Ocfs2: Add new OCFS2_IOC_INFO ioctl for ocfs2 v8.

The reason why we need this ioctl is to offer the none-privileged
end-user a possibility to get filesys info gathering.

We use OCFS2_IOC_INFO to manipulate the new ioctl, userspace passes a
structure to kernel containing an array of request pointers and request
count, such as,

* From userspace:

struct ocfs2_info_blocksize oib = {
        .ib_req = {
                .ir_magic = OCFS2_INFO_MAGIC,
                .ir_code = OCFS2_INFO_BLOCKSIZE,
                ...
        }
        ...
}

struct ocfs2_info_clustersize oic = {
        ...
}

uint64_t reqs[2] = {(unsigned long)&oib,
                    (unsigned long)&oic};

struct ocfs2_info info = {
        .oi_requests = reqs,
        .oi_count = 2,
}

ret = ioctl(fd, OCFS2_IOC_INFO, &info);

* In kernel:

Get the request pointers from *info*, then handle each request one bye one.

Idea here is to make the spearated request small enough to guarantee
a better backward&forward compatibility since a small piece of request
would be less likely to be broken if filesys on raw disk get changed.

Currently, the following 7 requests are supported per the requirement from
userspace tool o2info, and I believe it will grow over time:-)

        OCFS2_INFO_CLUSTERSIZE
        OCFS2_INFO_BLOCKSIZE
        OCFS2_INFO_MAXSLOTS
        OCFS2_INFO_LABEL
        OCFS2_INFO_UUID
        OCFS2_INFO_FS_FEATURES
        OCFS2_INFO_JOURNAL_SIZE

This ioctl is only specific to OCFS2.
Signed-off-by: NTristan Ye <tristan.ye@oracle.com>
Signed-off-by: NJoel Becker <joel.becker@oracle.com>

fs/ocfs2/ioctl.c

@@ -26,6 +26,26 @@
 
 #include <linux/ext2_fs.h>
 
+#define o2info_from_user(a, b)	\
+		copy_from_user(&(a), (b), sizeof(a))
+#define o2info_to_user(a, b)	\
+		copy_to_user((typeof(a) __user *)b, &(a), sizeof(a))
+
+/*
+ * This call is void because we are already reporting an error that may
+ * be -EFAULT.  The error will be returned from the ioctl(2) call.  It's
+ * just a best-effort to tell userspace that this request caused the error.
+ */
+static inline void __o2info_set_request_error(struct ocfs2_info_request *kreq,
+					struct ocfs2_info_request __user *req)
+{
+	kreq->ir_flags |= OCFS2_INFO_FL_ERROR;
+	(void)put_user(kreq->ir_flags, (__u32 __user *)&(req->ir_flags));
+}
+
+#define o2info_set_request_error(a, b) \
+		__o2info_set_request_error((struct ocfs2_info_request *)&(a), b)
+
 static int ocfs2_get_inode_attr(struct inode *inode, unsigned *flags)
 {
 	int status;
@@ -109,6 +129,328 @@ static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags,
 	return status;
 }
 
+int ocfs2_info_handle_blocksize(struct inode *inode,
+				struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_blocksize oib;
+
+	if (o2info_from_user(oib, req))
+		goto bail;
+
+	oib.ib_blocksize = inode->i_sb->s_blocksize;
+	oib.ib_req.ir_flags |= OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oib, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oib, req);
+
+	return status;
+}
+
+int ocfs2_info_handle_clustersize(struct inode *inode,
+				  struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_clustersize oic;
+	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+
+	if (o2info_from_user(oic, req))
+		goto bail;
+
+	oic.ic_clustersize = osb->s_clustersize;
+	oic.ic_req.ir_flags |= OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oic, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oic, req);
+
+	return status;
+}
+
+int ocfs2_info_handle_maxslots(struct inode *inode,
+			       struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_maxslots oim;
+	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+
+	if (o2info_from_user(oim, req))
+		goto bail;
+
+	oim.im_max_slots = osb->max_slots;
+	oim.im_req.ir_flags |= OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oim, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oim, req);
+
+	return status;
+}
+
+int ocfs2_info_handle_label(struct inode *inode,
+			    struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_label oil;
+	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+
+	if (o2info_from_user(oil, req))
+		goto bail;
+
+	memcpy(oil.il_label, osb->vol_label, OCFS2_MAX_VOL_LABEL_LEN);
+	oil.il_req.ir_flags |= OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oil, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oil, req);
+
+	return status;
+}
+
+int ocfs2_info_handle_uuid(struct inode *inode,
+			   struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_uuid oiu;
+	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+
+	if (o2info_from_user(oiu, req))
+		goto bail;
+
+	memcpy(oiu.iu_uuid_str, osb->uuid_str, OCFS2_TEXT_UUID_LEN + 1);
+	oiu.iu_req.ir_flags |= OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oiu, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oiu, req);
+
+	return status;
+}
+
+int ocfs2_info_handle_fs_features(struct inode *inode,
+				  struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_fs_features oif;
+	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+
+	if (o2info_from_user(oif, req))
+		goto bail;
+
+	oif.if_compat_features = osb->s_feature_compat;
+	oif.if_incompat_features = osb->s_feature_incompat;
+	oif.if_ro_compat_features = osb->s_feature_ro_compat;
+	oif.if_req.ir_flags |= OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oif, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oif, req);
+
+	return status;
+}
+
+int ocfs2_info_handle_journal_size(struct inode *inode,
+				   struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_journal_size oij;
+	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
+
+	if (o2info_from_user(oij, req))
+		goto bail;
+
+	oij.ij_journal_size = osb->journal->j_inode->i_size;
+
+	oij.ij_req.ir_flags |= OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oij, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oij, req);
+
+	return status;
+}
+
+int ocfs2_info_handle_unknown(struct inode *inode,
+			      struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_request oir;
+
+	if (o2info_from_user(oir, req))
+		goto bail;
+
+	oir.ir_flags &= ~OCFS2_INFO_FL_FILLED;
+
+	if (o2info_to_user(oir, req))
+		goto bail;
+
+	status = 0;
+bail:
+	if (status)
+		o2info_set_request_error(oir, req);
+
+	return status;
+}
+
+/*
+ * Validate and distinguish OCFS2_IOC_INFO requests.
+ *
+ * - validate the magic number.
+ * - distinguish different requests.
+ * - validate size of different requests.
+ */
+int ocfs2_info_handle_request(struct inode *inode,
+			      struct ocfs2_info_request __user *req)
+{
+	int status = -EFAULT;
+	struct ocfs2_info_request oir;
+
+	if (o2info_from_user(oir, req))
+		goto bail;
+
+	status = -EINVAL;
+	if (oir.ir_magic != OCFS2_INFO_MAGIC)
+		goto bail;
+
+	switch (oir.ir_code) {
+	case OCFS2_INFO_BLOCKSIZE:
+		if (oir.ir_size == sizeof(struct ocfs2_info_blocksize))
+			status = ocfs2_info_handle_blocksize(inode, req);
+		break;
+	case OCFS2_INFO_CLUSTERSIZE:
+		if (oir.ir_size == sizeof(struct ocfs2_info_clustersize))
+			status = ocfs2_info_handle_clustersize(inode, req);
+		break;
+	case OCFS2_INFO_MAXSLOTS:
+		if (oir.ir_size == sizeof(struct ocfs2_info_maxslots))
+			status = ocfs2_info_handle_maxslots(inode, req);
+		break;
+	case OCFS2_INFO_LABEL:
+		if (oir.ir_size == sizeof(struct ocfs2_info_label))
+			status = ocfs2_info_handle_label(inode, req);
+		break;
+	case OCFS2_INFO_UUID:
+		if (oir.ir_size == sizeof(struct ocfs2_info_uuid))
+			status = ocfs2_info_handle_uuid(inode, req);
+		break;
+	case OCFS2_INFO_FS_FEATURES:
+		if (oir.ir_size == sizeof(struct ocfs2_info_fs_features))
+			status = ocfs2_info_handle_fs_features(inode, req);
+		break;
+	case OCFS2_INFO_JOURNAL_SIZE:
+		if (oir.ir_size == sizeof(struct ocfs2_info_journal_size))
+			status = ocfs2_info_handle_journal_size(inode, req);
+		break;
+	default:
+		status = ocfs2_info_handle_unknown(inode, req);
+		break;
+	}
+
+bail:
+	return status;
+}
+
+int ocfs2_get_request_ptr(struct ocfs2_info *info, int idx,
+			  u64 *req_addr, int compat_flag)
+{
+	int status = -EFAULT;
+	u64 __user *bp = NULL;
+
+	if (compat_flag) {
+#ifdef CONFIG_COMPAT
+		/*
+		 * pointer bp stores the base address of a pointers array,
+		 * which collects all addresses of separate request.
+		 */
+		bp = (u64 __user *)(unsigned long)compat_ptr(info->oi_requests);
+#else
+		BUG();
+#endif
+	} else
+		bp = (u64 __user *)(unsigned long)(info->oi_requests);
+
+	if (o2info_from_user(*req_addr, bp + idx))
+		goto bail;
+
+	status = 0;
+bail:
+	return status;
+}
+
+/*
+ * OCFS2_IOC_INFO handles an array of requests passed from userspace.
+ *
+ * ocfs2_info_handle() recevies a large info aggregation, grab and
+ * validate the request count from header, then break it into small
+ * pieces, later specific handlers can handle them one by one.
+ *
+ * Idea here is to make each separate request small enough to ensure
+ * a better backward&forward compatibility, since a small piece of
+ * request will be less likely to be broken if disk layout get changed.
+ */
+int ocfs2_info_handle(struct inode *inode, struct ocfs2_info *info,
+		      int compat_flag)
+{
+	int i, status = 0;
+	u64 req_addr;
+	struct ocfs2_info_request __user *reqp;
+
+	if ((info->oi_count > OCFS2_INFO_MAX_REQUEST) ||
+	    (!info->oi_requests)) {
+		status = -EINVAL;
+		goto bail;
+	}
+
+	for (i = 0; i < info->oi_count; i++) {
+
+		status = ocfs2_get_request_ptr(info, i, &req_addr, compat_flag);
+		if (status)
+			break;
+
+		reqp = (struct ocfs2_info_request *)(unsigned long)req_addr;
+		if (!reqp) {
+			status = -EINVAL;
+			goto bail;
+		}
+
+		status = ocfs2_info_handle_request(inode, reqp);
+		if (status)
+			break;
+	}
+
+bail:
+	return status;
+}
+
 long ocfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	struct inode *inode = filp->f_path.dentry->d_inode;
@@ -120,6 +462,7 @@ long ocfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	struct reflink_arguments args;
 	const char *old_path, *new_path;
 	bool preserve;
+	struct ocfs2_info info;
 
 	switch (cmd) {
 	case OCFS2_IOC_GETFLAGS:
@@ -174,6 +517,12 @@ long ocfs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		preserve = (args.preserve != 0);
 
 		return ocfs2_reflink_ioctl(inode, old_path, new_path, preserve);
+	case OCFS2_IOC_INFO:
+		if (copy_from_user(&info, (struct ocfs2_info __user *)arg,
+				   sizeof(struct ocfs2_info)))
+			return -EFAULT;
+
+		return ocfs2_info_handle(inode, &info, 0);
 	default:
 		return -ENOTTY;
 	}
@@ -185,6 +534,7 @@ long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 	bool preserve;
 	struct reflink_arguments args;
 	struct inode *inode = file->f_path.dentry->d_inode;
+	struct ocfs2_info info;
 
 	switch (cmd) {
 	case OCFS2_IOC32_GETFLAGS:
@@ -209,6 +559,12 @@ long ocfs2_compat_ioctl(struct file *file, unsigned cmd, unsigned long arg)
 
 		return ocfs2_reflink_ioctl(inode, compat_ptr(args.old_path),
 					   compat_ptr(args.new_path), preserve);
+	case OCFS2_IOC_INFO:
+		if (copy_from_user(&info, (struct ocfs2_info __user *)arg,
+				   sizeof(struct ocfs2_info)))
+			return -EFAULT;
+
+		return ocfs2_info_handle(inode, &info, 1);
 	default:
 		return -ENOIOCTLCMD;
 	}

fs/ocfs2/ocfs2_ioctl.h

@@ -76,4 +76,99 @@ struct reflink_arguments {
 };
 #define OCFS2_IOC_REFLINK	_IOW('o', 4, struct reflink_arguments)
 
+/* Following definitions dedicated for ocfs2_info_request ioctls. */
+#define OCFS2_INFO_MAX_REQUEST		(50)
+#define OCFS2_TEXT_UUID_LEN		(OCFS2_VOL_UUID_LEN * 2)
+
+/* Magic number of all requests */
+#define OCFS2_INFO_MAGIC		(0x4F32494E)
+
+/*
+ * Always try to separate info request into small pieces to
+ * guarantee the backward&forward compatibility.
+ */
+struct ocfs2_info {
+	__u64 oi_requests;	/* Array of __u64 pointers to requests */
+	__u32 oi_count;		/* Number of requests in info_requests */
+	__u32 oi_pad;
+};
+
+struct ocfs2_info_request {
+/*00*/	__u32 ir_magic;	/* Magic number */
+	__u32 ir_code;	/* Info request code */
+	__u32 ir_size;	/* Size of request */
+	__u32 ir_flags;	/* Request flags */
+/*10*/			/* Request specific fields */
+};
+
+struct ocfs2_info_clustersize {
+	struct ocfs2_info_request ic_req;
+	__u32 ic_clustersize;
+	__u32 ic_pad;
+};
+
+struct ocfs2_info_blocksize {
+	struct ocfs2_info_request ib_req;
+	__u32 ib_blocksize;
+	__u32 ib_pad;
+};
+
+struct ocfs2_info_maxslots {
+	struct ocfs2_info_request im_req;
+	__u32 im_max_slots;
+	__u32 im_pad;
+};
+
+struct ocfs2_info_label {
+	struct ocfs2_info_request il_req;
+	__u8	il_label[OCFS2_MAX_VOL_LABEL_LEN];
+} __attribute__ ((packed));
+
+struct ocfs2_info_uuid {
+	struct ocfs2_info_request iu_req;
+	__u8	iu_uuid_str[OCFS2_TEXT_UUID_LEN + 1];
+} __attribute__ ((packed));
+
+struct ocfs2_info_fs_features {
+	struct ocfs2_info_request if_req;
+	__u32 if_compat_features;
+	__u32 if_incompat_features;
+	__u32 if_ro_compat_features;
+	__u32 if_pad;
+};
+
+struct ocfs2_info_journal_size {
+	struct ocfs2_info_request ij_req;
+	__u64 ij_journal_size;
+};
+
+/* Codes for ocfs2_info_request */
+enum ocfs2_info_type {
+	OCFS2_INFO_CLUSTERSIZE = 1,
+	OCFS2_INFO_BLOCKSIZE,
+	OCFS2_INFO_MAXSLOTS,
+	OCFS2_INFO_LABEL,
+	OCFS2_INFO_UUID,
+	OCFS2_INFO_FS_FEATURES,
+	OCFS2_INFO_JOURNAL_SIZE,
+	OCFS2_INFO_NUM_TYPES
+};
+
+/* Flags for struct ocfs2_info_request */
+/* Filled by the caller */
+#define OCFS2_INFO_FL_NON_COHERENT	(0x00000001)	/* Cluster coherency not
+							   required. This is a hint.
+							   It is up to ocfs2 whether
+							   the request can be fulfilled
+							   without locking. */
+/* Filled by ocfs2 */
+#define OCFS2_INFO_FL_FILLED		(0x40000000)	/* Filesystem understood
+							   this request and
+							   filled in the answer */
+
+#define OCFS2_INFO_FL_ERROR		(0x80000000)	/* Error happened during
+							   request handling. */
+
+#define OCFS2_IOC_INFO		_IOR('o', 5, struct ocfs2_info)
+
 #endif /* OCFS2_IOCTL_H */