Re: [RFC v4 03/15] vfs,hot_track: add the function for collecting I/O frequency

[Dave Chinner <david@fromorbit.com>]

On Sun, Oct 28, 2012 at 09:51:48PM +0800, Zhi Yong Wu wrote:
> On Sun, Oct 28, 2012 at 3:55 PM, Zheng Liu <gnehzuil.liu@gmail.com> wrote:
> > Hi Zhiyong,
> >
> > On Thu, Oct 25, 2012 at 11:08:55PM +0800, zwu.kernel@gmail.com wrote:
> > [snip]
> >> @@ -199,6 +342,54 @@ err:
> >>  }
> >>
> >>  /*
> >> + * Main function to update access frequency from read/writepage(s) hooks
> >> + */
> >> +inline void hot_update_freqs(struct inode *inode, u64 start,
> >> +                     u64 len, int rw)
> >
> > This function seems too big.  So we really need to inline this function?
> As Dave said in his comments, it will add a function call
> overhead even when tracking is not enabled. a static inline function
> will just result in no extra overhead other than the if
> statement....

I don't think I said that with respect to this code. I think I said
it w.r.t. a define or a small wrapper that decides to call
hot_update_freqs().  A static inline fucntion should only be a
couple of lines of code at most.

A static function, OTOH, can be inlined by the compiler if the
compiler thinks that is a win. But....

.....

> >> +EXPORT_SYMBOL_GPL(hot_update_freqs);

... it's an exported function, so it can't be inline or static, so
using "inline" is wrong whatever way you look at it. ;)

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com