brintos

brintos / linux-shallow public Read only

0
0
Text · 16.3 KiB · aa58603 Raw
534 lines · plain
1=========================================================2Cluster-wide Power-up/power-down race avoidance algorithm3=========================================================4 5This file documents the algorithm which is used to coordinate CPU and6cluster setup and teardown operations and to manage hardware coherency7controls safely.8 9The section "Rationale" explains what the algorithm is for and why it is10needed.  "Basic model" explains general concepts using a simplified view11of the system.  The other sections explain the actual details of the12algorithm in use.13 14 15Rationale16---------17 18In a system containing multiple CPUs, it is desirable to have the19ability to turn off individual CPUs when the system is idle, reducing20power consumption and thermal dissipation.21 22In a system containing multiple clusters of CPUs, it is also desirable23to have the ability to turn off entire clusters.24 25Turning entire clusters off and on is a risky business, because it26involves performing potentially destructive operations affecting a group27of independently running CPUs, while the OS continues to run.  This28means that we need some coordination in order to ensure that critical29cluster-level operations are only performed when it is truly safe to do30so.31 32Simple locking may not be sufficient to solve this problem, because33mechanisms like Linux spinlocks may rely on coherency mechanisms which34are not immediately enabled when a cluster powers up.  Since enabling or35disabling those mechanisms may itself be a non-atomic operation (such as36writing some hardware registers and invalidating large caches), other37methods of coordination are required in order to guarantee safe38power-down and power-up at the cluster level.39 40The mechanism presented in this document describes a coherent memory41based protocol for performing the needed coordination.  It aims to be as42lightweight as possible, while providing the required safety properties.43 44 45Basic model46-----------47 48Each cluster and CPU is assigned a state, as follows:49 50	- DOWN51	- COMING_UP52	- UP53	- GOING_DOWN54 55::56 57	    +---------> UP ----------+58	    |                        v59 60	COMING_UP                GOING_DOWN61 62	    ^                        |63	    +--------- DOWN <--------+64 65 66DOWN:67	The CPU or cluster is not coherent, and is either powered off or68	suspended, or is ready to be powered off or suspended.69 70COMING_UP:71	The CPU or cluster has committed to moving to the UP state.72	It may be part way through the process of initialisation and73	enabling coherency.74 75UP:76	The CPU or cluster is active and coherent at the hardware77	level.  A CPU in this state is not necessarily being used78	actively by the kernel.79 80GOING_DOWN:81	The CPU or cluster has committed to moving to the DOWN82	state.  It may be part way through the process of teardown and83	coherency exit.84 85 86Each CPU has one of these states assigned to it at any point in time.87The CPU states are described in the "CPU state" section, below.88 89Each cluster is also assigned a state, but it is necessary to split the90state value into two parts (the "cluster" state and "inbound" state) and91to introduce additional states in order to avoid races between different92CPUs in the cluster simultaneously modifying the state.  The cluster-93level states are described in the "Cluster state" section.94 95To help distinguish the CPU states from cluster states in this96discussion, the state names are given a `CPU_` prefix for the CPU states,97and a `CLUSTER_` or `INBOUND_` prefix for the cluster states.98 99 100CPU state101---------102 103In this algorithm, each individual core in a multi-core processor is104referred to as a "CPU".  CPUs are assumed to be single-threaded:105therefore, a CPU can only be doing one thing at a single point in time.106 107This means that CPUs fit the basic model closely.108 109The algorithm defines the following states for each CPU in the system:110 111	- CPU_DOWN112	- CPU_COMING_UP113	- CPU_UP114	- CPU_GOING_DOWN115 116::117 118	 cluster setup and119	CPU setup complete          policy decision120	      +-----------> CPU_UP ------------+121	      |                                v122 123	CPU_COMING_UP                   CPU_GOING_DOWN124 125	      ^                                |126	      +----------- CPU_DOWN <----------+127	 policy decision           CPU teardown complete128	or hardware event129 130 131The definitions of the four states correspond closely to the states of132the basic model.133 134Transitions between states occur as follows.135 136A trigger event (spontaneous) means that the CPU can transition to the137next state as a result of making local progress only, with no138requirement for any external event to happen.139 140 141CPU_DOWN:142	A CPU reaches the CPU_DOWN state when it is ready for143	power-down.  On reaching this state, the CPU will typically144	power itself down or suspend itself, via a WFI instruction or a145	firmware call.146 147	Next state:148		CPU_COMING_UP149	Conditions:150		none151 152	Trigger events:153		a) an explicit hardware power-up operation, resulting154		   from a policy decision on another CPU;155 156		b) a hardware event, such as an interrupt.157 158 159CPU_COMING_UP:160	A CPU cannot start participating in hardware coherency until the161	cluster is set up and coherent.  If the cluster is not ready,162	then the CPU will wait in the CPU_COMING_UP state until the163	cluster has been set up.164 165	Next state:166		CPU_UP167	Conditions:168		The CPU's parent cluster must be in CLUSTER_UP.169	Trigger events:170		Transition of the parent cluster to CLUSTER_UP.171 172	Refer to the "Cluster state" section for a description of the173	CLUSTER_UP state.174 175 176CPU_UP:177	When a CPU reaches the CPU_UP state, it is safe for the CPU to178	start participating in local coherency.179 180	This is done by jumping to the kernel's CPU resume code.181 182	Note that the definition of this state is slightly different183	from the basic model definition: CPU_UP does not mean that the184	CPU is coherent yet, but it does mean that it is safe to resume185	the kernel.  The kernel handles the rest of the resume186	procedure, so the remaining steps are not visible as part of the187	race avoidance algorithm.188 189	The CPU remains in this state until an explicit policy decision190	is made to shut down or suspend the CPU.191 192	Next state:193		CPU_GOING_DOWN194	Conditions:195		none196	Trigger events:197		explicit policy decision198 199 200CPU_GOING_DOWN:201	While in this state, the CPU exits coherency, including any202	operations required to achieve this (such as cleaning data203	caches).204 205	Next state:206		CPU_DOWN207	Conditions:208		local CPU teardown complete209	Trigger events:210		(spontaneous)211 212 213Cluster state214-------------215 216A cluster is a group of connected CPUs with some common resources.217Because a cluster contains multiple CPUs, it can be doing multiple218things at the same time.  This has some implications.  In particular, a219CPU can start up while another CPU is tearing the cluster down.220 221In this discussion, the "outbound side" is the view of the cluster state222as seen by a CPU tearing the cluster down.  The "inbound side" is the223view of the cluster state as seen by a CPU setting the CPU up.224 225In order to enable safe coordination in such situations, it is important226that a CPU which is setting up the cluster can advertise its state227independently of the CPU which is tearing down the cluster.  For this228reason, the cluster state is split into two parts:229 230	"cluster" state: The global state of the cluster; or the state231	on the outbound side:232 233		- CLUSTER_DOWN234		- CLUSTER_UP235		- CLUSTER_GOING_DOWN236 237	"inbound" state: The state of the cluster on the inbound side.238 239		- INBOUND_NOT_COMING_UP240		- INBOUND_COMING_UP241 242 243	The different pairings of these states results in six possible244	states for the cluster as a whole::245 246	                            CLUSTER_UP247	          +==========> INBOUND_NOT_COMING_UP -------------+248	          #                                               |249	                                                          |250	     CLUSTER_UP     <----+                                |251	  INBOUND_COMING_UP      |                                v252 253	          ^             CLUSTER_GOING_DOWN       CLUSTER_GOING_DOWN254	          #              INBOUND_COMING_UP <=== INBOUND_NOT_COMING_UP255 256	    CLUSTER_DOWN         |                                |257	  INBOUND_COMING_UP <----+                                |258	                                                          |259	          ^                                               |260	          +===========     CLUSTER_DOWN      <------------+261	                       INBOUND_NOT_COMING_UP262 263	Transitions -----> can only be made by the outbound CPU, and264	only involve changes to the "cluster" state.265 266	Transitions ===##> can only be made by the inbound CPU, and only267	involve changes to the "inbound" state, except where there is no268	further transition possible on the outbound side (i.e., the269	outbound CPU has put the cluster into the CLUSTER_DOWN state).270 271	The race avoidance algorithm does not provide a way to determine272	which exact CPUs within the cluster play these roles.  This must273	be decided in advance by some other means.  Refer to the section274	"Last man and first man selection" for more explanation.275 276 277	CLUSTER_DOWN/INBOUND_NOT_COMING_UP is the only state where the278	cluster can actually be powered down.279 280	The parallelism of the inbound and outbound CPUs is observed by281	the existence of two different paths from CLUSTER_GOING_DOWN/282	INBOUND_NOT_COMING_UP (corresponding to GOING_DOWN in the basic283	model) to CLUSTER_DOWN/INBOUND_COMING_UP (corresponding to284	COMING_UP in the basic model).  The second path avoids cluster285	teardown completely.286 287	CLUSTER_UP/INBOUND_COMING_UP is equivalent to UP in the basic288	model.  The final transition to CLUSTER_UP/INBOUND_NOT_COMING_UP289	is trivial and merely resets the state machine ready for the290	next cycle.291 292	Details of the allowable transitions follow.293 294	The next state in each case is notated295 296		<cluster state>/<inbound state> (<transitioner>)297 298	where the <transitioner> is the side on which the transition299	can occur; either the inbound or the outbound side.300 301 302CLUSTER_DOWN/INBOUND_NOT_COMING_UP:303	Next state:304		CLUSTER_DOWN/INBOUND_COMING_UP (inbound)305	Conditions:306		none307 308	Trigger events:309		a) an explicit hardware power-up operation, resulting310		   from a policy decision on another CPU;311 312		b) a hardware event, such as an interrupt.313 314 315CLUSTER_DOWN/INBOUND_COMING_UP:316 317	In this state, an inbound CPU sets up the cluster, including318	enabling of hardware coherency at the cluster level and any319	other operations (such as cache invalidation) which are required320	in order to achieve this.321 322	The purpose of this state is to do sufficient cluster-level323	setup to enable other CPUs in the cluster to enter coherency324	safely.325 326	Next state:327		CLUSTER_UP/INBOUND_COMING_UP (inbound)328	Conditions:329		cluster-level setup and hardware coherency complete330	Trigger events:331		(spontaneous)332 333 334CLUSTER_UP/INBOUND_COMING_UP:335 336	Cluster-level setup is complete and hardware coherency is337	enabled for the cluster.  Other CPUs in the cluster can safely338	enter coherency.339 340	This is a transient state, leading immediately to341	CLUSTER_UP/INBOUND_NOT_COMING_UP.  All other CPUs on the cluster342	should consider treat these two states as equivalent.343 344	Next state:345		CLUSTER_UP/INBOUND_NOT_COMING_UP (inbound)346	Conditions:347		none348	Trigger events:349		(spontaneous)350 351 352CLUSTER_UP/INBOUND_NOT_COMING_UP:353 354	Cluster-level setup is complete and hardware coherency is355	enabled for the cluster.  Other CPUs in the cluster can safely356	enter coherency.357 358	The cluster will remain in this state until a policy decision is359	made to power the cluster down.360 361	Next state:362		CLUSTER_GOING_DOWN/INBOUND_NOT_COMING_UP (outbound)363	Conditions:364		none365	Trigger events:366		policy decision to power down the cluster367 368 369CLUSTER_GOING_DOWN/INBOUND_NOT_COMING_UP:370 371	An outbound CPU is tearing the cluster down.  The selected CPU372	must wait in this state until all CPUs in the cluster are in the373	CPU_DOWN state.374 375	When all CPUs are in the CPU_DOWN state, the cluster can be torn376	down, for example by cleaning data caches and exiting377	cluster-level coherency.378 379	To avoid wasteful unnecessary teardown operations, the outbound380	should check the inbound cluster state for asynchronous381	transitions to INBOUND_COMING_UP.  Alternatively, individual382	CPUs can be checked for entry into CPU_COMING_UP or CPU_UP.383 384 385	Next states:386 387	CLUSTER_DOWN/INBOUND_NOT_COMING_UP (outbound)388		Conditions:389			cluster torn down and ready to power off390		Trigger events:391			(spontaneous)392 393	CLUSTER_GOING_DOWN/INBOUND_COMING_UP (inbound)394		Conditions:395			none396 397		Trigger events:398			a) an explicit hardware power-up operation,399			   resulting from a policy decision on another400			   CPU;401 402			b) a hardware event, such as an interrupt.403 404 405CLUSTER_GOING_DOWN/INBOUND_COMING_UP:406 407	The cluster is (or was) being torn down, but another CPU has408	come online in the meantime and is trying to set up the cluster409	again.410 411	If the outbound CPU observes this state, it has two choices:412 413		a) back out of teardown, restoring the cluster to the414		   CLUSTER_UP state;415 416		b) finish tearing the cluster down and put the cluster417		   in the CLUSTER_DOWN state; the inbound CPU will418		   set up the cluster again from there.419 420	Choice (a) permits the removal of some latency by avoiding421	unnecessary teardown and setup operations in situations where422	the cluster is not really going to be powered down.423 424 425	Next states:426 427	CLUSTER_UP/INBOUND_COMING_UP (outbound)428		Conditions:429				cluster-level setup and hardware430				coherency complete431 432		Trigger events:433				(spontaneous)434 435	CLUSTER_DOWN/INBOUND_COMING_UP (outbound)436		Conditions:437			cluster torn down and ready to power off438 439		Trigger events:440			(spontaneous)441 442 443Last man and First man selection444--------------------------------445 446The CPU which performs cluster tear-down operations on the outbound side447is commonly referred to as the "last man".448 449The CPU which performs cluster setup on the inbound side is commonly450referred to as the "first man".451 452The race avoidance algorithm documented above does not provide a453mechanism to choose which CPUs should play these roles.454 455 456Last man:457 458When shutting down the cluster, all the CPUs involved are initially459executing Linux and hence coherent.  Therefore, ordinary spinlocks can460be used to select a last man safely, before the CPUs become461non-coherent.462 463 464First man:465 466Because CPUs may power up asynchronously in response to external wake-up467events, a dynamic mechanism is needed to make sure that only one CPU468attempts to play the first man role and do the cluster-level469initialisation: any other CPUs must wait for this to complete before470proceeding.471 472Cluster-level initialisation may involve actions such as configuring473coherency controls in the bus fabric.474 475The current implementation in mcpm_head.S uses a separate mutual exclusion476mechanism to do this arbitration.  This mechanism is documented in477detail in vlocks.txt.478 479 480Features and Limitations481------------------------482 483Implementation:484 485	The current ARM-based implementation is split between486	arch/arm/common/mcpm_head.S (low-level inbound CPU operations) and487	arch/arm/common/mcpm_entry.c (everything else):488 489	__mcpm_cpu_going_down() signals the transition of a CPU to the490	CPU_GOING_DOWN state.491 492	__mcpm_cpu_down() signals the transition of a CPU to the CPU_DOWN493	state.494 495	A CPU transitions to CPU_COMING_UP and then to CPU_UP via the496	low-level power-up code in mcpm_head.S.  This could497	involve CPU-specific setup code, but in the current498	implementation it does not.499 500	__mcpm_outbound_enter_critical() and __mcpm_outbound_leave_critical()501	handle transitions from CLUSTER_UP to CLUSTER_GOING_DOWN502	and from there to CLUSTER_DOWN or back to CLUSTER_UP (in503	the case of an aborted cluster power-down).504 505	These functions are more complex than the __mcpm_cpu_*()506	functions due to the extra inter-CPU coordination which507	is needed for safe transitions at the cluster level.508 509	A cluster transitions from CLUSTER_DOWN back to CLUSTER_UP via510	the low-level power-up code in mcpm_head.S.  This511	typically involves platform-specific setup code,512	provided by the platform-specific power_up_setup513	function registered via mcpm_sync_init.514 515Deep topologies:516 517	As currently described and implemented, the algorithm does not518	support CPU topologies involving more than two levels (i.e.,519	clusters of clusters are not supported).  The algorithm could be520	extended by replicating the cluster-level states for the521	additional topological levels, and modifying the transition522	rules for the intermediate (non-outermost) cluster levels.523 524 525Colophon526--------527 528Originally created and documented by Dave Martin for Linaro Limited, in529collaboration with Nicolas Pitre and Achin Gupta.530 531Copyright (C) 2012-2013  Linaro Limited532Distributed under the terms of Version 2 of the GNU General Public533License, as defined in linux/COPYING.534