| Summary: | [3.3/3.4 regression] valid conversion static_cast<const unsigned int&>(lvalue-of-type-int) is rejected | ||
|---|---|---|---|
| Product: | gcc | Reporter: | Tim Bagot <timb> |
| Component: | c++ | Assignee: | Not yet assigned to anyone <unassigned> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | gcc-bugs, giovannibajo |
| Priority: | P2 | Keywords: | rejects-valid |
| Version: | 3.3 | ||
| Target Milestone: | 3.3.1 | ||
| Host: | i386-linux | Target: | i386-linux |
| Build: | i386-linux | Known to work: | |
| Known to fail: | Last reconfirmed: | 2003-05-22 22:27:55 | |
Confirmed. The code is illegal because the conversion is invalid (basically, you can't bind an int variable to an unsigned int reference), but the diagnostic is totally confusing. I don't think it is illegal, because the reference is to const unsigned int, not unsigned int. Looking at [8.5.3/5], the initialiser is an lvalue but the types are (as you point out) not reference compatible, and the initialiser does not have a class type; therefore the second top-level case applies: '- Otherwise the reference shall be to a non-volatile const type', which does hold. The initialiser is not an rvalue, so the final case applies: 'a temporary of type "cv1 T1" is created and initialised from the initialiser expression using the rules for a non-reference copy initialisation (8.5). The reference is then bound to the temporary.' In fact, initialising a named reference does work, fortunately providing a workaround: const unsigned int& y = x; This issue was discussed on comp.std.c++ here: http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF- 8&threadm=rpcadv8bdp0b5pgfo19831t7cu8v6vl4qc%404ax.com&rnum=1&prev=/&frame=on It appears to be a bug in GCC 3.3/3.4 and a regression with respect to 3.2. It's no more a diagnostic issue. Subject: Bug 10931 CVSROOT: /cvs/gcc Module name: gcc Changes by: mmitchel@gcc.gnu.org 2003-06-26 00:07:10 Modified files: gcc/cp : ChangeLog typeck.c cp-tree.h call.c gcc/testsuite : ChangeLog Added files: gcc/testsuite/g++.dg/expr: static_cast1.C Log message: PR c++/10931 * g++.dg/expr/static_cast1.C: New test. PR c++/10931 * call.c (convert_like): Pass issue_conversion_warnings. (convert_like_with_context): Likewise. (convert_like_real): Add issue_conversion_warnings parameter. (perform_direct_initialization_if_possible): New function. * cp-tree.h (perform_direct_initialization_if_possible): Declare it. * typeck.c (check_for_casting_away_constness): New function. (build_static_cast): Rewrite. Patches: http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.3455&r2=1.3456 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/typeck.c.diff?cvsroot=gcc&r1=1.467&r2=1.468 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/cp-tree.h.diff?cvsroot=gcc&r1=1.859&r2=1.860 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&r1=1.391&r2=1.392 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.2798&r2=1.2799 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/expr/static_cast1.C.diff?cvsroot=gcc&r1=NONE&r2=1.1 Fixed in GCC 3.3.1, GCC 3.4. Subject: Bug 10931 CVSROOT: /cvs/gcc Module name: gcc Branch: gcc-3_3-branch Changes by: mmitchel@gcc.gnu.org 2003-06-26 01:05:47 Modified files: gcc/cp : ChangeLog call.c cp-tree.h search.c typeck.c gcc/testsuite : ChangeLog gcc/testsuite/g++.dg/abi: mangle4.C gcc/testsuite/g++.dg/lookup: scoped1.C gcc/testsuite/g++.old-deja/g++.jason: access8.C gcc/testsuite/g++.old-deja/g++.martin: pmf1.C Added files: gcc/testsuite/g++.dg/expr: static_cast1.C gcc/testsuite/g++.dg/rtti: dyncast1.C Log message: PR c++/10990 * search.c (lookup_base_r): Rely on accessible_p, rather than trying to emulate that logic here. PR c++/10931 * call.c (convert_like): Pass issue_conversion_warnings. (convert_like_with_context): Likewise. (convert_like_real): Add issue_conversion_warnings parameter. (perform_direct_initialization_if_possible): New function. * cp-tree.h (perform_direct_initialization_if_possible): Declare it. * typeck.c (check_for_casting_away_constness): New function. (build_static_cast): Rewrite. PR c++/10990 * g++.dg/rtti/dyncast1.C: New test. * g++.dg/abi/mangle4.C: Correct base-specifier access. * g++.dg/lookup/scoped1.C: Remove XFAIL. * g++.old-deja/g++.martin/pmf1.C: Correct base-specifier access. PR c++/10931 * g++.dg/expr/static_cast1.C: New test. Patches: http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.3076.2.163&r2=1.3076.2.164 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/call.c.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.341.2.23&r2=1.341.2.24 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/cp-tree.h.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.776.2.23&r2=1.776.2.24 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/search.c.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.243.2.10&r2=1.243.2.11 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/typeck.c.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.436.2.8&r2=1.436.2.9 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.2261.2.202&r2=1.2261.2.203 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/abi/mangle4.C.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.1&r2=1.1.46.1 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/expr/static_cast1.C.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=NONE&r2=1.1.2.1 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/lookup/scoped1.C.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.1&r2=1.1.28.1 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/rtti/dyncast1.C.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=NONE&r2=1.1.2.1 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.old-deja/g++.jason/access8.C.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.4&r2=1.4.20.1 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.old-deja/g++.martin/pmf1.C.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.1&r2=1.1.64.1 |
This code: void foo(int x) { static_cast<const unsigned int&>(x); } gets: test.cpp:3: error: non-lvalue in unary `&' AFAICT there is nothing wrong with it, and it compiles fine before 3.3. (And what unary `&'?)