cvs commit: src/sys/conf files files.powerpc

[Marcel Moolenaar]

On Nov 22, 2005, at 12:01 PM, Wilko Bulte wrote:

> On Tue, Nov 22, 2005 at 12:16:01PM -0700, Scott Long wrote..
>> Wilko Bulte wrote:
>>
>>> On Tue, Nov 22, 2005 at 01:07:12PM -0500, John Baldwin wrote..
>>>
>>>> On Tuesday 22 November 2005 12:12 pm, Marius Strobl wrote:
>>>>
>>>>> marius      2005-11-22 17:12:49 UTC
>>>>>
>>>>> FreeBSD src repository
>>>>>
>>>>> Modified files:
>>>>>   sys/conf             files files.powerpc
>>>>> Log:
>>>>> Move zs.c from files to files.powerpc as zs(4) by now is only  
>>>>> supported
>>>>> on powerpc (more or less...). That way people updating from  
>>>>> FreeBSD 5
>>>>> to
>>>>> FreeBSD 6 and beyond on sparc64 will get an error from config 
>>>>> (8) rather
>>>>> than a mysterious compile error when they have a stale 'device  
>>>>> zs' in
>>>>> their kernel config file.
>>>>>
>>>>> MFC after:      2 weeks
>>>>
>>>> I think some Alpha machines (4100?) also use zs(4) for serial  
>>>> console.
>>>
>>>
>>> zs(4)? Hmm, the only oddball Alpha in this sense that comes to mind
>>> is Turbolaser (AlphaServer 8[24]00)
>>>
>>
>> Technically there is an esp driver now that could be used to  
>> revive TL
>> support, but unless there is someone actively interested in it
>> _right_now_ I don't think we should make anything contingent on it.
>
> Yes.  Last time I tried a CD boot on our Tlaser at work it just bombed
> out very early during boot.  I did not feel like digging into it,  
> especially
> given that I do not expect anyone to have a huge machine like that  
> at home
> to hack on (let alone pay for the electricity bill..)

Some trivial changes may be all that's needed. To illustrate:

Index: uart_cpu_alpha.c
===================================================================
RCS file: /nfs/home/ncvs/src/sys/dev/uart/uart_cpu_alpha.c,v
retrieving revision 1.11
diff -u -r1.11 uart_cpu_alpha.c
--- uart_cpu_alpha.c    2005/06/16 18:06:38     1.11
+++ uart_cpu_alpha.c    2005/11/23 03:18:03
@@ -80,7 +80,8 @@
                 if (ctb->ctb_term_type != CTB_PRINTERPORT)
                         return (ENXIO);
                 boothowto |= RB_SERIAL;
-               di->ops = uart_ns8250_ops;
+               di->ops = (hwrpb->rpb_type == ST_DEC_21000)
+                   ? uart_z8530_ops : uart_ns8250_ops;
                 di->bas.chan = 0;
                 di->bas.bst = uart_bus_space_io;
                 if (bus_space_map(di->bas.bst, 0x3f8, 8, 0, &di- 
 >bas.bsh) != 0)

The I/O port base and channel also need to be set correctly of
course but it looks like that can be hardcoded. This should be
all. With access to hardware I could give it a try if people
think it would be fun...

-- 
  Marcel Moolenaar         USPA: A-39004          marcel at xcllnt.net