8route wrote:
> Dear Atsushi:
> Hi!
> I found the reset switch on the TX3927 board can work if you switch it
> against the PCI interface when
> ==================================
> VFS: Mounted root (NFS filesystem).
> Freeing unused kernel memory: 44k freed
> ========================================
> ,usually I switch it towards the PCI interface.
>
> And I have fixed the source code of process.c as you said.
> >For TX3927, you must skip those two lines in exit_thread() and
> >flush_thread().
> >
> > set_cp0_status(ST0_CU1, ST0_CU1);
> > __asm__ __volatile__("cfc1\t$0,$31");
> >
> >
> But the problem still exists.The serial console output will stop at
> ==================================
> VFS: Mounted root (NFS filesystem).
> Freeing unused kernel memory: 44k freed
> ========================================
>
> I think that maybe the Linux system has booted up OK,but there is
> something wrong with the serial console because serial ports are integrated
> with TX3927.Do you agree?Please consider it in your latest patch.
> >By the way, now I'm planning to send patches for TX CPU boards
> >(including JMR3927) to oss.sgi.com. If you can wait a while, you can
> >try it.
> Thank you very much.
>
> 8route
> ajob4me@21cn.com
> 10/29/2001
I have already submitted a patch to Ralf for the JMR3927 and am waiting for
his reply.
We should coordinate to make sure we don't overwrite each other.
Alice
