Pocket Beagle SPI1 ChipSelect on silkscreen is not CS0 but CS1

52 views
Skip to first unread message

Henner Zeller

unread,
Oct 4, 2017, 7:33:47 AM10/4/17
to BeagleBoard
Hi,
Little heads-up: I just spent an hour or so to realize that the simple 'CS' (ChipSelect) printed on the silkscreen for SPI1 bus on the PocketBeagle is not CS0 as I expected, but CS1. So this is actually bus /dev/spidev2.1

Not sure if this was intended, but if it is, it should probably be marked on the silk-screen as CS1 to avoid future confusion; I can send a pull request for the board (in general, even CS0 should be marked as such)

If it is _not_ intended, the am335x-pocketbeagle.dts would need to be changed before capes start assuming this is CS1 (around line 389 Jason).

Cheers,
 Henner.

Robert Nelson

unread,
Oct 4, 2017, 9:53:13 AM10/4/17
to Beagle Board, Henner Zeller
HI Henner,

I've noticed this to while doing the overlays:

SPI1 CS1 is used as CS

https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/PB-SPI1-ETH-CLICK.dts#L83

SPI0 CS0 is used as CS

https://github.com/beagleboard/bb.org-overlays/blob/master/src/arm/PB-SPI0-ETH-CLICK.dts#L83

I haven't looked to closely, (just let my automated script setup the
pinmux's).. But if SPI1 (CS0) and SPI0 (CS1) is not on the header, we
should remove them from the base tree:

Regards,

--
Robert Nelson
https://rcn-ee.com/

Henner Zeller

unread,
Oct 31, 2017, 12:02:56 AM10/31/17
to Robert Nelson, Beagle Board
Yes, this would certainly less confusing to leave out the
non-available chip-selects in the /dev/spidev* devices.

Is there a reason why SPI0 shows up as /dev/spidev1.* and SPI1 as
/dev/spidev2.* ? This does look like an off-by-one. Is this
intentional ?

Thanks,
Henner.

Graham

unread,
Oct 31, 2017, 2:41:38 PM10/31/17
to BeagleBoard
Robert:

Are the default device tree pinmux connections for the PocketBeagle the functions 
named in the silkscreen labels on the bottom of the PocketBeagle?

So far this appears to be the case.  Any exceptions?

--- Graham

==

Robert Nelson

unread,
Oct 31, 2017, 3:00:24 PM10/31/17
to Beagle Board
On Tue, Oct 31, 2017 at 1:41 PM, Graham <gra...@flex-radio.com> wrote:
> Robert:
>
> Are the default device tree pinmux connections for the PocketBeagle the
> functions
> named in the silkscreen labels on the bottom of the PocketBeagle?

Yes,

>
> So far this appears to be the case. Any exceptions?

If not, let me know. ;) So we can fix it..

Jason Kridner

unread,
Oct 31, 2017, 4:13:11 PM10/31/17
to beagl...@googlegroups.com


> On Oct 31, 2017, at 2:59 PM, Robert Nelson <robert...@gmail.com> wrote:
>
>> On Tue, Oct 31, 2017 at 1:41 PM, Graham <gra...@flex-radio.com> wrote:
>> Robert:
>>
>> Are the default device tree pinmux connections for the PocketBeagle the
>> functions
>> named in the silkscreen labels on the bottom of the PocketBeagle?
>
> Yes,
>
>>
>> So far this appears to be the case. Any exceptions?
>
> If not, let me know. ;) So we can fix it..

48 should be labeled 44.

SPI1 is CS1, not CS0 (silk screen doesn’t say).


>
> Regards,
>
> --
> Robert Nelson
> https://rcn-ee.com/
>
> --
> For more options, visit http://beagleboard.org/discuss
> ---
> You received this message because you are subscribed to the Google Groups "BeagleBoard" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/CAOCHtYioaMyouYSSosFpBG%2BaEB7f6dGOhnjPTPa2ymA-5w8v8w%40mail.gmail.com.
> For more options, visit https://groups.google.com/d/optout.

Henner Zeller

unread,
Oct 31, 2017, 4:17:56 PM10/31/17
to Beagle Board
On 31 October 2017 at 13:12, Jason Kridner <jkri...@gmail.com> wrote:
>
>
>> On Oct 31, 2017, at 2:59 PM, Robert Nelson <robert...@gmail.com> wrote:
>>
>>> On Tue, Oct 31, 2017 at 1:41 PM, Graham <gra...@flex-radio.com> wrote:
>>> Robert:
>>>
>>> Are the default device tree pinmux connections for the PocketBeagle the
>>> functions
>>> named in the silkscreen labels on the bottom of the PocketBeagle?
>>
>> Yes,
>>
>>>
>>> So far this appears to be the case. Any exceptions?
>>
>> If not, let me know. ;) So we can fix it..
>
> 48 should be labeled 44.
>
> SPI1 is CS1, not CS0 (silk screen doesn’t say).

Actually, my pull request to the board fixed the chip-select printing:

https://github.com/beagleboard/pocketbeagle/pull/9
https://github.com/beagleboard/pocketbeagle/commit/1d96a30c13ed9ab13bd1372aa87de865339e5499


-h
Reply all
Reply to author
Forward
0 new messages