Lines Matching refs:scheduler

5 …scheduler attempts to achieve these goals by expecting all threads in the system to be tagged with…
8 …scheduler, timesharing is achieved by decaying the priority of threads depending on global system …
10 …scheduler is the timesharing algorithm for threads on a single cluster. The **Edge scheduler** ext…
14 …scheduler schedules groups of threads instead of individual threads. Breaking away from the tradit…
22 …kets for threads bounded to this hierarchy and cluster which allows the scheduler to timeshare eff…
26 …aring algorithm to allow the system to function similar to the existing scheduler from a higher le…
39 … lead to performance issues. In order to address that, the bucket level scheduler implements a roo…
51 …scheduler would continue to select this warping bucket over lower priority buckets. Once the warpi…
59 * Deadline based scheduling allows the scheduler to define strict bounds on worst case execution la…
62 * Since the bucket level scheduler deals with a fixed small number of runnable buckets in the worst…
70 …scheduler to decide which thread group should be selected next for execution. Each thread group wi…
74 …scheduler to honor priority differences specified from userspace via SPIs, priority boosts due to …
75 …scheduler calculates an interactivity score based on the ratio of voluntary blocking time and CPU …
76 …ype and criticality of work being done as part of the thread group. The scheduler uses these prope…
87 * Since the priority calculation is fairly cheap, the scheduler is able to maintain up-to-date info…
94 …scheduler decides which thread within a clutch bucket should be selected next for execution. Each …
98 …scheduler implements the Mach timesharing algorithm to decide which thread within the clutch bucke…
111 The per-bucket thread quantum allows the scheduler to bound the worst case execution latency for a …
117 The scheduler maintains a root level priority for the hierarchy in order to make decisions regardin…
160 * Every scheduler tick, snapshot the load for the clutch bucket
167 …scheduler implements all the necessary features needed for scheduling on multi-cluster asymmetric …
183 When the scheduler and performance controller for Skye (the first AMP platform) were being designed…
187 …scheduler uses **per-thread group recommendations** from the performance controller and **per-clus…
191 …scheduler expects the performance controller to specify a cluster recommendation for each thread g…
193 …scheduler looks at the preferred cluster recommendation of the sched\_clutch\_bucket_group it belo…
195 …nges the preferred cluster for a sched\_clutch\_bucket\_group, the Edge scheduler also provides an…
199 …scheduler uses the preferred cluster for the thread's sched\_clutch\_bucket_group. If the preferre…
203 The Edge scheduler maintains a thread migration graph where each node represents a cluster and each…
223 The Edge scheduler maintains a per-cluster scheduling latency metric which indicates the latency fo…
249 The `SCHED(steal_thread)` scheduler callout is invoked when the processor does not find any thread …
253 The Edge scheduler identifies clutch buckets (and correspondingly the threads in the clutch bucket)…
255 **Edge scheduler steal implementation**
257 The edge scheduler implements the steal operation via `sched_edge_processor_idle()`. This routine t…
266 …t from its preferred cluster type. If no such thread is found, then the scheduler aims to reduce t…
268 **Edge scheduler rebalance operation**
274 The Edge scheduler attempts to load balance cluster shared resource intensive threads across cluste…
283 …This policy is different from the default scheduling policy of the edge scheduler since this alway…
287 …scheduler implements a policy to round robining long running workload threads across clusters of v…