From mboxrd@z Thu Jan 1 00:00:00 1970 From: Santosh Shilimkar Subject: [PATCH 12/14] RDS: make sure rds_send_drop_to properly takes the m_rs_lock Date: Sat, 22 Aug 2015 15:45:33 -0700 Message-ID: <1440283535-4800-13-git-send-email-santosh.shilimkar@oracle.com> References: <1440283535-4800-1-git-send-email-santosh.shilimkar@oracle.com> Cc: linux-kernel@vger.kernel.org, davem@davemloft.net, ssantosh@kernel.org, Santosh Shilimkar To: netdev@vger.kernel.org Return-path: In-Reply-To: <1440283535-4800-1-git-send-email-santosh.shilimkar@oracle.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org rds_send_drop_to() is used during socket tear down to find all the messages on the socket and flush them . It can race with the acking code unless it takes the m_rs_lock on each and every message. This plugs a hole where we didn't take m_rs_lock on any message that didn't have the RDS_MSG_ON_CONN set. Taking m_rs_lock avoids double frees and other memory corruptions as the ack code trusts the message m_rs pointer on a socket that had actually been freed. We must take m_rs_lock to access m_rs. Because of lock nesting and rs access, we also need to acquire rs_lock. Reviewed-by: Ajaykumar Hotchandani Signed-off-by: Santosh Shilimkar Signed-off-by: Santosh Shilimkar --- net/rds/send.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) diff --git a/net/rds/send.c b/net/rds/send.c index 96ae38d..b0fe412 100644 --- a/net/rds/send.c +++ b/net/rds/send.c @@ -778,8 +778,22 @@ void rds_send_drop_to(struct rds_sock *rs, struct sockaddr_in *dest) while (!list_empty(&list)) { rm = list_entry(list.next, struct rds_message, m_sock_item); list_del_init(&rm->m_sock_item); - rds_message_wait(rm); + + /* just in case the code above skipped this message + * because RDS_MSG_ON_CONN wasn't set, run it again here + * taking m_rs_lock is the only thing that keeps us + * from racing with ack processing. + */ + spin_lock_irqsave(&rm->m_rs_lock, flags); + + spin_lock(&rs->rs_lock); + __rds_send_complete(rs, rm, RDS_RDMA_CANCELED); + spin_unlock(&rs->rs_lock); + + rm->m_rs = NULL; + spin_unlock_irqrestore(&rm->m_rs_lock, flags); + rds_message_put(rm); } } -- 1.9.1