Re: [patch 1/7] cpusets: add dirty map to struct address_space

[David Rientjes <rientjes@google.com>]

On Tue, 28 Oct 2008, Peter Zijlstra wrote:

> > This patch implements the management of dirty node maps for an address
> > space through the following functions:
> > 
> > cpuset_clear_dirty_nodes(mapping)	Clear the map of dirty nodes
> > 
> > cpuset_update_nodes(mapping, page)	Record a node in the dirty nodes
> > 					map
> > 
> > cpuset_init_dirty_nodes(mapping)	Initialization of the map
> > 
> > 
> > The dirty map may be stored either directly in the mapping (for NUMA
> > systems with less then BITS_PER_LONG nodes) or separately allocated for
> > systems with a large number of nodes (f.e. ia64 with 1024 nodes).
> > 
> > Updating the dirty map may involve allocating it first for large
> > configurations.  Therefore, we protect the allocation and setting of a
> > node in the map through the tree_lock.  The tree_lock is already taken
> > when a page is dirtied so there is no additional locking overhead if we
> > insert the updating of the nodemask there.
> 
> I find this usage of tree lock most bothersome, as my concurrent
> pagecache patches take the lock out. In which case this _does_ cause
> extra locking overhead.
> 

Yeah, if we don't serialize with tree_lock then we'll need to protect the 
attachment of mapping->dirty_nodes with a new spinlock in struct 
address_space (and only for configs where MAX_NUMNODES > BITS_PER_LONG). 
That locking overhead is negligible when mapping->dirty_nodes is non-NULL 
since there's no requirement to protect the setting of the node in the 
nodemask.

Are your concurrent pagecache patches in the latest mmotm?  If so, I can 
rebase this entire patchset off that.