On 6/15/23 10:34, Rob Herring wrote:
>> Something like this works for ahci_octeon:
>>
>> diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig
>> index 4572f837e504..6b115685a86e 100644
>> --- a/drivers/ata/Kconfig
>> +++ b/drivers/ata/Kconfig
>> @@ -239,7 +239,7 @@ config AHCI_MVEBU
>>
>> config AHCI_OCTEON
>> tristate "Cavium Octeon Soc Serial ATA"
>> - depends on SATA_AHCI_PLATFORM && MIPS && (CAVIUM_OCTEON_SOC || COMPILE_TEST)
>> + depends on SATA_AHCI_PLATFORM && ((MIPS && CAVIUM_OCTEON_SOC) ||
>> COMPILE_TEST)
>> default y
>> help
>> This option enables support for Cavium Octeon SoC Serial ATA.
>> diff --git a/drivers/ata/ahci_octeon.c b/drivers/ata/ahci_octeon.c
>> index 5021ab3ede49..2e6b28bf43f4 100644
>> --- a/drivers/ata/ahci_octeon.c
>> +++ b/drivers/ata/ahci_octeon.c
>> @@ -15,8 +15,13 @@
>> #include <linux/platform_device.h>
>> #include <linux/of_platform.h>
>>
>> +#ifdef CONFIG_MIPS
>> #include <asm/octeon/octeon.h>
>> #include <asm/bitfield.h>
>> +#else
>> +#define cvmx_readq_csr(csr) (0)
>> +#define cvmx_writeq_csr(csr, cfg) do { } while (0)
>> +#endif
>
> That's kind of ugly...
Yes :)
>
>>
>> #define CVMX_SATA_UCTL_SHIM_CFG 0xE8
>>
>> But it is not that simple for pata_octeon...
>
> I copied you on mine[1]. It doesn't build for !MIPS, but at least
> doesn't depend on a specific config. Should be good enough for 0-day
> to build somewhat quickly.
Yes, replied to that. The patches look good to me. An Ack from the MIPS folks on
patch 1 would be nice though.
>
> Rob
>
> [1] https://lore.kernel.org/all/20230614173633.2430653-2-robh@xxxxxxxxxx/
--
Damien Le Moal
Western Digital Research
