From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org, stable@vger.kernel.org Cc: Greg Kroah-Hartman , alan@lxorguk.ukuu.org.uk, Dan Carpenter , Venkat Venkatsubra , "David S. Miller" , James Morris , Jie Liu Subject: [ 79/85] RDS: fix rds-ping spinlock recursion Date: Thu, 25 Oct 2012 17:06:37 -0700 Message-Id: <20121026000039.055668300@linuxfoundation.org> In-Reply-To: <20121026000031.107227138@linuxfoundation.org> References: <20121026000031.107227138@linuxfoundation.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: 3.6-stable review patch. If anyone has any objections, please let me know. ------------------ From: "jeff.liu" [ Upstream commit 5175a5e76bbdf20a614fb47ce7a38f0f39e70226 ] This is the revised patch for fixing rds-ping spinlock recursion according to Venkat's suggestions. RDS ping/pong over TCP feature has been broken for years(2.6.39 to 3.6.0) since we have to set TCP cork and call kernel_sendmsg() between ping/pong which both need to lock "struct sock *sk". However, this lock has already been hold before rds_tcp_data_ready() callback is triggerred. As a result, we always facing spinlock resursion which would resulting in system panic. Given that RDS ping is only used to test the connectivity and not for serious performance measurements, we can queue the pong transmit to rds_wq as a delayed response. Reported-by: Dan Carpenter CC: Venkat Venkatsubra CC: David S. Miller CC: James Morris Signed-off-by: Jie Liu Signed-off-by: David S. Miller Signed-off-by: Greg Kroah-Hartman --- net/rds/send.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/net/rds/send.c +++ b/net/rds/send.c @@ -1122,7 +1122,7 @@ rds_send_pong(struct rds_connection *con rds_stats_inc(s_send_pong); if (!test_bit(RDS_LL_SEND_FULL, &conn->c_flags)) - rds_send_xmit(conn); + queue_delayed_work(rds_wq, &conn->c_send_w, 0); rds_message_put(rm); return 0;