> On Jul 4, 2014, at 9:41 PM, Bruce Mitchener <[email protected]> wrote:
> 
> A couple of questions about exception breakpoints:
> 
> 1) In SetExceptionBreakpoints in the various LanguageRuntime plugins, the C++ 
> one bails out if m_process is not valid, while the Objective C version 
> doesn't perform that check. Is the check necessary? Should both runtimes 
> perform it? (Should my Dylan language runtime plugin do it?)

You can't have a language runtime if you don't have a process (they get made 
with a pointer to their process, and are managed by the process.  So this check 
does not seem necessary.

> 
> 2) ThreadPlanCallFunction requires code for each language runtime currently 
> to deal with exception breakpoints (and a new member variable). Wouldn't it 
> be better to iterate over the active / available language runtime plugins 
> instead (using PluginManager::GetLanguageRuntimeInstances())

It would be fine to do this in a more abstract way, but you have to remember to 
only clear them if you set them.  Right now these are shared resources.

Jim

> 
> I'm happy to submit patches based on answers to the above.
> 
> Thanks,
> 
>  - Bruce
> 
> _______________________________________________
> lldb-dev mailing list
> [email protected]
> http://lists.cs.uiuc.edu/mailman/listinfo/lldb-dev

_______________________________________________
lldb-dev mailing list
[email protected]
http://lists.cs.uiuc.edu/mailman/listinfo/lldb-dev

Reply via email to