Messages in this thread | | | Date | Wed, 29 Jul 1998 23:57:31 -0700 (PDT) | From | Linus Torvalds <> | Subject | Re: PATCH: NFS client tcp hanging |
| |
On Wed, 29 Jul 1998, David S. Miller wrote: > > We have the rpciod kernel thread, why don't we use it for what it was > designed, to get into a context and do things such as creating and > opening sockets etc. ie. things which we need to do outside of a > plain interrupt context.
I'd certainly prefer that. Opening a socket from bh-land just isn't safe, whether it's the net-bh or anything else.
I applied Alan's patch, because it certainly can't make things any worse, but it still sounds (from Alan's and Davids emails - I haven't actually personally looked at what the TCP client needs to do) like we're doing something dodgy with bh handlers.
(And I don't think it has to be a rpciod thread, it could just as well be a "per-mount-TCP-thread" - threads are certainly low-cost enough that it might make sense to have one per mount).
Linus
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
| |