On Tue, Jun 15, 2010 at 07:10:26AM -0400, Christoph Hellwig wrote:

On Tue, Jun 15, 2010 at 08:55:38PM +1000, Nick Piggin wrote:

quoted

What I do in fsblock is to maintain a block-nr sorted tree of dirty
blocks. This works nicely because fsblock dirty state is properly
synchronized with page dirty state. So writeout can just walk this in
order and it provides pretty optimal submission pattern of any
interleavings of data and metadata. No need for buffer boundary or
hacks like that. (needs some intelligence for delalloc, though).

I think worrying about indirect blocks really doesn't matter much
these days.  For one thing extent based filesystems have a lot less
of these, and second for a journaling filesystem we only need to log
modification to the indirect blocks and not actually write them back
in place during the sync.  At least for XFS the actual writeback can
happen a lot later, as part of the ordered list of delwri buffers.

That's true, more importantly I meant any interleavings of data from
more than one file too.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>