Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756611AbZJVTMZ (ORCPT ); Thu, 22 Oct 2009 15:12:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755639AbZJVTMZ (ORCPT ); Thu, 22 Oct 2009 15:12:25 -0400 Received: from smtp-outbound-2.vmware.com ([65.115.85.73]:44803 "EHLO smtp-outbound-2.vmware.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753202AbZJVTMY convert rfc822-to-8bit (ORCPT ); Thu, 22 Oct 2009 15:12:24 -0400 From: Bela Lubkin To: "'Matt Domsch'" , Arjan van de Ven CC: Randy Dunlap , "Kok, Auke" , Corey Minyard , lkml , "discuss@LessWatts.org" , "openipmi-developer@lists.sourceforge.net" Date: Thu, 22 Oct 2009 12:12:42 -0700 Subject: RE: [Openipmi-developer] [Discuss] [PATCH] ipmi: use round_jiffies on timers to reduce timer overhead/wakeups Thread-Topic: [Openipmi-developer] [Discuss] [PATCH] ipmi: use round_jiffies on timers to reduce timer overhead/wakeups Thread-Index: AcpSx3xGEkG1GjydT6i3pbX/xv/paQAfSEYw Message-ID: References: <20091021102822.5b32b2dc.randy.dunlap@oracle.com> <20091021114210.4d7e1ea9@linux.intel.com> <4ADF57D7.7010808@intel.com> <20091021130348.cd521b0c.randy.dunlap@oracle.com> <4ADF6D76.7070409@acm.org> <4ADF75A2.50202@linux.intel.com> <20091022025013.GB20467@auslistsprd01.us.dell.com> In-Reply-To: <20091022025013.GB20467@auslistsprd01.us.dell.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1881 Lines: 41 Matt Domsch wrote: > Though I'm really curious that HP has a KCS+interrupt controller > available. That gives me hope that the industry-wide problems which > prevented Dell from doing likewise a couple years ago are now > resolved. I'll have my team look into it again. Can you expand on "industry-wide problems"? (Forced to share interrupts with a high rate device? Design your gizmo to support MSI/MSI-x. Add MSI support to ipmi_si if necessary...) As far as I can tell, HP has never shipped an interrupt-less BMC. Their current iLO2 BMC is KCS + interrupt. Their older design was SMIC + interrupt. Why does everyone use KCS when BT is obviously better? Can you have your team look into that as well? (Among the various goals here, I assume that BT -- with a single interrupt and a DMA transfer instead of shuffling bytes over I/O ports -- would cost less power. Not that the members of that list will receive this message: it bounces nonmembers.) It appears that BT designs usually include both BT & KCS programming interfaces to the same BMC. So perhaps there is some increased silicon complexity -- but c'mon, we're talking about a couple of silicon library macros here. Also, I see evidence of some Sun BMCs that have BT without KCS, so apparently it isn't required to pair them. Pairing is probably done for the benefit of certain dumb client software that assumes all BMCs are KCS. I say to heck with that SW. Any app running under an OS that provides an adequate driver [which includes at least Linux, Solaris, and -- I assume -- Win] shouldn't be thinking about the BMC programming interface at all. >Bela<-- 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/