From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-pg0-f66.google.com ([74.125.83.66]:34670 "EHLO mail-pg0-f66.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751266AbcLOMef (ORCPT ); Thu, 15 Dec 2016 07:34:35 -0500 Subject: Re: [PATCH] genirq/affinity: fix node generation from cpumask To: "Guilherme G. Piccoli" , tglx@linutronix.de, linux-kernel@vger.kernel.org References: <1481738472-2671-1-git-send-email-gpiccoli@linux.vnet.ibm.com> Cc: linux-pci@vger.kernel.org, hch@lst.de, linuxppc-dev@lists.ozlabs.org, gabriel@krisman.be From: Balbir Singh Message-ID: Date: Thu, 15 Dec 2016 23:34:21 +1100 MIME-Version: 1.0 In-Reply-To: <1481738472-2671-1-git-send-email-gpiccoli@linux.vnet.ibm.com> Content-Type: text/plain; charset=windows-1252 Sender: linux-pci-owner@vger.kernel.org List-ID: On 15/12/16 05:01, Guilherme G. Piccoli wrote: > Commit 34c3d9819fda ("genirq/affinity: Provide smarter irq spreading > infrastructure") introduced a better IRQ spreading mechanism, taking > account of the available NUMA nodes in the machine. > > Problem is that the algorithm of retrieving the nodemask iterates > "linearly" based on the number of online nodes - some architectures > present non-linear node distribution among the nodemask, like PowerPC. > If this is the case, the algorithm lead to a wrong node count number > and therefore to a bad/incomplete IRQ affinity distribution. > > For example, this problem were found in a machine with 128 CPUs and two > nodes, namely nodes 0 and 8 (instead of 0 and 1, if it was linearly > distributed). This led to a wrong affinity distribution which then led to > a bad mq allocation for nvme driver. > > Finally, we take the opportunity to fix a comment regarding the affinity > distribution when we have _more_ nodes than vectors. Very good catch! Acked-by: Balbir Singh