Re: [boost] gcc: timer patch, and ios_base workarounds

From: Beman Dawes (bdawes_at_[hidden])
Date: 2001-04-16 07:50:27

At 10:16 PM 4/15/2001, Greg Chicares wrote:

 >> What happens when gcc 3.0 ships?  Will we still have these variants?
 >
 >Yes we will. Both are actively maintained gcc subtargets for windows.
 >Cygwin uses a POSIX emulation layer that forces all programs linked
 >with it to be GPL, while mingw has no special POSIX support but is in
 >the public domain. Details at http://www.mingw.org/x86-win32-ports.shtml 
.
 >
 >I would be glad to perform whatever boost testing I can with the mingw
 >variant if you're interested in publishing the results. The outcome
 >really can be different with mingw vs. cygwin: e.g. they use different
 >C runtime libraries, neither of which is GNU glibc.

Don't they also use somewhat different snapshots for their builds?  For 
example, the Cygwin 2.95.2 I'm using is something like the 6th release of 
2.95.2.  I suppose that can cause differences in outcome.

Ideally, I should just add MinGW to the regular set regression test 
compilers, rather than Boost adding a separate table for one compiler.

Do you have any idea if MinGW and Cygwin coexist peacefully on the same 
machine?

Is the MinGw installation likely to be trouble free?

Does running the tests on MinGW result in any useful information?

--Beman