On Thu, May 26, 2005 at 02:37:35PM +0200, Sten Spans wrote: > On Thu, 26 May 2005, Bernd Walter wrote: > > >>> > >>>This is absolutely known - TCP/nfs has bugs in realigning packets. > >>>Don't use TCP on strong aligned architectures. > >> > >>Still a pr with a proper backtrace would be nice. > >>Or does one exist already ? > > > >Not that I know. > >I did know exactly when this happens years ago. > >The backtrace as such will not help you as the panic happens much > >later than the cause. > >IIRC the basic problem was that the realignment code only fixes > >a single missalignment, while theres a chance for more then one. > >Verify nfs_realign in nfsserver and nfsclient to get an idea. > >If you are interested - I've found a (non-working) patch that I wrote > >for it, but the intention of it should be clear. > > > > Sure. This is an nfsd specific problem, > Or does nfsclient have issues as well ? The code is the same in client and server - as well as the risk to get packets in that form from network. > I'll get a pr going to make sure that the issue > is documented, and possibly narrowed down enough for > other people to start painting a bikeshed about how > it should be fixed. OK. -- B.Walter BWCT http://www.bwct.de bernd_at_bwct.de info_at_bwct.deReceived on Thu May 26 2005 - 10:52:51 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:35 UTC