From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1765225AbYASEnm (ORCPT ); Fri, 18 Jan 2008 23:43:42 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1761410AbYASEne (ORCPT ); Fri, 18 Jan 2008 23:43:34 -0500 Received: from rv-out-0910.google.com ([209.85.198.185]:59388 "EHLO rv-out-0910.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760956AbYASEnd (ORCPT ); Fri, 18 Jan 2008 23:43:33 -0500 DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:to:subject:cc:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=jYzniKzRvBqoY6akJAaNOQbHz9JWoEkiGgr5YOe1Fo8S9WKY+DyuMgFxQow9NlaTUvkdYZJFJXRLMEN97sb0xJVsC1mp5zH/hxFy15G8EpE/lE0REP8VTtCDXh6V8Sr3d64bIR/agnSADUzkclaa87d4UuKqwY5TwtVom4fMkxk= Message-ID: <86802c440801182043l1f36086bq51d1fa0528e6bd74@mail.gmail.com> Date: Fri, 18 Jan 2008 20:43:33 -0800 From: "Yinghai Lu" To: "David Rientjes" Subject: Re: [PATCH 1/5] x86: Change size of node ids from u8 to u16 fixup Cc: travis@sgi.com, "Andrew Morton" , "Andi Kleen" , mingo@elte.hu, "Christoph Lameter" , linux-mm@kvack.org, linux-kernel@vger.kernel.org, "Eric Dumazet" In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline References: <20080118183011.354965000@sgi.com> <20080118183011.527888000@sgi.com> <86802c440801182003vd94044ex7fb13e61e5f79c81@mail.gmail.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Jan 18, 2008 8:36 PM, David Rientjes wrote: > On Fri, 18 Jan 2008, Yinghai Lu wrote: > > > > +#if MAX_NUMNODES > 256 > > > +typedef u16 numanode_t; > > > +#else > > > +typedef u8 numanode_t; > > > +#endif > > > + > > > #endif /* _LINUX_NUMA_H */ > > > > that is wrong, you can not change pxm_to_node_map from int to u8 or u16. > > > > Yeah, NID_INVAL is negative so no unsigned type will work here, > unfortunately. And that reduces the intended savings of your change since > the smaller type can only be used with a smaller CONFIG_NODES_SHIFT. > > > int acpi_map_pxm_to_node(int pxm) > > { > > int node = pxm_to_node_map[pxm]; > > > > if (node < 0){ > > if (nodes_weight(nodes_found_map) >= MAX_NUMNODES) > > return NID_INVAL; > > node = first_unset_node(nodes_found_map); > > __acpi_map_pxm_to_node(pxm, node); > > node_set(node, nodes_found_map); > > } > > > > return node; > > } > > > > node will will be always 255 or 65535 > > > > Right. > > > please keep that to int. > > > > I got > > SART: PXM 0 -> APIC 0 -> Node 255 > > SART: PXM 0 -> APIC 1 -> Node 255 > > SART: PXM 1 -> APIC 2 -> Node 255 > > SART: PXM 1 -> APIC 3 -> Node 255 > > > > I assume this is a typo and those proximity mappings are actually from the > SRAT. SRAT for processor only have PXM and APIC id. setup_node(pxm) will get node id for pxm, start from 0... > > if (node < 0){ > > if (nodes_weight(nodes_found_map) >= MAX_NUMNODES) > > return NID_INVAL; > > node = first_unset_node(nodes_found_map); > > __acpi_map_pxm_to_node(pxm, node); > > node_set(node, nodes_found_map); > > } YH