This is the mail archive of the
gdb-prs@sources.redhat.com
mailing list for the GDB project.
mi/1289: -break-insert with -c buggy
- From: alain at qnx dot com
- To: gdb-gnats at sources dot redhat dot com
- Cc: jld at ecoscentric dot com
- Date: 17 Jul 2003 20:31:21 -0000
- Subject: mi/1289: -break-insert with -c buggy
- Reply-to: alain at qnx dot com
>Number: 1289
>Category: mi
>Synopsis: -break-insert with -c buggy
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: unassigned
>State: open
>Class: change-request
>Submitter-Id: net
>Arrival-Date: Thu Jul 17 20:38:00 UTC 2003
>Closed-Date:
>Last-Modified:
>Originator: alain@qnx.com
>Release: unknown-1.0
>Organization:
>Environment:
GNU/Linux gdb 5.2.1
>Description:
This is an excerpt form the PR in Eclipse/CDT describing the
problem:
When a debug session is started using the "GDB Debugger", the breakpoints are set using the
GDB/MI command "-break-insert". Conditional breakpoints set this way appear to work
correctly.
If the condition is subsequently changed during a debug session using the breakpoint
properties dialog box, then a "-break-condition" command is issued by CDT. The breakpoint
may then be hit erroneously when the new condition is not satisfied. The behaviour is
consistent with the "-break-condition" command having the effect of erroneously removing the
breakpoint condition.
In brief:
-break-insert -c "i==10" hello.cc:23
then do
-break-condition 1 i==20
the breakpoint will always be hit even if i != 20
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted: