2Concurrency Managed Workqueue (cmwq)
4September, 2010 Tejun Heo <>
5            Florian Mickler <>
91. Introduction
102. Why cmwq?
113. The Design
124. Application Programming Interface (API)
135. Example Execution Scenarios
146. Guidelines
171. Introduction
19There are many cases where an asynchronous process execution context
20is needed and the workqueue (wq) API is the most commonly used
21mechanism for such cases.
23When such an asynchronous execution context is needed, a work item
24describing which function to execute is put on a queue. An
25independent thread serves as the asynchronous execution context. The
26queue is called workqueue and the thread is called worker.
28While there are work items on the workqueue the worker executes the
29functions associated with the work items one after the other. When
30there is no work item left on the workqueue the worker becomes idle.
31When a new work item gets queued, the worker begins executing again.
342. Why cmwq?
36In the original wq implementation, a multi threaded (MT) wq had one
37worker thread per CPU and a single threaded (ST) wq had one worker
38thread system-wide. A single MT wq needed to keep around the same
39number of workers as the number of CPUs. The kernel grew a lot of MT
40wq users over the years and with the number of CPU cores continuously
41rising, some systems saturated the default 32k PID space just booting
44Although MT wq wasted a lot of resource, the level of concurrency
45provided was unsatisfactory. The limitation was common to both ST and
46MT wq albeit less severe on MT. Each wq maintained its own separate
47worker pool. A MT wq could provide only one execution context per CPU
48while a ST wq one for the whole system. Work items had to compete for
49those very limited execution contexts leading to various problems
50including proneness to deadlocks around the single execution context.
52The tension between the provided level of concurrency and resource
53usage also forced its users to make unnecessary tradeoffs like libata
54choosing to use ST wq for polling PIOs and accepting an unnecessary
55limitation that no two polling PIOs can progress at the same time. As
56MT wq don't provide much better concurrency, users which require
57higher level of concurrency, like async or fscache, had to implement
58their own thread pool.
60Concurrency Managed Workqueue (cmwq) is a reimplementation of wq with
61focus on the following goals.
63* Maintain compatibility with the original workqueue API.
65* Use per-CPU unified worker pools shared by all wq to provide
66  flexible level of concurrency on demand without wasting a lot of
67  resource.
69* Automatically regulate worker pool and level of concurrency so that
70  the API users don't need to worry about such details.
733. The Design
75In order to ease the asynchronous execution of functions a new
76abstraction, the work item, is introduced.
78A work item is a simple struct that holds a pointer to the function
79that is to be executed asynchronously. Whenever a driver or subsystem
80wants a function to be executed asynchronously it has to set up a work
81item pointing to that function and queue that work item on a
84Special purpose threads, called worker threads, execute the functions
85off of the queue, one after the other. If no work is queued, the
86worker threads become idle. These worker threads are managed in so
87called thread-pools.
89The cmwq design differentiates between the user-facing workqueues that
90subsystems and drivers queue work items on and the backend mechanism
91which manages thread-pool and processes the queued work items.
93The backend is called gcwq. There is one gcwq for each possible CPU
94and one gcwq to serve work items queued on unbound workqueues.
96Subsystems and drivers can create and queue work items through special
97workqueue API functions as they see fit. They can influence some
98aspects of the way the work items are executed by setting flags on the
99workqueue they are putting the work item on. These flags include
100things like CPU locality, reentrancy, concurrency limits and more. To
101get a detailed overview refer to the API description of
102alloc_workqueue() below.
104When a work item is queued to a workqueue, the target gcwq is
105determined according to the queue parameters and workqueue attributes
106and appended on the shared worklist of the gcwq. For example, unless
107specifically overridden, a work item of a bound workqueue will be
108queued on the worklist of exactly that gcwq that is associated to the
109CPU the issuer is running on.
111For any worker pool implementation, managing the concurrency level
112(how many execution contexts are active) is an important issue. cmwq
113tries to keep the concurrency at a minimal but sufficient level.
114Minimal to save resources and sufficient in that the system is used at
115its full capacity.
117Each gcwq bound to an actual CPU implements concurrency management by
118hooking into the scheduler. The gcwq is notified whenever an active
119worker wakes up or sleeps and keeps track of the number of the
120currently runnable workers. Generally, work items are not expected to
121hog a CPU and consume many cycles. That means maintaining just enough
122concurrency to prevent work processing from stalling should be
123optimal. As long as there are one or more runnable workers on the
124CPU, the gcwq doesn't start execution of a new work, but, when the
125last running worker goes to sleep, it immediately schedules a new
126worker so that the CPU doesn't sit idle while there are pending work
127items. This allows using a minimal number of workers without losing
128execution bandwidth.
130Keeping idle workers around doesn't cost other than the memory space
131for kthreads, so cmwq holds onto idle ones for a while before killing
134For an unbound wq, the above concurrency management doesn't apply and
135the gcwq for the pseudo unbound CPU tries to start executing all work
136items as soon as possible. The responsibility of regulating
137concurrency level is on the users. There is also a flag to mark a
138bound wq to ignore the concurrency management. Please refer to the
139API section for details.
141Forward progress guarantee relies on that workers can be created when
142more execution contexts are necessary, which in turn is guaranteed
143through the use of rescue workers. All work items which might be used
144on code paths that handle memory reclaim are required to be queued on
145wq's that have a rescue-worker reserved for execution under memory
146pressure. Else it is possible that the thread-pool deadlocks waiting
147for execution contexts to free up.
1504. Application Programming Interface (API)
152alloc_workqueue() allocates a wq. The original create_*workqueue()
153functions are deprecated and scheduled for removal. alloc_workqueue()
154takes three arguments - @name, @flags and @max_active. @name is the
155name of the wq and also used as the name of the rescuer thread if
156there is one.
158A wq no longer manages execution resources but serves as a domain for
159forward progress guarantee, flush and work item attributes. @flags
160and @max_active control how work items are assigned execution
161resources, scheduled and executed.
167    By default, a wq guarantees non-reentrance only on the same
168    CPU. A work item may not be executed concurrently on the same
169    CPU by multiple workers but is allowed to be executed
170    concurrently on multiple CPUs. This flag makes sure
171    non-reentrance is enforced across all CPUs. Work items queued
172    to a non-reentrant wq are guaranteed to be executed by at most
173    one worker system-wide at any given time.
177    Work items queued to an unbound wq are served by a special
178    gcwq which hosts workers which are not bound to any specific
179    CPU. This makes the wq behave as a simple execution context
180    provider without concurrency management. The unbound gcwq
181    tries to start execution of work items as soon as possible.
182    Unbound wq sacrifices locality but is useful for the following
183    cases.
185    * Wide fluctuation in the concurrency level requirement is
186      expected and using bound wq may end up creating large number
187      of mostly unused workers across different CPUs as the issuer
188      hops through different CPUs.
190    * Long running CPU intensive workloads which can be better
191      managed by the system scheduler.
195    A freezable wq participates in the freeze phase of the system
196    suspend operations. Work items on the wq are drained and no
197    new work item starts execution until thawed.
201    All wq which might be used in the memory reclaim paths _MUST_
202    have this flag set. The wq is guaranteed to have at least one
203    execution context regardless of memory pressure.
207    Work items of a highpri wq are queued at the head of the
208    worklist of the target gcwq and start execution regardless of
209    the current concurrency level. In other words, highpri work
210    items will always start execution as soon as execution
211    resource is available.
213    Ordering among highpri work items is preserved - a highpri
214    work item queued after another highpri work item will start
215    execution after the earlier highpri work item starts.
217    Although highpri work items are not held back by other
218    runnable work items, they still contribute to the concurrency
219    level. Highpri work items in runnable state will prevent
220    non-highpri work items from starting execution.
222    This flag is meaningless for unbound wq.
226    Work items of a CPU intensive wq do not contribute to the
227    concurrency level. In other words, runnable CPU intensive
228    work items will not prevent other work items from starting
229    execution. This is useful for bound work items which are
230    expected to hog CPU cycles so that their execution is
231    regulated by the system scheduler.
233    Although CPU intensive work items don't contribute to the
234    concurrency level, start of their executions is still
235    regulated by the concurrency management and runnable
236    non-CPU-intensive work items can delay execution of CPU
237    intensive work items.
239    This flag is meaningless for unbound wq.
243    This combination makes the wq avoid interaction with
244    concurrency management completely and behave as a simple
245    per-CPU execution context provider. Work items queued on a
246    highpri CPU-intensive wq start execution as soon as resources
247    are available and don't affect execution of other work items.
251@max_active determines the maximum number of execution contexts per
252CPU which can be assigned to the work items of a wq. For example,
253with @max_active of 16, at most 16 work items of the wq can be
254executing at the same time per CPU.
256Currently, for a bound wq, the maximum limit for @max_active is 512
257and the default value used when 0 is specified is 256. For an unbound
258wq, the limit is higher of 512 and 4 * num_possible_cpus(). These
259values are chosen sufficiently high such that they are not the
260limiting factor while providing protection in runaway cases.
262The number of active work items of a wq is usually regulated by the
263users of the wq, more specifically, by how many work items the users
264may queue at the same time. Unless there is a specific need for
265throttling the number of active work items, specifying '0' is
268Some users depend on the strict execution ordering of ST wq. The
269combination of @max_active of 1 and WQ_UNBOUND is used to achieve this
270behavior. Work items on such wq are always queued to the unbound gcwq
271and only one work item can be active at any given time thus achieving
272the same ordering property as ST wq.
2755. Example Execution Scenarios
277The following example execution scenarios try to illustrate how cmwq
278behave under different configurations.
280 Work items w0, w1, w2 are queued to a bound wq q0 on the same CPU.
281 w0 burns CPU for 5ms then sleeps for 10ms then burns CPU for 5ms
282 again before finishing. w1 and w2 burn CPU for 5ms then sleep for
283 10ms.
285Ignoring all other tasks, works and processing overhead, and assuming
286simple FIFO scheduling, the following is one highly simplified version
287of possible sequences of events with the original wq.
290 0 w0 starts and burns CPU
291 5 w0 sleeps
292 15 w0 wakes up and burns CPU
293 20 w0 finishes
294 20 w1 starts and burns CPU
295 25 w1 sleeps
296 35 w1 wakes up and finishes
297 35 w2 starts and burns CPU
298 40 w2 sleeps
299 50 w2 wakes up and finishes
301And with cmwq with @max_active >= 3,
304 0 w0 starts and burns CPU
305 5 w0 sleeps
306 5 w1 starts and burns CPU
307 10 w1 sleeps
308 10 w2 starts and burns CPU
309 15 w2 sleeps
310 15 w0 wakes up and burns CPU
311 20 w0 finishes
312 20 w1 wakes up and finishes
313 25 w2 wakes up and finishes
315If @max_active == 2,
318 0 w0 starts and burns CPU
319 5 w0 sleeps
320 5 w1 starts and burns CPU
321 10 w1 sleeps
322 15 w0 wakes up and burns CPU
323 20 w0 finishes
324 20 w1 wakes up and finishes
325 20 w2 starts and burns CPU
326 25 w2 sleeps
327 35 w2 wakes up and finishes
329Now, let's assume w1 and w2 are queued to a different wq q1 which has
330WQ_HIGHPRI set,
333 0 w1 and w2 start and burn CPU
334 5 w1 sleeps
335 10 w2 sleeps
336 10 w0 starts and burns CPU
337 15 w0 sleeps
338 15 w1 wakes up and finishes
339 20 w2 wakes up and finishes
340 25 w0 wakes up and burns CPU
341 30 w0 finishes
343If q1 has WQ_CPU_INTENSIVE set,
346 0 w0 starts and burns CPU
347 5 w0 sleeps
348 5 w1 and w2 start and burn CPU
349 10 w1 sleeps
350 15 w2 sleeps
351 15 w0 wakes up and burns CPU
352 20 w0 finishes
353 20 w1 wakes up and finishes
354 25 w2 wakes up and finishes
3576. Guidelines
359* Do not forget to use WQ_MEM_RECLAIM if a wq may process work items
360  which are used during memory reclaim. Each wq with WQ_MEM_RECLAIM
361  set has an execution context reserved for it. If there is
362  dependency among multiple work items used during memory reclaim,
363  they should be queued to separate wq each with WQ_MEM_RECLAIM.
365* Unless strict ordering is required, there is no need to use ST wq.
367* Unless there is a specific need, using 0 for @max_active is
368  recommended. In most use cases, concurrency level usually stays
369  well under the default limit.
371* A wq serves as a domain for forward progress guarantee
372  (WQ_MEM_RECLAIM, flush and work item attributes. Work items which
373  are not involved in memory reclaim and don't need to be flushed as a
374  part of a group of work items, and don't require any special
375  attribute, can use one of the system wq. There is no difference in
376  execution characteristics between using a dedicated wq and a system
377  wq.
379* Unless work items are expected to consume a huge amount of CPU
380  cycles, using a bound wq is usually beneficial due to the increased
381  level of locality in wq operations and work item execution.

Archive Download this file