I rebase the pnfs tree onto v3.1-rc10 AND
merged Bruce's for-3.2 branch that includes major changes to the
stateid bookkeeping infrastructure.
Layout stateids now uses the new idr (radix tree) based mechanism
and therefore it's cleaner and faster.
DS stateids are still tracked in a global hash table (as they are
generated by the MDS)
I also seized the opportunity to clean up the pnfs patch series.
white space, use of inline, etc. no functional changes otherwise.
Tested to pass the connectathon tests over pnfsd-lexp.
Benny
On 2011-10-18 17:50, J. Bruce Fields wrote:
> On Tue, Oct 18, 2011 at 05:30:36PM -0700, Benny Halevy wrote:
>> DS stateids are still tracked in a global hash table (as they are
>> generated by the MDS)
>
> By DS stateid's you're talking about stateid's that a DS receives on IO?
yes.
>
> I hadn't thought about those....
>
> Does the DS need to do anything with them other than look them up and
> make a binary OK vs. BAD_STATEID decision?
>
Yes, basically.
Eventually, this needs to work with the infamous back-end protocol
to enforce client fencing.
Benny
>> I also seized the opportunity to clean up the pnfs patch series.
>> white space, use of inline, etc. no functional changes otherwise.
>>
>> Tested to pass the connectathon tests over pnfsd-lexp.
>
> Thanks!
>
> --b.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
On Tue, Oct 18, 2011 at 05:30:36PM -0700, Benny Halevy wrote:
> DS stateids are still tracked in a global hash table (as they are
> generated by the MDS)
By DS stateid's you're talking about stateid's that a DS receives on IO?
I hadn't thought about those....
Does the DS need to do anything with them other than look them up and
make a binary OK vs. BAD_STATEID decision?
> I also seized the opportunity to clean up the pnfs patch series.
> white space, use of inline, etc. no functional changes otherwise.
>
> Tested to pass the connectathon tests over pnfsd-lexp.
Thanks!
--b.