Sensor Value PropertiesChanged Events

[Patrick Williams]

On Tue, Feb 02, 2021 at 11:02:53AM +0100, Ambrozewicz, Adrian wrote:
> W dniu 2/2/2021 o 01:42, Bills, Jason M pisze:
>
> My trust in systemd D-Bus implementation is that signals are implemented 
> in optimal way and broker doesn't broadcast it to services without 
> proper 'match' defined. It should be checked though
> Moving to polling might in fact increase D-Bus utilization by 
> introducing message-response communication between producer and 
> consumer. In certain cases of sensors which tend to update slowly, 
> introducing a getter with faster interval would increase the traffic, if 
> the interval would be faster than the sensor update rate.

You raise a very good point here.  When I review code that contains a
signal match almost every time I see code with very little in the match
and doing a bunch of filtering in C++ code and I have to point this out.
It is like doing a `SELECT *` in SQL.

I wouldn't be surprised if there are lots of cases where we're effectively
broadcasting signals and then dropping them on the receive side rather than
allowing dbus-broker to filter out who gets them.  We might be able to
code something up in sdbusplus to warn when the filtering is likely
insufficient but that'd probably become a runtime check that gets lost
in a journal message.

-- 
Patrick Williams
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://lists.ozlabs.org/pipermail/openbmc/attachments/20210204/fd079ace/attachment.sig>