eric@nosun.west.sun.com (Eric Hanchrow) (11/28/89)
Here's a bizarre problem I noticed. I've been developing and testing a program on our Sun-4, running SunOS 4.0.3. As I've been going along, I've been running it under a debugger now and again, as problems show up. But the last time I made a fix to the code, recompiled, and started up the debugger, the debugger hung. "What debugger were you using?" you ask. I was using the Gnu debugger, release 3.2. I was compiling with the Gnu compiler, gcc version 1.35. Here's a transcript: Current directory is /home/eric/work/drivers/fx100/new/sun4/ GDB 3.2, Copyright (C) 1988 Free Software Foundation, Inc. There is ABSOLUTELY NO WARRANTY for GDB; type "info warranty" for details. GDB is free software and you are welcome to distribute copies of it under certain conditions; type "info copying" to see the conditions. Reading symbol data from /home/eric/work/drivers/fx100/new/sun4/fx100...done. Type "help" for a list of commands. (gdb) br main Reading in symbols for awk.c...done. Reading in symbols for bitmap.c...done. Reading in symbols for main.c...done. Breakpoint 1 at 0x6b84: file main.c, line 46. (gdb) run Starting program: /home/eric/work/drivers/fx100/new/sun4/fx100 And that's the last I hear from it. (I waited about ten minutes.) I don't think it was spending its time reading in a core file, because there wasn't a core file in that directory. And not only did the debugger process seem to hang, but I couldn't kill it with control-C. Or control-Z, either. And for that matter, `kill -9' didn't do it either. (As I write this, I have about a dozen of these never-say-die debug processes hanging around. They don't seem to use up any CPU cycles, and I don't want to reboot the machine unless I have to.) Now the interesting parts -- the problem does not occur on our RoadRunner (SunOS 4.0.1), or our Sun-3 (SunOS 3.5). And yet it *does* occur with dbx. I've tried all the combinations of platform/compiler/debugger/binding-option which make sense (e.g., gdb does not run on the 386i); in all cases, the attempts on the Sun-4 hung, and the attempts on the other platforms worked just fine. Does this sound familiar to anybody?? ----------------------- Date: 27 Nov 89 20:36:43 GMT From: "Joe Habermann" <haberman@cs.umn.edu> Subject: wanted: interface builder for SunView Does anyone know of an interface builder for SunView? What were looking for doesn't have to be Public Domain, but it must be well done. Yes, I know that Sun is saying that SunView is becoming XView. Ok, then, an XView interface builder. I'd especially like to hear from people have actually purchased such software and hear what they have to say about it, but I welcome replies from vendors as well. Sincere Thanks in Advance, Joe Habermann / haberman@umn-cs.cs.umn.edu / ..!rutgers!umn-cs!haberman
tro@uunet.uu.net (Tom Olin) (12/26/89)
I, too, have recently experienced dbx hanging while debugging in OpenWindows. There does not appear to be any specific event that causes it to hang. Often, it dies while I'm just looking and thinking. (Say, now there's an event!) Suddenly, the mouse freezes and everything stops moving, including xclock and perfmeter. I have to login on another terminal and kill -9 dbx. The environment: Sun-4/150, SunOS 4.0.3, OpenWindows 1.0 (FCS). I watch for clues but haven't noticed any yet.