[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#613634: [PATCH] vmalloc: eagerly clear ptes on vunmap



On Mon, 2011-05-02 at 11:57 +0100, Ian Campbell wrote:
> On Sun, 2011-05-01 at 16:51 +0100, Ben Hutchings wrote:
> > On Sun, 2011-05-01 at 12:33 +0100, Ian Campbell wrote:
> > > On Sat, 2011-04-30 at 17:34 +0100, Ben Hutchings wrote:
> > > > Ian, should we apply the backport from
> > > > 64141da587241301ce8638cc945f8b67853156ec to squeeze for bug #613634 et
> > > > al?
> > > 
> > > It's already in SVN for 2.6.32-34, seems to be in the changelog twice
> > > even, once closing #614400 and then again closing #613634, which are
> > > merged.
> > 
> > Oh right, it's applied to the xen-featureset kernels.
> 
> So it is, I'd forgotten that.
> 
> > But isn't it also
> > applicable to non-xen-featureset kernels which can run in domU?
> 
> Yes. I initially confused it with the vmalloc_sync_all fixes which
> really only impact domains running backend drivers (since that's the
> only real user of that code path). The vunmap thing is indeed more
> generically useful to fix so I think it should go into the generic
> flavours too.
> 
> Want me to take a look at that?

I'll check this in to svn shortly for 2.6.32-35. Unfortunately I missed
the 2.6.32-34 boat.

Ian.

-- 
Ian Campbell
Current Noise: Gojira - Ocean Planet

Isn't air travel wonderful?  Breakfast in London, dinner in New York,
luggage in Brazil.




Reply to: