Title: | AMIGA NOTES |
Notice: | Join us in the *NEW* conference - HYDRA::AMIGA_V2 |
Moderator: | HYDRA::MOORE |
Created: | Sat Apr 26 1986 |
Last Modified: | Wed Feb 05 1992 |
Last Successful Update: | Fri Jun 06 1997 |
Number of topics: | 5378 |
Total number of notes: | 38326 |
My A2000 has become very "crash-happy" lately. There have been a number of Task Helds at different times. Deluxe Paint III crashes a lot giving the guru code #00000003.0025F498. Sometimes, just opening a window causes the system to crash. This gives #00000003.002332E8. Anyone out there who can tell me what these guru numbers translate as? Any help appreciated. Randy
T.R | Title | User | Personal Name | Date | Lines |
---|---|---|---|---|---|
4390.1 | guru codes | HPSCAD::GATULIS | Frank Gatulis 297-6770 | Wed Jan 02 1991 14:13 | 5 |
See note: 1790.2 It has the codes listed. | |||||
4390.2 | Chip RAM Depletion? | ULTRA::KINDEL | Bill Kindel @ LTN1 | Wed Jan 02 1991 14:36 | 21 |
Re .0: > My A2000 has become very "crash-happy" lately. There have been a > number of Task Helds at different times. Deluxe Paint III crashes a > lot giving the guru code #00000003.0025F498. Sometimes, just opening > a window causes the system to crash. This gives #00000003.002332E8. > > Anyone out there who can tell me what these guru numbers translate as? I think the 00...03 in the upper longword is an address error. The lower longword gives the address of the offending instruction. The most probable cause is depletion of Chip RAM. If you open a shell window while things are still healthy, you may be able to do an AVAIL to confirm that suspicion when you get a Task Held indication. After a spate of similar problems, I downloaded ClockDJ and made it part of my startup so I continually show the time and the amount of available chip/fast RAM at the top of my workbench screen. If the clock stops, you're in BIG trouble. If the amount of Chip RAM drops below about 20KB, the Guru isn't far behind. | |||||
4390.3 | KALI::VISSER | Wed Jan 02 1991 14:52 | 5 | ||
'03.' is an address error, which is a trap for trying to access a word or long word with an odd address; I thought only a hardware problem, or really bad software could cause this. ??? John | |||||
4390.4 | Smashing low memory | TLE::RMEYERS | Randy Meyers | Wed Jan 02 1991 18:41 | 5 |
Re: .3 If a program ignores out of memory conditions, it will usually start scribbling all over low memory. The result can very easily cause the one of the operating system's pointers to be "odd." | |||||
4390.5 | PD checking program | SAC::EXTON_M | My 2 ECU's worth | Thu Jan 03 1991 03:49 | 9 |
I saw a reference to a program in Amiga Format which purported to check system memory about 50 times per second (and not cause much overhead). This would tend to catch low mem corruption near to the source. Can anyone provide further detail on this as the AF's belong to my Nephew. Malcolm |