We crash in the gimplifier because it sees a shift expression with different types of op1/op2. The problem arises in cp_fold: it gets "1 ? d.c : 0" of type int which is folded to "d.c", but it then sees that d.c is a bit-field with declared type long int, so if produces "(long int) d.c" -- but that differs from the original COND_EXPR type! Folding should not change the type of the CODE_EXPR, thus this fix.
Bootstrapped/regtested on x86_64-linux, ok for trunk/7/6? 2017-07-31 Marek Polacek <pola...@redhat.com> PR c++/81607 * cp-gimplify.c (cp_fold): If folding exposed a branch of a COND_EXPR, convert it to the original type of the COND_EXPR, if they differ. * g++.dg/other/bitfield6.C: New test. diff --git gcc/cp/cp-gimplify.c gcc/cp/cp-gimplify.c index a9563b1a8cd..9fc4ab18207 100644 --- gcc/cp/cp-gimplify.c +++ gcc/cp/cp-gimplify.c @@ -2314,9 +2314,9 @@ cp_fold (tree x) /* A COND_EXPR might have incompatible types in branches if one or both arms are bitfields. If folding exposed such a branch, fix it up. */ - if (TREE_CODE (x) != code) - if (tree type = is_bitfield_expr_with_lowered_type (x)) - x = fold_convert (type, x); + if (TREE_CODE (x) != code + && !useless_type_conversion_p (TREE_TYPE (org_x), TREE_TYPE (x))) + x = fold_convert (TREE_TYPE (org_x), x); break; diff --git gcc/testsuite/g++.dg/other/bitfield6.C gcc/testsuite/g++.dg/other/bitfield6.C index e69de29bb2d..c1e8a17989b 100644 --- gcc/testsuite/g++.dg/other/bitfield6.C +++ gcc/testsuite/g++.dg/other/bitfield6.C @@ -0,0 +1,9 @@ +// PR c++/81607 + +int a; + +struct b { + long c : 32; +} d; + +char f = (903092 ? int(d.c) : 0) << a; Marek