From: Abel Wu <wuyun.abel@bytedance.com>
To: Benjamin Segall <bsegall@google.com>,
Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org, vincent.guittot@linaro.org,
linux-kernel@vger.kernel.org, juri.lelli@redhat.com,
dietmar.eggemann@arm.com, rostedt@goodmis.org, mgorman@suse.de,
bristot@redhat.com, corbet@lwn.net, qyousef@layalina.io,
chris.hyser@oracle.com, patrick.bellasi@matbug.net,
pjt@google.com, pavel@ucw.cz, qperret@google.com,
tim.c.chen@linux.intel.com, joshdon@google.com, timj@gnu.org,
kprateek.nayak@amd.com, yu.c.chen@intel.com,
youssefesmat@chromium.org, joel@joelfernandes.org, efault@gmx.de,
tglx@linutronix.de
Subject: Re: [PATCH] sched/fair: fix pick_eevdf to always find the correct se
Date: Wed, 11 Oct 2023 20:12:09 +0800 [thread overview]
Message-ID: <6b606049-3412-437f-af25-a4c33139e2d8@bytedance.com> (raw)
In-Reply-To: <xm261qego72d.fsf_-_@google.com>
On 9/30/23 8:09 AM, Benjamin Segall Wrote:
> The old pick_eevdf could fail to find the actual earliest eligible
> deadline when it descended to the right looking for min_deadline, but it
> turned out that that min_deadline wasn't actually eligible. In that case
> we need to go back and search through any left branches we skipped
> looking for the actual best _eligible_ min_deadline.
>
> This is more expensive, but still O(log n), and at worst should only
> involve descending two branches of the rbtree.
>
> I've run this through a userspace stress test (thank you
> tools/lib/rbtree.c), so hopefully this implementation doesn't miss any
> corner cases.
>
> Fixes: 147f3efaa241 ("sched/fair: Implement an EEVDF-like scheduling policy")
> Signed-off-by: Ben Segall <bsegall@google.com>
> ---
> kernel/sched/fair.c | 72 ++++++++++++++++++++++++++++++++++++---------
> 1 file changed, 58 insertions(+), 14 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 0c31cda0712f..77e9440b8ab3 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -864,18 +864,20 @@ struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq)
> *
> * se->min_deadline = min(se->deadline, se->{left,right}->min_deadline)
> *
> * Which allows an EDF like search on (sub)trees.
> */
> -static struct sched_entity *pick_eevdf(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 *curr = cfs_rq->curr;
> struct sched_entity *best = NULL;
> + struct sched_entity *best_left = NULL;
>
> if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr)))
> curr = NULL;
> + best = curr;
>
> /*
> * Once selected, run a task until it either becomes non-eligible or
> * until it gets a new slice. See the HACK in set_next_entity().
> */
> @@ -892,45 +894,87 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
> node = node->rb_left;
> continue;
> }
>
> /*
> - * If this entity has an earlier deadline than the previous
> - * best, take this one. If it also has the earliest deadline
> - * of its subtree, we're done.
> + * Now we heap search eligible trees for the best (min_)deadline
> */
> - if (!best || deadline_gt(deadline, best, se)) {
> + if (!best || deadline_gt(deadline, best, se))
> best = se;
> - if (best->deadline == best->min_deadline)
> - break;
> - }
>
> /*
> - * If the earlest deadline in this subtree is in the fully
> - * eligible left half of our space, go there.
> + * Every se in a left branch is eligible, keep track of the
> + * branch with the best min_deadline
> */
> + if (node->rb_left) {
> + struct sched_entity *left = __node_2_se(node->rb_left);
> +
> + if (!best_left || deadline_gt(min_deadline, best_left, left))
> + best_left = left;
> +
> + /*
> + * min_deadline is in the left branch. rb_left and all
> + * descendants are eligible, so immediately switch to the second
> + * loop.
> + */
> + if (left->min_deadline == se->min_deadline)
> + break;
> + }
> +
> + /* min_deadline is at this node, no need to look right */
> + if (se->deadline == se->min_deadline)
> + break;
> +
> + /* else min_deadline is in the right branch. */
> + node = node->rb_right;
> + }
> +
> + /*
> + * We ran into an eligible node which is itself the best.
> + * (Or nr_running == 0 and both are NULL)
> + */
> + if (!best_left || (s64)(best_left->min_deadline - best->deadline) > 0)
> + return best;
> +
> + /*
> + * Now best_left and all of its children are eligible, and we are just
> + * looking for deadline == min_deadline
> + */
> + node = &best_left->run_node;
> + while (node) {
> + struct sched_entity *se = __node_2_se(node);
> +
> + /* min_deadline is the current node */
> + if (se->deadline == se->min_deadline)
> + return se;
IMHO it would be better tiebreak on vruntime by moving this hunk to ..
> +
> + /* min_deadline is in the left branch */
> if (node->rb_left &&
> __node_2_se(node->rb_left)->min_deadline == se->min_deadline) {
> node = node->rb_left;
> continue;
> }
.. here, thoughts?
>
> + /* else min_deadline is in the right branch */
> node = node->rb_right;
> }
> + return NULL;
Why not 'best'? Since ..
> +}
>
> - if (!best || (curr && deadline_gt(deadline, best, curr)))
> - best = curr;
> +static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq)
> +{
> + struct sched_entity *se = __pick_eevdf(cfs_rq);
>
> - if (unlikely(!best)) {
> + if (!se) {
> struct sched_entity *left = __pick_first_entity(cfs_rq);
.. cfs_rq->curr isn't considered here.
> if (left) {
> pr_err("EEVDF scheduling fail, picking leftmost\n");
> return left;
> }
> }
>
> - return best;
> + return se;
> }
>
> #ifdef CONFIG_SCHED_DEBUG
> struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq)
> {
The implementation of __pick_eevdf() now is quite complicated which
makes it really hard to maintain. I'm trying my best to refactor this
part, hopefully can do some help. Below is only for illustration, I
need to test more.
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 *curr = cfs_rq->curr;
struct sched_entity *best = NULL;
struct sched_entity *cand = NULL;
bool all_eligible = false;
if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr)))
curr = NULL;
/*
* Once selected, run a task until it either becomes non-eligible or
* until it gets a new slice. See the HACK in set_next_entity().
*/
if (sched_feat(RUN_TO_PARITY) && curr && curr->vlag == curr->deadline)
return curr;
while (node) {
struct sched_entity *se = __node_2_se(node);
/*
* If this entity is not eligible, try the left subtree.
*/
if (!all_eligible && !entity_eligible(cfs_rq, se)) {
node = node->rb_left;
continue;
}
if (node->rb_left) {
struct sched_entity *left = __node_2_se(node->rb_left);
BUG_ON(left->min_deadline < se->min_deadline);
/* Tiebreak on vruntime */
if (left->min_deadline == se->min_deadline) {
node = node->rb_left;
all_eligible = true;
continue;
}
if (!all_eligible) {
/*
* We're going to search right subtree and the one
* with min_deadline can be non-eligible, so record
* the left subtree as a candidate.
*/
if (!cand || deadline_gt(min_deadline, cand, left))
cand = left;
}
}
/* min_deadline is at this node, no need to look right */
if (se->deadline == se->min_deadline) {
best = se;
break;
}
node = node->rb_right;
if (!node && cand) {
node = cand;
all_eligible = true;
cand = NULL;
}
}
if (!best || (curr && deadline_gt(deadline, best, curr)))
best = curr;
return best;
}
next prev parent reply other threads:[~2023-10-11 12:15 UTC|newest]
Thread overview: 124+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-31 11:58 [PATCH 00/15] sched: EEVDF and latency-nice and/or slice-attr Peter Zijlstra
2023-05-31 11:58 ` [PATCH 01/15] sched/fair: Add avg_vruntime Peter Zijlstra
2023-06-02 13:51 ` Vincent Guittot
2023-06-02 14:27 ` Peter Zijlstra
2023-06-05 7:18 ` Vincent Guittot
2023-08-10 7:10 ` [tip: sched/core] sched/fair: Add cfs_rq::avg_vruntime tip-bot2 for Peter Zijlstra
2023-10-11 4:15 ` [PATCH 01/15] sched/fair: Add avg_vruntime Abel Wu
2023-10-11 7:30 ` Peter Zijlstra
2023-10-11 8:30 ` Abel Wu
2023-10-11 9:45 ` Peter Zijlstra
2023-10-11 10:05 ` Peter Zijlstra
2023-10-11 13:08 ` Peter Zijlstra
2023-05-31 11:58 ` [PATCH 02/15] sched/fair: Remove START_DEBIT Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] sched/fair: Remove sched_feat(START_DEBIT) tip-bot2 for Peter Zijlstra
2023-05-31 11:58 ` [PATCH 03/15] sched/fair: Add lag based placement Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2023-10-11 12:00 ` [PATCH 03/15] " Abel Wu
2023-10-11 13:24 ` Peter Zijlstra
2023-10-12 7:04 ` Abel Wu
2023-10-13 7:37 ` Peter Zijlstra
2023-10-13 8:14 ` Abel Wu
2023-10-12 19:15 ` Benjamin Segall
2023-10-12 22:34 ` Peter Zijlstra
2023-10-13 16:35 ` Peter Zijlstra
2023-10-14 8:08 ` Mike Galbraith
2023-10-13 14:34 ` Peter Zijlstra
2023-05-31 11:58 ` [PATCH 04/15] rbtree: Add rb_add_augmented_cached() helper Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2023-05-31 11:58 ` [PATCH 05/15] sched/fair: Implement an EEVDF like policy Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] sched/fair: Implement an EEVDF-like scheduling policy tip-bot2 for Peter Zijlstra
2023-09-29 21:40 ` [PATCH 05/15] sched/fair: Implement an EEVDF like policy Benjamin Segall
2023-10-02 17:39 ` Peter Zijlstra
2023-10-11 4:14 ` Abel Wu
2023-10-11 7:33 ` Peter Zijlstra
2023-10-11 11:49 ` Abel Wu
2023-09-30 0:09 ` [PATCH] sched/fair: fix pick_eevdf to always find the correct se Benjamin Segall
2023-10-03 10:42 ` [tip: sched/urgent] sched/fair: Fix pick_eevdf() tip-bot2 for Benjamin Segall
[not found] ` <CGME20231004203940eucas1p2f73b017497d1f4239a6e236fdb6019e2@eucas1p2.samsung.com>
2023-10-04 20:39 ` [PATCH] sched/fair: fix pick_eevdf to always find the correct se Marek Szyprowski
2023-10-09 7:53 ` [tip: sched/urgent] sched/eevdf: Fix pick_eevdf() tip-bot2 for Benjamin Segall
2023-10-11 12:12 ` Abel Wu [this message]
2023-10-11 13:14 ` [PATCH] sched/fair: fix pick_eevdf to always find the correct se Peter Zijlstra
2023-10-12 10:04 ` Abel Wu
2023-10-11 21:01 ` Benjamin Segall
2023-10-12 10:25 ` Abel Wu
2023-10-12 17:51 ` Benjamin Segall
2023-10-13 3:46 ` Abel Wu
2023-10-13 16:51 ` Benjamin Segall
2023-05-31 11:58 ` [PATCH 06/15] sched: Commit to lag based placement Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] sched/fair: " tip-bot2 for Peter Zijlstra
2023-05-31 11:58 ` [PATCH 07/15] sched/smp: Use lag to simplify cross-runqueue placement Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2023-09-12 15:32 ` [PATCH 07/15] " Sebastian Andrzej Siewior
2023-09-13 9:03 ` Peter Zijlstra
2023-10-04 1:17 ` [PATCH] sched/fair: Preserve PLACE_DEADLINE_INITIAL deadline Daniel Jordan
2023-10-04 13:09 ` [PATCH v2] " Daniel Jordan
2023-10-04 15:46 ` Chen Yu
2023-10-06 16:31 ` Daniel Jordan
2023-10-12 4:48 ` K Prateek Nayak
2023-10-05 5:56 ` [PATCH] " K Prateek Nayak
2023-10-06 16:35 ` Daniel Jordan
2023-10-06 16:48 ` [PATCH] sched/fair: Always update_curr() before placing at enqueue Daniel Jordan
2023-10-06 19:58 ` Peter Zijlstra
2023-10-18 0:43 ` Daniel Jordan
2023-10-16 5:39 ` K Prateek Nayak
2023-05-31 11:58 ` [PATCH 08/15] sched: Commit to EEVDF Peter Zijlstra
2023-06-16 21:23 ` Joel Fernandes
2023-06-22 12:01 ` Ingo Molnar
2023-06-22 13:11 ` Joel Fernandes
2023-08-10 7:10 ` [tip: sched/core] sched/fair: " tip-bot2 for Peter Zijlstra
2023-05-31 11:58 ` [PATCH 09/15] sched/debug: Rename min_granularity to base_slice Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] sched/debug: Rename sysctl_sched_min_granularity to sysctl_sched_base_slice tip-bot2 for Peter Zijlstra
2023-05-31 11:58 ` [PATCH 10/15] sched/fair: Propagate enqueue flags into place_entity() Peter Zijlstra
2023-08-10 7:10 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2023-05-31 11:58 ` [PATCH 11/15] sched/eevdf: Better handle mixed slice length Peter Zijlstra
2023-06-02 13:45 ` Vincent Guittot
2023-06-02 15:06 ` Peter Zijlstra
2023-06-10 6:34 ` Chen Yu
2023-06-10 11:22 ` Peter Zijlstra
2023-05-31 11:58 ` [RFC][PATCH 12/15] sched: Introduce latency-nice as a per-task attribute Peter Zijlstra
2023-05-31 11:58 ` [RFC][PATCH 13/15] sched/fair: Implement latency-nice Peter Zijlstra
2023-06-06 14:54 ` Vincent Guittot
2023-06-08 10:34 ` Peter Zijlstra
2023-06-08 12:44 ` Peter Zijlstra
2023-10-11 23:24 ` Benjamin Segall
2023-05-31 11:58 ` [RFC][PATCH 14/15] sched/fair: Add sched group latency support Peter Zijlstra
2023-05-31 11:58 ` [RFC][PATCH 15/15] sched/eevdf: Use sched_attr::sched_runtime to set request/slice Peter Zijlstra
2023-06-01 13:55 ` Vincent Guittot
2023-06-08 11:52 ` Peter Zijlstra
2023-08-24 0:52 ` [PATCH 00/15] sched: EEVDF and latency-nice and/or slice-attr Daniel Jordan
2023-09-06 13:13 ` Peter Zijlstra
2023-09-29 16:54 ` Youssef Esmat
2023-10-02 15:55 ` Youssef Esmat
2023-10-02 18:41 ` Peter Zijlstra
2023-10-05 12:05 ` Peter Zijlstra
2023-10-05 14:14 ` Peter Zijlstra
2023-10-05 14:42 ` Peter Zijlstra
2023-10-05 18:23 ` Youssef Esmat
2023-10-06 0:36 ` Youssef Esmat
2023-10-10 8:08 ` Peter Zijlstra
2023-10-07 22:04 ` Peter Zijlstra
2023-10-09 14:41 ` Peter Zijlstra
2023-10-10 0:51 ` Youssef Esmat
2023-10-10 8:01 ` Peter Zijlstra
2023-10-16 16:50 ` Peter Zijlstra
2023-10-05 7:31 [PATCH] sched/fair: fix pick_eevdf to always find the correct se Biju Das
2023-10-05 15:02 ` Peter Zijlstra
2023-10-05 15:08 ` Biju Das
2023-10-06 8:36 ` Marek Szyprowski
2023-10-06 9:38 ` Biju Das
[not found] ` <553e2ee4-ab3a-4635-a74f-0ba4cc03f3f9@samsung.com>
2023-10-06 10:31 ` Mike Galbraith
2023-10-06 14:00 ` Peter Zijlstra
2023-10-06 14:39 ` Mike Galbraith
2023-10-06 15:55 ` Peter Zijlstra
2023-10-06 16:54 ` Mike Galbraith
2023-10-06 19:24 ` Peter Zijlstra
2023-10-06 20:15 ` Marek Szyprowski
2023-10-06 23:27 ` Mike Galbraith
2023-10-07 0:37 ` Chen Yu
2023-10-07 6:26 ` Biju Das
2023-10-07 6:42 ` Chen Yu
2023-10-09 14:27 ` Phil Auld
2023-10-10 2:08 ` Chen Yu
2023-10-07 6:24 ` Biju Das
2023-10-05 15:11 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=6b606049-3412-437f-af25-a4c33139e2d8@bytedance.com \
--to=wuyun.abel@bytedance.com \
--cc=bristot@redhat.com \
--cc=bsegall@google.com \
--cc=chris.hyser@oracle.com \
--cc=corbet@lwn.net \
--cc=dietmar.eggemann@arm.com \
--cc=efault@gmx.de \
--cc=joel@joelfernandes.org \
--cc=joshdon@google.com \
--cc=juri.lelli@redhat.com \
--cc=kprateek.nayak@amd.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mgorman@suse.de \
--cc=mingo@kernel.org \
--cc=patrick.bellasi@matbug.net \
--cc=pavel@ucw.cz \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=qperret@google.com \
--cc=qyousef@layalina.io \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=tim.c.chen@linux.intel.com \
--cc=timj@gnu.org \
--cc=vincent.guittot@linaro.org \
--cc=youssefesmat@chromium.org \
--cc=yu.c.chen@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).