Several patches against linuxppc_devel-2.4.18-pre9

Dunka David..

The ethernet changes look fine. there was on other change regarding resetting the EMAC. The ethernet part can go in as is our we can wait until tomorrow when I send in that change + multi emac irq fix.

-armin

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Stefan, David & Andrew,

Thanks

I will remove the EMAC soft-reset.

--armin

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Hello

The attached patch fixes the following issues:

- drivers/net/ibm_ocp_enet.c
wrong prototypes + initialisation in debug code

- drivers/net/ibm_ocp_enet.h
missing prototypes for [get|set]_maltxdeir() missing prototypes for [get|set]_maltxctp1r()

- drivers/net/ibm_ocp_phy.c
workaround for MII-PHY communication bug reported by Andrew May

- arch/ppc/mm/fault.c
missing type cast

- arch/ppc/mm/cachemap.c
missing prototype for map_page() (added mmu_decl.h) 2 missing type casts

Dave

Armin and David,

I checked the EMAC/PHY patch on our board (CPCI-405 with Intel LXT971), and the driver seems to start without problems without ethernet cable connected, when the EMAC soft-reset command is removed from the init routine.

With the soft-reset in init still implemented, "fec_enet_mdio_read()" returns with timeout, waiting for transfer to complete. The next soft-reset in open leads to the same problem (timeout in mdio_read()), but doesn't lead to a kernel panik.

So Armin, could you please remove the soft-reset from init (if nobody objects, of cause).

Best regards,
Stefan Roese

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/