Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 13 Nov 2019 07:05:36 -0500 (EST)
From:   Mikulas Patocka <mpatocka@redhat.com>
To:     Nikos Tsironis <ntsironis@arrikto.com>
cc:     tglx@linutronix.de, linux-rt-users@vger.kernel.org,
        Mike Snitzer <msnitzer@redhat.com>,
        Scott Wood <swood@redhat.com>,
        Ilias Tsitsimpis <iliastsi@arrikto.com>, dm-devel@redhat.com,
        linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
        Daniel Wagner <dwagner@suse.de>,
        Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Subject: Re: [PATCH RT 2/2 v2] list_bl: avoid BUG when the list is not
 locked
In-Reply-To: <7020d479-e8c7-7249-c6cd-c6d01b01c92a@arrikto.com>
Message-ID: <alpine.LRH.2.02.1911130704430.28238@file01.intranet.prod.int.rdu2.redhat.com>
References: <alpine.LRH.2.02.1911121110430.12815@file01.intranet.prod.int.rdu2.redhat.com> <335dafcb-5e07-63ed-b288-196516170bde@arrikto.com> <alpine.LRH.2.02.1911130616240.20335@file01.intranet.prod.int.rdu2.redhat.com>
 <7020d479-e8c7-7249-c6cd-c6d01b01c92a@arrikto.com>
User-Agent: Alpine 2.02 (LRH 1266 2009-07-14)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=WINDOWS-1252
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


On Wed, 13 Nov 2019, Nikos Tsironis wrote:

> On 11/13/19 1:16 PM, Mikulas Patocka wrote:
> >=20
> >=20
> > On Wed, 13 Nov 2019, Nikos Tsironis wrote:
> >=20
> >> On 11/12/19 6:16 PM, Mikulas Patocka wrote:
> >>> list_bl would crash with BUG() if we used it without locking. dm-snap=
shot=20
> >>> uses its own locking on realtime kernels (it can't use list_bl becaus=
e=20
> >>> list_bl uses raw spinlock and dm-snapshot takes other non-raw spinloc=
ks=20
> >>> while holding bl_lock).
> >>>
> >>> To avoid this BUG, we must set LIST_BL_LOCKMASK =3D 0.
> >>>
> >>> This patch is intended only for the realtime kernel patchset, not for=
 the=20
> >>> upstream kernel.
> >>>
> >>> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
> >>>
> >>> Index: linux-rt-devel/include/linux/list_bl.h
> >>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
> >>> --- linux-rt-devel.orig/include/linux/list_bl.h=092019-11-07 14:01:51=
.000000000 +0100
> >>> +++ linux-rt-devel/include/linux/list_bl.h=092019-11-08 10:12:49.0000=
00000 +0100
> >>> @@ -19,7 +19,7 @@
> >>>   * some fast and compact auxiliary data.
> >>>   */
> >>> =20
> >>> -#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
> >>> +#if (defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)) && !defi=
ned(CONFIG_PREEMPT_RT_BASE)
> >>>  #define LIST_BL_LOCKMASK=091UL
> >>>  #else
> >>>  #define LIST_BL_LOCKMASK=090UL
> >>> @@ -161,9 +161,6 @@ static inline void hlist_bl_lock(struct
> >>>  =09bit_spin_lock(0, (unsigned long *)b);
> >>>  #else
> >>>  =09raw_spin_lock(&b->lock);
> >>> -#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
> >>> -=09__set_bit(0, (unsigned long *)b);
> >>> -#endif
> >>>  #endif
> >>>  }
> >>> =20
> >>
> >> Hi Mikulas,
> >>
> >> I think removing __set_bit()/__clear_bit() breaks hlist_bl_is_locked()=
,
> >> which is used by the RCU variant of list_bl.
> >>
> >> Nikos
> >=20
> > OK. so I can remove this part of the patch.
> >=20
>=20
> I think this causes another problem. LIST_BL_LOCKMASK is used in various
> functions to set/clear the lock bit, e.g. in hlist_bl_first(). So, if we
> lock the list through hlist_bl_lock(), thus setting the lock bit with
> __set_bit(), and then call hlist_bl_first() to get the first element,
> the returned pointer will be invalid. As LIST_BL_LOCKMASK is zero the
> least significant bit of the pointer will be 1.
>=20
> I think for dm-snapshot to work using its own locking, and without
> list_bl complaining, the following is sufficient:
>=20
> --- a/include/linux/list_bl.h
> +++ b/include/linux/list_bl.h
> @@ -25,7 +25,7 @@
>  #define LIST_BL_LOCKMASK       0UL
>  #endif
>=20
> -#ifdef CONFIG_DEBUG_LIST
> +#if defined(CONFIG_DEBUG_LIST) && !defined(CONFIG_PREEMPT_RT_BASE)
>  #define LIST_BL_BUG_ON(x) BUG_ON(x)
>  #else
>  #define LIST_BL_BUG_ON(x)
>=20
> Nikos

Yes - so, submit this.

Reviewed-by: Mikulas Patocka <mpatocka@redhat.com>

Mikulas