summaryrefslogtreecommitdiffstats
path: root/kernel/sched
diff options
context:
space:
mode:
authorAbel Wu <wuyun.abel@bytedance.com>2023-11-15 11:36:46 +0800
committerPeter Zijlstra <peterz@infradead.org>2023-11-15 09:57:47 +0100
commitee4373dc902c0a403dd084b254ce70a78f95466f (patch)
tree75eb05528687edac88215593be6bb7ba6d29c69a /kernel/sched
parent2227a957e1d5b1941be4e4207879ec74f4bb37f8 (diff)
downloadlinux-ee4373dc902c0a403dd084b254ce70a78f95466f.tar.gz
linux-ee4373dc902c0a403dd084b254ce70a78f95466f.tar.bz2
linux-ee4373dc902c0a403dd084b254ce70a78f95466f.zip
sched/eevdf: O(1) fastpath for task selection
Since the RB-tree is now sorted by deadline, let's first try the leftmost entity which has the earliest virtual deadline. I've done some benchmarks to see its effectiveness. All the benchmarks are done inside a normal cpu cgroup in a clean environment with cpu turbo disabled, on a dual-CPU Intel Xeon(R) Platinum 8260 with 2 NUMA nodes each of which has 24C/48T. hackbench: process/thread + pipe/socket + 1/2/4/8 groups netperf: TCP/UDP + STREAM/RR + 24/48/72/96/192 threads tbench: loopback 24/48/72/96/192 threads schbench: 1/2/4/8 mthreads direct: cfs_rq has only one entity parity: RUN_TO_PARITY fast: O(1) fastpath slow: heap search (%) direct parity fast slow hackbench 92.95 2.02 4.91 0.12 netperf 68.08 6.60 24.18 1.14 tbench 67.55 11.22 20.61 0.62 schbench 69.91 2.65 25.73 1.71 The above results indicate that this fastpath really makes task selection more efficient. Signed-off-by: Abel Wu <wuyun.abel@bytedance.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20231115033647.80785-4-wuyun.abel@bytedance.com
Diffstat (limited to 'kernel/sched')
-rw-r--r--kernel/sched/fair.c14
1 files changed, 11 insertions, 3 deletions
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 31bca05c3612..d3e045d80cab 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -878,6 +878,7 @@ struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq)
static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
{
struct rb_node *node = cfs_rq->tasks_timeline.rb_root.rb_node;
+ struct sched_entity *se = __pick_first_entity(cfs_rq);
struct sched_entity *curr = cfs_rq->curr;
struct sched_entity *best = NULL;
@@ -886,7 +887,7 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
* in this cfs_rq, saving some cycles.
*/
if (cfs_rq->nr_running == 1)
- return curr && curr->on_rq ? curr : __node_2_se(node);
+ return curr && curr->on_rq ? curr : se;
if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr)))
curr = NULL;
@@ -898,9 +899,14 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
if (sched_feat(RUN_TO_PARITY) && curr && curr->vlag == curr->deadline)
return curr;
+ /* Pick the leftmost entity if it's eligible */
+ if (se && entity_eligible(cfs_rq, se)) {
+ best = se;
+ goto found;
+ }
+
/* Heap search for the EEVD entity */
while (node) {
- struct sched_entity *se = __node_2_se(node);
struct rb_node *left = node->rb_left;
/*
@@ -913,6 +919,8 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
continue;
}
+ se = __node_2_se(node);
+
/*
* The left subtree either is empty or has no eligible
* entity, so check the current node since it is the one
@@ -925,7 +933,7 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
node = node->rb_right;
}
-
+found:
if (!best || (curr && entity_before(curr, best)))
best = curr;