Re: [PATCH] proc: pagemap: Hold mmap_sem during page walk

[Matt Mackall]

On Thu, 2010-04-01 at 15:34 +0900, KAMEZAWA Hiroyuki wrote:
> On Thu, 1 Apr 2010 15:09:56 +0900
> KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> wrote:
> 
> >
> > +		pm.pos = 0;
> > +		start_vaddr += PAGEMAP_WALK_SIZE;
> > +		end = start_vaddr + PAGEMAP_WALK_SIZE;
> 
> Sigh...this is bad..
> 
> ==
> From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
> 
> In initial design, walk_page_range() was designed just for walking page table and
> it didn't require mmap_sem. Now, find_vma() etc.. are used in walk_page_range()
> and we need mmap_sem around it.

This looks pretty reasonable. However, it also looks very similar to my
first version of pagemap (which started with double-buffering). It's
going to need re-testing to make sure it hasn't reintroduced any
wrapping, alignment, or off-by-one bugs that have already been ironed
out once or twice.

-- 
http://selenic.com : development and support for Mercurial and Linux


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/