Hello altogether,

thanky ou for all the tips you gave me setting up my 3/60 with NetBSD.
I have reset the si_options flag to 1, and the core dumps while starting X11
didn't occur any more.  I cannot say that I understand the link between these
two things, but at least it works :-).  One strange thing is still existent:
ldd prints junk about the X11 and Xt shared libraries:

twm:
        -lXmu.6 => /usr/X11R6/lib/libXmu.so.6.0 (0x2042000)
        -l.6 => /usr/X11R6/lib/libXt.so.6.0 (0x2052000)
        -l~ +.6 => /usr/X11R6/lib/libX11.so.6.0 (0x208e000)
        -lSM.6 => /usr/X11R6/lib/libSM.so.6.0 (0x20f6000)
        -lICE.6 => /usr/X11R6/lib/libICE.so.6.0 (0x20fe000)
        -lXext.6 => /usr/X11R6/lib/libXext.so.6.0 (0x210c000)
        -lc.12 => /usr/lib/libc.so.12.3 (0x2116000)
xclock:
        -lXaw.6 => /usr/X11R6/lib/libXaw.so.6.0 (0x2026000)
        -lXmu.6 => /usr/X11R6/lib/libXmu.so.6.0 (0x2056000)
        -lJQfRBARPB@PQPAD@Ln.6 => /usr/X11R6/lib/libXt.so.6.0 (0x2066000)
        -lRBARPB@PQPAD@Ln.6 => /usr/X11R6/lib/libX11.so.6.0 (0x20a2000)
        -lSM.6 => /usr/X11R6/lib/libSM.so.6.0 (0x210a000)
        -lICE.6 => /usr/X11R6/lib/libICE.so.6.0 (0x2112000)
        -lXext.6 => /usr/X11R6/lib/libXext.so.6.0 (0x2120000)
        -lc.12 => /usr/lib/libc.so.12.3 (0x212a000)
xterm:
        -lXaw.6 => /usr/X11R6/lib/libXaw.so.6.0 (0x2040000)
        -lXmu.6 => /usr/X11R6/lib/libXmu.so.6.0 (0x2070000)
        -l.6 => /usr/X11R6/lib/libXt.so.6.0 (0x2080000)
        -l???yb^|.6 => /usr/X11R6/lib/libX11.so.6.0 (0x20bc000)
        -lSM.6 => /usr/X11R6/lib/libSM.so.6.0 (0x2124000)
        -lICE.6 => /usr/X11R6/lib/libICE.so.6.0 (0x212c000)
        -lXext.6 => /usr/X11R6/lib/libXext.so.6.0 (0x213a000)
        -ltermcap.0 => /usr/lib/libtermcap.so.0.0 (0x2144000)
        -lc.12 => /usr/lib/libc.so.12.3 (0x2148000)

The output is different for different programs, but reproducable for a
single one.  It doesn't seem to influence the execution of programs.

Meanwhile, I also upgraded my system to a 24 MHz 68030/68882 combination.
25 MHz worked for about 15 minutes, but in a moment of heavy disk activity,
a SCSI DMA timeout occured :-( .  As my 68030 is a 33 MHz chip, I was also
curious and tried running the system at 33 MHz.  I had to replace the original
300ns boot EPROM with a faster one, after which the monitor worked.  However,
the system failed while loading the NetBSD kernel...seems the SCSI interface
is the limiting component in my machine.

Geert, do you have any detailed information why the 68030's data cache
wouldn't work with the Sun MMU?  I thought that I just would have to set
the enable flag in kernel initialization and to flush the cache on context
switches, about the same one has to do with the instruction cache.  It's a
pity that Sun doesn't have technical manuals on these machines any more...

Alfred Arnold

-- 
--------------------------------------------------------------------------------
Alfred Arnold                                 Phone: +49 (2461) 61-2434
Zentralinstitut fuer Angewandte Mathematik    Fax  : +49 (2461) 61-6656
Forschungszentrum Juelich                     E-Mail: a.arnold@kfa-juelich.de
D-52425 Juelich, Germany                      

Hi, I'm not a signature virus. Why don't you just copy me into your signature?
--------------------------------------------------------------------------------