lmi
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lmi] Problems with the Cygwin tests


From: Vadim Zeitlin
Subject: Re: [lmi] Problems with the Cygwin tests
Date: Fri, 10 May 2019 02:12:53 +0200

On Thu, 9 May 2019 21:47:08 +0000 Greg Chicares <address@hidden> wrote:

GC> On 2019-05-08 18:11, Greg Chicares wrote:
GC> > On 2019-05-07 15:05, Vadim Zeitlin wrote:
GC> [...]
GC> >> 5. The directory /opt/lmi/mingw/bin had to be added to the PATH for
GC> >>    xsltproc.
GC> > 
GC> > Fixed.
GC> 
GC> Looking back over this flurry of recent changes, I wonder
GC> why this one was necessary.

 Thanks for rechecking this, I actually wanted to do this myself, but
forgot about it. After doing it now, it turns out that this is not the best
way to solve the problem, we should add /opt/lmi/bin to the PATH instead.

GC> I find no 'xsltproc.exe' in the compiler tarball, so I guess the
GC> 'xsltproc.exe' we've built must depend on some compiler runtime file.

 Right.

GC> But why would that be? IIRC, libxslt is C, not C++, so it can't be
GC> libstdc++. So is the libgcc DLL the reason?

 Yes, it's libgcc_s_sjlj-1.dll and AFAIK there is nothing wrong with it
depending on it, but this DLL is already copied to /opt/lmi/bin, so we
should use it from there and not from compiler directory.

 Regards,
VZ


reply via email to

[Prev in Thread] Current Thread [Next in Thread]