linux-kernel - [tip:sched/core] sched: Fix SD_POWERSAVING_BALANCE|SD_PREFER_LOCAL vs SD_WAKE

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Thu, 17 Sep 2009 08:54:49 GMT
From:	tip-bot for Peter Zijlstra <a.p.zijlstra@...llo.nl>
To:	linux-tip-commits@...r.kernel.org
Cc:	linux-kernel@...r.kernel.org, hpa@...or.com, mingo@...hat.com,
	a.p.zijlstra@...llo.nl, tglx@...utronix.de, mingo@...e.hu
Subject: [tip:sched/core] sched: Fix SD_POWERSAVING_BALANCE|SD_PREFER_LOCAL vs SD_WAKE_AFFINE

Commit-ID:  29cd8bae396583a2ee9a3340db8c5102acf9f6fd
Gitweb:     http://git.kernel.org/tip/29cd8bae396583a2ee9a3340db8c5102acf9f6fd
Author:     Peter Zijlstra <a.p.zijlstra@...llo.nl>
AuthorDate: Thu, 17 Sep 2009 09:01:14 +0200
Committer:  Ingo Molnar <mingo@...e.hu>
CommitDate: Thu, 17 Sep 2009 10:40:31 +0200

sched: Fix SD_POWERSAVING_BALANCE|SD_PREFER_LOCAL vs SD_WAKE_AFFINE

The SD_POWERSAVING_BALANCE|SD_PREFER_LOCAL code can break out of
the domain iteration early, making us miss the SD_WAKE_AFFINE bits.

Fix this by continuing iteration until there is no need for a
larger domain.

This also cleans up the cgroup stuff a bit, but not having two
update_shares() invocations.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@...e.hu>


---
 kernel/sched_fair.c |   42 +++++++++++++++++++++++++++---------------
 1 files changed, 27 insertions(+), 15 deletions(-)

diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index ffee827..10d218a 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -1333,11 +1333,12 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
  */
 static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flags)
 {
-	struct sched_domain *tmp, *shares = NULL, *sd = NULL;
+	struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL;
 	int cpu = smp_processor_id();
 	int prev_cpu = task_cpu(p);
 	int new_cpu = cpu;
 	int want_affine = 0;
+	int want_sd = 1;
 	int sync = wake_flags & WF_SYNC;
 
 	if (sd_flag & SD_BALANCE_WAKE) {
@@ -1369,33 +1370,44 @@ static int select_task_rq_fair(struct task_struct *p, int sd_flag, int wake_flag
 				nr_running /= 2;
 
 			if (nr_running < capacity)
-				break;
+				want_sd = 0;
 		}
 
 		if (want_affine && (tmp->flags & SD_WAKE_AFFINE) &&
 		    cpumask_test_cpu(prev_cpu, sched_domain_span(tmp))) {
 
-			if (sched_feat(LB_SHARES_UPDATE)) {
-				update_shares(tmp);
-				shares = tmp;
-			}
-
-			if (wake_affine(tmp, p, sync)) {
-				new_cpu = cpu;
-				goto out;
-			}
-
+			affine_sd = tmp;
 			want_affine = 0;
 		}
 
+		if (!want_sd && !want_affine)
+			break;
+
 		if (!(tmp->flags & sd_flag))
 			continue;
 
-		sd = tmp;
+		if (want_sd)
+			sd = tmp;
+	}
+
+	if (sched_feat(LB_SHARES_UPDATE)) {
+		/*
+		 * Pick the largest domain to update shares over
+		 */
+		tmp = sd;
+		if (affine_sd && (!tmp ||
+				  cpumask_weight(sched_domain_span(affine_sd)) >
+				  cpumask_weight(sched_domain_span(sd))))
+			tmp = affine_sd;
+
+		if (tmp)
+			update_shares(tmp);
 	}
 
-	if (sd && sd != shares && sched_feat(LB_SHARES_UPDATE))
-		update_shares(sd);
+	if (affine_sd && wake_affine(affine_sd, p, sync)) {
+		new_cpu = cpu;
+		goto out;
+	}
 
 	while (sd) {
 		int load_idx = sd->forkexec_idx;
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/