virtualization.lists.linux-foundation.org archive mirror
 help / color / mirror / Atom feed
* Re: [PATCH net-next v7] vhost_net: avoid tx queue stuck when sendmsg fails
       [not found] <1610685980-38608-1-git-send-email-wangyunjian@huawei.com>
@ 2021-01-15  6:05 ` Jason Wang
  2021-01-15 18:15   ` Willem de Bruijn
  2021-01-19  9:55 ` Michael S. Tsirkin
       [not found] ` <20210118143329.08cc14a6@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
  2 siblings, 1 reply; 4+ messages in thread
From: Jason Wang @ 2021-01-15  6:05 UTC (permalink / raw)
  To: wangyunjian, netdev
  Cc: willemdebruijn.kernel, mst, jerry.lilijun, virtualization,
	chenchanghu, brian.huangbin, xudingke


On 2021/1/15 下午12:46, wangyunjian wrote:
> From: Yunjian Wang <wangyunjian@huawei.com>
>
> Currently the driver doesn't drop a packet which can't be sent by tun
> (e.g bad packet). In this case, the driver will always process the
> same packet lead to the tx queue stuck.
>
> To fix this issue:
> 1. in the case of persistent failure (e.g bad packet), the driver
>     can skip this descriptor by ignoring the error.
> 2. in the case of transient failure (e.g -ENOBUFS, -EAGAIN and -ENOMEM),
>     the driver schedules the worker to try again.
>
> Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>


Acked-by: Jason Wang <jasowang@redhat.com>


> ---
> v7:
>     * code rebase
> v6:
>     * update code styles and commit log
> ---
>   drivers/vhost/net.c | 26 ++++++++++++++------------
>   1 file changed, 14 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
> index 3b744031ec8f..df82b124170e 100644
> --- a/drivers/vhost/net.c
> +++ b/drivers/vhost/net.c
> @@ -828,14 +828,15 @@ static void handle_tx_copy(struct vhost_net *net, struct socket *sock)
>   				msg.msg_flags &= ~MSG_MORE;
>   		}
>   
> -		/* TODO: Check specific error and bomb out unless ENOBUFS? */
>   		err = sock->ops->sendmsg(sock, &msg, len);
>   		if (unlikely(err < 0)) {
> -			vhost_discard_vq_desc(vq, 1);
> -			vhost_net_enable_vq(net, vq);
> -			break;
> -		}
> -		if (err != len)
> +			if (err == -EAGAIN || err == -ENOMEM || err == -ENOBUFS) {
> +				vhost_discard_vq_desc(vq, 1);
> +				vhost_net_enable_vq(net, vq);
> +				break;
> +			}
> +			pr_debug("Fail to send packet: err %d", err);
> +		} else if (unlikely(err != len))
>   			pr_debug("Truncated TX packet: len %d != %zd\n",
>   				 err, len);
>   done:
> @@ -924,7 +925,6 @@ static void handle_tx_zerocopy(struct vhost_net *net, struct socket *sock)
>   			msg.msg_flags &= ~MSG_MORE;
>   		}
>   
> -		/* TODO: Check specific error and bomb out unless ENOBUFS? */
>   		err = sock->ops->sendmsg(sock, &msg, len);
>   		if (unlikely(err < 0)) {
>   			if (zcopy_used) {
> @@ -933,11 +933,13 @@ static void handle_tx_zerocopy(struct vhost_net *net, struct socket *sock)
>   				nvq->upend_idx = ((unsigned)nvq->upend_idx - 1)
>   					% UIO_MAXIOV;
>   			}
> -			vhost_discard_vq_desc(vq, 1);
> -			vhost_net_enable_vq(net, vq);
> -			break;
> -		}
> -		if (err != len)
> +			if (err == -EAGAIN || err == -ENOMEM || err == -ENOBUFS) {
> +				vhost_discard_vq_desc(vq, 1);
> +				vhost_net_enable_vq(net, vq);
> +				break;
> +			}
> +			pr_debug("Fail to send packet: err %d", err);
> +		} else if (unlikely(err != len))
>   			pr_debug("Truncated TX packet: "
>   				 " len %d != %zd\n", err, len);
>   		if (!zcopy_used)

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH net-next v7] vhost_net: avoid tx queue stuck when sendmsg fails
  2021-01-15  6:05 ` [PATCH net-next v7] vhost_net: avoid tx queue stuck when sendmsg fails Jason Wang
@ 2021-01-15 18:15   ` Willem de Bruijn
  0 siblings, 0 replies; 4+ messages in thread
From: Willem de Bruijn @ 2021-01-15 18:15 UTC (permalink / raw)
  To: Jason Wang
  Cc: Willem de Bruijn, Michael S. Tsirkin, Network Development,
	wangyunjian, Lilijun (Jerry),
	virtualization, xudingke, huangbin (J),
	chenchanghu

On Fri, Jan 15, 2021 at 1:12 AM Jason Wang <jasowang@redhat.com> wrote:
>
>
> On 2021/1/15 下午12:46, wangyunjian wrote:
> > From: Yunjian Wang <wangyunjian@huawei.com>
> >
> > Currently the driver doesn't drop a packet which can't be sent by tun
> > (e.g bad packet). In this case, the driver will always process the
> > same packet lead to the tx queue stuck.
> >
> > To fix this issue:
> > 1. in the case of persistent failure (e.g bad packet), the driver
> >     can skip this descriptor by ignoring the error.
> > 2. in the case of transient failure (e.g -ENOBUFS, -EAGAIN and -ENOMEM),
> >     the driver schedules the worker to try again.
> >
> > Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
>
>
> Acked-by: Jason Wang <jasowang@redhat.com>

Acked-by: Willem de Bruijn <willemb@google.com>
_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH net-next v7] vhost_net: avoid tx queue stuck when sendmsg fails
       [not found] <1610685980-38608-1-git-send-email-wangyunjian@huawei.com>
  2021-01-15  6:05 ` [PATCH net-next v7] vhost_net: avoid tx queue stuck when sendmsg fails Jason Wang
@ 2021-01-19  9:55 ` Michael S. Tsirkin
       [not found] ` <20210118143329.08cc14a6@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
  2 siblings, 0 replies; 4+ messages in thread
From: Michael S. Tsirkin @ 2021-01-19  9:55 UTC (permalink / raw)
  To: wangyunjian
  Cc: willemdebruijn.kernel, netdev, jerry.lilijun, virtualization,
	xudingke, brian.huangbin, chenchanghu

On Fri, Jan 15, 2021 at 12:46:20PM +0800, wangyunjian wrote:
> From: Yunjian Wang <wangyunjian@huawei.com>
> 
> Currently the driver doesn't drop a packet which can't be sent by tun
> (e.g bad packet). In this case, the driver will always process the
> same packet lead to the tx queue stuck.
> 
> To fix this issue:
> 1. in the case of persistent failure (e.g bad packet), the driver
>    can skip this descriptor by ignoring the error.
> 2. in the case of transient failure (e.g -ENOBUFS, -EAGAIN and -ENOMEM),
>    the driver schedules the worker to try again.
> 
> Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>

Acked-by: Michael S. Tsirkin <mst@redhat.com>

> ---
> v7:
>    * code rebase
> v6:
>    * update code styles and commit log
> ---
>  drivers/vhost/net.c | 26 ++++++++++++++------------
>  1 file changed, 14 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/vhost/net.c b/drivers/vhost/net.c
> index 3b744031ec8f..df82b124170e 100644
> --- a/drivers/vhost/net.c
> +++ b/drivers/vhost/net.c
> @@ -828,14 +828,15 @@ static void handle_tx_copy(struct vhost_net *net, struct socket *sock)
>  				msg.msg_flags &= ~MSG_MORE;
>  		}
>  
> -		/* TODO: Check specific error and bomb out unless ENOBUFS? */
>  		err = sock->ops->sendmsg(sock, &msg, len);
>  		if (unlikely(err < 0)) {
> -			vhost_discard_vq_desc(vq, 1);
> -			vhost_net_enable_vq(net, vq);
> -			break;
> -		}
> -		if (err != len)
> +			if (err == -EAGAIN || err == -ENOMEM || err == -ENOBUFS) {
> +				vhost_discard_vq_desc(vq, 1);
> +				vhost_net_enable_vq(net, vq);
> +				break;
> +			}
> +			pr_debug("Fail to send packet: err %d", err);
> +		} else if (unlikely(err != len))
>  			pr_debug("Truncated TX packet: len %d != %zd\n",
>  				 err, len);
>  done:
> @@ -924,7 +925,6 @@ static void handle_tx_zerocopy(struct vhost_net *net, struct socket *sock)
>  			msg.msg_flags &= ~MSG_MORE;
>  		}
>  
> -		/* TODO: Check specific error and bomb out unless ENOBUFS? */
>  		err = sock->ops->sendmsg(sock, &msg, len);
>  		if (unlikely(err < 0)) {
>  			if (zcopy_used) {
> @@ -933,11 +933,13 @@ static void handle_tx_zerocopy(struct vhost_net *net, struct socket *sock)
>  				nvq->upend_idx = ((unsigned)nvq->upend_idx - 1)
>  					% UIO_MAXIOV;
>  			}
> -			vhost_discard_vq_desc(vq, 1);
> -			vhost_net_enable_vq(net, vq);
> -			break;
> -		}
> -		if (err != len)
> +			if (err == -EAGAIN || err == -ENOMEM || err == -ENOBUFS) {
> +				vhost_discard_vq_desc(vq, 1);
> +				vhost_net_enable_vq(net, vq);
> +				break;
> +			}
> +			pr_debug("Fail to send packet: err %d", err);
> +		} else if (unlikely(err != len))
>  			pr_debug("Truncated TX packet: "
>  				 " len %d != %zd\n", err, len);
>  		if (!zcopy_used)
> -- 
> 2.23.0

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH net-next v7] vhost_net: avoid tx queue stuck when sendmsg fails
       [not found] ` <20210118143329.08cc14a6@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
@ 2021-01-19  9:56   ` Michael S. Tsirkin
  0 siblings, 0 replies; 4+ messages in thread
From: Michael S. Tsirkin @ 2021-01-19  9:56 UTC (permalink / raw)
  To: Jakub Kicinski
  Cc: willemdebruijn.kernel, netdev, wangyunjian, jerry.lilijun,
	virtualization, xudingke, brian.huangbin, chenchanghu

On Mon, Jan 18, 2021 at 02:33:29PM -0800, Jakub Kicinski wrote:
> On Fri, 15 Jan 2021 12:46:20 +0800 wangyunjian wrote:
> > From: Yunjian Wang <wangyunjian@huawei.com>
> > 
> > Currently the driver doesn't drop a packet which can't be sent by tun
> > (e.g bad packet). In this case, the driver will always process the
> > same packet lead to the tx queue stuck.
> > 
> > To fix this issue:
> > 1. in the case of persistent failure (e.g bad packet), the driver
> >    can skip this descriptor by ignoring the error.
> > 2. in the case of transient failure (e.g -ENOBUFS, -EAGAIN and -ENOMEM),
> >    the driver schedules the worker to try again.
> > 
> > Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
> 
> Michael, LMK if you want to have a closer look otherwise I'll apply
> tomorrow.

Thanks for the reminder. Acked.

-- 
MST

_______________________________________________
Virtualization mailing list
Virtualization@lists.linux-foundation.org
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2021-01-19  9:57 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <1610685980-38608-1-git-send-email-wangyunjian@huawei.com>
2021-01-15  6:05 ` [PATCH net-next v7] vhost_net: avoid tx queue stuck when sendmsg fails Jason Wang
2021-01-15 18:15   ` Willem de Bruijn
2021-01-19  9:55 ` Michael S. Tsirkin
     [not found] ` <20210118143329.08cc14a6@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
2021-01-19  9:56   ` Michael S. Tsirkin

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).