All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Bills, Jason M" <jason.m.bills@linux.intel.com>
To: "openbmc@lists.ozlabs.org" <openbmc@lists.ozlabs.org>
Subject: Sensor Value PropertiesChanged Events
Date: Mon, 1 Feb 2021 16:42:33 -0800	[thread overview]
Message-ID: <31abd546-4538-ecf0-134e-b8e48e75b3ad@linux.intel.com> (raw)

Hi All,

There is an issue and idea that James Feist and I chatted about to maybe 
relieve some of our D-Bus traffic.

A major contributor to our D-Bus traffic (as seen in dbus-monitor) is 
the polling sensors updating the xyz.openbmc_project.Sensor.Value.Value 
property on each polling loop, which generates a PropertiesChanged 
signal for every sensor on every polling loop (once per second?).

The concern is that more important D-Bus messages could be getting 
delayed as D-Bus processes these Sensor Value signals.

The idea to fix this is to change the sensors with a custom getter on 
the Value property, so the last read can be pulled from D-Bus using a 
get-property call, but it would no longer signal a PropertiesChanged event.

I pushed a proposed change here: 
https://gerrit.openbmc-project.xyz/c/openbmc/dbus-sensors/+/40199.

Our original assumption was that nobody was matching on this 
PropertiesChanged signal for the Value property; however, it was pointed 
out to me today, that PID control has a match for it and may be using it.

So, I wanted to start a broader community discussion about this issue:

1. Is this a real concern or are PropertiesChanged signals so 
lightweight that removing them won't help with D-Bus load?

2. Does anyone need to match on sensor Value property updates or is 
reading them with get-property enough?

3. Does PID control use the Value match?  If so and there are benefits 
to removing these signals, could PID control manage without them?


As a side note, I still have two remaining services that publish 
PropertiesChanged events on sensor Value properties:

PWM Sensors.  I have a proposed (and untested) change here: 
gerrit.openbmc-project.xyz/c/openbmc/dbus-sensors/+/40200.

A Power sensor, that I will track down based on this discussion.

Thanks!
-Jason

             reply	other threads:[~2021-02-02  0:44 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-02-02  0:42 Bills, Jason M [this message]
2021-02-02  1:26 ` Sensor Value PropertiesChanged Events Ed Tanous
2021-02-02 10:02   ` Ambrozewicz, Adrian
2021-02-04 15:55     ` Patrick Williams
2021-02-02 20:39   ` Bills, Jason M
2021-02-03  0:11   ` Andrew Jeffery
2021-02-03  1:28     ` Andrew Jeffery

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=31abd546-4538-ecf0-134e-b8e48e75b3ad@linux.intel.com \
    --to=jason.m.bills@linux.intel.com \
    --cc=openbmc@lists.ozlabs.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.