Andrea Righi <andrea.righi@xxxxxxxxxxxxx> writes:
> On Wed, Apr 15, 2020 at 03:44:08PM +0800, Huang, Ying wrote:
>> Andrea Righi <andrea.righi@xxxxxxxxxxxxx> writes:
>>
>> > mm/swapfile.c | 4 +++-
>> > 1 file changed, 3 insertions(+), 1 deletion(-)
>> >
>> > diff --git a/mm/swapfile.c b/mm/swapfile.c
>> > index 9fd47e6f7a86..cb9eb517178d 100644
>> > --- a/mm/swapfile.c
>> > +++ b/mm/swapfile.c
>> > @@ -1944,7 +1944,9 @@ static int unuse_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
>> > vmf.pmd = pmd;
>> > last_ra = atomic_read(&last_readahead_pages);
>> > atomic_set(&swapin_readahead_hits, last_ra);
>>
>> You need to remove the above 2 lines firstly.
>
> Meh... too much enthusiasm, and I definitely need more coffee this
> morning. Here's the right patch applied:
>
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 5871a2aa86a5..8b38441b66fa 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -1940,7 +1940,9 @@ static int unuse_pte_range(struct vm_area_struct *vma, pmd_t *pmd,
> vmf.vma = vma;
> vmf.address = addr;
> vmf.pmd = pmd;
> - page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE, &vmf);
> + page = lookup_swap_cache(entry, vma, addr);
> + if (!page)
> + page = swapin_readahead(entry, GFP_HIGHUSER_MOVABLE, &vmf);
The vmf assignment can be moved inside "if" block. Otherwise the patch
looks good to me.
> if (!page) {
> if (*swap_map == 0 || *swap_map == SWAP_MAP_BAD)
> goto try_next;
>
> And following the right results:
>
> r::swapin_nr_pages(unsigned long offset):unsigned long:$retval
> COUNT EVENT
> 1618 $retval = 1
> 4960 $retval = 2
> 41315 $retval = 4
> 103521 $retval = 8
>
> swapoff time: 12.19s
>
> So, not as good as the fixed-size readahead, but it's definitely an
> improvement, considering that the swapoff time is ~22s without this
> patch applied.
>
> I think this change can be a simple and reasonable compromise.
Yes. I think so too.
Best Regards,
Huang, Ying
> Thanks again and sorry for the noise,
> -Andrea
