This is the mail archive of the gcc@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: My opinions on tree-level and RTL-level optimization


    You seem to be confused.  We've known *why* CSE does stuff that GCSE
    doesn't catch for almost as long as we've had GCSE.

    It's because CSE *doesn't just do CSE*!  It does value numbering, and
    a bunch of other things, which are not really implemented at the RTL
    level as seperate passes, 

Well, sure, but most of the benefit of running those is within a basic
block.  I don't see why the combination of a global just-CSE with the
current intra-block code wouldn't be effective.

    Also, the viewpoint that absolutely everything CSE currently does
    needs to be done in order to remove CSE is wrong.

I'm not talking about removing CSE.  Indeed, the part of CSE that
chooses the best operand from a cost point of view likely needs to stay
forever.  I was just talking about removing the following of jumps.

    The correct viewpoint is "we shouldn't remove CSE until every
    *profitable* transformation it makes is subsumed by something else".

And, as I understand it, the claim is that this is not yet true for the
following of jumps and my question is why.


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