| Summary: | gdb invoked from cdt crashes when the path to the project is long enough | ||
|---|---|---|---|
| Product: | [Tools] CDT | Reporter: | David Komanek <komanek> |
| Component: | cdt-debug | Assignee: | cdt-debug-inbox <cdt-debug-inbox> |
| Status: | NEW --- | QA Contact: | Jonah Graham <jonah> |
| Severity: | normal | ||
| Priority: | P3 | CC: | elaskavaia.cdt, komanek, pawel.1.piech |
| Version: | 6.0.2 | ||
| Target Milestone: | --- | ||
| Hardware: | PC | ||
| OS: | Windows Vista | ||
| Whiteboard: | |||
|
Description
David Komanek
Probably the same experience by another user: http://www.eclipse.org/forums/index.php?t=tree&th=164529&#page_top some more hints: - there is a possibility it does not depend on the length of the pathm but on the number of cerain characters in the path (backslash, underscore ? - it seems to be related to Windows Vista - this problem occurs on my Windows Vista 64bit, but I am not experiencing this with Windows 7 64bit If gdb crashes why you sending bug to cdt? Should be gdb bug. Because if I invoke it manually it works fine, but if it is called from Eclipse (debugging the same executable), it does not work well. It is somehow dependent on the path to the Eclipse project in the filesystem, so I suppose, there are some parameters or environment values passed from cdt to the debugger which are causing it. I agree that gdb should give a well formulated response to this, but at least there is a difference using gdb from cdt or by manual invocation. You can post debug console here (Enable vebose console mode in the Debug tab and post gdb console). But if app is crashing is it app problem - no matter how "incorrect" are arguments are it should not crash. I´m sorry, I don´t see such an option. Hopefully, someone smarter than me will be able to reproduce this and point the gdb community. Since I have the workaround by moving the project directory to a shorter filesystem path, I can live with it. Thank you for your assistance. |