NUMA topologies where the shortest path between some two nodes requires three or more hops (i.e. diameter > 2) end up being misrepresented in the scheduler topology structures. This is currently detected when booting a kernel with CONFIG_SCHED_DEBUG=y + sched_debug on the cmdline, although this will only yield a warning about sched_group spans not matching sched_domain spans: ERROR: groups don't span domain->span Add an explicit warning for that case, triggered regardless of CONFIG_SCHED_DEBUG, and decorate it with an appropriate comment. The topology described in the comment can be booted up on QEMU by appending the following to your usual QEMU incantation: -smp cores=4 \ -numa node,cpus=0,nodeid=0 -numa node,cpus=1,nodeid=1, \ -numa node,cpus=2,nodeid=2, -numa node,cpus=3,nodeid=3, \ -numa dist,src=0,dst=1,val=20, -numa dist,src=0,dst=2,val=30, \ -numa dist,src=0,dst=3,val=40, -numa dist,src=1,dst=2,val=20, \ -numa dist,src=1,dst=3,val=30, -numa dist,src=2,dst=3,val=20 A somewhat more realistic topology (6-node mesh) with the same affliction can be conjured with: -smp cores=6 \ -numa node,cpus=0,nodeid=0 -numa node,cpus=1,nodeid=1, \ -numa node,cpus=2,nodeid=2, -numa node,cpus=3,nodeid=3, \ -numa node,cpus=4,nodeid=4, -numa node,cpus=5,nodeid=5, \ -numa dist,src=0,dst=1,val=20, -numa dist,src=0,dst=2,val=30, \ -numa dist,src=0,dst=3,val=40, -numa dist,src=0,dst=4,val=30, \ -numa dist,src=0,dst=5,val=20, \ -numa dist,src=1,dst=2,val=20, -numa dist,src=1,dst=3,val=30, \ -numa dist,src=1,dst=4,val=20, -numa dist,src=1,dst=5,val=30, \ -numa dist,src=2,dst=3,val=20, -numa dist,src=2,dst=4,val=30, \ -numa dist,src=2,dst=5,val=40, \ -numa dist,src=3,dst=4,val=20, -numa dist,src=3,dst=5,val=30, \ -numa dist,src=4,dst=5,val=20
Showing the changes to kernel (1 of 1 file). Show all
Unified · Split
| @@ -675,6 +675,7 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) · 674 unmodified lines | ||
| 675 | 675 | { |
| 676 | 676 | struct rq *rq = cpu_rq(cpu); |
| 677 | 677 | struct sched_domain *tmp; |
| 678 | + int numa_distance = 0; | |
| 678 | 679 | |
| 679 | 680 | /* Remove the sched domains which do not contribute to scheduling. */ |
| 680 | 681 | for (tmp = sd; tmp; ) { |
| @@ -706,6 +707,38 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) · 25 unmodified lines | ||
| 706 | 707 | sd->child = NULL; |
| 707 | 708 | } |
| 708 | 709 | |
| 710 | + for (tmp = sd; tmp; tmp = tmp->parent) | |
| 711 | + numa_distance += !!(tmp->flags & SD_NUMA); | |
| 712 | + | |
| 713 | + /* | |
| 714 | + * FIXME: Diameter >=3 is misrepresented. | |
| 715 | + * | |
| 716 | + * Smallest diameter=3 topology is: | |
| 717 | + * | |
| 718 | + * node 0 1 2 3 | |
| 719 | + * 0: 10 20 30 40 | |
| 720 | + * 1: 20 10 20 30 | |
| 721 | + * 2: 30 20 10 20 | |
| 722 | + * 3: 40 30 20 10 | |
| 723 | + * | |
| 724 | + * 0 --- 1 --- 2 --- 3 | |
| 725 | + * | |
| 726 | + * NUMA-3 0-3 N/A N/A 0-3 | |
| 727 | + * groups: {0-2},{1-3} {1-3},{0-2} | |
| 728 | + * | |
| 729 | + * NUMA-2 0-2 0-3 0-3 1-3 | |
| 730 | + * groups: {0-1},{1-3} {0-2},{2-3} {1-3},{0-1} {2-3},{0-2} | |
| 731 | + * | |
| 732 | + * NUMA-1 0-1 0-2 1-3 2-3 | |
| 733 | + * groups: {0},{1} {1},{2},{0} {2},{3},{1} {3},{2} | |
| 734 | + * | |
| 735 | + * NUMA-0 0 1 2 3 | |
| 736 | + * | |
| 737 | + * The NUMA-2 groups for nodes 0 and 3 are obviously buggered, as the | |
| 738 | + * group span isn't a subset of the domain span. | |
| 739 | + */ | |
| 740 | + WARN_ONCE(numa_distance > 2, "Shortest NUMA path spans too many nodes\n"); | |
| 741 | + | |
| 709 | 742 | sched_domain_debug(sd, cpu); |
| 710 | 743 | |
| 711 | 744 | rq_attach_root(rq, rd); |