce parents between drained_begin/end Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 11/31] block: Decouple throttling from BlockDriverState Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 12/31] block: Remove bdrv_move_feature_fields() Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 13/31] Revert "block: Forbid I/O throttling on nodes with multiple parents for 2.6" Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 14/31] block: Don't check throttled reqs in bdrv_requests_pending() Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 15/31] block: Use BdrvChild callbacks for change_media/resize Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 16/31] block: User BdrvChild callback for device name Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 17/31] blockjob: Don't set iostatus of target Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 18/31] blockjob: Don't touch BDS iostatus Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 19/31] block: Remove bdrv_aio_multiwrite() Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 20/31] block: Add bdrv_has_blk() Kevin Wolf
2016-05-19 15:21 ` [Qemu-devel] [PULL 21/31] block: Avoid bs->blk in bdrv_next() Kevin Wolf
2016-05-20 7:54 ` Paolo Bonzini
2016-05-20 8:05 ` Kevin Wolf
2016-05-20 8:10 ` Kevin Wolf [this message]
2016-05-20 9:39 ` Paolo Bonzini
2016-05-20 10:26 ` Kevin Wolf
2016-05-20 10:46 ` Paolo Bonzini
2016-05-19 15:22 ` [Qemu-devel] [PULL 22/31] block: Don't return throttling info in query-named-block-nodes Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 23/31] block: Remove BlockDriverState.blk Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 24/31] block: Propagate AioContext change to all children Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 25/31] qcow2: fix condition in is_zero_cluster Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 26/31] qcow2: Fix write_zeroes with partially allocated backing file cluster Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 27/31] qemu-iotests: Some more write_zeroes tests Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 28/31] block: clarify error message for qmp-eject Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 29/31] qemu-io: Fix recent UI updates Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 30/31] qemu-iotests: Simplify 109 with unaligned qemu-img compare Kevin Wolf
2016-05-19 15:22 ` [Qemu-devel] [PULL 31/31] qemu-iotests: Fix regression in 136 on aio_read invalid Kevin Wolf
2016-05-19 16:40 ` [Qemu-devel] [PULL 00/31] Block layer patches Peter Maydell
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=20160520081017.GC4861@noname.redhat.com \
--to=kwolf@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
/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).