From: "Nikunj A. Dadhania" <nikunj@linux.vnet.ibm.com>
To: Nikhil Rao <ncrao@google.com>, Ingo Molnar <mingo@elte.hu>,
Peter Zijlstra <peterz@infradead.org>
Cc: Paul Turner <pjt@google.com>, Mike Galbraith <efault@gmx.de>,
linux-kernel@vger.kernel.org, Nikhil Rao <ncrao@google.com>,
Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com>,
Bharata B Rao <bharata@linux.vnet.ibm.com>
Subject: Re: [RFC][PATCH 00/18] Increase resolution of load weights
Date: Thu, 28 Apr 2011 12:37:27 +0530 [thread overview]
Message-ID: <m3tydi6ehc.fsf@IBM-009124035060.in.ibm.com> (raw)
In-Reply-To: <1303332697-16426-1-git-send-email-ncrao@google.com>
[-- Attachment #1: Type: text/plain, Size: 2270 bytes --]
On Wed, 20 Apr 2011 13:51:19 -0700, Nikhil Rao <ncrao@google.com> wrote:
> Hi All,
>
> I have attached an early version of a RFC patchset to increase resolution of
> sched entity load weights. This RFC introduces SCHED_LOAD_RESOLUTION which
> scales NICE_0_LOAD by a factor of 1024. The scaling is done internally and should
> be completely invisible to the user.
>
> Why do we need this?
> This extra resolution allows us to scale on two dimensions - number of cpus and
> the depth of hierarchies. It also allows for proper load balancing of low weight
> task groups (for eg., nice+19 on autogroup).
>
> One of the big roadblocks for increasing resolution is the use of unsigned long
> for load.weight, which on 32-bit architectures can overflow with ~48 max-weight
> sched entities. In this RFC we convert all uses of load.weight to u64. This is
> still a work-in-progress and I have listed some of the issues I am still
> investigating below.
>
> I would like to get some feedback on the direction of this patchset. Please let
> me know if there are alternative ways of doing this, and I'll be happy to
> explore them as well.
>
> The patchset applies cleanly to v2.6.39-rc4. It compiles for i386 and boots on
> x86_64. Beyond the basic checks, it has not been well tested yet.
>
> Major TODOs:
> - Detect overflow in update shares calculations (time * load), and set load_avg
> to maximum possible value (~0ULL).
> - tg->task_weight uses an atomic which needs to be updates to 64-bit on 32-bit
> machines. Might need to add a lock to protect this instead of atomic ops.
> - Check wake-affine math and effective load calculations for overflows.
> - Needs more testing and need to ensure fairness/balancing is not broken.
>
Hi Nikhil,
I did a quick test for creating 600 cpu hog tasks with and without this
patches on a 16cpu machine(x86_64) and I am seeing some mis-behaviour:
Base kernel - 2.6.39-rc4
[root@krm1 ~]# time -p ./test
real 43.54
user 0.12
sys 1.05
[root@krm1 ~]#
Base + patches
[root@krm1 ~]# time -p ./test
Takes almost infinity, after 2 minutes I see only 16 tasks created
viewed from another ssh session to the machine:
[root@krm1 ~]# ps ax | grep dowhile$ | wc -l
16
[root@krm1 ~]#
I am attaching the script.
Regards,
Nikunj
[-- Attachment #2: create_loop_tasks.sh --]
[-- Type: application/x-sh, Size: 253 bytes --]
next prev parent reply other threads:[~2011-04-28 7:07 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-04-20 20:51 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 01/18] sched: introduce SCHED_POWER_SCALE to scale cpu_power calculations Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 02/18] sched: increase SCHED_LOAD_SCALE resolution Nikhil Rao
2011-04-28 9:54 ` Nikunj A. Dadhania
2011-04-28 17:11 ` Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 03/18] sched: use u64 for load_weight fields Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 04/18] sched: update cpu_load to be u64 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 05/18] sched: update this_cpu_load() to return u64 value Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 06/18] sched: update source_load(), target_load() and weighted_cpuload() to use u64 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 07/18] sched: update find_idlest_cpu() to use u64 for load Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 08/18] sched: update find_idlest_group() to use u64 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 09/18] sched: update division in cpu_avg_load_per_task to use div_u64 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 10/18] sched: update wake_affine path to use u64, s64 for weights Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 11/18] sched: update update_sg_lb_stats() to use u64 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 12/18] sched: Update update_sd_lb_stats() " Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 13/18] sched: update f_b_g() to use u64 for weights Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 14/18] sched: change type of imbalance to be u64 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 15/18] sched: update h_load to use u64 Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 16/18] sched: update move_task() and helper functions to use u64 for weights Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 17/18] sched: update f_b_q() to use u64 for weighted cpuload Nikhil Rao
2011-04-20 20:51 ` [RFC][Patch 18/18] sched: update shares distribution to use u64 Nikhil Rao
2011-04-21 6:16 ` [RFC][PATCH 00/18] Increase resolution of load weights Ingo Molnar
2011-04-21 16:32 ` Peter Zijlstra
2011-04-26 16:11 ` Nikhil Rao
2011-04-21 16:40 ` Peter Zijlstra
2011-04-28 7:07 ` Nikunj A. Dadhania [this message]
2011-04-28 11:48 ` Nikunj A. Dadhania
2011-04-28 12:12 ` Srivatsa Vaddagiri
2011-04-28 18:33 ` Nikhil Rao
2011-04-28 18:51 ` Paul Turner
2011-04-28 18:53 ` Paul Turner
2011-04-28 21:27 ` Nikhil Rao
2011-04-29 16:55 ` Paul Turner
2011-04-28 18:20 ` Nikhil Rao
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=m3tydi6ehc.fsf@IBM-009124035060.in.ibm.com \
--to=nikunj@linux.vnet.ibm.com \
--cc=bharata@linux.vnet.ibm.com \
--cc=efault@gmx.de \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=ncrao@google.com \
--cc=peterz@infradead.org \
--cc=pjt@google.com \
--cc=vatsa@linux.vnet.ibm.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
all inboxes | Powered by JetHome®