Andy Valencia wrote:
>>This one at $FF00 is the process for screen #0... it exists, and is
>>runnable, so now I need to find out what it's doing. I'm going to
>>put together some instructions that'll let you dump out the return
>>stack for this guy, so I can start to guess what he's doing instead
>>of sitting in the CLI waiting for keystrokes.
>
>
> Ok, here we go...
>
> $FF00 4 dumpw ( will dump 4 words in hex )
> The fourth word is the return stack, probably $FE90
> $FE90 24 dumpw ( dump out 24 words of the return stack)
> Now start decoding words starting from the third line, fourth hex
> number on that line.
> also extensions ( prval is in the extensions vocab )
> $12D6AC (prval) ( on my system, this word was 12D6AC, so I type it in )
> ( hex notation and have (prval) print it symbolically )
> Now do the same for successive words; this is a poor man's stack backtrace
> for whatever's going on with screen 0. I found 6-8 words to be enough;
> don't be freaked if some values look weird from (prval), because the
> return stack holds various values, not just return addresses.
>
> Andy
>
Andy,
Obtained the following
4th word was FE88 when I did $FF00 4 dumpw
Translations of return stack info from address $FE88
127B28 (resume)+8
E910
127CFC (runit)+20
127D80 (scan)+108
127D80 (sched)+60
4
3
127DC8 (sched)+24
127E7C (mpause)+44
1016C0 @execute+28
104A98 pause+16
12DDB0 waitdrq+28
12DF1C ideio+336
8
0
12DF14 ideio+328
12E11C init-ideoff+44
12E1D4 boot-ide+36
1016C0 @execute+28
104D74 cold?+512
104F10 cold+48
2E255
83F975C0
F889FCC4
Hope this gives you some idea of what is happening.
One other point, the caps lock doesn't work - even though shift does.
Hex numbers must be entered in uppercase.
regards
Bruce
Received on Mon, 17 Jan 2005 01:45:02 +1100
This archive was generated by hypermail 2.1.8 : Tue Sep 26 2006 - 09:03:04 PDT