Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
RE: [cdt-dev] CDT Summit Report

Hi Andrew,
 
Not yet. We didn't get very much time on build as we did on the other topics. I think the summary of it is to redo the Project model to simplify it and to break the dependency on managed build that was introduced with it in 4.0. We also need to redo the interfaces to Scanner Discovery to make it easier to extend, and to break apart the build output scanning and the compiler built-in calculation which are two separate concerns.
 
We didn't discuss at all what to do with managed build itself. I think we still have issues with extensibility there and it would be nice if you could defined a complete managed build definition in Java. But we don't have anyone investing in that area at the moment.
 
I'll do a more complete write up on the wiki and put items into the plan.
 
Cheers,
Doug.


From: cdt-dev-bounces@xxxxxxxxxxx [mailto:cdt-dev-bounces@xxxxxxxxxxx] On Behalf Of Andrew Gvozdev
Sent: Monday, September 29, 2008 11:02 PM
To: CDT General developers list.
Subject: Re: [cdt-dev] CDT Summit Report

Hi Doug,
There are very detailed notes for Debug and Indexing on Wiki, are there any notes on Build session?

Thanks,
Andrew

On Fri, Sep 26, 2008 at 12:14 PM, Schaefer, Doug <Doug.Schaefer@xxxxxxxxxxxxx> wrote:
Hey gang,
 
I thought I'd update everyone on the CDT Summit. I think it was one of our best summits ever. It was smaller, but that helped keep us focused on making decisions and allowed us to dive deep into the topic areas we need to.
 
We captured a lot on the wiki and will be adding things there as we unwind from the week.
 
We spent the first day mainly on the development process we use in our day-to-day activity with the CDT. I think we have a renewed focus on quality over content that will help us make the final touches to improve the CDT architecture and to help address the backlog and new defects coming in.
 
We were able to decide on the CDT componentization issue that we started a month or so ago, which is great news and the team worked hard on coming up with the result. We'll have 15 components, which may seem like a lot, but it will help us focus on those 15 areas to ensure we address quality issues and to help focus our future feature developement areas.
 
On Wednesday, we broke out into breakout sessions, one for indexing and related features, and one for debug. The indexing guys did some deep dives into specific areas that we need to address to improved the quality of the information you can get out of the parser framework including indexing. There was also some good discussion on how we can get information out of code that is inactive due to the preprocessor, capturing as much multi-configuration information as practically possible.
 
The debugger gang talked mainly about DSF and what needs to be done to it to bring it into the CDT fold. There is some work to bring the DSF GDB integration up to feature parity with CDI GDB. We also talked about improvements to debugging with disassembly and how we want to do the disassembly in the editor panes in particular.
 
On Thursday, we did a quick trip through the e4 Resources work and CDT Build. Both of these are areas where we need to clean up the architecture. The Resources work will done in the e4 project but it's important to ensure it meets the needs of the CDT community, as well as others of course. The Build issues are mainly the Project Model and how to simplify it and to decouple standard build from managed build. I think we have a good understanding of what we want to accomplish there, we just need people to work on it.
 
We also talked about the fact for each vendor contributing to CDT, we know of at least 3 others that commercially benefit from the CDT yet don't contribute to it. That's pretty frustrating for us that do and we need to find a way to bring them into the fold. And we need to get some buzz going for the CDT again and do things like more articles and webinars and presenting more at conferences. The CDT is pretty good these days, but if our community starts taking that for granted, it'll stagnate as a project and we don't want that.
 
Anyway, it was a great week and a lot of good came out of it. It should be a good year for the CDT. Thanks to everyone who made the effort, the employers for paying for the trip, and the employees for taking the time away from their lives to attend. And thanks again to our sponsors, Wind River, IBM, and the Eclipse Foundation for making this possible.

Cheers,
Doug.

_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev



Back to the top