From mboxrd@z Thu Jan 1 00:00:00 1970 From: Evgeniy Polyakov Subject: Re: question about softirqs Date: Wed, 13 May 2009 09:55:11 +0400 Message-ID: <20090513055511.GA8529@ioremap.net> References: <18948.63755.279732.294842@cargo.ozlabs.ibm.com> <20090508.234815.127227651.davem@davemloft.net> <4A086DB2.8040703@nortel.com> <20090511.162436.193717082.davem@davemloft.net> <4A08C62F.1050105@nortel.com> <20090512081237.GA16403@elte.hu> <1242119578.11251.321.camel@twins> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Ingo Molnar , Chris Friesen , Thomas Gleixner , Steven Rostedt , David Miller , linuxppc-dev@ozlabs.org, paulus@samba.org, netdev@vger.kernel.org To: Peter Zijlstra Return-path: Received: from cs-studio.ru ([195.178.208.66]:50998 "EHLO tservice.net.ru" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751165AbZEMFzT (ORCPT ); Wed, 13 May 2009 01:55:19 -0400 Content-Disposition: inline In-Reply-To: <1242119578.11251.321.camel@twins> Sender: netdev-owner@vger.kernel.org List-ID: Hi. On Tue, May 12, 2009 at 11:12:58AM +0200, Peter Zijlstra (a.p.zijlstra@chello.nl) wrote: > Wouldn't the even better solution be to get rid of softirqs > all-together? And move tasklets into some thread context? Only if we are ready to fix 7 times rescheduling regressions compared to kernel threads (work queue actually). At least that's how tasklet behaved compared to work queue 1.5 years ago in the simplest and quite naive test where tasklet/work rescheduled iself number of times: http://marc.info/?l=linux-crypto-vger&m=119462472517405&w=2 -- Evgeniy Polyakov From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from tservice.net.ru (matrixpower.ru [195.178.208.66]) by ozlabs.org (Postfix) with ESMTP id 831A2DDF9B for ; Wed, 13 May 2009 16:15:01 +1000 (EST) Date: Wed, 13 May 2009 09:55:11 +0400 From: Evgeniy Polyakov To: Peter Zijlstra Subject: Re: question about softirqs Message-ID: <20090513055511.GA8529@ioremap.net> References: <18948.63755.279732.294842@cargo.ozlabs.ibm.com> <20090508.234815.127227651.davem@davemloft.net> <4A086DB2.8040703@nortel.com> <20090511.162436.193717082.davem@davemloft.net> <4A08C62F.1050105@nortel.com> <20090512081237.GA16403@elte.hu> <1242119578.11251.321.camel@twins> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <1242119578.11251.321.camel@twins> Cc: netdev@vger.kernel.org, Steven Rostedt , David Miller , linuxppc-dev@ozlabs.org, paulus@samba.org, Ingo Molnar , Thomas Gleixner List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Hi. On Tue, May 12, 2009 at 11:12:58AM +0200, Peter Zijlstra (a.p.zijlstra@chello.nl) wrote: > Wouldn't the even better solution be to get rid of softirqs > all-together? And move tasklets into some thread context? Only if we are ready to fix 7 times rescheduling regressions compared to kernel threads (work queue actually). At least that's how tasklet behaved compared to work queue 1.5 years ago in the simplest and quite naive test where tasklet/work rescheduled iself number of times: http://marc.info/?l=linux-crypto-vger&m=119462472517405&w=2 -- Evgeniy Polyakov