On Wednesday 08 April 2009 11:10:10 Roland Dreier wrote:
>  > The debug messaage wrongly prints the address of a local variable. Also
>  > when INFINIBAND_NES_DEBUG is not set, gcc emits an unused variable
>  > warning. Fix it.
>  >
>  >    nes_debug(NES_DBG_CM, "Unable to find listener for %pI4:%x\n",
>  > -            &tmp_addr, dst_port);
>  > +            cpu_to_be32(dst_addr), dst_port);
>
> My understanding is that %pI4 wants a pointer (as all %p formats do) --
> and every other use of %pI4 in the kernel that I looked as is passing a
> pointer to printk.  Have you tested this patch with NES debug on?  I
> would expect gcc to warn about passing a non-pointer to a %p format.
>

Yes, I was little hasty. Here is the corrected patch. Replaced %p with %x.

Thanks
Nikanth

The debug messaage wrongly prints the address of a local variable. Also when
INFINIBAND_NES_DEBUG is not set, gcc emits an unused variable warning. Fix it.

Signed-off-by: Nikanth Karthikesan <[email protected]>

---

diff --git a/drivers/infiniband/hw/nes/nes_cm.c 
b/drivers/infiniband/hw/nes/nes_cm.c
index 5242515..14ffada 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -859,7 +859,6 @@ static struct nes_cm_listener *find_listener(struct 
nes_cm_core *cm_core,
 {
        unsigned long flags;
        struct nes_cm_listener *listen_node;
-       __be32 tmp_addr = cpu_to_be32(dst_addr);
 
        /* walk list and find cm_node associated with this session ID */
        spin_lock_irqsave(&cm_core->listen_list_lock, flags);
@@ -876,8 +875,8 @@ static struct nes_cm_listener *find_listener(struct 
nes_cm_core *cm_core,
        }
        spin_unlock_irqrestore(&cm_core->listen_list_lock, flags);
 
-       nes_debug(NES_DBG_CM, "Unable to find listener for %pI4:%x\n",
-                 &tmp_addr, dst_port);
+       nes_debug(NES_DBG_CM, "Unable to find listener for %xI4:%x\n",
+                 cpu_to_be32(dst_addr), dst_port);
 
        /* no listener */
        return NULL;

_______________________________________________
general mailing list
[email protected]
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general

To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general

Reply via email to