netdev - Re: [PATCH v2 net-next 2/6] net: dsa: propagate flags down towards drivers

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20230328114943.4mibmn2icutcio4m@skbuf>
Date:   Tue, 28 Mar 2023 14:49:43 +0300
From:   Vladimir Oltean <olteanv@...il.com>
To:     Hans Schultz <netdev@...io-technology.com>
Cc:     davem@...emloft.net, kuba@...nel.org, netdev@...r.kernel.org,
        Florian Fainelli <f.fainelli@...il.com>,
        Andrew Lunn <andrew@...n.ch>,
        Eric Dumazet <edumazet@...gle.com>,
        Paolo Abeni <pabeni@...hat.com>,
        Kurt Kanzenbach <kurt@...utronix.de>,
        Hauke Mehrtens <hauke@...ke-m.de>,
        Woojung Huh <woojung.huh@...rochip.com>,
        "maintainer:MICROCHIP KSZ SERIES ETHERNET SWITCH DRIVER" 
        <UNGLinuxDriver@...rochip.com>, Sean Wang <sean.wang@...iatek.com>,
        Landen Chao <Landen.Chao@...iatek.com>,
        DENG Qingfang <dqfext@...il.com>,
        Matthias Brugger <matthias.bgg@...il.com>,
        AngeloGioacchino Del Regno 
        <angelogioacchino.delregno@...labora.com>,
        Claudiu Manoil <claudiu.manoil@....com>,
        Alexandre Belloni <alexandre.belloni@...tlin.com>,
        Clément Léger <clement.leger@...tlin.com>,
        Jiri Pirko <jiri@...nulli.us>,
        Ivan Vecera <ivecera@...hat.com>,
        Roopa Prabhu <roopa@...dia.com>,
        Nikolay Aleksandrov <razor@...ckwall.org>,
        Shuah Khan <shuah@...nel.org>,
        Christian Marangi <ansuelsmth@...il.com>,
        Ido Schimmel <idosch@...dia.com>,
        open list <linux-kernel@...r.kernel.org>,
        "moderated list:ARM/Mediatek SoC support" 
        <linux-arm-kernel@...ts.infradead.org>,
        "moderated list:ARM/Mediatek SoC support" 
        <linux-mediatek@...ts.infradead.org>,
        "open list:RENESAS RZ/N1 A5PSW SWITCH DRIVER" 
        <linux-renesas-soc@...r.kernel.org>,
        "moderated list:ETHERNET BRIDGE" <bridge@...ts.linux-foundation.org>,
        "open list:KERNEL SELFTEST FRAMEWORK" 
        <linux-kselftest@...r.kernel.org>
Subject: Re: [PATCH v2 net-next 2/6] net: dsa: propagate flags down towards
 drivers

On Tue, Mar 28, 2023 at 01:04:23PM +0200, Hans Schultz wrote:
> On Tue, Mar 28, 2023 at 01:59, Vladimir Oltean <olteanv@...il.com> wrote:
> >
> > which idea is that, again?
> 
> So I cannot us the offloaded flag as it is added by DSA in the common
> case when using 'bridge fdb replace ... dynamic'.

Why not? I find it reasonable that the software bridge does not age out
a dynamic FDB entry that is offloaded to hardware... the hardware should
do that ("dynamic" being the key). At least, I find it more reasonable
than the current behavior, where the bridge notifies dynamic FDB entries
to switchdev, but doesn't say they're dynamic, and switchdev treats them
as static, so they don't roam from one bridge port to another until
software sees a packet with that MAC DA, and they have the potential of
blocking traffic because of that.

If for some reason you do think that behavior is useful and still want
to keep it (I'm not sure I would), I would consider extending struct
switchdev_notifier_fdb_info with a "bool pls_dont_age_out", and I would
make dsa_fdb_offload_notify() set this to true if the driver did
actually install the dynamic FDB entry as dynamic in the ATU.

> 
> The idea is then to use the ext_learn flag instead, which is not aged by
> the bridge. To do this the driver (mv88e6xxx) will send a
> SWITCHDEV_FDB_ADD_TO_BRIDGE switchdev event when the new dynamic flag is
> true. The function sending this event will then be named
> mv88e6xxx_add_fdb_synth_learned() in
> drivers/net/dsa/mv88e6xxx/switchdev.c, replacing the
> mv88e6xxx_set_fdb_offloaded() function but in most part the same
> content, just another event type.

Basically you're suggesting that the hardware driver, after receiving a
SWITCHDEV_FDB_ADD_TO_DEVICE and responding to it with SWITCHDEV_FDB_OFFLOADED,
emits a SWITCHDEV_FDB_ADD_TO_BRIDGE which takes over that software
bridge FDB entry, with the advantage that the new one already has the
semantics of not being aged out by the software bridge.

hmmm... I'd say that the flow should work even with a single notifier
emitted from the driver side, which would be SWITCHDEV_FDB_OFFLOADED,
perhaps annotated with some qualifiers that would inform the bridge a
certain behavior is required. Although, as mentioned, I think that in
principle, "pls_dont_age_out" should be unnecessary, because it just
papers over the issue that switchdev drivers treat static and dynamic
FDB entries just the same, and "pls_dont_age_out" would be the
differentiator for an issue that should have been solved elsewhere, as
it could lead to other problems of its own.

Basically we're designing around a workaround to a problem to which
we're turning a blind eye. These are my 2c.