On Fri, Mar 6, 2009 at 9:15 AM, Andre Noll <maan@xxxxxxxxxxxxxxx> wrote:
> On 17:24, Dan Williams wrote:
>> +raid10_size(mddev_t *mddev, sector_t sectors, int raid_disks)
>> +{
>> + sector_t size;
>> + int chunk_shift;
>> + conf_t *conf = mddev_to_conf(mddev);
>> + int chunk_size = mddev->chunk_size;
>> +
>> + if (!raid_disks)
>> + raid_disks = mddev->raid_disks;
>> + if (!sectors)
>> + sectors = mddev->dev_sectors;
>> +
>> + chunk_shift = ffz(~chunk_size) - 9;
>> + size = sectors >> chunk_shift;
>> + sector_div(size, conf->far_copies);
>> + size = size * raid_disks;
>> + sector_div(size, conf->near_copies);
>> +
>> + return size << chunk_shift;
>> +}
>
> Is there a reason you are recomputing chunk_shift instead of using
> conf->chunk_shift.
>
This is a leftover from an earlier version of the patch where I was
allowing chunk_size to specified. Since we do not support chunk_size
reshaping I dropped the extra argument, will clean this up as well.
>> +static sector_t
>> +raid5_size(mddev_t *mddev, sector_t sectors, int raid_disks)
>> +{
>> + raid5_conf_t *conf = mddev_to_conf(mddev);
>> + int chunk_size = mddev->chunk_size;
>> +
>> + if (!sectors)
>> + sectors = mddev->dev_sectors;
>> + if (!raid_disks)
>> + raid_disks = conf->previous_raid_disks;
>> +
>> + sectors &= ~(chunk_size / 512 - 1);
>
> Don't we need a cast to sector_t here?
>
Yes.
Thanks,
Dan
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
