GCC -mno-cygwin vs mingw32-gcc cross environment.

Paul Garceau pgarceau@qwest.net
Wed Apr 25 02:28:00 GMT 2001

On 24 Apr 2001, at 20:18, the Illustrious Christopher Faylor wrote:

> On Wed, Apr 25, 2001 at 09:43:59AM +1000, Robert Collins wrote:
> >> Given the above...when I say branch, I am saying create a cvs branch
> >> that adds i686-pc-mingw32-gcc as a subset of Cygwin when Cygwins
> >> -mno- cygwin switch is used...
> >
> >I don't see how that will help. CVS branchs allow for concurrent
> >development. I.E. If I ask chris nicely he may give me a new branch on
> >cvs.sources.redhat.com for me to really push the pthread code ahead -
> >and it would only be needed if there where 3 or 4 folk all working on
> >the same project but we didn't want to break CVS HEAD at all.
> >
> >Please explain a little more clearly for me :]
> And for me.  I don't know what CVS branches have to do with this
> discussion.

	Apparently they have absolutely nothing to do with
 anything re: this discussion.

	All I was trying to suggest was the addition of a new directory/folder
that is specific to -mno-cygwin stuff (i686-pc-mingw32-gcc).
	Nothing more since I fail to see any reason behind forcing cross-
compiling outside of the fact that it makes things more complicated. 

	In fact I really don't seem much reason behind cross-compiling to 
begin with, especially where mingw/-mno-cygwin is concerned.
	Mingw is designed to support win32api based OS, if you haven't got a 
win32api OS, then mingw is pretty much useless since you don't have any 
means to test your source code.
	You want to build mingw executables? Then get the mingw tool set and 
build your executables...why waste time making something already quite 
complex (Cygwin) even more so...?

	I am shutting up now...sorry for any confusion I may have created.

> cgf


		Paul G.

Nothing real can be threatened.
    Nothing unreal exists.

More information about the Cygwin-apps mailing list