Re: [PATCH v3 3/3] ata: libata: Print horkages applied to devices

[Damien Le Moal <dlemoal@kernel.org>]

On 7/23/24 06:30, Igor Pylypiv wrote:
>>  static const struct ata_dev_horkage_entry ata_dev_horkages[] = {
>> @@ -4266,21 +4329,24 @@ static const struct ata_dev_horkage_entry ata_dev_horkages[] = {
>>  	{ }
>>  };
>>  
>> -static unsigned long ata_dev_horkage(const struct ata_device *dev)
>> +static unsigned int ata_dev_horkage(const struct ata_device *dev)
>>  {
>>  	unsigned char model_num[ATA_ID_PROD_LEN + 1];
>>  	unsigned char model_rev[ATA_ID_FW_REV_LEN + 1];
>>  	const struct ata_dev_horkage_entry *ad = ata_dev_horkages;
>>  
>> +	/* dev->horkage is an unsigned int. */
>> +	BUILD_BUG_ON(__ATA_HORKAGE_MAX > 31);
> 
> Should this check be '__ATA_HORKAGE_MAX > 32'?
> 
> When __ATA_HORKAGE_MAX is 32 then the last horkage bit will be 31.

Oops... Yes, of course you are right. Good catch.

>> +enum ata_horkage {
>> +	__ATA_HORKAGE_DIAGNOSTIC,	/* Failed boot diag */
>> +	__ATA_HORKAGE_NODMA,		/* DMA problems */
>> +	__ATA_HORKAGE_NONCQ,		/* Don't use NCQ */
>> +	__ATA_HORKAGE_MAX_SEC_128,	/* Limit max sects to 128 */
>> +	__ATA_HORKAGE_BROKEN_HPA,	/* Broken HPA */
>> +	__ATA_HORKAGE_DISABLE,		/* Disable it */
>> +	__ATA_HORKAGE_HPA_SIZE,		/* native size off by one */
>> +	__ATA_HORKAGE_IVB,		/* cbl det validity bit bugs */
>> +	__ATA_HORKAGE_STUCK_ERR,	/* stuck ERR on next PACKET */
>> +	__ATA_HORKAGE_BRIDGE_OK,	/* no bridge limits */
>> +	__ATA_HORKAGE_ATAPI_MOD16_DMA,	/* use ATAPI DMA for commands
>> +					    not multiple of 16 bytes */
>> +	__ATA_HORKAGE_FIRMWARE_WARN,	/* firmware update warning */
>> +	__ATA_HORKAGE_1_5_GBPS,		/* force 1.5 Gbps */
>> +	__ATA_HORKAGE_NOSETXFER,		/* skip SETXFER, SATA only */
> nit: extra tab                          ^^^^^^^^

Yep. Fixed.

-- 
Damien Le Moal
Western Digital Research