On Mon, 13 Apr 2020 16:19:40 +0200
Alexander Potapenko <glider@xxxxxxxxxx> wrote:
> On Wed, Apr 8, 2020 at 6:01 PM Andrey Konovalov <andreyknvl@xxxxxxxxxx> wrote:
> >
> > On Wed, Mar 25, 2020 at 5:14 PM <glider@xxxxxxxxxx> wrote:
> > >
> > > Disable the efficient 8-byte reading under KMSAN to avoid false positives.
> >
> > Another user of read_word_at_a_time() is dentry_string_cmp() in
> > dcache.c, should we disable it there as well?
> >
>
> I think we'd better disable DCACHE_WORD_ACCESS if KMSAN is enabled.
> Will do that in v6.
>
> > >
> > > Signed-off-by: Alexander Potapenko <glider@xxxxxxxxxx>
> > > To: Alexander Potapenko <glider@xxxxxxxxxx>
> > > Cc: Vegard Nossum <vegard.nossum@xxxxxxxxxx>
> > > Cc: Dmitry Vyukov <dvyukov@xxxxxxxxxx>
> > > Cc: Marco Elver <elver@xxxxxxxxxx>
> > > Cc: Andrey Konovalov <andreyknvl@xxxxxxxxxx>
> > > Cc: linux-mm@xxxxxxxxx
> > >
> > > ---
> > >
> > > v4:
> > > - actually disable the optimization under KMSAN via max=0
> > > - use IS_ENABLED as requested by Marco Elver
> > >
> > > Change-Id: I25d1acf5c3df6eff85894cd94f5ddbe93308271c
> > > ---
> > > lib/string.c | 8 ++++++++
> > > 1 file changed, 8 insertions(+)
> > >
> > > diff --git a/lib/string.c b/lib/string.c
> > > index 6012c385fb314..fec929e70f1a5 100644
> > > --- a/lib/string.c
> > > +++ b/lib/string.c
> > > @@ -202,6 +202,14 @@ ssize_t strscpy(char *dest, const char *src, size_t count)
> > > max = 0;
> > > #endif
> > >
> > > + /*
> > > + * read_word_at_a_time() below may read uninitialized bytes after the
> > > + * trailing zero and use them in comparisons. Disable this optimization
> > > + * under KMSAN to prevent false positive reports.
> > > + */
> > > + if (IS_ENABLED(CONFIG_KMSAN))
> > > + max = 0;
> > > +
Instead of disabling the optimization, can't you make KMSAN aware of the
"has_zero()" call (which I believe is the part that will trip up KMSAN) and
just not report it?
-- Steve
> > > while (max >= sizeof(unsigned long)) {
> > > unsigned long c, data;
> > >
> > > --
> > > 2.25.1.696.g5e7596f4ac-goog
> > >
>
>
>
