[Fwd: Re: SMP Panic caused by [PATCH] sched: consolidate sched domains]

[James Bottomley <James.Bottomley@SteelEye.com>]

[-- Attachment #1: Type: text/plain, Size: 242 bytes --]

Could we get this in please?  The current screw up in the scheduling
domain patch means that any architecture that actually hotplugs CPUs
will crash in find_busiest_group() ... and I notice this has just bitten
the z Series people...

James


[-- Attachment #2: Forwarded message - Re: SMP Panic caused by [PATCH] sched: consolidate sched domains --]
[-- Type: message/rfc822, Size: 3528 bytes --]

From: William Lee Irwin III <wli@holomorphy.com>
To: James Bottomley <James.Bottomley@steeleye.com>
Cc: Jesse Barnes <jbarnes@engr.sgi.com>, Andrew Morton <akpm@osdl.org>, Linus Torvalds <torvalds@osdl.org>, Matthew Dobson <colpatch@us.ibm.com>, Nick Piggin <nickpiggin@yahoo.com.au>, Linux Kernel <linux-kernel@vger.kernel.org>
Subject: Re: SMP Panic caused by [PATCH] sched: consolidate sched domains
Date: Sun, 29 Aug 2004 10:50:58 -0700
Message-ID: <20040829175058.GP5492@holomorphy.com>

On Sun, Aug 29, 2004 at 10:40:39AM -0700, William Lee Irwin III wrote:
> Okay, if you prefer the #ifdef:

And for the other half of it:


Index: wait-2.6.9-rc1-mm1/kernel/sched.c
===================================================================
--- wait-2.6.9-rc1-mm1.orig/kernel/sched.c	2004-08-28 11:41:47.000000000 -0700
+++ wait-2.6.9-rc1-mm1/kernel/sched.c	2004-08-29 10:46:52.543081208 -0700
@@ -4224,7 +4224,11 @@
 		sd = &per_cpu(phys_domains, i);
 		group = cpu_to_phys_group(i);
 		*sd = SD_CPU_INIT;
+#ifdef CONFIG_NUMA
 		sd->span = nodemask;
+#else
+		sd->span = cpu_possible_map;
+#endif
 		sd->parent = p;
 		sd->groups = &sched_group_phys[group];
 
@@ -4262,6 +4266,7 @@
 						&cpu_to_isolated_group);
 	}
 
+#ifdef CONFIG_NUMA
 	/* Set up physical groups */
 	for (i = 0; i < MAX_NUMNODES; i++) {
 		cpumask_t nodemask = node_to_cpumask(i);
@@ -4273,6 +4278,10 @@
 		init_sched_build_groups(sched_group_phys, nodemask,
 						&cpu_to_phys_group);
 	}
+#else
+	init_sched_build_groups(sched_group_phys, cpu_possible_map,
+							&cpu_to_phys_group);
+#endif
 
 #ifdef CONFIG_NUMA
 	/* Set up node groups */