netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Wei Liu <wei.liu2@citrix.com>
To: annie li <annie.li@oracle.com>
Cc: Wei Liu <wei.liu2@citrix.com>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
	"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
	"konrad.wilk@oracle.com" <konrad.wilk@oracle.com>,
	"jbeulich@suse.com" <jbeulich@suse.com>,
	Ian Campbell <Ian.Campbell@citrix.com>,
	"wdauchy@gmail.com" <wdauchy@gmail.com>,
	David Vrabel <david.vrabel@citrix.com>
Subject: Re: [PATCH V5 6/7] xen-netback: coalesce slots in TX path and fix regressions
Date: Mon, 22 Apr 2013 10:28:25 +0100	[thread overview]
Message-ID: <20130422092825.GN5815@zion.uk.xensource.com> (raw)
In-Reply-To: <517462D2.8040209@oracle.com>

On Sun, Apr 21, 2013 at 11:06:10PM +0100, annie li wrote:
[...]
> > -		if (unlikely(frags >= MAX_SKB_FRAGS)) {
> > -			netdev_err(vif->dev, "Too many frags\n");
> > +		/* This guest is really using too many slots and
> > +		 * considered malicious.
> > +		 */
> > +		if (unlikely(slots >= max_skb_slots)) {
> > +			netdev_err(vif->dev,
> > +				   "Malicious frontend using %d slots, threshold %u\n",
> > +				   slots, max_skb_slots);
> >   			netbk_fatal_tx_err(vif);
> >   			return -E2BIG;
> 
> It is possible that vif is freed when packet size is less than 64K here 
> but slots required >= max_skb_slots. Alough max_skb_slots can be 
> configured, this kind of packets would be dropped in following if 
> (!drop_err && slots >= XEN_NETIF_NR_SLOTS_MIN).
> 

Yes, because slots required >= max_skb_slots is considered malicious
behavior. The size of the packet doesn't matter here.

> >   		}
> >   
> > -		memcpy(txp, RING_GET_REQUEST(&vif->tx, cons + frags),
> > +		/* Xen network protocol had implicit dependency on
> > +		 * MAX_SKB_FRAGS. XEN_NETIF_NR_SLOTS_MIN is set to the
> > +		 * historical MAX_SKB_FRAGS value 18 to honor the same
> > +		 * behavior as before. Any packet using more than 18
> > +		 * slots but less than max_skb_slots slots is dropped
> > +		 */
> > +		if (!drop_err && slots >= XEN_NETIF_NR_SLOTS_MIN) {
> > +			if (net_ratelimit())
> > +				netdev_dbg(vif->dev,
> > +					   "Too many slots (%d) exceeding limit (%d), dropping packet\n",
> > +					   slots, XEN_NETIF_NR_SLOTS_MIN);
> > +			drop_err = -E2BIG;
> 
> It is possible to drop packets like above(size < 64K && slot >= 
> XEN_NETIF_NR_SLOTS_MIN). I do not know how frequently this kind of 
> packets appear, maybe some SKBs with compound page(size < 64K && slot >= 
> XEN_NETIF_NR_SLOTS_MIN) are dropped here?
> 

I don't know either, but this is the default behavior of netback so here
we retain the same behavior here to avoid further regression.


Wei.

  reply	other threads:[~2013-04-22  9:28 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-15 17:06 [PATCH V5 0/7] Bundle fixes for Xen netfront /netback Wei Liu
2013-04-15 17:06 ` [PATCH V5 1/7] xen-netfront: remove unused variable `extra' Wei Liu
2013-04-15 17:06 ` [PATCH V5 2/7] xen-netfront: frags -> slots in xennet_get_responses Wei Liu
2013-04-15 17:06 ` [PATCH V5 3/7] xen-netback: remove skb in xen_netbk_alloc_page Wei Liu
2013-04-15 17:06 ` [PATCH V5 4/7] xen-netfront: frags -> slots in log message Wei Liu
2013-04-17 13:48   ` Ian Campbell
2013-04-15 17:06 ` [PATCH V5 5/7] xen-netfront: reduce gso_max_size to account for max TCP header Wei Liu
2013-04-17 13:51   ` Ian Campbell
2013-04-15 17:06 ` [PATCH V5 6/7] xen-netback: coalesce slots in TX path and fix regressions Wei Liu
2013-04-17 13:53   ` Ian Campbell
2013-04-21 22:06   ` annie li
2013-04-22  9:28     ` Wei Liu [this message]
2013-04-30 18:06   ` [Xen-devel] " annie li
2013-05-01  9:22     ` Wei Liu
2013-04-15 17:06 ` [PATCH V5 7/7] xen-netback: don't disconnect frontend when seeing oversize packet Wei Liu
2013-04-15 17:38 ` [PATCH V5 0/7] Bundle fixes for Xen netfront /netback David Miller
2013-04-17 13:49   ` Ian Campbell
2013-04-17 17:05     ` David Miller

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=20130422092825.GN5815@zion.uk.xensource.com \
    --to=wei.liu2@citrix.com \
    --cc=Ian.Campbell@citrix.com \
    --cc=annie.li@oracle.com \
    --cc=david.vrabel@citrix.com \
    --cc=jbeulich@suse.com \
    --cc=konrad.wilk@oracle.com \
    --cc=netdev@vger.kernel.org \
    --cc=wdauchy@gmail.com \
    --cc=xen-devel@lists.xen.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).