On 2018/10/12 19:23, Hannes Reinecke wrote:
> On 10/12/18 12:08 PM, Damien Le Moal wrote:
>> The unsigned 32 bits overflow check for the zone size value is already
>> done within sd_zbc_check_zones() with the test:
>>
>> } else if (logical_to_sectors(sdkp->device, zone_blocks) > UINT_MAX) {
>>
>> so there is no need to check again for an out of range value in
>> sd_zbc_read_zones(). Simplify the code and fix sd_zbc_check_zones()
>> error return to -EFBIG instead of -ENODEV if the zone size is too large.
>> Change the return type of sd_zbc_check_zones() to an int for the error
>> code and return the zone size (zone_blocks) through a u32 pointer to
>> avoid overflowing the signed 32 return value.
>>
>> Signed-off-by: Damien Le Moal <damien.lemoal@xxxxxxx>
>> ---
>> drivers/scsi/sd_zbc.c | 19 ++++++++-----------
>> 1 file changed, 8 insertions(+), 11 deletions(-)
>>
>> diff --git a/drivers/scsi/sd_zbc.c b/drivers/scsi/sd_zbc.c
>> index ca73c46931c0..0678e1e108b0 100644
>> --- a/drivers/scsi/sd_zbc.c
>> +++ b/drivers/scsi/sd_zbc.c
>> @@ -373,7 +373,7 @@ static int sd_zbc_check_zoned_characteristics(struct scsi_disk *sdkp,
>> * Returns the zone size in number of blocks upon success or an error code
>> * upon failure.
>> */
>> -static s64 sd_zbc_check_zones(struct scsi_disk *sdkp)
>> +static int sd_zbc_check_zones(struct scsi_disk *sdkp, u32 *zblocks)
>> {
>> u64 zone_blocks = 0;
>> sector_t max_lba, block = 0;
>
> Actually I thought to just change the 's32' to 'int', and not adding
> another parameter; but anyway.
Yes, I understood that. But since chunk_sectors is unsigned int, zone_blocks has
to be too and so returning that through an int would be asking for troubles. If
we ever have see a drive with a 2G LBA zone size that is :)
I thought it was cleaner this way.
>
> Reviewed-by: Hannes Reinecke <hare@xxxxxxxx>
Thanks !
>
> Hannes
>
--
Damien Le Moal
Western Digital Research
