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

Re: [Cluster-devel] [PATCH 001/008] configure makefile cleanup



Robert Peterson wrote:
> Fabio M. Di Nitto wrote:
>> gfs and gfs-kernel do not build anylonger from HEAD. gfs-kernel requires
>> dlm that's not built and gfs requires iddev that's not part of this branch
>> anymore.
>>
>> dlm needs to build before group. The latter requires to link with dlm.
>>
>> enter group in distclean and add a distclean target for group/Makefile.
>>
>> don't configure cmirror since we do not build it (Or perhaps fix the Makefile?).
>>
>> gfs2 is on the way to be merged upstream. gnbd should probably do the same and
>> already part of the ubuntu kernel tree. So stop building it from here.
>>
>> fix clean/make target in gfs-kernel since gulm is long way gone.
> Hi Fabio,
> 
> NACK.  Actually, I've been working to get gfs and gfs-kernel to build 
> properly
> with the upstream kernel and new infrastructure changes. 
> I've got a gfs-kernel that compiles now (but needs other things, regarding
> the locking infrastructure), and I'm working on getting the gfs tools to 
> use
> udev's libvolume_id rather than iddev, as we did for the gfs2 userspace 
> tools.

Ok perfect! I wasn't aware of general plan and directions (or I might have just
missed them) so i did prefer to raise the issue.

> So hopefully in the next few business days, I'll have these compiling 
> properly
> and committed to CVS.  Until then, I apologize for the compile problems.

It's not a problem. In Ubuntu we are dropping gfs for the next release to focus
on gfs2.
My main concern was for CVS users that builds from the top level dir.

> You're right that the gfs2-kernel stuff should not be compiled out of CVS,
> because that's all in a public git tree now, and that's where people 
> should get it.
> I'm not sure about the plans for gnbd at the moment, so I'll defer to 
> someone else.

Agreed 100%.

Thanks!
Fabio

-- 
I'm going to make him an offer he can't refuse.


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