| /linux-6.12.1/tools/workqueue/ |
| D | wq_monitor.py | 7 This is a drgn script to monitor workqueues. For more info on drgn, visit
23 CMW/RPR For per-cpu workqueues, the number of concurrency-management
25 unbound workqueues, the number of times a worker was repatriated
54 workqueues = prog['workqueues'] variable
154 for wq in list_for_each_entry('struct workqueue_struct', workqueues.address_of_(), 'list'):
|
| D | wq_dump.py | 13 Shows the CPUs that can be used for unbound workqueues and how they will be
36 Lists all workqueues along with their type and worker pool association. For
96 workqueues = prog['workqueues'] variable
181 for wq in list_for_each_entry('struct workqueue_struct', workqueues.address_of_(), 'list'):
202 for wq in list_for_each_entry('struct workqueue_struct', workqueues.address_of_(), 'list'):
241 for wq in list_for_each_entry('struct workqueue_struct', workqueues.address_of_(), 'list'):
|
| /linux-6.12.1/Documentation/driver-api/crypto/iaa/ |
| D | iaa-crypto.rst | 93 With scalable mode, both shared and dedicated workqueues are available
116 With legacy mode, only dedicated workqueues are available for use.
126 With no IOMMU mode, only dedicated workqueues are available for use.
138 the user must first disable the IAA devices and workqueues, reset the
151 workqueues.
173 IAA workqueues need to be bound to the iaa_crypto driver.
195 driver, the available workqueues are 'rebalanced' such that work
215 When there are no IAA workqueues bound to the driver, the IAA crypto
303 The devices and workqueues are also enabled and therefore the driver
312 the enabled devices and workqueues must first be disabled. In order
[all …]
|
| /linux-6.12.1/include/linux/ |
| D | workqueue.h | 379 * Per-cpu workqueues are generally preferred because they tend to
381 * workqueues exclude the scheduler from choosing the CPU to
393 * Workqueues marked with WQ_POWER_EFFICIENT are per-cpu by default
395 * specified. Per-cpu workqueues which are identified to
428 * System-wide workqueues which are always present.
482 * For unbound workqueues, @max_active limits the number of in-flight work items
520 * workqueues created with the same purpose and to avoid leaking a lockdep_map
540 * Useful for workqueues created with the same purpose and to avoid leaking a
559 * implemented as unbound workqueues with @max_active of one.
753 * Detect attempt to flush system-wide workqueues at compile time when possible.
[all …]
|
| D | kfence.h | 71 * up the allocation gate timer, and requires that workqueues are available.
|
| /linux-6.12.1/tools/testing/selftests/bpf/progs/ |
| D | wq.c | 142 /* test that workqueues can be used from an array */
162 /* test that workqueues can be used from a hashmap */
172 /* test that workqueues can be used from a hashmap with NO_PREALLOC. */
182 /* test that workqueues can be used from a LRU map */
|
| /linux-6.12.1/Documentation/core-api/ |
| D | workqueue.rst | 82 For threaded workqueues, special purpose threads, called [k]workers, execute
87 The cmwq design differentiates between the user-facing workqueues that
93 worker-pools to serve work items queued on unbound workqueues - the
96 BH workqueues use the same framework. However, as there can only be one
140 For unbound workqueues, the number of backing pools is dynamic.
176 BH workqueues can be considered a convenience interface to softirq. BH
177 workqueues are always per-CPU and all BH work items are executed in the
180 All BH workqueues must have 0 ``max_active`` and ``WQ_HIGHPRI`` is the
246 unbound workqueues.
384 worker on the same CPU. This makes unbound workqueues behave as per-cpu
[all …]
|
| /linux-6.12.1/kernel/ |
| D | workqueue.c | 22 * pools for workqueues which are not bound to any specific CPU - the
337 struct list_head list; /* PR: list of all workqueues */
374 * the workqueues list without grabbing wq_pool_mutex.
375 * This is used to dump all workqueues from sysrq.
386 * Each pod type describes how CPUs should be grouped for unbound workqueues.
440 static DEFINE_MUTEX(wq_pool_mutex); /* protects pools and workqueues list */
446 static LIST_HEAD(workqueues); /* PR: list of all workqueues */
1294 * workqueues as appropriate. To avoid flooding the console, each violating work
1546 * - %NULL for per-cpu workqueues as they don't need to use shared nr_active.
1806 * This function should only be called for ordered workqueues where only the
[all …]
|
| D | umh.c | 148 * possibly reduced affinity (eg: per-cpu workqueues). We don't want
151 * Unbound workqueues provide such wide affinity and allow to block on
154 * Besides, workqueues provide the privilege level that caller might not have
394 * asynchronously if wait is not set, and runs as a child of system workqueues.
|
| D | async.c | 354 * unbound workqueues can handle only upto min_active interdependent in async_init()
|
| /linux-6.12.1/kernel/power/ |
| D | Kconfig | 336 Per-cpu workqueues are generally preferred because they show
338 per-cpu workqueues tend to be more power hungry than unbound
339 workqueues.
342 per-cpu workqueues which were observed to contribute
|
| /linux-6.12.1/Documentation/admin-guide/ |
| D | kernel-per-CPU-kthreads.rst | 149 2. Convert all drivers that you must use from tasklets to workqueues.
236 files. The set of WQ_SYSFS workqueues can be displayed using
237 "ls /sys/devices/virtual/workqueue". That said, the workqueues
239 sprinkling WQ_SYSFS across all the workqueues. The reason for
|
| D | xfs.rst | 507 XFS uses kernel workqueues to parallelize metadata update processes. This
516 The control knobs for a filesystem's workqueues are organized by task at hand
534 The interesting knobs for XFS workqueues are as follows:
|
| /linux-6.12.1/Documentation/arch/x86/ |
| D | sva.rst | 31 Shared Hardware Workqueues
153 * Devices have a limited number (~10's to 1000's) of hardware workqueues.
154 The device driver manages allocating hardware workqueues.
184 * How are shared workqueues different?
|
| /linux-6.12.1/Documentation/RCU/Design/Expedited-Grace-Periods/ |
| D | Expedited-Grace-Periods.rst | 402 Use of Workqueues
409 workqueues (see Documentation/core-api/workqueue.rst).
462 The use of workqueues has the advantage that the expedited grace-period
464 corresponding disadvantage that workqueues cannot be used until they are
472 the use of workqueues. However, the requesting task is only required to
475 workqueues are used.
|
| /linux-6.12.1/fs/btrfs/ |
| D | volumes.h | 311 * Checksum mode - offload it to workqueues or do it synchronously in
317 * Do it synchronously if the checksum is fast, or offload to workqueues
321 /* Always offload checksum to workqueues. */
323 /* Never offload checksum to workqueues. */
|
| /linux-6.12.1/drivers/net/wwan/t7xx/ |
| D | t7xx_modem_ops.h | 34 * @HIF_EX_CLEARQ_DONE: Disable RX, flush TX/RX workqueues and clear RX.
|
| D | t7xx_modem_ops.c | 274 * HIF_EX_CLEARQ_DONE: Disable RX, flush TX/RX workqueues and clear RX.
293 * +------------------+ : Flush TX/RX workqueues
|
| /linux-6.12.1/drivers/watchdog/ |
| D | softdog.c | 94 * Also, executing it in system-wide workqueues blocks in softdog_fire()
|
| /linux-6.12.1/kernel/sched/ |
| D | isolation.c | 4 * any CPU: unbound workqueues, timers, kthreads and any offloadable work.
|
| /linux-6.12.1/Documentation/admin-guide/device-mapper/ |
| D | dm-crypt.rst | 117 Set dm-crypt workqueues and the writer thread to high priority. This
|
| /linux-6.12.1/drivers/bus/ |
| D | mips_cdmm.c | 133 * workqueues. For the standard driver callbacks we need a work function
573 * workqueues. For the CPU callbacks, they need to be called for all devices on
|
| /linux-6.12.1/kernel/rcu/ |
| D | tree_exp.h | 539 // Workqueues should not be signaled. in synchronize_rcu_expedited_wait_once()
542 WARN_ON(t < 0); /* workqueues should not be signaled. */ in synchronize_rcu_expedited_wait_once()
704 * workqueues and mid-boot-time tasks.
|
| /linux-6.12.1/Documentation/devicetree/bindings/media/ |
| D | mediatek,vcodec-subdev-decoder.yaml | 38 There are two workqueues in parent device: lat workqueue and core workqueue. They are used
|
| /linux-6.12.1/fs/xfs/scrub/ |
| D | nlinks_repair.c | 342 * We can only push the inactivation workqueues with an empty in xrep_nlinks()
|