Re: [PATCH 1.1 1/1] arcmsr: Notify has sense data report

[éæé]

There is no this case.
We always have auto request sense to device in case of CHECK_CONDITION status.

2014-11-28 18:14 GMT+08:00 Hannes Reinecke <hare@xxxxxxx>:
> On 11/28/2014 11:07 AM, Ching Huang wrote:
>> From: Ching Huang <ching2048@xxxxxxxxxxxx>
>>
>> Setting command status with 'DRIVER_SENSE' except 'CHECK_CONDITION' if we have sense data.
>>
>> Signed-off-by: Ching Huang <ching2048@xxxxxxxxxxxx>
>> ---
>>
>> This patch is relative to http://git.infradead.org/users/hch/scsi-queue.git/tree/refs/heads/drivers-for-3.18:/drivers/scsi/arcmsr
>>
>>
>> diff -uprN a/drivers/scsi/arcmsr/arcmsr.h b/drivers/scsi/arcmsr/arcmsr.h
>> --- a/drivers/scsi/arcmsr/arcmsr.h    2014-09-19 11:04:41.845100000 +0800
>> +++ b/drivers/scsi/arcmsr/arcmsr.h    2014-11-28 10:54:30.000000000 +0800
>> @@ -52,7 +52,7 @@ struct device_attribute;
>>       #define ARCMSR_MAX_FREECCB_NUM  320
>>  #define ARCMSR_MAX_OUTSTANDING_CMD   255
>>  #endif
>> -#define ARCMSR_DRIVER_VERSION                "v1.30.00.04-20140919"
>> +#define ARCMSR_DRIVER_VERSION                "v1.30.00.05-20141128"
>>  #define ARCMSR_SCSI_INITIATOR_ID                                             255
>>  #define ARCMSR_MAX_XFER_SECTORS                                                      512
>>  #define ARCMSR_MAX_XFER_SECTORS_B                                            4096
>> diff -uprN a/drivers/scsi/arcmsr/arcmsr_hba.c b/drivers/scsi/arcmsr/arcmsr_hba.c
>> --- a/drivers/scsi/arcmsr/arcmsr_hba.c        2014-09-19 11:01:26.990090000 +0800
>> +++ b/drivers/scsi/arcmsr/arcmsr_hba.c        2014-11-28 17:51:04.000000000 +0800
>> @@ -989,16 +989,15 @@ static void arcmsr_report_sense_info(str
>>
>>       struct scsi_cmnd *pcmd = ccb->pcmd;
>>       struct SENSE_DATA *sensebuffer = (struct SENSE_DATA *)pcmd->sense_buffer;
>> -     pcmd->result = DID_OK << 16;
>> -     if (sensebuffer) {
>> -             int sense_data_length =
>> +
>> +     pcmd->result = (DRIVER_SENSE << 24) | (DID_OK << 16) | (CHECK_CONDITION << 1);
>> +     int sense_data_length =
>>                       sizeof(struct SENSE_DATA) < SCSI_SENSE_BUFFERSIZE
>>                       ? sizeof(struct SENSE_DATA) : SCSI_SENSE_BUFFERSIZE;
>> -             memset(sensebuffer, 0, SCSI_SENSE_BUFFERSIZE);
>> -             memcpy(sensebuffer, ccb->arcmsr_cdb.SenseData, sense_data_length);
>> -             sensebuffer->ErrorCode = SCSI_SENSE_CURRENT_ERRORS;
>> -             sensebuffer->Valid = 1;
>> -     }
>> +     memset(sensebuffer, 0, SCSI_SENSE_BUFFERSIZE);
>> +     memcpy(sensebuffer, ccb->arcmsr_cdb.SenseData, sense_data_length);
>> +     sensebuffer->ErrorCode = SCSI_SENSE_CURRENT_ERRORS;
>> +     sensebuffer->Valid = 1;
>>  }
>>
>>  static u32 arcmsr_disable_outbound_ints(struct AdapterControlBlock *acb)
>>
>>
> Hmm. Sorry to be picky, but:
>
> Is there a chance the _no_ sense data will be provided?
> If so we'd end up starting the error handler, calling REQUEST SENSE
> (which will fail, too), and finally offlining the device.
> I'd rather provide a default sense code for these cases.
>
> Plus please use SAM_STAT_CHECK_CONDITION instead of
> CHECK_CONDITION << 1.
>
> Cheers,
>
> Hannes
> --
> Dr. Hannes Reinecke                   zSeries & Storage
> hare@xxxxxxx                          +49 911 74053 688
> SUSE LINUX GmbH, Maxfeldstr. 5, 90409 NÃrnberg
> GF: J. Hawn, J. Guild, F. ImendÃrffer, HRB 21284 (AG NÃrnberg)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/