This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: debugging info considered harmful to lto.
On Mon, Jul 23, 2007 at 08:56:11PM +0200, Jan Hubicka wrote:
> > On Mon, Jul 23, 2007 at 07:33:46PM +0200, Jan Hubicka wrote:
> > It's mostly supposed to be a space optimization. We get e.g. the
> > names and (sometimes) types of local variables from the origin copy,
> > and only need a location at each inlining site. I'm sure there are
> > other things we could do with this information, that rely on knowing
> > what variables are the same from the user's point of view.
>
> Thanks for explanation - the space optimization seems relatively
> chalenging to implement, in particular because the variables in scope
> might change in between the time abstract copy is output and the time
> the block referencing to the block via abstract pointer is output.
>
> How things are compared to match and where the optimization is done?
I don't know anything about how this works in GCC; I was describing
how GDB handles it. The blocks don't have any obvious IDs, so
removing blocks with no declarations seems safe to me.
--
Daniel Jacobowitz
CodeSourcery