From: Dave Chinner <david@fromorbit.com>
To: linux-fsdevel@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Subject: [PATCH 16/18] fs: Make iunique independent of inode_lock
Date: Fri, 8 Oct 2010 16:21:30 +1100 [thread overview]
Message-ID: <1286515292-15882-17-git-send-email-david@fromorbit.com> (raw)
In-Reply-To: <1286515292-15882-1-git-send-email-david@fromorbit.com>
From: Nick Piggin <npiggin@suse.de>
Before removing the inode_lock, the iunique counter needs to be made
independent of the inode_lock. Add a new lock to protect the iunique
counter and nest it inside the inode_lock to provide the same
protection that the inode_lock currently provides.
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Dave Chinner <dchinner@redhat.com>
---
fs/inode.c | 33 ++++++++++++++++++++++++++++-----
1 files changed, 28 insertions(+), 5 deletions(-)
diff --git a/fs/inode.c b/fs/inode.c
index 13e1325..4ec360e 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1070,6 +1070,30 @@ static struct inode *get_new_inode_fast(struct super_block *sb,
return inode;
}
+/*
+ * search the inode cache for a matching inode number.
+ * If we find one, then the inode number we are trying to
+ * allocate is not unique and so we should not use it.
+ *
+ * Returns 1 if the inode number is unique, 0 if it is not.
+ */
+static int test_inode_iunique(struct super_block * sb, unsigned long ino)
+{
+ struct inode_hash_bucket *b = inode_hashtable + hash(sb, ino);
+ struct hlist_bl_node *node;
+ struct inode *inode;
+
+ spin_lock_bucket(b);
+ hlist_bl_for_each_entry(inode, node, &b->head, i_hash) {
+ if (inode->i_ino == ino && inode->i_sb == sb) {
+ spin_unlock_bucket(b);
+ return 0;
+ }
+ }
+ spin_unlock_bucket(b);
+ return 1;
+}
+
/**
* iunique - get a unique inode number
* @sb: superblock
@@ -1091,19 +1115,18 @@ ino_t iunique(struct super_block *sb, ino_t max_reserved)
* error if st_ino won't fit in target struct field. Use 32bit counter
* here to attempt to avoid that.
*/
+ static DEFINE_SPINLOCK(unique_lock);
static unsigned int counter;
- struct inode *inode;
- struct inode_hash_bucket *b;
ino_t res;
spin_lock(&inode_lock);
+ spin_lock(&unique_lock);
do {
if (counter <= max_reserved)
counter = max_reserved + 1;
res = counter++;
- b = inode_hashtable + hash(sb, res);
- inode = find_inode_fast(sb, b, res);
- } while (inode != NULL);
+ } while (!test_inode_iunique(sb, res));
+ spin_unlock(&unique_lock);
spin_unlock(&inode_lock);
return res;
--
1.7.1
next prev parent reply other threads:[~2010-10-08 5:21 UTC|newest]
Thread overview: 162+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-08 5:21 fs: Inode cache scalability V2 Dave Chinner
2010-10-08 5:21 ` [PATCH 01/18] kernel: add bl_list Dave Chinner
2010-10-08 8:18 ` Andi Kleen
2010-10-08 10:33 ` Dave Chinner
2010-10-08 5:21 ` [PATCH 02/18] fs: Convert nr_inodes and nr_unused to per-cpu counters Dave Chinner
2010-10-08 7:01 ` Christoph Hellwig
2010-10-08 5:21 ` [PATCH 03/18] fs: keep inode with backing-dev Dave Chinner
2010-10-08 7:01 ` Christoph Hellwig
2010-10-08 7:27 ` Dave Chinner
2010-10-08 5:21 ` [PATCH 04/18] fs: Implement lazy LRU updates for inodes Dave Chinner
2010-10-08 7:08 ` Christoph Hellwig
2010-10-08 7:31 ` Dave Chinner
2010-10-08 9:08 ` Al Viro
2010-10-08 9:51 ` Dave Chinner
2010-10-08 5:21 ` [PATCH 05/18] fs: inode split IO and LRU lists Dave Chinner
2010-10-08 7:14 ` Christoph Hellwig
2010-10-08 7:38 ` Dave Chinner
2010-10-08 9:16 ` Al Viro
2010-10-08 9:58 ` Dave Chinner
2010-10-08 5:21 ` [PATCH 06/18] fs: Clean up inode reference counting Dave Chinner
2010-10-08 7:20 ` Christoph Hellwig
2010-10-08 7:46 ` Dave Chinner
2010-10-08 8:15 ` Christoph Hellwig
2010-10-08 5:21 ` [PATCH 07/18] exofs: use iput() for inode reference count decrements Dave Chinner
2010-10-08 7:21 ` Christoph Hellwig
2010-10-16 7:56 ` Nick Piggin
2010-10-16 16:29 ` Christoph Hellwig
2010-10-17 15:41 ` Boaz Harrosh
2010-10-08 5:21 ` [PATCH 08/18] fs: add inode reference coutn read accessor Dave Chinner
2010-10-08 7:24 ` Christoph Hellwig
2010-10-08 5:21 ` [PATCH 09/18] fs: rework icount to be a locked variable Dave Chinner
2010-10-08 7:27 ` Christoph Hellwig
2010-10-08 7:50 ` Dave Chinner
2010-10-08 8:17 ` Christoph Hellwig
2010-10-08 13:16 ` Chris Mason
2010-10-08 9:32 ` Al Viro
2010-10-08 10:15 ` Dave Chinner
2010-10-08 13:14 ` Chris Mason
2010-10-08 13:53 ` Christoph Hellwig
2010-10-08 14:09 ` Dave Chinner
2010-10-08 5:21 ` [PATCH 10/18] fs: Factor inode hash operations into functions Dave Chinner
2010-10-08 7:29 ` Christoph Hellwig
2010-10-08 9:41 ` Al Viro
2010-10-08 5:21 ` [PATCH 11/18] fs: Introduce per-bucket inode hash locks Dave Chinner
2010-10-08 7:33 ` Christoph Hellwig
2010-10-08 7:51 ` Dave Chinner
2010-10-08 9:49 ` Al Viro
2010-10-08 9:51 ` Christoph Hellwig
2010-10-08 13:43 ` Christoph Hellwig
2010-10-08 14:17 ` Dave Chinner
2010-10-08 18:54 ` Christoph Hellwig
2010-10-16 7:57 ` Nick Piggin
2010-10-16 16:16 ` Christoph Hellwig
2010-10-16 17:12 ` Nick Piggin
2010-10-17 0:45 ` Christoph Hellwig
2010-10-17 2:06 ` Nick Piggin
2010-10-17 0:46 ` Dave Chinner
2010-10-17 2:25 ` Nick Piggin
2010-10-18 16:16 ` Andi Kleen
2010-10-18 16:21 ` Christoph Hellwig
2010-10-19 7:00 ` Nick Piggin
2010-10-19 16:50 ` Christoph Hellwig
2010-10-20 3:11 ` Nick Piggin
2010-10-24 15:44 ` Thomas Gleixner
2010-10-24 21:17 ` Nick Piggin
2010-10-25 4:41 ` Thomas Gleixner
2010-10-25 7:04 ` Thomas Gleixner
2010-10-26 0:12 ` Nick Piggin
2010-10-26 0:06 ` Nick Piggin
2010-10-08 5:21 ` [PATCH 12/18] fs: add a per-superblock lock for the inode list Dave Chinner
2010-10-08 7:35 ` Christoph Hellwig
2010-10-08 5:21 ` [PATCH 13/18] fs: split locking of inode writeback and LRU lists Dave Chinner
2010-10-08 7:42 ` Christoph Hellwig
2010-10-08 8:00 ` Dave Chinner
2010-10-08 8:18 ` Christoph Hellwig
2010-10-16 7:57 ` Nick Piggin
2010-10-16 16:20 ` Christoph Hellwig
2010-10-16 17:19 ` Nick Piggin
2010-10-17 1:00 ` Dave Chinner
2010-10-17 2:20 ` Nick Piggin
2010-10-08 5:21 ` [PATCH 14/18] fs: Protect inode->i_state with th einode->i_lock Dave Chinner
2010-10-08 7:49 ` Christoph Hellwig
2010-10-08 8:04 ` Dave Chinner
2010-10-08 8:18 ` Christoph Hellwig
2010-10-16 7:57 ` Nick Piggin
2010-10-16 16:19 ` Christoph Hellwig
2010-10-09 8:05 ` Christoph Hellwig
2010-10-09 14:52 ` Matthew Wilcox
2010-10-10 2:01 ` Dave Chinner
2010-10-08 5:21 ` [PATCH 15/18] fs: introduce a per-cpu last_ino allocator Dave Chinner
2010-10-08 7:53 ` Christoph Hellwig
2010-10-08 8:05 ` Dave Chinner
2010-10-08 8:22 ` Andi Kleen
2010-10-08 8:44 ` Christoph Hellwig
2010-10-08 9:58 ` Al Viro
2010-10-08 10:09 ` Andi Kleen
2010-10-08 10:19 ` Al Viro
2010-10-08 10:20 ` Eric Dumazet
2010-10-08 9:56 ` Al Viro
2010-10-08 10:03 ` Christoph Hellwig
2010-10-08 10:20 ` Eric Dumazet
2010-10-08 13:48 ` Christoph Hellwig
2010-10-08 14:06 ` Eric Dumazet
2010-10-08 19:10 ` Christoph Hellwig
2010-10-09 17:14 ` Matthew Wilcox
2010-10-16 7:57 ` Nick Piggin
2010-10-16 16:22 ` Christoph Hellwig
2010-10-16 17:21 ` Nick Piggin
2010-10-08 5:21 ` Dave Chinner [this message]
2010-10-08 7:55 ` [PATCH 16/18] fs: Make iunique independent of inode_lock Christoph Hellwig
2010-10-08 8:06 ` Dave Chinner
2010-10-08 8:19 ` Christoph Hellwig
2010-10-08 5:21 ` [PATCH 17/18] fs: icache remove inode_lock Dave Chinner
2010-10-08 8:03 ` Christoph Hellwig
2010-10-08 8:09 ` Dave Chinner
2010-10-13 7:20 ` Nick Piggin
2010-10-13 7:27 ` Nick Piggin
2010-10-13 11:28 ` Christoph Hellwig
2010-10-13 12:03 ` Nick Piggin
2010-10-13 12:20 ` Christoph Hellwig
2010-10-13 12:25 ` Nick Piggin
2010-10-13 10:42 ` Eric Dumazet
2010-10-13 12:07 ` Nick Piggin
2010-10-13 11:25 ` Christoph Hellwig
2010-10-13 12:30 ` Nick Piggin
2010-10-13 23:23 ` Dave Chinner
2010-10-14 9:06 ` Nick Piggin
2010-10-14 9:13 ` Nick Piggin
2010-10-14 14:41 ` Christoph Hellwig
2010-10-15 0:14 ` Nick Piggin
2010-10-15 3:13 ` Dave Chinner
2010-10-15 3:30 ` Nick Piggin
2010-10-15 3:44 ` Nick Piggin
2010-10-15 6:41 ` Nick Piggin
2010-10-15 10:59 ` Dave Chinner
2010-10-15 13:03 ` Nick Piggin
2010-10-15 13:29 ` Nick Piggin
2010-10-15 17:33 ` Nick Piggin
2010-10-15 17:52 ` Christoph Hellwig
2010-10-15 18:02 ` Nick Piggin
2010-10-15 18:14 ` Nick Piggin
2010-10-16 2:09 ` Nick Piggin
2010-10-15 14:11 ` Nick Piggin
2010-10-15 20:50 ` Nick Piggin
2010-10-15 20:56 ` Nick Piggin
2010-10-15 4:04 ` Nick Piggin
2010-10-15 11:33 ` Dave Chinner
2010-10-15 13:14 ` Nick Piggin
2010-10-15 15:38 ` Nick Piggin
2010-10-16 7:57 ` Nick Piggin
2010-10-08 5:21 ` [PATCH 18/18] fs: Reduce inode I_FREEING and factor inode disposal Dave Chinner
2010-10-08 8:11 ` Christoph Hellwig
2010-10-08 10:18 ` Al Viro
2010-10-08 10:52 ` Dave Chinner
2010-10-08 12:10 ` Al Viro
2010-10-08 13:55 ` Dave Chinner
2010-10-09 17:22 ` Matthew Wilcox
2010-10-09 8:08 ` [PATCH 19/18] fs: split __inode_add_to_list Christoph Hellwig
2010-10-12 10:47 ` Dave Chinner
2010-10-12 11:31 ` Christoph Hellwig
2010-10-12 12:05 ` Dave Chinner
2010-10-09 11:18 ` [PATCH 20/18] fs: do not assign default i_ino in new_inode Christoph Hellwig
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=1286515292-15882-17-git-send-email-david@fromorbit.com \
--to=david@fromorbit.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@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 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).