The patch titled
Subject: mm: ksm: do not block on page lock when searching stable tree
has been added to the -mm tree. Its filename is
mm-ksm-do-not-block-on-page-lock-when-searching-stable-tree.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-ksm-do-not-block-on-page-lock-when-searching-stable-tree.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-ksm-do-not-block-on-page-lock-when-searching-stable-tree.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Yang Shi <yang.shi@xxxxxxxxxxxxxxxxx>
Subject: mm: ksm: do not block on page lock when searching stable tree
ksmd needs to search the stable tree to look for the suitable KSM page,
but the KSM page might be locked for a while due to i.e. KSM page rmap
walk. Basically it is not a big deal since commit 2c653d0ee2ae ("ksm:
introduce ksm_max_page_sharing per page deduplication limit"), since
max_page_sharing limits the number of shared KSM pages.
But it still sounds not worth waiting for the lock, the page can be skip,
then try to merge it in the next scan to avoid potential stall if its
content is still intact.
Introduce trylock mode to get_ksm_page() to not block on page lock, like
what try_to_merge_one_page() does. And, define three possible operations
(nolock, lock and trylock) as enum type to avoid stacking up bools and
make the code more readable.
Return -EBUSY if trylock fails, since NULL means not find suitable KSM
page, which is a valid case.
With the default max_page_sharing setting (256), there is almost no
observed change comparing lock vs trylock.
However, with ksm02 of LTP, the reduced ksmd full scan time can be
observed, which has set max_page_sharing to 786432. With lock version,
ksmd may tak 10s - 11s to run two full scans, with trylock version ksmd
may take 8s - 11s to run two full scans. And, the number of pages_sharing
and pages_to_scan keep same. Basically, this change has no harm.
Link: http://lkml.kernel.org/r/1548793753-62377-1-git-send-email-yang.shi@xxxxxxxxxxxxxxxxx
Signed-off-by: Yang Shi <yang.shi@xxxxxxxxxxxxxxxxx>
Suggested-by: John Hubbard <jhubbard@xxxxxxxxxx>
Reviewed-by: Kirill Tkhai <ktkhai@xxxxxxxxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/ksm.c | 46 ++++++++++++++++++++++++++++++++++++----------
1 file changed, 36 insertions(+), 10 deletions(-)
--- a/mm/ksm.c~mm-ksm-do-not-block-on-page-lock-when-searching-stable-tree
+++ a/mm/ksm.c
@@ -667,6 +667,12 @@ static void remove_node_from_stable_tree
free_stable_node(stable_node);
}
+enum get_ksm_page_flags {
+ GET_KSM_PAGE_NOLOCK,
+ GET_KSM_PAGE_LOCK,
+ GET_KSM_PAGE_TRYLOCK
+};
+
/*
* get_ksm_page: checks if the page indicated by the stable node
* is still its ksm page, despite having held no reference to it.
@@ -686,7 +692,8 @@ static void remove_node_from_stable_tree
* a page to put something that might look like our key in page->mapping.
* is on its way to being freed; but it is an anomaly to bear in mind.
*/
-static struct page *get_ksm_page(struct stable_node *stable_node, bool lock_it)
+static struct page *get_ksm_page(struct stable_node *stable_node,
+ enum get_ksm_page_flags flags)
{
struct page *page;
void *expected_mapping;
@@ -729,8 +736,15 @@ again:
goto stale;
}
- if (lock_it) {
+ if (flags == GET_KSM_PAGE_TRYLOCK) {
+ if (!trylock_page(page)) {
+ put_page(page);
+ return ERR_PTR(-EBUSY);
+ }
+ } else if (flags == GET_KSM_PAGE_LOCK)
lock_page(page);
+
+ if (flags != GET_KSM_PAGE_NOLOCK) {
if (READ_ONCE(page->mapping) != expected_mapping) {
unlock_page(page);
put_page(page);
@@ -764,7 +778,7 @@ static void remove_rmap_item_from_tree(s
struct page *page;
stable_node = rmap_item->head;
- page = get_ksm_page(stable_node, true);
+ page = get_ksm_page(stable_node, GET_KSM_PAGE_LOCK);
if (!page)
goto out;
@@ -864,7 +878,7 @@ static int remove_stable_node(struct sta
struct page *page;
int err;
- page = get_ksm_page(stable_node, true);
+ page = get_ksm_page(stable_node, GET_KSM_PAGE_LOCK);
if (!page) {
/*
* get_ksm_page did remove_node_from_stable_tree itself.
@@ -1386,7 +1400,7 @@ static struct page *stable_node_dup(stru
* stable_node parameter itself will be freed from
* under us if it returns NULL.
*/
- _tree_page = get_ksm_page(dup, false);
+ _tree_page = get_ksm_page(dup, GET_KSM_PAGE_NOLOCK);
if (!_tree_page)
continue;
nr += 1;
@@ -1509,7 +1523,7 @@ static struct page *__stable_node_chain(
if (!is_stable_node_chain(stable_node)) {
if (is_page_sharing_candidate(stable_node)) {
*_stable_node_dup = stable_node;
- return get_ksm_page(stable_node, false);
+ return get_ksm_page(stable_node, GET_KSM_PAGE_NOLOCK);
}
/*
* _stable_node_dup set to NULL means the stable_node
@@ -1614,7 +1628,8 @@ again:
* wrprotected at all times. Any will work
* fine to continue the walk.
*/
- tree_page = get_ksm_page(stable_node_any, false);
+ tree_page = get_ksm_page(stable_node_any,
+ GET_KSM_PAGE_NOLOCK);
}
VM_BUG_ON(!stable_node_dup ^ !!stable_node_any);
if (!tree_page) {
@@ -1674,7 +1689,12 @@ again:
* It would be more elegant to return stable_node
* than kpage, but that involves more changes.
*/
- tree_page = get_ksm_page(stable_node_dup, true);
+ tree_page = get_ksm_page(stable_node_dup,
+ GET_KSM_PAGE_TRYLOCK);
+
+ if (PTR_ERR(tree_page) == -EBUSY)
+ return ERR_PTR(-EBUSY);
+
if (unlikely(!tree_page))
/*
* The tree may have been rebalanced,
@@ -1843,7 +1863,8 @@ again:
* wrprotected at all times. Any will work
* fine to continue the walk.
*/
- tree_page = get_ksm_page(stable_node_any, false);
+ tree_page = get_ksm_page(stable_node_any,
+ GET_KSM_PAGE_NOLOCK);
}
VM_BUG_ON(!stable_node_dup ^ !!stable_node_any);
if (!tree_page) {
@@ -2061,6 +2082,10 @@ static void cmp_and_merge_page(struct pa
/* We first start with searching the page inside the stable tree */
kpage = stable_tree_search(page);
+
+ if (PTR_ERR(kpage) == -EBUSY)
+ return;
+
if (kpage == page && rmap_item->head == stable_node) {
put_page(kpage);
return;
@@ -2243,7 +2268,8 @@ static struct rmap_item *scan_get_next_r
list_for_each_entry_safe(stable_node, next,
&migrate_nodes, list) {
- page = get_ksm_page(stable_node, false);
+ page = get_ksm_page(stable_node,
+ GET_KSM_PAGE_NOLOCK);
if (page)
put_page(page);
cond_resched();
_
Patches currently in -mm which might be from yang.shi@xxxxxxxxxxxxxxxxx are
mm-swap-check-if-swap-backing-device-is-congested-or-not.patch
mm-swap-check-if-swap-backing-device-is-congested-or-not-fix-2.patch
mm-swap-add-comment-for-swap_vma_readahead.patch
mm-swap-use-mem_cgroup_is_root-instead-of-deferencing-css-parent.patch
mm-vmscan-do-not-iterate-all-mem-cgroups-for-global-direct-reclaim.patch
mm-ksm-do-not-block-on-page-lock-when-searching-stable-tree.patch
