Re: [PATCH] scsi: fix scsi_mode_sense()

[Damien Le Moal <Damien.LeMoal@wdc.com>]

On 2021/08/19 18:30, Niklas Cassel wrote:
> On Thu, Aug 19, 2021 at 04:37:50PM +0900, Damien Le Moal wrote:
>> The allocation length field of the MODE SENSE 10 command is 16-bits,
>> occupying bytes 7 and 8 of the CDB. With this command, access to mode
>> pages larger than 255 bytes is thus possible. Make sure that
>> scsi_mode_sense() code reflects this by initializing the allocation
>> length field using put_unaligned_be16() instead of directly setting
>> only byte 8 of the CDB with the buffer length.
>>
>> While at it, also fix the folowing:
>> * use get_unaligned_be16() to retrieve the mode data length and block
>>   descriptor length fields of the mode sense reply header instead of
>>   using an open coded calculation.
>> * Fix the kdoc dbd argument explanation: the DBD bit stands for
>>   Disable Block Descriptor, which is the opposite of what the dbd
>>   argument description was.
>>
>> Signed-off-by: Damien Le Moal <damien.lemoal@wdc.com>
>> ---
>>  drivers/scsi/scsi_lib.c | 9 ++++-----
>>  1 file changed, 4 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
>> index 7456a26aef51..92db00d81733 100644
>> --- a/drivers/scsi/scsi_lib.c
>> +++ b/drivers/scsi/scsi_lib.c
>> @@ -2070,7 +2070,7 @@ EXPORT_SYMBOL_GPL(scsi_mode_select);
>>  /**
>>   *	scsi_mode_sense - issue a mode sense, falling back from 10 to six bytes if necessary.
>>   *	@sdev:	SCSI device to be queried
>> - *	@dbd:	set if mode sense will allow block descriptors to be returned
>> + *	@dbd:	set to prevent mode sense from returning block descriptors
>>   *	@modepage: mode page being requested
>>   *	@buffer: request buffer (may not be smaller than eight bytes)
>>   *	@len:	length of request buffer.
>> @@ -2112,7 +2112,7 @@ scsi_mode_sense(struct scsi_device *sdev, int dbd, int modepage,
>>  			len = 8;
>>  
>>  		cmd[0] = MODE_SENSE_10;
>> -		cmd[8] = len;
>> +		put_unaligned_be16(len, &cmd[7]);
>>  		header_length = 8;
>>  	} else {
>>  		if (len < 4)
>> @@ -2166,12 +2166,11 @@ scsi_mode_sense(struct scsi_device *sdev, int dbd, int modepage,
>>  		data->longlba = 0;
>>  		data->block_descriptor_length = 0;
>>  	} else if (use_10_for_ms) {
>> -		data->length = buffer[0]*256 + buffer[1] + 2;
>> +		data->length = get_unaligned_be16(&buffer[0]) + 2;
>>  		data->medium_type = buffer[2];
>>  		data->device_specific = buffer[3];
>>  		data->longlba = buffer[4] & 0x01;
>> -		data->block_descriptor_length = buffer[6]*256
>> -			+ buffer[7];
>> +		data->block_descriptor_length = get_unaligned_be16(&buffer[6]);
>>  	} else {
>>  		data->length = buffer[0] + 1;
>>  		data->medium_type = buffer[1];
> 
> (Nit:
> When the subject contains "fix", I think that people automatically look
> for a Fixes-tag. However, AFAIK, there isn't a caller that sends in a
> length > 1 byte. So a slightly better subject might have been:
> "scsi: allow scsi_mode_sense() to read more than 255 bytes")

Indeed, that is better.

Martin,

Can you fix that up ? Or should I resend ?

There is also the fact that the argument len is not being checked against 255.
So if the device does not have use_10_for_ms set, weird things can happen.
I did not add the check because this code has been like this since a long time
and no problem is being reported...

> 
> Reviewed-by: Niklas Cassel <niklas.cassel@wdc.com>
> 


-- 
Damien Le Moal
Western Digital Research