Re: [PATCH] fs/binfmt_elf.c: allocate initialized memory in fill_thread_core_info()

[Kees Cook <keescook@chromium.org>]

On Mon, Apr 20, 2020 at 03:33:52PM -0700, Andrew Morton wrote:
> On Sun, 19 Apr 2020 12:08:48 +0200 glider@google.com wrote:
> 
> > KMSAN reported uninitialized data being written to disk when dumping
> > core. As a result, several kilobytes of kmalloc memory may be written to
> > the core file and then read by a non-privileged user.

Ewww. That's been there for 12 years. Did something change in
regset_size() or regset->get()? Do you know what leaves the hole?

> > 
> > ...
> >
> > --- a/fs/binfmt_elf.c
> > +++ b/fs/binfmt_elf.c
> > @@ -1733,7 +1733,7 @@ static int fill_thread_core_info(struct elf_thread_core_info *t,
> >  		    (!regset->active || regset->active(t->task, regset) > 0)) {
> >  			int ret;
> >  			size_t size = regset_size(t->task, regset);
> > -			void *data = kmalloc(size, GFP_KERNEL);
> > +			void *data = kzalloc(size, GFP_KERNEL);
> >  			if (unlikely(!data))
> >  				return 0;
> >  			ret = regset->get(t->task, regset,
> 
> This seems to be a quite easy way of exposing quite a large amount of
> kernel memory contents, so I think I'll add a cc:stable to this patch?

Yes please.

Acked-by: Kees Cook <keescook@chromium.org>

-Kees

-- 
Kees Cook