From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from merlin.infradead.org ([205.233.59.134]:42362 "EHLO merlin.infradead.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751012AbdGNXRk (ORCPT ); Fri, 14 Jul 2017 19:17:40 -0400 Subject: Re: EPOLLET behavior and performance To: "Brian C. Anderson" , linux-fsdevel@vger.kernel.org Cc: Victor Miura References: From: Randy Dunlap Message-ID: <66862a8e-bdf3-af5c-335f-a65d2d4c6975@infradead.org> Date: Fri, 14 Jul 2017 16:17:37 -0700 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US Content-Transfer-Encoding: 7bit Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On 07/14/2017 03:45 PM, Brian C. Anderson wrote: > Hi, > > Hoping I have the right email list for this topic. fsdevel makes sense, but linux-kernel would probably provide more eyes on it. [snip] > A few questions: > 1) Do you see any problems with the approach I'm taking? > 2) How concerning is backwards compatibility; especially regarding > user code that may not handle being notified of EPOLLIN when the file > isn't actually readable anymore. > 3) A ~30% improvement is larger than I would have expected. Any clues > what might be going on? I find it hard to believe improved cache > locality explains it all. > 4) Are there existing performance tests I can run the patch against? You could try some of Davide Libenzi's epoll test programs, although they need some updating before they are usable. See: http://xmailserver.org/epoll_test.c {internals test} http://xmailserver.org/totalmess.c {multithread test} and http://marc.info/?l=linux-ia64&m=108458304114939&w=2 {performance} -- ~Randy