Re: [git patches] xfs and block fixes for virtually indexed arches

[James Bottomley <James.Bottomley@suse.de>]

On Fri, 2009-12-18 at 19:24 +0900, FUJITA Tomonori wrote:
> On Fri, 18 Dec 2009 11:01:29 +0100
> James Bottomley <James.Bottomley@suse.de> wrote:
> 
> > > Yeah, but now only XFS passes vmap'ed pages to the block layer. Isn't
> > > it better to wait until we have real users of the API?
> > 
> > XFS is a real user ... the XFS filesystem is our most trusted code base
> > that can break the 8TB limit, which hard disks are already at.  Ext4 may
> > be ready, but it's not universally present in enterprise distros like
> > XFS.
> 
> XFS already has the own code to handle that, which works fine (with
> your patchset except for 5/6 for the block layer). Not much motivation
> for XFS to move to the generic API?

Right, but it's for completeness.  If we decide to allow vmap buffers,
then only supporting them on certain paths is a recipe for confusion in
a year's time when someone assumes we support vmap buffers on all block
paths; a bit like the current confusion over what we support ....

> > > > That would ensure the architecturally
> > > > correct flushing of the aliases, and would satisfy the expectations of
> > > > blk_rq_map_kern().  The down side is that vmap/vmalloc set up and clear
> > > > page tables, which isn't necessary and might impact performance (xfs
> > > > people?)
> > > 
> > > btw, I'm not sure that the existing blk_rq_map_* API isn't fit well to
> > > file systems since blk_rq_map_user and blk_rq_map_kern takes a request
> > > structure.
> > 
> > OK, so that was illustrative.  The meat of the change is at the bio
> > layer anyway (fss tend to speak bios).
> 
> Yeah, I think so, it's up to Jens to add new APIs for vmap there.

Agreed.

James