Hi
Il sab 18 giu 2022, 23:56 Heiko Thiery heiko.thiery@gmail.com ha scritto:
Hi Michael, Hi Marek,
Am Mi., 15. Juni 2022 um 08:43 Uhr schrieb Michael Nazzareno Trimarchi michael@amarulasolutions.com:
Hi Heiko
On Wed, Jun 15, 2022 at 8:23 AM Heiko Thiery heiko.thiery@gmail.com
wrote:
Hi Marek,
[SNIP]
diff --git a/board/freescale/imx8mn_evk/spl.c
b/board/freescale/imx8mn_evk/spl.c
index 14cb51368f..0d9909a662 100644 --- a/board/freescale/imx8mn_evk/spl.c +++ b/board/freescale/imx8mn_evk/spl.c @@ -83,6 +83,15 @@ int power_init_board(void) #ifdef CONFIG_IMX8MN_LOW_DRIVE_MODE /* Set VDD_SOC/VDD_DRAM to 0.8v for low drive mode */ pmic_reg_write(dev, PCA9450_BUCK1OUT_DVS0, 0x10); +#elif defined(CONFIG_TARGET_IMX8MN_DDR3L_EVK)
/* Set VDD_SOC to 0.85v for DDR3L at 1600MTS */pmic_reg_write(dev, PCA9450_BUCK1OUT_DVS0, 0x14);/* Disable the BUCK2 */pmic_reg_write(dev, PCA9450_BUCK2CTRL, 0x48);/* Set NVCC_DRAM to 1.35v */pmic_reg_write(dev, PCA9450_BUCK6OUT, 0x1E);#else
All this part is not done by the spl pmic driver?
I saw that you added the PCA9450 driver. Do you know if this initialization can be done by the driver when CONFIG_SPL_DM_REGULATOR is enabled? If I see this correctly, it can't be done. Is that correct?
+&i2c1 {
u-boot,dm-spl;+};
+&{/soc@0/bus@30800000/i2c@30a20000/pmic@4b} {
u-boot,dm-spl;+};
+&{/soc@0/bus@30800000/i2c@30a20000/pmic@4b/regulators} {
u-boot,dm-spl;+};
+&pinctrl_i2c1 {
u-boot,dm-spl;+};
+&pinctrl_pmic {
u-boot,dm-spl;+};
Maybe something like this should work. Now question is about should be done in pre-reloc or not
It took me a little while to understand what was meant by this. In the meantime I could change the PMIC/Regulator initialization to DT/DM and can make the setting so. What is not clear to me yet is at which point this should be done (regulators_enable_boot_on()). Currently I do it in the board/spl specific board_init_f().
Do any of you have any advice here?
After having a discussion with you, i found that maybe i have some problem and some of the setting should be done before ddr inizialization. I will take a look on it on afternoon. I know marek was working on some way to probe driver during binding and we need even to be sure that this can happen before memory training.
Michael
Thanks
Heiko
Michael
-- Heiko
-- Michael Nazzareno Trimarchi Co-Founder & Chief Executive Officer M. +39 347 913 2170 michael@amarulasolutions.com __________________________________
Amarula Solutions BV Joop Geesinkweg 125, 1114 AB, Amsterdam, NL T. +31 (0)85 111 9172 info@amarulasolutions.com www.amarulasolutions.com