Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751766AbZL0May (ORCPT ); Sun, 27 Dec 2009 07:30:54 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751559AbZL0Max (ORCPT ); Sun, 27 Dec 2009 07:30:53 -0500 Received: from crmm.lgl.lu ([158.64.72.228]:59223 "EHLO lll.lu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751515AbZL0Max (ORCPT ); Sun, 27 Dec 2009 07:30:53 -0500 Message-ID: <4B3752F5.3080205@knaff.lu> Date: Sun, 27 Dec 2009 13:28:37 +0100 From: Alain Knaff User-Agent: Thunderbird 2.0.0.23 (X11/20090817) MIME-Version: 1.0 To: Andi Kleen CC: Arjan van de Ven , Linus Torvalds , Mark Hounschell , "Pallipadi, Venkatesh" , "dmarkh@cfl.rr.com" , Linux Kernel Mailing List , "fdutils@fdutils.linux.lu" , "Li, Shaohua" , Ingo Molnar , morgan@physics.ucla.edu, JONES@UINPLA.NPL.UIUC.EDU Subject: Re: [Fdutils] DMA cache consistency bug introduced in 2.6.28 References: <4B3214EC.6020308@compro.net> <6598A4E21F1DB24D80BF72956484F59802EFD1C6@orsmsx001.amr.corp.intel.com> <4B32386B.2060509@compro.net> <87bphpd4rt.fsf@basil.nowhere.org> <20091223170832.GH20539@basil.fritz.box> <20091225132116.6c9089f2@infradead.org> <20091225203304.GA22306@basil.fritz.box> <20091226103835.68479100@infradead.org> <20091226164029.GA10256@basil.fritz.box> In-Reply-To: <20091226164029.GA10256@basil.fritz.box> X-Enigmail-Version: 0.95.7 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1301 Lines: 37 Andi Kleen wrote: >>> Does pmqos work with apci=off etc.? >> yes >> >>> I didn't think it shut down >>> the classic "HLT" idle, does it? >> it does if you specify a latency of 0; it will then go into the >> spin-only state until you give up your latency requirement > > I looked at it this evening, but it seems like pm_qos is not > interrupt safe (e.g. calls blocking notifiers) and floppy currently does > enable/disable_hlt from interrupts and bottom halves. > > Would need some more infrastructure work or restructuring > of the floppy driver. > > -Andi disable_hlt/enable_hlt was only needed to work around a bug on TM4000 (Texas Instrument) Laptops which were popular around 1994 / 1995. Basically, as soon as the CPU went into hlt() state, so did the DMA controller, either causing a really slow transfer, or (worse) a buffer over/underrun which failed the operation. On hardware unaffected by this particular bug (which would be most hardware around now, 14 years after the fact...), these calls can safely be removed. Regards, Alain -- 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/