Re: [PATCH v3 3/6] rtc: goldfish: Support platform data for non-DT probing

[Heinrich Schuchardt <heinrich.schuchardt@canonical.com>]

On 12/30/25 17:01, Kuan-Wei Chiu wrote:
> Currently, the Goldfish RTC driver exclusively relies on device tree
> to retrieve the base address, failing immediately if dev_read_addr()
> returns FDT_ADDR_T_NONE. This restriction prevents the driver from
> being used on platforms that instantiate devices via U_BOOT_DRVINFO()
> instead of device tree, such as the QEMU m68k virt machine.
> 
> Add support for platform data to address this limitation. Update the
> probe function to fall back to retrieving the base address from
> struct goldfish_rtc_plat if the device tree address is unavailable.
> Introduce a new header file include/goldfish_rtc.h to define the
> platform data structure.
> 
> Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
> ---
> Changes in v3:
> - New patch.
> 
> It appears there is no specific maintainer listed for this driver?
> 
>   drivers/rtc/goldfish_rtc.c | 10 ++++++++--
>   include/goldfish_rtc.h     | 15 +++++++++++++++
>   2 files changed, 23 insertions(+), 2 deletions(-)
>   create mode 100644 include/goldfish_rtc.h
> 
> diff --git a/drivers/rtc/goldfish_rtc.c b/drivers/rtc/goldfish_rtc.c
> index e63a2766c76..900daabccb9 100644
> --- a/drivers/rtc/goldfish_rtc.c
> +++ b/drivers/rtc/goldfish_rtc.c
> @@ -9,6 +9,7 @@
>   
>   #include <div64.h>
>   #include <dm.h>
> +#include <goldfish_rtc.h>
>   #include <mapmem.h>
>   #include <rtc.h>
>   #include <linux/io.h>
> @@ -77,11 +78,16 @@ static int goldfish_rtc_set(struct udevice *dev, const struct rtc_time *time)
>   static int goldfish_rtc_probe(struct udevice *dev)
>   {
>   	struct goldfish_rtc *priv = dev_get_priv(dev);
> +	struct goldfish_rtc_plat *plat;
>   	fdt_addr_t addr;
>   
>   	addr = dev_read_addr(dev);
> -	if (addr == FDT_ADDR_T_NONE)
> -		return -EINVAL;
> +	if (addr == FDT_ADDR_T_NONE) {
> +		plat = dev_get_plat(dev);
> +		if (!plat)
> +			return -EINVAL;
> +		addr = (fdt_addr_t)plat->base;

addr is not a pointer, it is an address in the sandbox virtual memory 
address space. This is why map_sysmem() is used below.

The suggested code fails to build on sandbox_defconfig with the driver 
enabled:

drivers/rtc/goldfish_rtc.c: In function ‘goldfish_rtc_probe’:
drivers/rtc/goldfish_rtc.c:89:24: warning: cast from pointer to integer 
of different size [-Wpointer-to-int-cast]
    89 |                 addr = (fdt_addr_t)plat->base;
       |

Please, use

	priv->base = plat->base;

If plat->base is NULL, please, throw -EINVAL.

Best regards

Heinrich

> +	}
>   	priv->base = map_sysmem(addr, 0x20);
>   
>   	return 0;
> diff --git a/include/goldfish_rtc.h b/include/goldfish_rtc.h
> new file mode 100644
> index 00000000000..bb113a0bd79
> --- /dev/null
> +++ b/include/goldfish_rtc.h
> @@ -0,0 +1,15 @@
> +/* SPDX-License-Identifier: GPL-2.0-or-later */
> +/*
> + * Copyright (C) 2025, Kuan-Wei Chiu <visitorckw@gmail.com>
> + */
> +
> +#ifndef _GOLDFISH_RTC_H_
> +#define _GOLDFISH_RTC_H_
> +
> +#include <linux/types.h>
> +
> +struct goldfish_rtc_plat {
> +	void __iomem *base;
> +};
> +
> +#endif /* _GOLDFISH_RTC_H_ */