x86: SMP topology warning with HP ProLiant DL385 G7

[Thomas Zimmermann <tzimmermann@suse.de>]

Hi,

I get a kernel error about the topology begin incorrect. See below for 
the message.

[    1.400582] ------------[ cut here ]------------
[    1.400582] sched: CPU #8's smt-sibling CPU #0 is not on the same 
node! [node: 1 != 0]. Ignoring dependency.
[    1.400582] WARNING: arch/x86/kernel/smpboot.c:329 at 
topology_sane.isra.0+0x161/0x310, CPU#8: swapper/8/0
[    1.400582] Modules linked in:
[    1.400582] CPU: 8 UID: 0 PID: 0 Comm: swapper/8 Not tainted 
7.0.0-rc7-1-default+ #4879 PREEMPT(full)
[    1.400582] Hardware name: HP ProLiant DL385 G7, BIOS A18 12/20/2010
[    1.400582] RIP: 0010:topology_sane.isra.0+0x16a/0x310
[    1.400582] Code: 48 89 f2 83 e2 07 83 c2 03 40 38 fa 7c 09 40 84 ff 
0f 85 ec 00 00 00 88 04 24 48 8d 3d ef ea aa 04 44 8b 06 4c 89 da 44 89 
d6 <67> 48 0f b9 3a 0f b6 04 24 48 83 c4 30 5b 5d c3 cc cc cc cc 489
[    1.400582] RSP: 0000:ffff8882401dfdf0 EFLAGS: 00010046
[    1.400582] RAX: ffff8881d6405e00 RBX: ffff8881d6405e00 RCX: 
0000000000000000
[    1.400582] RDX: ffffffff8d65d9e0 RSI: 0000000000000008 RDI: 
ffffffff8ef22c40
[    1.400582] RBP: ffff888246370000 R08: 0000000000000001 R09: 
0000000000000000
[    1.400582] R10: 0000000000000008 R11: ffffffff8d65d9e0 R12: 
ffff888339e2f0e0
[    1.400582] R13: dffffc0000000000 R14: 0000000000000000 R15: 
ffffffff8febf1ec
[    1.400582] FS:  0000000000000000(0000) GS:ffff8883a9f70000(0000) 
knlGS:0000000000000000
[    1.400582] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    1.400582] CR2: 0000000000000000 CR3: 00000002c8496000 CR4: 
00000000000006f0
[    1.400582] Call Trace:
[    1.400582]  <TASK>
[    1.400582]  set_cpu_sibling_map+0x1286/0x1c70
[    1.400582]  ? get_cpu_address_sizes+0x313/0x320
[    1.400582]  start_secondary+0xc5/0x290
[    1.400582]  ? __pfx_start_secondary+0x10/0x10
[    1.400582]  ? soft_restart_cpu+0x14/0x14
[    1.400582]  common_startup_64+0x13e/0x141
[    1.400582]  </TASK>
[    1.400582] irq event stamp: 0
[    1.400582] hardirqs last  enabled at (0): [<0000000000000000>] 0x0
[    1.400582] hardirqs last disabled at (0): [<ffffffff8a5182d7>] 
copy_process+0x1637/0x5400
[    1.400582] softirqs last  enabled at (0): [<ffffffff8a51832f>] 
copy_process+0x168f/0x5400
[    1.400582] softirqs last disabled at (0): [<0000000000000000>] 0x0
[    1.400582] ---[ end trace 0000000000000000 ]---

There are also plenty of these warnings nearby:

[    1.400582] [Firmware Bug]: CPU   1: APIC ID mismatch. CPUID: 0x0010 
APIC: 0x0020
[    1.400582] [Firmware Bug]: CPU   2: APIC ID mismatch. CPUID: 0x0001 
APIC: 0x0011
[    1.400582] [Firmware Bug]: CPU   3: APIC ID mismatch. CPUID: 0x0011 
APIC: 0x0021
[    1.400582] [Firmware Bug]: CPU   4: APIC ID mismatch. CPUID: 0x0002 
APIC: 0x0012
[    1.400582] [Firmware Bug]: CPU   5: APIC ID mismatch. CPUID: 0x0012 
APIC: 0x0022
[    1.400582] [Firmware Bug]: CPU   6: APIC ID mismatch. CPUID: 0x0003 
APIC: 0x0013
[    1.400582] [Firmware Bug]: CPU   7: APIC ID mismatch. CPUID: 0x0013 
APIC: 0x0023
[    1.400582] [Firmware Bug]: CPU   8: APIC ID mismatch. CPUID: 0x0004 
APIC: 0x0014

Seen with the 7.0-rc7, but probably happens with any other Linux version 
as well. Let me know, if I can help resolving this issue.

Best regards
Thomas


-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)