Re: [gccsdk] Building the libraries

On Sun, Nov 18, 2012 at 02:59:57PM +0100, John Tytgat wrote:
> > At first I suspected 'env/ro-make' but it turned out that wasn't
> > involved (yet). The error was actually caused by line 16 of
> > 'autobuilder/libraries/gtk/libglib2.0-0/setvars' which reads:
> > 'make -j$AB_CPUS install' (a configure/make sequence in a setvars file?
> > really?).
> > At that point the variable AB_CPUS contains X (for 'guess number of
> > CPUs') so that will never work. I (temporarily) changed that line to
> > 'make -j$NUMCPUS install' and libglib2.0-0 was eventually built
> > successfully.
>
> Theo recently added automatic -j option adding to the MAKEFLAGS. I don't
> think a 'setvars' file should explicitely add it anymore.
>
> Theo, can you confirm tis ?

The reason for the make sequence in the setvars file is that libglib depends
on having a native (ie x86) copy of some of the build tools that it uses to
build itself. It's quite distro-specific as to whether they can just be
installed as packages, so my solution was to build libglib twice - once
natively, to get the tools, and then again using the cross-compiler. The
make in setvars is the first native build.

Looks like my change to CPU guessing post-dated this change in
libglib/setvars. It's not vital to add it either way, but does speed up the
compile. I'm not sure if MAKEFLAGS is automatically picked up on a native
build, though.

Theo

_______________________________________________
GCCSDK mailing list gcc@gccsdk.riscos.info
Bugzilla: http://www.riscos.info/bugzilla/index.cgi
List Info: http://www.riscos.info/mailman/listinfo/gcc
Main Page: http://www.riscos.info/index.php/GCCSDK