Re: Debian on mac68k (Mac IIfx update)
On Wed, 2 Nov 2016, Mac User wrote:
> On 11/2/16 12:59 AM, Finn Thain wrote:
> > Does the output from "hwclock -r" agree with Mac OS on your IIfx? If
> > so, you could get your hwclock init script to run "hwclock --hctosys"
> > before the fsck and mount scripts execute.
> No, "hwclock -r" reports "Cannot access the Hardware Clock via any known
> method. Use the --debug option to see the details of our search for an
> access method." I also got this error in Debian 3.1 with kernel 2.2.25.
> With --debug, it reports "no such device" for /dev/rtc (though the file
> /dev/rtc exists). Perhaps it's because I'm still using a Debian 3.1
> filesystem? Maybe one of the CONFIG_RTC_* options for the kernel would
> work? As long as I don't go back and forth between kernel 2.2.25 and
> 4.1.35, it should be ok (4.1.35 only forced a full fsck check once).
I don't have a Debian Sarge system to look at so I'd have to experiment to
figure this out. I suggest,
# mkdir /root/dev.save
# mv /dev/rtc* /root/dev.save
# mknod /dev/rtc c 10 135
# hwclock -r
If that doesn't work, you could then try,
# rm /dev/rtc
# mknod /dev/rtc c 254 0
# hwclock -r
> On the plus side, with 4.1.35, wall clock time no longer lags noticeably
> when CPU usage goes up. Maybe that will fix some other subtle problems,
> such as SCSI timeouts with two or more disks connected (I'll check on
> that at some point).
I can't imagine that clock drift would cause SCSI timeouts. However, the
timeouts can readily be explained by serious bugs in the Mac port of Linux
2.2, some of which relate to interrupt handling and the 5380 SCSI driver.
If you want to know the details, the fixes for these bugs can be seen in
the git commit logs going back to v2.6.
> > Regarding the esp_scsi bug that affects your 660av, would you be able
> > to send a close-up photo of the SCSI chip in your 660av?
> see attached.. I believe the relevant chip is "AM79C950K" or
> "AM79C950KC". The writing on it is just barely visible. According to
> "The Q660AV and Q840AV do not have a distinct SCSI chip. They use the
> AMD AM79C950K. That is code named CURIO by Apple. It is a combination
> ethernet, serial port, SCSI chip. It basically combines the ethernet
> MAC, 85C30 serial chip and the 53C96 or 53C94 SCSI chip into one
> component. I have never found a datasheet for it."
Thanks for that. Apple's developer note says the Curio includes a 53C94
core, so it should be a lot like the other Quadras (except Q900/950 which
I gather has 53C96 chips).
This is a different device to those used on Amiga Zorro boards, which
seems to argue against chip errata being the cause of the "Reconnect
IRQ2 timeout" bug.