qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Max Reitz <mreitz@redhat.com>
To: Alberto Garcia <berto@igalia.com>, qemu-block@nongnu.org
Cc: Kevin Wolf <kwolf@redhat.com>,
	qemu-devel@nongnu.org, Markus Armbruster <armbru@redhat.com>,
	Stefan Hajnoczi <stefanha@redhat.com>,
	Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v6 12/15] blockdev: Keep track of monitor-owned BDS
Date: Mon, 9 Nov 2015 17:28:18 +0100	[thread overview]
Message-ID: <5640C9A2.7060908@redhat.com> (raw)
In-Reply-To: <w51si4fw6v4.fsf@maestria.local.igalia.com>

[-- Attachment #1: Type: text/plain, Size: 1874 bytes --]

On 09.11.2015 16:05, Alberto Garcia wrote:
> On Wed 04 Nov 2015 07:57:44 PM CET, Max Reitz wrote:
>> @@ -3519,11 +3537,18 @@ void qmp_x_blockdev_del(bool has_id, const char *id,
>>                         bdrv_get_device_or_node_name(bs));
>>              goto out;
>>          }
>> +
>> +        if (!blk && !bs->monitor_list.tqe_prev) {
>> +            error_setg(errp, "Node %s is not owned by the monitor",
>> +                       bs->node_name);
>> +            goto out;
>> +        }
>>      }
>>  
>>      if (blk) {
>>          blk_unref(blk);
>>      } else {
>> +        QTAILQ_REMOVE(&monitor_bdrv_states, bs, monitor_list);
>>          bdrv_unref(bs);
>>      }
> 
> blk_unref(blk) will also unref the BDS (if there's any), so you also
> need to update monitor_bdrv_states in that case, don't you?

If we get to blk_unref(blk), that means there is a BB attached to that
BDS (if there is any). If that BDS is monitor-owned, that means its
refcount is at least 2 (one from the BB, one from the monitor).
Therefore, blockdev-del will fail before that point.

> Anyway, wouldn't it make more sense to do this in bdrv_delete() ?

No, I don't think so. The monitor_bdrv_states is not a plain list of
BDSs, but actually contains strong references. Every time a BDS is
entered there, that reference should be counted (which we don't actually
have to call bdrv_ref() for since we're just keeping the initial
refcount of 1 in qmp_blockdev_add()), and every time a BDS is removed,
its reference should be dropped, just like is done here.

That's the issue I had with this implementation of blockdev-del. It's
just dropping references without actually having those references. Now
we at least have the reference through monitor_bdrv_states, and in the
future, we'll have such a list of monitor references for BBs, too.

Max


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]

  parent reply	other threads:[~2015-11-09 16:28 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-04 18:57 [Qemu-devel] [PATCH v6 for-2.6 00/15] block: Rework bdrv_close_all() Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 01/15] blockdev: Add missing bdrv_unref() in drive-backup Max Reitz
2015-11-09 13:21   ` Alberto Garcia
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 02/15] blockjob: Call bdrv_unref() on creation error Max Reitz
2015-11-09 13:23   ` Alberto Garcia
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 03/15] block: Release dirty bitmaps in bdrv_close() Max Reitz
2015-11-06 18:59   ` [Qemu-devel] [Qemu-block] " John Snow
2015-11-09 16:21     ` Max Reitz
2015-11-09 21:00       ` Max Reitz
2015-11-09 16:04   ` [Qemu-devel] " Kevin Wolf
2015-11-09 16:47     ` Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 04/15] iotests: Move _filter_nbd into common.filter Max Reitz
2015-11-09 16:04   ` Kevin Wolf
2015-11-09 18:17     ` Max Reitz
2015-11-09 18:20       ` Max Reitz
2015-11-10 10:25       ` Kevin Wolf
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 05/15] iotests: Make redirecting qemu's stderr optional Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 06/15] iotests: Add test for eject under NBD server Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 07/15] block: Move BDS close notifiers into BB Max Reitz
2015-11-09 16:04   ` Kevin Wolf
2015-11-09 16:50     ` Max Reitz
2015-11-09 16:59       ` Kevin Wolf
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 08/15] block: Use blk_remove_bs() in blk_delete() Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 09/15] blockdev: Use blk_remove_bs() in do_drive_del() Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 10/15] block: Make bdrv_close() static Max Reitz
2015-11-09 13:25   ` Alberto Garcia
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 11/15] block: Add list of all BlockDriverStates Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 12/15] blockdev: Keep track of monitor-owned BDS Max Reitz
2015-11-09 15:05   ` Alberto Garcia
2015-11-09 16:26     ` Kevin Wolf
2015-11-09 16:38       ` Alberto Garcia
2015-11-09 16:28     ` Max Reitz [this message]
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 13/15] block: Add blk_remove_all_bs() Max Reitz
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 14/15] block: Rewrite bdrv_close_all() Max Reitz
2015-11-05 17:15   ` Paolo Bonzini
2015-11-05 17:37     ` Max Reitz
2015-11-05 17:40       ` Paolo Bonzini
2015-11-05 17:44         ` Eric Blake
2015-11-05 17:54           ` Paolo Bonzini
2015-11-04 18:57 ` [Qemu-devel] [PATCH v6 15/15] iotests: Add test for multiple BB on BDS tree Max Reitz
2015-11-09 16:03 ` [Qemu-devel] [PATCH v6 for-2.6 00/15] block: Rework bdrv_close_all() Kevin Wolf

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=5640C9A2.7060908@redhat.com \
    --to=mreitz@redhat.com \
    --cc=armbru@redhat.com \
    --cc=berto@igalia.com \
    --cc=kwolf@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-block@nongnu.org \
    --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).