mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Krzysztof Mazur <krzysiek@podlesie.net>
To: "Chas Williams (CONTRACTOR)" <chas@cmf.nrl.navy.mil>
Cc: davem@davemloft.net, dwmw2@infradead.org, netdev@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2 2/3] pppoatm: fix race condition with destroying of vcc
Date: Wed, 31 Oct 2012 11:22:10 +0100	[thread overview]
Message-ID: <20121031102210.GA15077@shrek.podlesie.net> (raw)
In-Reply-To: <20121031094147.GA1004@shrek.podlesie.net>

On Wed, Oct 31, 2012 at 10:41:47AM +0100, Krzysztof Mazur wrote:
> 
> I think that we should add a wrapper to vcc->send(), based on
> fixed pppoatm_send(), that performs required checks and takes the ATM socket
> lock.
> 

I'm sending initial version of such wrapper and update to pppoatm.
Untested but the code is just copied from pppoatm_send.

In final series I will fix some old &sk_atm(ATM_SKB(skb)->vcc)-like
code from original version, before moving to vcc_send_bh(), but
it's just an initial idea for some comments.

Krzysiek

diff --git a/net/atm/common.c b/net/atm/common.c
index 0c0ad93..e0602d2 100644
--- a/net/atm/common.c
+++ b/net/atm/common.c
@@ -558,6 +558,32 @@ int vcc_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
 	return copied;
 }
 
+int vcc_send_bh(struct atm_vcc *vcc, struct sk_buff *skb)
+{
+	int ret;
+
+	bh_lock_sock(sk_atm(vcc));
+	ret = -EAGAIN;
+	if (sock_owned_by_user(sk_atm(vcc)))
+		goto out;
+	if (test_bit(ATM_VF_RELEASED, &vcc->flags)
+			|| test_bit(ATM_VF_CLOSE, &vcc->flags)
+			|| !test_bit(ATM_VF_READY, &vcc->flags))
+		goto out;
+
+	if (sk_wmem_alloc_get(sk_atm(vcc)) && !atm_may_send(vcc, skb->truesize))
+		goto out;
+
+	atomic_add(skb->truesize, &sk_atm(ATM_SKB(skb)->vcc)->sk_wmem_alloc);
+	ATM_SKB(skb)->atm_options = ATM_SKB(skb)->vcc->atm_options;
+	pr_debug("atm_skb(%p)->vcc(%p)->dev(%p)\n",
+		 skb, ATM_SKB(skb)->vcc, ATM_SKB(skb)->vcc->dev);
+	ret = ATM_SKB(skb)->vcc->send(ATM_SKB(skb)->vcc, skb);
+out:
+	bh_unlock_sock(sk_atm(vcc));
+	return ret;
+}
+
 int vcc_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
 		size_t total_len)
 {
diff --git a/net/atm/common.h b/net/atm/common.h
index cc3c2da..3a1c340 100644
--- a/net/atm/common.h
+++ b/net/atm/common.h
@@ -15,6 +15,7 @@ int vcc_release(struct socket *sock);
 int vcc_connect(struct socket *sock, int itf, short vpi, int vci);
 int vcc_recvmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *msg,
 		size_t size, int flags);
+int vcc_send_bh(struct atm_vcc *vcc, struct sk_buff *skb);
 int vcc_sendmsg(struct kiocb *iocb, struct socket *sock, struct msghdr *m,
 		size_t total_len);
 unsigned int vcc_poll(struct file *file, struct socket *sock, poll_table *wait);
diff --git a/net/atm/pppoatm.c b/net/atm/pppoatm.c
index 5fc335a..7612f18 100644
--- a/net/atm/pppoatm.c
+++ b/net/atm/pppoatm.c
@@ -296,31 +296,10 @@ static int pppoatm_send(struct ppp_channel *chan, struct sk_buff *skb)
 	}
 
 	vcc = ATM_SKB(skb)->vcc;
-	bh_lock_sock(sk_atm(vcc));
-	if (sock_owned_by_user(sk_atm(vcc)))
-		goto nospace_unlock_sock;
-	if (test_bit(ATM_VF_RELEASED, &vcc->flags)
-			|| test_bit(ATM_VF_CLOSE, &vcc->flags)
-			|| !test_bit(ATM_VF_READY, &vcc->flags))
-		goto nospace_unlock_sock;
-
-	/*
-	 * It's not clear that we need to bother with using atm_may_send()
-	 * to check we don't exceed sk->sk_sndbuf.
-	 */
-	if (sk_wmem_alloc_get(sk_atm(vcc)) && !atm_may_send(vcc, skb->truesize))
-		goto nospace_unlock_sock;
-
-	atomic_add(skb->truesize, &sk_atm(ATM_SKB(skb)->vcc)->sk_wmem_alloc);
-	ATM_SKB(skb)->atm_options = ATM_SKB(skb)->vcc->atm_options;
-	pr_debug("atm_skb(%p)->vcc(%p)->dev(%p)\n",
-		 skb, ATM_SKB(skb)->vcc, ATM_SKB(skb)->vcc->dev);
-	ret = ATM_SKB(skb)->vcc->send(ATM_SKB(skb)->vcc, skb)
-	    ? DROP_PACKET : 1;
-	bh_unlock_sock(sk_atm(vcc));
-	return ret;
-nospace_unlock_sock:
-	bh_unlock_sock(sk_atm(vcc));
+	ret = vcc_send_bh(vcc, skb);
+	if (ret == -EAGAIN)
+		goto nospace;
+	return ret ? DROP_PACKET : 1;
 nospace:
 	/*
 	 * We don't have space to send this SKB now, but we might have

  reply	other threads:[~2012-10-31 10:22 UTC|newest]

Thread overview: 89+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-22 17:14 [PATCH v2 1/3] pppoatm: don't send frames to destroyed vcc Krzysztof Mazur
2012-10-22 17:14 ` [PATCH v2 2/3] pppoatm: fix race condition with destroying of vcc Krzysztof Mazur
2012-10-30  9:37   ` David Woodhouse
2012-10-30 19:07     ` Krzysztof Mazur
2012-10-30 19:52       ` Krzysztof Mazur
2012-10-31 10:16         ` David Woodhouse
2012-10-31 11:30           ` Krzysztof Mazur
2012-10-31 11:52             ` David Woodhouse
2012-10-30 14:26   ` Chas Williams (CONTRACTOR)
2012-10-30 18:20     ` Krzysztof Mazur
2012-10-31  9:41       ` Krzysztof Mazur
2012-10-31 10:22         ` Krzysztof Mazur [this message]
2012-10-31 20:03         ` chas williams - CONTRACTOR
2012-10-31 22:04           ` Krzysztof Mazur
2012-11-01 14:26             ` chas williams - CONTRACTOR
2012-11-02  9:40               ` Krzysztof Mazur
2012-11-02 10:54                 ` Krzysztof Mazur
2012-10-22 17:14 ` [PATCH v2 3/3] pppoatm: protect against freeing " Krzysztof Mazur
2012-10-30  9:39   ` David Woodhouse
2012-10-30 19:26     ` Krzysztof Mazur
2012-11-27 17:16   ` David Woodhouse
2012-11-27 17:39     ` Krzysztof Mazur
2012-11-27 18:02       ` David Woodhouse
2012-11-27 18:28         ` Krzysztof Mazur
2012-11-28 20:18           ` Krzysztof Mazur
2012-11-28 20:44             ` David Woodhouse
2012-11-28 21:24               ` Krzysztof Mazur
2012-11-28 21:20             ` chas williams - CONTRACTOR
2012-11-28 21:45               ` [PATCH] atm: introduce vcc_pop() Krzysztof Mazur
2012-11-28 21:59                 ` chas williams - CONTRACTOR
2012-11-28 22:10                   ` Krzysztof Mazur
2012-11-28 22:33                     ` [PATCH] atm: introduce vcc_pop_skb() Krzysztof Mazur
2012-12-03 13:22                       ` David Woodhouse
2012-12-03 20:11                         ` Krzysztof Mazur
2012-11-27 18:39         ` [PATCH v2 3/3] pppoatm: protect against freeing of vcc Krzysztof Mazur
2012-11-27 18:54         ` chas williams - CONTRACTOR
2012-11-27 22:36           ` [PATCH] solos-pci: Wait for pending TX to complete when releasing vcc David Woodhouse
2012-11-27 23:28             ` [PATCH] br2684: don't send frames on not-ready vcc David Woodhouse
2012-11-27 23:51               ` Krzysztof Mazur
2012-11-28  0:54                 ` David Woodhouse
2012-11-28  8:08                   ` Krzysztof Mazur
2012-11-28  9:58                     ` David Woodhouse
2012-11-28 16:41               ` David Miller
2012-11-28 17:01                 ` David Woodhouse
2012-11-28 17:04                   ` David Miller
2012-11-28 17:09                     ` David Woodhouse
2012-11-28 17:11                       ` David Miller
2012-11-30  1:18                       ` Nathan Williams
2012-11-30  1:34                         ` David Woodhouse
2012-11-28  9:21           ` [PATCH v2 3/3] pppoatm: protect against freeing of vcc David Laight
2012-11-28 10:04             ` Krzysztof Mazur
2012-11-28 10:24               ` David Woodhouse
2012-11-28 15:18                 ` chas williams - CONTRACTOR
2012-11-28 22:18             ` David Woodhouse
2012-11-29 10:57               ` Krzysztof Mazur
2012-11-29 11:55                 ` David Woodhouse
2012-11-29 12:43                   ` [PATCH] solos-pci: don't call vcc->pop() after pclose() Krzysztof Mazur
2012-11-29 12:57                     ` David Woodhouse
2012-11-29 13:20                       ` Krzysztof Mazur
2012-11-29 14:42                         ` David Woodhouse
2012-11-29 14:55                           ` Krzysztof Mazur
2012-11-29 14:41                     ` chas williams - CONTRACTOR
2012-11-29 14:29                 ` [PATCH v2 3/3] pppoatm: protect against freeing of vcc chas williams - CONTRACTOR
2012-11-29 15:09               ` Krzysztof Mazur
2012-11-29 15:47                 ` David Woodhouse
2012-11-29 15:59                   ` chas williams - CONTRACTOR
2012-11-29 16:24                     ` David Woodhouse
2012-11-29 17:17                       ` chas williams - CONTRACTOR
2012-11-29 18:11                         ` David Woodhouse
2012-11-29 18:29                           ` chas williams - CONTRACTOR
2012-11-29 22:17                             ` David Woodhouse
2012-11-30  1:38                               ` Chas Williams (CONTRACTOR)
2012-11-30  1:57                                 ` David Woodhouse
2012-11-30  8:25                                   ` David Woodhouse
2012-11-30  9:53                                     ` Krzysztof Mazur
2012-11-30 12:10                                       ` David Woodhouse
2012-11-30 16:23                                         ` David Woodhouse
2012-11-30 17:00                                           ` Krzysztof Mazur
2012-11-30 18:33                                             ` David Woodhouse
2012-11-30 17:12                                           ` chas williams - CONTRACTOR
2012-11-30 17:39                                             ` Krzysztof Mazur
2012-11-29 16:28                   ` Krzysztof Mazur
2012-11-29 15:37               ` chas williams - CONTRACTOR
2012-11-29 15:59                 ` David Woodhouse
2012-11-29 16:11                   ` chas williams - CONTRACTOR
2012-10-23  6:52 ` [PATCH v2 1/3] pppoatm: don't send frames to destroyed vcc David Miller
2012-10-23  8:12   ` David Woodhouse
2012-10-30  9:35 ` David Woodhouse
2012-10-30 20:19   ` Krzysztof Mazur

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=20121031102210.GA15077@shrek.podlesie.net \
    --to=krzysiek@podlesie.net \
    --cc=chas@cmf.nrl.navy.mil \
    --cc=davem@davemloft.net \
    --cc=dwmw2@infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=netdev@vger.kernel.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

all inboxes | Powered by JetHome®