Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!bbn!apple!well!shf
From: shf@well.UUCP (Stuart H. Ferguson)
Newsgroups: comp.sys.amiga.tech
Subject: Re: Debugging libraries with SDB.
Message-ID: <11050@well.UUCP>
Date: 21 Mar 89 23:30:57 GMT
References: <10964@well.UUCP> <3640@mit-amt> <15885@oberon.USC.EDU>
Reply-To: shf@well.UUCP (Stuart H. Ferguson)
Organization: The Blue Planet
Lines: 24

+-- papa@pollux.usc.edu (Marco Papa) writes:
| >In article <10964@well.UUCP> shf@well.UUCP (Stuart H. Ferguson) writes:
| >> [...] It turns out that you *can* debug libraries with
| >>SDB.  SDB can maintain two sets of symbol and source code information, and
| >>if a ".dbg" file is created for the library, SDB can trace library calls
| >>in source mode.
| Stuart, have you done it yourself?  I have tried with a "device" and
| got nowhere. sdb always says that it cannot find the device, when I know
| that mydev.dbg file is in the current dir and the device is loaded (after
| the OpenDevice). Has anybody been able to do this REALLY?
 
Yes, I did it.  I had to try several permutations to get it to work, but
the one that did was having the "iffparse.library" loaded in memory, having
the "iffparse.dbg" file on the current directory and having the sources
located in a subdirectory of the current directory.  I used the -sPATH
option on SDB to make the source subdirectory available, and did 
"ll iffparse" in SDB.  Other combinations of names didn't work.

Once the .dbg was loaded, SDB would switch back and forth between symbol
tables with "ax".  Worked well.  I haven't tried it with a device
driver, so I can't say how that works.
-- 
		Stuart Ferguson		(shf@well.UUCP)
		Action by HAVOC