[pnfs] 2.6.17 pnfs as client module

[Marc Eshel]

The reference to inode_lock was added by me with the recall layout patch. 
The reason I did not see the error that Dean got is because I was running 
on a UP where spin_lock is a no-op. I am not sure if exporting inode_lock 
is the right thing to do, maybe in the future we will have all the layouts 
on some other chain (cache) so we will not have to scan the all the 
inodes.
Marc. 

pnfs-bounces at linux-nfs.org wrote on 07/28/2006 12:21:40 PM:

> yes, i've got a patch to send into trond - it's a generic nfs issue, the 

> inode_lock is grabbed in fs/nfs/delegation.c. i'll check in the fix
> 
> -->Andy
> 
> > Has anyone tried compiling nfs on the client as a module?  I'm getting 

> > the following error.  I assume it is in the base and not part of our 
> > pNFS changes.
> > ...
> > Root device is (254, 0)
> > Boot sector 512 bytes.
> > Setup is 6904 bytes.
> > System is 1725 kB
> > Kernel: arch/i386/boot/bzImage is ready  (#1)
> >   Building modules, stage 2.
> >   MODPOST
> > WARNING: "inode_lock" [fs/nfs/nfs.ko] undefined!
> >   CC      fs/nfs/nfs.mod.o
> >   CC      fs/nfs/nfslayoutdriver.mod.o
> >   LD [M]  fs/nfs/nfs.ko
> >   LD [M]  fs/nfs/nfslayoutdriver.ko
> > 
> > 
> > -- 
> > Dean Hildebrand
> > Ph.D. Candidate
> > University of Michigan
> > 
> > _______________________________________________
> > pNFS mailing list
> > pNFS at linux-nfs.org
> > http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs
> 
> 
> _______________________________________________
> pNFS mailing list
> pNFS at linux-nfs.org
> http://linux-nfs.org/cgi-bin/mailman/listinfo/pnfs