mm/mempolicy.c: convert the shared_policy lock to a rwlock
When running the SPECint_rate gcc on some very large boxes it was
noticed that the system was spending lots of time in
mpol_shared_policy_lookup(). The gamess benchmark can also show it and
is what I mostly used to chase down the issue since the setup for that I
found to be easier.
To be clear the binaries were on tmpfs because of disk I/O requirements.
We then used text replication to avoid icache misses and having all the
copies banging on the memory where the instruction code resides. This
results in us hitting a bottleneck in mpol_shared_policy_lookup() since
lookup is serialised by the shared_policy lock.
I have only reproduced this on very large (3k+ cores) boxes. The
problem starts showing up at just a few hundred ranks getting worse
until it threatens to livelock once it gets large enough. For example
on the gamess benchmark at 128 ranks this area consumes only ~1% of
time, at 512 ranks it consumes nearly 13%, and at 2k ranks it is over
90%.
To alleviate the contention in this area I converted the spinlock to an
rwlock. This allows a large number of lookups to happen simultaneously.
The results were quite good reducing this consumtion at max ranks to
around 2%.
[akpm@linux-foundation.org: tidy up code comments]
Signed-off-by: Nathan Zimmer <nzimmer@sgi.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Nadia Yvette Chambers <nyc@holomorphy.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index f6820ec..a1cb8fd 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -738,7 +738,7 @@
/*
* The policy is initialized here even if we are creating a
* private inode because initialization simply creates an
- * an empty rb tree and calls spin_lock_init(), later when we
+ * an empty rb tree and calls rwlock_init(), later when we
* call mpol_free_shared_policy() it will just return because
* the rb tree will still be empty.
*/
diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h
index 3d385c8..2696c1f 100644
--- a/include/linux/mempolicy.h
+++ b/include/linux/mempolicy.h
@@ -122,7 +122,7 @@
struct shared_policy {
struct rb_root root;
- spinlock_t lock;
+ rwlock_t lock;
};
int vma_dup_policy(struct vm_area_struct *src, struct vm_area_struct *dst);
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 87a1779..d8caff0 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -2142,12 +2142,14 @@
*
* Remember policies even when nobody has shared memory mapped.
* The policies are kept in Red-Black tree linked from the inode.
- * They are protected by the sp->lock spinlock, which should be held
+ * They are protected by the sp->lock rwlock, which should be held
* for any accesses to the tree.
*/
-/* lookup first element intersecting start-end */
-/* Caller holds sp->lock */
+/*
+ * lookup first element intersecting start-end. Caller holds sp->lock for
+ * reading or for writing
+ */
static struct sp_node *
sp_lookup(struct shared_policy *sp, unsigned long start, unsigned long end)
{
@@ -2178,8 +2180,10 @@
return rb_entry(n, struct sp_node, nd);
}
-/* Insert a new shared policy into the list. */
-/* Caller holds sp->lock */
+/*
+ * Insert a new shared policy into the list. Caller holds sp->lock for
+ * writing.
+ */
static void sp_insert(struct shared_policy *sp, struct sp_node *new)
{
struct rb_node **p = &sp->root.rb_node;
@@ -2211,13 +2215,13 @@
if (!sp->root.rb_node)
return NULL;
- spin_lock(&sp->lock);
+ read_lock(&sp->lock);
sn = sp_lookup(sp, idx, idx+1);
if (sn) {
mpol_get(sn->policy);
pol = sn->policy;
}
- spin_unlock(&sp->lock);
+ read_unlock(&sp->lock);
return pol;
}
@@ -2360,7 +2364,7 @@
int ret = 0;
restart:
- spin_lock(&sp->lock);
+ write_lock(&sp->lock);
n = sp_lookup(sp, start, end);
/* Take care of old policies in the same range. */
while (n && n->start < end) {
@@ -2393,7 +2397,7 @@
}
if (new)
sp_insert(sp, new);
- spin_unlock(&sp->lock);
+ write_unlock(&sp->lock);
ret = 0;
err_out:
@@ -2405,7 +2409,7 @@
return ret;
alloc_new:
- spin_unlock(&sp->lock);
+ write_unlock(&sp->lock);
ret = -ENOMEM;
n_new = kmem_cache_alloc(sn_cache, GFP_KERNEL);
if (!n_new)
@@ -2431,7 +2435,7 @@
int ret;
sp->root = RB_ROOT; /* empty tree == default mempolicy */
- spin_lock_init(&sp->lock);
+ rwlock_init(&sp->lock);
if (mpol) {
struct vm_area_struct pvma;
@@ -2497,14 +2501,14 @@
if (!p->root.rb_node)
return;
- spin_lock(&p->lock);
+ write_lock(&p->lock);
next = rb_first(&p->root);
while (next) {
n = rb_entry(next, struct sp_node, nd);
next = rb_next(&n->nd);
sp_delete(p, n);
}
- spin_unlock(&p->lock);
+ write_unlock(&p->lock);
}
#ifdef CONFIG_NUMA_BALANCING