This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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: PR 23551: why should we coalesce inlined variables?


On Mon, 2007-07-09 at 12:05 -0300, Alexandre Oliva wrote:
> On Jul  9, 2007, "Richard Guenther" <richard.guenther@gmail.com> wrote:
> 
> > We probably will need to end up with a list of debug symbols per ssa-name.
> 
> Not only that.  We need to be able to mark the ranges in which each
> variable is associated with that value.  Without the ranges, we can't
> generate accurate debug information.
> 

The ranges are the ssa_name range itself.  There will be some minor
issues to deal with when dealing with ssa_name overlap, but its nothing
we can't sort out when -g is specified, and will therefore not cost
computation time when -g isnt specified.  So it wont change the
generated code, it just might mean a little more conflict resolution to
separate out when each of the  multiple ssa_names attached actually
expire. It will just require a little processing and thought.

And its WAY better than what we have now. 

Andrew


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