Re: [PATCH] x86: uaccess: fix regression in unsafe_get_user

[Arthur Gautier]

On Mon, Feb 18, 2019 at 09:51:50PM +0000, Arthur Gautier wrote:
> On Mon, Feb 18, 2019 at 11:15:44AM -0800, Andy Lutomirski wrote:
> > This seems like it's just papering over the underlying problem: with
> > Jann's new checks in place, strncpy_from_user() is simply buggy.  Does
> > the patch below look decent?  It's only compile-tested, but it's
> > conceptually straightforward.  I was hoping I could get rid of the
> > check-maximum-address stuff, but it's needed for architectures where
> > the user range is adjacent to the kernel range (i.e. not x86_64).
> 
> I'm unable to trigger the BUG I had with my initramfs with this patch
> applied. Thanks!
> 

Hello All,

Just a followup on this issue, I'm still able to reproduce the original
issue with:
    truncate -s 8388313 a
    SECONDFILENAME=bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb
    truncate -s 10 $SECONDFILENAME
    echo "a\n$SECONDFILENAME" | cpio -o --format=newc | lz4 -l > initrd.img.lz4

I think Andy submitted a patch Feb 25 2019, but I was not copied on it
(I believe it was sent to x86@xxxxxxxxxx) and I don't know which fate it
had.

Any chance we could have a look again?

Thanks a lot!

-- 
\o/ Arthur
 G  Gandi.net