[Devel] [PATCH RHEL7 COMMIT] ms/mm/rmap: extend rmap_walk_xxx() to cope with different cases
Konstantin Khorenko
khorenko at virtuozzo.com
Tue Dec 8 06:15:56 PST 2015
The commit is pushed to "branch-rh7-3.10.0-229.7.2.vz7.9.x-ovz" and will appear at https://src.openvz.org/scm/ovz/vzkernel.git
after rh7-3.10.0-229.7.2.vz7.9.15
------>
commit b0b85ef542d082b5eaf67e6c666fb2e56dd3b099
Author: Vladimir Davydov <vdavydov at virtuozzo.com>
Date: Tue Dec 8 18:15:56 2015 +0400
ms/mm/rmap: extend rmap_walk_xxx() to cope with different cases
Patchset description:
rmap_walk() present in RH7 requires the caller to either hold mmap_sem
or pin the page's anon_vma. page_idle_clear_pte_refs does neither. As a
result, it might end up trying to lock/unlock anon_vma which has already
been freed and possibly reallocated. This won't do any good.
Let's pull the new version of rmap_walk() from upstream, which allows to
specify a custom anon_vma lock function and use it in page_idle code to
avoid this issue. This patch puts page_idle in sync with upstream.
I hope this will fix:
https://jira.sw.ru/browse/PSBM-42015
Joonsoo Kim (3):
mm/rmap: factor lock function out of rmap_walk_anon()
mm/rmap: make rmap_walk to get the rmap_walk_control argument
mm/rmap: extend rmap_walk_xxx() to cope with different cases
Vladimir Davydov (1):
mm: page_idle: look up page anon_vma carefully when checking references
============================
This patch description:
From: Joonsoo Kim <iamjoonsoo.kim at lge.com>
There are a lot of common parts in traversing functions, but there are
also a little of uncommon parts in it. By assigning proper function
pointer on each rmap_walker_control, we can handle these difference
correctly.
Following are differences we should handle.
1. difference of lock function in anon mapping case
2. nonlinear handling in file mapping case
3. prechecked condition:
checking memcg in page_referenced(),
checking VM_SHARE in page_mkclean()
checking temporary vma in try_to_unmap()
4. exit condition:
checking page_mapped() in try_to_unmap()
So, in this patch, I introduce 4 function pointers to handle above
differences.
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim at lge.com>
Cc: Naoya Horiguchi <n-horiguchi at ah.jp.nec.com>
Cc: Mel Gorman <mgorman at suse.de>
Cc: Hugh Dickins <hughd at google.com>
Cc: Rik van Riel <riel at redhat.com>
Cc: Ingo Molnar <mingo at kernel.org>
Cc: Hillf Danton <dhillf at gmail.com>
Signed-off-by: Andrew Morton <akpm at linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds at linux-foundation.org>
(cherry picked from commit 0dd1c7bbce8d1d142bb25aefaa50262dfd77cb78)
Signed-off-by: Vladimir Davydov <vdavydov at virtuozzo.com>
---
include/linux/rmap.h | 15 +++++++++++++++
mm/ksm.c | 7 +++++++
mm/rmap.c | 37 +++++++++++++++++++++++++++++--------
3 files changed, 51 insertions(+), 8 deletions(-)
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 083b1df..2491eae 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -237,10 +237,25 @@ extern struct anon_vma *page_lock_anon_vma_read(struct page *page);
extern void page_unlock_anon_vma_read(struct anon_vma *anon_vma);
int page_mapped_in_vma(struct page *page, struct vm_area_struct *vma);
+/*
+ * rmap_walk_control: To control rmap traversing for specific needs
+ *
+ * arg: passed to rmap_one() and invalid_vma()
+ * rmap_one: executed on each vma where page is mapped
+ * done: for checking traversing termination condition
+ * file_nonlinear: for handling file nonlinear mapping
+ * anon_lock: for getting anon_lock by optimized way rather than default
+ * invalid_vma: for skipping uninterested vma
+ */
struct rmap_walk_control {
void *arg;
int (*rmap_one)(struct page *page, struct vm_area_struct *vma,
unsigned long addr, void *arg);
+ int (*done)(struct page *page);
+ int (*file_nonlinear)(struct page *, struct address_space *,
+ struct vm_area_struct *vma);
+ struct anon_vma *(*anon_lock)(struct page *page);
+ bool (*invalid_vma)(struct vm_area_struct *vma, void *arg);
};
/*
diff --git a/mm/ksm.c b/mm/ksm.c
index ff2d5a3..e6bedcc 100644
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -2038,12 +2038,19 @@ again:
if ((rmap_item->mm == vma->vm_mm) == search_new_forks)
continue;
+ if (rwc->invalid_vma && rwc->invalid_vma(vma, rwc->arg))
+ continue;
+
ret = rwc->rmap_one(page, vma,
rmap_item->address, rwc->arg);
if (ret != SWAP_AGAIN) {
anon_vma_unlock_read(anon_vma);
goto out;
}
+ if (rwc->done && rwc->done(page)) {
+ anon_vma_unlock_read(anon_vma);
+ goto out;
+ }
}
anon_vma_unlock_read(anon_vma);
}
diff --git a/mm/rmap.c b/mm/rmap.c
index 362c5e8..3eda1ed 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1749,10 +1749,14 @@ void __put_anon_vma(struct anon_vma *anon_vma)
}
#ifdef CONFIG_MIGRATION
-static struct anon_vma *rmap_walk_anon_lock(struct page *page)
+static struct anon_vma *rmap_walk_anon_lock(struct page *page,
+ struct rmap_walk_control *rwc)
{
struct anon_vma *anon_vma;
+ if (rwc->anon_lock)
+ return rwc->anon_lock(page);
+
/*
* Note: remove_migration_ptes() cannot use page_lock_anon_vma_read()
* because that depends on page_mapped(); but not all its usages
@@ -1778,16 +1782,22 @@ static int rmap_walk_anon(struct page *page, struct rmap_walk_control *rwc)
struct anon_vma_chain *avc;
int ret = SWAP_AGAIN;
- anon_vma = rmap_walk_anon_lock(page);
+ anon_vma = rmap_walk_anon_lock(page, rwc);
if (!anon_vma)
return ret;
anon_vma_interval_tree_foreach(avc, &anon_vma->rb_root, pgoff, pgoff) {
struct vm_area_struct *vma = avc->vma;
unsigned long address = vma_address(page, vma);
+
+ if (rwc->invalid_vma && rwc->invalid_vma(vma, rwc->arg))
+ continue;
+
ret = rwc->rmap_one(page, vma, address, rwc->arg);
if (ret != SWAP_AGAIN)
break;
+ if (rwc->done && rwc->done(page))
+ break;
}
anon_vma_unlock_read(anon_vma);
return ret;
@@ -1805,15 +1815,26 @@ static int rmap_walk_file(struct page *page, struct rmap_walk_control *rwc)
mutex_lock(&mapping->i_mmap_mutex);
vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
unsigned long address = vma_address(page, vma);
+
+ if (rwc->invalid_vma && rwc->invalid_vma(vma, rwc->arg))
+ continue;
+
ret = rwc->rmap_one(page, vma, address, rwc->arg);
if (ret != SWAP_AGAIN)
- break;
+ goto done;
+ if (rwc->done && rwc->done(page))
+ goto done;
}
- /*
- * No nonlinear handling: being always shared, nonlinear vmas
- * never contain migration ptes. Decide what to do about this
- * limitation to linear when we need rmap_walk() on nonlinear.
- */
+
+ if (!rwc->file_nonlinear)
+ goto done;
+
+ if (list_empty(&mapping->i_mmap_nonlinear))
+ goto done;
+
+ ret = rwc->file_nonlinear(page, mapping, vma);
+
+done:
mutex_unlock(&mapping->i_mmap_mutex);
return ret;
}
More information about the Devel
mailing list