From: Sage Weil <sage@newdream.net>
Date: Mon, 16 Mar 2009 22:29:29 +0000 (-0700)
Subject: kclient: rename listener -> unsafe; fix fsync; simplify dir_fsync
X-Git-Tag: v0.7.2~160
X-Git-Url: http://git-server-git.apps.pok.os.sepia.ceph.com/?a=commitdiff_plain;h=3ea98116ba2bf1d35e727f7d40f2afb58852391c;p=ceph.git

kclient: rename listener -> unsafe; fix fsync; simplify dir_fsync
---

diff --git a/src/kernel/dir.c b/src/kernel/dir.c
index 7540f5a80c73..e31e3a756a21 100644
--- a/src/kernel/dir.c
+++ b/src/kernel/dir.c
@@ -800,52 +800,50 @@ static ssize_t ceph_read_dir(struct file *file, char __user *buf, size_t size,
 	return size - left;
 }
 
-static int ceph_dir_fsync(struct file *file, struct dentry *dentry, int datasync)
+static int ceph_dir_fsync(struct file *file, struct dentry *dentry,
+			  int datasync)
 {
 	struct inode *inode = dentry->d_inode;
 	struct ceph_inode_info *ci = ceph_inode(inode);
-	int ret, err;
-	struct list_head *p, *n, *head;
+	struct list_head *head = &ci->i_unsafe_dirops;
 	struct ceph_mds_request *req;
 	u64 last_tid;
+	int ret = 0;
 
-	ret = 0;
-	dout(10, "sync on directory\n");
-
-	head = &ci->i_listener_list;
-
-	spin_lock(&ci->i_listener_lock);
-
+	dout(10, "dir_fsync %p\n", inode);
+	spin_lock(&ci->i_unsafe_lock);
 	if (list_empty(head))
 		goto out;
 
 	req = list_entry(head->prev,
-			 struct ceph_mds_request, r_listener_item);
+			 struct ceph_mds_request, r_unsafe_dir_item);
 	last_tid = req->r_tid;
 
-	list_for_each_safe(p, n, head) {
-		req = list_entry(p, struct ceph_mds_request, r_listener_item);
-
-		/* avoid starvation */
-		if (req->r_tid > last_tid)
-			goto out;
-
+	do {
 		ceph_mdsc_get_request(req);
-		spin_unlock(&ci->i_listener_lock);
+		spin_unlock(&ci->i_unsafe_lock);
+		dout(10, "dir_fsync %p wait on tid %llu (until %llu)\n",
+		     inode, req->r_tid, last_tid);
 		if (req->r_timeout) {
-			err = wait_for_completion_timeout(&req->r_safe_completion,
-							req->r_timeout);
-			if (err == 0)
+			ret = wait_for_completion_timeout(&req->r_safe_completion,
+							  req->r_timeout);
+			if (ret > 0)
+				ret = 0;
+			else if (ret == 0)
 				ret = -EIO;  /* timed out */
 		} else {
 			wait_for_completion(&req->r_safe_completion);
 		}
-		spin_lock(&ci->i_listener_lock);
-
+		spin_lock(&ci->i_unsafe_lock);
 		ceph_mdsc_put_request(req);
-	}
+
+		if (ret || list_empty(head))
+			break;
+		req = list_entry(head->next,
+				 struct ceph_mds_request, r_unsafe_dir_item);
+	} while (req->r_tid < last_tid);
 out:
-	spin_unlock(&ci->i_listener_lock);
+	spin_unlock(&ci->i_unsafe_lock);
 	return ret;
 }
 
diff --git a/src/kernel/file.c b/src/kernel/file.c
index d027d71fa9d9..0c379bd18871 100644
--- a/src/kernel/file.c
+++ b/src/kernel/file.c
@@ -412,9 +412,9 @@ static void sync_write_commit(struct ceph_osd_request *req)
 	struct ceph_inode_info *ci = ceph_inode(req->r_inode);
 
 	dout(10, "sync_write_commit %p tid %llu\n", req, req->r_tid);
-	spin_lock(&ci->i_listener_lock);
+	spin_lock(&ci->i_unsafe_lock);
 	list_del_init(&req->r_unsafe_item);
-	spin_unlock(&ci->i_listener_lock);
+	spin_unlock(&ci->i_unsafe_lock);
 	ceph_put_cap_refs(ci, CEPH_CAP_FILE_WR);
 }
 
@@ -431,7 +431,7 @@ static void sync_write_wait(struct inode *inode)
        struct ceph_osd_request *req;
        u64 last_tid;
 
-       spin_lock(&ci->i_listener_lock);
+       spin_lock(&ci->i_unsafe_lock);
        if (list_empty(head))
 	       goto out;
 
@@ -442,22 +442,24 @@ static void sync_write_wait(struct inode *inode)
 
        do {
 	       ceph_osdc_get_request(req);
-	       spin_unlock(&ci->i_listener_lock);
+	       spin_unlock(&ci->i_unsafe_lock);
 	       dout(10, "sync_write_wait on tid %llu (until %llu)\n",
 		    req->r_tid, last_tid);
 	       wait_for_completion(&req->r_safe_completion);
-	       spin_lock(&ci->i_listener_lock);
+	       spin_lock(&ci->i_unsafe_lock);
 	       ceph_osdc_put_request(req);
 
 	       /*
 		* from here on look at first entry in chain, since we
 		* only want to wait for anything older than last_tid
 		*/
+	       if (list_empty(head))
+		       break;
 	       req = list_entry(head->next, struct ceph_osd_request,
 	                        r_unsafe_item);
-       } while (req->r_tid <= last_tid);
+       } while (req->r_tid < last_tid);
 out:
-       spin_unlock(&ci->i_listener_lock);
+       spin_unlock(&ci->i_unsafe_lock);
 }
 
 /*
@@ -560,9 +562,9 @@ more:
 			 * Add to inode unsafe list only after we
 			 * start_request so that a tid has been assigned.
 			 */
-			spin_lock(&ci->i_listener_lock);
+			spin_lock(&ci->i_unsafe_lock);
 			list_add(&ci->i_unsafe_writes, &req->r_unsafe_item);
-			spin_unlock(&ci->i_listener_lock);
+			spin_unlock(&ci->i_unsafe_lock);
 			ceph_get_more_cap_refs(ci, CEPH_CAP_FILE_WR);
 		}
 		ret = ceph_osdc_wait_request(&client->osdc, req);
@@ -733,7 +735,7 @@ static int ceph_fsync(struct file *file, struct dentry *dentry, int datasync)
 	struct inode *inode = dentry->d_inode;
 	int ret;
 
-	dout(10, "fsync on inode %p\n", inode);
+	dout(10, "fsync %p\n", inode);
 	sync_write_wait(inode);
 
 	ret = filemap_write_and_wait(inode->i_mapping);
diff --git a/src/kernel/inode.c b/src/kernel/inode.c
index bedf122a0d21..0754f274e87f 100644
--- a/src/kernel/inode.c
+++ b/src/kernel/inode.c
@@ -288,7 +288,10 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	ci->i_wrbuffer_ref_head = 0;
 	ci->i_rdcache_gen = 0;
 	ci->i_rdcache_revoking = 0;
+
 	INIT_LIST_HEAD(&ci->i_unsafe_writes);
+	INIT_LIST_HEAD(&ci->i_unsafe_dirops);
+	spin_lock_init(&ci->i_unsafe_lock);
 
 	ci->i_snap_realm = NULL;
 	INIT_LIST_HEAD(&ci->i_snap_realm_item);
@@ -299,9 +302,6 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 
 	INIT_WORK(&ci->i_vmtruncate_work, ceph_vmtruncate_work);
 
-	INIT_LIST_HEAD(&ci->i_listener_list);
-	spin_lock_init(&ci->i_listener_lock);
-
 	return &ci->vfs_inode;
 }
 
diff --git a/src/kernel/mds_client.c b/src/kernel/mds_client.c
index 1eb0210b2f13..6959d461bab7 100644
--- a/src/kernel/mds_client.c
+++ b/src/kernel/mds_client.c
@@ -441,10 +441,10 @@ static void __register_request(struct ceph_mds_client *mdsc,
 	if (listener) {
 		struct ceph_inode_info *ci = ceph_inode(listener);
 		
-		spin_lock(&ci->i_listener_lock);
-		req->r_listener = listener;
-		list_add_tail(&req->r_listener_item, &ci->i_listener_list);
-		spin_unlock(&ci->i_listener_lock);
+		spin_lock(&ci->i_unsafe_lock);
+		req->r_unsafe_dir = listener;
+		list_add_tail(&req->r_unsafe_dir_item, &ci->i_unsafe_dirops);
+		spin_unlock(&ci->i_unsafe_lock);
 	}
 
 	ceph_sysfs_mds_req_init(mdsc, req);
@@ -457,12 +457,12 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
 	radix_tree_delete(&mdsc->request_tree, req->r_tid);
 	ceph_mdsc_put_request(req);
 
-	if (req->r_listener) {
-		struct ceph_inode_info *ci = ceph_inode(req->r_listener);
+	if (req->r_unsafe_dir) {
+		struct ceph_inode_info *ci = ceph_inode(req->r_unsafe_dir);
 
-		spin_lock(&ci->i_listener_lock);
-		list_del_init(&req->r_listener_item);
-		spin_unlock(&ci->i_listener_lock);
+		spin_lock(&ci->i_unsafe_lock);
+		list_del_init(&req->r_unsafe_dir_item);
+		spin_unlock(&ci->i_unsafe_lock);
 	}
 
 	ceph_sysfs_mds_req_cleanup(req);
@@ -836,7 +836,7 @@ ceph_mdsc_create_request(struct ceph_mds_client *mdsc, int op,
 		return ERR_PTR(-ENOMEM);
 	req->r_started = jiffies;
 	req->r_resend_mds = -1;
-	INIT_LIST_HEAD(&req->r_listener_item);
+	INIT_LIST_HEAD(&req->r_unsafe_dir_item);
 	req->r_fmode = -1;
 	atomic_set(&req->r_ref, 1);  /* one for request_tree, one for caller */
 	INIT_LIST_HEAD(&req->r_wait);
diff --git a/src/kernel/mds_client.h b/src/kernel/mds_client.h
index 9d25a9d004de..0af5851231f0 100644
--- a/src/kernel/mds_client.h
+++ b/src/kernel/mds_client.h
@@ -179,8 +179,8 @@ struct ceph_mds_request {
 	u32 r_direct_hash;      /* choose dir frag based on this dentry hash */
 	bool r_direct_is_hash;  /* true if r_direct_hash is valid */
 
-	struct inode	*r_listener;
-	struct list_head r_listener_item;
+	struct inode	*r_unsafe_dir;
+	struct list_head r_unsafe_dir_item;
 
 	/* references to the trailing dentry and inode from parsing the
 	 * mds response.  also used to feed a VFS-provided dentry into
diff --git a/src/kernel/super.h b/src/kernel/super.h
index 5caeae807d93..403f83a65545 100644
--- a/src/kernel/super.h
+++ b/src/kernel/super.h
@@ -281,7 +281,10 @@ struct ceph_inode_info {
 				   If it's non-zero, we _may_ have cached
 				   pages. */
 	u32 i_rdcache_revoking; /* RDCACHE gen to async invalidate, if any */
+
 	struct list_head i_unsafe_writes; /* uncommitted sync writes */
+	struct list_head i_unsafe_dirops; /* uncommitted mds dir ops */
+	spinlock_t i_unsafe_lock;
 
 	struct ceph_snap_realm *i_snap_realm; /* snap realm (if caps) */
 	struct list_head i_snap_realm_item;
@@ -292,9 +295,6 @@ struct ceph_inode_info {
 
 	struct work_struct i_vmtruncate_work;
 
-	struct list_head i_listener_list; /* requests we pend on */
-	spinlock_t i_listener_lock;
-
 	struct inode vfs_inode; /* at end */
 };