Re: [PATCH v9 6/7] arm64: dts: ti: k3-j721e-sk: Add TPS6594 family PMICs

From: Neha Malcom Francis
Date: Fri Dec 08 2023 - 00:40:26 EST


Hi Nishanth

On 07/12/23 19:19, Nishanth Menon wrote:
On 11:01-20231207, Neha Malcom Francis wrote:
Hi Nishanth, Udit,

On 07/12/23 10:12, Kumar, Udit wrote:

On 12/5/2023 8:46 PM, Nishanth Menon wrote:
On 15:04-20231205, Neha Malcom Francis wrote:
This patch adds support for TPS6594 PMIC family on wakeup I2C0 bus.
These devices provide regulators (bucks and LDOs), but also GPIOs, a
RTC, a watchdog, an ESM (Error Signal Monitor) which monitors the SoC
error output signal, and a PFSM (Pre-configurable Finite State Machine)
which manages the operational modes of the PMIC.

Signed-off-by: Neha Malcom Francis <n-francis@xxxxxx>
---
  arch/arm64/boot/dts/ti/k3-j721e-sk.dts | 158 +++++++++++++++++++++++++
  1 file changed, 158 insertions(+)

diff --git a/arch/arm64/boot/dts/ti/k3-j721e-sk.dts
b/arch/arm64/boot/dts/ti/k3-j721e-sk.dts
index 42fe8eee9ec8..e600825f7e78 100644
--- a/arch/arm64/boot/dts/ti/k3-j721e-sk.dts
+++ b/arch/arm64/boot/dts/ti/k3-j721e-sk.dts
@@ -459,6 +459,13 @@ J721E_IOPAD(0x234, PIN_INPUT, 7) /* (U3)
EXT_REFCLK1.GPIO1_12 */
  };
  &wkup_pmx0 {
+    pmic_irq_pins_default: pmic-irq-default-pins {
+        bootph-pre-ram;
+        pinctrl-single,pins = <
+            J721E_WKUP_IOPAD(0x0cc, PIN_INPUT, 7) /* (G28) WKUP_GPIO0_7 */
+        >;
+    };
+
      mcu_cpsw_pins_default: mcu-cpsw-default-pins {
          pinctrl-single,pins = <
              J721E_WKUP_IOPAD(0x84, PIN_INPUT, 0) /* (B24) MCU_RGMII1_RD0 */
@@ -560,6 +567,157 @@ eeprom@51 {
          compatible = "atmel,24c512";
          reg = <0x51>;
      };
+
+    tps659413: pmic@48 {
+        bootph-pre-ram;
only for the leaf nodes. See
https://libera.irclog.whitequark.org/armlinux/2023-10-19


AFAIK, please correct me, u-boot still needs in all nodes ?


That's what I believe as well, is it better to have only the leaf nodes in
kernel and have U-Boot DTSI handle the parent bootph properties? If so I'll
send out v10 making change accordingly.


Yes, u-boot today needs it in all nodes. BUT, u-boot needs to be fixed in
line to obey the rules of the schema convention that Rob clarified in
the discussion above.

The other choice is NOT to introduce new bootph properties till u-boot
is fixed up (this is also why I haven't sent out further updates for
bootph properties for kernel in this cycle).


I think we can have it following kernel convention here and fix up in the U-Boot DTSI for this series since PMIC nodes are needed. Thanks!

--
Thanking You
Neha Malcom Francis