[Ocfs2-devel] [PATCH 1/4] ocfs2/trivial: Move 'wanted' into parens of ocfs2_resmap_resv_bits.

[Tao Ma <tao.ma@oracle.com>]

Hi Mark,

On 08/14/2010 07:16 AM, Mark Fasheh wrote:
> Hi Tao,
>
> On Mon, Aug 02, 2010 at 11:02:12AM +0800, Tao Ma wrote:
>> The first time I read the function ocfs2_resmap_resv_bits, I consider
>> about what 'wanted' will be used and consider about the comments.
>> Then I find it is only used if the reservation is empty. ;)
>>
>> So we'd better move it to the parens so that it make the code more
>> readable, what's more, ocfs2_resmap_resv_bits is used so frequently
>> and we should save some cpus. The corresponding BUG_ON is also moved
>> into parens since it is only meaningful after we reinit the resv.
>>
>> Cc: Mark Fasheh<mfasheh@suse.com>
>> Signed-off-by: Tao Ma<tao.ma@oracle.com>
>> ---
>>   fs/ocfs2/reservations.c |   26 ++++++++++++--------------
>>   1 files changed, 12 insertions(+), 14 deletions(-)
>>
>> diff --git a/fs/ocfs2/reservations.c b/fs/ocfs2/reservations.c
>> index d8b6e42..567c1a0 100644
>> --- a/fs/ocfs2/reservations.c
>> +++ b/fs/ocfs2/reservations.c
>> @@ -732,25 +732,23 @@ int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap,
>>   			   struct ocfs2_alloc_reservation *resv,
>>   			   int *cstart, int *clen)
>>   {
>> -	unsigned int wanted = *clen;
>> -
>>   	if (resv == NULL || ocfs2_resmap_disabled(resmap))
>>   		return -ENOSPC;
>>
>>   	spin_lock(&resv_lock);
>>
>> -	/*
>> -	 * We don't want to over-allocate for temporary
>> -	 * windows. Otherwise, we run the risk of fragmenting the
>> -	 * allocation space.
>> -	 */
>> -	wanted = ocfs2_resv_window_bits(resmap, resv);
>> -	if ((resv->r_flags&  OCFS2_RESV_FLAG_TMP) || wanted<  *clen)
>> -		wanted = *clen;
>> -
>>   	if (ocfs2_resv_empty(resv)) {
>> -		mlog(0, "empty reservation, find new window\n");
>> +		/*
>> +		 * We don't want to over-allocate for temporary
>> +		 * windows. Otherwise, we run the risk of fragmenting the
>> +		 * allocation space.
>> +		 */
>> +		unsigned int wanted = ocfs2_resv_window_bits(resmap, resv);
>> +
>> +		if ((resv->r_flags&  OCFS2_RESV_FLAG_TMP) || wanted<  *clen)
>> +			wanted = *clen;
>>
>> +		mlog(0, "empty reservation, find new window\n");
>>   		/*
>>   		 * Try to get a window here. If it works, we must fall
>>   		 * through and test the bitmap . This avoids some
>> @@ -759,9 +757,9 @@ int ocfs2_resmap_resv_bits(struct ocfs2_reservation_map *resmap,
>>   		 * that inode.
>>   		 */
>>   		ocfs2_resv_find_window(resmap, resv, wanted);
>> -	}
>>
>> -	BUG_ON(ocfs2_resv_empty(resv));
>> +		BUG_ON(ocfs2_resv_empty(resv));
>> +	}
>
> Can we leave the BUG_ON() outside the if (ocfs2_resv_empty(...)) { } block?
> You're technically correct in that the only possibility right now is if
> there's a bug in ocfs2_resv_find_window(). However, I think it still belongs
> outside the block because we're returning an allocation at that point. The
> code is self documenting then - "don't let it get here unless resv->r_start
> and resv->r_len mean something". That way any future changes to the function
> will be forced to take this BUG_ON() into consideration, and we are then
> less likely to accidentally corrupt data.
Actually I have thought of it. But please note that the if check is 
ocfs2_resv_empty. So does it look a little bit strange that:
if (ocfs2_resv_empty()) {
}

BUG_ON(ocfs2_resv_empty());

We have just checked ocfs2_resv_empty and now we BUG_ON it again?

Regards,
Tao