Re: [PATCH] vfs: Add a trace point in the mark_inode_dirty function

[Arjan van de Ven <arjan@infradead.org>]

On Sun, 25 Oct 2009 23:03:21 -0700
Andrew Morton <akpm@linux-foundation.org> wrote:

> > ...
> >
> > @@ -1071,6 +1072,8 @@ void __mark_inode_dirty(struct inode *inode,
> > int flags) if ((inode->i_state & flags) == flags)
> >  		return;
> >  
> > +	trace_dirty_inode(inode, current);
> > +
> >  	if (unlikely(block_dump))
> >  		block_dump___mark_inode_dirty(inode);
> >  
> 
> Doesn't powertop also want to know who is spinning up the disk via
> buffered reads, direct-io reads and direct-io writes?
> 
> That's why the block_dump hook in submit_bio() is there.

there is already an existing trace event for this in the block layer...
I was planning to use block_bio_queue for this.

-- 
Arjan van de Ven 	Intel Open Source Technology Centre
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org