linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: shaohui.zheng@intel.com
To: akpm@linux-foundation.org, linux-mm@kvack.org
Cc: linux-kernel@vger.kernel.org, haicheng.li@linux.intel.com,
	lethal@linux-sh.org, ak@linux.intel.com,
	shaohui.zheng@linux.intel.com, Sam Ravnborg <sam@ravnborg.org>,
	Shaohui Zheng <shaohui.zheng@intel.com>
Subject: [6/8,v3] NUMA Hotplug Emulator: Fake CPU socket with logical CPU on x86
Date: Wed, 17 Nov 2010 10:08:05 +0800	[thread overview]
Message-ID: <20101117021000.846715655@intel.com> (raw)
In-Reply-To: 20101117020759.016741414@intel.com

[-- Attachment #1: 006-hotplug-emulator-fake_socket_with_logic_cpu_on_x86.patch --]
[-- Type: text/plain, Size: 7938 bytes --]

From: Shaohui Zheng <shaohui.zheng@intel.com>

When hotplug a CPU with emulator, we are using a logical CPU to emulate the
CPU hotplug process. For the CPU supported SMT, some logical CPUs are in the
same socket, but it may located in different NUMA node after we have emulator.
it misleads the scheduling domain to build the incorrect hierarchy, and it
causes the following call trace when rebalance the scheduling domain:

divide error: 0000 [#1] SMP 
last sysfs file: /sys/devices/system/cpu/cpu8/online
CPU 0 
Modules linked in: fbcon tileblit font bitblit softcursor radeon ttm drm_kms_helper e1000e usbhid via_rhine mii drm i2c_algo_bit igb dca
Pid: 0, comm: swapper Not tainted 2.6.32hpe #78 X8DTN
RIP: 0010:[<ffffffff81051da5>]  [<ffffffff81051da5>] find_busiest_group+0x6c5/0xa10
RSP: 0018:ffff880028203c30  EFLAGS: 00010246
RAX: 0000000000000000 RBX: 0000000000015ac0 RCX: 0000000000000000
RDX: 0000000000000000 RSI: ffff880277e8cfa0 RDI: 0000000000000000
RBP: ffff880028203dc0 R08: ffff880277e8cfa0 R09: 0000000000000040
R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
FS:  0000000000000000(0000) GS:ffff880028200000(0000) knlGS:0000000000000000
CS:  0010 DS: 0018 ES: 0018 CR0: 000000008005003b
CR2: 00007f16cfc85770 CR3: 0000000001001000 CR4: 00000000000006f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process swapper (pid: 0, threadinfo ffffffff81822000, task ffffffff8184a600)
Stack:
 ffff880028203d60 ffff880028203cd0 ffff8801c204ff08 ffff880028203e38
<0> 0101ffff81018c59 ffff880028203e44 00000001810806bd ffff8801c204fe00
<0> 0000000528200000 ffffffff00000000 0000000000000018 0000000000015ac0
Call Trace:
 <IRQ> 
 [<ffffffff81088ee0>] ? tick_dev_program_event+0x40/0xd0
 [<ffffffff81053b2c>] rebalance_domains+0x17c/0x570
 [<ffffffff81018c89>] ? read_tsc+0x9/0x20
 [<ffffffff81088ee0>] ? tick_dev_program_event+0x40/0xd0
 [<ffffffff810569ed>] run_rebalance_domains+0xbd/0xf0
 [<ffffffff8106471f>] __do_softirq+0xaf/0x1e0
 [<ffffffff810b7d18>] ? handle_IRQ_event+0x58/0x160
 [<ffffffff810130ac>] call_softirq+0x1c/0x30
 [<ffffffff81014a85>] do_softirq+0x65/0xa0
 [<ffffffff810645cd>] irq_exit+0x7d/0x90
 [<ffffffff81013ff0>] do_IRQ+0x70/0xe0
 [<ffffffff810128d3>] ret_from_intr+0x0/0x11
 <EOI> 
 [<ffffffff8133387f>] ? acpi_idle_enter_bm+0x281/0x2b5
 [<ffffffff81333878>] ? acpi_idle_enter_bm+0x27a/0x2b5
 [<ffffffff8145dc8f>] ? cpuidle_idle_call+0x9f/0x130
 [<ffffffff81010e2b>] ? cpu_idle+0xab/0x100
 [<ffffffff8158aee6>] ? rest_init+0x66/0x70
 [<ffffffff81905d90>] ? start_kernel+0x3e3/0x3ef
 [<ffffffff8190533a>] ? x86_64_start_reservations+0x125/0x129
 [<ffffffff81905438>] ? x86_64_start_kernel+0xfa/0x109
Code: 00 00 e9 4c fb ff ff 0f 1f 80 00 00 00 00 48 8b b5 d8 fe ff ff 48 8b 45 a8 4d 29 ef 8b 56 08 48 c1 e0 0a 49 89 f0 48 89 d7 31 d2 <48> f7 f7 31 d2 48 89 45 a0 8b 76 08 4c 89 f0 48 c1 e0 0a 48 f7 
RIP  [<ffffffff81051da5>] find_busiest_group+0x6c5/0xa10
 RSP <ffff880028203c30>

Solution:

We put the logical CPU into a fake CPU socket, and assign it an unique
 phys_proc_id. For the fake socket, we put one logical CPU in only. This
method fixes the above bug.

CC: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: Shaohui Zheng <shaohui.zheng@intel.com>
---
Index: linux-hpe4/arch/x86/include/asm/processor.h
===================================================================
--- linux-hpe4.orig/arch/x86/include/asm/processor.h	2010-11-17 09:00:51.354100239 +0800
+++ linux-hpe4/arch/x86/include/asm/processor.h	2010-11-17 09:01:10.222837594 +0800
@@ -113,6 +113,15 @@
 	/* Index into per_cpu list: */
 	u16			cpu_index;
 #endif
+
+#ifdef CONFIG_ARCH_CPU_PROBE_RELEASE
+	/*
+	 * Use a logic cpu to emulate a physical cpu's hotplug. We put the
+	 * logical cpu into a fake socket, assign a fake physical id to it,
+	 * and create a fake core.
+	 */
+	__u8		cpu_probe_on; /* A flag to enable cpu probe/release */
+#endif
 } __attribute__((__aligned__(SMP_CACHE_BYTES)));
 
 #define X86_VENDOR_INTEL	0
Index: linux-hpe4/arch/x86/kernel/smpboot.c
===================================================================
--- linux-hpe4.orig/arch/x86/kernel/smpboot.c	2010-11-17 09:01:10.202837209 +0800
+++ linux-hpe4/arch/x86/kernel/smpboot.c	2010-11-17 09:01:10.222837594 +0800
@@ -97,6 +97,7 @@
  */
 static DEFINE_MUTEX(x86_cpu_hotplug_driver_mutex);
 
+#ifdef CONFIG_ARCH_CPU_PROBE_RELEASE
 void cpu_hotplug_driver_lock()
 {
         mutex_lock(&x86_cpu_hotplug_driver_mutex);
@@ -106,6 +107,7 @@
 {
         mutex_unlock(&x86_cpu_hotplug_driver_mutex);
 }
+#endif
 
 #else
 static struct task_struct *idle_thread_array[NR_CPUS] __cpuinitdata ;
@@ -198,6 +200,8 @@
 {
 	int cpuid, phys_id;
 	unsigned long timeout;
+	u8 cpu_probe_on = 0;
+	struct cpuinfo_x86 *c;
 
 	/*
 	 * If waken up by an INIT in an 82489DX configuration
@@ -277,7 +281,20 @@
 	/*
 	 * Save our processor parameters
 	 */
+	c = &cpu_data(cpuid);
+#ifdef CONFIG_ARCH_CPU_PROBE_RELEASE
+	cpu_probe_on = c->cpu_probe_on;
+	phys_id = c->phys_proc_id;
+#endif
+
 	smp_store_cpu_info(cpuid);
+#ifdef CONFIG_ARCH_CPU_PROBE_RELEASE
+	if (cpu_probe_on) {
+		c->phys_proc_id = phys_id; /* restore the fake phys_proc_id */
+		c->cpu_core_id = 0; /* force the logical cpu to core 0 */
+		c->cpu_probe_on = cpu_probe_on;
+	}
+#endif
 
 	notify_cpu_starting(cpuid);
 
@@ -400,6 +417,11 @@
 {
 	int i;
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
+	int cpu_probe_on = 0;
+
+#ifdef CONFIG_ARCH_CPU_PROBE_RELEASE
+	cpu_probe_on = c->cpu_probe_on;
+#endif
 
 	cpumask_set_cpu(cpu, cpu_sibling_setup_mask);
 
@@ -431,7 +453,8 @@
 
 	for_each_cpu(i, cpu_sibling_setup_mask) {
 		if (per_cpu(cpu_llc_id, cpu) != BAD_APICID &&
-		    per_cpu(cpu_llc_id, cpu) == per_cpu(cpu_llc_id, i)) {
+		    per_cpu(cpu_llc_id, cpu) == per_cpu(cpu_llc_id, i) &&
+			cpu_probe_on == 0) {
 			cpumask_set_cpu(i, c->llc_shared_map);
 			cpumask_set_cpu(cpu, cpu_data(i).llc_shared_map);
 		}
Index: linux-hpe4/arch/x86/kernel/topology.c
===================================================================
--- linux-hpe4.orig/arch/x86/kernel/topology.c	2010-11-17 09:01:10.202837209 +0800
+++ linux-hpe4/arch/x86/kernel/topology.c	2010-11-17 09:01:10.222837594 +0800
@@ -70,6 +70,36 @@
 }
 EXPORT_SYMBOL(arch_unregister_cpu);
 
+#ifdef CONFIG_ARCH_CPU_PROBE_RELEASE
+/*
+ * Put the logical cpu into a new sokect, and encapsule it into core 0.
+ */
+static void fake_cpu_socket_info(int cpu)
+{
+	struct cpuinfo_x86 *c = &cpu_data(cpu);
+	int i, phys_id = 0;
+
+	/* calculate the max phys_id */
+	for_each_present_cpu(i) {
+		struct cpuinfo_x86 *c = &cpu_data(i);
+		if (phys_id < c->phys_proc_id)
+			phys_id = c->phys_proc_id;
+	}
+
+	c->phys_proc_id = phys_id + 1; /* pick up a unused phys_proc_id */
+	c->cpu_core_id = 0; /* always put the logical cpu to core 0 */
+	c->cpu_probe_on = 1;
+}
+
+static void clear_cpu_socket_info(int cpu)
+{
+	struct cpuinfo_x86 *c = &cpu_data(cpu);
+	c->phys_proc_id = 0;
+	c->cpu_core_id = 0;
+	c->cpu_probe_on = 0;
+}
+
+
 ssize_t arch_cpu_probe(const char *buf, size_t count)
 {
 	int nid = 0;
@@ -109,6 +139,7 @@
 	/* register cpu */
 	arch_register_cpu_node(selected, nid);
 	acpi_map_lsapic_emu(selected, nid);
+	fake_cpu_socket_info(selected);
 
 	return count;
 }
@@ -132,10 +163,13 @@
 
 	arch_unregister_cpu(cpu);
 	acpi_unmap_lsapic(cpu);
+	clear_cpu_socket_info(cpu);
+	set_cpu_present(cpu, true);
 
 	return count;
 }
 EXPORT_SYMBOL(arch_cpu_release);
+#endif CONFIG_ARCH_CPU_PROBE_RELEASE
 
 #else /* CONFIG_HOTPLUG_CPU */
 

-- 
Thanks & Regards,
Shaohui


--
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/ .
Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2010-11-17  4:45 UTC|newest]

Thread overview: 70+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-17  2:07 [0/8,v3] NUMA Hotplug Emulator - Introduction & Feedbacks shaohui.zheng
2010-11-17  2:08 ` [1/8,v3] NUMA Hotplug Emulator: add function to hide memory region via e820 table shaohui.zheng
2010-11-17  8:16   ` David Rientjes
2010-11-18  9:20     ` Shaohui Zheng
2010-11-18 21:16       ` David Rientjes
2010-11-19  0:12         ` Shaohui Zheng
2010-11-21  0:45           ` David Rientjes
2010-11-21 14:00             ` Américo Wang
2010-11-21 21:33               ` David Rientjes
2010-11-17  2:08 ` [2/8,v3] NUMA Hotplug Emulator: infrastructure of NUMA hotplug emulation shaohui.zheng
2010-11-17  8:16   ` David Rientjes
2010-11-17  7:51     ` Shaohui Zheng
2010-11-17 21:10       ` David Rientjes
2010-11-18  4:14         ` Shaohui Zheng
2010-11-18  6:27           ` Paul Mundt
2010-11-18  5:27             ` Shaohui Zheng
2010-11-18 21:24               ` David Rientjes
2010-11-19  0:32                 ` Shaohui Zheng
2010-11-21  0:48                   ` David Rientjes
2010-11-21  2:28                     ` [patch 1/2] x86: add numa=possible command line option David Rientjes
2010-11-21  2:28                       ` [patch 2/2] mm: add node hotplug emulation David Rientjes
2010-11-21 17:34                         ` Greg KH
2010-11-21 21:48                           ` David Rientjes
2010-11-21 23:08                             ` [patch 2/2 v2] " David Rientjes
2010-11-22  0:56                               ` Greg KH
2010-11-28  1:52                                 ` David Rientjes
2010-11-28  5:17                                   ` Greg KH
2010-11-30  0:04                                     ` David Rientjes
2010-11-21 14:26                       ` [patch 1/2] x86: add numa=possible command line option Américo Wang
2010-11-21 21:46                         ` David Rientjes
2010-11-22 15:43                           ` Américo Wang
2010-11-21 15:14                     ` [2/8,v3] NUMA Hotplug Emulator: infrastructure of NUMA hotplug emulation Li, Haicheng
2010-11-21 21:42                       ` David Rientjes
2010-11-18 21:19           ` David Rientjes
2010-11-17  2:08 ` [3/8,v3] NUMA Hotplug Emulator: Userland interface to hotplug-add fake offlined nodes shaohui.zheng
2010-11-17  8:16   ` David Rientjes
2010-11-17  2:08 ` [4/8,v3] NUMA Hotplug Emulator: Abstract cpu register functions shaohui.zheng
2010-11-17  2:08 ` [5/8,v3] NUMA Hotplug Emulator: support cpu probe/release in x86 shaohui.zheng
2010-11-21 14:45   ` Américo Wang
2010-11-22  0:01     ` Shaohui Zheng
2010-11-22 15:51       ` Américo Wang
2010-11-22 23:29         ` Shaohui Zheng
2010-11-17  2:08 ` shaohui.zheng [this message]
2010-11-17  2:08 ` [7/8,v3] NUMA Hotplug Emulator: extend memory probe interface to support NUMA shaohui.zheng
2010-11-17 18:50   ` Dave Hansen
2010-11-17 21:18     ` David Rientjes
2010-11-17 21:55       ` Dave Hansen
2010-11-17 22:44         ` David Rientjes
2010-11-17 23:00           ` Dave Hansen
2010-11-17 23:17             ` David Rientjes
2010-11-18 16:59           ` Aaron Durbin
2010-11-18  4:48       ` Shaohui Zheng
2010-11-18  6:24         ` Paul Mundt
2010-11-18 21:28           ` David Rientjes
2010-11-18 21:31         ` David Rientjes
2010-11-18  4:36     ` Shaohui Zheng
2010-11-19  7:51     ` Shaohui Zheng
2010-11-19 16:36       ` Dave Hansen
2010-11-17  2:08 ` [8/8,v3] NUMA Hotplug Emulator: documentation shaohui.zheng
2010-11-17 23:06   ` Randy Dunlap
2010-11-18  2:31     ` Shaohui Zheng
2010-11-21 15:03   ` Américo Wang
2010-11-21 15:16     ` Li, Haicheng
2010-11-21 23:33     ` Shaohui Zheng
2010-11-22 16:04       ` Américo Wang
2010-11-22 23:23         ` Shaohui Zheng
2010-11-17  5:22 ` [0/8,v3] NUMA Hotplug Emulator - Introduction & Feedbacks Paul Mundt
2010-11-19  5:54   ` Shaohui Zheng
2010-11-17  9:26 ` Yinghai Lu
2010-11-18  2:03   ` Shaohui Zheng

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20101117021000.846715655@intel.com \
    --to=shaohui.zheng@intel.com \
    --cc=ak@linux.intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=haicheng.li@linux.intel.com \
    --cc=lethal@linux-sh.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=sam@ravnborg.org \
    --cc=shaohui.zheng@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).