Subject: Re: sq1: ...cannot share CPU interrupts
To: Thomas Fritz <tf@slash10.com>
From: Rafal Boni <rafal@attbi.com>
List: port-sgimips
Date: 09/16/2002 22:18:33
In message <3D85F44E.60403@slash10.com>, you write: 

-> Hi there!
-> 
-> I just installed NetBSD1.6 on a SGI Challenge S. (Took some time to find 
-> the sets on releng.netbsd.org, which really work fine. Some hints on
-> www.netbsd.org for these snapshot would be great for other newbies... :)

Now that 1.6 release is official, the sets should be moving to some better
announced/more popular locations.

-> I have the system up and running, the only problem I have is, when I 
-> want to compile the device sq1, which is on hpc1, into the kernel.
-> 
-> Booting this kernel produces this output:
-> 
[...]
-> hpc1 at gio0 addr 0x1fb00000: SGI HPC3
-> zsc at hpc1 offset 0x59830 not configured
-> sq1 at hpc1 offset 0x54000panic: cannot share CPU interrupts <----------
-> halting...
-> 
-> Does someone have a hint for me on that topic?

Feed the developers beer so they fix up the code? 8-)

Seriously, this is a shortcoming of the Indy/Indigo2/Challense S/M code...
the interrupt handling isn't flexible enough to share interrupts between
devices and even if that *did* work, I'm not sure more work wouldn't be
required to get the Mezzanine SCSI/ethernet board running (ether may be
easy, SCSI uses a different chip for which apparently *NO* docs exist).

It's on my list of things to do, and I'm going to have to re-write the
O2 interrupt code anyway, so I'll try and do that in a way that the IP22
machines can also benefit.  OTOH, I've had little free time for hacking
lately, so I'm not promising any dates in the near future 8-)

--rafal

----
Rafal Boni                                                     rafal@attbi.com
  We are all worms.  But I do believe I am a glowworm.  -- Winston Churchill