2013/7/31 Earnie Boyd <[email protected]>: > On Wed, Jul 31, 2013 at 3:47 PM, Kai Tietz <[email protected]> wrote: >> 2013/7/31 Baruch Burstein <[email protected]>: >>> On Wed, Jul 31, 2013 at 9:49 PM, Kai Tietz <[email protected]> wrote: >>>> >>>> 2013/7/31 Earnie Boyd <[email protected]>: >>>> > On Wed, Jul 31, 2013 at 8:21 AM, Kai Tietz wrote: >>>> >> >>>> >> As there never was, and won't be in next future a target with ending >>>> >> ...mingw64 in triplet. >>>> >> >>>> > >>>> > That statement maybe here but not at MinGW.org. The x86_64-pc-mingw64 >>>> > triplet is contained in the config master repository. >>>> >>>> That you added this triplet to this file doesn't make it right. Just >>>> mingw.org (as venture) want to support that triplet at some day, but >>>> no upstream toolchain-part (gcc, binutils, gdb, and dependent >>>> libraries) supports it. >>> >>> >>> I understood from JonY's answer that these triplets are arbitrary and don't >>> really mean anything other then helping keep cross-compilers organized. So >>> what do you mean that no upstream toolchain-part (gcc, binutils, gdb, and >>> dependent libraries) supports it? >> >> What I am referring to is the 'x86_64-pc-mingw64' triplet used in this >> config.guess file. Actual there is no support for the mingw64 triplet >> part. Neither in binutils, nor in gcc. So this 64 is bogus and >> should be replaced by 32. > > That doesn't mean they cannot in the future and one must start with > the config master. This also supports MSYSTEM=MINGW64 and > config.guess getting the build system correct for x86_64-pc-mingw64 > with a 64 bit MSYS.
So, it doesn't exist. But there is already something existing, and I don't see why your future plans should be here, if something already covers it. >> Longterm this mingw32 and mingw64 should be replaced simply by mingw. >> this 32/64 is just bogus and is already covered by the architecture >> part of triplet (i?86/x86_64). > > See the discussion beginning with > https://sourceforge.net/p/mingw/mailman/message/31095908/ Sorry, this discussion was held already upstream on gcc in 2005-2006. And honestly I am not that much interested in what you decide on your mailing-list. Also it is of not much interest that you created some private branches of binutils/gcc etc. So you can keep your changes to config.guess private, too. > Synopsis is we want to retain mingw32 and mingw64 even though I wanted > to drop 32 and 64. > > Secondarily I would like to see the default of i386 be changed to i686 > in config.sub. Well, on that I agree. i686 should be suitable for modern systems. If somebody requires configure for an older architecture, then OP should explicit specify triplet. > -- > Earnie > -- https://sites.google.com/site/earnieboyd Kai ------------------------------------------------------------------------------ Get your SQL database under version control now! Version control is standard for application code, but databases havent caught up. So what steps can you take to put your SQL databases under version control? Why should you start doing it? Read more to find out. http://pubads.g.doubleclick.net/gampad/clk?id=49501711&iu=/4140/ostg.clktrk _______________________________________________ Mingw-w64-public mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mingw-w64-public
