orangefs: implement statx

Fortunately OrangeFS has had a getattr request mask for a long time.

The server basically has two difficulty levels for attributes.  Fetching
any attribute except size requires communicating with the metadata
server for that handle.  Since all the attributes are right there, it
makes sense to return them all.  Fetching the size requires
communicating with every I/O server (that the file is distributed
across).  Therefore if asked for anything except size, get everything
except size, and if asked for size, get everything.

Signed-off-by: Martin Brandenburg <martin@omnibond.com>
Signed-off-by: Mike Marshall <hubcap@omnibond.com>

5 files changed, 64 insertions, 28 deletions

diff --git a/fs/orangefs/file.c b/fs/orangefs/file.c
index e6bbc80..b421df1 100644
--- a/fs/orangefs/file.c
+++ b/fs/orangefs/file.c
@@ -475,7 +475,8 @@ static ssize_t orangefs_file_write_iter(struct kiocb *iocb, struct iov_iter *ite
 
 	/* Make sure generic_write_checks sees an up to date inode size. */
 	if (file->f_flags & O_APPEND) {
-		rc = orangefs_inode_getattr(file->f_mapping->host, 0, 1);
+		rc = orangefs_inode_getattr(file->f_mapping->host, 0, 1,
+		    STATX_SIZE);
 		if (rc == -ESTALE)
 			rc = -EIO;
 		if (rc) {
@@ -693,7 +694,8 @@ static loff_t orangefs_file_llseek(struct file *file, loff_t offset, int origin)
 		 * NOTE: We are only interested in file size here,
 		 * so we set mask accordingly.
 		 */
-		ret = orangefs_inode_getattr(file->f_mapping->host, 0, 1);
+		ret = orangefs_inode_getattr(file->f_mapping->host, 0, 1,
+		    STATX_SIZE);
 		if (ret == -ESTALE)
 			ret = -EIO;
 		if (ret) {
diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c
index a304bf3..8baf545 100644
--- a/fs/orangefs/inode.c
+++ b/fs/orangefs/inode.c
@@ -161,7 +161,7 @@ static int orangefs_setattr_size(struct inode *inode, struct iattr *iattr)
 		     iattr->ia_size);
 
 	/* Ensure that we have a up to date size, so we know if it changed. */
-	ret = orangefs_inode_getattr(inode, 0, 1);
+	ret = orangefs_inode_getattr(inode, 0, 1, STATX_SIZE);
 	if (ret == -ESTALE)
 		ret = -EIO;
 	if (ret) {
@@ -256,13 +256,19 @@ int orangefs_getattr(const struct path *path, struct kstat *stat,
 		     "orangefs_getattr: called on %pd\n",
 		     path->dentry);
 
-	ret = orangefs_inode_getattr(inode, 0, 0);
+	ret = orangefs_inode_getattr(inode, 0, 0, request_mask);
 	if (ret == 0) {
 		generic_fillattr(inode, stat);
 
 		/* override block size reported to stat */
 		orangefs_inode = ORANGEFS_I(inode);
 		stat->blksize = orangefs_inode->blksize;
+
+		if (request_mask & STATX_SIZE)
+			stat->result_mask = STATX_BASIC_STATS;
+		else
+			stat->result_mask = STATX_BASIC_STATS &
+			    ~STATX_SIZE;
 	}
 	return ret;
 }
@@ -277,7 +283,7 @@ int orangefs_permission(struct inode *inode, int mask)
 	gossip_debug(GOSSIP_INODE_DEBUG, "%s: refreshing\n", __func__);
 
 	/* Make sure the permission (and other common attrs) are up to date. */
-	ret = orangefs_inode_getattr(inode, 0, 0);
+	ret = orangefs_inode_getattr(inode, 0, 0, STATX_MODE);
 	if (ret < 0)
 		return ret;
 
@@ -375,7 +381,7 @@ struct inode *orangefs_iget(struct super_block *sb, struct orangefs_object_kref
 	if (!inode || !(inode->i_state & I_NEW))
 		return inode;
 
-	error = orangefs_inode_getattr(inode, 1, 1);
+	error = orangefs_inode_getattr(inode, 1, 1, STATX_ALL);
 	if (error) {
 		iget_failed(inode);
 		return ERR_PTR(error);
@@ -420,7 +426,7 @@ struct inode *orangefs_new_inode(struct super_block *sb, struct inode *dir,
 	orangefs_set_inode(inode, ref);
 	inode->i_ino = hash;	/* needed for stat etc */
 
-	error = orangefs_inode_getattr(inode, 1, 1);
+	error = orangefs_inode_getattr(inode, 1, 1, STATX_ALL);
 	if (error)
 		goto out_iput;
 
diff --git a/fs/orangefs/namei.c b/fs/orangefs/namei.c
index 7c31593..478e88b 100644
--- a/fs/orangefs/namei.c
+++ b/fs/orangefs/namei.c
@@ -74,6 +74,7 @@ static int orangefs_create(struct inode *dir,
 	unlock_new_inode(inode);
 	orangefs_set_timeout(dentry);
 	ORANGEFS_I(inode)->getattr_time = jiffies - 1;
+	ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS;
 
 	gossip_debug(GOSSIP_NAME_DEBUG,
 		     "%s: dentry instantiated for %pd\n",
@@ -322,6 +323,7 @@ static int orangefs_symlink(struct inode *dir,
 	unlock_new_inode(inode);
 	orangefs_set_timeout(dentry);
 	ORANGEFS_I(inode)->getattr_time = jiffies - 1;
+	ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS;
 
 	gossip_debug(GOSSIP_NAME_DEBUG,
 		     "Inode (Symlink) %pU -> %pd\n",
@@ -386,6 +388,7 @@ static int orangefs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode
 	unlock_new_inode(inode);
 	orangefs_set_timeout(dentry);
 	ORANGEFS_I(inode)->getattr_time = jiffies - 1;
+	ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS;
 
 	gossip_debug(GOSSIP_NAME_DEBUG,
 		     "Inode (Directory) %pU -> %pd\n",
diff --git a/fs/orangefs/orangefs-kernel.h b/fs/orangefs/orangefs-kernel.h
index d9b050b..ea0ce50 100644
--- a/fs/orangefs/orangefs-kernel.h
+++ b/fs/orangefs/orangefs-kernel.h
@@ -215,6 +215,7 @@ struct orangefs_inode_s {
 	unsigned long pinode_flags;
 
 	unsigned long getattr_time;
+	u32 getattr_mask;
 };
 
 #define P_ATIME_FLAG 0
@@ -495,7 +496,8 @@ int orangefs_inode_setxattr(struct inode *inode,
 			 size_t size,
 			 int flags);
 
-int orangefs_inode_getattr(struct inode *inode, int new, int bypass);
+int orangefs_inode_getattr(struct inode *inode, int new, int bypass,
+    u32 request_mask);
 
 int orangefs_inode_check_changed(struct inode *inode);
 
diff --git a/fs/orangefs/orangefs-utils.c b/fs/orangefs/orangefs-utils.c
index 9b96b99..fcbf4e5 100644
--- a/fs/orangefs/orangefs-utils.c
+++ b/fs/orangefs/orangefs-utils.c
@@ -251,7 +251,8 @@ static int orangefs_inode_is_stale(struct inode *inode, int new,
 	return 0;
 }
 
-int orangefs_inode_getattr(struct inode *inode, int new, int bypass)
+int orangefs_inode_getattr(struct inode *inode, int new, int bypass,
+    u32 request_mask)
 {
 	struct orangefs_inode_s *orangefs_inode = ORANGEFS_I(inode);
 	struct orangefs_kernel_op_s *new_op;
@@ -262,7 +263,13 @@ int orangefs_inode_getattr(struct inode *inode, int new, int bypass)
 	    get_khandle_from_ino(inode));
 
 	if (!new && !bypass) {
-		if (time_before(jiffies, orangefs_inode->getattr_time))
+		/*
+		 * Must have all the attributes in the mask and be within cache
+		 * time.
+		 */
+		if ((request_mask & orangefs_inode->getattr_mask) ==
+		    request_mask &&
+		    time_before(jiffies, orangefs_inode->getattr_time))
 			return 0;
 	}
 
@@ -270,7 +277,15 @@ int orangefs_inode_getattr(struct inode *inode, int new, int bypass)
 	if (!new_op)
 		return -ENOMEM;
 	new_op->upcall.req.getattr.refn = orangefs_inode->refn;
-	new_op->upcall.req.getattr.mask = ORANGEFS_ATTR_SYS_ALL_NOHINT;
+	/*
+	 * Size is the hardest attribute to get.  The incremental cost of any
+	 * other attribute is essentially zero.
+	 */
+	if (request_mask & STATX_SIZE || new)
+		new_op->upcall.req.getattr.mask = ORANGEFS_ATTR_SYS_ALL_NOHINT;
+	else
+		new_op->upcall.req.getattr.mask =
+		    ORANGEFS_ATTR_SYS_ALL_NOHINT & ~ORANGEFS_ATTR_SYS_SIZE;
 
 	ret = service_operation(new_op, __func__,
 	    get_interruptible_flag(inode));
@@ -291,25 +306,29 @@ int orangefs_inode_getattr(struct inode *inode, int new, int bypass)
 	case S_IFREG:
 		inode->i_flags = orangefs_inode_flags(&new_op->
 		    downcall.resp.getattr.attributes);
-		inode_size = (loff_t)new_op->
-		    downcall.resp.getattr.attributes.size;
-		rounded_up_size =
-		    (inode_size + (4096 - (inode_size % 4096)));
-		inode->i_size = inode_size;
-		orangefs_inode->blksize =
-		    new_op->downcall.resp.getattr.attributes.blksize;
-		spin_lock(&inode->i_lock);
-		inode->i_bytes = inode_size;
-		inode->i_blocks =
-		    (unsigned long)(rounded_up_size / 512);
-		spin_unlock(&inode->i_lock);
+		if (request_mask & STATX_SIZE || new) {
+			inode_size = (loff_t)new_op->
+			    downcall.resp.getattr.attributes.size;
+			rounded_up_size =
+			    (inode_size + (4096 - (inode_size % 4096)));
+			inode->i_size = inode_size;
+			orangefs_inode->blksize =
+			    new_op->downcall.resp.getattr.attributes.blksize;
+			spin_lock(&inode->i_lock);
+			inode->i_bytes = inode_size;
+			inode->i_blocks =
+			    (unsigned long)(rounded_up_size / 512);
+			spin_unlock(&inode->i_lock);
+		}
 		break;
 	case S_IFDIR:
-		inode->i_size = PAGE_SIZE;
-		orangefs_inode->blksize = i_blocksize(inode);
-		spin_lock(&inode->i_lock);
-		inode_set_bytes(inode, inode->i_size);
-		spin_unlock(&inode->i_lock);
+		if (request_mask & STATX_SIZE || new) {
+			inode->i_size = PAGE_SIZE;
+			orangefs_inode->blksize = i_blocksize(inode);
+			spin_lock(&inode->i_lock);
+			inode_set_bytes(inode, inode->i_size);
+			spin_unlock(&inode->i_lock);
+		}
 		set_nlink(inode, 1);
 		break;
 	case S_IFLNK:
@@ -349,6 +368,10 @@ int orangefs_inode_getattr(struct inode *inode, int new, int bypass)
 
 	orangefs_inode->getattr_time = jiffies +
 	    orangefs_getattr_timeout_msecs*HZ/1000;
+	if (request_mask & STATX_SIZE || new)
+		orangefs_inode->getattr_mask = STATX_BASIC_STATS;
+	else
+		orangefs_inode->getattr_mask = STATX_BASIC_STATS & ~STATX_SIZE;
 	ret = 0;
 out:
 	op_release(new_op);