Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <CANX2M5ZRrRA64k0hOif02TjmY9kbbO2aCBPyq79es34RXZ=cAw@mail.gmail.com>
 <CAG_fn=V57m0om5HUHHFOQr9R9TWHtfm4+jO96Smf+Q+XjRkxtQ@mail.gmail.com> <CANX2M5bWPMDJGgD=xq33A3p96ii3wBOuy9UKYAstX4psdAGrrA@mail.gmail.com>
In-Reply-To: <CANX2M5bWPMDJGgD=xq33A3p96ii3wBOuy9UKYAstX4psdAGrrA@mail.gmail.com>
From:   Alexander Potapenko <glider@google.com>
Date:   Thu, 13 Apr 2023 11:42:01 +0200
Message-ID: <CAG_fn=WOZ2RTyTYrgV3JhsqESf-YCmj_FmbzJZdJogpZk8SnOw@mail.gmail.com>
Subject: Re: Possible incorrect handling of fault injection inside KMSAN instrumentation
To:     Dipanjan Das <mail.dipanjan.das@gmail.com>
Cc:     Marco Elver <elver@google.com>, Dmitry Vyukov <dvyukov@google.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        kasan-dev@googlegroups.com, linux-mm@kvack.org,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        syzkaller <syzkaller@googlegroups.com>,
        Marius Fleischer <fleischermarius@googlemail.com>,
        Priyanka Bose <its.priyanka.bose@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Wed, Apr 12, 2023 at 8:24=E2=80=AFPM Dipanjan Das
<mail.dipanjan.das@gmail.com> wrote:
>
> On Wed, Apr 12, 2023 at 7:39=E2=80=AFAM Alexander Potapenko <glider@googl=
e.com> wrote:
>
> > Here are two patches that fix the problem:
> >  - https://github.com/google/kmsan/commit/b793a6d5a1c1258326b0f53d6e3ac=
8aa3eeb3499
> > - for kmsan_vmap_pages_range_noflush();
> >  - https://github.com/google/kmsan/commit/cb9e33e0cd7ff735bc302ff69c022=
74f24060cff
> > - for kmsan_ioremap_page_range()
> >
> > Can you please try them out?
>
> The second patch needs a small modification.
>
> The return value of `__vmap_pages_range_noflush` at Line 181
> (https://github.com/google/kmsan/commit/cb9e33e0cd7ff735bc302ff69c02274f2=
4060cff#diff-6c23520766ef70571c16b74ed93474716645c7ba81dc07028c076b6fd5ad27=
31R181)
> should also be assigned to `mapped`. With this modification, the patch
> works.

Good catch, thanks!
I'll send an updated version.