>From: Rod Ross <rodross <at> myrealbox.com>
>Date: 2007-11-10 01:51:27 GMT
>On Thu, Nov 08, 2007 at 11:51:11PM -0500, Alex Kenis wrote:
>> Apple's developer notes were
>> pretty detailed up through the 601 cpu era... then they get more
>> general after that, so I can only really speculate until I find
>> anything solid.
>I know the latest patches work for the kernel 126.96.36.199 (see my uname below).
>So, at least add them. Alex, If you think you can help with 601 errors
>and 2.6.* or linking errors please do so.
>Linux rod 188.8.131.52 #1 Sun Oct 7 22:02:59 EDT 2007 ppc GNU/Linux
>This SF.net email is sponsored by: Splunk Inc.
>Still grepping through log files to find problems? Stop.
>Now Search log events and configuration files using AJAX and a browser.
I am currently running Linux debian 2.4.35 #1...
I will give the newest 2.4 version a try soon. I tinkered around yesterday with the standard Debian 2.4.27-nubus kernel, but hit a few bugs with it... which were actually reported to them, but they were ignored until they were no longer relevant (which makes no sense to me since they dropped official nubus support after that... you think they would have fixed the last supported version for posterity's sake). i also played with the 68k Emile booter to see if it is a viable option for PPC, but hit some strange errors that i did not have the time to mess with.
Anyway, i looked at the 2.6 errors and i have no idea how to even start. Do you have links to any discussions about it that i can take a peek at? for now, I am going to tinker with a few of the unresolved powerbook issues like sound and floppy support and those mysterious ASIC's, as well as picking at migrating the fremebuffer driver... not that i think i can think of anything that has not been contemplated before, but maybe if i chase the commands around enough in MacOS, I can find something useful. i spent the better part of today digging up all my old Mac developer tools and debuggers/decompilers, and digging through Texas Instruments' & LSI old literature (hoping for come info on an old I/O chip similar to the Whitney) and ancient A/UX and MkLinux code. A/UX proved to be fairly useless since it seems that there is very little direct hardware interaction, and it seems to run more like a MacOS app than an OS. i did find more data sheets for the asco 2300 to add to the asc3550o sheet that was posted before: http://mysite.verizon.net/vzeqncxc/nubus/asco2300.pdf
My plan is to look at the code for machines that work, and then chase the functions around on one of my Powerbooks, see how it interacts with the IOKit driver, and do some probing with Apple's nubus 'slots' program and a few others and then peek at the addressesses and memory in macsbug, etc. I am not looking for anything specific... just trends so that maybe I can see how/if the Whitney buffers the singer chip or allows a DMA-like function like the O'Hare chips do with the next generation of powerbooks, so if anyone wants me to find something specific, let me know.