[U-Boot] imx6qp wandboard getting stuck in U-Boot

[Fabio Estevam]

On Wed, Sep 27, 2017 at 1:27 PM, Fabio Estevam <festevam at gmail.com> wrote:
> Hi,
>
> With the two attached U-Boot patches plus the dts one I get the
> following on imx6qp wandboard:
>
> U-Boot SPL 2017.09-37395-g28ca7c0-dirty (Sep 27 2017 - 11:04:41)
> Trying to boot from MMC1
>
>
> U-Boot 2017.09-37395-g28ca7c0-dirty (Sep 27 2017 - 11:04:41 -0300)
>
> CPU:   Freescale i.MX6QP rev1.0 at 792 MHz
> Reset cause: POR
> I2C:   ready
> DRAM:  2 GiB
> PMIC:  PFUZE100 ID=0x10
> MMC:   FSL_SDHC: 0, FSL_SDHC: 1
> No panel detected: default to HDMI
> Display: HDMI (1024x768)
> In:    serial
> Out:   serial
> Err:   serial
> Board: Wandboard rev D1
> Net:   FEC [PRIME]
> Hit any key to stop autoboot:  0
> switch to partitions #0, OK
> mmc0 is current device
> Scanning mmc 0:1...
> Found U-Boot script /boot/boot.scr
> 336 bytes read in 69 ms (3.9 KiB/s)
> ## Executing script at 12000000
> 38550 bytes read in 83 ms (453.1 KiB/s)
> 6378872 bytes read in 367 ms (16.6 MiB/s)
> ## Flattened Device Tree blob at 18000000
>    Booting using the fdt blob at 0x18000000
>    Using Device Tree in place at 18000000, end 1800c695
> (Hangs here)

Looks like a clock issue. If I do this change:

--- a/board/wandboard/spl.c
+++ b/board/wandboard/spl.c
@@ -258,13 +258,13 @@ static void ccgr_init(void)
 {
        struct mxc_ccm_reg *ccm = (struct mxc_ccm_reg *)CCM_BASE_ADDR;

-       writel(0x00C03F3F, &ccm->CCGR0);
-       writel(0x0030FC03, &ccm->CCGR1);
-       writel(0x0FFFC000, &ccm->CCGR2);
-       writel(0x3FF00000, &ccm->CCGR3);
-       writel(0x00FFF300, &ccm->CCGR4);
-       writel(0x0F0000C3, &ccm->CCGR5);
-       writel(0x000003FF, &ccm->CCGR6);
+       writel(0xffffffff, &ccm->CCGR0);
+       writel(0xffffffff, &ccm->CCGR1);
+       writel(0xffffffff, &ccm->CCGR2);
+       writel(0xffffffff, &ccm->CCGR3);
+       writel(0xffffffff, &ccm->CCGR4);
+       writel(0xffffffff, &ccm->CCGR5);
+       writel(0xffffffff, &ccm->CCGR6);
 }

Then I am able to boot the kernel fine. Will try to find out what is
the missing clock for the imx6qp case.