Return-Path: linux-nfs-owner@vger.kernel.org Received: from fieldses.org ([174.143.236.118]:49373 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757651AbaLIREu (ORCPT ); Tue, 9 Dec 2014 12:04:50 -0500 Date: Tue, 9 Dec 2014 12:04:48 -0500 From: "J. Bruce Fields" To: Ben Myers Cc: Andrew Dahl , Jeff Layton , Trond Myklebust , Chris Worley , linux-nfs@vger.kernel.org Subject: Re: [PATCH 3/4] sunrpc: convert to lockless lookup of queued server threads Message-ID: <20141209170447.GH20526@fieldses.org> References: <1416597571-4265-1-git-send-email-jlayton@primarydata.com> <1416597571-4265-4-git-send-email-jlayton@primarydata.com> <20141201234759.GF30749@fieldses.org> <20141202065750.283704a7@tlielax.poochiereds.net> <20141202071422.5b01585d@tlielax.poochiereds.net> <20141202165023.GA9195@fieldses.org> <20141202185358.GH11444@sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20141202185358.GH11444@sgi.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Tue, Dec 02, 2014 at 12:53:58PM -0600, Ben Myers wrote: > Hey Bruce, > > On Tue, Dec 02, 2014 at 11:50:24AM -0500, J. Bruce Fields wrote: > > On Tue, Dec 02, 2014 at 07:14:22AM -0500, Jeff Layton wrote: > > > On Tue, 2 Dec 2014 06:57:50 -0500 > > > Jeff Layton wrote: > > > > > > > On Mon, 1 Dec 2014 19:38:19 -0500 > > > > Trond Myklebust wrote: > > > > > > > > > On Mon, Dec 1, 2014 at 6:47 PM, J. Bruce Fields wrote: > > > > > > I find it hard to think about how we expect this to affect performance. > > > > > > So it comes down to the observed results, I guess, but just trying to > > > > > > get an idea: > > > > > > > > > > > > - this eliminates sp_lock. I think the original idea here was > > > > > > that if interrupts could be routed correctly then there > > > > > > shouldn't normally be cross-cpu contention on this lock. Do > > > > > > we understand why that didn't pan out? Is hardware capable of > > > > > > doing this really rare, or is it just too hard to configure it > > > > > > correctly? > > > > > > > > > > One problem is that a 1MB incoming write will generate a lot of > > > > > interrupts. While that is not so noticeable on a 1GigE network, it is > > > > > on a 40GigE network. The other thing you should note is that this > > > > > workload was generated with ~100 clients pounding on that server, so > > > > > there are a fair amount of TCP connections to service in parallel. > > > > > Playing with the interrupt routing doesn't necessarily help you so > > > > > much when all those connections are hot. > > > > > > > > > > > In principle though, the percpu pool_mode should have alleviated the > > > contention on the sp_lock. When an interrupt comes in, the xprt gets > > > queued to its pool. If there is a pool for each cpu then there should > > > be no sp_lock contention. The pernode pool mode might also have > > > alleviated the lock contention to a lesser degree in a NUMA > > > configuration. > > > > > > Do we understand why that didn't help? > > > > Yes, the lots-of-interrupts-per-rpc problem strikes me as a separate if > > not entirely orthogonal problem. > > > > (And I thought it should be addressable separately; Trond and I talked > > about this in Westford. I think it currently wakes a thread to handle > > each individual tcp segment--but shouldn't it be able to do all the data > > copying in the interrupt and wait to wake up a thread until it's got the > > entire rpc?) > > > > > In any case, I think that doing this with RCU is still preferable. > > > We're walking a very short list, so doing it lockless is still a > > > good idea to improve performance without needing to use the percpu > > > pool_mode. > > > > I find that entirely plausible. > > > > Maybe it would help to ask SGI people. Cc'ing Ben Myers in hopes he > > could point us to the right person. > > > > It'd be interesting to know: > > > > - are they using the svc_pool stuff? > > - if not, why not? > > - if so: > > - can they explain how they configure systems to take > > advantage of it? > > - do they have any recent results showing how it helps? > > - could they test Jeff's patches for performance > > regressions? > > > > Anyway, I'm off for now, back to work Thursday. > > > > --b. > > Andrew Dahl is the right person. Cc'd. Thanks! I'm less worried about Jeff's particular changes here, but I would still really love to see answers to the above questions. We've had a couple cases now of people trying to use the pool_modes for performance tuning without good results, and I'd like to figure out what's happening. If this keeps up then we may end up just breaking them by accident (if we haven't already). --b.