Home
last modified time | relevance | path

Searched refs:scheduler (Results 1 – 25 of 26) sorted by relevance

12

/xnu-11417.121.6/doc/scheduler/
H A Dsched_clutch_edge.md7scheduler attempts to achieve these goals by expecting all threads in the system to be tagged with…
10scheduler, timesharing is achieved by decaying the priority of threads depending on global system …
12scheduler is the timesharing algorithm for threads on a single cluster. The **Edge scheduler** ext…
16scheduler schedules groups of threads instead of individual threads. Breaking away from the tradit…
24 …kets for threads bounded to this hierarchy and cluster which allows the scheduler to timeshare eff…
28 …aring algorithm to allow the system to function similar to the existing scheduler from a higher le…
41 … lead to performance issues. In order to address that, the bucket level scheduler implements a roo…
53scheduler would continue to select this warping bucket over lower priority buckets. Once the warpi…
61 * Deadline based scheduling allows the scheduler to define strict bounds on worst case execution la…
64 * Since the bucket level scheduler deals with a fixed small number of runnable buckets in the worst…
[all …]
/xnu-11417.121.6/tests/sched/sched_test_harness/
H A DREADME.md4scheduler test harness makes it possible to test scheduler policy code deterministically from with…
7scheduler code this way has been in unraveling dependencies that would grow the amount and complex…
19 …eue_harness.h` to validate implementations of thread timesharing in the scheduler. For example, te…
22 …le, tests can create a mock HW topology and validate which clusters the scheduler would send certa…
25 …to specially annotate test output and to prepend the name of a specific scheduler policy-under-tes…
28 Clutch and Edge are the two scheduler policies currently testable using the harness. They each publ…
/xnu-11417.121.6/tests/
H A Daqm_qdelay_utun.c141 static uint32_t scheduler; variable
166 scheduler = ifcqs->ifqs_scheduler; in aqmstats_setup()
226 if (scheduler == PKTSCHEDT_FQ_CODEL) {
245 if (scheduler == PKTSCHEDT_FQ_CODEL) {
H A Dstackshot_block_owner_14362384.m34 T_META_NAMESPACE("xnu.scheduler"),
/xnu-11417.121.6/tools/lldbmacros/
H A Dulock.py2 from scheduler import GetRecentTimestamp
H A Dtaskinfo.py2 from scheduler import GetRecentTimestamp
H A Dmemorystatus.py5 from scheduler import GetRecentTimestamp
H A Dvm_pageout.py10 from scheduler import GetRecentTimestamp
H A DMakefile70 scheduler.py \
H A Dworkqueue.py3 from scheduler import GetRecentTimestamp
H A Dmisc.py8 from scheduler import *
H A Dktrace.py5 from scheduler import *
H A Dxnu.py1631 from scheduler import *
H A Dprocess.py1602 from scheduler import GetSchedMostRecentDispatch
/xnu-11417.121.6/config/
H A DMASTER.x86_6495 options CONFIG_SCHED_SMT # SMT-aware scheduler policies # <config_sched_smt>
H A DMASTER671 # Timeshare scheduler implementations
/xnu-11417.121.6/bsd/net/pktsched/
H A Dpktsched.c116 pktsched_setup(struct ifclassq *ifq, u_int32_t scheduler, u_int32_t sflags, in pktsched_setup() argument
127 if (ifq->ifcq_type == scheduler) { in pktsched_setup()
/xnu-11417.121.6/doc/lifecycle/
H A Dstartup.md75 - Rank 2: `ml_io_timeouts_init` for scheduler hygiene.
/xnu-11417.121.6/tests/skywalk/
H A Dskywalk_test_common.c2994 u_int32_t scheduler; in skt_aqstatpr() local
3032 scheduler = ifcqs->ifqs_scheduler; in skt_aqstatpr()
3043 for (n = 0; n < IFCQ_SC_MAX && scheduler != PKTSCHEDT_NONE; n++) { in skt_aqstatpr()
3052 switch (scheduler) { in skt_aqstatpr()
/xnu-11417.121.6/osfmk/kern/
H A Dsched_prim.h854 #error Enable at least one scheduler algorithm in osfmk/conf/MASTER.XXX
H A Drecount.c518 #pragma mark - scheduler
H A Dexclaves.c1217 #pragma mark exclaves scheduler communication
/xnu-11417.121.6/doc/observability/
H A Dmt_stackshot.md36 the scheduler and with interrupts disabled, it does not use "actual" threads to
/xnu-11417.121.6/bsd/net/
H A Dpfvar.h1768 u_int32_t scheduler; /* scheduler type */ member
2015 u_int8_t scheduler; member
/xnu-11417.121.6/doc/vm/
H A Dmemorystatus.md76 … use finer grain locking or at least not share the global lock with the scheduler. See [rdar://363…

12