Path: utzoo!censor!geac!torsqnt!lethe!yunexus!ists!helios.physics.utoronto.ca!news-server.csri.toronto.edu!rutgers!usc!samsung!umich!sharkey!tygra!dave
From: dave@tygra.ddmi.com (David Conrad)
Newsgroups: comp.lang.pascal
Subject: Re: Turbo 6.0: problem with inthtegrated debugger.
Summary: No code, no breakpoint.  No breakpoint, no run-to.
Keywords: debugger, breakpoint, code generation
Message-ID: <1991Jan10.084610.4035@tygra.ddmi.com>
Date: 10 Jan 91 08:46:10 GMT
References: <2394@bnlux0.bnl.gov> <1991Jan9.183514.7754@isis.cs.du.edu>
Organization: Moore Software, Detroit, MI
Lines: 42

In article <1991Jan9.183514.7754@isis.cs.du.edu> nlarson@isis.UUCP (Neil Larson) writes:
>In article <2394@bnlux0.bnl.gov> kushmer@bnlux0.bnl.gov (christopher kushmerick) writes:
>>I am having what appears to be recurring and intermittant problems
>>with the integrated debugger. I am trying to use the F-4 (run to  cursor)
>>command, and it reports that there is no code generated for this line.
>
>calculatecrc16(packet,crcval);
>if crcval = 0 then
>   crcvalerror := true;
>calculatecrc16(packet,crcval);
>
>the compiler will not create identical machine code for both calls to the 
>procedure calls, but will just jump to the previous call.

What??!?  Unh-uh.  The if statement would get executed twice.  The compiler
won't do anything like that, thank von Neuman.
 
There are some lines of source code for which no instructions are generated:
 
repeat                  {this doesn't *do* anything, just a marker}
  ch := readkey;        {calls readkey, places return value in ch}
  x := x + 5;           {load x into register, add five, store to x}
until ch = #27;         {load ch, cmp with $1b, jne to top of loop}
 
Now, the way the debugger sets breakpoints is by substituting an interrupt
instruction for the first instruction at a particular line of source.  The
way F4-Go to cursor is implemented is by setting a breakpoint at the line
to go to.  But if no instructions are generated by the line, then there is
nowhere to put the interrupt instruction.  Hence the message.
 
Now a *really* smart debugger would set a breakpoint at the next line which
generates some code and set a flag somewhere indicating the actual source
line to position the cursor at when the breakpoint was encountered.  But
that's a hefty kludge to workaround a pretty minor annoyance, eh?
--
David R. Conrad
dave@tygra.ddmi.com
-- 
=  CAT-TALK Conferencing Network, Computer Conferencing and File Archive  =
-  1-313-343-0800, 300/1200/2400/9600 baud, 8/N/1. New users use 'new'    - 
=  as a login id.  AVAILABLE VIA PC-PURSUIT!!! (City code "MIDET")        =
   E-MAIL Address: dave@DDMI.COM