netdev
[Top] [All Lists]

Re: [PATCH] IPV6_CHECKSUM socket option can corrupt kernel memory

To: herbert@xxxxxxxxxxxxxxxxxxx, davem@xxxxxxxxxxxxx
Subject: Re: [PATCH] IPV6_CHECKSUM socket option can corrupt kernel memory
From: YOSHIFUJI Hideaki / 吉藤英明 <yoshfuji@xxxxxxxxxxxxxx>
Date: Thu, 14 Apr 2005 21:34:32 +0900 (JST)
Cc: dlstevens@xxxxxxxxxx, netdev@xxxxxxxxxxx, yoshfuji@xxxxxxxxxxxxxx
In-reply-to: <E1DM3QD-00045e-00@gondolin.me.apana.org.au>
Organization: USAGI Project
References: <OF618A0572.CA254DA0-ON88256FE1.0080DB81-88256FE1.0082220B@us.ibm.com> <E1DM3QD-00045e-00@gondolin.me.apana.org.au>
Sender: netdev-bounce@xxxxxxxxxxx
In article <E1DM3QD-00045e-00@xxxxxxxxxxxxxxxxxxxxxxxx> (at Thu, 14 Apr 2005 
22:26:13 +1000), Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx> says:

> So here is a patch that introduces skb_store_bits -- the opposite of
> skb_copy_bits, and uses them to read/write the csum field in rawv6.
> 
> Signed-off-by: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>

(I re-review this later, but anyway,) I basically agree;
this is what I wanted.

Signed-off-by: Hideaki YOSHIFUJI <yoshfuji@xxxxxxxxxxxxxx>

--yoshfuji

<Prev in Thread] Current Thread [Next in Thread>