Re: [patch 03/13] io-accounting: write accounting

[Andrew Morton <akpm@osdl.org>]

On Wed, 13 Dec 2006 00:45:50 -0800
Suleiman Souhlal <ssouhlal@FreeBSD.org> wrote:

> akpm@osdl.org wrote:
> > From: Andrew Morton <akpm@osdl.org>
> > 
> > Accounting writes is fairly simple: whenever a process flips a page from clean
> > to dirty, we accuse it of having caused a write to underlying storage of
> > PAGE_CACHE_SIZE bytes.
> 
> On architectures where dirtying a page doesn't cause a page fault (like i386), couldn't you end up billing the wrong process (in fact, I think that even on other archituctures set_page_dirty() doesn't get called immediately in the page fault handler)? 

Yes, that would be a problem in 2.6.18 and earlier.

In 2.6.19 and later, we do take a fault when transitioning a page from
pte-clean to pte-dirty.  That was done to get the dirty-page accounting
right - to avoid the all-of-memory-is-dirty-but-the-kernel-doesn't-know-it
problem.


> AFAICS, set_page_dirty() is mostly called when trying to unmap a page when trying to shrink LRU lists, and there is no guarantee that this happens under the process that dirtied it (in fact, the set_page_dirty() is often done by kswapd).

hm, that code is still there in zap_pte_range().  If all is well, that
set_page_dirty() call should never return true.  Peter did, you ever test
for that?

(Well, it might return true in rare races, because zap_pte_range() doesn't
lock the pages)