From: Ahmad Fatoum <a.fatoum@pengutronix.de>
To: Marco Felsch <m.felsch@pengutronix.de>, barebox@lists.infradead.org
Subject: Re: [PATCH] ARM: i.MX6: boot: detect USB serial downloader more reliable
Date: Wed, 11 Aug 2021 20:30:53 +0200 [thread overview]
Message-ID: <241f3046-fdaf-ccd9-bc43-8c6879da75d8@pengutronix.de> (raw)
In-Reply-To: <20210811173624.28061-1-m.felsch@pengutronix.de>
On 11.08.21 19:36, Marco Felsch wrote:
> The problem with the BootROM is that the SCR registers are not set
s/SCR/SRC/ ?
> accordingly in case of a failed primary boot. E.g. if the device is
> configured to boot from an eMMC and the eMMC is empty or image is
> corrupt, the BootROM goes into 'recovery boot mode' (reference manual
> Figure 8-1) and the last possible recovery option is the serial
> downloader. In such case the SRC registers still indicate that the
> device was booted from an eMMC instead of serial-download.
>
> This commit ports the U-Boot commit [1] with slightly adaptions suggested
> by Ahmad to Barebox. Also we need to reorder the imx6_init() else we
> reset the otg-controller to early.
>
> [1] https://source.denx.de/u-boot/u-boot/-/commit/e203dcf23e9eabc2e4f3d0b079457cd1516f2081
>
> Signed-off-by: Marco Felsch <m.felsch@pengutronix.de>
> ---
> arch/arm/mach-imx/boot.c | 27 +++++++++++++++++++++++++++
> arch/arm/mach-imx/imx6.c | 4 ++--
> include/soc/fsl/fsl_udc.h | 11 +++++++++++
> 3 files changed, 40 insertions(+), 2 deletions(-)
>
> diff --git a/arch/arm/mach-imx/boot.c b/arch/arm/mach-imx/boot.c
> index 2b66bbf71e..fcb3c10064 100644
> --- a/arch/arm/mach-imx/boot.c
> +++ b/arch/arm/mach-imx/boot.c
> @@ -8,6 +8,7 @@
> #include <magicvar.h>
>
> #include <io.h>
> +#include <mach/clock-imx6.h>
> #include <mach/generic.h>
> #include <mach/imx25-regs.h>
> #include <mach/imx27-regs.h>
> @@ -23,6 +24,7 @@
> #include <mach/imx8mq.h>
> #include <mach/imx6.h>
>
> +#include <soc/fsl/fsl_udc.h>
>
> static void
> imx_boot_save_loc(void (*get_boot_source)(enum bootsource *, int *))
> @@ -397,6 +399,11 @@ static u32 imx6_get_src_boot_mode(void __iomem *src_base)
> return readl(src_base + IMX6_SRC_SBMR1);
> }
>
> +static inline bool imx6_usboh3_clk_active(void)
> +{
> + return (readl(MXC_CCM_CCGR6) & 0x3) == 0x3;
> +}
> +
> void imx6_get_boot_source(enum bootsource *src, int *instance)
> {
> void __iomem *src_base = IOMEM(MX6_SRC_BASE_ADDR);
> @@ -410,6 +417,26 @@ void imx6_get_boot_source(enum bootsource *src, int *instance)
>
> bootsrc = imx53_bootsource_internal(bootmode);
>
> + /*
> + * imx6_bootsource_serial() can't detect cases where the boot ROM
> + * decided to use the serial downloader as a fall back (primary
> + * boot source failed).
> + *
> + * Infer that the boot ROM used the USB serial downloader by
> + * checking whether both the UDC and the clock enabling access
> + * to its MMIO region are currently active...
> + * This assumes:
> + * - On fresh boots, PBL doesn't itself start a stopped UDC
> + * - In barebox proper, boot source is saved before the UDC driver
> + * may enable the UDC
> + */
> +
> + if (imx6_usboh3_clk_active() &&
> + is_chipidea_udc_running(IOMEM(MX6_OTG_BASE_ADDR))) {
> + *src = BOOTSOURCE_SERIAL;
> + return;
> + }
> +
> if (imx6_bootsource_serial(sbmr2) ||
> imx6_bootsource_serial_forced(bootsrc)) {
> *src = BOOTSOURCE_SERIAL;
> diff --git a/arch/arm/mach-imx/imx6.c b/arch/arm/mach-imx/imx6.c
> index 9ccb391384..fa4cb093c3 100644
> --- a/arch/arm/mach-imx/imx6.c
> +++ b/arch/arm/mach-imx/imx6.c
> @@ -205,10 +205,10 @@ int imx6_init(void)
> void __iomem *src = IOMEM(MX6_SRC_BASE_ADDR);
> u64 mx6_uid;
>
> - imx6_init_lowlevel();
> -
> imx6_boot_save_loc();
>
> + imx6_init_lowlevel();
A comment inside imx6_init_lowlevel hinting at the dependency w.r.t
imx6_boot_save_loc would be nice.
Apart from these nitpicks:
Reviewed-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
> +
> mx6_silicon_revision = imx6_cpu_revision();
> mx6_uid = imx6_uid();
>
> diff --git a/include/soc/fsl/fsl_udc.h b/include/soc/fsl/fsl_udc.h
> index b983f714c5..0b409a9f6b 100644
> --- a/include/soc/fsl/fsl_udc.h
> +++ b/include/soc/fsl/fsl_udc.h
> @@ -1,6 +1,9 @@
> #ifndef __FSL_UDC_H
> #define __FSL_UDC_H
>
> +#include <linux/types.h>
> +#include <io.h>
> +
> /* USB DR device mode registers (Little Endian) */
> struct usb_dr_device {
> /* Capability register */
> @@ -380,4 +383,12 @@ int imx_barebox_start_usb(void __iomem *dr, void *dest);
> int imx8mm_barebox_load_usb(void *dest);
> int imx8mm_barebox_start_usb(void *dest);
>
> +static inline bool is_chipidea_udc_running(void __iomem *dr)
> +{
> + struct usb_dr_device __iomem *dr_regs = dr;
> +
> + return (readl(&dr_regs->usbmode) & USB_MODE_CTRL_MODE_DEVICE)
> + && (readl(&dr_regs->usbcmd) & USB_CMD_RUN_STOP);
> +}
> +
> #endif /* __FSL_UDC_H */
>
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox
next prev parent reply other threads:[~2021-08-11 18:32 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-11 17:36 Marco Felsch
2021-08-11 18:30 ` Ahmad Fatoum [this message]
2021-08-12 7:38 ` Marco Felsch
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=241f3046-fdaf-ccd9-bc43-8c6879da75d8@pengutronix.de \
--to=a.fatoum@pengutronix.de \
--cc=barebox@lists.infradead.org \
--cc=m.felsch@pengutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox