[openfabrics-ewg] Rollup patch for ipath and OFED

[Woodruff, Robert J]

Michael wrote, 
Quoting r. Woodruff, Robert J <robert.j.woodruff at intel.com>:
>> Bottom line is we either need to keep the code in the trunk up to
date
>> or remove it, but having multiple data bases with different versions
>> is somewhat confusing. 

>Since kernel level code is in kernel.org git trees,
>as long as we do keep kernel code in subversion this automatically
>implies multiple different databases. No?

>-- 
>MST

Yes, there are multiple databases and this is confusing. However,
I don't think there are any backport patches and such in the kernel.org
trees,
so we need some database somewhere that has the latest version of code
under development
that people can test against, otherwise it seems likely that people
would be developing and testing against different branches and when
time comes to integrate it all for the next kernel release, there could
be
issues that arise. 

In the past, we always kept the latest development version of the code
in SVN
and could easily build it to test. I have said this in the past but will
say it again, if people want to use git instead of SVN for the kernel
code,
then that is fine, but if that is to be the case, then we should have
one git tree somewhere that has all the latest code that we can pull
from
and then remove the kernel code from SVN to avoid confusion.

woody