This is the mail archive of the
archer@sourceware.org
mailing list for the Archer project.
Re: feature idea: upto
> We discussed this a little. My recollection is that Daniel tried this
> but concluded that MI clients get confused when gdb stops and the
> selected frame is not the newest frame. So, some kind of MI extension
> or change is needed.
For "catch throw" I can sort of see the point of the "inner" frame. But
perhaps it can be presented better. That is, it really is the frame where
the exception exists and you can find the exception object via ABI rules.
In the calling frame, it's really just some caller that constructed an
object and is passing it down. This has already happened, so the work of
that "throw" statement (i.e. the object construction) is indeed not where
you stopped. I think of it as a bit like a signal frame. It is really
there and distinct, but the next one out is the interesting one to think
about.