Re: [PATCH v3 3/9] mm: slightly clarify KSM logic in do_swap_page()

[David Hildenbrand <david@redhat.com>]

On 09.03.22 19:48, Yang Shi wrote:
> On Mon, Jan 31, 2022 at 8:33 AM David Hildenbrand <david@redhat.com> wrote:
>>
>> Let's make it clearer that KSM might only have to copy a page
>> in case we have a page in the swapcache, not if we allocated a fresh
>> page and bypassed the swapcache. While at it, add a comment why this is
>> usually necessary and merge the two swapcache conditions.
>>
>> Signed-off-by: David Hildenbrand <david@redhat.com>
>> ---
>>  mm/memory.c | 38 +++++++++++++++++++++++---------------
>>  1 file changed, 23 insertions(+), 15 deletions(-)
>>
>> diff --git a/mm/memory.c b/mm/memory.c
>> index 923165b4c27e..3c91294cca98 100644
>> --- a/mm/memory.c
>> +++ b/mm/memory.c
>> @@ -3615,21 +3615,29 @@ vm_fault_t do_swap_page(struct vm_fault *vmf)
>>                 goto out_release;
>>         }
>>
>> -       /*
>> -        * Make sure try_to_free_swap or reuse_swap_page or swapoff did not
> 
> We could remove the reference to "reuse_swap_page", right?
>
Yes, I noticed this a couple of days ago as well and already have a
patch prepared for that ("mm: adjust stale comment in do_swap_page()
mentioning reuse_swap_page()" at
https://github.com/davidhildenbrand/linux/commits/cow_fixes_part_3)

If Andrew wants, we can fix that up directly before sending upstream or
I'll simply include that patch when sending out part2 v2.

(I want to avoid sending another series just for this)

Thanks!

-- 
Thanks,

David / dhildenb