netdev - Re: [PATCH net] net: sungem: fix rx checksum support

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:   Wed, 20 Jun 2018 13:22:40 +0200
From:   Mathieu Malaterre <malat@...ian.org>
To:     "David S. Miller" <davem@...emloft.net>
Cc:     Eric Dumazet <edumazet@...gle.com>, netdev@...r.kernel.org,
        Meelis Roos <mroos@...ux.ee>, schwab@...ux-m68k.org,
        eric.dumazet@...il.com
Subject: Re: [PATCH net] net: sungem: fix rx checksum support

On Wed, Jun 20, 2018 at 7:31 AM David Miller <davem@...emloft.net> wrote:
>
> From: Eric Dumazet <edumazet@...gle.com>
> Date: Tue, 19 Jun 2018 19:18:50 -0700
>
> > After commit 88078d98d1bb ("net: pskb_trim_rcsum() and CHECKSUM_COMPLETE
> > are friends"), sungem owners reported the infamous "eth0: hw csum failure"
> > message.
> >
> > CHECKSUM_COMPLETE has in fact never worked for this driver, but this
> > was masked by the fact that upper stacks had to strip the FCS, and
> > therefore skb->ip_summed was set back to CHECKSUM_NONE before
> > my recent change.
> >
> > Driver configures a number of bytes to skip when the chip computes
> > the checksum, and for some reason only half of the Ethernet header
> > was skipped.
> >
> > Then a second problem is that we should strip the FCS by default,
> > unless the driver is updated to eventually support NETIF_F_RXFCS in
> > the future.
> >
> > Finally, a driver should check if NETIF_F_RXCSUM feature is enabled
> > or not, so that the admin can turn off rx checksum if wanted.
> >
> > Many thanks to Andreas Schwab and Mathieu Malaterre for their
> > help in debugging this issue.
> >
> > Signed-off-by: Eric Dumazet <edumazet@...gle.com>
> > Reported-by: Meelis Roos <mroos@...ux.ee>
> > Reported-by: Mathieu Malaterre <malat@...ian.org>
> > Reported-by: Andreas Schwab <schwab@...ux-m68k.org>
> > Tested-by: Andreas Schwab <schwab@...ux-m68k.org>
>
> Applied and queued up for -stable, thanks Eric.

Thanks for the stable tag.

IMHO the commit message should have also reference commit 7ce5a27f2ef8
("Revert "net: Handle CHECKSUM_COMPLETE more adequately in
pskb_trim_rcsum().""). Which means that commit 88078d98d1bb ("net:
pskb_trim_rcsum() and CHECKSUM_COMPLETE are friends") should be ready
for stable also, since it seems that the only driver responsible for
the revert was sungem.

my 2cts