This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fix bootstrap on SPARC/Solaris with RTL checking
> > 2007-06-15 Eric Botcazou <ebotcazou@libertysurf.fr>
> >
> > * df-scan.c (df_get_call_refs): Be prepared for MEMs inside CLOBBERs.
>
> Can a dataflow maintainer take a look at it?
> http://gcc.gnu.org/ml/gcc-patches/2007-06/msg01093.html
> I think I can self-approve it, but there might be subtleties lurking that I'm
> not aware of. Thanks in advance.
The PR32374 has occurence of similar problem, just in more slipperly
case - here we emit clobber of memory slot referenced via
virtual_stack_regs_reg and instantiate_virtual_regs skips the clobber.
Originally I think we got rid of it at nearest dead code removal since
clobbers of memory don't seem to have good meaning, but now the clobber
confuses dataflow by exposing the virtual register after
instantiate_virtual_regs.
Perhaps rather than teaching dataflow to tolerate clobVbers of memory
and perhaps the dead but still appearing virtuals or teaching
instantiate to instantiate virutals, we can just avoid the clobbers of
memory from being emitted at first place?
Honza
>
> --
> Eric Botcazou