Re: [PATCH 0/2 v4] sched: Rewrite per entity runnable load average tracking

From: Yuyang Du <yuyang.du@intel.com>
To: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: mingo@redhat.com, peterz@infradead.org,
	linux-kernel@vger.kernel.org, pjt@google.com, bsegall@google.com,
	arjan.van.de.ven@intel.com, len.brown@intel.com,
	rafael.j.wysocki@intel.com, alan.cox@intel.com,
	mark.gross@intel.com, fengguang.wu@intel.com
Subject: Re: [PATCH 0/2 v4] sched: Rewrite per entity runnable load average tracking
Date: Mon, 28 Jul 2014 03:34:53 +0800	[thread overview]
Message-ID: <20140727193453.GA23367@intel.com> (raw)
In-Reply-To: <1405835183.6512.37.camel@marge.simpson.net>

Thanks a lot, Mike.

Ben asked for this test, but actually I don't know how to get pipe-test, still
not even after google it.

On Sun, Jul 20, 2014 at 07:46:23AM +0200, Mike Galbraith wrote:
> On Fri, 2014-07-18 at 07:26 +0800, Yuyang Du wrote: 
> > Thanks to Morten, Ben, and Fengguang.
> > 
> > v4 changes:
> > 
> > - Insert memory barrier before writing cfs_rq->load_last_update_copy.
> > - Fix typos.
> 
> My little desktop box says lovely minus signs have had their usual
> effect on the general case (cgroups enabled but not in use). 
> 
> pipe-test scheduling cross core - full fastpath
> 3.0.101-default        3.753363 usecs/loop -- avg 3.770737 530.4 KHz   1.000
> 3.1.10-default         3.723843 usecs/loop -- avg 3.716058 538.2 KHz   1.014
> 3.2.51-default         3.728060 usecs/loop -- avg 3.710372 539.0 KHz   1.016
> 3.3.8-default          3.906174 usecs/loop -- avg 3.900399 512.8 KHz    .966
> 3.4.97-default         3.864158 usecs/loop -- avg 3.865281 517.4 KHz    .975
> 3.5.7-default          3.967481 usecs/loop -- avg 3.962757 504.7 KHz    .951
> 3.6.11-default         3.851186 usecs/loop -- avg 3.845321 520.1 KHz    .980
> 3.7.10-default         3.777869 usecs/loop -- avg 3.776913 529.5 KHz    .998
> 3.8.13-default         4.049927 usecs/loop -- avg 4.041905 494.8 KHz    .932
> 3.9.11-default         3.973046 usecs/loop -- avg 3.974208 503.2 KHz    .948
> 3.10.27-default        4.189598 usecs/loop -- avg 4.189298 477.4 KHz    .900
> 3.11.10-default        4.293870 usecs/loop -- avg 4.297979 465.3 KHz    .877
> 3.12.24-default        4.321570 usecs/loop -- avg 4.321961 462.8 KHz    .872
> 3.13.11-default        4.137845 usecs/loop -- avg 4.134863 483.7 KHz    .911
> 3.14.10-default        4.145348 usecs/loop -- avg 4.139987 483.1 KHz    .910            
> 3.15.4-default         4.355594 usecs/loop -- avg 4.351961 459.6 KHz    .866             
> 3.16.0-default         4.537279 usecs/loop -- avg 4.543532 440.2 KHz    .829     1.000   
> 3.16.0-default+v4      4.343542 usecs/loop -- avg 4.318803 463.1 KHz    .873     1.052
> 
> Extending max depth to 5, cost of depth++ seemingly did not change
> despite repeatable dip at depth = 3 (gremlins at play).
> 
> mount -t cgroup o cpu none /cgroups
> mkdir -p /cgroups/a/b/c/d/e
> 
> cgexec -g cpu:a pipe-test 1
> 3.16.0-default         5.016373 usecs/loop -- avg 5.021115 398.3 KHz   1.000
> 3.16.0-default+v4      4.978625 usecs/loop -- avg 4.977381 401.8 KHz   1.008
> 
> cgexec -g cpu:a/b pipe-test 1
> 3.16.0-default         5.543566 usecs/loop -- avg 5.565475 359.4 KHz   1.000
> 3.16.0-default+v4      5.597399 usecs/loop -- avg 5.570444 359.0 KHz    .998
> 
> cgexec -g cpu:a/b/c pipe-test 1
> 3.16.0-default         6.092256 usecs/loop -- avg 6.094186 328.2 KHz   1.000
> 3.16.0-default+v4      6.294858 usecs/loop -- avg 6.338453 315.5 KHz    .961
> 
> cgexec -g cpu:a/b/c/d pipe-test 1
> 3.16.0-default         6.719044 usecs/loop -- avg 6.717118 297.7 KHz   1.000
> 3.16.0-default+v4      6.788559 usecs/loop -- avg 6.710102 298.1 KHz   1.001
> 
> cgexec -g cpu:a/b/c/d/e pipe-test 1
> 3.16.0-default         7.186431 usecs/loop -- avg 7.194884 278.0 KHz   1.000
> 3.16.0-default+v4      7.368443 usecs/loop -- avg 7.250371 275.8 KHz    .992
> 

So the result is flat compared to before or a pass?

Thanks,
Yuyang