[U-Boot-Users] Temperature sensor on Sequoia board

[U-Boot-Users] Temperature sensor on Sequoia board

[Larry Johnson]

Hi Stefan and group,

When I was checking the LM75/AD7414 driver using my Sequoia board, I
found that the AD7414 on the Sequoia does not appear to update its
temperature as its documentation describes.  I had to reset U-Boot to
get the temperature reading to change.  Also, every since I have had
that board, the first temperature reading reported by U-Boot after
applying power has always been incorrect.  I didn't find anything in
the AD7414 documentation that would explain this.

Has anyone seen this behavior on other Sequoia boards?

Best regards,
Larry

[U-Boot-Users] Temperature sensor on Sequoia board

[Mike Nuss]

Larry Johnson wrote:
> 
> Hi Stefan and group,
> 
> When I was checking the LM75/AD7414 driver using my Sequoia board, I
> found that the AD7414 on the Sequoia does not appear to update its
> temperature as its documentation describes.  I had to reset U-Boot to
> get the temperature reading to change.  Also, every since I have had
> that board, the first temperature reading reported by U-Boot after
> applying power has always been incorrect.  I didn't find anything in
> the AD7414 documentation that would explain this.
> 
> Has anyone seen this behavior on other Sequoia boards?

Yes. In my experience, the reason the first reading is often incorrect is simply because the monitoring can take a little time to start. I've seen the same behavior on the adm1026 and adt7462 also.

I definitely see temperature updates using the Linux driver for ad7414 on Sequoia, but in U-Boot I wouldn't expect it to change much anyway.

Mike

[U-Boot-Users] Temperature sensor on Sequoia board

[Stefan Roese]

On Thursday 21 February 2008, Mike Nuss wrote:
> > When I was checking the LM75/AD7414 driver using my Sequoia board, I
> > found that the AD7414 on the Sequoia does not appear to update its
> > temperature as its documentation describes.  I had to reset U-Boot to
> > get the temperature reading to change.  Also, every since I have had
> > that board, the first temperature reading reported by U-Boot after
> > applying power has always been incorrect.  I didn't find anything in
> > the AD7414 documentation that would explain this.
> >
> > Has anyone seen this behavior on other Sequoia boards?

Yes. I've seen this before, but forgot to take care of it.

> Yes. In my experience, the reason the first reading is often incorrect is
> simply because the monitoring can take a little time to start. I've seen
> the same behavior on the adm1026 and adt7462 also.

No. IIRC, then reading the temperatures starts working correctly after writing 
to the I2C device. So a simple "iprobe" will get it working. This is no 
solution of course.

> I definitely see temperature updates using the Linux driver for ad7414 on
> Sequoia, but in U-Boot I wouldn't expect it to change much anyway.

Yes, it works in Linux all the time, since Linux does an probe of the I2C 
devices itself. But we should really fix this in U-Boot too. Greatly 
appreciated, if somebody would take care of this issue. :)

Thanks.

Best regards,
Stefan

=====================================================================
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-0 Fax: +49-8142-66989-80  Email: office at denx.de
=====================================================================