[PATCH v22.11] vhost: fix missing spinlock unlock

Xueming(Steven) Li xuemingl at nvidia.com
Mon Jan 22 14:10:51 CET 2024


Hi Hailin,

> -----Original Message-----
> From: Xueming(Steven) Li <xuemingl at nvidia.com>
> Sent: 1/11/2024 19:11
> To: Xu, HailinX <hailinx.xu at intel.com>
> Cc: Maxime Coquelin <maxime.coquelin at redhat.com>; stable at dpdk.org;
> ktraynor at redhat.com; dmarchan at redhat.com
> Subject: RE: [PATCH v22.11] vhost: fix missing spinlock unlock
> 
> Thanks Maxime!
> 
> Hailin, could you please confirm if it fixes
> https://bugs.dpdk.org/show_bug.cgi?id=1338?

The CI passed with this fix, w/o your further confirmation, I'd assume it fixed.
I'll send out 22.11 release if no objections.

> 
> > -----Original Message-----
> > From: Maxime Coquelin <maxime.coquelin at redhat.com>
> > Sent: 1/11/2024 18:52
> > To: stable at dpdk.org; Xueming(Steven) Li <xuemingl at nvidia.com>;
> > ktraynor at redhat.com; dmarchan at redhat.com
> > Cc: Maxime Coquelin <maxime.coquelin at redhat.com>
> > Subject: [PATCH v22.11] vhost: fix missing spinlock unlock
> >
> > Two regressions were introduced when backporting below
> > patch:
> > b4c4e5675c85 ("vhost: fix missing lock protection in power monitor
> > API")
> >
> > First, rte_vhost_get_monitor_addr did not release the lock in the success case.
> > Then, rte_rwlock_read_lock() was converted to rte_spinlock_trylock()
> > instead of rte_spinlock_lock().
> >
> > This patch addresses both of these issues.
> >
> > Fixes: a07736eb68da ("vhost: fix missing lock protection in power
> > monitor API")
> > Cc: stable at dpdk.org
> >
> > Signed-off-by: Maxime Coquelin <maxime.coquelin at redhat.com>
> > ---
> >  lib/vhost/vhost.c | 12 ++++++------
> >  1 file changed, 6 insertions(+), 6 deletions(-)
> >
> > diff --git a/lib/vhost/vhost.c b/lib/vhost/vhost.c index
> > 40ac350e21..9e28198528
> > 100644
> > --- a/lib/vhost/vhost.c
> > +++ b/lib/vhost/vhost.c
> > @@ -2060,6 +2060,7 @@ rte_vhost_get_monitor_addr(int vid, uint16_t
> > queue_id,  {
> >  	struct virtio_net *dev = get_device(vid);
> >  	struct vhost_virtqueue *vq;
> > +	int ret = 0;
> >
> >  	if (dev == NULL)
> >  		return -1;
> > @@ -2070,11 +2071,12 @@ rte_vhost_get_monitor_addr(int vid, uint16_t
> > queue_id,
> >  	if (vq == NULL)
> >  		return -1;
> >
> > -	if (!rte_spinlock_trylock(&vq->access_lock))
> > -		return -1;
> > +	rte_spinlock_lock(&vq->access_lock);
> >
> > -	if (unlikely(!vq->access_ok))
> > +	if (unlikely(!vq->access_ok)) {
> > +		ret = -1;
> >  		goto out_unlock;
> > +	}
> >
> >  	if (vq_is_packed(dev)) {
> >  		struct vring_packed_desc *desc;
> > @@ -2095,12 +2097,10 @@ rte_vhost_get_monitor_addr(int vid, uint16_t
> > queue_id,
> >  		pmc->match = 0;
> >  	}
> >
> > -	return 0;
> > -
> >  out_unlock:
> >  	rte_spinlock_unlock(&vq->access_lock);
> >
> > -	return -1;
> > +	return ret;
> >  }
> >
> >
> > --
> > 2.43.0



More information about the stable mailing list