On Wed, Nov 6, 2013 at 11:31 PM, Richard Sandiford <rdsandif...@googlemail.com> wrote: > Ping.
Ok. Thanks, Richard. > Richard Sandiford <rdsandif...@googlemail.com> writes: >> As discussed on gcc@, integer_onep is supposed to return false for >> a nonzero 1-bit value if the type is signed. >> >> Tested on x86_64-linux-gnu and powerpc64-linux-gnu. OK to install? >> >> Thanks, >> Richard >> >> >> Index: gcc/tree.c >> =================================================================== >> --- gcc/tree.c 2013-10-29 19:19:27.623468618 +0000 >> +++ gcc/tree.c 2013-11-02 17:25:06.499657501 +0000 >> @@ -2092,7 +2092,7 @@ integer_onep (const_tree expr) >> switch (TREE_CODE (expr)) >> { >> case INTEGER_CST: >> - return wi::eq_p (expr, 1); >> + return wi::eq_p (wi::to_widest (expr), 1); >> case COMPLEX_CST: >> return (integer_onep (TREE_REALPART (expr)) >> && integer_zerop (TREE_IMAGPART (expr)));