linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Glauber Costa <glommer@openvz.org>
To: linux-mm@kvack.org
Cc: Andrew Morton <akpm@linux-foundation.org>,
	Mel Gorman <mgorman@suse.de>,
	cgroups@vger.kernel.org, kamezawa.hiroyu@jp.fujitsu.com,
	Johannes Weiner <hannes@cmpxchg.org>,
	Michal Hocko <mhocko@suse.cz>,
	hughd@google.com, Greg Thelen <gthelen@google.com>,
	linux-fsdevel@vger.kernel.org, Dave Chinner <dchinner@redhat.com>,
	Glauber Costa <glommer@openvz.org>
Subject: [PATCH v5 05/31] dcache: remove dentries from LRU before putting on dispose list
Date: Thu,  9 May 2013 10:06:22 +0400	[thread overview]
Message-ID: <1368079608-5611-6-git-send-email-glommer@openvz.org> (raw)
In-Reply-To: <1368079608-5611-1-git-send-email-glommer@openvz.org>

From: Dave Chinner <dchinner@redhat.com>

One of the big problems with modifying the way the dcache shrinker
and LRU implementation works is that the LRU is abused in several
ways. One of these is shrink_dentry_list().

Basically, we can move a dentry off the LRU onto a different list
without doing any accounting changes, and then use dentry_lru_prune()
to remove it from what-ever list it is now on to do the LRU
accounting at that point.

This makes it -really hard- to change the LRU implementation. The
use of the per-sb LRU lock serialises movement of the dentries
between the different lists and the removal of them, and this is the
only reason that it works. If we want to break up the dentry LRU
lock and lists into, say, per-node lists, we remove the only
serialisation that allows this lru list/dispose list abuse to work.

To make this work effectively, the dispose list has to be isolated
from the LRU list - dentries have to be removed from the LRU
*before* being placed on the dispose list. This means that the LRU
accounting and isolation is completed before disposal is started,
and that means we can change the LRU implementation freely in
future.

This means that dentries *must* be marked with DCACHE_SHRINK_LIST
when they are placed on the dispose list so that we don't think that
parent dentries found in try_prune_one_dentry() are on the LRU when
the are actually on the dispose list. This would result in
accounting the dentry to the LRU a second time. Hence
dentry_lru_prune() has to handle the DCACHE_SHRINK_LIST case
differently because the dentry isn't on the LRU list.

[ v2: don't decrement nr unused twice, spotted by Sha Zhengju ]
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Glauber Costa <glommer@openvz.org>
Acked-by: Mel Gorman <mgorman@suse.de>
---
 fs/dcache.c | 71 ++++++++++++++++++++++++++++++++++++++++++++++++++++---------
 1 file changed, 61 insertions(+), 10 deletions(-)

diff --git a/fs/dcache.c b/fs/dcache.c
index 6649764..26fd63d 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -331,7 +331,6 @@ static void dentry_lru_add(struct dentry *dentry)
 static void __dentry_lru_del(struct dentry *dentry)
 {
 	list_del_init(&dentry->d_lru);
-	dentry->d_flags &= ~DCACHE_SHRINK_LIST;
 	dentry->d_sb->s_nr_dentry_unused--;
 	this_cpu_dec(nr_dentry_unused);
 }
@@ -341,6 +340,8 @@ static void __dentry_lru_del(struct dentry *dentry)
  */
 static void dentry_lru_del(struct dentry *dentry)
 {
+	BUG_ON(dentry->d_flags & DCACHE_SHRINK_LIST);
+
 	if (!list_empty(&dentry->d_lru)) {
 		spin_lock(&dentry->d_sb->s_dentry_lru_lock);
 		__dentry_lru_del(dentry);
@@ -352,6 +353,12 @@ static void dentry_lru_del(struct dentry *dentry)
  * Remove a dentry that is unreferenced and about to be pruned
  * (unhashed and destroyed) from the LRU, and inform the file system.
  * This wrapper should be called _prior_ to unhashing a victim dentry.
+ *
+ * Check that the dentry really is on the LRU as it may be on a private dispose
+ * list and in that case we do not want to call the generic LRU removal
+ * functions. This typically happens when shrink_dcache_sb() clears the LRU in
+ * one go and then try_prune_one_dentry() walks back up the parent chain finding
+ * dentries that are also on the dispose list.
  */
 static void dentry_lru_prune(struct dentry *dentry)
 {
@@ -359,21 +366,28 @@ static void dentry_lru_prune(struct dentry *dentry)
 		if (dentry->d_flags & DCACHE_OP_PRUNE)
 			dentry->d_op->d_prune(dentry);
 
-		spin_lock(&dentry->d_sb->s_dentry_lru_lock);
-		__dentry_lru_del(dentry);
-		spin_unlock(&dentry->d_sb->s_dentry_lru_lock);
+		if ((dentry->d_flags & DCACHE_SHRINK_LIST))
+			list_del_init(&dentry->d_lru);
+		else {
+			spin_lock(&dentry->d_sb->s_dentry_lru_lock);
+			__dentry_lru_del(dentry);
+			spin_unlock(&dentry->d_sb->s_dentry_lru_lock);
+		}
+		dentry->d_flags &= ~DCACHE_SHRINK_LIST;
 	}
 }
 
 static void dentry_lru_move_list(struct dentry *dentry, struct list_head *list)
 {
+	BUG_ON(dentry->d_flags & DCACHE_SHRINK_LIST);
+
 	spin_lock(&dentry->d_sb->s_dentry_lru_lock);
 	if (list_empty(&dentry->d_lru)) {
 		list_add_tail(&dentry->d_lru, list);
-		dentry->d_sb->s_nr_dentry_unused++;
-		this_cpu_inc(nr_dentry_unused);
 	} else {
 		list_move_tail(&dentry->d_lru, list);
+		dentry->d_sb->s_nr_dentry_unused--;
+		this_cpu_dec(nr_dentry_unused);
 	}
 	spin_unlock(&dentry->d_sb->s_dentry_lru_lock);
 }
@@ -815,12 +829,18 @@ static void shrink_dentry_list(struct list_head *list)
 		}
 
 		/*
+		 * The dispose list is isolated and dentries are not accounted
+		 * to the LRU here, so we can simply remove it from the list
+		 * here regardless of whether it is referenced or not.
+		 */
+		list_del_init(&dentry->d_lru);
+
+		/*
 		 * We found an inuse dentry which was not removed from
-		 * the LRU because of laziness during lookup.  Do not free
-		 * it - just keep it off the LRU list.
+		 * the LRU because of laziness during lookup. Do not free it.
 		 */
 		if (dentry->d_count) {
-			dentry_lru_del(dentry);
+			dentry->d_flags &= ~DCACHE_SHRINK_LIST;
 			spin_unlock(&dentry->d_lock);
 			continue;
 		}
@@ -872,6 +892,8 @@ relock:
 		} else {
 			list_move_tail(&dentry->d_lru, &tmp);
 			dentry->d_flags |= DCACHE_SHRINK_LIST;
+			this_cpu_dec(nr_dentry_unused);
+			sb->s_nr_dentry_unused--;
 			spin_unlock(&dentry->d_lock);
 			if (!--count)
 				break;
@@ -885,6 +907,27 @@ relock:
 	shrink_dentry_list(&tmp);
 }
 
+/*
+ * Mark all the dentries as on being the dispose list so we don't think they are
+ * still on the LRU if we try to kill them from ascending the parent chain in
+ * try_prune_one_dentry() rather than directly from the dispose list.
+ */
+static void
+shrink_dcache_list(
+	struct list_head *dispose)
+{
+	struct dentry *dentry;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(dentry, dispose, d_lru) {
+		spin_lock(&dentry->d_lock);
+		dentry->d_flags |= DCACHE_SHRINK_LIST;
+		spin_unlock(&dentry->d_lock);
+	}
+	rcu_read_unlock();
+	shrink_dentry_list(dispose);
+}
+
 /**
  * shrink_dcache_sb - shrink dcache for a superblock
  * @sb: superblock
@@ -899,8 +942,16 @@ void shrink_dcache_sb(struct super_block *sb)
 	spin_lock(&sb->s_dentry_lru_lock);
 	while (!list_empty(&sb->s_dentry_lru)) {
 		list_splice_init(&sb->s_dentry_lru, &tmp);
+
+		/*
+		 * account for removal here so we don't need to handle it later
+		 * even though the dentry is no longer on the lru list.
+		 */
+		this_cpu_sub(nr_dentry_unused, sb->s_nr_dentry_unused);
+		sb->s_nr_dentry_unused = 0;
+
 		spin_unlock(&sb->s_dentry_lru_lock);
-		shrink_dentry_list(&tmp);
+		shrink_dcache_list(&tmp);
 		spin_lock(&sb->s_dentry_lru_lock);
 	}
 	spin_unlock(&sb->s_dentry_lru_lock);
-- 
1.8.1.4

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2013-05-09  6:06 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-09  6:06 [PATCH v5 00/31] kmemcg shrinkers Glauber Costa
2013-05-09  6:06 ` [PATCH v5 01/31] super: fix calculation of shrinkable objects for small numbers Glauber Costa
2013-05-09  6:06 ` [PATCH v5 02/31] vmscan: take at least one pass with shrinkers Glauber Costa
2013-05-09 11:12   ` Mel Gorman
2013-05-09 11:28     ` Glauber Costa
2013-05-09 11:35       ` Glauber Costa
2013-05-09  6:06 ` [PATCH v5 03/31] dcache: convert dentry_stat.nr_unused to per-cpu counters Glauber Costa
2013-05-09  6:06 ` [PATCH v5 04/31] dentry: move to per-sb LRU locks Glauber Costa
2013-05-10  5:29   ` Dave Chinner
2013-05-10  8:16     ` Dave Chinner
2013-05-09  6:06 ` Glauber Costa [this message]
2013-05-09  6:06 ` [PATCH v5 06/31] mm: new shrinker API Glauber Costa
2013-05-09 13:30   ` Mel Gorman
2013-05-09  6:06 ` [PATCH v5 07/31] shrinker: convert superblock shrinkers to new API Glauber Costa
2013-05-09 13:33   ` Mel Gorman
2013-05-09  6:06 ` [PATCH v5 08/31] list: add a new LRU list type Glauber Costa
2013-05-09 13:37   ` Mel Gorman
2013-05-09 21:02     ` Glauber Costa
2013-05-10  9:21       ` Mel Gorman
2013-05-10  9:56         ` Glauber Costa
2013-05-10 10:01           ` Mel Gorman
2013-05-09  6:06 ` [PATCH v5 09/31] inode: convert inode lru list to generic lru list code Glauber Costa
2013-05-09  6:06 ` [PATCH v5 10/31] dcache: convert to use new lru list infrastructure Glauber Costa
2013-05-09  6:06 ` [PATCH v5 11/31] list_lru: per-node " Glauber Costa
2013-05-09 13:42   ` Mel Gorman
2013-05-09 21:05     ` Glauber Costa
2013-05-09  6:06 ` [PATCH v5 12/31] shrinker: add node awareness Glauber Costa
2013-05-09  6:06 ` [PATCH v5 13/31] fs: convert inode and dentry shrinking to be node aware Glauber Costa
2013-05-09  6:06 ` [PATCH v5 14/31] xfs: convert buftarg LRU to generic code Glauber Costa
2013-05-09 13:43   ` Mel Gorman
2013-05-09  6:06 ` [PATCH v5 15/31] xfs: convert dquot cache lru to list_lru Glauber Costa
2013-05-09  6:06 ` [PATCH v5 16/31] fs: convert fs shrinkers to new scan/count API Glauber Costa
2013-05-09  6:06 ` [PATCH v5 17/31] drivers: convert shrinkers to new count/scan API Glauber Costa
2013-05-09 13:52   ` Mel Gorman
2013-05-09 21:19     ` Glauber Costa
2013-05-10  9:00       ` Mel Gorman
2013-05-09  6:06 ` [PATCH v5 18/31] shrinker: convert remaining shrinkers to " Glauber Costa
2013-05-09  6:06 ` [PATCH v5 19/31] hugepage: convert huge zero page shrinker to new shrinker API Glauber Costa
2013-05-10  1:24   ` Kirill A. Shutemov
2013-05-09  6:06 ` [PATCH v5 20/31] shrinker: Kill old ->shrink API Glauber Costa
2013-05-09 13:53   ` Mel Gorman
2013-05-09  6:06 ` [PATCH v5 21/31] vmscan: also shrink slab in memcg pressure Glauber Costa
2013-05-09  6:06 ` [PATCH v5 22/31] memcg,list_lru: duplicate LRUs upon kmemcg creation Glauber Costa
2013-05-09  6:06 ` [PATCH v5 23/31] lru: add an element to a memcg list Glauber Costa
2013-05-09  6:06 ` [PATCH v5 24/31] list_lru: per-memcg walks Glauber Costa
2013-05-09  6:06 ` [PATCH v5 25/31] memcg: per-memcg kmem shrinking Glauber Costa
2013-05-09  6:06 ` [PATCH v5 26/31] memcg: scan cache objects hierarchically Glauber Costa
2013-05-09  6:06 ` [PATCH v5 27/31] super: targeted memcg reclaim Glauber Costa
2013-05-09  6:06 ` [PATCH v5 28/31] memcg: move initialization to memcg creation Glauber Costa
2013-05-09  6:06 ` [PATCH v5 29/31] vmpressure: in-kernel notifications Glauber Costa
2013-05-09  6:06 ` [PATCH v5 30/31] memcg: reap dead memcgs upon global memory pressure Glauber Costa
2013-05-09  6:06 ` [PATCH v5 31/31] memcg: debugging facility to access dangling memcgs Glauber Costa
2013-05-09 10:55 ` [PATCH v5 00/31] kmemcg shrinkers Mel Gorman
2013-05-09 11:34   ` Glauber Costa
2013-05-09 13:18   ` Dave Chinner
2013-05-09 14:03     ` Mel Gorman
2013-05-09 21:24       ` Glauber Costa
  -- strict thread matches above, loose matches on Subject: below --
2013-05-08 20:22 Glauber Costa
2013-05-08 20:22 ` [PATCH v5 05/31] dcache: remove dentries from LRU before putting on dispose list Glauber Costa

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=1368079608-5611-6-git-send-email-glommer@openvz.org \
    --to=glommer@openvz.org \
    --cc=akpm@linux-foundation.org \
    --cc=cgroups@vger.kernel.org \
    --cc=dchinner@redhat.com \
    --cc=gthelen@google.com \
    --cc=hannes@cmpxchg.org \
    --cc=hughd@google.com \
    --cc=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mgorman@suse.de \
    --cc=mhocko@suse.cz \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).