[Mingw-w64-public] CRT version of official builds

Hi all, I did notice that in recent official builds (aka mingw-builds), while the buid-info.txt shows that they are configured with "--rt-version=v4" option, the actual version of CRT is 5 according to the definitions of __MINGW64_* macros in _mingw_mac.h file: #define __MINGW64_VERSION_MAJOR 5 #

Re: [Mingw-w64-public] Bug in math.h: mathematical constants undefined if __STRICT_ANSI__ and _POSIX_C_SOURCE or _POSIX_SOURCE is defined

On 9/10/2015 22:42, Thomas Uhle wrote: > If compiling with 'gcc -std=c++11' for instance, gcc defines > __STRICT_ANSI__ and, thus, the mathematical constants in math.h that are > also defined in POSIX.1 (see > http://pubs.opengroup.org/onlinepubs/9699919799/basedefs/math.h.html for > reference) are

Re: [Mingw-w64-public] segfaults after longjmp on 64bit Windows w/SJLJ exceptions

2015-09-10 23:16 GMT+02:00 Kevin Ushey : > Hi everyone, > > I'm part of a team attempting to update the toolchain used to compile and > build R and R packages for Windows, and we've been having some issues. > > Previously, we used a toolchain built with gcc 4.6.3 (SJLJ exceptions, win32 > threads)

Re: [Mingw-w64-public] gcc 5.2.1 libgomp testsuite failures with latest runtime and winpthread

-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 Ping! Am 29.08.2015 um 23:00 schrieb Kai Tietz: > > Am 09.08.2015 19:17 schrieb "Mattias Engdegård" >: >> >> 7 aug 2015 kl. 14.50 skrev Rainer Emrich >: >> >> > This fixes the issue, lib

Re: [Mingw-w64-public] gcc 5.2.1 libgomp testsuite failures with latest runtime and winpthread

Hi, I committed code to tolerate sloppy code to our master. Kai 2015-09-11 10:19 GMT+02:00 Adrien Nader : > On Fri, Sep 11, 2015, Michel Zou wrote: >> >> You can ping the gcc dudes instead ! >> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67141 > > Considering the issue, I doubt that only libgo

Re: [Mingw-w64-public] gcc 5.2.1 libgomp testsuite failures with latest runtime and winpthread

On Fri, Sep 11, 2015, Michel Zou wrote: > > You can ping the gcc dudes instead ! > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67141 Considering the issue, I doubt that only libgomp has the sloppy coding. :) -- Adrien Nader

Re: [Mingw-w64-public] gcc 5.2.1 libgomp testsuite failures with latest runtime and winpthread

You can ping the gcc dudes instead ! https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67141 > Date: Fri, 11 Sep 2015 09:54:16 +0200 > From: sf.rai...@emrich-ebersheim.de > To: mingw-w64-public@lists.sourceforge.net > Subject: Re: [Mingw-w64-public] gcc 5.2.1 libgomp testsuite failures with > latest r

Re: [Mingw-w64-public] gcc 5.2.1 libgomp testsuite failures with latest runtime and winpthread

Ping! Am 29.08.2015 um 23:00 schrieb Kai Tietz: > > Am 09.08.2015 19:17 schrieb "Mattias Engdegård" >: >> >> 7 aug 2015 kl. 14.50 skrev Rainer Emrich >: >> >> > This fixes the issue, libgom.sum attached. Can we have this in head and in