| Message ID | AANLkTinHQ=OmmyPAV6fpHiFAsOxtyn69nH1mU0-RYQOa@mail.gmail.com |
|---|---|
| State | New |
| Headers | show |
On Sat, Nov 20, 2010 at 8:03 PM, Uros Bizjak <ubizjak@gmail.com> wrote: > Hello! > > Attached patch fixes PR 43057 by unsharing arg0 argument. > > 2010-11-20 Uros Bizjak <ubizjak@gmail.com> > > PR middle-end/43057 > * fold-const.c (fold_ternary_loc) <COND_EXPR>: Unshare arg0 when > converting A ? 1 : 0 to A. > > Patch was bootstrapped and regression tested on x86_64-pc-linux-gnu > {,-m32}. FWIW, "--enable-checking=all" bootstrap is still running and > it already compiled up to stage-2 gcc/dse.o. I will leave the > bootstrap overnight to see if it breaks (it probably won't). > > OK for mainline and 4.5? Ok. Thanks, Richard. > Uros. >
Index: fold-const.c =================================================================== --- fold-const.c (revision 166974) +++ fold-const.c (working copy) @@ -13415,7 +13415,7 @@ fold_ternary_loc (location_t loc, enum t a COND, which will recurse. In that case, the COND_EXPR is probably the best choice, so leave it alone. */ && type == TREE_TYPE (arg0)) - return pedantic_non_lvalue_loc (loc, arg0); + return pedantic_non_lvalue_loc (loc, unshare_expr (arg0)); /* Convert A ? 0 : 1 to !A. This prefers the use of NOT_EXPR over COND_EXPR in cases such as floating point comparisons. */
Hello! Attached patch fixes PR 43057 by unsharing arg0 argument. 2010-11-20 Uros Bizjak <ubizjak@gmail.com> PR middle-end/43057 * fold-const.c (fold_ternary_loc) <COND_EXPR>: Unshare arg0 when converting A ? 1 : 0 to A. Patch was bootstrapped and regression tested on x86_64-pc-linux-gnu {,-m32}. FWIW, "--enable-checking=all" bootstrap is still running and it already compiled up to stage-2 gcc/dse.o. I will leave the bootstrap overnight to see if it breaks (it probably won't). OK for mainline and 4.5? Uros.