| RE: [cdt-dev] CDT 3.0 RC3 Release Schedule |
|
My vote is for #2… I’m happy with the APIs I think. The screaming of not being able to walk the stack will far outweigh the screaming of those that don’t want us to change the APIs so late.
I’d prefer if also we implemented the fix for 3.0 and delayed the build for the fix but I don’t know how everyone feels about that.
___________________________________________
Chris Recoskie Software Designer IDE Frameworks Group Texas Instruments, Toronto
From: cdt-dev-bounces@xxxxxxxxxxx
[mailto:cdt-dev-bounces@xxxxxxxxxxx] On
Behalf Of Sebastien Marineau
Hi all,
with respect to 104421 (registers in the debug view), Alain, Mikhail and Dave have been working on the suggestion in yesterday's email (option #2) of implementing the CDI change (but not the full fix), to avoid a CDI breakage in the future.
This is the options we have now:
1. We ship what we have today, and break CDI in the next bugfix release to address the issue. This is the least risky option but is may not be the most desirable from an API compatibility standpoint.
2. We change the CDI interface and "plumb" around it on the debug UI and back-end. The original bug is not fixed, but we can provide the fix in the next release without any CDI changes. The impact of going this route: - Alain has posted on the debug-dev mailing list the suggested CDI changes. We would need people to review this ASAP (Freescale, TI, etc) to ensure we are satisfied with the new API. - Implement the change tonight/tomorrow, which means holding off on the final RC at least for a day. - This introduces some risk with last-minute changes.
What are people's thoughts on this?
As for the C++ browser (102195), Alain is working on disabling it.
Sebastien
|