From mboxrd@z Thu Jan 1 00:00:00 1970 From: Piotr =?utf-8?B?RGHFgmVr?= Subject: Re: Hammer vs Jewel librbd performance testing and git bisection results Date: Wed, 11 May 2016 17:22:32 +0200 Message-ID: <20160511152232.GC6902@predictor> References: <20160511133504.GB6902@predictor> <751229226.29976609.1462974560840.JavaMail.zimbra@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from predictor.org.pl ([185.5.97.54]:52842 "EHLO predictor.org.pl" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932181AbcEKPU1 (ORCPT ); Wed, 11 May 2016 11:20:27 -0400 Content-Disposition: inline In-Reply-To: <751229226.29976609.1462974560840.JavaMail.zimbra@redhat.com> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Matt Benjamin Cc: ceph-devel@vger.kernel.org On Wed, May 11, 2016 at 09:49:20AM -0400, Matt Benjamin wrote: > > > syNM5pr-ZwBuR7lqnphEd-4kQUid0C9eRyta3ohOA/edit?usp=3Dsharing > > >=20 > > > There are several commits of interest that have a noticeable effe= ct > > > on 128K sequential read performance: > > >=20 > > > [..] > > > 2) https://github.com/ceph/ceph/commit/c474ee42 > > >=20 > > > This commit had a very large impact, reducing performance by anot= her > > > 20-25%. > >=20 > > https://github.com/ceph/ceph/commit/c474ee42#diff-254555dde8dcfb7fb= 908791ab8214b92R318 > > I would check if temporarily forcing unique_lock_name() to return i= ts arg > > (or other constant) would change things. If so, probably a more eff= icient way > > to construct unique lock name may be in order. >=20 > ++ >=20 > Naively, too, what unique_lock_name is doing amounts to: 1) creating= extra of [a small] std::string [better fixed, but not a likely root ca= use?] 2) using Utils::stringify to hook ostream operators in the type p= assed in, and doing that on a new sstream; =20 I don't thing "stringify" alone is a source of problems. This looks lik= e a convenience function to make object dumping easier, so no real point in changing it, rather than that... > Maybe we should look horizontally at either speeding up or finding al= ternatives to stringify in other places? =2E.. it's usage should be limited to non-performance-critical paths. In this particular case, probably checking for lockdep and refusing to = act=20 when it is disabled sounds like the least intrusive way to fix things. --=20 Piotr Da=C5=82ek branch@predictor.org.pl http://blog.predictor.org.pl -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html