On Wed, Apr 27, 2011 at 10:26 PM, Richard Guenther <richard.guent...@gmail.com> wrote: > On Wed, Apr 27, 2011 at 10:03 PM, Kai Tietz <ktiet...@googlemail.com> wrote: >> Hello, >> >> this patch adds the ability to gimple-fold to operate for truth and/or >> operations >> on folded binary-and/or optimized truth trees - as done by fold-const. As >> fold >> converts trivial operations like (A && B) to (A & B) != 0, in most cases >> further >> folding of truth and/or trees wasn't done. >> folding will be again detected. >> >> ChangeLog gcc/ >> >> 2011-04-27 Kai Tietz >> >> * gimple-fold.c (is_bit_ior_and): New helper function. >> (fold_equal_and_or): New helper function for truth && >> and || logic folding with treating binary optimization. >> (maybe_fold_and_comparisons): Folding for and. >> (maybe_fold_or_comparisons): Folding for or. >> >> ChangeLog gcc/testsuite/ >> >> 2011-04-27 Kai Tietz >> >> * gcc.dg/binop-tand1.c: New. >> * gcc.dg/binop-tand2.c: New. >> * gcc.dg/binop-tor1.c: New. >> * gcc.dg/binop-tor2.c: New. >> >> Tested for x86_64-pc-linux-gnu and x86_64-w64-mingw32. Ok for apply? > > No. This is certainly the wrong place to do tree combining.
Either look into extending tree-ssa-forwprop.c or work on http://gcc.gnu.org/ml/gcc-patches/2011-03/msg01099.html, thus fold to piecewise gimple expressions so that the machinery can also be used from value-numbering. Richard.
