The patch titled
Subject: mm: account pglazyfreed exactly
has been added to the -mm tree. Its filename is
mm-support-madvisemadv_free-fix-2.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mm-support-madvisemadv_free-fix-2.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mm-support-madvisemadv_free-fix-2.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/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Minchan Kim <minchan@xxxxxxxxxx>
Subject: mm: account pglazyfreed exactly
If anon pages are zapped by unmapping between page_mapped check and
try_to_unmap in shrink_page_list, they could be !PG_dirty although thre
are not MADV_FREEed pages so that VM accoutns it as pglazyfreed wrongly.
To fix, this patch counts the number of lazyfree ptes in try_to_unmap_one
and try_to_unmap returns SWAP_LZFREE only if the count is not zero, page
is !PG_dirty and SWAP_SUCCESS.
Signed-off-by: Minchan Kim <minchan@xxxxxxxxxx>
Cc: Hugh Dickins <hughd@xxxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Rik van Riel <riel@xxxxxxxxxx>
Cc: Jason Evans <je@xxxxxx>
Cc: Daniel Micay <danielmicay@xxxxxxxxx>
Cc: "Kirill A. Shutemov" <kirill@xxxxxxxxxxxxx>
Cc: Shaohua Li <shli@xxxxxxxxxx>
Cc: Michal Hocko <mhocko@xxxxxxx>
Cc: <yalin.wang2010@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/rmap.h | 1 +
mm/rmap.c | 29 +++++++++++++++++++++++++----
mm/vmscan.c | 8 ++++++--
3 files changed, 32 insertions(+), 6 deletions(-)
diff -puN include/linux/rmap.h~mm-support-madvisemadv_free-fix-2 include/linux/rmap.h
--- a/include/linux/rmap.h~mm-support-madvisemadv_free-fix-2
+++ a/include/linux/rmap.h
@@ -312,5 +312,6 @@ static inline int page_mkclean(struct pa
#define SWAP_AGAIN 1
#define SWAP_FAIL 2
#define SWAP_MLOCK 3
+#define SWAP_LZFREE 4
#endif /* _LINUX_RMAP_H */
diff -puN mm/rmap.c~mm-support-madvisemadv_free-fix-2 mm/rmap.c
--- a/mm/rmap.c~mm-support-madvisemadv_free-fix-2
+++ a/mm/rmap.c
@@ -1408,6 +1408,11 @@ void page_remove_rmap(struct page *page,
*/
}
+struct rmap_private {
+ enum ttu_flags flags;
+ int lazyfreed;
+};
+
/*
* @arg: enum ttu_flags will be passed to this argument
*/
@@ -1419,7 +1424,8 @@ static int try_to_unmap_one(struct page
pte_t pteval;
spinlock_t *ptl;
int ret = SWAP_AGAIN;
- enum ttu_flags flags = (enum ttu_flags)arg;
+ struct rmap_private *rp = arg;
+ enum ttu_flags flags = rp->flags;
/* munlock has nothing to gain from examining un-locked vmas */
if ((flags & TTU_MUNLOCK) && !(vma->vm_flags & VM_LOCKED))
@@ -1515,6 +1521,7 @@ static int try_to_unmap_one(struct page
if (!PageDirty(page) && (flags & TTU_LZFREE)) {
/* It's a freeable page by MADV_FREE */
dec_mm_counter(mm, MM_ANONPAGES);
+ rp->lazyfreed++;
goto discard;
}
@@ -1591,9 +1598,14 @@ static int page_not_mapped(struct page *
int try_to_unmap(struct page *page, enum ttu_flags flags)
{
int ret;
+ struct rmap_private rp = {
+ .flags = flags,
+ .lazyfreed = 0,
+ };
+
struct rmap_walk_control rwc = {
.rmap_one = try_to_unmap_one,
- .arg = (void *)flags,
+ .arg = (void *)&rp,
.done = page_not_mapped,
.anon_lock = page_lock_anon_vma_read,
};
@@ -1613,8 +1625,11 @@ int try_to_unmap(struct page *page, enum
ret = rmap_walk(page, &rwc);
- if (ret != SWAP_MLOCK && !page_mapped(page))
+ if (ret != SWAP_MLOCK && !page_mapped(page)) {
ret = SWAP_SUCCESS;
+ if (rp.lazyfreed && !PageDirty(page))
+ ret = SWAP_LZFREE;
+ }
return ret;
}
@@ -1636,9 +1651,15 @@ int try_to_unmap(struct page *page, enum
int try_to_munlock(struct page *page)
{
int ret;
+
+ struct rmap_private rp = {
+ .flags = TTU_MUNLOCK,
+ .lazyfreed = 0,
+ };
+
struct rmap_walk_control rwc = {
.rmap_one = try_to_unmap_one,
- .arg = (void *)TTU_MUNLOCK,
+ .arg = (void *)&rp,
.done = page_not_mapped,
.anon_lock = page_lock_anon_vma_read,
diff -puN mm/vmscan.c~mm-support-madvisemadv_free-fix-2 mm/vmscan.c
--- a/mm/vmscan.c~mm-support-madvisemadv_free-fix-2
+++ a/mm/vmscan.c
@@ -909,6 +909,7 @@ static unsigned long shrink_page_list(st
enum page_references references = PAGEREF_RECLAIM_CLEAN;
bool dirty, writeback;
bool lazyfree = false;
+ int ret = SWAP_SUCCESS;
cond_resched();
@@ -1064,7 +1065,7 @@ static unsigned long shrink_page_list(st
* processes. Try to unmap it here.
*/
if (page_mapped(page) && mapping) {
- switch (try_to_unmap(page, lazyfree ?
+ switch (ret = try_to_unmap(page, lazyfree ?
(ttu_flags | TTU_BATCH_FLUSH | TTU_LZFREE) :
(ttu_flags | TTU_BATCH_FLUSH))) {
case SWAP_FAIL:
@@ -1073,6 +1074,8 @@ static unsigned long shrink_page_list(st
goto keep_locked;
case SWAP_MLOCK:
goto cull_mlocked;
+ case SWAP_LZFREE:
+ goto lazyfree;
case SWAP_SUCCESS:
; /* try to free the page below */
}
@@ -1179,6 +1182,7 @@ static unsigned long shrink_page_list(st
}
}
+lazyfree:
if (!mapping || !__remove_mapping(mapping, page, true))
goto keep_locked;
@@ -1191,7 +1195,7 @@ static unsigned long shrink_page_list(st
*/
__ClearPageLocked(page);
free_it:
- if (lazyfree && !PageDirty(page))
+ if (ret == SWAP_LZFREE)
count_vm_event(PGLAZYFREED);
nr_reclaimed++;
_
Patches currently in -mm which might be from minchan@xxxxxxxxxx are
zram-pass-gfp-from-zcomp-frontend-to-backend.patch
mm-support-madvisemadv_free.patch
mm-support-madvisemadv_free-fix.patch
mm-support-madvisemadv_free-fix-2.patch
mm-define-madv_free-for-some-arches.patch
mm-free-swp_entry-in-madvise_free.patch
mm-move-lazily-freed-pages-to-inactive-list.patch
mm-mark-stable-page-dirty-in-ksm.patch
x86-add-pmd_-for-thp.patch
sparc-add-pmd_-for-thp.patch
powerpc-add-pmd_-for-thp.patch
arm-add-pmd_mkclean-for-thp.patch
arm64-add-pmd_mkclean-for-thp.patch
mm-dont-split-thp-page-when-syscall-is-called.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
