[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170502145224.GC4626@worktop.programming.kicks-ass.net>
Date: Tue, 2 May 2017 16:52:24 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Rik van Riel <riel@...hat.com>
Cc: mingo@...nel.org, lvenanci@...hat.com, lwang@...hat.com,
efault@....de, tglx@...utronix.de, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 06/14] sched/topology,debug: Verify the first group
matches the child domain
On Mon, May 01, 2017 at 05:13:26PM -0400, Rik van Riel wrote:
> On Fri, 2017-04-28 at 15:20 +0200, Peter Zijlstra wrote:
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
>
> This could use a changelog.
Yes indeed... I put off writing one because $hard, and clearly I forgot
entirely :-/
How's this?
---
Subject: sched/topology,debug: Verify the first group matches the child domain
From: Peter Zijlstra <peterz@...radead.org>
Date: Fri Apr 14 18:20:48 CEST 2017
We want sched_groups to be sibling child domains (or individual CPUs
when there are no child domains). Furthermore, since the first group
of a domain should include the CPU of that domain, the first group of
each domain should match the child domain.
Verify this is indeed so.
Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
---
kernel/sched/topology.c | 6 ++++++
1 file changed, 6 insertions(+)
--- a/kernel/sched/topology.c
+++ b/kernel/sched/topology.c
@@ -93,6 +93,12 @@ static int sched_domain_debug_one(struct
group->sgc->capacity);
}
+ if (group == sd->groups && sd->child &&
+ !cpumask_equal(sched_domain_span(sd->child),
+ sched_group_cpus(group))) {
+ printk(KERN_ERR "ERROR: domain->groups does not match domain->child\n");
+ }
+
group = group->next;
if (group != sd->groups)
Powered by blists - more mailing lists