[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-mac68k
Subject:    [linux-mac68k] Various things (Debian install, Powerbooks available, Quadra SCSI)...
From:       Michael Schmitz <SCHMITZ () lcbvax ! cchem ! berkeley ! edu>
Date:       1998-08-21 4:37:11
[Download RAW message or body]

Hi

to keep this list from getting too inactive, a few update bits:

- Debian installation stuff has been placed on wave.lm.com (the resurfaced
FTP site) in MacLinux/latest, please get
ftp://wave.lm.com/incoming/Debian-m68k-2.0-Mac-Install.html
and just follow the instructions if you want to install Debian 2.0. I've
mentioned that a few times on c.o.l.m86k but didn't announce it here IIRC.

- Pat Eyler contacted me about Powerbooks that his company is decommissioning
and he'd like to contribute to forward Linux development. The models are 170,
and maybe 145 and 160. Powerbook ADB development is what they would be useful
for, I guess. Contact Pat (peyler@nortom.com) or me for more details.
(and thanks to Alan for making the contact; I've read your Linux Expo paper
about good excuses carefully, Alan :-)

- Quadra SCSI: I've royally messed up my source trying to beat the SCSI code 
into a shape that won't break other platforms if our changes get integrated,
but it looks as if I see light, for the Quadra 800 SCSI at least (but that
beast wasn't very problematic anyway). To cut a long story short, the Mac code
can use the vanilla esp_intr handler and doesn't need the patch Dave sent to
the list for the 640 (removing the dma_irq_p() test from esp_exec_cmd() to
get more than one SCSI command to survive). For the 800 (and most of the 
machines using the same flavor of SCSI hardware) all it takes is making 
dma_irq_p() test for the interrupt flag in the ESP status register, and nothing
else. Once I've got the 800 to fully behave again, I'll post a kernel that 
uses the same test on the dafb_dma_irq_p() function so we can see if that's
also sufficient for the 700, 900, 640 and similar boxes.

All in all, we could use a poll of the stability of the current Quadra SCSI
driver: on what machines does it work at all, where does it fail?

- Another thing is ongoing merge of ADB code bits with the PPC port, we've
started to eliminate duplicate code in the ADB mouse drivers and that seems to 
play nicely, further things down the line are merging the hardware independent 
part of the ADB driver (which has ADB device address resolution in PPC) and
maybe the keyboard driver. Some of this may not happen before 2.2 (Alan: if you 
have already done things in that vein, please send a diff to Dave, Paul or me).

	Michael

[prev in list] [next in list] [prev in thread] [next in thread]