drbd: Introduce protocol version 100 headers

The 8 byte header finally becomes too small. With the protocol 100 header we
have 16 bit for the volume number, proper 32 bit for the data length, and
32 bit for further extensions in the future.

Previous versions of drbd are using version 80 headers for all packets
short enough for protocol 80.  They support both header versions in
worker context, but only version 80 headers in asynchronous context.
For backwards compatibility, continue to use version 80 headers for
short packets before protocol version 100.

From protocol version 100 on, use the same header version for all
packets.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>

drivers/block/drbd/drbd_int.h

@@ -307,6 +307,14 @@ struct p_header95 {
 	u32	  length;	/* Use only 24 bits of that. Ignore the highest 8 bit. */
 } __packed;
 
+struct p_header100 {
+	u32	  magic;
+	u16	  volume;
+	u16	  command;
+	u32	  length;
+	u32	  pad;
+} __packed;
+
 extern unsigned int drbd_header_size(struct drbd_tconn *tconn);
 
 /* these defines must not be changed without changing the protocol version */

drivers/block/drbd/drbd_main.c

@@ -698,9 +698,15 @@ void drbd_thread_current_set_cpu(struct drbd_thread *thi)
  */
 unsigned int drbd_header_size(struct drbd_tconn *tconn)
 {
-	BUILD_BUG_ON(sizeof(struct p_header80) != sizeof(struct p_header95));
+	if (tconn->agreed_pro_version >= 100) {
+		BUILD_BUG_ON(!IS_ALIGNED(sizeof(struct p_header100), 8));
+		return sizeof(struct p_header100);
+	} else {
+		BUILD_BUG_ON(sizeof(struct p_header80) !=
+			     sizeof(struct p_header95));
 		BUILD_BUG_ON(!IS_ALIGNED(sizeof(struct p_header80), 8));
 		return sizeof(struct p_header80);
+	}
 }
 
 static unsigned int prepare_header80(struct p_header80 *h, enum drbd_packet cmd, int size)
@@ -719,10 +725,24 @@ static unsigned int prepare_header95(struct p_header95 *h, enum drbd_packet cmd,
 	return sizeof(struct p_header95);
 }
 
-static unsigned int prepare_header(struct drbd_tconn *tconn, int vnr, void *buffer,
-				   enum drbd_packet cmd, int size)
+static unsigned int prepare_header100(struct p_header100 *h, enum drbd_packet cmd,
+				      int size, int vnr)
+{
+	h->magic = cpu_to_be32(DRBD_MAGIC_100);
+	h->volume = cpu_to_be16(vnr);
+	h->command = cpu_to_be16(cmd);
+	h->length = cpu_to_be32(size);
+	h->pad = 0;
+	return sizeof(struct p_header100);
+}
+
+static unsigned int prepare_header(struct drbd_tconn *tconn, int vnr,
+				   void *buffer, enum drbd_packet cmd, int size)
 {
-	if (tconn->agreed_pro_version >= 95)
+	if (tconn->agreed_pro_version >= 100)
+		return prepare_header100(buffer, cmd, size, vnr);
+	else if (tconn->agreed_pro_version >= 95 &&
+		 size > DRBD_MAX_SIZE_H80_PACKET)
 		return prepare_header95(buffer, cmd, size);
 	else
 		return prepare_header80(buffer, cmd, size);

drivers/block/drbd/drbd_nl.c

@@ -2833,8 +2833,7 @@ int drbd_adm_add_minor(struct sk_buff *skb, struct genl_info *info)
 		retcode = ERR_INVALID_REQUEST;
 		goto out;
 	}
-	/* FIXME we need a define here */
-	if (adm_ctx.volume >= 256) {
+	if (adm_ctx.volume > DRBD_VOLUME_MAX) {
 		drbd_msg_put_info("requested volume id out of range");
 		retcode = ERR_INVALID_REQUEST;
 		goto out;

drivers/block/drbd/drbd_receiver.c

@@ -983,7 +983,17 @@ static int decode_header(struct drbd_tconn *tconn, void *header, struct packet_i
 {
 	unsigned int header_size = drbd_header_size(tconn);
 
-	if (header_size == sizeof(struct p_header95) &&
+	if (header_size == sizeof(struct p_header100) &&
+	    *(__be32 *)header == cpu_to_be32(DRBD_MAGIC_100)) {
+		struct p_header100 *h = header;
+		if (h->pad != 0) {
+			conn_err(tconn, "Header padding is not zero\n");
+			return -EINVAL;
+		}
+		pi->vnr = be16_to_cpu(h->volume);
+		pi->cmd = be16_to_cpu(h->command);
+		pi->size = be32_to_cpu(h->length);
+	} else if (header_size == sizeof(struct p_header95) &&
 		   *(__be16 *)header == cpu_to_be16(DRBD_MAGIC_BIG)) {
 		struct p_header95 *h = header;
 

include/linux/drbd.h

@@ -341,6 +341,7 @@ enum drbd_timeout_flag {
 
 #define DRBD_MAGIC 0x83740267
 #define DRBD_MAGIC_BIG 0x835a
+#define DRBD_MAGIC_100 0x8620ec20
 
 /* how I came up with this magic?
  * base64 decode "actlog==" ;) */

include/linux/drbd_genl.h

@@ -95,8 +95,6 @@ GENL_struct(DRBD_NLA_CFG_REPLY, 1, drbd_cfg_reply,
  * and/or the replication group (aka resource) name,
  * and the volume id within the resource. */
 GENL_struct(DRBD_NLA_CFG_CONTEXT, 2, drbd_cfg_context,
-		/* currently only 256 volumes per group,
-		 * but maybe we still change that */
 	__u32_field(1, GENLA_F_MANDATORY,	ctx_volume)
 	__str_field(2, GENLA_F_MANDATORY,	ctx_conn_name, 128)
 )

include/linux/drbd_limits.h

@@ -19,6 +19,8 @@
 #define DRBD_MINOR_COUNT_MAX 256
 #define DRBD_MINOR_COUNT_DEF 32
 
+#define DRBD_VOLUME_MAX 65535
+
 #define DRBD_DIALOG_REFRESH_MIN 0
 #define DRBD_DIALOG_REFRESH_MAX 600