Hi Andrei,
--------------------------------------------------
From: "Andrei Emeltchenko" <andrei.emeltchenko.news@xxxxxxxxx>
Sent: Thursday, July 19, 2012 5:10 PM
To: "Jaganath Kanakkassery" <jaganath.k@xxxxxxxxxxx>
Cc: <linux-bluetooth@xxxxxxxxxxxxxxx>; "Johan Hedberg"
<johan.hedberg@xxxxxxxxx>; "Gustavo Padovan" <gustavo@xxxxxxxxxxx>
Subject: Re: [PATCH] Bluetooth: Fix socket not getting freed if l2cap
channel create fails
Hi Jaganath,
On Thu, Jul 19, 2012 at 04:50:16PM +0530, Jaganath Kanakkassery wrote:
Hi Andrei,
--------------------------------------------------
From: "Andrei Emeltchenko" <andrei.emeltchenko.news@xxxxxxxxx>
Sent: Thursday, July 19, 2012 1:22 PM
To: "Jaganath Kanakkassery" <jaganath.k@xxxxxxxxxxx>
Cc: <linux-bluetooth@xxxxxxxxxxxxxxx>
Subject: Re: [PATCH] Bluetooth: Fix socket not getting freed if
l2cap channel create fails
>Hi Jaganath,
>
>On Thu, Jul 19, 2012 at 12:54:04PM +0530, Jaganath Kanakkassery wrote:
>>If l2cap_chan_create() fails then it will return from l2cap_sock_kill
>>since zapped flag of sk is reset.
>>
>>Signed-off-by: Jaganath Kanakkassery <jaganath.k@xxxxxxxxxxx>
>>---
>> net/bluetooth/l2cap_sock.c | 2 +-
>> 1 files changed, 1 insertions(+), 1 deletions(-)
>>
>>diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c
>>index 79350d1..419857d 100644
>>--- a/net/bluetooth/l2cap_sock.c
>>+++ b/net/bluetooth/l2cap_sock.c
>>@@ -1174,7 +1174,7 @@ static struct sock
>>*l2cap_sock_alloc(struct net *net, struct socket *sock, int p
>>
>> chan = l2cap_chan_create();
>> if (!chan) {
>>- l2cap_sock_kill(sk);
>>+ sk_free(sk);
>
>Could you consider using sock_put which will call sk_free,
>maybe we need to add also sock_orphan?
Ok, Actually I used sk_free since there is not refcount increase at
this point
Have you tested it? It shall be 1, set by sock_init_data.
Yes it is 1. So even if we use sock_put() , it will decrement the refcount
and call sk_free().
and also I found the same code in rfcomm_sock_alloc().
So should I fix it in RFCOMM also?
I think using sock_put would be the right approach. Maybe maintainers
could comment here?
Ok, I will wait for maintainers comments.
Thanks,
Jaganath
--
To unsubscribe from this list: send the line "unsubscribe linux-bluetooth" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html