[PATCH v2] ARM: dts: ux500: Fix BT+WLAN on Samsung phones

[Stephan Gerhold]

On Mon, Mar 15, 2021 at 12:07:50PM +0100, Linus Walleij wrote:
> Some of the Samsung phones had set GPIO 215 as voltage
> regulator GPIO, while in fact this is always on GPIO 222.
> 
> Using GPIO 215 meant that the Bluetooth would not come
> on properly since both WLAN and Bluetooth use GPIO 222
> to power up. The Bluetooth part refers to this GPIO
> line as "shutdown-gpios" which is kind of true, but
> fails to mention that it will also shut down the WLAN
> when it is disabled, which was causing bugs.
> 

Hmm, I'm a bit confused here. The datasheet for BCM4334 [1] that I was
looking at back when I added WiFi/BT to samsung-golden says:

"Section 22: Power-Up Sequence and Timing
 Note: The WL_REG_ON and BT_REG_ON signals are ORed in the BCM4334. [...]
 If two independent host GPIOs are used (one for WL_REG_ON and one for
 BT_REG_ON), then only one of the two signals needs to be high to enable
 the BCM4334 regulators."

and the "Description of Control Signals" part after that repeats the
same, and even shows diagrams that only one of them should be high to
enable only WiFi or only BT.

According to the schematics for samsung-golden:
  - GPIO215 goes to WL_REG_ON of BCM4334
  - GPIO222 goes to BT_REG_ON of BCM4334

So I would say the device tree is correct the way it is, unless the
datasheet is wrong... I wonder if there is some other problem here,
I think WiFi/BT used to probe just fine on samsung-golden...

Stephan

[1]: https://www.cypress.com/file/298706/download