qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Michael S. Tsirkin" <mst@redhat.com>
To: "Xueming(Steven) Li" <xuemingl@nvidia.com>
Cc: "zhangyuwei.9149@bytedance.com" <zhangyuwei.9149@bytedance.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"tiwei.bie@intel.com" <tiwei.bie@intel.com>,
	"qemu-stable@nongnu.org" <qemu-stable@nongnu.org>
Subject: Re: [PATCH v3 1/2] vhost-user: fix VirtQ notifier cleanup
Date: Tue, 19 Oct 2021 02:57:35 -0400	[thread overview]
Message-ID: <20211019025722-mutt-send-email-mst@kernel.org> (raw)
In-Reply-To: <4a1739ac3cdb895e41f7554865d5e1df4d70658c.camel@nvidia.com>

On Tue, Oct 19, 2021 at 06:45:19AM +0000, Xueming(Steven) Li wrote:
> On Tue, 2021-10-19 at 02:15 -0400, Michael S. Tsirkin wrote:
> > On Fri, Oct 08, 2021 at 03:58:04PM +0800, Xueming Li wrote:
> > > When vhost-user device cleanup and unmmap notifier address, VM cpu
> > > thread that writing the notifier failed with accessing invalid address.
> > > 
> > > To avoid this concurrent issue, wait memory flatview update by draining
> > > rcu callbacks, then unmap notifiers.
> > > 
> > > Fixes: 44866521bd6e ("vhost-user: support registering external host notifiers")
> > > Cc: tiwei.bie@intel.com
> > > Cc: qemu-stable@nongnu.org
> > > Cc: Yuwei Zhang <zhangyuwei.9149@bytedance.com>
> > > Signed-off-by: Xueming Li <xuemingl@nvidia.com>
> > > ---
> > >  hw/virtio/vhost-user.c | 21 ++++++++++++++-------
> > >  1 file changed, 14 insertions(+), 7 deletions(-)
> > > 
> > > diff --git a/hw/virtio/vhost-user.c b/hw/virtio/vhost-user.c
> > > index bf6e50223c..b2e948bdc7 100644
> > > --- a/hw/virtio/vhost-user.c
> > > +++ b/hw/virtio/vhost-user.c
> > > @@ -1165,6 +1165,12 @@ static void vhost_user_host_notifier_remove(struct vhost_dev *dev,
> > >  
> > >      if (n->addr && n->set) {
> > >          virtio_queue_set_host_notifier_mr(vdev, queue_idx, &n->mr, false);
> > > +        if (!qemu_in_vcpu_thread()) { /* Avoid vCPU dead lock. */
> > > +            /* Wait for VM threads accessing old flatview which contains notifier. */
> > > +            drain_call_rcu();
> > > +        }
> > > +        munmap(n->addr, qemu_real_host_page_size);
> > > +        n->addr = NULL;
> > >          n->set = false;
> > >      }
> > >  }
> > 
> > 
> > ../hw/virtio/vhost-user.c: In function ‘vhost_user_host_notifier_remove’:
> > ../hw/virtio/vhost-user.c:1168:14: error: implicit declaration of function ‘qemu_in_vcpu_thread’ [-Werror=implicit-function-declaration]
> >  1168 |         if (!qemu_in_vcpu_thread()) { /* Avoid vCPU dead lock. */
> >       |              ^~~~~~~~~~~~~~~~~~~
> > ../hw/virtio/vhost-user.c:1168:14: error: nested extern declaration of ‘qemu_in_vcpu_thread’ [-Werror=nested-externs]
> > cc1: all warnings being treated as errors
> > ninja: build stopped: subcommand failed.
> > make[1]: *** [Makefile:162: run-ninja] Error 1
> > make[1]: Leaving directory '/scm/qemu/build'
> > make: *** [GNUmakefile:11: all] Error 2
> > 
> > 
> > Although the following patch fixes it, bisect is broken.
> 
> Yes, really an issue, v4 posted, thanks!

Pls address the comment on 2/2 too.

> > 
> > 
> > > @@ -1502,12 +1508,7 @@ static int vhost_user_slave_handle_vring_host_notifier(struct vhost_dev *dev,
> > >  
> > >      n = &user->notifier[queue_idx];
> > >  
> > > -    if (n->addr) {
> > > -        virtio_queue_set_host_notifier_mr(vdev, queue_idx, &n->mr, false);
> > > -        object_unparent(OBJECT(&n->mr));
> > > -        munmap(n->addr, page_size);
> > > -        n->addr = NULL;
> > > -    }
> > > +    vhost_user_host_notifier_remove(dev, queue_idx);
> > >  
> > >      if (area->u64 & VHOST_USER_VRING_NOFD_MASK) {
> > >          return 0;
> > > @@ -2485,11 +2486,17 @@ void vhost_user_cleanup(VhostUserState *user)
> > >      for (i = 0; i < VIRTIO_QUEUE_MAX; i++) {
> > >          if (user->notifier[i].addr) {
> > >              object_unparent(OBJECT(&user->notifier[i].mr));
> > > +        }
> > > +    }
> > > +    memory_region_transaction_commit();
> > > +    /* Wait for VM threads accessing old flatview which contains notifier. */
> > > +    drain_call_rcu();
> > > +    for (i = 0; i < VIRTIO_QUEUE_MAX; i++) {
> > > +        if (user->notifier[i].addr) {
> > >              munmap(user->notifier[i].addr, qemu_real_host_page_size);
> > >              user->notifier[i].addr = NULL;
> > >          }
> > >      }
> > > -    memory_region_transaction_commit();
> > >      user->chr = NULL;
> > >  }
> > >  
> > > -- 
> > > 2.33.0
> > 
> 



  reply	other threads:[~2021-10-19  7:03 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-08  7:58 [PATCH v3 0/2] Improve vhost-user VQ notifier unmap Xueming Li
2021-10-08  7:58 ` [PATCH v3 1/2] vhost-user: fix VirtQ notifier cleanup Xueming Li
2021-10-19  6:15   ` Michael S. Tsirkin
2021-10-19  6:45     ` Xueming(Steven) Li
2021-10-19  6:57       ` Michael S. Tsirkin [this message]
2021-10-08  7:58 ` [PATCH v3 2/2] vhost-user: remove VirtQ notifier restore Xueming Li
2021-10-19  6:37   ` Michael S. Tsirkin
2021-10-19  7:21     ` Xueming(Steven) Li
2021-10-19  7:24       ` Michael S. Tsirkin
2021-10-19  8:00         ` Xueming(Steven) Li

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=20211019025722-mutt-send-email-mst@kernel.org \
    --to=mst@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-stable@nongnu.org \
    --cc=tiwei.bie@intel.com \
    --cc=xuemingl@nvidia.com \
    --cc=zhangyuwei.9149@bytedance.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).