From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Mon, 12 Nov 2007 11:17:39 -0800 (PST) From: Christoph Lameter Subject: Re: Page allocator: Clean up pcp draining functions In-Reply-To: <20071112160451.GC6653@skynet.ie> Message-ID: References: <20071112160451.GC6653@skynet.ie> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-linux-mm@kvack.org Return-Path: To: Mel Gorman Cc: akpm@linux-foundation.org, linux-mm@kvack.org, Yasunori Goto , "Rafael J. Wysocki" List-ID: On Mon, 12 Nov 2007, Mel Gorman wrote: > Reflecting the comment, perhaps the following would not hurt? > > VM_BUG_ON(cpu != smp_processor_id() && cpu_online(cpu)) Well we need to check first with the hotplug developers if the cpu is already marked off line when this function is called. > > if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) { > > - local_irq_disable(); > > - __drain_pages(cpu); > > + drain_pages(cpu); > > + > > + /* > > + * Spill the event counters of the dead processor > > + * into the current processors event counters. > > + * This artificially elevates the count of the current > > + * processor. > > + */ > > This comment addition does not appear to be related to the rest of the > patch. Its related to the action of vm_events_fold_cpu which is not that unproblematic since the numbers indicate now that more events occurred on this processor than what actually occurred. > Acked-by: Mel Gorman Thanks. -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org