Re: testing newer toolchain components

[Dan Kegel <dank at kegel dot com>]

Robert P. J. Day wrote:
>   even though my sh3 build seems to work, if i wanted to test newer
> toolchain components and i have some machine cycles to spare, is there
> anything untoward about just bumping up a component version number and
> letting it build?
>
>   for example, sh3 has the combination:
>
> 	gcc-3.4.1
> 	glibc-2.3.3
> 	binutils-2.15
>
> just for fun, i could bump gcc up to 3.4.2 or 3.4.3 and see what
> happens?  is there a rigorous recipe for doing this kind of testing,
> or does one just crank up the version number(s) and let it rip to see
> what happens?

Just crank it up and see what breaks.  Let's say you move to gcc-3.4.3.
The first thing that might happen is it might not build;
that could either be a new problem, or an old one that was
worked around by a patch in crosstool's patches/gcc-3.4.1 directory.

The second thing that might happen is it might not produce fully
working binaries.  Again, that might be caused by a new problem,
or by an old one that was worked around by a patch in crosstool's patches/gcc-3.4.1 directory.

You should run the gcc and glibc regression test suites, or whatever
test suites you have handy, before and after the upgrade when you experiment like this,
since a working toolchain is by no means assured.
- Dan

--
Trying to get a job as a c++ developer?  See http://kegel.com/academy/getting-hired.html

------
Want more information?  See the CrossGCC FAQ, http://www.objsw.com/CrossGCC/
Want to unsubscribe? Send a note to crossgcc-unsubscribe@sources.redhat.com