Re: The pnfs-submit tree with renaming done.

Re: The pnfs-submit tree with renaming done.

[Benny Halevy]

Hi Andy, 

Thanks for the tarball you sent with the pnfs-submit patchset
with renamed types.

I merged into the tree and applied the renames onto the post submit patches
and obj/block layout drivers.

I added some renames for the callback code to match
the existing naming convention and did a bit of cleanup of the patchset
and the post submit series, including splicing the tail of the submit patchset
into the post submit branch head.

I revived the "pnfsd, pnfs: protocol level pnfs constants" preliminary patch
as it contains no-brainer nfsv4.1 definitions that Trond already agreed
can be submitted as a whole and not per use.
This simplifies rebasing of the pnfs-submit branch as pnfs-submit-all
and makes more sense overall IMO.

I also resurrected Bruce's stats code in the "remove this unused code"
patch, and added a question mark for this patch as I think it requires
more through review to agree exactly what should go away for good.
(Some of it is obvious and I should've probably put it aside...)

I released this clean version as pnfs-all-2.6.35-2010-08-24
ad I'm working on rebasing onto 2.6.36-rc1.

At this point, I'd like to close the 2.6.35 stream for new features
and release only fixes to it from now on.

Benny

Re: The pnfs-submit tree with renaming done.

[William A. (Andy) Adamson]

Thanks Benny - looks great :)

-->Andy

On Tue, Aug 24, 2010 at 10:42 AM, Benny Halevy <bhalevy@panasas.com> wrote:
> Hi Andy,
>
> Thanks for the tarball you sent with the pnfs-submit patchset
> with renamed types.
>
> I merged into the tree and applied the renames onto the post submit patches
> and obj/block layout drivers.
>
> I added some renames for the callback code to match
> the existing naming convention and did a bit of cleanup of the patchset
> and the post submit series, including splicing the tail of the submit patchset
> into the post submit branch head.
>
> I revived the "pnfsd, pnfs: protocol level pnfs constants" preliminary patch
> as it contains no-brainer nfsv4.1 definitions that Trond already agreed
> can be submitted as a whole and not per use.
> This simplifies rebasing of the pnfs-submit branch as pnfs-submit-all
> and makes more sense overall IMO.
>
> I also resurrected Bruce's stats code in the "remove this unused code"
> patch, and added a question mark for this patch as I think it requires
> more through review to agree exactly what should go away for good.
> (Some of it is obvious and I should've probably put it aside...)
>
> I released this clean version as pnfs-all-2.6.35-2010-08-24
> ad I'm working on rebasing onto 2.6.36-rc1.
>
> At this point, I'd like to close the 2.6.35 stream for new features
> and release only fixes to it from now on.
>
> Benny
>