Re: [PATCH] fat: Simplify calc_fat_clusters code

[Seunghun Lee <waydi1@gmail.com>]

On 09/03/2014 12:37 AM, OGAWA Hirofumi wrote:
> Seunghun Lee <waydi1@gmail.com> writes:
>
>> diff --git a/fs/fat/inode.c b/fs/fat/inode.c
>> index 756aead..6992dea 100644
>> --- a/fs/fat/inode.c
>> +++ b/fs/fat/inode.c
>> @@ -1307,12 +1307,9 @@ static unsigned long calc_fat_clusters(struct super_block *sb)
>>  	struct msdos_sb_info *sbi = MSDOS_SB(sb);
>>  
>>  	/* Divide first to avoid overflow */
>> -	if (sbi->fat_bits != 12) {
>> -		unsigned long ent_per_sec = sb->s_blocksize * 8 / sbi->fat_bits;
>> -		return ent_per_sec * sbi->fat_length;
>> -	}
>> +	unsigned long ent_per_sec = sb->s_blocksize * 8 / sbi->fat_bits;
>>  
>> -	return sbi->fat_length * sb->s_blocksize * 8 / sbi->fat_bits;
>> +	return ent_per_sec * sbi->fat_length;
>>  }
> When sbi->fat_bits == 12, it doesn't work, right? (there is the remainder)
>
> Thanks.
Yes, you are right.

I didn't think about the remainder.

Thanks.