On Fri, Jan 8, 2021 at 1:22 AM Tony Lindgren [off-list ref] wrote:

* H. Nikolaus Schaller [off-list ref] [201230 13:29]:

quoted

quoted

Am 30.12.2020 um 13:55 schrieb Adam Ford [off-list ref]:
On Wed, Dec 30, 2020 at 2:43 AM Tony Lindgren [off-list ref] wrote:

quoted

At least for 4430, trying to use the single conversion mode eventually
hangs the thermal sensor. This can be quite easily seen with errors:

thermal thermal_zone0: failed to read out thermal zone (-5)

...

quoted

quoted

I don't have an OMAP4, but if you want, I can test a DM3730.

Indeed I remember a similar discussion from the DM3730 [1]. temp values were
always those from the last measurement. E.g. the first one was done
during (cold) boot and the first request after 10 minutes did show a
quite cold system... The next one did show a hot system independent
of what had been between (suspend or high activity).

It seems as if it was even reproducible with a very old kernel on a BeagleBoard.
So it is quite fundamental.

We tried to fix it but did not come to a solution [2]. So we opened an issue
in our tracker [3] and decided to stay with continuous conversion although this
raises idle mode processor load.

Hmm so maybe eocz high always times out in single mode since it also
triggers at least on dra7?

Yes it would be great if you guys can the $subject patch a try at
least on your omap36xx and omap5 boards and see if you see eocz
time out warnings in dmesg.

I should be able to try it on the dm3730 logicpd-torpedo kit this weekend.

adam

Regards,

Tony