Re: [PATCH -mm -v4 03/21] mm, THP, swap: Support PMD swap mapping in swap_duplicate()

[Dave Hansen]

> Yes.  Boolean parameter isn't good at most times.  Matthew Wilcox
> suggested to use
> 
>         swap_duplicate(&entry, HPAGE_PMD_NR);
> 
> vs.
> 
>         swap_duplicate(&entry, 1);
> 
> He thinks this makes the interface more flexible to support other swap
> entry size in the future.  What do you think about that?

That looks great to me too.

>>>  		if (likely(!non_swap_entry(entry))) {
>>> -			if (swap_duplicate(entry) < 0)
>>> +			if (swap_duplicate(&entry, false) < 0)
>>>  				return entry.val;
>>>  
>>>  			/* make sure dst_mm is on swapoff's mmlist. */
>>
>> I'll also point out that in a multi-hundred-line patch, adding arguments
>> to a existing function would not be something I'd try to include in the
>> patch.  I'd break it out separately unless absolutely necessary.
> 
> You mean add another patch, which only adds arguments to the function,
> but not change the body of the function?

Yes.  Or, just add the non-THP-swap version first.