Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933914AbXF2Qfj (ORCPT ); Fri, 29 Jun 2007 12:35:39 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S933410AbXF2QfR (ORCPT ); Fri, 29 Jun 2007 12:35:17 -0400 Received: from minus.inr.ac.ru ([194.67.69.97]:43800 "HELO ms2.inr.ac.ru" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with SMTP id S933073AbXF2QfP (ORCPT ); Fri, 29 Jun 2007 12:35:15 -0400 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=ms2.inr.ac.ru; b=T7yun7uuxNVK8xMuhlHw1f8h8Y8CDOONVo+aKDxi2oq4BbKHUp9V3R/aGLOCbnSrZpboJIj3T29wOcepPY4v5Yj2/kMJd/MzA14yIonGuHC9/exq930c0g0B6Rmnk7xwk8803hkx8/A7ihTYluhau6mnW4FG+W4g4vYV2bt+kO0=; Date: Fri, 29 Jun 2007 20:34:09 +0400 From: Alexey Kuznetsov To: Duncan Sands Cc: Ingo Molnar , Jeff Garzik , Linus Torvalds , Steven Rostedt , LKML , Andrew Morton , Thomas Gleixner , Christoph Hellwig , john stultz , Oleg Nesterov , "Paul E. McKenney" , Dipankar Sarma , "David S. Miller" , matthew.wilcox@hp.com Subject: Re: [RFC PATCH 0/6] Convert all tasklets to workqueues Message-ID: <20070629163409.GB17709@ms2.inr.ac.ru> References: <20070622040014.234651401@goodmis.org> <200706291348.21193.duncan.sands@math.u-psud.fr> <20070629133658.GB19021@ms2.inr.ac.ru> <200706291601.41680.duncan.sands@math.u-psud.fr> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200706291601.41680.duncan.sands@math.u-psud.fr> User-Agent: Mutt/1.5.6i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 905 Lines: 27 Hello! > What changed? softirq remains raised for such tasklet. Old times softirq was processed once per invocation, in schedule and on syscall exit and this was relatively harmless. Since softirqs are very weakly moderated, it results in strong cpu hogging. > And can it be fixed? With current tasklets this can be fixed only introducing additional synchronization cost to tasklet schedule. Not good. Better to fix caller. If Ingo knows about this, I guess it is already fixed in -rt tasklet or can be fixed. They are really flexible and extensible: -rt tasklet cost is so high, that even another thousand of cycles will remain unnnoticed. :-) Alexey - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/