Hi Robin,

Am Freitag, den 27.07.2018, 01:51 +0000 schrieb Robin Gong:
[...]

quoted

quoted

quoted

---
?Documentation/devicetree/bindings/clock/imx6q-clock.txt | 8 ++++++++
?1 file changed, 8 insertions(+)

diff --git a/Documentation/devicetree/bindings/clock/imx6q-clock.txt

b/Documentation/devicetree/bindings/clock/imx6q-clock.txt
index a45ca67a9d5f..e1308346e00d 100644

--- a/Documentation/devicetree/bindings/clock/imx6q-clock.txt
+++ b/Documentation/devicetree/bindings/clock/imx6q-clock.txt

@@ -6,6 +6,14 @@ Required properties:

?- interrupts: Should contain CCM interrupt
?- #clock-cells: Should be <1>

+Optional properties:
+- fsl,pmic-stby-poweroff: Configure CCM to assert PMIC_STBY_REQ
+signal
+??on power off.
+??Use this property if the SoC should be powered off by external
+power
+??management IC (PMIC) triggered via PMIC_STBY_REQ signal.

PMIC_ON_REQ didn't connect to any pin of PMIC in your case?

No. First, it was only one customer specific issue. After some research I found
even publicly available boards (for example RioTboard) which has same/similar
design. After seeing this in imx6 documentation as valid power off way, I have
no doubts - there should be even more devices doin this in the wild.

Not sure why the customer didn't follow reference design, since PMIC_ON_REQ can
provide power off/on feature by pressing ONOFF key which connected ONOFF
pin of i.mx6(ONOFF will toggle PMIC_ON_REQ to power off/on PMIC). The official
power off/on way (PMIC_ON_REQ) can also power off all power rails of PFUZE except
snvs as your patch did on PMIC_STBY_REQ.??PMIC_STBY_REQ is used to notify pmic
switch power mode (PFM/APS) or decrease voltage to save power in kernel suspend,
not power off....I am not sure if we need this patchset to 'workaround' the board issue.

Not all boards follow the reference design, that's a fact of life.

Please look at the i.MX6Q reference manual. The sequence implemented in
this patchset can be found as a valid way to power off the system in
"60.4.3 Power mode transitions" "Normal ON to OFF with external PMIC",
so there is hardly any way to argue that this is a board specific
quirk. This is one of the Freescale/NXP recommended sequences to turn
off the system.

Regards,
Lucas