On Tue, Aug 5, 2014 at 4:42 PM, Bruce Fields <bfields@xxxxxxxxxxxx> wrote: > On Mon, Aug 04, 2014 at 09:36:41AM -0400, Bruce Fields wrote: >> On Sun, Aug 03, 2014 at 01:03:02PM -0400, Trond Myklebust wrote: >> > When testing knfsd on 10GigE and 40GigE networks, we've been hitting a few >> > scalability issues that are seriously affecting performance. The main issue >> > was scalability of the pool->sp_lock, but we also hit a couple of things >> > like the use of ioctls in the receive fast path, as well as some issues >> > with heuristic tests being performed twice in the same path. >> > >> > This is not urgent, and can definitely be delayed until the 3.18 merge >> > window, however since the performance gains were significant over NFSv3, >> > I thought I'd share now. >> >> Thanks! On a quick skim these and the DRC patches look good. >> >> My first priority is to see if I can merge the last of the state locking >> patches which are already in nfsd-next, so yes I'll probably queue these >> up for 3.18 later. > > Though note I already applied the first three--I'm assuming the versions > of those posted here are the same as the ones previously applied. (See > git://linux-nfs.org/~bfields/linux.git for-3.17.) > Yes. Those first 3 should be identical to the patches that I posted earlier. -- Trond Myklebust Linux NFS client maintainer, PrimaryData trond.myklebust@xxxxxxxxxxxxxxx -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
