The patch titled
mm: always lock the root (oldest) anon_vma
has been added to the -mm tree. Its filename is
mm-always-lock-the-root-oldest-anon_vma.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 ***
See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find
out what to do about this
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: mm: always lock the root (oldest) anon_vma
From: Rik van Riel <riel@xxxxxxxxxx>
Always (and only) lock the root (oldest) anon_vma whenever we do something
in an anon_vma. The recently introduced anon_vma scalability is due to
the rmap code scanning only the VMAs that need to be scanned. Many common
operations still took the anon_vma lock on the root anon_vma, so always
taking that lock is not expected to introduce any scalability issues.
However, always taking the same lock does mean we only need to take one
lock, which means rmap_walk on pages from any anon_vma in the vma is
excluded from occurring during an munmap, expand_stack or other operation
that needs to exclude rmap_walk and similar functions.
Also add the proper locking to vma_adjust.
Signed-off-by: Rik van Riel <riel@xxxxxxxxxx>
Tested-by: Larry Woodman <lwoodman@xxxxxxxxxx>
Acked-by: Larry Woodman <lwoodman@xxxxxxxxxx>
Reviewed-by: Minchan Kim <minchan.kim@xxxxxxxxx>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Acked-by: Mel Gorman <mel@xxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/rmap.h | 8 ++++----
mm/ksm.c | 2 +-
mm/migrate.c | 2 +-
mm/mmap.c | 30 ++++++++++++++++++++++--------
4 files changed, 28 insertions(+), 14 deletions(-)
diff -puN include/linux/rmap.h~mm-always-lock-the-root-oldest-anon_vma include/linux/rmap.h
--- a/include/linux/rmap.h~mm-always-lock-the-root-oldest-anon_vma
+++ a/include/linux/rmap.h
@@ -104,24 +104,24 @@ static inline void vma_lock_anon_vma(str
{
struct anon_vma *anon_vma = vma->anon_vma;
if (anon_vma)
- spin_lock(&anon_vma->lock);
+ spin_lock(&anon_vma->root->lock);
}
static inline void vma_unlock_anon_vma(struct vm_area_struct *vma)
{
struct anon_vma *anon_vma = vma->anon_vma;
if (anon_vma)
- spin_unlock(&anon_vma->lock);
+ spin_unlock(&anon_vma->root->lock);
}
static inline void anon_vma_lock(struct anon_vma *anon_vma)
{
- spin_lock(&anon_vma->lock);
+ spin_lock(&anon_vma->root->lock);
}
static inline void anon_vma_unlock(struct anon_vma *anon_vma)
{
- spin_unlock(&anon_vma->lock);
+ spin_unlock(&anon_vma->root->lock);
}
/*
diff -puN mm/ksm.c~mm-always-lock-the-root-oldest-anon_vma mm/ksm.c
--- a/mm/ksm.c~mm-always-lock-the-root-oldest-anon_vma
+++ a/mm/ksm.c
@@ -325,7 +325,7 @@ static void drop_anon_vma(struct rmap_it
{
struct anon_vma *anon_vma = rmap_item->anon_vma;
- if (atomic_dec_and_lock(&anon_vma->external_refcount, &anon_vma->lock)) {
+ if (atomic_dec_and_lock(&anon_vma->external_refcount, &anon_vma->root->lock)) {
int empty = list_empty(&anon_vma->head);
anon_vma_unlock(anon_vma);
if (empty)
diff -puN mm/migrate.c~mm-always-lock-the-root-oldest-anon_vma mm/migrate.c
--- a/mm/migrate.c~mm-always-lock-the-root-oldest-anon_vma
+++ a/mm/migrate.c
@@ -682,7 +682,7 @@ skip_unmap:
rcu_unlock:
/* Drop an anon_vma reference if we took one */
- if (anon_vma && atomic_dec_and_lock(&anon_vma->external_refcount, &anon_vma->lock)) {
+ if (anon_vma && atomic_dec_and_lock(&anon_vma->external_refcount, &anon_vma->root->lock)) {
int empty = list_empty(&anon_vma->head);
anon_vma_unlock(anon_vma);
if (empty)
diff -puN mm/mmap.c~mm-always-lock-the-root-oldest-anon_vma mm/mmap.c
--- a/mm/mmap.c~mm-always-lock-the-root-oldest-anon_vma
+++ a/mm/mmap.c
@@ -506,6 +506,7 @@ int vma_adjust(struct vm_area_struct *vm
struct vm_area_struct *importer = NULL;
struct address_space *mapping = NULL;
struct prio_tree_root *root = NULL;
+ struct anon_vma *anon_vma = NULL;
struct file *file = vma->vm_file;
long adjust_next = 0;
int remove_next = 0;
@@ -578,6 +579,17 @@ again: remove_next = 1 + (end > next->
}
}
+ /*
+ * When changing only vma->vm_end, we don't really need anon_vma
+ * lock. This is a fairly rare case by itself, but the anon_vma
+ * lock may be shared between many sibling processes. Skipping
+ * the lock for brk adjustments makes a difference sometimes.
+ */
+ if (vma->anon_vma && (insert || importer || start != vma->vm_start)) {
+ anon_vma = vma->anon_vma;
+ anon_vma_lock(anon_vma);
+ }
+
if (root) {
flush_dcache_mmap_lock(mapping);
vma_prio_tree_remove(vma, root);
@@ -617,6 +629,8 @@ again: remove_next = 1 + (end > next->
__insert_vm_struct(mm, insert);
}
+ if (anon_vma)
+ anon_vma_unlock(anon_vma);
if (mapping)
spin_unlock(&mapping->i_mmap_lock);
@@ -2466,23 +2480,23 @@ static DEFINE_MUTEX(mm_all_locks_mutex);
static void vm_lock_anon_vma(struct mm_struct *mm, struct anon_vma *anon_vma)
{
- if (!test_bit(0, (unsigned long *) &anon_vma->head.next)) {
+ if (!test_bit(0, (unsigned long *) &anon_vma->root->head.next)) {
/*
* The LSB of head.next can't change from under us
* because we hold the mm_all_locks_mutex.
*/
- spin_lock_nest_lock(&anon_vma->lock, &mm->mmap_sem);
+ spin_lock_nest_lock(&anon_vma->root->lock, &mm->mmap_sem);
/*
* We can safely modify head.next after taking the
- * anon_vma->lock. If some other vma in this mm shares
+ * anon_vma->root->lock. If some other vma in this mm shares
* the same anon_vma we won't take it again.
*
* No need of atomic instructions here, head.next
* can't change from under us thanks to the
- * anon_vma->lock.
+ * anon_vma->root->lock.
*/
if (__test_and_set_bit(0, (unsigned long *)
- &anon_vma->head.next))
+ &anon_vma->root->head.next))
BUG();
}
}
@@ -2573,7 +2587,7 @@ out_unlock:
static void vm_unlock_anon_vma(struct anon_vma *anon_vma)
{
- if (test_bit(0, (unsigned long *) &anon_vma->head.next)) {
+ if (test_bit(0, (unsigned long *) &anon_vma->root->head.next)) {
/*
* The LSB of head.next can't change to 0 from under
* us because we hold the mm_all_locks_mutex.
@@ -2584,10 +2598,10 @@ static void vm_unlock_anon_vma(struct an
*
* No need of atomic instructions here, head.next
* can't change from under us until we release the
- * anon_vma->lock.
+ * anon_vma->root->lock.
*/
if (!__test_and_clear_bit(0, (unsigned long *)
- &anon_vma->head.next))
+ &anon_vma->root->head.next))
BUG();
anon_vma_unlock(anon_vma);
}
_
Patches currently in -mm which might be from riel@xxxxxxxxxx are
mm-revalidate-anon_vma-in-page_lock_anon_vma.patch
mm-rename-anon_vma_lock-to-vma_lock_anon_vma.patch
mm-change-direct-call-of-spin_lockanon_vma-lock-to-inline-function.patch
mm-track-the-root-oldest-anon_vma.patch
mm-track-the-root-oldest-anon_vma-fix.patch
mm-always-lock-the-root-oldest-anon_vma.patch
mm-extend-ksm-refcounts-to-the-anon_vma-root.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
