From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: [RFC net-next PATCH V1 9/9] net: frag remove readers-writer lock (hack) Date: Sun, 25 Nov 2012 22:03:16 -0800 Message-ID: <20121125220316.7bbc2287@nehalam.linuxnetplumber.net> References: <20121123130749.18764.25962.stgit@dragon> <20121123130847.18764.87682.stgit@dragon> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: Eric Dumazet , "David S. Miller" , Florian Westphal , netdev@vger.kernel.org, Pablo Neira Ayuso , Thomas Graf , Cong Wang , "Patrick McHardy" , "Paul E. McKenney" , Herbert Xu To: Jesper Dangaard Brouer Return-path: Received: from mail.vyatta.com ([76.74.103.46]:50345 "EHLO mail.vyatta.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752311Ab2KZGE0 (ORCPT ); Mon, 26 Nov 2012 01:04:26 -0500 In-Reply-To: <20121123130847.18764.87682.stgit@dragon> Sender: netdev-owner@vger.kernel.org List-ID: On Fri, 23 Nov 2012 14:08:47 +0100 Jesper Dangaard Brouer wrote: > Do NOT APPLY this patch. > > After all the other patches, the rw_lock is now the contention point. Reader-writer locks are significantly slower than a simple spinlock. Unless the reader holds the lock for "significant" number of cycles, a spin lock will be faster. Did you try just changing it to a spinlock?