qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Lukas Straub <lukasstraub2@web.de>
To: "Zhang, Chen" <chen.zhang@intel.com>
Cc: "Marc-André Lureau" <marcandre.lureau@redhat.com>,
	"Jason Wang" <jasowang@redhat.com>,
	qemu-devel <qemu-devel@nongnu.org>,
	"Li Zhijian" <lizhijian@cn.fujitsu.com>,
	"Paolo Bonzini" <pbonzini@redhat.com>
Subject: Re: [PATCH v4 6/6] net/colo-compare.c: Correct ordering in complete and finalize
Date: Thu, 7 May 2020 18:09:09 +0200	[thread overview]
Message-ID: <20200507180909.06165172@luklap> (raw)
In-Reply-To: <83c26ab24d28480d94c45c0a0f49d130@intel.com>

[-- Attachment #1: Type: text/plain, Size: 4463 bytes --]

On Thu, 7 May 2020 13:26:11 +0000
"Zhang, Chen" <chen.zhang@intel.com> wrote:

> > -----Original Message-----
> > From: Lukas Straub <lukasstraub2@web.de>
> > Sent: Monday, May 4, 2020 6:28 PM
> > To: qemu-devel <qemu-devel@nongnu.org>
> > Cc: Zhang, Chen <chen.zhang@intel.com>; Li Zhijian
> > <lizhijian@cn.fujitsu.com>; Jason Wang <jasowang@redhat.com>; Marc-
> > André Lureau <marcandre.lureau@redhat.com>; Paolo Bonzini
> > <pbonzini@redhat.com>
> > Subject: [PATCH v4 6/6] net/colo-compare.c: Correct ordering in complete
> > and finalize
> > 
> > In colo_compare_complete, insert CompareState into net_compares only
> > after everything has been initialized.
> > In colo_compare_finalize, remove CompareState from net_compares before
> > anything is deinitialized.  
> 
> S/deinitialized/finalized
> 
> It looks no dependences on each step on initialization and finalization.
> Do you means we just need add/remove each colo-compare module at last in logic?

Yes. While I didn't see any crashes here, there is the possibility that if colo-compare is removed during checkpoint, the destroyed event_bh is called from colo_notify_compares_event. Same with colo_compare_complete (very unlikely) if colo-compare is created while colo is running, colo_notify_compares_event may call the uninitialized event_bh.

Regards,
Lukas Straub

> Or current code have some issue?
> 
> Thanks
> Zhang Chen
> 
> > 
> > Signed-off-by: Lukas Straub <lukasstraub2@web.de>
> > ---
> >  net/colo-compare.c | 45 +++++++++++++++++++++++----------------------
> >  1 file changed, 23 insertions(+), 22 deletions(-)
> > 
> > diff --git a/net/colo-compare.c b/net/colo-compare.c index
> > c7572d75e9..6f80bcece6 100644
> > --- a/net/colo-compare.c
> > +++ b/net/colo-compare.c
> > @@ -1283,15 +1283,6 @@ static void
> > colo_compare_complete(UserCreatable *uc, Error **errp)
> >                             s->vnet_hdr);
> >      }
> > 
> > -    qemu_mutex_lock(&colo_compare_mutex);
> > -    if (!colo_compare_active) {
> > -        qemu_mutex_init(&event_mtx);
> > -        qemu_cond_init(&event_complete_cond);
> > -        colo_compare_active = true;
> > -    }
> > -    QTAILQ_INSERT_TAIL(&net_compares, s, next);
> > -    qemu_mutex_unlock(&colo_compare_mutex);
> > -
> >      s->out_sendco.s = s;
> >      s->out_sendco.chr = &s->chr_out;
> >      s->out_sendco.notify_remote_frame = false; @@ -1312,6 +1303,16 @@
> > static void colo_compare_complete(UserCreatable *uc, Error **errp)
> >                                                        connection_destroy);
> > 
> >      colo_compare_iothread(s);
> > +
> > +    qemu_mutex_lock(&colo_compare_mutex);
> > +    if (!colo_compare_active) {
> > +        qemu_mutex_init(&event_mtx);
> > +        qemu_cond_init(&event_complete_cond);
> > +        colo_compare_active = true;
> > +    }
> > +    QTAILQ_INSERT_TAIL(&net_compares, s, next);
> > +    qemu_mutex_unlock(&colo_compare_mutex);
> > +
> >      return;
> >  }
> > 
> > @@ -1384,19 +1385,6 @@ static void colo_compare_finalize(Object *obj)
> >      CompareState *s = COLO_COMPARE(obj);
> >      CompareState *tmp = NULL;
> > 
> > -    qemu_chr_fe_deinit(&s->chr_pri_in, false);
> > -    qemu_chr_fe_deinit(&s->chr_sec_in, false);
> > -    qemu_chr_fe_deinit(&s->chr_out, false);
> > -    if (s->notify_dev) {
> > -        qemu_chr_fe_deinit(&s->chr_notify_dev, false);
> > -    }
> > -
> > -    if (s->iothread) {
> > -        colo_compare_timer_del(s);
> > -    }
> > -
> > -    qemu_bh_delete(s->event_bh);
> > -
> >      qemu_mutex_lock(&colo_compare_mutex);
> >      QTAILQ_FOREACH(tmp, &net_compares, next) {
> >          if (tmp == s) {
> > @@ -1411,6 +1399,19 @@ static void colo_compare_finalize(Object *obj)
> >      }
> >      qemu_mutex_unlock(&colo_compare_mutex);
> > 
> > +    qemu_chr_fe_deinit(&s->chr_pri_in, false);
> > +    qemu_chr_fe_deinit(&s->chr_sec_in, false);
> > +    qemu_chr_fe_deinit(&s->chr_out, false);
> > +    if (s->notify_dev) {
> > +        qemu_chr_fe_deinit(&s->chr_notify_dev, false);
> > +    }
> > +
> > +    if (s->iothread) {
> > +        colo_compare_timer_del(s);
> > +    }
> > +
> > +    qemu_bh_delete(s->event_bh);
> > +
> >      AioContext *ctx = iothread_get_aio_context(s->iothread);
> >      aio_context_acquire(ctx);
> >      AIO_WAIT_WHILE(ctx, !s->out_sendco.done);
> > --
> > 2.20.1  


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

  reply	other threads:[~2020-05-07 16:13 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-04 10:27 [PATCH v4 0/6] colo-compare bugfixes Lukas Straub
2020-05-04 10:28 ` [PATCH v4 1/6] net/colo-compare.c: Create event_bh with the right AioContext Lukas Straub
2020-05-04 10:28 ` [PATCH v4 2/6] chardev/char.c: Use qemu_co_sleep_ns if in coroutine Lukas Straub
2020-05-04 10:28 ` [PATCH v4 3/6] net/colo-compare.c: Fix deadlock in compare_chr_send Lukas Straub
2020-05-07 11:00   ` Zhang, Chen
2020-05-07 15:51     ` Lukas Straub
2020-05-08  2:19       ` Zhang, Chen
2020-05-08  6:08         ` Lukas Straub
2020-05-08  6:28           ` Zhang, Chen
2020-05-08  7:56             ` Lukas Straub
2020-05-11  8:30               ` Zhang, Chen
2020-05-04 10:28 ` [PATCH v4 4/6] net/colo-compare.c: Only hexdump packets if tracing is enabled Lukas Straub
2020-05-04 11:27   ` Philippe Mathieu-Daudé
2020-05-04 11:58     ` Philippe Mathieu-Daudé
2020-05-04 10:28 ` [PATCH v4 5/6] net/colo-compare.c, softmmu/vl.c: Check that colo-compare is active Lukas Straub
2020-05-07 11:38   ` Zhang, Chen
2020-05-07 15:54     ` Lukas Straub
2020-05-08  2:26       ` Zhang, Chen
2020-05-08  3:55         ` Derek Su
2020-05-08  6:10         ` Lukas Straub
2020-05-08  6:50           ` Zhang, Chen
2020-05-09 12:21             ` Lukas Straub
2020-05-11  8:49               ` Zhang, Chen
2020-05-12 17:28             ` Dr. David Alan Gilbert
2020-05-04 10:28 ` [PATCH v4 6/6] net/colo-compare.c: Correct ordering in complete and finalize Lukas Straub
2020-05-07 13:26   ` Zhang, Chen
2020-05-07 16:09     ` Lukas Straub [this message]
2020-05-08  3:01       ` Zhang, Chen
2020-05-11  8:33       ` Zhang, Chen
2020-05-15  9:15 ` [PATCH v4 0/6] colo-compare bugfixes Zhang, Chen

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=20200507180909.06165172@luklap \
    --to=lukasstraub2@web.de \
    --cc=chen.zhang@intel.com \
    --cc=jasowang@redhat.com \
    --cc=lizhijian@cn.fujitsu.com \
    --cc=marcandre.lureau@redhat.com \
    --cc=pbonzini@redhat.com \
    --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).