[Prism54-devel] [Bug 80] kernel BUG at drivers/net/wireless/prism54/oid_mgt.c:333!

[bugzilla-daemon@mcgrof.com]

http://prism54.org/cgi-bin/bugzilla/show_bug.cgi?id=80

slts@free.fr changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED



------- Additional Comments From slts@free.fr  2004-04-27 17:09 -------
Several comments : 
>  I'm not able to get my prism54 based card working :( Tried lot of
>kernels, but still no success. I've latest firmware and kernel below
>is 2.6.5 with several patches (they are not touching network subsystem
>or prism54 driver).
>My card is: 2.4GHz 54Mbps Wireless Cardbus Adapter (EZ Connect g)
>              SMC Networks, Model No: SMC2835W

>kernel BUG at drivers/net/wireless/prism54/oid_mgt.c:333!

this is an old version. -> INVALID

>but there is the
>SMC2802W V2 card mentioned and my card is SMC2835W V2. Looks like
>V2 & newmac problem.

True. Your card won't work with this driver. -> INVALID

Your log with latest CVS looks normal, given that you have a newmac card.
The "eth2: errant PIMFOR application frame" is maybe somewhat unusual.

For the last part : After a "kernel BUG", your kernel is almost killed. So the 

>Apr 25 22:37:45 echelon kernel: Unable to handle kernel paging request at
>virtual address 00050042
>...

and the crash are consequences of the "kernel BUG at
/home/robert/cvs/prism54-ng/ksrc/islpci_dev.c:804!".
But this last one looks like a VALID bug. Are you able to reproduce it ?

I'd like to see the log of this Kernel BUG with the following change in
islpci_dev.c : 

-------------------------------patch-----------------------------------

--- ksrc/islpci_dev.c   26 Apr 2004 10:09:58 -0000      1.77
+++ ksrc/islpci_dev.c   27 Apr 2004 17:04:22 -0000
@@ -795,10 +795,10 @@
                        priv->state = new_state;
                break;
        };
-#if 0
+
        printk(KERN_DEBUG "%s: state transition %d -> %d (off#%d)\n",
               priv->ndev->name, old_state, new_state, priv->state_off);
-#endif
+
 
        /* invariants */
        BUG_ON(priv->state_off < 0);


-----------------------------------------------------------------------------


...And justice for all



------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.