netdev - Re: [PATCH v4 net-next 02/12] ipv6: add IFLA_GSO_IPV6_MAX

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: Windows password security audit tool. GUI, reports in PDF.

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

Date:   Fri, 6 May 2022 15:25:31 -0700
From:   Eric Dumazet <edumazet@...gle.com>
To:     Alexander Duyck <alexander.duyck@...il.com>
Cc:     Eric Dumazet <eric.dumazet@...il.com>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Paolo Abeni <pabeni@...hat.com>,
        netdev <netdev@...r.kernel.org>, Coco Li <lixiaoyan@...gle.com>
Subject: Re: [PATCH v4 net-next 02/12] ipv6: add IFLA_GSO_IPV6_MAX_SIZE

On Fri, May 6, 2022 at 3:16 PM Alexander Duyck
<alexander.duyck@...il.com> wrote:
>
> On Fri, May 6, 2022 at 2:50 PM Eric Dumazet <edumazet@...gle.com> wrote:
> >
> > On Fri, May 6, 2022 at 2:37 PM Alexander Duyck
> > <alexander.duyck@...il.com> wrote:
> > >
> > > On Fri, May 6, 2022 at 2:20 PM Eric Dumazet <edumazet@...gle.com> wrote:
> > > >
> > > > On Fri, May 6, 2022 at 1:48 PM Alexander H Duyck
> > > > <alexander.duyck@...il.com> wrote:
> > > > >
> > > > > On Fri, 2022-05-06 at 08:30 -0700, Eric Dumazet wrote:
> > > > > > From: Coco Li <lixiaoyan@...gle.com>
> > > > > >
> > > > > > This enables ipv6/TCP stacks to build TSO packets bigger than
> > > > > > 64KB if the driver is LSOv2 compatible.
> > > > > >
> > > > > > This patch introduces new variable gso_ipv6_max_size
> > > > > > that is modifiable through ip link.
> > > > > >
> > > > > > ip link set dev eth0 gso_ipv6_max_size 185000
> > > > > >
> > > > > > User input is capped by driver limit (tso_max_size)
> > > > > >
> > > > > > Signed-off-by: Coco Li <lixiaoyan@...gle.com>
> > > > > > Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> > > > >
> > > > > So I am still not a fan of adding all this extra tooling to make an
> > > > > attribute that is just being applied to one protocol. Why not just
> > > > > allow gso_max_size to extend beyond 64K and only limit it by
> > > > > tso_max_size?
> > > >
> > > > Answer is easy, and documented in our paper. Please read it.
> > >
> > > I have read it.
> > >
> > > > We do not want to enable BIG TCP for IPv4, this breaks user space badly.
> > > >
> > > > I do not want to break tcpdump just because some people think TCP just works.
> > >
> > > The changes I suggested don't enable it for IPv4. What your current
> > > code is doing now is using dev->gso_max_size and if it is the correct
> > > IPv6 type you are using dev->gso_ipv6_max_size. What I am saying is
> > > that instead of adding yet another netdev control you should just make
> > > it so that we retain existing behavior when gso_max_size is less than
> > > GSO_MAX_SIZE, and when it exceeds it all non-ipv6 types max out at
> > > GSO_MAX_SIZE and only the ipv6 type packets use gso_max_size when you
> > > exceed GSO_MAX_SIZE.
> >
> > gso_max_size can not exceed GSO_MAX_SIZE.
> > This will break many drivers.
> > I do not want to change hundreds of them.
>
> Most drivers will not be impacted because they cannot exceed
> tso_max_size. The tso_max_size is the limit, not GSO_MAX_SIZE. Last I
> knew this patch set is overwriting that value to increase it beyond
> the legacy limits.
>
> Right now the check is:
> if (max_size > GSO_MAX_SIZE || || max_size > dev->tso_max_size)
>

This is Jakub patch, already merged. Nothing to do with BIG TCP ?


> What I am suggesting is that tso_max_size be used as the only limit,
> which is already defaulted to cap out at TSO_LEGACY_MAX_SIZE. So just
> remove the "max_size > GSO_MAX_SIZE ||" portion of the call. Then when
> you call netif_set_tso_max_size in the driver to enable jumbograms you
> are good to set gso_max_size to something larger than the standard
> 65536.

Again, this was Jakub patch, right ?

>
> > Look, we chose this implementation so that chances of breaking things
> > are very small.
> > I understand this is frustrating, but I suggest you take the
> > responsibility of breaking things,
> > and not add this on us.
>
> What I have been trying to point out is your patch set will break things.
>

Can you give a concrete example ?

> For all those cases out there where people are using gso_max_size to
> limit things you just poked a hole in that for IPv6 cases. What I am
> suggesting is that we don't do that as it will be likely to trigger a
> number of problems for people.

No, because legacy drivers won't  use BIG TCP.

dev->tso_max_size will be <= 65536 for them.

Look at netif_set_gso_ipv6_max_size() logic.

>
> The primary reason gso_max_size was added was because there are cases
> out there where doing too big of a TSO was breaking things. For
> devices that are being used for LSOv2 I highly doubt they need to
> worry about cases less than 65536. As such they can just max out at
> 65536 for all non-IPv6 traffic and instead use gso_max_size as the
> limit for the IPv6/TSO case.

I think we disagree completely.