blob: 4868c34f75090fe75b4f6bb28f70c7638540c4cd [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>
Heiko Carstens255acee2006-02-17 13:52:46 -080014 s390:
15 Heiko Carstens <heiko.carstens@de.ibm.com>
Ashok Rajc8094062006-01-08 01:03:17 -080016
17Authors: Ashok Raj <ashok.raj@intel.com>
18Lots of feedback: Nathan Lynch <nathanl@austin.ibm.com>,
19 Joel Schopp <jschopp@austin.ibm.com>
20
21Introduction
22
23Modern advances in system architectures have introduced advanced error
24reporting and correction capabilities in processors. CPU architectures permit
25partitioning support, where compute resources of a single CPU could be made
26available to virtual machine environments. There are couple OEMS that
27support NUMA hardware which are hot pluggable as well, where physical
28node insertion and removal require support for CPU hotplug.
29
30Such advances require CPUs available to a kernel to be removed either for
31provisioning reasons, or for RAS purposes to keep an offending CPU off
32system execution path. Hence the need for CPU hotplug support in the
33Linux kernel.
34
35A more novel use of CPU-hotplug support is its use today in suspend
36resume support for SMP. Dual-core and HT support makes even
37a laptop run SMP kernels which didn't support these methods. SMP support
38for suspend/resume is a work in progress.
39
40General Stuff about CPU Hotplug
41--------------------------------
42
43Command Line Switches
44---------------------
45maxcpus=n Restrict boot time cpus to n. Say if you have 4 cpus, using
46 maxcpus=2 will only boot 2. You can choose to bring the
47 other cpus later online, read FAQ's for more info.
48
Satoru Takeuchica926e82006-10-19 23:29:06 -070049additional_cpus=n (*) Use this to limit hotpluggable cpus. This option sets
Heiko Carstens255acee2006-02-17 13:52:46 -080050 cpu_possible_map = cpu_present_map + additional_cpus
Ashok Raj8f8b1132006-02-16 14:01:48 -080051
Heiko Carstens6303dbf2006-02-20 18:27:58 -080052(*) Option valid only for following architectures
53- x86_64, ia64, s390
54
Ashok Raj8f8b1132006-02-16 14:01:48 -080055ia64 and x86_64 use the number of disabled local apics in ACPI tables MADT
56to determine the number of potentially hot-pluggable cpus. The implementation
57should only rely on this to count the #of cpus, but *MUST* not rely on the
58apicid values in those tables for disabled apics. In the event BIOS doesnt
59mark such hot-pluggable cpus as disabled entries, one could use this
60parameter "additional_cpus=x" to represent those cpus in the cpu_possible_map.
61
Heiko Carstens6303dbf2006-02-20 18:27:58 -080062s390 uses the number of cpus it detects at IPL time to also the number of bits
63in cpu_possible_map. If it is desired to add additional cpus at a later time
64the number should be specified using this option or the possible_cpus option.
Ashok Raj8f8b1132006-02-16 14:01:48 -080065
Heiko Carstens37a33022006-02-17 13:52:47 -080066possible_cpus=n [s390 only] use this to set hotpluggable cpus.
67 This option sets possible_cpus bits in
68 cpu_possible_map. Thus keeping the numbers of bits set
69 constant even if the machine gets rebooted.
70 This option overrides additional_cpus.
71
Ashok Rajc8094062006-01-08 01:03:17 -080072CPU maps and such
73-----------------
74[More on cpumaps and primitive to manipulate, please check
75include/linux/cpumask.h that has more descriptive text.]
76
77cpu_possible_map: Bitmap of possible CPUs that can ever be available in the
78system. This is used to allocate some boot time memory for per_cpu variables
79that aren't designed to grow/shrink as CPUs are made available or removed.
80Once set during boot time discovery phase, the map is static, i.e no bits
81are added or removed anytime. Trimming it accurately for your system needs
82upfront can save some boot time memory. See below for how we use heuristics
83in x86_64 case to keep this under check.
84
85cpu_online_map: Bitmap of all CPUs currently online. Its set in __cpu_up()
86after a cpu is available for kernel scheduling and ready to receive
87interrupts from devices. Its cleared when a cpu is brought down using
88__cpu_disable(), before which all OS services including interrupts are
89migrated to another target CPU.
90
91cpu_present_map: Bitmap of CPUs currently present in the system. Not all
92of them may be online. When physical hotplug is processed by the relevant
93subsystem (e.g ACPI) can change and new bit either be added or removed
94from the map depending on the event is hot-add/hot-remove. There are currently
95no locking rules as of now. Typical usage is to init topology during boot,
96at which time hotplug is disabled.
97
98You really dont need to manipulate any of the system cpu maps. They should
99be read-only for most use. When setting up per-cpu resources almost always use
KAMEZAWA Hiroyuki3c30a752006-03-28 01:56:39 -0800100cpu_possible_map/for_each_possible_cpu() to iterate.
Ashok Rajc8094062006-01-08 01:03:17 -0800101
102Never use anything other than cpumask_t to represent bitmap of CPUs.
103
Satoru Takeuchica926e82006-10-19 23:29:06 -0700104 #include <linux/cpumask.h>
Ashok Rajc8094062006-01-08 01:03:17 -0800105
Satoru Takeuchica926e82006-10-19 23:29:06 -0700106 for_each_possible_cpu - Iterate over cpu_possible_map
107 for_each_online_cpu - Iterate over cpu_online_map
108 for_each_present_cpu - Iterate over cpu_present_map
109 for_each_cpu_mask(x,mask) - Iterate over some random collection of cpu mask.
Ashok Rajc8094062006-01-08 01:03:17 -0800110
Satoru Takeuchica926e82006-10-19 23:29:06 -0700111 #include <linux/cpu.h>
112 lock_cpu_hotplug() and unlock_cpu_hotplug():
Ashok Rajc8094062006-01-08 01:03:17 -0800113
114The above calls are used to inhibit cpu hotplug operations. While holding the
115cpucontrol mutex, cpu_online_map will not change. If you merely need to avoid
116cpus going away, you could also use preempt_disable() and preempt_enable()
117for those sections. Just remember the critical section cannot call any
118function that can sleep or schedule this process away. The preempt_disable()
119will work as long as stop_machine_run() is used to take a cpu down.
120
121CPU Hotplug - Frequently Asked Questions.
122
Satoru Takeuchica926e82006-10-19 23:29:06 -0700123Q: How to enable my kernel to support CPU hotplug?
Ashok Rajc8094062006-01-08 01:03:17 -0800124A: When doing make defconfig, Enable CPU hotplug support
125
126 "Processor type and Features" -> Support for Hotpluggable CPUs
127
128Make sure that you have CONFIG_HOTPLUG, and CONFIG_SMP turned on as well.
129
130You would need to enable CONFIG_HOTPLUG_CPU for SMP suspend/resume support
131as well.
132
133Q: What architectures support CPU hotplug?
134A: As of 2.6.14, the following architectures support CPU hotplug.
135
136i386 (Intel), ppc, ppc64, parisc, s390, ia64 and x86_64
137
138Q: How to test if hotplug is supported on the newly built kernel?
139A: You should now notice an entry in sysfs.
140
141Check if sysfs is mounted, using the "mount" command. You should notice
142an entry as shown below in the output.
143
Satoru Takeuchica926e82006-10-19 23:29:06 -0700144 ....
145 none on /sys type sysfs (rw)
146 ....
Ashok Rajc8094062006-01-08 01:03:17 -0800147
Satoru Takeuchica926e82006-10-19 23:29:06 -0700148If this is not mounted, do the following.
Ashok Rajc8094062006-01-08 01:03:17 -0800149
Satoru Takeuchica926e82006-10-19 23:29:06 -0700150 #mkdir /sysfs
151 #mount -t sysfs sys /sys
Ashok Rajc8094062006-01-08 01:03:17 -0800152
Satoru Takeuchica926e82006-10-19 23:29:06 -0700153Now you should see entries for all present cpu, the following is an example
Ashok Rajc8094062006-01-08 01:03:17 -0800154in a 8-way system.
155
Satoru Takeuchica926e82006-10-19 23:29:06 -0700156 #pwd
157 #/sys/devices/system/cpu
158 #ls -l
159 total 0
160 drwxr-xr-x 10 root root 0 Sep 19 07:44 .
161 drwxr-xr-x 13 root root 0 Sep 19 07:45 ..
162 drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu0
163 drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu1
164 drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu2
165 drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu3
166 drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu4
167 drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu5
168 drwxr-xr-x 3 root root 0 Sep 19 07:44 cpu6
169 drwxr-xr-x 3 root root 0 Sep 19 07:48 cpu7
Ashok Rajc8094062006-01-08 01:03:17 -0800170
171Under each directory you would find an "online" file which is the control
172file to logically online/offline a processor.
173
174Q: Does hot-add/hot-remove refer to physical add/remove of cpus?
175A: The usage of hot-add/remove may not be very consistently used in the code.
Satoru Takeuchica926e82006-10-19 23:29:06 -0700176CONFIG_HOTPLUG_CPU enables logical online/offline capability in the kernel.
Ashok Rajc8094062006-01-08 01:03:17 -0800177To support physical addition/removal, one would need some BIOS hooks and
178the platform should have something like an attention button in PCI hotplug.
179CONFIG_ACPI_HOTPLUG_CPU enables ACPI support for physical add/remove of CPUs.
180
181Q: How do i logically offline a CPU?
182A: Do the following.
183
Satoru Takeuchica926e82006-10-19 23:29:06 -0700184 #echo 0 > /sys/devices/system/cpu/cpuX/online
Ashok Rajc8094062006-01-08 01:03:17 -0800185
Satoru Takeuchica926e82006-10-19 23:29:06 -0700186Once the logical offline is successful, check
Ashok Rajc8094062006-01-08 01:03:17 -0800187
Satoru Takeuchica926e82006-10-19 23:29:06 -0700188 #cat /proc/interrupts
Ashok Rajc8094062006-01-08 01:03:17 -0800189
Satoru Takeuchica926e82006-10-19 23:29:06 -0700190You should now not see the CPU that you removed. Also online file will report
Ashok Rajc8094062006-01-08 01:03:17 -0800191the state as 0 when a cpu if offline and 1 when its online.
192
Satoru Takeuchica926e82006-10-19 23:29:06 -0700193 #To display the current cpu state.
194 #cat /sys/devices/system/cpu/cpuX/online
Ashok Rajc8094062006-01-08 01:03:17 -0800195
196Q: Why cant i remove CPU0 on some systems?
197A: Some architectures may have some special dependency on a certain CPU.
198
199For e.g in IA64 platforms we have ability to sent platform interrupts to the
200OS. a.k.a Corrected Platform Error Interrupts (CPEI). In current ACPI
201specifications, we didn't have a way to change the target CPU. Hence if the
202current ACPI version doesn't support such re-direction, we disable that CPU
203by making it not-removable.
204
205In such cases you will also notice that the online file is missing under cpu0.
206
207Q: How do i find out if a particular CPU is not removable?
208A: Depending on the implementation, some architectures may show this by the
209absence of the "online" file. This is done if it can be determined ahead of
210time that this CPU cannot be removed.
211
212In some situations, this can be a run time check, i.e if you try to remove the
213last CPU, this will not be permitted. You can find such failures by
214investigating the return value of the "echo" command.
215
216Q: What happens when a CPU is being logically offlined?
217A: The following happen, listed in no particular order :-)
218
219- A notification is sent to in-kernel registered modules by sending an event
220 CPU_DOWN_PREPARE
221- All process is migrated away from this outgoing CPU to a new CPU
222- All interrupts targeted to this CPU is migrated to a new CPU
223- timers/bottom half/task lets are also migrated to a new CPU
224- Once all services are migrated, kernel calls an arch specific routine
225 __cpu_disable() to perform arch specific cleanup.
226- Once this is successful, an event for successful cleanup is sent by an event
227 CPU_DEAD.
228
229 "It is expected that each service cleans up when the CPU_DOWN_PREPARE
230 notifier is called, when CPU_DEAD is called its expected there is nothing
231 running on behalf of this CPU that was offlined"
232
233Q: If i have some kernel code that needs to be aware of CPU arrival and
234 departure, how to i arrange for proper notification?
235A: This is what you would need in your kernel code to receive notifications.
236
Satoru Takeuchica926e82006-10-19 23:29:06 -0700237 #include <linux/cpu.h>
238 static int __cpuinit foobar_cpu_callback(struct notifier_block *nfb,
Ashok Rajc8094062006-01-08 01:03:17 -0800239 unsigned long action, void *hcpu)
240 {
241 unsigned int cpu = (unsigned long)hcpu;
242
243 switch (action) {
244 case CPU_ONLINE:
245 foobar_online_action(cpu);
246 break;
247 case CPU_DEAD:
248 foobar_dead_action(cpu);
249 break;
250 }
251 return NOTIFY_OK;
252 }
253
Chandra Seetharaman7c7165c2006-07-30 03:03:36 -0700254 static struct notifier_block __cpuinitdata foobar_cpu_notifer =
Ashok Rajc8094062006-01-08 01:03:17 -0800255 {
256 .notifier_call = foobar_cpu_callback,
257 };
258
Chandra Seetharaman7c7165c2006-07-30 03:03:36 -0700259You need to call register_cpu_notifier() from your init function.
260Init functions could be of two types:
2611. early init (init function called when only the boot processor is online).
2622. late init (init function called _after_ all the CPUs are online).
Ashok Rajc8094062006-01-08 01:03:17 -0800263
Chandra Seetharaman7c7165c2006-07-30 03:03:36 -0700264For the first case, you should add the following to your init function
Ashok Rajc8094062006-01-08 01:03:17 -0800265
266 register_cpu_notifier(&foobar_cpu_notifier);
267
Chandra Seetharaman7c7165c2006-07-30 03:03:36 -0700268For the second case, you should add the following to your init function
269
270 register_hotcpu_notifier(&foobar_cpu_notifier);
271
Ashok Rajc8094062006-01-08 01:03:17 -0800272You can fail PREPARE notifiers if something doesn't work to prepare resources.
273This will stop the activity and send a following CANCELED event back.
274
275CPU_DEAD should not be failed, its just a goodness indication, but bad
276things will happen if a notifier in path sent a BAD notify code.
277
278Q: I don't see my action being called for all CPUs already up and running?
279A: Yes, CPU notifiers are called only when new CPUs are on-lined or offlined.
280 If you need to perform some action for each cpu already in the system, then
281
Satoru Takeuchica926e82006-10-19 23:29:06 -0700282 for_each_online_cpu(i) {
Ashok Rajc8094062006-01-08 01:03:17 -0800283 foobar_cpu_callback(&foobar_cpu_notifier, CPU_UP_PREPARE, i);
Satoru Takeuchica926e82006-10-19 23:29:06 -0700284 foobar_cpu_callback(&foobar_cpu_notifier, CPU_ONLINE, i);
285 }
Ashok Rajc8094062006-01-08 01:03:17 -0800286
287Q: If i would like to develop cpu hotplug support for a new architecture,
288 what do i need at a minimum?
289A: The following are what is required for CPU hotplug infrastructure to work
290 correctly.
291
292 - Make sure you have an entry in Kconfig to enable CONFIG_HOTPLUG_CPU
293 - __cpu_up() - Arch interface to bring up a CPU
294 - __cpu_disable() - Arch interface to shutdown a CPU, no more interrupts
295 can be handled by the kernel after the routine
296 returns. Including local APIC timers etc are
297 shutdown.
298 - __cpu_die() - This actually supposed to ensure death of the CPU.
299 Actually look at some example code in other arch
300 that implement CPU hotplug. The processor is taken
301 down from the idle() loop for that specific
302 architecture. __cpu_die() typically waits for some
303 per_cpu state to be set, to ensure the processor
304 dead routine is called to be sure positively.
305
306Q: I need to ensure that a particular cpu is not removed when there is some
307 work specific to this cpu is in progress.
308A: First switch the current thread context to preferred cpu
309
Satoru Takeuchica926e82006-10-19 23:29:06 -0700310 int my_func_on_cpu(int cpu)
311 {
312 cpumask_t saved_mask, new_mask = CPU_MASK_NONE;
313 int curr_cpu, err = 0;
Ashok Rajc8094062006-01-08 01:03:17 -0800314
Satoru Takeuchica926e82006-10-19 23:29:06 -0700315 saved_mask = current->cpus_allowed;
316 cpu_set(cpu, new_mask);
317 err = set_cpus_allowed(current, new_mask);
Ashok Rajc8094062006-01-08 01:03:17 -0800318
Satoru Takeuchica926e82006-10-19 23:29:06 -0700319 if (err)
320 return err;
Ashok Rajc8094062006-01-08 01:03:17 -0800321
Satoru Takeuchica926e82006-10-19 23:29:06 -0700322 /*
323 * If we got scheduled out just after the return from
324 * set_cpus_allowed() before running the work, this ensures
325 * we stay locked.
326 */
327 curr_cpu = get_cpu();
Ashok Rajc8094062006-01-08 01:03:17 -0800328
Satoru Takeuchica926e82006-10-19 23:29:06 -0700329 if (curr_cpu != cpu) {
330 err = -EAGAIN;
331 goto ret;
332 } else {
333 /*
334 * Do work : But cant sleep, since get_cpu() disables preempt
335 */
336 }
337 ret:
338 put_cpu();
339 set_cpus_allowed(current, saved_mask);
340 return err;
341 }
Ashok Rajc8094062006-01-08 01:03:17 -0800342
343
344Q: How do we determine how many CPUs are available for hotplug.
345A: There is no clear spec defined way from ACPI that can give us that
346 information today. Based on some input from Natalie of Unisys,
347 that the ACPI MADT (Multiple APIC Description Tables) marks those possible
348 CPUs in a system with disabled status.
349
350 Andi implemented some simple heuristics that count the number of disabled
351 CPUs in MADT as hotpluggable CPUS. In the case there are no disabled CPUS
352 we assume 1/2 the number of CPUs currently present can be hotplugged.
353
354 Caveat: Today's ACPI MADT can only provide 256 entries since the apicid field
355 in MADT is only 8 bits.
356
357User Space Notification
358
359Hotplug support for devices is common in Linux today. Its being used today to
360support automatic configuration of network, usb and pci devices. A hotplug
361event can be used to invoke an agent script to perform the configuration task.
362
363You can add /etc/hotplug/cpu.agent to handle hotplug notification user space
364scripts.
365
366 #!/bin/bash
367 # $Id: cpu.agent
368 # Kernel hotplug params include:
369 #ACTION=%s [online or offline]
370 #DEVPATH=%s
371 #
372 cd /etc/hotplug
373 . ./hotplug.functions
374
375 case $ACTION in
376 online)
377 echo `date` ":cpu.agent" add cpu >> /tmp/hotplug.txt
378 ;;
379 offline)
380 echo `date` ":cpu.agent" remove cpu >>/tmp/hotplug.txt
381 ;;
382 *)
383 debug_mesg CPU $ACTION event not supported
384 exit 1
385 ;;
386 esac