[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20170428103341.63fcmyz2dt53zsrc@hirez.programming.kicks-ass.net>
Date: Fri, 28 Apr 2017 12:33:41 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Lauro Venancio <lvenanci@...hat.com>
Cc: lwang@...hat.com, riel@...hat.com, Mike Galbraith <efault@....de>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...nel.org>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 0/4] sched/topology: fix overlap group capacity and
balance cpu
On Wed, Apr 26, 2017 at 02:59:09PM -0300, Lauro Venancio wrote:
> Maybe, to make the debug messages clear, you could print "balance_cpu:
> %d" instead of "capacity(%d): " and assert (sgc id == balance cpu). You
> would need to restart the sgc id count on every topology level.
Right, that's a good suggestion. I've also completely revamped printing
the groups, because they got really long and it was very hard to
visually separate the individual groups on a line.
I've ended up with the following:
[] CPU0 attaching sched-domain(s):
[] domain-0: span=0,4 level=DIE
[] groups: 0:{ span=0 }, 4:{ span=4 }
[] domain-1: span=0-1,3-5,7 level=NUMA
[] groups: 0:{ span=0,4 mask=0,4 cap=2048 }, 1:{ span=1,5 mask=1,5 cap=2048 }, 3:{ span=3,7 mask=3,7 cap=2048 }
[] domain-2: span=0-7 level=NUMA
[] groups: 0:{ span=0-1,3-5,7 mask=0,4 cap=6144 }, 2:{ span=1-3,5-7 mask=2,6 cap=6144 }
[] CPU1 attaching sched-domain(s):
[] domain-0: span=1,5 level=DIE
[] groups: 1:{ span=1 }, 5:{ span=5 }
[] domain-1: span=0-2,4-6 level=NUMA
[] groups: 1:{ span=1,5 mask=1,5 cap=2048 }, 2:{ span=2,6 mask=2,6 cap=2048 }, 4:{ span=0,4 mask=0,4 cap=2048 }
[] domain-2: span=0-7 level=NUMA
[] groups: 1:{ span=0-2,4-6 mask=1,5 cap=6144 }, 3:{ span=0,2-4,6-7 mask=3,7 cap=6144 }
Powered by blists - more mailing lists