This is the mail archive of the
cygwin
mailing list for the Cygwin project.
Re: [ANNOUNCEMENT] Updated: mingw64-{i686,x86_64}-gcc-5.4.0-3 (x86/x86_64)
- From: Tony Kelman <tony at kelman dot net>
- To: "cygwin at cygwin dot com" <cygwin at cygwin dot com>
- Date: Fri, 11 Nov 2016 14:06:31 +0000
- Subject: Re: [ANNOUNCEMENT] Updated: mingw64-{i686,x86_64}-gcc-5.4.0-3 (x86/x86_64)
- Authentication-results: sourceware.org; auth=none
- Authentication-results: cygwin.com; dkim=none (message not signed) header.d=none;cygwin.com; dmarc=none action=none header.from=kelman.net;
- References: <announce.5825AD20.7030804@gmail.com>
- Spamdiagnosticmetadata: NSPM
- Spamdiagnosticoutput: 1:99
> I have just updated the mingw-w64 cross compilers Cygwin 32bit and
> 64bit to mingw64-{i686,x86_64}-gcc-5.4.0-3.
Hi Jon,
What will be left as prev? 5.4.0-2, or the last 4.9 build? Can you
help raise https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77333? The
interprocedural analysis that was added for gcc 5 is deeply buggy on
mingw targets, resulting in miscompiling C++ libraries like LLVM.
This is keeping me on gcc 4 indefinitely until it gets fixed. I may
have to abandon gcc and switch to clang due to this, but there are
many libraries that have broken build systems that behave incorrectly
when you try to use clang on Windows.
I'll use time machine if that's my only way to keep getting gcc 4,
but this bug in gcc should really be fixed or it's going to start
showing up in cygwin-mingw-cross and msys2 packages.
Thanks,
Tony
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple