From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roland Dreier Subject: Re: [PATCH 3/14] nes: connection manager routines Date: Thu, 09 Aug 2007 10:37:10 -0700 Message-ID: References: <200708080050.l780oGxo004694@neteffect.com> <46B9207D.6000905@garzik.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: ggrundstrom@neteffect.com, ewg@lists.openfabrics.org, netdev@vger.kernel.org To: Jeff Garzik Return-path: Received: from sj-iport-1-in.cisco.com ([171.71.176.70]:14602 "EHLO sj-iport-1.cisco.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S965964AbXHIRhT (ORCPT ); Thu, 9 Aug 2007 13:37:19 -0400 In-Reply-To: <46B9207D.6000905@garzik.org> (Jeff Garzik's message of "Tue, 07 Aug 2007 21:46:37 -0400") Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org > > +atomic_t cm_connects; > > +atomic_t cm_accepts; > > +atomic_t cm_disconnects; > > +atomic_t cm_closes; > > +atomic_t cm_connecteds; > > +atomic_t cm_connect_reqs; > > +atomic_t cm_rejects; > > do you really want to take the hit of a LOCK prefix each time you > increment a stat??? I think these are once-per-connection things, so the overhead is not that bad. On the other hand there's probably another lock you have to take anyway so maybe you can just define that these are protected by some lock you take anyway. Also, these names seem kind of generic to be global -- a nes_ prefix is probably appropriate. - R.