Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755475Ab0DOGMJ (ORCPT ); Thu, 15 Apr 2010 02:12:09 -0400 Received: from ey-out-2122.google.com ([74.125.78.24]:38744 "EHLO ey-out-2122.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755293Ab0DOGMF (ORCPT ); Thu, 15 Apr 2010 02:12:05 -0400 DomainKey-Signature: a=rsa-sha1; c=nofws; d=googlemail.com; s=gamma; h=date:from:to:cc:subject:message-id:mime-version:content-type :content-disposition:user-agent; b=xaBBwNLFsWL/xImmfNElqGsMH32n2phLslL8X6VGLjxId9zq1a6NZ8j4gAOrqZNZF4 CG+gAITT+K8gJ0mh8mhAmh17w0cXsndjJnclr/ZfnuAxJYGIwTGGgpuoyuBgZ22be7uv M5Tc2+d/10T1v34aYvlOrywi9zshte71Bd7g4= Date: Thu, 15 Apr 2010 08:12:00 +0200 From: Andreas Herrmann To: Christoph Lameter , Lee Schermerhorn Cc: Andi Kleen , Ingo Molnar , Greg Kroah-Hartman , linux-kernel@vger.kernel.org Subject: useless node/has_cpu sysfs attribute Message-ID: <20100415061200.GC21517@alberich.amd.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1378 Lines: 44 Hi, commit 99dcc3e5a94ed491fbef402831d8c0bbb267f995 (this_cpu: Page allocator conversion) removed the call node_set_state(node, N_CPU); /* this node has a cpu */ (in fact process_zones() where this call happened was removed). Thus the sysfs attribute /sys/devices/system/node/has_cpu doesn't show anything anymore. Before this commit the attribute usually listed all nodes ("having a CPU") in the system, e.g. worms linux # cat /sys/devices/system/node/has_cpu 0-7 This attribute never reflected any changes triggered by CPU hotplug. I.e. you could set all CPUs of a node offline but the node still was listed in this attribute. (because no hotplug notifiers exist that would call node_set_state(node, N_CPU)/node_clear_state(node,N_CPU)). Do we need this attribute? Who relies on information from this attribute? I think if we do not need the attribute we should remove it. But if there is use for it we need to fix it and make it CPU hotplug aware. Opinions? Thanks, Andreas PS: The attribute was introduced with commit bde631a51876f23e9bbdce43f02b7232502c151e (mm: add node states sysfs class attributeS) -- 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/