All of lore.kernel.org
 help / color / mirror / Atom feed
From: Anna Schumaker <Anna.Schumaker@netapp.com>
To: <Trond.Myklebust@primarydata.com>, <linux-nfs@vger.kernel.org>
Cc: <Anna.Schumaker@netapp.com>
Subject: [PATCH v4 2/2] NFS: Reduce time spent holding the i_mutex during fallocate()
Date: Mon, 16 Mar 2015 14:06:24 -0400	[thread overview]
Message-ID: <1426529184-8794-3-git-send-email-Anna.Schumaker@Netapp.com> (raw)
In-Reply-To: <1426529184-8794-1-git-send-email-Anna.Schumaker@Netapp.com>

At the very least, we should not be taking the i_mutex until after
checking if the server even supports ALLOCATE or DEALLOCATE, allowing
v4.0 or v4.1 to exit without potentially waiting on a lock.

Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
---
 fs/nfs/nfs42proc.c | 8 ++++++++
 fs/nfs/nfs4file.c  | 9 ++-------
 2 files changed, 10 insertions(+), 7 deletions(-)

diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index b9aa6bb..3a9e752 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -96,9 +96,13 @@ int nfs42_proc_allocate(struct file *filep, loff_t offset, loff_t len)
 	if (!nfs_server_capable(inode, NFS_CAP_ALLOCATE))
 		return -EOPNOTSUPP;
 
+	mutex_lock(&inode->i_mutex);
+
 	err = nfs42_proc_fallocate(&msg, filep, offset, len);
 	if (err == -EOPNOTSUPP)
 		NFS_SERVER(inode)->caps &= ~NFS_CAP_ALLOCATE;
+
+	mutex_unlock(&inode->i_mutex);
 	return err;
 }
 
@@ -114,11 +118,15 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
 		return -EOPNOTSUPP;
 
 	nfs_wb_all(inode);
+	mutex_lock(&inode->i_mutex);
+
 	err = nfs42_proc_fallocate(&msg, filep, offset, len);
 	if (err == 0)
 		truncate_pagecache_range(inode, offset, (offset + len) -1);
 	if (err == -EOPNOTSUPP)
 		NFS_SERVER(inode)->caps &= ~NFS_CAP_DEALLOCATE;
+
+	mutex_unlock(&inode->i_mutex);
 	return err;
 }
 
diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c
index d09c689..1d1848e 100644
--- a/fs/nfs/nfs4file.c
+++ b/fs/nfs/nfs4file.c
@@ -152,14 +152,9 @@ static long nfs42_fallocate(struct file *filep, int mode, loff_t offset, loff_t
 	if (ret < 0)
 		return ret;
 
-	mutex_lock(&inode->i_mutex);
 	if (mode & FALLOC_FL_PUNCH_HOLE)
-		ret = nfs42_proc_deallocate(filep, offset, len);
-	else
-		ret = nfs42_proc_allocate(filep, offset, len);
-	mutex_unlock(&inode->i_mutex);
-
-	return ret;
+		return nfs42_proc_deallocate(filep, offset, len);
+	return nfs42_proc_allocate(filep, offset, len);
 }
 #endif /* CONFIG_NFS_V4_2 */
 
-- 
2.3.3


      parent reply	other threads:[~2015-03-16 18:06 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-16 18:06 [PATCH v4 0/2] NFS: ALLOCATE and DEALLOCATE improvemnts Anna Schumaker
2015-03-16 18:06 ` [PATCH v4 1/2] NFS: Don't zap caches on fallocate() Anna Schumaker
2015-03-16 18:06 ` Anna Schumaker [this message]

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=1426529184-8794-3-git-send-email-Anna.Schumaker@Netapp.com \
    --to=anna.schumaker@netapp.com \
    --cc=Trond.Myklebust@primarydata.com \
    --cc=linux-nfs@vger.kernel.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.