home *** CD-ROM | disk | FTP | other *** search
Text File | 1989-07-18 | 2.2 KB | 53 lines | [TEXT/GEOL] |
- Item forwarded by BURBECK.S to MACAPP.TEST
-
- Item 4809312 13-July-89 21:01
-
- From: D2605 Xanadu Operating,Marc Stiegler, PRT
-
- To: FRIEDRICH1 Friedrich, Steve
-
- cc: BURBECK.S Burbeck, Steve
- MACAPP.ADMIN Design Methodology, Joel Norvell
-
- Sub: APDA release MacApp death
-
- Steve,
- I just got a copy of the new (2.0ß9-1) APDA release MacApp and I have 2
- significant problems with it. First, when my appl is the first one run after a
- boot, it (sometimes - as in twice) gets a bus error (caught by the MacApp
- debugger) in the TApplication.IApplication method while initializing the
- menu's. The address referenced is db6db6 (i think). Everything is fine on the
- second run, or if something is run before it (which I usually do). I couldn't
- get it to fail again when I was writing this message so I don't have more
- accurate information. (Next time, I'll write it down...)
- The second problem is the real one. My application uses MacTCP to connect
- to a Sun running a server. After the first packet is received from the Sun
- (which is sent automatically after a Connect request is honored) the program
- now bombs. The MacApp debugger prints out:
- Broke at Begin UNPATCHTRAP Seg#:1
- Command [ ... ]:|
-
- after which the system breaks into MacsBug, which prints out:
- Spurious interrupt or Unitialized Interrupt Vector at: 2E680061
- 2E680061 * ORI.B #$4E,D0 ; 'N' | 0000 004E
-
- at which point the system is totally locked up, no response. Reset switch is
- required.
- As might be surmised, this all worked fine under the previous version of
- MacApp that I was using (2.0ß8.1). I have been trying to track down the exact
- source of the problem, but as you might guess, it is rather time consuming
- because I have to step through the application carefully, and reboot often. I
- will continue to try to locate the exact source, but I wondered about 2 things:
-
-
- a) has anyone else reported funny things like this?
- b) where there any changes to the patch, debug, or failure code that might
- cause a change in interrupt handling? (i did some compares but found no
- important(?) differences)
-
- Thanks,
- --Hugh Hoover
- Xanadu Operating Company
-
-
-