sched/topology: Restore SD_PREFER_SIBLING on MC domains

Peter Zijlstra · Ingo Molnar · commit ed4ad1ca08a5 · 2017-10-10T11:45:27.000+02:00
The normal x86_topology on NHM+ machines degenerates because the MC and CPU domains are of the same size, therefore MC inherits SD_PREFER_SIBLING from CPU (which then gets taken out). The result is that we'll spread tasks across the first NUMA level in order to maximize cache utilization. However, for the x86_numa_in_package_topology we loose the CPU domain, and we'll not have SD_PREFER_SIBLING set anywhere, giving a distinct difference in behaviour. Commit: 8e7fbcb ("sched: Remove stale power aware scheduling remnants and dysfunctional knobs") made a fail by not preserving the SD_PREFER_SIBLING for the !power_saving case on both CPU and MC. Then commit: 6956dc5 ("sched/numa: Add SD_PERFER_SIBLING to CPU domain") adds it back to the CPU but not MC. Restore that now, such that we get consistent spreading behaviour wrt L3 and NUMA. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: Ingo Molnar <mingo@kernel.org>
diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
@@ -1157,6 +1157,7 @@ sd_init(struct sched_domain_topology_level *tl,
 		sd->smt_gain = 1178; /* ~15% */
 
 	} else if (sd->flags & SD_SHARE_PKG_RESOURCES) {
+		sd->flags |= SD_PREFER_SIBLING;
 		sd->imbalance_pct = 117;
 		sd->cache_nice_tries = 1;
 		sd->busy_idx = 2;
