All of lore.kernel.org
 help / color / mirror / Atom feed
From: Al Viro <viro@ZenIV.linux.org.uk>
To: linux-fsdevel@vger.kernel.org
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
	linux-kernel@vger.kernel.org, Stephen Tweedie <sct@redhat.com>,
	Jeremy Eder <jeder@redhat.com>
Subject: [PATCH 5/5][RFC][CFT] resizable namespace.c hashes
Date: Wed, 5 Mar 2014 03:52:19 +0000	[thread overview]
Message-ID: <20140305035219.GE26528@ZenIV.linux.org.uk> (raw)
In-Reply-To: <20140305034751.GW18016@ZenIV.linux.org.uk>

* switch allocation to alloc_large_system_hash()
* make sizes overridable by boot parameters (mhash_entries=, mphash_entries=)
* switch mountpoint_hashtable from list_head to hlist_head

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/mount.h     |    2 +-
 fs/namespace.c |   81 ++++++++++++++++++++++++++++++++++++++++----------------
 2 files changed, 59 insertions(+), 24 deletions(-)

diff --git a/fs/mount.h b/fs/mount.h
index 61e1c05..46790ae1 100644
--- a/fs/mount.h
+++ b/fs/mount.h
@@ -19,7 +19,7 @@ struct mnt_pcp {
 };
 
 struct mountpoint {
-	struct list_head m_hash;
+	struct hlist_node m_hash;
 	struct dentry *m_dentry;
 	int m_count;
 };
diff --git a/fs/namespace.c b/fs/namespace.c
index 00f12af..6be899f 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -23,11 +23,34 @@
 #include <linux/uaccess.h>
 #include <linux/proc_ns.h>
 #include <linux/magic.h>
+#include <linux/bootmem.h>
 #include "pnode.h"
 #include "internal.h"
 
-#define HASH_SHIFT ilog2(PAGE_SIZE / sizeof(struct list_head))
-#define HASH_SIZE (1UL << HASH_SHIFT)
+static unsigned int m_hash_mask __read_mostly;
+static unsigned int m_hash_shift __read_mostly;
+static unsigned int mp_hash_mask __read_mostly;
+static unsigned int mp_hash_shift __read_mostly;
+
+static __initdata unsigned long mhash_entries;
+static int __init set_mhash_entries(char *str)
+{
+	if (!str)
+		return 0;
+	mhash_entries = simple_strtoul(str, &str, 0);
+	return 1;
+}
+__setup("mhash_entries=", set_mhash_entries);
+
+static __initdata unsigned long mphash_entries;
+static int __init set_mphash_entries(char *str)
+{
+	if (!str)
+		return 0;
+	mphash_entries = simple_strtoul(str, &str, 0);
+	return 1;
+}
+__setup("mphash_entries=", set_mphash_entries);
 
 static u64 event;
 static DEFINE_IDA(mnt_id_ida);
@@ -37,7 +60,7 @@ static int mnt_id_start = 0;
 static int mnt_group_start = 1;
 
 static struct list_head *mount_hashtable __read_mostly;
-static struct list_head *mountpoint_hashtable __read_mostly;
+static struct hlist_head *mountpoint_hashtable __read_mostly;
 static struct kmem_cache *mnt_cache __read_mostly;
 static DECLARE_RWSEM(namespace_sem);
 
@@ -55,12 +78,19 @@ EXPORT_SYMBOL_GPL(fs_kobj);
  */
 __cacheline_aligned_in_smp DEFINE_SEQLOCK(mount_lock);
 
-static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry)
+static inline struct list_head *m_hash(struct vfsmount *mnt, struct dentry *dentry)
 {
 	unsigned long tmp = ((unsigned long)mnt / L1_CACHE_BYTES);
 	tmp += ((unsigned long)dentry / L1_CACHE_BYTES);
-	tmp = tmp + (tmp >> HASH_SHIFT);
-	return tmp & (HASH_SIZE - 1);
+	tmp = tmp + (tmp >> m_hash_shift);
+	return &mount_hashtable[tmp & m_hash_mask];
+}
+
+static inline struct hlist_head *mp_hash(struct dentry *dentry)
+{
+	unsigned long tmp = ((unsigned long)dentry / L1_CACHE_BYTES);
+	tmp = tmp + (tmp >> mp_hash_shift);
+	return &mountpoint_hashtable[tmp & mp_hash_mask];
 }
 
 /*
@@ -575,7 +605,7 @@ bool legitimize_mnt(struct vfsmount *bastard, unsigned seq)
  */
 struct mount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry)
 {
-	struct list_head *head = mount_hashtable + hash(mnt, dentry);
+	struct list_head *head = m_hash(mnt, dentry);
 	struct mount *p;
 
 	list_for_each_entry_rcu(p, head, mnt_hash)
@@ -590,7 +620,7 @@ struct mount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry)
  */
 struct mount *__lookup_mnt_last(struct vfsmount *mnt, struct dentry *dentry)
 {
-	struct list_head *head = mount_hashtable + hash(mnt, dentry);
+	struct list_head *head = m_hash(mnt, dentry);
 	struct mount *p;
 
 	list_for_each_entry_reverse(p, head, mnt_hash)
@@ -633,11 +663,11 @@ struct vfsmount *lookup_mnt(struct path *path)
 
 static struct mountpoint *new_mountpoint(struct dentry *dentry)
 {
-	struct list_head *chain = mountpoint_hashtable + hash(NULL, dentry);
+	struct hlist_head *chain = mp_hash(dentry);
 	struct mountpoint *mp;
 	int ret;
 
-	list_for_each_entry(mp, chain, m_hash) {
+	hlist_for_each_entry(mp, chain, m_hash) {
 		if (mp->m_dentry == dentry) {
 			/* might be worth a WARN_ON() */
 			if (d_unlinked(dentry))
@@ -659,7 +689,7 @@ static struct mountpoint *new_mountpoint(struct dentry *dentry)
 
 	mp->m_dentry = dentry;
 	mp->m_count = 1;
-	list_add(&mp->m_hash, chain);
+	hlist_add_head(&mp->m_hash, chain);
 	return mp;
 }
 
@@ -670,7 +700,7 @@ static void put_mountpoint(struct mountpoint *mp)
 		spin_lock(&dentry->d_lock);
 		dentry->d_flags &= ~DCACHE_MOUNTED;
 		spin_unlock(&dentry->d_lock);
-		list_del(&mp->m_hash);
+		hlist_del(&mp->m_hash);
 		kfree(mp);
 	}
 }
@@ -739,8 +769,7 @@ static void attach_mnt(struct mount *mnt,
 			struct mountpoint *mp)
 {
 	mnt_set_mountpoint(parent, mp, mnt);
-	list_add_tail(&mnt->mnt_hash, mount_hashtable +
-			hash(&parent->mnt, mp->m_dentry));
+	list_add_tail(&mnt->mnt_hash, m_hash(&parent->mnt, mp->m_dentry));
 	list_add_tail(&mnt->mnt_child, &parent->mnt_mounts);
 }
 
@@ -762,8 +791,8 @@ static void commit_tree(struct mount *mnt)
 
 	list_splice(&head, n->list.prev);
 
-	list_add_tail(&mnt->mnt_hash, mount_hashtable +
-				hash(&parent->mnt, mnt->mnt_mountpoint));
+	list_add_tail(&mnt->mnt_hash,
+				m_hash(&parent->mnt, mnt->mnt_mountpoint));
 	list_add_tail(&mnt->mnt_child, &parent->mnt_mounts);
 	touch_mnt_namespace(n);
 }
@@ -2794,18 +2823,24 @@ void __init mnt_init(void)
 	mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct mount),
 			0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL);
 
-	mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC);
-	mountpoint_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC);
+	mount_hashtable = alloc_large_system_hash("Mount-cache",
+				sizeof(struct list_head),
+				mhash_entries, 19,
+				0,
+				&m_hash_shift, &m_hash_mask, 0, 0);
+	mountpoint_hashtable = alloc_large_system_hash("Mountpoint-cache",
+				sizeof(struct hlist_head),
+				mphash_entries, 19,
+				0,
+				&mp_hash_shift, &mp_hash_mask, 0, 0);
 
 	if (!mount_hashtable || !mountpoint_hashtable)
 		panic("Failed to allocate mount hash table\n");
 
-	printk(KERN_INFO "Mount-cache hash table entries: %lu\n", HASH_SIZE);
-
-	for (u = 0; u < HASH_SIZE; u++)
+	for (u = 0; u <= m_hash_mask; u++)
 		INIT_LIST_HEAD(&mount_hashtable[u]);
-	for (u = 0; u < HASH_SIZE; u++)
-		INIT_LIST_HEAD(&mountpoint_hashtable[u]);
+	for (u = 0; u <= mp_hash_mask; u++)
+		INIT_HLIST_HEAD(&mountpoint_hashtable[u]);
 
 	kernfs_init();
 
-- 
1.7.10.4

  parent reply	other threads:[~2014-03-05  3:52 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-03-05  3:47 [PATCHES][RFC][CFT] scalability fixes for shitloads of mounts Al Viro
2014-03-05  3:49 ` [PATCH 1/5][RFC][CFT] percpu fixes, part 1 Al Viro
2014-03-06 19:20   ` Tejun Heo
2014-03-06 20:30     ` Al Viro
2014-03-06 20:47       ` Tejun Heo
2014-03-07  2:52         ` Al Viro
2014-03-07 12:30           ` Tejun Heo
2014-03-14 18:45           ` Al Viro
2014-03-14 18:47             ` Tejun Heo
2014-03-14 18:53               ` Al Viro
2014-03-17 20:12                 ` [PATCH percpu/for-3.15] percpu: allocation size should be even Tejun Heo
2014-03-05  3:50 ` [PATCH 2/5][RFC][CFT] fold pcpu_split_block() into the only caller Al Viro
2014-03-06 19:21   ` Tejun Heo
2014-03-05  3:51 ` [PATCH 3/5][RFC][CFT] smarter propagate_mnt() Al Viro
2014-03-05  3:51 ` [PATCH 4/5][RFC][CFT] reduce m_start() cost Al Viro
2014-03-05  3:52 ` Al Viro [this message]
2014-03-07 17:17   ` [PATCH 5/5][RFC][CFT] resizable namespace.c hashes Andi Kleen
2014-03-07 18:38     ` Al Viro

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20140305035219.GE26528@ZenIV.linux.org.uk \
    --to=viro@zeniv.linux.org.uk \
    --cc=jeder@redhat.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sct@redhat.com \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.