[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference turris::debug

Title:DEBUG
Notice:Updated locations for reporting QARs -- see note 834.1
Moderator:LOWFAT::DIETER
Created:Fri Jan 24 1986
Last Modified:Wed Jun 04 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:1868
Total number of notes:8200

1858.0. "Ada source line wonrg/incorrect w DEBUG V7.1C Alpha" by CSC32::D_SANFORD () Thu Apr 10 1997 14:45

    OpenVMS Alpha V6.2
    Debug V7.1C
    DEC Ada V3.3

    This is a long shot!  The customer says everything we want to
    hear, so hard to see if this is user induced.  The code below
    is not actual code, just a customer making the problem sound
    straight forward, but at the same time can't duplicate in a
    simple example.  I beleive this also fails using character 
    cell interface, but they only use the GUI.

    The user claims, while using the GUI, that the debugger source 
    window indicates the wrong source line.  Usually only off 3-4
    lines.  For example:

	->	100	something := 1;
		102	if not ... then
		103	   Text_IO.Put_line ("whatever");
		104	end if;
		105	Text_IO.Put_Line ("Done");

    When he STEPs, the "Done" is printed to SYS$OUTPUT.  He claims the
    source file is the original, not purged renamed or whatever.

    I asked him to try:

	DBG> SET SOURCE []

    To force debugger to reload source lines, not sure what affect this
    will have.

    He said the DEBUG V6.2 has the same problem, but this is all here
    say.  The code is large, defense related so wouldn't get it, and a
    simple example does not fail.

    Any tricks to determine if this problem is user induced?  Looked
    at relateds notes 1796 & 1781.

    Regards, -drew
    
T.RTitleUserPersonal
Name
DateLines
1858.1LOWFAT::DIETERThu Apr 10 1997 15:0318
Hi Drew -

V6.2 _did_ have a problem with "wrong source lines" for which there
is an ECO.  The problem was fixed in V7.1 and V7.1C.  Can you verify
that he has installed the V6.2 ECO (to correct the problem with the
V6.2 debugger) and that he is actually using the V7.1C debugger when 
he says he is?

For V7.1, is he compiling his code /NOOPT?

Sometimes it helps to display the instruction view/display, with line
numbers, to see which instructions associated with which lines are 
actually getting executed.  

BTW, the V6.2 "wrong source line" problem was also character cell...

Mary