| To: | "David S. Miller" <davem@xxxxxxxxxx> |
|---|---|
| Subject: | Re: oops in tcp_v4_rcv. |
| From: | Arnaldo Carvalho de Melo <acme@xxxxxxxxxxxxxxxx> |
| Date: | Wed, 28 May 2003 22:50:20 -0300 |
| Cc: | manfred@xxxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx, akpm@xxxxxxxxx |
| In-reply-to: | <20030528.184054.78710412.davem@redhat.com> |
| Organization: | Conectiva S.A. |
| References: | <3ED54DBC.4020203@colorfullife.com> <3ED55F4D.1070306@colorfullife.com> <20030528.184054.78710412.davem@redhat.com> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mutt/1.5.4i |
Em Wed, May 28, 2003 at 06:40:54PM -0700, David S. Miller escreveu: > From: Manfred Spraul <manfred@xxxxxxxxxxxxxxxx> > Date: Thu, 29 May 2003 03:15:57 +0200 > > I think I understand now what causes the crash: > The tcp_ehash assumes that the entries are of the type 'struct inet_sock'. > But the actual entry is of the type tcp_tw_bucket. And 'sk->inet.daddr' > is not shared between both structures. > > Thanks for figuring this out. Indeed, I had suspected the > sock layout change Arnaldo did early in 2.5.x as the main > possible suspect. > > I'll try to fix this. I'm as well looking at this, longstanding bug :-\ - Arnaldo |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: oops in tcp_v4_rcv., David S. Miller |
|---|---|
| Next by Date: | Re: oops in tcp_v4_rcv., David S. Miller |
| Previous by Thread: | Re: oops in tcp_v4_rcv., David S. Miller |
| Next by Thread: | Re: oops in tcp_v4_rcv., David S. Miller |
| Indexes: | [Date] [Thread] [Top] [All Lists] |