Date: Wed, 31 Jan 2024 09:41:01 +0000 From: bugzilla-noreply@freebsd.org To: toolchain@FreeBSD.org Subject: [Bug 276738] clang: static_assert conflicts with -std=c++98 -pedantic-errors Message-ID: <bug-276738-29464-lFWCVE41oM@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-276738-29464@https.bugs.freebsd.org/bugzilla/> References: <bug-276738-29464@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D276738 Warner Losh <imp@FreeBSD.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |imp@FreeBSD.org --- Comment #4 from Warner Losh <imp@FreeBSD.org> --- For C, there's certain features that we try to make available in a compatib= le way to old standards (eg __restrict to compile it away). I think the logic here's not right for old C++ standards. So the trouble is that llvm's /usr/include/c++/v1/__config # define static_assert(...) _Static_assert(__VA_ARGS__) when compiling for < C++11 (at least that's my reading of it, it's a twisty maze so it may be in a branch not taken. That's the only place I could see = any redirection that would be affected by the #undef in the example. sys/cdefs.h defines _Static_assert in an old-school C way for C++ < C++11, = so this feature is visible, but badly implemented for C++. I suspect that both from reading this code and the #undef in the example me= ans we should simply not define _Static_assert for C++ at all. The compiler will either implement it or not, depending on its wishes and give an error if no= t. assert.h already tries to do the right thing by not defining it for C++ at = all. --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-276738-29464-lFWCVE41oM>