From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ingo Molnar Subject: Re: [patch 0/2] Run interrupt handlers always with interrupts disabled Date: Fri, 26 Mar 2010 10:20:23 +0100 Message-ID: <20100326092023.GH27394@elte.hu> References: <20100326000325.917127328@linutronix.de> <20100326081457.GA23570@flint.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mx2.mail.elte.hu ([157.181.151.9]:46067 "EHLO mx2.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751438Ab0CZJWm (ORCPT ); Fri, 26 Mar 2010 05:22:42 -0400 Content-Disposition: inline In-Reply-To: <20100326081457.GA23570@flint.arm.linux.org.uk> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Thomas Gleixner , Nicolas Pitre , Linus Torvalds , LKML , linux-arch@vger.kernel.org, Andrew * Russell King wrote: > On Fri, Mar 26, 2010 at 12:06:44AM -0000, Thomas Gleixner wrote: > > The following patch series removes the IRQF_DISABLED functionality > > from the core interrupt code and runs all interrupt handlers with > > interrupts disabled. > > As was covered in previous discussions, what about drivers such as SMC91x > which take a long time to retrieve packets from the hardware? Always running > handlers with IRQs disabled will kill things such as serial on these > platforms. As long as it's rare (which it is) i dont see a problem: you can enable interrupts in the handler by using local_irq_enable(), like the IDE PIO drivers do. That way it's documented a bit better as well, because it shows the precise source of the latency, with a big comment explaining it, etc. Thanks, Ingo From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx2.mail.elte.hu ([157.181.151.9]:46067 "EHLO mx2.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751438Ab0CZJWm (ORCPT ); Fri, 26 Mar 2010 05:22:42 -0400 Date: Fri, 26 Mar 2010 10:20:23 +0100 From: Ingo Molnar Subject: Re: [patch 0/2] Run interrupt handlers always with interrupts disabled Message-ID: <20100326092023.GH27394@elte.hu> References: <20100326000325.917127328@linutronix.de> <20100326081457.GA23570@flint.arm.linux.org.uk> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20100326081457.GA23570@flint.arm.linux.org.uk> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Thomas Gleixner , Nicolas Pitre , Linus Torvalds , LKML , linux-arch@vger.kernel.org, Andrew Morton , Peter Zijlstra , Alan Cox , Andi Kleen , David Miller , Greg Kroah-Hartman , Arnaldo Carvalho de Melo , Russell King Message-ID: <20100326092023.P2h2NiIucHxeH7y7I7H8aAe9cHocuO4EfuA33T97GSM@z> * Russell King wrote: > On Fri, Mar 26, 2010 at 12:06:44AM -0000, Thomas Gleixner wrote: > > The following patch series removes the IRQF_DISABLED functionality > > from the core interrupt code and runs all interrupt handlers with > > interrupts disabled. > > As was covered in previous discussions, what about drivers such as SMC91x > which take a long time to retrieve packets from the hardware? Always running > handlers with IRQs disabled will kill things such as serial on these > platforms. As long as it's rare (which it is) i dont see a problem: you can enable interrupts in the handler by using local_irq_enable(), like the IDE PIO drivers do. That way it's documented a bit better as well, because it shows the precise source of the latency, with a big comment explaining it, etc. Thanks, Ingo