Subject: Re: sun4c vs 1.6
To: None <port-sparc@netbsd.org>
From: Greg A. Woods <woods@weird.com>
List: port-sparc
Date: 10/23/2002 18:49:00
[ On Wednesday, October 23, 2002 at 13:23:47 (-0700), Aaron J. Grier wrote: ]
> Subject: sun4c vs 1.6
>
> I just installed 1.6 on an IPX which I plan to use for firewalling
> duties, but remembered some discussion a while back regarding problems
> with the software cache flush and the 1+.  does this affect the IPX as
> well?

As Manuel said it shouldn't be a problem on the IPX....

> cpu0 at mainbus0: cache chip bug; trap page uncached: W8601/8701 or MB86903 @ 40 MHz, on-chip FPU

interesting... none of my machines trigger that one.....

> I have another IPX with dead NVRAM (I'll be replacing it soon) which has
> a weitek rather than fujitsu CPU -- would it be any safer?

The one with the Weitek might be faster!  ;-)

> or has this all been fixed and I have nothing to worry about?  (I'm sure
> Greg will speak up at this point...)

The pressure has gone away at my end -- I'm typing on an SS20 now, and
will soon be swapping back down to an SS5 with the SS20 turning into a
colour workstation as I try to port the Xfree86 ZX (sunleo) 24-bit
support to NetBSD (unless someone pipes up and suggests that the SS5
w/microSPARC-II @85MHz will be faster at Sbus graphics than an SS20
w/SM50 SuperSPARC.  All I need now is a decently long 13w3 to 4-BNC
cable and some much more convenient way to arrange two large keyboards
and mice near my colour monitor!  ;-)

At this point even my ss2 will be relegated to just testing kernels,
installs, and binary packages as I now also have another SS20 to replace
it with.

I may not even have any use at all any more for the SS1 and SS1+, though
I could certainly still use them to help test any fixes to this software
cache flush bug.....

-- 
								Greg A. Woods

+1 416 218-0098;            <g.a.woods@ieee.org>;           <woods@robohack.ca>
Planix, Inc. <woods@planix.com>; VE3TCP; Secrets of the Weird <woods@weird.com>