On Mon, Aug 22, 2022 at 04:29:05PM +0200, Jens Wiklander wrote:
> On Mon, Aug 22, 2022 at 3:32 PM Greg KH <gregkh@xxxxxxxxxxxxxxxxxxx> wrote:
> >
> > On Mon, Aug 22, 2022 at 03:12:27PM +0200, Jens Wiklander wrote:
> > > commit 573ae4f13f630d6660008f1974c0a8a29c30e18a upstream.
> > >
> > > With special lengths supplied by user space, tee_shm_register() has
> > > an integer overflow when calculating the number of pages covered by a
> > > supplied user space memory region.
> > >
> > > This may cause pin_user_pages_fast() to do a NULL pointer dereference.
> > >
> > > Fix this by adding an an explicit call to access_ok() in
> > > tee_ioctl_shm_register() to catch an invalid user space address early.
> > >
> > > Fixes: 033ddf12bcf5 ("tee: add register user memory")
> > > Cc: stable@xxxxxxxxxxxxxxx # 5.4
> > > Cc: stable@xxxxxxxxxxxxxxx # 5.10
> > > Reported-by: Nimish Mishra <neelam.nimish@xxxxxxxxx>
> > > Reported-by: Anirban Chakraborty <ch.anirban00727@xxxxxxxxx>
> > > Reported-by: Debdeep Mukhopadhyay <debdeep.mukhopadhyay@xxxxxxxxx>
> > > Suggested-by: Jerome Forissier <jerome.forissier@xxxxxxxxxx>
> > > Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
> > > [JW: backport to stable 5.4 and 5.10 + update commit message]
> >
> > You already sent me a 5.4 version here:
> > https://lore.kernel.org/r/20220822092621.3691771-1-jens.wiklander@xxxxxxxxxx
> >
> > And I applied that.
> >
> > And for 5.10, it's already in the tree as commit 578c349570d2 ("tee: add
> > overflow check in register_shm_helper()") and was in the 5.10.137
> > release.
> >
> > > Signed-off-by: Jens Wiklander <jens.wiklander@xxxxxxxxxx>
> > > ---
> > > drivers/tee/tee_core.c | 3 +++
> > > 1 file changed, 3 insertions(+)
> > >
> > > diff --git a/drivers/tee/tee_core.c b/drivers/tee/tee_core.c
> > > index a7ccd4d2bd10..2db144d2d26f 100644
> > > --- a/drivers/tee/tee_core.c
> > > +++ b/drivers/tee/tee_core.c
> > > @@ -182,6 +182,9 @@ tee_ioctl_shm_register(struct tee_context *ctx,
> > > if (data.flags)
> > > return -EINVAL;
> > >
> > > + if (!access_ok((void __user *)(unsigned long)data.addr, data.length))
> > > + return -EFAULT;
> >
> > What I took in 5.10.137 was:
> >
> > + if (!access_ok((void __user *)addr, length))
> > + return ERR_PTR(-EFAULT);
> >
> > Should I fix it up to look like what you sent here instead?
>
> Yes, please.
Ok, no, that does not work on 5.10.y at all, it blows up with the
obvious issue that there is no data pointer in this function. It's also
in a different file, drivers/tee/tee_shm.c
So I'm going to leave 5.10.y alone for now, I think it's fixed.
I've taken your 5.4 patch that you sent previously, and still need a
4.19.y change (and any older kernels that are also vulnerable.)
thanks,
greg k-h
