1 ====================================================================
2 Interaction of Suspend code (S3) with the CPU hotplug infrastructure
3 ====================================================================
5 (C) 2011 - 2014 Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
8 I. Differences between CPU hotplug and Suspend-to-RAM
9 ======================================================
11 How does the regular CPU hotplug code differ from how the Suspend-to-RAM
12 infrastructure uses it internally? And where do they share common code?
14 Well, a picture is worth a thousand words... So ASCII art follows :-)
16 [This depicts the current design in the kernel, and focusses only on the
17 interactions involving the freezer and CPU hotplug and also tries to explain
18 the locking involved. It outlines the notifications involved as well.
19 But please note that here, only the call paths are illustrated, with the aim
20 of describing where they take different paths and where they share code.
21 What happens when regular CPU hotplug and Suspend-to-RAM race with each other
22 is not depicted here.]
24 On a high level, the suspend-resume cycle goes like this::
26 |Freeze| -> |Disable nonboot| -> |Do suspend| -> |Enable nonboot| -> |Thaw |
27 |tasks | | cpus | | | | cpus | |tasks|
40 Acquire system_transition_mutex lock
43 Send PM_SUSPEND_PREPARE
51 disable_nonboot_cpus()
55 Acquire cpu_add_remove_lock
58 Iterate over CURRENTLY
65 | [This takes cpuhotplug.lock |
66 Common | before taking down the CPU |
67 code | and releases it when done] | O
68 | While it is at it, notifications |
69 | are sent when notable events occur, |
70 ======> by running all registered callbacks. |
75 Note down these cpus in | P
76 frozen_cpus mask ----------
79 Disable regular cpu hotplug
80 by increasing cpu_hotplug_disabled
83 Release cpu_add_remove_lock
86 /* disable_nonboot_cpus() complete */
93 Resuming back is likewise, with the counterparts being (in the order of
94 execution during resume):
96 * enable_nonboot_cpus() which involves::
98 | Acquire cpu_add_remove_lock
99 | Decrease cpu_hotplug_disabled, thereby enabling regular cpu hotplug
100 | Call _cpu_up() [for all those cpus in the frozen_cpus mask, in a loop]
101 | Release cpu_add_remove_lock
105 * send PM_POST_SUSPEND notifications
106 * Release system_transition_mutex lock.
109 It is to be noted here that the system_transition_mutex lock is acquired at the
110 very beginning, when we are just starting out to suspend, and then released only
111 after the entire cycle is complete (i.e., suspend + resume).
117 Regular CPU hotplug call path
118 -----------------------------
121 /sys/devices/system/cpu/cpu*/online
129 Acquire cpu_add_remove_lock
132 If cpu_hotplug_disabled > 0
138 | [This takes cpuhotplug.lock
139 Common | before taking down the CPU
140 code | and releases it when done]
141 | While it is at it, notifications
142 | are sent when notable events occur,
143 ======> by running all registered callbacks.
147 Release cpu_add_remove_lock
153 So, as can be seen from the two diagrams (the parts marked as "Common code"),
154 regular CPU hotplug and the suspend code path converge at the _cpu_down() and
155 _cpu_up() functions. They differ in the arguments passed to these functions,
156 in that during regular CPU hotplug, 0 is passed for the 'tasks_frozen'
157 argument. But during suspend, since the tasks are already frozen by the time
158 the non-boot CPUs are offlined or onlined, the _cpu_*() functions are called
159 with the 'tasks_frozen' argument set to 1.
160 [See below for some known issues regarding this.]
163 Important files and functions/entry points:
164 -------------------------------------------
166 - kernel/power/process.c : freeze_processes(), thaw_processes()
167 - kernel/power/suspend.c : suspend_prepare(), suspend_enter(), suspend_finish()
168 - kernel/cpu.c: cpu_[up|down](), _cpu_[up|down](),
169 [disable|enable]_nonboot_cpus()
173 II. What are the issues involved in CPU hotplug?
174 ------------------------------------------------
176 There are some interesting situations involving CPU hotplug and microcode
177 update on the CPUs, as discussed below:
179 [Please bear in mind that the kernel requests the microcode images from
180 userspace, using the request_firmware() function defined in
181 drivers/base/firmware_loader/main.c]
184 a. When all the CPUs are identical:
186 This is the most common situation and it is quite straightforward: we want
187 to apply the same microcode revision to each of the CPUs.
188 To give an example of x86, the collect_cpu_info() function defined in
189 arch/x86/kernel/microcode_core.c helps in discovering the type of the CPU
190 and thereby in applying the correct microcode revision to it.
191 But note that the kernel does not maintain a common microcode image for the
192 all CPUs, in order to handle case 'b' described below.
195 b. When some of the CPUs are different than the rest:
197 In this case since we probably need to apply different microcode revisions
198 to different CPUs, the kernel maintains a copy of the correct microcode
199 image for each CPU (after appropriate CPU type/model discovery using
200 functions such as collect_cpu_info()).
203 c. When a CPU is physically hot-unplugged and a new (and possibly different
204 type of) CPU is hot-plugged into the system:
206 In the current design of the kernel, whenever a CPU is taken offline during
207 a regular CPU hotplug operation, upon receiving the CPU_DEAD notification
208 (which is sent by the CPU hotplug code), the microcode update driver's
209 callback for that event reacts by freeing the kernel's copy of the
210 microcode image for that CPU.
212 Hence, when a new CPU is brought online, since the kernel finds that it
213 doesn't have the microcode image, it does the CPU type/model discovery
214 afresh and then requests the userspace for the appropriate microcode image
215 for that CPU, which is subsequently applied.
217 For example, in x86, the mc_cpu_callback() function (which is the microcode
218 update driver's callback registered for CPU hotplug events) calls
219 microcode_update_cpu() which would call microcode_init_cpu() in this case,
220 instead of microcode_resume_cpu() when it finds that the kernel doesn't
221 have a valid microcode image. This ensures that the CPU type/model
222 discovery is performed and the right microcode is applied to the CPU after
223 getting it from userspace.
226 d. Handling microcode update during suspend/hibernate:
228 Strictly speaking, during a CPU hotplug operation which does not involve
229 physically removing or inserting CPUs, the CPUs are not actually powered
230 off during a CPU offline. They are just put to the lowest C-states possible.
231 Hence, in such a case, it is not really necessary to re-apply microcode
232 when the CPUs are brought back online, since they wouldn't have lost the
233 image during the CPU offline operation.
235 This is the usual scenario encountered during a resume after a suspend.
236 However, in the case of hibernation, since all the CPUs are completely
237 powered off, during restore it becomes necessary to apply the microcode
238 images to all the CPUs.
240 [Note that we don't expect someone to physically pull out nodes and insert
241 nodes with a different type of CPUs in-between a suspend-resume or a
242 hibernate/restore cycle.]
244 In the current design of the kernel however, during a CPU offline operation
245 as part of the suspend/hibernate cycle (cpuhp_tasks_frozen is set),
246 the existing copy of microcode image in the kernel is not freed up.
247 And during the CPU online operations (during resume/restore), since the
248 kernel finds that it already has copies of the microcode images for all the
249 CPUs, it just applies them to the CPUs, avoiding any re-discovery of CPU
250 type/model and the need for validating whether the microcode revisions are
251 right for the CPUs or not (due to the above assumption that physical CPU
252 hotplug will not be done in-between suspend/resume or hibernate/restore
259 Are there any known problems when regular CPU hotplug and suspend race
262 Yes, they are listed below:
264 1. When invoking regular CPU hotplug, the 'tasks_frozen' argument passed to
265 the _cpu_down() and _cpu_up() functions is *always* 0.
266 This might not reflect the true current state of the system, since the
267 tasks could have been frozen by an out-of-band event such as a suspend
268 operation in progress. Hence, the cpuhp_tasks_frozen variable will not
269 reflect the frozen state and the CPU hotplug callbacks which evaluate
270 that variable might execute the wrong code path.
272 2. If a regular CPU hotplug stress test happens to race with the freezer due
273 to a suspend operation in progress at the same time, then we could hit the
274 situation described below:
276 * A regular cpu online operation continues its journey from userspace
277 into the kernel, since the freezing has not yet begun.
278 * Then freezer gets to work and freezes userspace.
279 * If cpu online has not yet completed the microcode update stuff by now,
280 it will now start waiting on the frozen userspace in the
281 TASK_UNINTERRUPTIBLE state, in order to get the microcode image.
282 * Now the freezer continues and tries to freeze the remaining tasks. But
283 due to this wait mentioned above, the freezer won't be able to freeze
284 the cpu online hotplug task and hence freezing of tasks fails.
286 As a result of this task freezing failure, the suspend operation gets