Lines Matching full:topology
3 * Scheduler topology setup/handling methods
765 * NUMA topology (first read the regular topology blurb below)
775 * which represents a 4 node ring topology like:
787 * For the above NUMA topology that gives 3 levels:
798 * As can be seen; things don't nicely line up as with the regular topology.
801 * the topology.
828 * Another 'interesting' topology is:
844 * This topology is asymmetric, nodes 1,2 are fully connected, but nodes 0,3
969 * As we are referencing sgc across different topology level, we need in find_descended_sibling()
1021 * Smallest diameter=3 topology is: in build_overlap_sched_groups()
1077 * Package topology (also see the load-balance blurb in fair.c)
1080 * topology features. By default (default_topology[]) these include:
1095 * denote the ever growing mask of CPUs belonging to that level of topology.
1119 * topology levels, while sched_group moves you sideways through it, at child
1233 * unless there are asymmetries in the topology. If there are asymmetries,
1348 * with state indicating CPU topology changes.
1490 * SD_flags allowed in topology descriptions.
1492 * These flags are purely descriptive of the topology and do not prescribe
1500 * Odd one out, which beside describing the topology has a quirk also
1533 "wrong sd_flags in topology description\n")) in sd_init()
1624 * Topology list, bottom-up.
1695 * A system can have three types of NUMA topology:
1700 * The difference between a glueless mesh topology and a backplane
1701 * topology lies in whether communication between not directly
1706 * The type of topology can be discerned with the following tests:
1859 /* Compute default topology size */ in sched_init_numa()
1868 * Copy the default topology bits.. in sched_init_numa()
1942 * A new node has been brought up, potentially changing the topology in __sched_domains_numa_masks_set()
1980 * sched_numa_find_closest() - given the NUMA topology, find the cpu
2119 pr_err("BUG: arch topology borken\n"); in build_sched_domain()
2137 * Ensure topology masks are sane, i.e. there are no conflicts (overlaps) for
2138 * any two given CPUs at this (non-NUMA) topology level.
2160 * match the topology we're about to build, but that can only in topology_span_sane()
2288 * CPU core maps. It is supposed to return 1 if the topology changed