Re: new kernel hangs on Indy R4000

[Frank Wille <frank%phoenix.owl.de@localhost>]

Michael wrote:

> I've seen that too, I found the cause yesterday. Apparently some IOC  
> initialization screws with newport's drawing engine. Here's what I did:
> 
> Index: ioc.c
> ===================================================================
> RCS file: /cvsroot/src/sys/arch/sgimips/ioc/ioc.c,v
> retrieving revision 1.7
> diff -u -w -r1.7 ioc.c
> - --- ioc.c       17 Oct 2007 19:57:05 -0000      1.7
> +++ ioc.c       25 Feb 2009 23:16:46 -0000
> @@ -140,10 +140,13 @@
>                           IOC_WRITE_ENET_AUTO | IOC_WRITE_ENET_UTP |
>                           IOC_WRITE_PC_UART2 | IOC_WRITE_PC_UART1);
> 
> +/* XXX: the firmware should have taken care of this already */
> +#if 0
>         if (mach_subtype == MACH_SGI_IP22_GUINNESS) {
>                 bus_space_write_4(sc->sc_iot, sc->sc_ioh, IOC_GCSEL,  
> 0xff);
>                 bus_space_write_4(sc->sc_iot, sc->sc_ioh, IOC_GCREG,  
> 0xff);
>         }
> +#endif
> 
>  #if defined(BLINK)
>         ioc_blink(sc);
> 
> if it works for you and doesn't have any adverse side effects I'll  
> commit it.

Yes. Works fine!


Next problem:
Did anybody manage to mount a file system with recent kernels? NFS-mounting
a root file system with GENERIC_IP2x failed. Same problem with the internal
disk image of the INSTALL_IP2x kernel:
[...]
boot device: sq0
root on md0a dumps on md0b
no file system for md0 (dev 0x200)
cannot mount root, error = 79


-- 
    _  Frank Wille (frank%phoenix.owl.de@localhost)
 _ //  http://sun.hasenbraten.de/~frank/
 \X/   Phx @ #AmigaGer