Re: [PATCH v5 04/13] riscv: sbi: add SBI FWFT extension calls

["Clément Léger" <cleger@rivosinc.com>]

On 24/04/2025 13:06, Andrew Jones wrote:
> On Thu, Apr 17, 2025 at 02:19:51PM +0200, Clément Léger wrote:
>> Add FWFT extension calls. This will be ratified in SBI V3.0 hence, it is
>> provided as a separate commit that can be left out if needed.
>>
>> Signed-off-by: Clément Léger <cleger@rivosinc.com>
>> ---
>>  arch/riscv/kernel/sbi.c | 20 +++++++++++++++++++-
>>  1 file changed, 19 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/riscv/kernel/sbi.c b/arch/riscv/kernel/sbi.c
>> index 379981c2bb21..7b062189b184 100644
>> --- a/arch/riscv/kernel/sbi.c
>> +++ b/arch/riscv/kernel/sbi.c
>> @@ -299,6 +299,8 @@ static int __sbi_rfence_v02(int fid, const struct cpumask *cpu_mask,
>>  	return 0;
>>  }
>>  
>> +static bool sbi_fwft_supported;
> 
> At some point we may want an SBI extension bitmap, but this is only the
> second SBI extension supported boolean that I'm aware of, so I guess we're
> still OK for now.

That seems reasonable to have a bitmap rather than duplicating
*ext*_supported. If that's something that bothers you, I can take care
of it and use a bitmap. SSE will also have a sse_supported boolean but
in it's own driver file though.

> 
>> +
>>  /**
>>   * sbi_fwft_set() - Set a feature on the local hart
>>   * @feature: The feature ID to be set
>> @@ -309,7 +311,15 @@ static int __sbi_rfence_v02(int fid, const struct cpumask *cpu_mask,
>>   */
>>  int sbi_fwft_set(u32 feature, unsigned long value, unsigned long flags)
>>  {
>> -	return -EOPNOTSUPP;
>> +	struct sbiret ret;
>> +
>> +	if (!sbi_fwft_supported)
>> +		return -EOPNOTSUPP;
>> +
>> +	ret = sbi_ecall(SBI_EXT_FWFT, SBI_EXT_FWFT_SET,
>> +			feature, value, flags, 0, 0, 0);
>> +
>> +	return sbi_err_map_linux_errno(ret.error);
>>  }
>>  
>>  struct fwft_set_req {
>> @@ -348,6 +358,9 @@ int sbi_fwft_local_set_cpumask(const cpumask_t *mask, u32 feature,
>>  		.error = ATOMIC_INIT(0),
>>  	};
>>  
>> +	if (!sbi_fwft_supported)
>> +		return -EOPNOTSUPP;
>> +
>>  	if (feature & SBI_FWFT_GLOBAL_FEATURE_BIT)
>>  		return -EINVAL;
>>  
>> @@ -679,6 +692,11 @@ void __init sbi_init(void)
>>  			pr_info("SBI DBCN extension detected\n");
>>  			sbi_debug_console_available = true;
>>  		}
>> +		if ((sbi_spec_version >= sbi_mk_version(3, 0)) &&
>> +		    (sbi_probe_extension(SBI_EXT_FWFT) > 0)) {
> 
> Unnecessary (), but I see it's consistent with the expressions above.

I can fix the other ones as well in another commit.

> 
>> +			pr_info("SBI FWFT extension detected\n");
>> +			sbi_fwft_supported = true;
>> +		}
>>  	} else {
>>  		__sbi_set_timer = __sbi_set_timer_v01;
>>  		__sbi_send_ipi	= __sbi_send_ipi_v01;
>> -- 
>> 2.49.0
>>

Thanks,

Clément

> 
> Besides the () nit
> 
> Reviewed-by: Andrew Jones <ajones@ventanamicro.com>