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: Non-call exceptions versus cse


> >     2002-11-25  Andrew Haley  <aph@redhat.com>
> > 
> > 	* cse.c (cse_insn): Don't cse past a basic block boundary.
> > 
> > This seems to be *greatly* limiting CSE.  I'd suggest some "ligher" change.
> > The whole point of CSE is that it is *not* limited by basic block boundaries.
> I am just looking on how hard would be to make cse_basic_block to use
> CFG instead of inventing it by it's own.  Then we can thread trapping
> insns as jump insns easilly.  This is bit crazy, as CSE notion of basic
> block is superblock, so I need to keep CSE basic block code and only use
> current CFG to search for basic block boundaries.  That should be
> non-intrusive enought for mainline.
Hmm, this won't help, as CSE is looking for superblock and that one is
quite safe to recongnize using labels...

Honza
> 
> Honza


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