Am Montag, 3. September 2012 06:40:04 UTC+2 schrieb gilles:

> I think it's too early in the boot process to be the same xenix
> bug that affects Xprofile and needs a patch.

That's right. None of the three byte patterns that have to be replaced occurs in the boot block.

With IDEfile's trace mode[1] enabled I can seen every HD access on a RS232 terminal. When the system hangs, it reads block 0000 and nothing else. No other read attempt is made before the "75" is reported.

Following James' advice I tried "apple-3" at the "startup from" menu -- and it works!. Using "apple-1" for the internal 10MB crashes, using "apple-3" works, that's reproducable. Evidence for the "uninitialized register" theory.

> I'll ask my son to explore my other laptop at home to find the code

I would appreciate it. Otherwise James would have to feed his disassembler with these 512 bytes.


James, do you know which files are affected by the Xenix patches? (Of course) I forgot to patch my boot disk before installing, therefore I have to apply the patches to my hard drive. Knowing the filenames would narrow the block ranges that have to be scanned for the byte sequences[2].


Now UniPlus is running, Xenix is running, so next one is GEMdos? And for me, fixing my Widgets. Two have been rescued by re-formatting with BLU, one has a too many spares (it will become a 9.9 MB unit with 511 cylinders and modified firmware[3]), and the others really need repairs on component level.


Best regards
Patrick



[1] enter "l" at the RS-232 terminal to toggle IDEfile trace mode on/off. Each access is reported like "Host reads block 0815". That significantly slows down the system, but produces nice boot traces.

[2] how do I find file unix.o on my disk? With trace mode enabled, "cat unix.o" and watch the terminal output. The first accesses may be from the cat command itself, but the rest belongs to the file data.

[3] my re-assembly of the Widget firmware ist at bitsavers.org. These files can be assembled with the famous as assembler http://john.ccac.rwth-aachen.de:8000/as/ . Feel free to patch anything defined in the "defs.inc" file. Now we are able to re-format the drives, so you cannot damage anything.



Another one: the Widget servo board is able to return 4 status bytes. Does anyone have a description what these bytes mean? This information is missing in the  different "Servo Functional Objective" documents available at bitsavers.org and the Widget firmware itself uses only the SERVOERR line as feedback.



--
-----
You received this message because you are a member of the LisaList group.
The group FAQ is at http://lowendmac.com/lists/lisa.html
To post to this group, send email to lisalist@googlegroups.com
To leave this group, send email to lisalist+unsubscribe@googlegroups.com
For more options, visit this group at http://groups.google.com/group/lisalist