From 571dba52a34015a5a7aa5d480a86936878444a6f Mon Sep 17 00:00:00 2001
From: Greg Farnum <gregf@hq.newdream.net>
Date: Fri, 24 Sep 2010 14:56:40 -0700
Subject: [PATCH] ceph: add CEPH_MDS_OP_SETDIRLAYOUT and associated ioctl.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/ioctl.c              | 66 ++++++++++++++++++++++++++++++++++++
 fs/ceph/ioctl.h              |  4 ++-
 include/linux/ceph/ceph_fs.h |  1 +
 3 files changed, 70 insertions(+), 1 deletion(-)

diff --git a/fs/ceph/ioctl.c b/fs/ceph/ioctl.c
index 899578b0c46b..8888c9ba68db 100644
--- a/fs/ceph/ioctl.c
+++ b/fs/ceph/ioctl.c
@@ -91,6 +91,68 @@ static long ceph_ioctl_set_layout(struct file *file, void __user *arg)
 	return err;
 }
 
+/*
+ * Set a layout policy on a directory inode. All items in the tree
+ * rooted at this inode will inherit this layout on creation,
+ * (It doesn't apply retroactively )
+ * unless a subdirectory has its own layout policy.
+ */
+static long ceph_ioctl_set_layout_policy (struct file *file, void __user *arg)
+{
+	struct inode *inode = file->f_dentry->d_inode;
+	struct ceph_mds_request *req;
+	struct ceph_ioctl_layout l;
+	int err, i;
+	struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc;
+
+	/* copy and validate */
+	if (copy_from_user(&l, arg, sizeof(l)))
+		return -EFAULT;
+
+	if ((l.object_size & ~PAGE_MASK) ||
+	    (l.stripe_unit & ~PAGE_MASK) ||
+	    !l.stripe_unit ||
+	    (l.object_size &&
+	        (unsigned)l.object_size % (unsigned)l.stripe_unit))
+		return -EINVAL;
+
+	/* make sure it's a valid data pool */
+	if (l.data_pool > 0) {
+		mutex_lock(&mdsc->mutex);
+		err = -EINVAL;
+		for (i = 0; i < mdsc->mdsmap->m_num_data_pg_pools; i++)
+			if (mdsc->mdsmap->m_data_pg_pools[i] == l.data_pool) {
+				err = 0;
+				break;
+			}
+		mutex_unlock(&mdsc->mutex);
+		if (err)
+			return err;
+	}
+
+	req = ceph_mdsc_create_request(mdsc, CEPH_MDS_OP_SETDIRLAYOUT,
+				       USE_AUTH_MDS);
+
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+	req->r_inode = igrab(inode);
+
+	req->r_args.setlayout.layout.fl_stripe_unit =
+			cpu_to_le32(l.stripe_unit);
+	req->r_args.setlayout.layout.fl_stripe_count =
+			cpu_to_le32(l.stripe_count);
+	req->r_args.setlayout.layout.fl_object_size =
+			cpu_to_le32(l.object_size);
+	req->r_args.setlayout.layout.fl_pg_pool =
+			cpu_to_le32(l.data_pool);
+	req->r_args.setlayout.layout.fl_pg_preferred =
+			cpu_to_le32(l.preferred_osd);
+
+	err = ceph_mdsc_do_request(mdsc, inode, req);
+	ceph_mdsc_put_request(req);
+	return err;
+}
+
 /*
  * Return object name, size/offset information, and location (OSD
  * number, network address) for a given file offset.
@@ -177,11 +239,15 @@ long ceph_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case CEPH_IOC_SET_LAYOUT:
 		return ceph_ioctl_set_layout(file, (void __user *)arg);
 
+	case CEPH_IOC_SET_LAYOUT_POLICY:
+		return ceph_ioctl_set_layout_policy(file, (void __user *)arg);
+
 	case CEPH_IOC_GET_DATALOC:
 		return ceph_ioctl_get_dataloc(file, (void __user *)arg);
 
 	case CEPH_IOC_LAZYIO:
 		return ceph_ioctl_lazyio(file);
 	}
+
 	return -ENOTTY;
 }
diff --git a/fs/ceph/ioctl.h b/fs/ceph/ioctl.h
index 88451a3b6857..a6ce54e94eb5 100644
--- a/fs/ceph/ioctl.h
+++ b/fs/ceph/ioctl.h
@@ -4,7 +4,7 @@
 #include <linux/ioctl.h>
 #include <linux/types.h>
 
-#define CEPH_IOCTL_MAGIC 0x97
+#define CEPH_IOCTL_MAGIC 0x98
 
 /* just use u64 to align sanely on all archs */
 struct ceph_ioctl_layout {
@@ -17,6 +17,8 @@ struct ceph_ioctl_layout {
 				   struct ceph_ioctl_layout)
 #define CEPH_IOC_SET_LAYOUT _IOW(CEPH_IOCTL_MAGIC, 2,		\
 				   struct ceph_ioctl_layout)
+#define CEPH_IOC_SET_LAYOUT_POLICY _IOW(CEPH_IOCTL_MAGIC, 5,	\
+				   struct ceph_ioctl_layout)
 
 /*
  * Extract identity, address of the OSD and object storing a given
diff --git a/include/linux/ceph/ceph_fs.h b/include/linux/ceph/ceph_fs.h
index d5619ac86711..c3c74aef289d 100644
--- a/include/linux/ceph/ceph_fs.h
+++ b/include/linux/ceph/ceph_fs.h
@@ -299,6 +299,7 @@ enum {
 	CEPH_MDS_OP_SETATTR    = 0x01108,
 	CEPH_MDS_OP_SETFILELOCK= 0x01109,
 	CEPH_MDS_OP_GETFILELOCK= 0x00110,
+	CEPH_MDS_OP_SETDIRLAYOUT=0x0110a,
 
 	CEPH_MDS_OP_MKNOD      = 0x01201,
 	CEPH_MDS_OP_LINK       = 0x01202,