Re: Bug in the debugger?

[Neil Palmer <neilpalmer400mr@xxxxxxxx>]

Report it to IBM.
 
Neil Palmer, Cambridge, Ontario, Canada
 
(This account not monitored for personal mail,
remove the last two letters before @ for that)


----- Original Message ----
From: M Lazarus <mlazarus@xxxxxxxx>
To: MIDRANGE-l@xxxxxxxxxxxx
Sent: Friday, October 20, 2006 12:23:26 PM
Subject: Bug in the debugger?


A colleague was testing the new CL enhancements and came across what
seems to be a bug in the debugger.

Here's the test code:
             PGM        PARM(&QUALOBJ)
             DCL        VAR(&QUALOBJ) TYPE(*CHAR) LEN(20)
             DCL        VAR(&MAINOBJ) TYPE(*CHAR) LEN(20)
             DCL        VAR(&QUAL01) TYPE(*CHAR) STG(*DEFINED) +
                          LEN(10) DEFVAR(&QUALOBJ 1)
             DCL        VAR(&QUAL02) TYPE(*CHAR) STG(*DEFINED) +
                          LEN(10) DEFVAR(&QUALOBJ 11)

             CHGVAR     VAR(&MAINOBJ) VALUE(&QUALOBJ)
             SNDPGMMSG  MSG(&QUAL01) TOPGMQ(*PRV (*PGMBDY))
             SNDPGMMSG  MSG(&QUAL02) TOPGMQ(*PRV (*PGMBDY))

             ENDPGM

CALL TEST PARM( '1234567890abcdefghij' )

The messages generated contain the correct info:
1234567890
abcdefghij

Doing the same CALL, step through via debug.
Both &QUAL01 and &QUAL02 will show 1234567890 throughout the entire run.

This happens in both the green screen and WDSC.

Program is compiled with no optimization.  Is this a debugger bug?

-mark
-- .com/midrange-l.