Re: uw_install_context() and GDB

[Phil Muldoon <pmuldoon at redhat dot com>]

Daniel Jacobowitz wrote:
> On Thu, Aug 14, 2008 at 05:46:32PM +0100, Phil Muldoon wrote:
>   
> > The first was modifying the "next" command to place an  
> > implementation-internal breakpoint in __cxa_begin_catch, and comparing  
> > frames. (Is the frame  that "next"  was performed on  above the  
> > current-frame? If so stop). But this method would skip any user-written  
> > destructors in the _UA_CLEANUP_PHASE. If any user-written destructors  
> > exist in the frame "next" was run on, or any frame inbetween that and the 
> > exception-handler, then the inferior should be stopped there - at the 
> > first one.  This would be the proper behaviour for "next".  Placing a 
> > breakpoint in __cxa_begin_catch would skip those. So I do not think this 
> > method is suitable.
> >     
>
> I think you have to justify that assertion ("this would be the proper
> behavior").  

Quite right. That sentence really should have been prefixed by "In my 
opinion" ;)

> Why would it be?  They're like signal handlers or
> function calls; breakpoints in them would stop the program, but I
> don't see a reason "next" should see them

I take your point well. But if these are user-written destructors, and 
they are being executed on the journey to the exception handler, 
shouldn't "next" return control here? There is a little bit of irony. In 
a conversation I was having recently, I was making a case of "next" over 
a throw always returning control at the corresponding "catch", and 
ignoring the destructors. I changed my mind when it was suggested that 
significant and important work relevant to the code a user has written 
happen in destructors. But quite right, I glossed over this, and should 
make room for conversation on it.

Regards

Phil