Re: Nick's core remove PageReserved broke vmware...

[Gleb Natapov <gleb@minantech.com>]

On Mon, Nov 14, 2005 at 04:57:08PM +0200, Michael S. Tsirkin wrote:
> Quoting r. Hugh Dickins <hugh@veritas.com>:
> > Subject: Re: Nick's core remove PageReserved broke vmware...
> > 
> > On Mon, 14 Nov 2005, Michael S. Tsirkin wrote:
> > > Quoting Gleb Natapov <gleb@minantech.com>:
> > > > On Mon, Nov 14, 2005 at 02:25:35PM +0200, Michael S. Tsirkin wrote:
> > > > > 
> > > > > There's one thing that I have thought about: what happens
> > > > > if I set DONTFORK on a page which already has COW set
> > > > > (e.g. after fork)?
> > > > > 
> > > > > It seems that the right thing would be to force a page copy -
> > > > > otherwise the page can get copied on write.
> >
> > > Should we worry about this?
> > 
> > About what?
> 
> For pages which hardware will only read, not write,
> hardware driver does get_user_pages with write cleared.
> 
It looks like openib always pass 1 as write. I think this is
exactly for this reason.

> This means that COW may remain set.
> 
> -- 
> MST

--
			Gleb.