All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vivek Goyal <vgoyal@redhat.com>
To: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	miklos@szeredi.hu
Cc: virtio-fs@redhat.com, vgoyal@redhat.com, stefanha@redhat.com,
	dgilbert@redhat.com, Miklos Szeredi <mszeredi@redhat.com>
Subject: [PATCH 01/13] fuse: delete dentry if timeout is zero
Date: Wed, 21 Aug 2019 13:37:30 -0400	[thread overview]
Message-ID: <20190821173742.24574-2-vgoyal@redhat.com> (raw)
In-Reply-To: <20190821173742.24574-1-vgoyal@redhat.com>

From: Miklos Szeredi <mszeredi@redhat.com>

Don't hold onto dentry in lru list if need to re-lookup it anyway at next
access.

More advanced version of this patch would periodically flush out dentries
from the lru which have gone stale.

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
---
 fs/fuse/dir.c | 26 +++++++++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index dd0f64f7bc06..fd8636e67ae9 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -29,12 +29,26 @@ union fuse_dentry {
 	struct rcu_head rcu;
 };
 
-static inline void fuse_dentry_settime(struct dentry *entry, u64 time)
+static void fuse_dentry_settime(struct dentry *dentry, u64 time)
 {
-	((union fuse_dentry *) entry->d_fsdata)->time = time;
+	/*
+	 * Mess with DCACHE_OP_DELETE because dput() will be faster without it.
+	 *  Don't care about races, either way it's just an optimization
+	 */
+	if ((time && (dentry->d_flags & DCACHE_OP_DELETE)) ||
+	    (!time && !(dentry->d_flags & DCACHE_OP_DELETE))) {
+		spin_lock(&dentry->d_lock);
+		if (time)
+			dentry->d_flags &= ~DCACHE_OP_DELETE;
+		else
+			dentry->d_flags |= DCACHE_OP_DELETE;
+		spin_unlock(&dentry->d_lock);
+	}
+
+	((union fuse_dentry *) dentry->d_fsdata)->time = time;
 }
 
-static inline u64 fuse_dentry_time(struct dentry *entry)
+static inline u64 fuse_dentry_time(const struct dentry *entry)
 {
 	return ((union fuse_dentry *) entry->d_fsdata)->time;
 }
@@ -255,8 +269,14 @@ static void fuse_dentry_release(struct dentry *dentry)
 	kfree_rcu(fd, rcu);
 }
 
+static int fuse_dentry_delete(const struct dentry *dentry)
+{
+	return time_before64(fuse_dentry_time(dentry), get_jiffies_64());
+}
+
 const struct dentry_operations fuse_dentry_operations = {
 	.d_revalidate	= fuse_dentry_revalidate,
+	.d_delete	= fuse_dentry_delete,
 	.d_init		= fuse_dentry_init,
 	.d_release	= fuse_dentry_release,
 };
-- 
2.20.1


WARNING: multiple messages have this Message-ID (diff)
From: Vivek Goyal <vgoyal@redhat.com>
To: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
	miklos@szeredi.hu
Cc: virtio-fs@redhat.com, Miklos Szeredi <mszeredi@redhat.com>,
	vgoyal@redhat.com
Subject: [Virtio-fs] [PATCH 01/13] fuse: delete dentry if timeout is zero
Date: Wed, 21 Aug 2019 13:37:30 -0400	[thread overview]
Message-ID: <20190821173742.24574-2-vgoyal@redhat.com> (raw)
In-Reply-To: <20190821173742.24574-1-vgoyal@redhat.com>

From: Miklos Szeredi <mszeredi@redhat.com>

Don't hold onto dentry in lru list if need to re-lookup it anyway at next
access.

More advanced version of this patch would periodically flush out dentries
from the lru which have gone stale.

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
---
 fs/fuse/dir.c | 26 +++++++++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index dd0f64f7bc06..fd8636e67ae9 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -29,12 +29,26 @@ union fuse_dentry {
 	struct rcu_head rcu;
 };
 
-static inline void fuse_dentry_settime(struct dentry *entry, u64 time)
+static void fuse_dentry_settime(struct dentry *dentry, u64 time)
 {
-	((union fuse_dentry *) entry->d_fsdata)->time = time;
+	/*
+	 * Mess with DCACHE_OP_DELETE because dput() will be faster without it.
+	 *  Don't care about races, either way it's just an optimization
+	 */
+	if ((time && (dentry->d_flags & DCACHE_OP_DELETE)) ||
+	    (!time && !(dentry->d_flags & DCACHE_OP_DELETE))) {
+		spin_lock(&dentry->d_lock);
+		if (time)
+			dentry->d_flags &= ~DCACHE_OP_DELETE;
+		else
+			dentry->d_flags |= DCACHE_OP_DELETE;
+		spin_unlock(&dentry->d_lock);
+	}
+
+	((union fuse_dentry *) dentry->d_fsdata)->time = time;
 }
 
-static inline u64 fuse_dentry_time(struct dentry *entry)
+static inline u64 fuse_dentry_time(const struct dentry *entry)
 {
 	return ((union fuse_dentry *) entry->d_fsdata)->time;
 }
@@ -255,8 +269,14 @@ static void fuse_dentry_release(struct dentry *dentry)
 	kfree_rcu(fd, rcu);
 }
 
+static int fuse_dentry_delete(const struct dentry *dentry)
+{
+	return time_before64(fuse_dentry_time(dentry), get_jiffies_64());
+}
+
 const struct dentry_operations fuse_dentry_operations = {
 	.d_revalidate	= fuse_dentry_revalidate,
+	.d_delete	= fuse_dentry_delete,
 	.d_init		= fuse_dentry_init,
 	.d_release	= fuse_dentry_release,
 };
-- 
2.20.1


  reply	other threads:[~2019-08-21 17:39 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-08-21 17:37 [PATCH v3 00/13] virtio-fs: shared file system for virtual machines Vivek Goyal
2019-08-21 17:37 ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` Vivek Goyal [this message]
2019-08-21 17:37   ` [Virtio-fs] [PATCH 01/13] fuse: delete dentry if timeout is zero Vivek Goyal
2019-08-21 17:37 ` [PATCH 02/13] fuse: Use default_file_splice_read for direct IO Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-28  7:45   ` Miklos Szeredi
2019-08-28  7:45     ` [Virtio-fs] " Miklos Szeredi
2019-08-28 12:27     ` Vivek Goyal
2019-08-28 12:27       ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 03/13] fuse: export fuse_end_request() Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 04/13] fuse: export fuse_len_args() Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 05/13] fuse: Export fuse_send_init_request() Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 06/13] fuse: export fuse_get_unique() Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 07/13] Export fuse_dequeue_forget() function Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 08/13] fuse: extract fuse_fill_super_common() Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 09/13] fuse: add fuse_iqueue_ops callbacks Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 10/13] fuse: Separate fuse device allocation and installation in fuse_conn Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 11/13] virtio_fs: add skeleton virtio_fs.ko module Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 12/13] virtio-fs: Do not provide abort interface in fusectl Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-21 17:37 ` [PATCH 13/13] init/do_mounts.c: add virtio_fs root fs support Vivek Goyal
2019-08-21 17:37   ` [Virtio-fs] " Vivek Goyal
2019-08-29  9:28 ` [PATCH v3 00/13] virtio-fs: shared file system for virtual machines Miklos Szeredi
2019-08-29  9:28   ` [Virtio-fs] " Miklos Szeredi
2019-08-29 11:58   ` Vivek Goyal
2019-08-29 11:58     ` [Virtio-fs] " Vivek Goyal
2019-08-29 12:35   ` Stefan Hajnoczi
2019-08-29 12:35     ` [Virtio-fs] " Stefan Hajnoczi
2019-08-29 13:29   ` Vivek Goyal
2019-08-29 13:29     ` [Virtio-fs] " Vivek Goyal
2019-08-29 13:41     ` Miklos Szeredi
2019-08-29 13:41       ` [Virtio-fs] " Miklos Szeredi
2019-08-29 14:31       ` Vivek Goyal
2019-08-29 14:31         ` [Virtio-fs] " Vivek Goyal
2019-08-29 14:47         ` Miklos Szeredi
2019-08-29 14:47           ` [Virtio-fs] " Miklos Szeredi
2019-08-29 16:01   ` Vivek Goyal
2019-08-29 16:01     ` [Virtio-fs] " Vivek Goyal
2019-08-31  5:46     ` Miklos Szeredi
2019-08-31  5:46       ` [Virtio-fs] " Miklos Szeredi
2019-09-03  8:05 ` Miklos Szeredi
2019-09-03  8:05   ` [Virtio-fs] " Miklos Szeredi
2019-09-03  8:31   ` Michael S. Tsirkin
2019-09-03  8:31   ` Michael S. Tsirkin
2019-09-03  8:31     ` [Virtio-fs] " Michael S. Tsirkin
2019-09-03  9:17     ` Miklos Szeredi
2019-09-03  9:17       ` [Virtio-fs] " Miklos Szeredi
2019-09-04 15:54       ` Stefan Hajnoczi
2019-09-04 15:54         ` [Virtio-fs] " Stefan Hajnoczi
2019-09-04 15:54       ` Stefan Hajnoczi
2019-09-03 14:07     ` Vivek Goyal
2019-09-03 14:07       ` [Virtio-fs] " Vivek Goyal
2019-09-03 14:12       ` Michael S. Tsirkin
2019-09-03 14:12         ` [Virtio-fs] " Michael S. Tsirkin
2019-09-03 14:18         ` Vivek Goyal
2019-09-03 14:18           ` [Virtio-fs] " Vivek Goyal
2019-09-03 17:15           ` Michael S. Tsirkin
2019-09-03 17:15             ` [Virtio-fs] " Michael S. Tsirkin
2019-09-03 17:15           ` Michael S. Tsirkin
2019-09-03 14:18         ` Vivek Goyal
2019-09-03 14:12       ` Michael S. Tsirkin
2019-09-03 14:07     ` Vivek Goyal

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=20190821173742.24574-2-vgoyal@redhat.com \
    --to=vgoyal@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miklos@szeredi.hu \
    --cc=mszeredi@redhat.com \
    --cc=stefanha@redhat.com \
    --cc=virtio-fs@redhat.com \
    /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.