Recommended GCC version?

[Andrew Haley]

David Holmes - Sun Microsystems wrote:

> Andrew John Hughes said the following on 06/20/08 09:35:
>> I would hope one of the side effects of moving the JDK from a
>> proprietary to a community-based Free Software model would be that it
>> gets built, run and tested on a much wider range of platforms and
>> compilers.  This has already started to happen.
>
> Agreed - this is a good end goal. Meanwhile there are some
> practicalities to address.
>
>> The reality is that people aren't going to download and build a
>> specific copy of GCC just for OpenJDK, and distros will certainly want
>> it to build with the GCC they use for everything else.
>
> But are the Distros expecting/assuming that everything will work fine
> with their version of GCC?

Yes, of course, and if it doesn't then we'll either fix GCC or
OpenJDK.

> Who is expected to have done the testing?

There are two levels of testing here: that of the core package as
written, and that of a specific build, for which we need to do
regression testing.  There is nothing special about Java in this
regard: every package has the same problem.  The packagers of the
distros have been doing this stuff for many years.

> If something goes wrong who would be expected to fix it? Would the
> Distros patch the OpenJDK code with a workaround for their GCC
> version? Or would they grab a known working GCC version and rebuild
> using that?

We'll patch OpenJDK or GCC, depending on which is at fault.  We will
not build a special GCC in order to build OpenJDK; that would be a
pain and contrary to packaging guidelines.

> Right now the reality is that these alternate compiler versions have
> not undergone extensive testing for the OpenJDK.

I guess it all depends on what you mean by "extensive".  We've run all the
test suites to which we have access and a good deal of application testing
too.

> Over time that will hopefully change, but for now - caveat emptor!

Of course.  One of the problems here is that the distro packagers need
access to regression tests, some of which (AIUI) are still Sun
confidential.

Andrew.