This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: exception handling predicament


On Fri, Aug 19, 2011 at 11:39:59AM -0700, David Miller wrote:
> From: Jakub Jelinek <jakub@redhat.com>
> Date: Fri, 19 Aug 2011 20:22:08 +0200
> 
> > GCC doesn't have a -fasynchronous-exceptions option which would
> > consider every single insn to be able to trap (and I think it would be hard
> > to implement it).
> 
> I don't think GCC should implement this, but why would it be so hard?

Because then every insn/stmt needs to end the basic block and have EH edge,
so far more optimization passes would actually need to update EH when doing
any changes.

> All we care about is the scope of the __cleanup__ variable, and the
> basic blocks encompasing that is very well defined.

This really isn't specific to __cleanup__, if you use C++ and destructors
it is exactly the same.

	Jakub


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]