The patch titled
Subject: mm/mempolicy.c: convert the shared_policy lock to a rwlock
has been added to the -mm tree. Its filename is
mempolicy-convert-the-shared_policy-lock-to-a-rwlock.patch
This patch should soon appear at
http://ozlabs.org/~akpm/mmots/broken-out/mempolicy-convert-the-shared_policy-lock-to-a-rwlock.patch
and later at
http://ozlabs.org/~akpm/mmotm/broken-out/mempolicy-convert-the-shared_policy-lock-to-a-rwlock.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: Nathan Zimmer <nzimmer@xxxxxxx>
Subject: 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%.
Signed-off-by: Nathan Zimmer <nzimmer@xxxxxxx>
Acked-by: David Rientjes <rientjes@xxxxxxxxxx>
Cc: Nadia Yvette Chambers <nyc@xxxxxxxxxxxxxx>
Cc: Naoya Horiguchi <n-horiguchi@xxxxxxxxxxxxx>
Cc: Mel Gorman <mgorman@xxxxxxx>
Cc: "Aneesh Kumar K.V" <aneesh.kumar@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/hugetlbfs/inode.c | 2 +-
include/linux/mempolicy.h | 2 +-
mm/mempolicy.c | 20 ++++++++++----------
3 files changed, 12 insertions(+), 12 deletions(-)
diff -puN fs/hugetlbfs/inode.c~mempolicy-convert-the-shared_policy-lock-to-a-rwlock fs/hugetlbfs/inode.c
--- a/fs/hugetlbfs/inode.c~mempolicy-convert-the-shared_policy-lock-to-a-rwlock
+++ a/fs/hugetlbfs/inode.c
@@ -739,7 +739,7 @@ static struct inode *hugetlbfs_get_inode
/*
* 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 -puN include/linux/mempolicy.h~mempolicy-convert-the-shared_policy-lock-to-a-rwlock include/linux/mempolicy.h
--- a/include/linux/mempolicy.h~mempolicy-convert-the-shared_policy-lock-to-a-rwlock
+++ a/include/linux/mempolicy.h
@@ -122,7 +122,7 @@ struct sp_node {
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 -puN mm/mempolicy.c~mempolicy-convert-the-shared_policy-lock-to-a-rwlock mm/mempolicy.c
--- a/mm/mempolicy.c~mempolicy-convert-the-shared_policy-lock-to-a-rwlock
+++ a/mm/mempolicy.c
@@ -2142,7 +2142,7 @@ bool __mpol_equal(struct mempolicy *a, s
*
* 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.
*/
@@ -2179,7 +2179,7 @@ sp_lookup(struct shared_policy *sp, unsi
}
/* Insert a new shared policy into the list. */
-/* Caller holds sp->lock */
+/* Caller holds the write of sp->lock */
static void sp_insert(struct shared_policy *sp, struct sp_node *new)
{
struct rb_node **p = &sp->root.rb_node;
@@ -2211,13 +2211,13 @@ mpol_shared_policy_lookup(struct shared_
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 +2360,7 @@ static int shared_policy_replace(struct
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 +2393,7 @@ restart:
}
if (new)
sp_insert(sp, new);
- spin_unlock(&sp->lock);
+ write_unlock(&sp->lock);
ret = 0;
err_out:
@@ -2405,7 +2405,7 @@ err_out:
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 +2431,7 @@ void mpol_shared_policy_init(struct shar
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 +2497,14 @@ void mpol_free_shared_policy(struct shar
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
_
Patches currently in -mm which might be from nzimmer@xxxxxxx are
mempolicy-convert-the-shared_policy-lock-to-a-rwlock.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
