1==================================================================== 2Interaction of Suspend code (S3) with the CPU hotplug infrastructure 3==================================================================== 4 5(C) 2011 - 2014 Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> 6 7 8I. Differences between CPU hotplug and Suspend-to-RAM 9====================================================== 10 11How does the regular CPU hotplug code differ from how the Suspend-to-RAM 12infrastructure uses it internally? And where do they share common code? 13 14Well, a picture is worth a thousand words... So ASCII art follows :-) 15 16[This depicts the current design in the kernel, and focusses only on the 17interactions involving the freezer and CPU hotplug and also tries to explain 18the locking involved. It outlines the notifications involved as well. 19But please note that here, only the call paths are illustrated, with the aim 20of describing where they take different paths and where they share code. 21What happens when regular CPU hotplug and Suspend-to-RAM race with each other 22is not depicted here.] 23 24On a high level, the suspend-resume cycle goes like this:: 25 26 |Freeze| -> |Disable nonboot| -> |Do suspend| -> |Enable nonboot| -> |Thaw | 27 |tasks | | cpus | | | | cpus | |tasks| 28 29 30More details follow:: 31 32 Suspend call path 33 ----------------- 34 35 Write 'mem' to 36 /sys/power/state 37 sysfs file 38 | 39 v 40 Acquire system_transition_mutex lock 41 | 42 v 43 Send PM_SUSPEND_PREPARE 44 notifications 45 | 46 v 47 Freeze tasks 48 | 49 | 50 v 51 disable_nonboot_cpus() 52 /* start */ 53 | 54 v 55 Acquire cpu_add_remove_lock 56 | 57 v 58 Iterate over CURRENTLY 59 online CPUs 60 | 61 | 62 | ---------- 63 v | L 64 ======> _cpu_down() | 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. | 71 | | O 72 | | 73 | | 74 v | 75 Note down these cpus in | P 76 frozen_cpus mask ---------- 77 | 78 v 79 Disable regular cpu hotplug 80 by increasing cpu_hotplug_disabled 81 | 82 v 83 Release cpu_add_remove_lock 84 | 85 v 86 /* disable_nonboot_cpus() complete */ 87 | 88 v 89 Do suspend 90 91 92 93Resuming back is likewise, with the counterparts being (in the order of 94execution during resume): 95 96* enable_nonboot_cpus() which involves:: 97 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 102 v 103 104* thaw tasks 105* send PM_POST_SUSPEND notifications 106* Release system_transition_mutex lock. 107 108 109It is to be noted here that the system_transition_mutex lock is acquired at the very 110beginning, when we are just starting out to suspend, and then released only 111after the entire cycle is complete (i.e., suspend + resume). 112 113:: 114 115 116 117 Regular CPU hotplug call path 118 ----------------------------- 119 120 Write 0 (or 1) to 121 /sys/devices/system/cpu/cpu*/online 122 sysfs file 123 | 124 | 125 v 126 cpu_down() 127 | 128 v 129 Acquire cpu_add_remove_lock 130 | 131 v 132 If cpu_hotplug_disabled > 0 133 return gracefully 134 | 135 | 136 v 137 ======> _cpu_down() 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. 144 | 145 | 146 v 147 Release cpu_add_remove_lock 148 [That's it!, for 149 regular CPU hotplug] 150 151 152 153So, as can be seen from the two diagrams (the parts marked as "Common code"), 154regular 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, 156in that during regular CPU hotplug, 0 is passed for the 'tasks_frozen' 157argument. But during suspend, since the tasks are already frozen by the time 158the non-boot CPUs are offlined or onlined, the _cpu_*() functions are called 159with the 'tasks_frozen' argument set to 1. 160[See below for some known issues regarding this.] 161 162 163Important files and functions/entry points: 164------------------------------------------- 165 166- 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](), [disable|enable]_nonboot_cpus() 169 170 171 172II. What are the issues involved in CPU hotplug? 173------------------------------------------------ 174 175There are some interesting situations involving CPU hotplug and microcode 176update on the CPUs, as discussed below: 177 178[Please bear in mind that the kernel requests the microcode images from 179userspace, using the request_firmware() function defined in 180drivers/base/firmware_loader/main.c] 181 182 183a. When all the CPUs are identical: 184 185 This is the most common situation and it is quite straightforward: we want 186 to apply the same microcode revision to each of the CPUs. 187 To give an example of x86, the collect_cpu_info() function defined in 188 arch/x86/kernel/microcode_core.c helps in discovering the type of the CPU 189 and thereby in applying the correct microcode revision to it. 190 But note that the kernel does not maintain a common microcode image for the 191 all CPUs, in order to handle case 'b' described below. 192 193 194b. When some of the CPUs are different than the rest: 195 196 In this case since we probably need to apply different microcode revisions 197 to different CPUs, the kernel maintains a copy of the correct microcode 198 image for each CPU (after appropriate CPU type/model discovery using 199 functions such as collect_cpu_info()). 200 201 202c. When a CPU is physically hot-unplugged and a new (and possibly different 203 type of) CPU is hot-plugged into the system: 204 205 In the current design of the kernel, whenever a CPU is taken offline during 206 a regular CPU hotplug operation, upon receiving the CPU_DEAD notification 207 (which is sent by the CPU hotplug code), the microcode update driver's 208 callback for that event reacts by freeing the kernel's copy of the 209 microcode image for that CPU. 210 211 Hence, when a new CPU is brought online, since the kernel finds that it 212 doesn't have the microcode image, it does the CPU type/model discovery 213 afresh and then requests the userspace for the appropriate microcode image 214 for that CPU, which is subsequently applied. 215 216 For example, in x86, the mc_cpu_callback() function (which is the microcode 217 update driver's callback registered for CPU hotplug events) calls 218 microcode_update_cpu() which would call microcode_init_cpu() in this case, 219 instead of microcode_resume_cpu() when it finds that the kernel doesn't 220 have a valid microcode image. This ensures that the CPU type/model 221 discovery is performed and the right microcode is applied to the CPU after 222 getting it from userspace. 223 224 225d. Handling microcode update during suspend/hibernate: 226 227 Strictly speaking, during a CPU hotplug operation which does not involve 228 physically removing or inserting CPUs, the CPUs are not actually powered 229 off during a CPU offline. They are just put to the lowest C-states possible. 230 Hence, in such a case, it is not really necessary to re-apply microcode 231 when the CPUs are brought back online, since they wouldn't have lost the 232 image during the CPU offline operation. 233 234 This is the usual scenario encountered during a resume after a suspend. 235 However, in the case of hibernation, since all the CPUs are completely 236 powered off, during restore it becomes necessary to apply the microcode 237 images to all the CPUs. 238 239 [Note that we don't expect someone to physically pull out nodes and insert 240 nodes with a different type of CPUs in-between a suspend-resume or a 241 hibernate/restore cycle.] 242 243 In the current design of the kernel however, during a CPU offline operation 244 as part of the suspend/hibernate cycle (cpuhp_tasks_frozen is set), 245 the existing copy of microcode image in the kernel is not freed up. 246 And during the CPU online operations (during resume/restore), since the 247 kernel finds that it already has copies of the microcode images for all the 248 CPUs, it just applies them to the CPUs, avoiding any re-discovery of CPU 249 type/model and the need for validating whether the microcode revisions are 250 right for the CPUs or not (due to the above assumption that physical CPU 251 hotplug will not be done in-between suspend/resume or hibernate/restore 252 cycles). 253 254 255III. Known problems 256=================== 257 258Are there any known problems when regular CPU hotplug and suspend race 259with each other? 260 261Yes, they are listed below: 262 2631. When invoking regular CPU hotplug, the 'tasks_frozen' argument passed to 264 the _cpu_down() and _cpu_up() functions is *always* 0. 265 This might not reflect the true current state of the system, since the 266 tasks could have been frozen by an out-of-band event such as a suspend 267 operation in progress. Hence, the cpuhp_tasks_frozen variable will not 268 reflect the frozen state and the CPU hotplug callbacks which evaluate 269 that variable might execute the wrong code path. 270 2712. If a regular CPU hotplug stress test happens to race with the freezer due 272 to a suspend operation in progress at the same time, then we could hit the 273 situation described below: 274 275 * A regular cpu online operation continues its journey from userspace 276 into the kernel, since the freezing has not yet begun. 277 * Then freezer gets to work and freezes userspace. 278 * If cpu online has not yet completed the microcode update stuff by now, 279 it will now start waiting on the frozen userspace in the 280 TASK_UNINTERRUPTIBLE state, in order to get the microcode image. 281 * Now the freezer continues and tries to freeze the remaining tasks. But 282 due to this wait mentioned above, the freezer won't be able to freeze 283 the cpu online hotplug task and hence freezing of tasks fails. 284 285 As a result of this task freezing failure, the suspend operation gets 286 aborted. 287