Re: [PATCH v4 01/11] include: fdtdec: decouple fdt_addr_t and phys_addr_t size

[Johan Jonker <jbx6244@gmail.com>]

On 2/23/23 01:47, Kever Yang wrote:
> Hi Johan,
> 
>     This update will cause below error on evb-ast2500/2600:

Hi Kever, Simon,

For "spi-aspeed-smc.c" I have submitted a patch proposal.
The changes in patch 1 possible involves a lot more boards that might need a change to which the external MAINTAINERS may or may not respond to.
As that becomes a lengthy time consuming process. Is it maybe possible to skip patch 1 for now and get at least the more essential patches 2-11 merged?
Please advise what's the best option here to get forward.

Johan


> 
> $ tools/buildman/buildman -o ${UBOOT_TRAVIS_BUILD_DIR} -w -E -W -e --board ${TEST_PY_BD} ${OVERRIDE}
> Building current source for 1 boards (1 thread, 24 jobs per thread)
>        arm:  +   evb-ast2600
> +drivers/spi/spi-aspeed-smc.c: In function 'apseed_spi_of_to_plat':
> +drivers/spi/spi-aspeed-smc.c:1154:22: error: format '%lx' expects argument of type 'long unsigned int', but argument 4 has type 'fdt_size_t' {aka 'unsigned int'} [-Werror=format=]
> + 1154 |         dev_dbg(bus, "ctrl_base = 0x%x, ahb_base = 0x%p, size = 0x%lx\n",
> +      | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> + 1155 |                 (u32)priv->regs, plat->ahb_base, plat->ahb_sz);
> +      | ~~~~~~~~~~~~
> +      |                                                      |
> +      | fdt_size_t {aka unsigned int}
> +include/dm/device_compat.h:50:24: note: in definition of macro 'dev_printk_emit'
> +   50 |                 printf(fmt, ##__VA_ARGS__); \
> +      |                        ^~~
> +include/dm/device_compat.h:117:9: note: in expansion of macro '__dev_printk'
> +  117 |         __dev_printk(LOGL_DEBUG, dev, fmt, ##__VA_ARGS__)
> +      |         ^~~~~~~~~~~~
> +drivers/spi/spi-aspeed-smc.c:1154:9: note: in expansion of macro 'dev_dbg'
> +      |         ^~~~~~~
> +drivers/spi/spi-aspeed-smc.c:1154:69: note: format string is defined here
> + | ~~^
> + | |
> + | long unsigned int
> + | %x
> +In file included from include/linux/printk.h:4,
> +                 from include/linux/kernel.h:5,
> +                 from arch/arm/include/asm/io.h:14,
> +                 from drivers/spi/spi-aspeed-smc.c:13:
> +include/log.h:222:24: note: in definition of macro 'log'
> +  222 |                 printf(_fmt, ##_args); \
> +      |                        ^~~~
> +include/dm/device_compat.h:85:17: note: in expansion of macro 'dev_printk_emit'
> +   85 |                 dev_printk_emit(LOG_CATEGORY, level, fmt, ##__VA_ARGS__); \
> +      |                 ^~~~~~~~~~~~~~~
> +include/dm/device_compat.h:54:24: note: in definition of macro 'dev_printk_emit'
> +   54 |                 printf(fmt, ##__VA_ARGS__); \
> +include/dm/device_compat.h:91:41: error: format '%lx' expects argument of type 'long unsigned int', but argument 6 has type 'fdt_size_t' {aka 'unsigned int'} [-Werror=format=]
> +   91 |                                         "%s %s: " fmt, \
> +      |                                         ^~~~~~~~~
> +include/dm/device_compat.h:90:25: note: in expansion of macro 'dev_printk_emit'
> +   90 | dev_printk_emit(__dev->driver->id, level, \
> +      |                         ^~~~~~~~~~~~~~~
> +include/dm/device_compat.h:96:41: error: format '%lx' expects argument of type 'long unsigned int', but argument 6 has type 'fdt_size_t' {aka 'unsigned int'} [-Werror=format=]
> +   96 |                                         "%s %s: " fmt, \
> +include/dm/device_compat.h:95:25: note: in expansion of macro 'dev_printk_emit'
> +   95 |                         dev_printk_emit(LOG_CATEGORY, level, \
> +cc1: all warnings being treated as errors
> +make[3]: *** [scripts/Makefile.build:256: drivers/spi/spi-aspeed-smc.o] Error 1
> +make[2]: *** [scripts/Makefile.build:397: drivers/spi] Error 2
> +make[1]: *** [Makefile:1846: drivers] Error 2
> +make: *** [Makefile:177: sub-make] Error 2
>     0    0    1 /1              evb-ast2600
> 
> 
> Thanks,
> 
> - Kever
> 
> On 2023/2/18 23:26, Johan Jonker wrote:
>> The DT specification supports CPUs with both 32-bit and 64-bit addressing
>> capabilities. In U-boot the fdt_addr_t and phys_addr_t size are coupled
>> by a typedef. The MTD NAND drivers for 32-bit CPU's can describe partitions
>> with a 64-bit reg property. These partitions synced from Linux end up with
>> the wrong offset and sizes when only the lower 32-bit is passed.
>> Decouple the fdt_addr_t and phys_addr_t size as they don't necessary
>> match.
>>
>> Signed-off-by: Johan Jonker <jbx6244@gmail.com>
>> Reviewed-by: Simon Glass <sjg@chromium.org>
>> Reviewed-by: Kever Yang <kever.yang@rock-chips.com>
>> ---
>>
>> Changed V2:
>>    reword
>>
>> ---
>>
>> Note:
>>    Most drivers still assume that FDT and CPU capabilities are identical.
>>    In order to use these variables a cast is needed.
>> ---
>>   Kconfig          |  8 ++++++++
>>   include/fdtdec.h | 13 +++++++++----
>>   2 files changed, 17 insertions(+), 4 deletions(-)
>>
>> diff --git a/Kconfig b/Kconfig
>> index a75cce7e..7697dade 100644
>> --- a/Kconfig
>> +++ b/Kconfig
>> @@ -422,11 +422,19 @@ endif # EXPERT
>>
>>   config PHYS_64BIT
>>       bool "64bit physical address support"
>> +    select FDT_64BIT
>>       help
>>         Say Y here to support 64bit physical memory address.
>>         This can be used not only for 64bit SoCs, but also for
>>         large physical address extension on 32bit SoCs.
>>
>> +config FDT_64BIT
>> +    bool "64bit fdt address support"
>> +    help
>> +      Say Y here to support 64bit fdt addresses.
>> +      This can be used not only for 64bit SoCs, but also
>> +      for large address extensions on 32bit SoCs.
>> +
>>   config HAS_ROM
>>       bool
>>       select BINMAN
>> diff --git a/include/fdtdec.h b/include/fdtdec.h
>> index 12355afd..af29ac0c 100644
>> --- a/include/fdtdec.h
>> +++ b/include/fdtdec.h
>> @@ -18,15 +18,18 @@
>>   #include <pci.h>
>>
>>   /*
>> - * A typedef for a physical address. Note that fdt data is always big
>> + * Support for 64bit fdt addresses.
>> + * This can be used not only for 64bit SoCs, but also
>> + * for large address extensions on 32bit SoCs.
>> + * Note that fdt data is always big
>>    * endian even on a litle endian machine.
>>    */
>> -typedef phys_addr_t fdt_addr_t;
>> -typedef phys_size_t fdt_size_t;
>>
>>   #define FDT_SIZE_T_NONE (-1U)
>>
>> -#ifdef CONFIG_PHYS_64BIT
>> +#ifdef CONFIG_FDT_64BIT
>> +typedef u64 fdt_addr_t;
>> +typedef u64 fdt_size_t;
>>   #define FDT_ADDR_T_NONE ((ulong)(-1))
>>
>>   #define fdt_addr_to_cpu(reg) be64_to_cpu(reg)
>> @@ -35,6 +38,8 @@ typedef phys_size_t fdt_size_t;
>>   #define cpu_to_fdt_size(reg) cpu_to_be64(reg)
>>   typedef fdt64_t fdt_val_t;
>>   #else
>> +typedef u32 fdt_addr_t;
>> +typedef u32 fdt_size_t;
>>   #define FDT_ADDR_T_NONE (-1U)
>>
>>   #define fdt_addr_to_cpu(reg) be32_to_cpu(reg)
>> -- 
>> 2.20.1
>>