David, I am running an iperf test with 200 client threads and according to my knowledge the server opens only 1 socket for all the threads.. So this might be the same bug, what do you think?
Thanks a lot, Adhiraj. On 9/9/05, David S. Miller <[EMAIL PROTECTED]> wrote: > From: Adhiraj Joshi <[EMAIL PROTECTED]> > Date: Fri, 9 Sep 2005 11:35:27 +0530 > > > Hi, > > > > There was a bug some time back in a Must-Fix list regarding UDP > > applications going in dead lock. > > (http://kernel.org/pub/linux/kernel/people/akpm/must-fix/must-fix-2.txt) > > Here is an extract: > > -------------------------------------------------------------------------------------------------------------------------------------------------- > > > > - UDP apps can in theory deadlock, because the ip_append_data path can end > > up sleeping while the socket lock is held. > > > > It is OK to sleep with the socket held held, normally. But in this case > > the sleep happens while waiting for socket memory/space to become > > > > available, if another context needs to take the socket lock to free up the > > space we could hang. > > > > -------------------------------------------------------------------------------------------------------------------------------------------------- > > > > Is the fix present in the kernel now? I faced a problem similar to this one. > > This bug is almost entirely theoretical and nearly impossible to > trigger with any current code path. At a minimum you would need to > have two thread simultaneously sending over the same UDP socket. > - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
