This is the mail archive of the
mailing list for the GCC project.
Re: stepanov on sparc; sparc backend scheduling issues
Paolo Carlini <firstname.lastname@example.org> writes:
> Daniel Berlin wrote:
> > Well, one thing talked about (struct splitting) *is* basically scalar
> > replacement of aggregates.
> > At least, for dwarf2, we can now track these values if someone wanted
> > to.
> > You can do it with the location lists, and DW_OP_piece.
> Excuse my naivete: you mean that, due to work carried out for the dwarf2 debugging
> format, most of the infrastructure necessary for this kind of optimization is ready
> to use???
I mean that in the threads you posteed, someone pointed out that doing
this type of thing really screws up debugging info, if you have no way
to track the location throughout the compiler, to inform the debugger.
We didn't before.
We basically can now.
"When I was a kid, I went to the store and asked the guy, "Do you
have any toy train schedules?"