Re: [PATCH v2] fs: select: fix information leak to userspace

[Vasiliy Kulikov <segoon@openwall.com>]

On Sun, Nov 14, 2010 at 18:06 -0800, Andrew Morton wrote:
> On Sun, 14 Nov 2010 12:25:33 +0300 Vasiliy Kulikov <segoon@openwall.com> wrote:
> 
> > On some architectures __kernel_suseconds_t is int.  On these archs
> > struct timeval has padding bytes at the end.  This struct is copied to
> > userspace with these padding bytes uninitialized.  This leads to leaking
> > of contents of kernel stack memory.
> > 
> > Signed-off-by: Vasiliy Kulikov <segoon@openwall.com>
> > ---
> >  Patch v1 used memset(), it was waste of cycles on almost all archs.
> > 
> >  Compile tested.
> > 
> >  fs/select.c |    7 ++++---
> >  1 files changed, 4 insertions(+), 3 deletions(-)
> > 
> > diff --git a/fs/select.c b/fs/select.c
> > index b7b10aa..43d4805 100644
> > --- a/fs/select.c
> > +++ b/fs/select.c
> > @@ -288,7 +288,6 @@ static int poll_select_copy_remaining(struct timespec *end_time, void __user *p,
> >  				      int timeval, int ret)
> >  {
> >  	struct timespec rts;
> > -	struct timeval rtv;
> >  
> >  	if (!p)
> >  		return ret;
> > @@ -306,8 +305,10 @@ static int poll_select_copy_remaining(struct timespec *end_time, void __user *p,
> >  		rts.tv_sec = rts.tv_nsec = 0;
> >  
> >  	if (timeval) {
> > -		rtv.tv_sec = rts.tv_sec;
> > -		rtv.tv_usec = rts.tv_nsec / NSEC_PER_USEC;
> > +		struct timeval rtv = {
> > +			.tv_sec = rts.tv_sec,
> > +			.tv_usec = rts.tv_nsec / NSEC_PER_USEC
> > +		};
> >  
> >  		if (!copy_to_user(p, &rtv, sizeof(rtv)))
> >  			return ret;
> 
> Please check the assembly code - this will still leave four bytes of
> uninitalised stack data in 'rtv', surely.

This concrete c code generates movl + movq, movl would zero unnamed
4 bytes.  However, I cannot find whether this behavior is guaranteed...

-- 
Vasiliy Kulikov