Re: [PATCH v1 05/15] mm/rmap: convert RMAP flags to a proper distinct rmap_t type

[David Hildenbrand]

On 08.03.22 18:15, Nadav Amit wrote:
> 
> 
>> On Mar 8, 2022, at 6:14 AM, David Hildenbrand <david@xxxxxxxxxx> wrote:
>>
>> We want to pass the flags to more than one anon rmap function, getting
>> rid of special "do_page_add_anon_rmap()". So let's pass around a distinct
>> __bitwise type and refine documentation.
>>
>> Signed-off-by: David Hildenbrand <david@xxxxxxxxxx>
>> ---
>> include/linux/rmap.h | 22 ++++++++++++++++++----
>> mm/memory.c          |  6 +++---
>> mm/rmap.c            |  7 ++++---
>> 3 files changed, 25 insertions(+), 10 deletions(-)
>>
>> diff --git a/include/linux/rmap.h b/include/linux/rmap.h
>> index 92c3585b8c6a..49f6b208938c 100644
>> --- a/include/linux/rmap.h
>> +++ b/include/linux/rmap.h
>> @@ -158,9 +158,23 @@ static inline void anon_vma_merge(struct vm_area_struct *vma,
>>
>> struct anon_vma *page_get_anon_vma(struct page *page);
>>
>> -/* bitflags for do_page_add_anon_rmap() */
>> -#define RMAP_EXCLUSIVE 0x01
>> -#define RMAP_COMPOUND 0x02
>> +/* RMAP flags, currently only relevant for some anon rmap operations. */
>> +typedef int __bitwise rmap_t;
>> +
>> +/*
>> + * No special request: if the page is a subpage of a compound page, it is
>> + * mapped via a PTE. The mapped (sub)page is possibly shared between processes.
>> + */
>> +#define RMAP_NONE		((__force rmap_t)0)
>> +
>> +/* The (sub)page is exclusive to a single process. */
>> +#define RMAP_EXCLUSIVE		((__force rmap_t)BIT(0))
>> +
>> +/*
>> + * The compound page is not mapped via PTEs, but instead via a single PMD and
>> + * should be accounted accordingly.
>> + */
>> +#define RMAP_COMPOUND		((__force rmap_t)BIT(1))
> 

Hi Nadav,

> I was once shouted at for a similar suggestion, but I am going to try
> once more… If you already define a new type, why not to use bitfields?

I don't have a strong opinion, however, I'd prefer keeping it consistent
with existing ways of passing flags.

Personally, I like __bitwise because it just behave the way we're used
to pass flags -- with additional type safety.

Especially once eventually passing many flags (like we do with GFP),
bitfields might turn out rather nasty -- IMHO.


Thanks!

-- 
Thanks,

David / dhildenb