|To:||Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>|
|From:||David Coulson <david@xxxxxxxxxxxxxxxx>|
|Date:||Tue, 25 Jan 2005 09:16:37 -0500|
|Cc:||Thomas Graf <tgraf@xxxxxxx>, "David S. Miller" <davem@xxxxxxxxxxxxx>, kaber@xxxxxxxxx, netdev@xxxxxxxxxxx|
|References:||<20050124005348.GL23931@xxxxxxxxxxxxxx> <E1Cst4o-0007bD-00@xxxxxxxxxxxxxxxxxxxxxxxx> <20050123202715.281ac87c.davem@xxxxxxxxxxxxx> <20050124121610.GP23931@xxxxxxxxxxxxxx> <41F50B6C.6010107@xxxxxxxxxxxxxxxx> <20050124151510.GV23931@xxxxxxxxxxxxxx> <20050124225423.GA15405@xxxxxxxxxxxxxxxxxxx>|
|User-agent:||Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.7.5) Gecko/20041217|
Herbert Xu wrote:
OK, I think I've found the problem. It's a totally innocuous bug in ip_fragment/ip6_fragment. When we're in the fast path and use the pre-existing frag_list skb's, we forgot to clear ip_summed.
Okay - I applied Herbert's patch to my kernels and it all looks good now. Thomas showed me how to generate fragged packets using hping2 (handy tool!) and replicate the kernel failure.
I slammed it with hping2 for a while, and didn't experience any kernel errors or see any debug information from previous patches. Everything is working as it should be, and I verified I saw the fragged packets in tcpdump to make sure they were being routed properly.
:-) David -- David J. Coulson email: david@xxxxxxxxxxxxxxxx web: http://www.davidcoulson.net/ phone: (216) 920-3100 / (216) 258-4942
|<Prev in Thread]||Current Thread||[Next in Thread>|