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: [patch] Speed up operand_equal_p during SSA optimizations.


    We have code to do another pointer comparison after STRIP_NOPS.

    > (3) Why isn't the "&&" an "||"?  If either operand is an SSA_NAME, the
    > only way they can be the same is if both are identical.

    A good point.

Ah, but if you made that change, then you *do* have to move it past
the STRIP_NOPS call ...

    I guess so, but I don't want to complicate the new check.  After all,
    this is a hack.  Without my patch, operand_equal_p works just as well,
    only slower.

Can you even measure the speedup?  All these cases at the front of
the function produce more complexity and slow down the other cases.

I'm really not sure this is worthwhile.


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