This is the mail archive of the cygwin mailing list for the Cygwin project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [ANNOUNCEMENT] New package: mingw64-x86_64-gcc-4.5.1-1


On Sep 14 18:15, JonY wrote:
> On 9/14/2010 16:03, Corinna Vinschen wrote:
> >On Sep 14 15:30, JonY wrote:
> >>On 9/14/2010 15:29, Charles Wilson wrote:
> >>>I don't know about Andy, but I sure do -- and I can reproduce his
> >>>problem.  I suspect there is a "bug" in how the cross tool locates the
> >>>	/usr/x86_64-w64-mingw32/bin
> >>>directory, given the mount structure:
> >>>	/usr/bin = /bin
> >>>	/usr/lib = /lib
> >>>BUT
> >>>	/usr/x86_64-w64-mingw32 != /x86_64-w64-mingw32
> >>>
> >>>because if I do THIS:
> >>>mount -o bind /usr/x86_64-w64-mingw32 /x86_64-w64-mingw32
> >>>
> >>>then
> >>>   /bin/x86_64-w64-mingw32-gcc -o foo foo.c
> >>>works, just as if I had invoked
> >>>   x86_64-w64-mingw32-gcc -o foo foo.c
> >>>
> >>>I say this is a "bug" in quotes, because...well, I'm not sure it fits
> >>>the definition. It's *our* fault we use a wacky mount structure on cygwin...
> >>>
> >>>--
> >>>Chuck
> >>>
> >>
> >>So, if /usr/x86_64-w64-mingw32 actually exists, it works?
> >>
> >>This looks bad, nonetheless.
> >>
> >>Maybe we can fix cygwin by only redirecting known directories like,
> >>/usr/bin and /usr/lib to those in /.
> >
> >Cygwin doesn't redirect any /usr dirs to /.  There are default mount
> >points for /usr/bin ->  /bin and /usr/lib ->  lib.  That's all.  The
> >problematic path is generated in gcc itself.
> >
> >
> >Corinna
> >
> 
> What do you suggest the fix should be?

I really don't know, but it's certainly not a fix in Cygwin.  The fact
that /usr/bin is a mount point to /bin is nothing which wouldn't be 
allowed under Linux as well.  There's a default mechanism in gcc which
evaluates the paths to the tools, headers and libdb dirs by massaging
the $prefix and $exec_prefix values in some way.  I assume you can
work around this issue by using slightly different values, but I'm
not fluent enough with the path evaluationin gcc to suggest the
correct solution, for a given value of "correct".


Corinna

-- 
Corinna Vinschen                  Please, send mails regarding Cygwin to
Cygwin Project Co-Leader          cygwin AT cygwin DOT com
Red Hat

--
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


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]