| Re: [cdt-dev] Invalid Breakpoint Handling |
Hi Tracy, Thank you for looking at the patch. I have been trying to address Marc Khouzam’s comments regarding error reporting but haven’t been able to find a better solution. I agree, we can commit the patch as it is and file a separate Bugzilla for the error reporting issue. I will update the patch and push the latest version to Gerrit as soon as I can. Mikhail From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Tracy Miranda Hi all, Following up from a discussion at the call yesterday related to CDT's handling of invalid breakpoints. Jonah & I had a review of the most recent gerrit for https://bugs.eclipse.org/bugs/show_bug.cgi?id=367256. It looks good, it makes that specific case much better & clearer to the user, so thanks Mikhail! The comment related to the new GDB console is a separate concern, down to GDB behaviour (and captured in this bug Marc already reported: https://sourceware.org/bugzilla/show_bug.cgi?id=20959) In general, the handling of invalid breakpoints is pretty unkind to our users. Part of the problem with fixing it is some of the issues lie with GDB. Plus there is alot of nuance with different gdb versions, different console behaviour and types of launches (e.g. remote). Nevertheless distilling the various issues and tackling them one by one seems like a good way to go for now. To sum up, it would be great to get this change in and then build on that to keep improving the user experience for invalid bps. This will be key to having good multicore breakpoint support. Tracy |