Port-macppc archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: OldWorld floppy and serial port work
Actually, my bad... turns out it was the other end of the serial line that's
the problem. :-) If I just use a Mini-DIN8->DB9 conversion and run it to a
*real* serial port on my Windows machine, it behaves; the Keyspan adaptor must
be having issues (it's been otherwise well-behaved so far, though). I just
took the macppc FAQ about the serial ports at its word, and I suppose that's
what I get. We might want to update that sometime?
Back to floppy-world... :-)
- Dave
On Jul 8, 2010, at 3:46 PM, Michael wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hello,
>
> On Jul 8, 2010, at 12:46 PM, David Riley wrote:
>
>> Well, that's the problem; I have a 9600, which is probably somewhat
>> different than the serial port on the G4 (which is built in, but not exposed
>> by default, and I wonder if it's different hardware). All the way back to
>> the original Mac, they've used a Z8530 (just like Sun and everyone else who
>> wanted a *real* serial chip), and mac68k works fine. For me, if I look at a
>> man page, it stops responding to serial commands (though it'll still output
>> log messages, which is puzzling). The system log doesn't output any debug
>> info, so I'll probably have to start from the top.
>
> I would suspect interrupt problems. The G4 has a completely different
> interrupt controller ( OpenPIC instead of Apple's own ) which should not lose
> interrupts but Apple's might. If we enable an interrupt line that's already
> asserted it will not cause an interrupt for example. There is code in place
> to check this condition, it works fine with other devices but occasionally
> weird things happen - for example I get lost interrupt messages from the IDE
> chip on my beige G3 on first access to the disk with no further ill effects.
> And this does not happen with any other Apple IDE chip so it's probably a
> Heathrow thing.
> So, first thing I'd do is to run systat 1 vmstat and check if the serial
> port's interrupts still fire when thinks lock up and you keep sending. Since
> it happens on a G4 as well it's very likely not the actual interrupt
> dispatching code but something in the macppc-specific portion of the zs
> driver. In other words, what Donald Lee said.
>
> have fun
> Michael
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.7 (Darwin)
>
> iQEVAwUBTDYq+MpnzkX8Yg2nAQL0uAgAuusDMh/0zs2P2K6mO7tWUJCDLeFKoAqt
> y9fvifqchysiN9pcV9LHEiiRf1BqEAB4xyxIfdoAtC+pBsAD+FlKlXZ03IQc9Wew
> DYet0L0cas/+I3+p/ySvsMzAK7hZwvKYm0gOy9lYoDF9ZskGkdOZSsXg00jPZWHl
> 99KtkEdhB7n0ZeLEHLn91H5I4rNZbN3s8mbBFtrU7YXQl7XT0eBF5GQRew/xPG6K
> YtXnufoHiZQ8O1IwIgmDIhoWC5RCL8pVdqNRycFYMieNW80PY5mICuLruZq2oh5k
> LUSaq3CRKIJKUklOlgBvY4UgxSE+Tzfh4AdrDGEB7rspfVRXPgg8/w==
> =E7YL
> -----END PGP SIGNATURE-----
Home |
Main Index |
Thread Index |
Old Index