[<prev] [next>] [day] [month] [year] [list]
Message-ID: <20060715000458.GB9334@gondor.apana.org.au>
Date: Sat, 15 Jul 2006 10:04:58 +1000
From: Herbert Xu <herbert@...dor.apana.org.au>
To: Andrew Morton <akpm@...l.org>
Cc: Reuben Farrelly <reuben-lkml@...b.net>,
linux-kernel@...r.kernel.org, netdev@...r.kernel.org
Subject: Re: 2.6.18-rc1-mm2
On Fri, Jul 14, 2006 at 12:05:51AM -0700, Andrew Morton wrote:
>
> > Call Trace:
> > [<ffffffff8026963e>] show_trace+0xae/0x265
> > [<ffffffff8026980a>] dump_stack+0x15/0x1b
> > [<ffffffff8043ba7b>] skb_checksum_help+0x61/0x126
> > [<ffffffff8802f35f>] :iptable_nat:ip_nat_fn+0x5f/0x1d2
This is tell you that there is a bug in ip_nat_fn in that it completes
the partial checksum even for TSO packets which will cause them to go
out with bogus checksums.
The warning also indicates that the system has detected this and has
worked around it by recomputing the partial checksum after NAT.
The warning is here so someone can fix NAT to not trash the partial
checksum. It would also tell us if anyone else breaks checksums in
this way.
I've already made the warning appear only once per-boot so I'd really
like to keep it in until
1) NAT is fixed.
2) We're reasonably sure there's nothing else doing this.
Prior to this change your TSO packets would've gone out with corrupted
checksums silently. Essentially TSO would only slow your machine down
since every transmission it makes has to be retransmitted as non-TSO.
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@...dor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists