lkml.org 
[lkml]   [2021]   [Mar]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: [External] [PATCH 5/8] hugetlb: call update_and_free_page without hugetlb_lock
On Thu, Mar 25, 2021 at 8:29 AM Mike Kravetz <mike.kravetz@oracle.com> wrote:
>
> With the introduction of remove_hugetlb_page(), there is no need for
> update_and_free_page to hold the hugetlb lock. Change all callers to
> drop the lock before calling.
>
> With additional code modifications, this will allow loops which decrease
> the huge page pool to drop the hugetlb_lock with each page to reduce
> long hold times.
>
> The ugly unlock/lock cycle in free_pool_huge_page will be removed in
> a subsequent patch which restructures free_pool_huge_page.
>
> Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>

Reviewed-by: Muchun Song <songmuchun@bytedance.com>

Some nits below.

> ---
> mm/hugetlb.c | 34 +++++++++++++++++++++++++++++-----
> 1 file changed, 29 insertions(+), 5 deletions(-)
>
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 3938ec086b5c..fae7f034d1eb 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -1450,16 +1450,18 @@ static void __free_huge_page(struct page *page)
>
> if (HPageTemporary(page)) {
> remove_hugetlb_page(h, page, false);
> + spin_unlock(&hugetlb_lock);
> update_and_free_page(h, page);
> } else if (h->surplus_huge_pages_node[nid]) {
> /* remove the page from active list */
> remove_hugetlb_page(h, page, true);
> + spin_unlock(&hugetlb_lock);
> update_and_free_page(h, page);
> } else {
> arch_clear_hugepage_flags(page);
> enqueue_huge_page(h, page);
> + spin_unlock(&hugetlb_lock);
> }
> - spin_unlock(&hugetlb_lock);
> }
>
> /*
> @@ -1740,7 +1742,13 @@ static int free_pool_huge_page(struct hstate *h, nodemask_t *nodes_allowed,
> list_entry(h->hugepage_freelists[node].next,
> struct page, lru);
> remove_hugetlb_page(h, page, acct_surplus);
> + /*
> + * unlock/lock around update_and_free_page is temporary
> + * and will be removed with subsequent patch.
> + */
> + spin_unlock(&hugetlb_lock);
> update_and_free_page(h, page);
> + spin_lock(&hugetlb_lock);
> ret = 1;
> break;
> }
> @@ -1809,8 +1817,9 @@ int dissolve_free_huge_page(struct page *page)
> }
> remove_hugetlb_page(h, page, false);
> h->max_huge_pages--;
> + spin_unlock(&hugetlb_lock);
> update_and_free_page(h, head);
> - rc = 0;
> + return 0;
> }
> out:
> spin_unlock(&hugetlb_lock);
> @@ -2563,22 +2572,37 @@ static void try_to_free_low(struct hstate *h, unsigned long count,
> nodemask_t *nodes_allowed)
> {
> int i;
> + struct list_head page_list;

I prefer to use LIST_HEAD(page_list) directly.

> + struct page *page, *next;
>
> if (hstate_is_gigantic(h))
> return;
>
> + /*
> + * Collect pages to be freed on a list, and free after dropping lock
> + */
> + INIT_LIST_HEAD(&page_list);
> for_each_node_mask(i, *nodes_allowed) {
> - struct page *page, *next;
> struct list_head *freel = &h->hugepage_freelists[i];
> list_for_each_entry_safe(page, next, freel, lru) {
> if (count >= h->nr_huge_pages)
> - return;
> + goto out;
> if (PageHighMem(page))
> continue;
> remove_hugetlb_page(h, page, false);
> - update_and_free_page(h, page);
> + INIT_LIST_HEAD(&page->lru);

As Michal pointed out that this is superfluous.

> + list_add(&page->lru, &page_list);
> }
> }
> +
> +out:
> + spin_unlock(&hugetlb_lock);
> + list_for_each_entry_safe(page, next, &page_list, lru) {
> + list_del(&page->lru);

It looks like list_del() is also superfluous. Should we remove it?

Thanks.

> + update_and_free_page(h, page);
> + cond_resched();
> + }
> + spin_lock(&hugetlb_lock);
> }
> #else
> static inline void try_to_free_low(struct hstate *h, unsigned long count,
> --
> 2.30.2
>

\
 
 \ /
  Last update: 2021-03-27 07:56    [W:0.403 / U:0.768 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site