From: Eric Dumazet <eric.dumazet@gmail.com>
To: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>,
netdev@vger.kernel.org, xma@us.ibm.com, davem@davemloft.net,
linux-kernel@vger.kernel.org, ebiederm@xmission.com
Subject: Re: [PATCH 2/6] macvtap: zerocopy: fix truesize underestimation
Date: Mon, 16 Apr 2012 10:49:53 +0200 [thread overview]
Message-ID: <1334566193.28012.66.camel@edumazet-glaptop> (raw)
In-Reply-To: <20120416071423.GA25396@redhat.com>
On Mon, 2012-04-16 at 10:14 +0300, Michael S. Tsirkin wrote:
> On Mon, Apr 16, 2012 at 02:07:59PM +0800, Jason Wang wrote:
> > As the skb fragment were pinned/built from user pages, we should
> > account the page instead of length for truesize.
> >
> > Signed-off-by: Jason Wang <jasowang@redhat.com>
>
> I'm not sure this is right: the skb does *not* consume the
> whole page, userspace uses the rest of the page
> for other skbs. So we'll end up accounting for the
> same page twice.
> Eric, what's the right thing to do here in your opinion?
Problem is we dont know for sure userspace wont free pages right after
this syscall. So an evil application could consume more kernel memory
than what socket limit allowed.
Its same problem with vmsplice(mem -> pipe) + splice(pipe -> socket)
When we clone skb with frags, resulting skb will have same truesize,
even if the pages are shared ...
next prev parent reply other threads:[~2012-04-16 8:49 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-04-16 6:07 [PATCH 1/6] macvtap: zerocopy: fix offset calculation when building skb Jason Wang
2012-04-16 6:07 ` [PATCH 2/6] macvtap: zerocopy: fix truesize underestimation Jason Wang
2012-04-16 7:14 ` Michael S. Tsirkin
2012-04-16 8:23 ` Jason Wang
2012-04-16 8:49 ` Eric Dumazet [this message]
2012-04-16 13:25 ` Michael S. Tsirkin
2012-04-16 6:08 ` [PATCH 3/6] macvtap: zerocopy: validate vector length before pinning user pages Jason Wang
2012-04-16 6:53 ` Eric Dumazet
2012-04-16 8:21 ` Jason Wang
2012-04-17 5:33 ` Eric Dumazet
2012-04-17 5:43 ` Michael S. Tsirkin
2012-04-17 6:19 ` Michael S. Tsirkin
2012-04-16 7:58 ` Michael S. Tsirkin
2012-04-16 6:08 ` [PATCH 4/6] macvtap: zerocopy: set SKBTX_DEV_ZEROCOPY only when skb is built successfully Jason Wang
2012-04-16 6:08 ` [PATCH 5/6] vhost_net: fix use after free of vq->ubufs Jason Wang
2012-04-16 13:28 ` Michael S. Tsirkin
2012-04-17 3:19 ` Jason Wang
2012-04-17 10:22 ` Michael S. Tsirkin
2012-04-17 10:47 ` Jason Wang
2012-04-16 6:08 ` [PATCH 6/6] vhost_net: don't poll on -EFAULT Jason Wang
2012-04-16 7:16 ` Michael S. Tsirkin
2012-04-16 8:28 ` Jason Wang
2012-04-16 13:39 ` Michael S. Tsirkin
2012-04-17 3:27 ` Jason Wang
2012-04-17 4:57 ` Michael S. Tsirkin
2012-04-17 5:54 ` Jason Wang
2012-04-17 6:07 ` Michael S. Tsirkin
2012-04-17 6:30 ` Jason Wang
2012-04-17 10:18 ` Michael S. Tsirkin
2012-04-17 10:46 ` Jason Wang
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=1334566193.28012.66.camel@edumazet-glaptop \
--to=eric.dumazet@gmail.com \
--cc=davem@davemloft.net \
--cc=ebiederm@xmission.com \
--cc=jasowang@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mst@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=xma@us.ibm.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
all inboxes | Powered by JetHome®