Re: [PATCH v2 1/9] mm/hugetlb: remove checking hstate_is_gigantic() in return_unused_surplus_pages()

[HORIGUCHI NAOYA(堀口 直也)]

On Fri, Jun 24, 2022 at 04:15:19PM +0800, Miaohe Lin wrote:
> On 2022/6/24 16:03, Muchun Song wrote:
> > On Fri, Jun 24, 2022 at 10:25:48AM +0800, Miaohe Lin wrote:
> >> On 2022/6/24 7:51, Naoya Horiguchi wrote:
> >>> From: Naoya Horiguchi <naoya.horiguchi@xxxxxxx>
> >>>
> >>> I found a weird state of 1GB hugepage pool, caused by the following
> >>> procedure:
> >>>
> >>>   - run a process reserving all free 1GB hugepages,
> >>>   - shrink free 1GB hugepage pool to zero (i.e. writing 0 to
> >>>     /sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages), then
> >>>   - kill the reserving process.
> >>>
> >>> , then all the hugepages are free *and* surplus at the same time.
> >>>
> >>>   $ cat /sys/kernel/mm/hugepages/hugepages-1048576kB/nr_hugepages
> >>>   3
> >>>   $ cat /sys/kernel/mm/hugepages/hugepages-1048576kB/free_hugepages
> >>>   3
> >>>   $ cat /sys/kernel/mm/hugepages/hugepages-1048576kB/resv_hugepages
> >>>   0
> >>>   $ cat /sys/kernel/mm/hugepages/hugepages-1048576kB/surplus_hugepages
> >>>   3
> >>>
> >>> This state is resolved by reserving and allocating the pages then
> >>> freeing them again, so this seems not to result in serious problem.
> >>> But it's a little surprizing (shrinking pool suddenly fails).
> >>>
> >>> This behavior is caused by hstate_is_gigantic() check in
> >>> return_unused_surplus_pages(). This was introduced so long ago in 2008
> >>> by commit aa888a74977a ("hugetlb: support larger than MAX_ORDER"), and
> >>> it seems to me that this check is no longer unnecessary. Let's remove it.
> >>
> >> s/unnecessary/necessary/

Thanks, I fixed it.

> >>
> >>>
> >>> Signed-off-by: Naoya Horiguchi <naoya.horiguchi@xxxxxxx>
> >>> ---
> >>>  mm/hugetlb.c | 4 ----
> >>>  1 file changed, 4 deletions(-)
> >>>
> >>> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> >>> index a57e1be41401..c538278170a2 100644
> >>> --- a/mm/hugetlb.c
> >>> +++ b/mm/hugetlb.c
> >>> @@ -2432,10 +2432,6 @@ static void return_unused_surplus_pages(struct hstate *h,
> >>>  	/* Uncommit the reservation */
> >>>  	h->resv_huge_pages -= unused_resv_pages;
> >>>  
> >>> -	/* Cannot return gigantic pages currently */
> >>> -	if (hstate_is_gigantic(h))
> >>> -		goto out;
> >>> -
> >>
> >> IIUC it might be better to do the below check:
> >> 	/*
> >> 	 * Cannot return gigantic pages currently if runtime gigantic page
> >> 	 * allocation is not supported.
> >> 	 */
> >> 	if (hstate_is_gigantic(h) && !gigantic_page_runtime_supported())
> >> 		goto out;
> >>
> > 
> > The change looks good to me. However, the comments above is unnecessary
> > since gigantic_page_runtime_supported() is straightforward.
> 
> Agree. The comments can be removed.

Thank you, both. Adding !gigantic_page_runtime_supported without comment
makes sense to me.

Thanks,
Naoya Horiguchi