Re: [dpdk-dev] [PATCH v1] vhost: add sanity check for resubmiting reqs in split ring

["Xia, Chenbo" <chenbo.xia@intel.com>]

Hi Changpeng,

> -----Original Message-----
> From: Maxime Coquelin <maxime.coquelin@redhat.com>
> Sent: Thursday, October 14, 2021 4:26 PM
> To: Li Feng <fengli@smartx.com>; Xia, Chenbo <chenbo.xia@intel.com>
> Cc: dev@dpdk.org
> Subject: Re: [PATCH v1] vhost: add sanity check for resubmiting reqs in split
> ring
> 
> 
> 
> On 10/14/21 10:17, Maxime Coquelin wrote:
> > Hi Li,
> >
> > Adding Jin Yu who introduced this function.
> 
> Looks like Jin Yu has left Intel, Chenbo, could you find someone from
> the Intel SPDK team to look at it?

Could you or your team member help check this?

Thanks,
Chenbo

> 
> > On 8/27/21 07:12, Li Feng wrote:
> >> When getting reqs from the avail ring, the id may exceed inflight
> >> queue size. Then the dpdk will crash forever.
> >
> > You need to add Fixes tag and Cc stable@dpdk.org so that it can be
> > backported.
> >
> >> Signed-off-by: Li Feng <fengli@smartx.com>
> >> ---
> >>   lib/vhost/vhost_user.c | 10 ++++++++--
> >>   1 file changed, 8 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/lib/vhost/vhost_user.c b/lib/vhost/vhost_user.c
> >> index 29a4c9af60..f09d0f6a48 100644
> >> --- a/lib/vhost/vhost_user.c
> >> +++ b/lib/vhost/vhost_user.c
> >> @@ -1823,8 +1823,14 @@ vhost_check_queue_inflights_split(struct
> >> virtio_net *dev,
> >>       last_io = inflight_split->last_inflight_io;
> >>       if (inflight_split->used_idx != used->idx) {
> >> -        inflight_split->desc[last_io].inflight = 0;
> >> -        rte_atomic_thread_fence(__ATOMIC_SEQ_CST);
> >> +        if (unlikely(last_io >= inflight_split->desc_num)) {
> >> +            VHOST_LOG_CONFIG(ERR, "last_inflight_io '%"PRIu16"'
> >> exceeds inflight "
> >> +                "queue size (%"PRIu16").\n", last_io,
> >> +                inflight_split->desc_num);
> >
> > If such error happens, shouldn't we return RTE_VHOST_MSG_RESULT_ERR
> > instead of just logging an error?
> >
> >> +        } else {
> >> +            inflight_split->desc[last_io].inflight = 0;
> >> +            rte_atomic_thread_fence(__ATOMIC_SEQ_CST);
> >> +        }
> >>           inflight_split->used_idx = used->idx;
> >>       }
> >>
> >
> > Regards,
> > Maxime