phosphor-hwmon bottleneck potential

[Patrick Williams]

On Thu, May 04, 2017 at 09:11:49AM -0700, Patrick Venture wrote:
> Without multi-threading to remove the timing bottleneck on my platform, and
> the inability to provide other more direct interfaces for controlling fans,
> I'll be unable to use phosphor-hwmon as a primary source for
> reading/writing sensor data and will be unable to upstream any effort i
> make to write a pluggable thermal controller.

From what you wrote earlier, it seems like the constraint has nothing to
do with phosphor-hwmon but with the hwmon subsystem for your particular
driver?  How does a "direct interface" avoid the hwmon latency?  I'm 99%
certain that it isn't dbus itself contributing to the 8 second latency
you are seeing here.  Are you arriving at a different conclusion?

-- 
Patrick Williams
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: Digital signature
URL: <http://lists.ozlabs.org/pipermail/openbmc/attachments/20170505/823a9e01/attachment.sig>