blob: e05278087ffaa04c90afd11bf85ff5629f814507 [file] [log] [blame]
Ashok Rajc8094062006-01-08 01:03:17 -08001 CPU hotplug Support in Linux(tm) Kernel
2
3 Maintainers:
4 CPU Hotplug Core:
5 Rusty Russell <rusty@rustycorp.com.au>
6 Srivatsa Vaddagiri <vatsa@in.ibm.com>
7 i386:
8 Zwane Mwaikambo <zwane@arm.linux.org.uk>
9 ppc64:
10 Nathan Lynch <nathanl@austin.ibm.com>
11 Joel Schopp <jschopp@austin.ibm.com>
12 ia64/x86_64:
13 Ashok Raj <ashok.raj@intel.com>
14
15Authors: Ashok Raj <ashok.raj@intel.com>
16Lots of feedback: Nathan Lynch <nathanl@austin.ibm.com>,
17 Joel Schopp <jschopp@austin.ibm.com>
18
19Introduction
20
21Modern advances in system architectures have introduced advanced error
22reporting and correction capabilities in processors. CPU architectures permit
23partitioning support, where compute resources of a single CPU could be made
24available to virtual machine environments. There are couple OEMS that
25support NUMA hardware which are hot pluggable as well, where physical
26node insertion and removal require support for CPU hotplug.
27
28Such advances require CPUs available to a kernel to be removed either for
29provisioning reasons, or for RAS purposes to keep an offending CPU off
30system execution path. Hence the need for CPU hotplug support in the
31Linux kernel.
32
33A more novel use of CPU-hotplug support is its use today in suspend
34resume support for SMP. Dual-core and HT support makes even
35a laptop run SMP kernels which didn't support these methods. SMP support
36for suspend/resume is a work in progress.
37
38General Stuff about CPU Hotplug
39--------------------------------
40
41Command Line Switches
42---------------------
43maxcpus=n Restrict boot time cpus to n. Say if you have 4 cpus, using
44 maxcpus=2 will only boot 2. You can choose to bring the
45 other cpus later online, read FAQ's for more info.
46
Ashok Raj8f8b11382006-02-16 14:01:48 -080047additional_cpus*=n Use this to limit hotpluggable cpus. This option sets
Ashok Rajc8094062006-01-08 01:03:17 -080048 cpu_possible_map = cpu_present_map + additional_cpus
49
Ashok Raj8f8b11382006-02-16 14:01:48 -080050(*) Option valid only for following architectures
51- x86_64, ia64
52
53ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT
54to determine the number of potentially hot-pluggable cpus. The implementation
55should only rely on this to count the #of cpus, but *MUST* not rely on the
56apicid values in those tables for disabled apics. In the event BIOS doesnt
57mark such hot-pluggable cpus as disabled entries, one could use this
58parameter "additional_cpus=x" to represent those cpus in the cpu_possible_map.
59
60
Ashok Rajc8094062006-01-08 01:03:17 -080061CPU maps and such
62-----------------
63[More on cpumaps and primitive to manipulate, please check
64include/linux/cpumask.h that has more descriptive text.]
65
66cpu_possible_map: Bitmap of possible CPUs that can ever be available in the
67system. This is used to allocate some boot time memory for per_cpu variables
68that aren't designed to grow/shrink as CPUs are made available or removed.
69Once set during boot time discovery phase, the map is static, i.e no bits
70are added or removed anytime. Trimming it accurately for your system needs
71upfront can save some boot time memory. See below for how we use heuristics
72in x86_64 case to keep this under check.
73
74cpu_online_map: Bitmap of all CPUs currently online. Its set in __cpu_up()
75after a cpu is available for kernel scheduling and ready to receive
76interrupts from devices. Its cleared when a cpu is brought down using
77__cpu_disable(), before which all OS services including interrupts are
78migrated to another target CPU.
79
80cpu_present_map: Bitmap of CPUs currently present in the system. Not all
81of them may be online. When physical hotplug is processed by the relevant
82subsystem (e.g ACPI) can change and new bit either be added or removed
83from the map depending on the event is hot-add/hot-remove. There are currently
84no locking rules as of now. Typical usage is to init topology during boot,
85at which time hotplug is disabled.
86
87You really dont need to manipulate any of the system cpu maps. They should
88be read-only for most use. When setting up per-cpu resources almost always use
89cpu_possible_map/for_each_cpu() to iterate.
90
91Never use anything other than cpumask_t to represent bitmap of CPUs.
92
93#include <linux/cpumask.h>
94
95for_each_cpu - Iterate over cpu_possible_map
96for_each_online_cpu - Iterate over cpu_online_map
97for_each_present_cpu - Iterate over cpu_present_map
98for_each_cpu_mask(x,mask) - Iterate over some random collection of cpu mask.
99
100#include <linux/cpu.h>
101lock_cpu_hotplug() and unlock_cpu_hotplug():
102
103The above calls are used to inhibit cpu hotplug operations. While holding the
104cpucontrol mutex, cpu_online_map will not change. If you merely need to avoid
105cpus going away, you could also use preempt_disable() and preempt_enable()
106for those sections. Just remember the critical section cannot call any
107function that can sleep or schedule this process away. The preempt_disable()
108will work as long as stop_machine_run() is used to take a cpu down.
109
110CPU Hotplug - Frequently Asked Questions.
111
112Q: How to i enable my kernel to support CPU hotplug?
113A: When doing make defconfig, Enable CPU hotplug support
114
115 "Processor type and Features" -> Support for Hotpluggable CPUs
116
117Make sure that you have CONFIG_HOTPLUG, and CONFIG_SMP turned on as well.
118
119You would need to enable CONFIG_HOTPLUG_CPU for SMP suspend/resume support
120as well.
121
122Q: What architectures support CPU hotplug?
123A: As of 2.6.14, the following architectures support CPU hotplug.
124
125i386 (Intel), ppc, ppc64, parisc, s390, ia64 and x86_64
126
127Q: How to test if hotplug is supported on the newly built kernel?
128A: You should now notice an entry in sysfs.
129
130Check if sysfs is mounted, using the "mount" command. You should notice
131an entry as shown below in the output.
132
133....
134none on /sys type sysfs (rw)
135....
136
137if this is not mounted, do the following.
138
139#mkdir /sysfs
140#mount -t sysfs sys /sys
141
142now you should see entries for all present cpu, the following is an example
143in a 8-way system.
144
145#pwd
146#/sys/devices/system/cpu
147#ls -l
148total 0
149drwxr-xr-x 10 root root 0 Sep 19 07:44 .
150drwxr-xr-x 13 root root 0 Sep 19 07:45 ..
151drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu0
152drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu1
153drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu2
154drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu3
155drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu4
156drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu5
157drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu6
158drwxr-xr-x 3 root root 0 Sep 19 07:48 cpu7
159
160Under each directory you would find an "online" file which is the control
161file to logically online/offline a processor.
162
163Q: Does hot-add/hot-remove refer to physical add/remove of cpus?
164A: The usage of hot-add/remove may not be very consistently used in the code.
165CONFIG_CPU_HOTPLUG enables logical online/offline capability in the kernel.
166To support physical addition/removal, one would need some BIOS hooks and
167the platform should have something like an attention button in PCI hotplug.
168CONFIG_ACPI_HOTPLUG_CPU enables ACPI support for physical add/remove of CPUs.
169
170Q: How do i logically offline a CPU?
171A: Do the following.
172
173#echo 0 > /sys/devices/system/cpu/cpuX/online
174
175once the logical offline is successful, check
176
177#cat /proc/interrupts
178
179you should now not see the CPU that you removed. Also online file will report
180the state as 0 when a cpu if offline and 1 when its online.
181
182#To display the current cpu state.
183#cat /sys/devices/system/cpu/cpuX/online
184
185Q: Why cant i remove CPU0 on some systems?
186A: Some architectures may have some special dependency on a certain CPU.
187
188For e.g in IA64 platforms we have ability to sent platform interrupts to the
189OS. a.k.a Corrected Platform Error Interrupts (CPEI). In current ACPI
190specifications, we didn't have a way to change the target CPU. Hence if the
191current ACPI version doesn't support such re-direction, we disable that CPU
192by making it not-removable.
193
194In such cases you will also notice that the online file is missing under cpu0.
195
196Q: How do i find out if a particular CPU is not removable?
197A: Depending on the implementation, some architectures may show this by the
198absence of the "online" file. This is done if it can be determined ahead of
199time that this CPU cannot be removed.
200
201In some situations, this can be a run time check, i.e if you try to remove the
202last CPU, this will not be permitted. You can find such failures by
203investigating the return value of the "echo" command.
204
205Q: What happens when a CPU is being logically offlined?
206A: The following happen, listed in no particular order :-)
207
208- A notification is sent to in-kernel registered modules by sending an event
209 CPU_DOWN_PREPARE
210- All process is migrated away from this outgoing CPU to a new CPU
211- All interrupts targeted to this CPU is migrated to a new CPU
212- timers/bottom half/task lets are also migrated to a new CPU
213- Once all services are migrated, kernel calls an arch specific routine
214 __cpu_disable() to perform arch specific cleanup.
215- Once this is successful, an event for successful cleanup is sent by an event
216 CPU_DEAD.
217
218 "It is expected that each service cleans up when the CPU_DOWN_PREPARE
219 notifier is called, when CPU_DEAD is called its expected there is nothing
220 running on behalf of this CPU that was offlined"
221
222Q: If i have some kernel code that needs to be aware of CPU arrival and
223 departure, how to i arrange for proper notification?
224A: This is what you would need in your kernel code to receive notifications.
225
226 #include <linux/cpu.h>
227 static int __cpuinit foobar_cpu_callback(struct notifier_block *nfb,
228 unsigned long action, void *hcpu)
229 {
230 unsigned int cpu = (unsigned long)hcpu;
231
232 switch (action) {
233 case CPU_ONLINE:
234 foobar_online_action(cpu);
235 break;
236 case CPU_DEAD:
237 foobar_dead_action(cpu);
238 break;
239 }
240 return NOTIFY_OK;
241 }
242
243 static struct notifier_block foobar_cpu_notifer =
244 {
245 .notifier_call = foobar_cpu_callback,
246 };
247
248
249In your init function,
250
251 register_cpu_notifier(&foobar_cpu_notifier);
252
253You can fail PREPARE notifiers if something doesn't work to prepare resources.
254This will stop the activity and send a following CANCELED event back.
255
256CPU_DEAD should not be failed, its just a goodness indication, but bad
257things will happen if a notifier in path sent a BAD notify code.
258
259Q: I don't see my action being called for all CPUs already up and running?
260A: Yes, CPU notifiers are called only when new CPUs are on-lined or offlined.
261 If you need to perform some action for each cpu already in the system, then
262
263 for_each_online_cpu(i) {
264 foobar_cpu_callback(&foobar_cpu_notifier, CPU_UP_PREPARE, i);
265 foobar_cpu_callback(&foobar-cpu_notifier, CPU_ONLINE, i);
266 }
267
268Q: If i would like to develop cpu hotplug support for a new architecture,
269 what do i need at a minimum?
270A: The following are what is required for CPU hotplug infrastructure to work
271 correctly.
272
273 - Make sure you have an entry in Kconfig to enable CONFIG_HOTPLUG_CPU
274 - __cpu_up() - Arch interface to bring up a CPU
275 - __cpu_disable() - Arch interface to shutdown a CPU, no more interrupts
276 can be handled by the kernel after the routine
277 returns. Including local APIC timers etc are
278 shutdown.
279 - __cpu_die() - This actually supposed to ensure death of the CPU.
280 Actually look at some example code in other arch
281 that implement CPU hotplug. The processor is taken
282 down from the idle() loop for that specific
283 architecture. __cpu_die() typically waits for some
284 per_cpu state to be set, to ensure the processor
285 dead routine is called to be sure positively.
286
287Q: I need to ensure that a particular cpu is not removed when there is some
288 work specific to this cpu is in progress.
289A: First switch the current thread context to preferred cpu
290
291 int my_func_on_cpu(int cpu)
292 {
293 cpumask_t saved_mask, new_mask = CPU_MASK_NONE;
294 int curr_cpu, err = 0;
295
296 saved_mask = current->cpus_allowed;
297 cpu_set(cpu, new_mask);
298 err = set_cpus_allowed(current, new_mask);
299
300 if (err)
301 return err;
302
303 /*
304 * If we got scheduled out just after the return from
305 * set_cpus_allowed() before running the work, this ensures
306 * we stay locked.
307 */
308 curr_cpu = get_cpu();
309
310 if (curr_cpu != cpu) {
311 err = -EAGAIN;
312 goto ret;
313 } else {
314 /*
315 * Do work : But cant sleep, since get_cpu() disables preempt
316 */
317 }
318 ret:
319 put_cpu();
320 set_cpus_allowed(current, saved_mask);
321 return err;
322 }
323
324
325Q: How do we determine how many CPUs are available for hotplug.
326A: There is no clear spec defined way from ACPI that can give us that
327 information today. Based on some input from Natalie of Unisys,
328 that the ACPI MADT (Multiple APIC Description Tables) marks those possible
329 CPUs in a system with disabled status.
330
331 Andi implemented some simple heuristics that count the number of disabled
332 CPUs in MADT as hotpluggable CPUS. In the case there are no disabled CPUS
333 we assume 1/2 the number of CPUs currently present can be hotplugged.
334
335 Caveat: Today's ACPI MADT can only provide 256 entries since the apicid field
336 in MADT is only 8 bits.
337
338User Space Notification
339
340Hotplug support for devices is common in Linux today. Its being used today to
341support automatic configuration of network, usb and pci devices. A hotplug
342event can be used to invoke an agent script to perform the configuration task.
343
344You can add /etc/hotplug/cpu.agent to handle hotplug notification user space
345scripts.
346
347 #!/bin/bash
348 # $Id: cpu.agent
349 # Kernel hotplug params include:
350 #ACTION=%s [online or offline]
351 #DEVPATH=%s
352 #
353 cd /etc/hotplug
354 . ./hotplug.functions
355
356 case $ACTION in
357 online)
358 echo `date` ":cpu.agent" add cpu >> /tmp/hotplug.txt
359 ;;
360 offline)
361 echo `date` ":cpu.agent" remove cpu >>/tmp/hotplug.txt
362 ;;
363 *)
364 debug_mesg CPU $ACTION event not supported
365 exit 1
366 ;;
367 esac