On Tue, Apr 14, 2020 at 04:10:35PM +0100, Will Deacon wrote:
> On Tue, Apr 14, 2020 at 03:53:45PM +0100, Vincenzo Frascino wrote:
> > On 4/14/20 2:27 PM, Mark Rutland wrote:
> > > On Tue, Apr 14, 2020 at 01:50:38PM +0100, Vincenzo Frascino wrote:
> > >> On 4/14/20 11:42 AM, Mark Rutland wrote:
> > >>> The aarch32_vdso_pages[] array never has entries allocated in the C_VVAR
> > >>> or C_VDSO slots, and as the array is zero initialized these contain
> > >>> NULL.
> > >>>
> > >>> However in __aarch32_alloc_vdso_pages() when
> > >>> aarch32_alloc_kuser_vdso_page() fails we attempt to free the page whose
> > >>> struct page is at NULL, which is obviously nonsensical.
> > >>
> > >> Could you please explain why do you think that free(NULL) is "nonsensical"?
> > >
> > > Regardless of the below, can you please explain why it is sensical? I'm
> > > struggling to follow your argument here.
> >
> > free(NULL) is a no-operation ("no action occurs") according to the C standard
> > (ISO-IEC 9899 paragraph 7.20.3.2). Hence this should not cause any bug if the
> > allocator is correctly implemented. From what I can see the implementation of
> > the page allocator honors this assumption.
> >
> > Since you say it is a bug (providing evidence), we might have to investigate
> > because probably there is an issue somewhere else.
>
> Not sure why you feel the need to throw the C standard around -- the patch
> from Mark looks obviously like the right thing to do to me, so:
>
> Acked-by: Will Deacon <will@xxxxxxxxxx>
>
> Catalin -- please take this one as a fix so that I can queue the rest of
> the patches for 5.8 once it's hit mainline.
I queued this patch for -rc2. Thanks.
--
Catalin
