[ofa-general] RE: [PATCH] rdma_cm: fix port type (fix bug 557)

[Sean Hefty]

Can you see if this patch also fixes the problem?  I'd like to keep
next_port defined as an int to match the idr_get_new_above() prototype
and sysctl_local_port_range definition.

If this fixes the problem, we should add it to OFED and queue it for
2.6.23.
---

next_port should be between sysctl_local_port_range[0] and [1].  However,
it is initially set to a random value.  If the value is negative, next_port
can fall outside of this range because of the % operator returning a
negative value.

Signed-off-by: Sean Hefty <sean.hefty at intel.com>
---

 drivers/infiniband/core/cma.c |    4 ++--
 1 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index eb15119..b0831cb 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -2772,8 +2772,8 @@ static int cma_init(void)
 	int ret;
 
 	get_random_bytes(&next_port, sizeof next_port);
-	next_port = (next_port % (sysctl_local_port_range[1] -
-				  sysctl_local_port_range[0])) +
+	next_port = ((unsigned int) next_port % 
+		    (sysctl_local_port_range[1] - sysctl_local_port_range[0])) +
 		    sysctl_local_port_range[0];
 	cma_wq = create_singlethread_workqueue("rdma_cm");
 	if (!cma_wq)