From 2f2dc053404febedc9c273452d9d518fb31fde72 Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@newdream.net>
Date: Tue, 6 Oct 2009 11:31:09 -0700
Subject: ceph: MDS client

The MDS (metadata server) client is responsible for submitting
requests to the MDS cluster and parsing the response.  We decide which
MDS to submit each request to based on cached information about the
current partition of the directory hierarchy across the cluster.  A
stateful session is opened with each MDS before we submit requests to
it, and a mutex is used to control the ordering of messages within
each session.

An MDS request may generate two responses.  The first indicates the
operation was a success and returns any result.  A second reply is
sent when the operation commits to disk.  Note that locking on the MDS
ensures that the results of updates are visible only to the updating
client before the operation commits.  Requests are linked to the
containing directory so that an fsync will wait for them to commit.

If an MDS fails and/or recovers, we resubmit requests as needed.  We
also reconnect existing capabilities to a recovering MDS to
reestablish that shared session state.  Old dentry leases are
invalidated.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/mdsmap.c | 166 +++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 166 insertions(+)
 create mode 100644 fs/ceph/mdsmap.c

(limited to 'fs/ceph/mdsmap.c')

diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
new file mode 100644
index 000000000000..15913cbeb289
--- /dev/null
+++ b/fs/ceph/mdsmap.c
@@ -0,0 +1,166 @@
+#include "ceph_debug.h"
+
+#include <linux/bug.h>
+#include <linux/err.h>
+#include <linux/random.h>
+#include <linux/slab.h>
+#include <linux/types.h>
+
+#include "mdsmap.h"
+#include "messenger.h"
+#include "decode.h"
+
+#include "super.h"
+
+
+/*
+ * choose a random mds that is "up" (i.e. has a state > 0), or -1.
+ */
+int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m)
+{
+	int n = 0;
+	int i;
+	char r;
+
+	/* count */
+	for (i = 0; i < m->m_max_mds; i++)
+		if (m->m_info[i].state > 0)
+			n++;
+	if (n == 0)
+		return -1;
+
+	/* pick */
+	get_random_bytes(&r, 1);
+	n = r % n;
+	i = 0;
+	for (i = 0; n > 0; i++, n--)
+		while (m->m_info[i].state <= 0)
+			i++;
+
+	return i;
+}
+
+/*
+ * Decode an MDS map
+ *
+ * Ignore any fields we don't care about (there are quite a few of
+ * them).
+ */
+struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
+{
+	struct ceph_mdsmap *m;
+	int i, j, n;
+	int err = -EINVAL;
+	u16 version;
+
+	m = kzalloc(sizeof(*m), GFP_NOFS);
+	if (m == NULL)
+		return ERR_PTR(-ENOMEM);
+
+	ceph_decode_16_safe(p, end, version, bad);
+
+	ceph_decode_need(p, end, 8*sizeof(u32) + sizeof(u64), bad);
+	ceph_decode_32(p, m->m_epoch);
+	ceph_decode_32(p, m->m_client_epoch);
+	ceph_decode_32(p, m->m_last_failure);
+	ceph_decode_32(p, m->m_root);
+	ceph_decode_32(p, m->m_session_timeout);
+	ceph_decode_32(p, m->m_session_autoclose);
+	ceph_decode_64(p, m->m_max_file_size);
+	ceph_decode_32(p, m->m_max_mds);
+
+	m->m_info = kcalloc(m->m_max_mds, sizeof(*m->m_info), GFP_NOFS);
+	if (m->m_info == NULL)
+		goto badmem;
+
+	/* pick out active nodes from mds_info (state > 0) */
+	ceph_decode_32(p, n);
+	for (i = 0; i < n; i++) {
+		u32 namelen;
+		s32 mds, inc, state;
+		u64 state_seq;
+		u8 infoversion;
+		struct ceph_entity_addr addr;
+		u32 num_export_targets;
+		void *pexport_targets = NULL;
+
+		ceph_decode_need(p, end, sizeof(addr) + 1 + sizeof(u32), bad);
+		*p += sizeof(addr);          /* skip addr key */
+		ceph_decode_8(p, infoversion);
+		ceph_decode_32(p, namelen);  /* skip mds name */
+		*p += namelen;
+
+		ceph_decode_need(p, end,
+				 5*sizeof(u32) + sizeof(u64) +
+				 sizeof(addr) + sizeof(struct ceph_timespec),
+				 bad);
+		ceph_decode_32(p, mds);
+		ceph_decode_32(p, inc);
+		ceph_decode_32(p, state);
+		ceph_decode_64(p, state_seq);
+		ceph_decode_copy(p, &addr, sizeof(addr));
+		*p += sizeof(struct ceph_timespec);
+		*p += sizeof(u32);
+		ceph_decode_32_safe(p, end, namelen, bad);
+		*p += sizeof(namelen);
+		if (infoversion >= 2) {
+			ceph_decode_32_safe(p, end, num_export_targets, bad);
+			pexport_targets = *p;
+			*p += sizeof(num_export_targets * sizeof(u32));
+		} else {
+			num_export_targets = 0;
+		}
+
+		dout("mdsmap_decode %d/%d mds%d.%d %s %s\n",
+		     i+1, n, mds, inc, pr_addr(&addr.in_addr),
+		     ceph_mds_state_name(state));
+		if (mds >= 0 && mds < m->m_max_mds && state > 0) {
+			m->m_info[mds].state = state;
+			m->m_info[mds].addr = addr;
+			m->m_info[mds].num_export_targets = num_export_targets;
+			if (num_export_targets) {
+				m->m_info[mds].export_targets =
+					kcalloc(num_export_targets, sizeof(u32),
+						GFP_NOFS);
+				for (j = 0; j < num_export_targets; j++)
+					ceph_decode_32(&pexport_targets,
+					      m->m_info[mds].export_targets[j]);
+			} else {
+				m->m_info[mds].export_targets = NULL;
+			}
+		}
+	}
+
+	/* pg_pools */
+	ceph_decode_32_safe(p, end, n, bad);
+	m->m_num_data_pg_pools = n;
+	m->m_data_pg_pools = kcalloc(n, sizeof(u32), GFP_NOFS);
+	if (!m->m_data_pg_pools)
+		goto badmem;
+	ceph_decode_need(p, end, sizeof(u32)*(n+1), bad);
+	for (i = 0; i < n; i++)
+		ceph_decode_32(p, m->m_data_pg_pools[i]);
+	ceph_decode_32(p, m->m_cas_pg_pool);
+
+	/* ok, we don't care about the rest. */
+	dout("mdsmap_decode success epoch %u\n", m->m_epoch);
+	return m;
+
+badmem:
+	err = -ENOMEM;
+bad:
+	pr_err("corrupt mdsmap\n");
+	ceph_mdsmap_destroy(m);
+	return ERR_PTR(-EINVAL);
+}
+
+void ceph_mdsmap_destroy(struct ceph_mdsmap *m)
+{
+	int i;
+
+	for (i = 0; i < m->m_max_mds; i++)
+		kfree(m->m_info[i].export_targets);
+	kfree(m->m_info);
+	kfree(m->m_data_pg_pools);
+	kfree(m);
+}
-- 
cgit v1.2.3


From e251e288082d5e89604eee1fef0c31bed1fe8f02 Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@newdream.net>
Date: Wed, 7 Oct 2009 16:38:19 -0700
Subject: ceph: fix mdsmap decoding when multiple mds's are present

A misplaced sizeof() around namelen was throwing things off.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/mdsmap.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

(limited to 'fs/ceph/mdsmap.c')

diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 15913cbeb289..09180d8fafe4 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -85,28 +85,28 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		void *pexport_targets = NULL;
 
 		ceph_decode_need(p, end, sizeof(addr) + 1 + sizeof(u32), bad);
-		*p += sizeof(addr);          /* skip addr key */
+		ceph_decode_copy(p, &addr, sizeof(addr));
 		ceph_decode_8(p, infoversion);
 		ceph_decode_32(p, namelen);  /* skip mds name */
 		*p += namelen;
 
 		ceph_decode_need(p, end,
-				 5*sizeof(u32) + sizeof(u64) +
+				 4*sizeof(u32) + sizeof(u64) +
 				 sizeof(addr) + sizeof(struct ceph_timespec),
 				 bad);
 		ceph_decode_32(p, mds);
 		ceph_decode_32(p, inc);
 		ceph_decode_32(p, state);
 		ceph_decode_64(p, state_seq);
-		ceph_decode_copy(p, &addr, sizeof(addr));
+		*p += sizeof(addr);
 		*p += sizeof(struct ceph_timespec);
 		*p += sizeof(u32);
 		ceph_decode_32_safe(p, end, namelen, bad);
-		*p += sizeof(namelen);
+		*p += namelen;
 		if (infoversion >= 2) {
 			ceph_decode_32_safe(p, end, num_export_targets, bad);
 			pexport_targets = *p;
-			*p += sizeof(num_export_targets * sizeof(u32));
+			*p += num_export_targets * sizeof(u32);
 		} else {
 			num_export_targets = 0;
 		}
-- 
cgit v1.2.3


From c89136ea4253c73e89e97f5138bb22d97ad9f564 Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@newdream.net>
Date: Wed, 14 Oct 2009 09:59:09 -0700
Subject: ceph: convert encode/decode macros to inlines

This avoids the fugly pass by reference and makes the code a bit easier
to read.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/decode.h     | 129 ++++++++++++++++++++++++++-------------------------
 fs/ceph/mds_client.c |  20 ++++----
 fs/ceph/mdsmap.c     |  38 +++++++--------
 fs/ceph/mon_client.c |   4 +-
 fs/ceph/osd_client.c |   8 ++--
 fs/ceph/osdmap.c     |  70 ++++++++++++++--------------
 6 files changed, 137 insertions(+), 132 deletions(-)

(limited to 'fs/ceph/mdsmap.c')

diff --git a/fs/ceph/decode.h b/fs/ceph/decode.h
index fc2769df062d..91179fb2cc3f 100644
--- a/fs/ceph/decode.h
+++ b/fs/ceph/decode.h
@@ -3,12 +3,44 @@
 
 #include <asm/unaligned.h>
 
+#include "types.h"
+
 /*
  * in all cases,
  *   void **p     pointer to position pointer
  *   void *end    pointer to end of buffer (last byte + 1)
  */
 
+static inline u64 ceph_decode_64(void **p)
+{
+	u64 v = get_unaligned_le64(*p);
+	*p += sizeof(u64);
+	return v;
+}
+static inline u32 ceph_decode_32(void **p)
+{
+	u32 v = get_unaligned_le32(*p);
+	*p += sizeof(u32);
+	return v;
+}
+static inline u16 ceph_decode_16(void **p)
+{
+	u16 v = get_unaligned_le16(*p);
+	*p += sizeof(u16);
+	return v;
+}
+static inline u8 ceph_decode_8(void **p)
+{
+	u8 v = *(u8 *)*p;
+	(*p)++;
+	return v;
+}
+static inline void ceph_decode_copy(void **p, void *pv, size_t n)
+{
+	memcpy(pv, *p, n);
+	*p += n;
+}
+
 /*
  * bounds check input.
  */
@@ -18,48 +50,20 @@
 			goto bad;			\
 	} while (0)
 
-#define ceph_decode_64(p, v)					\
-	do {							\
-		v = get_unaligned_le64(*(p));			\
-		*(p) += sizeof(u64);				\
-	} while (0)
-#define ceph_decode_32(p, v)					\
-	do {							\
-		v = get_unaligned_le32(*(p));			\
-		*(p) += sizeof(u32);				\
-	} while (0)
-#define ceph_decode_16(p, v)					\
-	do {							\
-		v = get_unaligned_le16(*(p));			\
-		*(p) += sizeof(u16);				\
-	} while (0)
-#define ceph_decode_8(p, v)				\
-	do {						\
-		v = *(u8 *)*(p);			\
-		(*p)++;					\
-	} while (0)
-
-#define ceph_decode_copy(p, pv, n)			\
-	do {						\
-		memcpy(pv, *(p), n);			\
-		*(p) += n;				\
-	} while (0)
-
-/* bounds check too */
 #define ceph_decode_64_safe(p, end, v, bad)			\
 	do {							\
 		ceph_decode_need(p, end, sizeof(u64), bad);	\
-		ceph_decode_64(p, v);				\
+		v = ceph_decode_64(p);				\
 	} while (0)
 #define ceph_decode_32_safe(p, end, v, bad)			\
 	do {							\
 		ceph_decode_need(p, end, sizeof(u32), bad);	\
-		ceph_decode_32(p, v);				\
+		v = ceph_decode_32(p);				\
 	} while (0)
 #define ceph_decode_16_safe(p, end, v, bad)			\
 	do {							\
 		ceph_decode_need(p, end, sizeof(u16), bad);	\
-		ceph_decode_16(p, v);				\
+		v = ceph_decode_16(p);				\
 	} while (0)
 
 #define ceph_decode_copy_safe(p, end, pv, n, bad)		\
@@ -71,41 +75,42 @@
 /*
  * struct ceph_timespec <-> struct timespec
  */
-#define ceph_decode_timespec(ts, tv)					\
-	do {								\
-		(ts)->tv_sec = le32_to_cpu((tv)->tv_sec);		\
-		(ts)->tv_nsec = le32_to_cpu((tv)->tv_nsec);		\
-	} while (0)
-#define ceph_encode_timespec(tv, ts)				\
-	do {							\
-		(tv)->tv_sec = cpu_to_le32((ts)->tv_sec);	\
-		(tv)->tv_nsec = cpu_to_le32((ts)->tv_nsec);	\
-	} while (0)
-
+static inline void ceph_decode_timespec(struct timespec *ts,
+					struct ceph_timespec *tv)
+{
+	ts->tv_sec = le32_to_cpu(tv->tv_sec);
+	ts->tv_nsec = le32_to_cpu(tv->tv_nsec);
+}
+static inline void ceph_encode_timespec(struct ceph_timespec *tv,
+					struct timespec *ts)
+{
+	tv->tv_sec = cpu_to_le32(ts->tv_sec);
+	tv->tv_nsec = cpu_to_le32(ts->tv_nsec);
+}
 
 /*
  * encoders
  */
-#define ceph_encode_64(p, v)						\
-	do {								\
-		put_unaligned_le64(v, (__le64 *)*(p));			\
-		*(p) += sizeof(u64);					\
-	} while (0)
-#define ceph_encode_32(p, v)					\
-	do {							\
-		put_unaligned_le32(v, (__le32 *)*(p));		\
-		*(p) += sizeof(u32);				\
-	} while (0)
-#define ceph_encode_16(p, v)					\
-	do {							\
-		put_unaligned_le16(v), (__le16 *)*(p));		\
-	*(p) += sizeof(u16);					\
-	} while (0)
-#define ceph_encode_8(p, v)			  \
-	do {					  \
-		*(u8 *)*(p) = v;		  \
-		(*(p))++;			  \
-	} while (0)
+static inline void ceph_encode_64(void **p, u64 v)
+{
+	put_unaligned_le64(v, (__le64 *)*p);
+	*p += sizeof(u64);
+}
+static inline void ceph_encode_32(void **p, u32 v)
+{
+	put_unaligned_le32(v, (__le32 *)*p);
+	*p += sizeof(u32);
+}
+static inline void ceph_encode_16(void **p, u16 v)
+{
+	put_unaligned_le16(v, (__le16 *)*p);
+	*p += sizeof(u16);
+}
+static inline void ceph_encode_8(void **p, u8 v)
+{
+	*(u8 *)*p = v;
+	(*p)++;
+}
 
 /*
  * filepath, string encoders
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index de8ba4a242ca..2b19da31a8b3 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -136,9 +136,9 @@ static int parse_reply_info_dir(void **p, void *end,
 		goto bad;
 
 	ceph_decode_need(p, end, sizeof(num) + 2, bad);
-	ceph_decode_32(p, num);
-	ceph_decode_8(p, info->dir_end);
-	ceph_decode_8(p, info->dir_complete);
+	num = ceph_decode_32(p);
+	info->dir_end = ceph_decode_8(p);
+	info->dir_complete = ceph_decode_8(p);
 	if (num == 0)
 		goto done;
 
@@ -160,7 +160,7 @@ static int parse_reply_info_dir(void **p, void *end,
 	while (num) {
 		/* dentry */
 		ceph_decode_need(p, end, sizeof(u32)*2, bad);
-		ceph_decode_32(p, info->dir_dname_len[i]);
+		info->dir_dname_len[i] = ceph_decode_32(p);
 		ceph_decode_need(p, end, info->dir_dname_len[i], bad);
 		info->dir_dname[i] = *p;
 		*p += info->dir_dname_len[i];
@@ -1791,10 +1791,10 @@ static void handle_forward(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
 	from_mds = le64_to_cpu(msg->hdr.src.name.num);
 
 	ceph_decode_need(&p, end, sizeof(u64)+2*sizeof(u32), bad);
-	ceph_decode_64(&p, tid);
-	ceph_decode_32(&p, next_mds);
-	ceph_decode_32(&p, fwd_seq);
-	ceph_decode_8(&p, must_resend);
+	tid = ceph_decode_64(&p);
+	next_mds = ceph_decode_32(&p);
+	fwd_seq = ceph_decode_32(&p);
+	must_resend = ceph_decode_8(&p);
 
 	WARN_ON(must_resend);  /* shouldn't happen. */
 
@@ -2783,8 +2783,8 @@ void ceph_mdsc_handle_map(struct ceph_mds_client *mdsc, struct ceph_msg *msg)
 		pr_err("got mdsmap with wrong fsid\n");
 		return;
 	}
-	ceph_decode_32(&p, epoch);
-	ceph_decode_32(&p, maplen);
+	epoch = ceph_decode_32(&p);
+	maplen = ceph_decode_32(&p);
 	dout("handle_map epoch %u len %d\n", epoch, (int)maplen);
 
 	/* do we need it? */
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 09180d8fafe4..80daea064470 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -60,21 +60,21 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 	ceph_decode_16_safe(p, end, version, bad);
 
 	ceph_decode_need(p, end, 8*sizeof(u32) + sizeof(u64), bad);
-	ceph_decode_32(p, m->m_epoch);
-	ceph_decode_32(p, m->m_client_epoch);
-	ceph_decode_32(p, m->m_last_failure);
-	ceph_decode_32(p, m->m_root);
-	ceph_decode_32(p, m->m_session_timeout);
-	ceph_decode_32(p, m->m_session_autoclose);
-	ceph_decode_64(p, m->m_max_file_size);
-	ceph_decode_32(p, m->m_max_mds);
+	m->m_epoch = ceph_decode_32(p);
+	m->m_client_epoch = ceph_decode_32(p);
+	m->m_last_failure = ceph_decode_32(p);
+	m->m_root = ceph_decode_32(p);
+	m->m_session_timeout = ceph_decode_32(p);
+	m->m_session_autoclose = ceph_decode_32(p);
+	m->m_max_file_size = ceph_decode_64(p);
+	m->m_max_mds = ceph_decode_32(p);
 
 	m->m_info = kcalloc(m->m_max_mds, sizeof(*m->m_info), GFP_NOFS);
 	if (m->m_info == NULL)
 		goto badmem;
 
 	/* pick out active nodes from mds_info (state > 0) */
-	ceph_decode_32(p, n);
+	n = ceph_decode_32(p);
 	for (i = 0; i < n; i++) {
 		u32 namelen;
 		s32 mds, inc, state;
@@ -86,18 +86,18 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 
 		ceph_decode_need(p, end, sizeof(addr) + 1 + sizeof(u32), bad);
 		ceph_decode_copy(p, &addr, sizeof(addr));
-		ceph_decode_8(p, infoversion);
-		ceph_decode_32(p, namelen);  /* skip mds name */
+		infoversion = ceph_decode_8(p);
+		namelen = ceph_decode_32(p);  /* skip mds name */
 		*p += namelen;
 
 		ceph_decode_need(p, end,
 				 4*sizeof(u32) + sizeof(u64) +
 				 sizeof(addr) + sizeof(struct ceph_timespec),
 				 bad);
-		ceph_decode_32(p, mds);
-		ceph_decode_32(p, inc);
-		ceph_decode_32(p, state);
-		ceph_decode_64(p, state_seq);
+		mds = ceph_decode_32(p);
+		inc = ceph_decode_32(p);
+		state = ceph_decode_32(p);
+		state_seq = ceph_decode_64(p);
 		*p += sizeof(addr);
 		*p += sizeof(struct ceph_timespec);
 		*p += sizeof(u32);
@@ -123,8 +123,8 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 					kcalloc(num_export_targets, sizeof(u32),
 						GFP_NOFS);
 				for (j = 0; j < num_export_targets; j++)
-					ceph_decode_32(&pexport_targets,
-					      m->m_info[mds].export_targets[j]);
+					m->m_info[mds].export_targets[j] =
+					       ceph_decode_32(&pexport_targets);
 			} else {
 				m->m_info[mds].export_targets = NULL;
 			}
@@ -139,8 +139,8 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		goto badmem;
 	ceph_decode_need(p, end, sizeof(u32)*(n+1), bad);
 	for (i = 0; i < n; i++)
-		ceph_decode_32(p, m->m_data_pg_pools[i]);
-	ceph_decode_32(p, m->m_cas_pg_pool);
+		m->m_data_pg_pools[i] = ceph_decode_32(p);
+	m->m_cas_pg_pool = ceph_decode_32(p);
 
 	/* ok, we don't care about the rest. */
 	dout("mdsmap_decode success epoch %u\n", m->m_epoch);
diff --git a/fs/ceph/mon_client.c b/fs/ceph/mon_client.c
index d6c8e783467e..bea2be9077e4 100644
--- a/fs/ceph/mon_client.c
+++ b/fs/ceph/mon_client.c
@@ -45,9 +45,9 @@ struct ceph_monmap *ceph_monmap_decode(void *p, void *end)
 
 	ceph_decode_need(&p, end, sizeof(fsid) + 2*sizeof(u32), bad);
 	ceph_decode_copy(&p, &fsid, sizeof(fsid));
-	ceph_decode_32(&p, epoch);
+	epoch = ceph_decode_32(&p);
 
-	ceph_decode_32(&p, num_mon);
+	num_mon = ceph_decode_32(&p);
 	ceph_decode_need(&p, end, num_mon*sizeof(m->mon_inst[0]), bad);
 
 	if (num_mon >= CEPH_MAX_MON)
diff --git a/fs/ceph/osd_client.c b/fs/ceph/osd_client.c
index 8e33928647f4..bbd9a5d23712 100644
--- a/fs/ceph/osd_client.c
+++ b/fs/ceph/osd_client.c
@@ -894,8 +894,8 @@ void ceph_osdc_handle_map(struct ceph_osd_client *osdc, struct ceph_msg *msg)
 	dout(" %d inc maps\n", nr_maps);
 	while (nr_maps > 0) {
 		ceph_decode_need(&p, end, 2*sizeof(u32), bad);
-		ceph_decode_32(&p, epoch);
-		ceph_decode_32(&p, maplen);
+		epoch = ceph_decode_32(&p);
+		maplen = ceph_decode_32(&p);
 		ceph_decode_need(&p, end, maplen, bad);
 		next = p + maplen;
 		if (osdc->osdmap && osdc->osdmap->epoch+1 == epoch) {
@@ -927,8 +927,8 @@ void ceph_osdc_handle_map(struct ceph_osd_client *osdc, struct ceph_msg *msg)
 	dout(" %d full maps\n", nr_maps);
 	while (nr_maps) {
 		ceph_decode_need(&p, end, 2*sizeof(u32), bad);
-		ceph_decode_32(&p, epoch);
-		ceph_decode_32(&p, maplen);
+		epoch = ceph_decode_32(&p);
+		maplen = ceph_decode_32(&p);
 		ceph_decode_need(&p, end, maplen, bad);
 		if (nr_maps > 1) {
 			dout("skipping non-latest full map %u len %d\n",
diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c
index 342e5f80996b..6f0aeff4185a 100644
--- a/fs/ceph/osdmap.c
+++ b/fs/ceph/osdmap.c
@@ -67,7 +67,7 @@ static int crush_decode_uniform_bucket(void **p, void *end,
 {
 	dout("crush_decode_uniform_bucket %p to %p\n", *p, end);
 	ceph_decode_need(p, end, (1+b->h.size) * sizeof(u32), bad);
-	ceph_decode_32(p, b->item_weight);
+	b->item_weight = ceph_decode_32(p);
 	return 0;
 bad:
 	return -EINVAL;
@@ -86,8 +86,8 @@ static int crush_decode_list_bucket(void **p, void *end,
 		return -ENOMEM;
 	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
 	for (j = 0; j < b->h.size; j++) {
-		ceph_decode_32(p, b->item_weights[j]);
-		ceph_decode_32(p, b->sum_weights[j]);
+		b->item_weights[j] = ceph_decode_32(p);
+		b->sum_weights[j] = ceph_decode_32(p);
 	}
 	return 0;
 bad:
@@ -105,7 +105,7 @@ static int crush_decode_tree_bucket(void **p, void *end,
 		return -ENOMEM;
 	ceph_decode_need(p, end, b->num_nodes * sizeof(u32), bad);
 	for (j = 0; j < b->num_nodes; j++)
-		ceph_decode_32(p, b->node_weights[j]);
+		b->node_weights[j] = ceph_decode_32(p);
 	return 0;
 bad:
 	return -EINVAL;
@@ -124,8 +124,8 @@ static int crush_decode_straw_bucket(void **p, void *end,
 		return -ENOMEM;
 	ceph_decode_need(p, end, 2 * b->h.size * sizeof(u32), bad);
 	for (j = 0; j < b->h.size; j++) {
-		ceph_decode_32(p, b->item_weights[j]);
-		ceph_decode_32(p, b->straws[j]);
+		b->item_weights[j] = ceph_decode_32(p);
+		b->straws[j] = ceph_decode_32(p);
 	}
 	return 0;
 bad:
@@ -148,15 +148,15 @@ static struct crush_map *crush_decode(void *pbyval, void *end)
 		return ERR_PTR(-ENOMEM);
 
 	ceph_decode_need(p, end, 4*sizeof(u32), bad);
-	ceph_decode_32(p, magic);
+	magic = ceph_decode_32(p);
 	if (magic != CRUSH_MAGIC) {
 		pr_err("crush_decode magic %x != current %x\n",
 		       (unsigned)magic, (unsigned)CRUSH_MAGIC);
 		goto bad;
 	}
-	ceph_decode_32(p, c->max_buckets);
-	ceph_decode_32(p, c->max_rules);
-	ceph_decode_32(p, c->max_devices);
+	c->max_buckets = ceph_decode_32(p);
+	c->max_rules = ceph_decode_32(p);
+	c->max_devices = ceph_decode_32(p);
 
 	c->device_parents = kcalloc(c->max_devices, sizeof(u32), GFP_NOFS);
 	if (c->device_parents == NULL)
@@ -208,11 +208,11 @@ static struct crush_map *crush_decode(void *pbyval, void *end)
 			goto badmem;
 
 		ceph_decode_need(p, end, 4*sizeof(u32), bad);
-		ceph_decode_32(p, b->id);
-		ceph_decode_16(p, b->type);
-		ceph_decode_16(p, b->alg);
-		ceph_decode_32(p, b->weight);
-		ceph_decode_32(p, b->size);
+		b->id = ceph_decode_32(p);
+		b->type = ceph_decode_16(p);
+		b->alg = ceph_decode_16(p);
+		b->weight = ceph_decode_32(p);
+		b->size = ceph_decode_32(p);
 
 		dout("crush_decode bucket size %d off %x %p to %p\n",
 		     b->size, (int)(*p-start), *p, end);
@@ -227,7 +227,7 @@ static struct crush_map *crush_decode(void *pbyval, void *end)
 
 		ceph_decode_need(p, end, b->size*sizeof(u32), bad);
 		for (j = 0; j < b->size; j++)
-			ceph_decode_32(p, b->items[j]);
+			b->items[j] = ceph_decode_32(p);
 
 		switch (b->alg) {
 		case CRUSH_BUCKET_UNIFORM:
@@ -290,9 +290,9 @@ static struct crush_map *crush_decode(void *pbyval, void *end)
 		ceph_decode_copy_safe(p, end, &r->mask, 4, bad); /* 4 u8's */
 		ceph_decode_need(p, end, r->len*3*sizeof(u32), bad);
 		for (j = 0; j < r->len; j++) {
-			ceph_decode_32(p, r->steps[j].op);
-			ceph_decode_32(p, r->steps[j].arg1);
-			ceph_decode_32(p, r->steps[j].arg2);
+			r->steps[j].op = ceph_decode_32(p);
+			r->steps[j].arg1 = ceph_decode_32(p);
+			r->steps[j].arg2 = ceph_decode_32(p);
 		}
 	}
 
@@ -411,11 +411,11 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 
 	ceph_decode_need(p, end, 2*sizeof(u64)+6*sizeof(u32), bad);
 	ceph_decode_copy(p, &map->fsid, sizeof(map->fsid));
-	ceph_decode_32(p, map->epoch);
+	map->epoch = ceph_decode_32(p);
 	ceph_decode_copy(p, &map->created, sizeof(map->created));
 	ceph_decode_copy(p, &map->modified, sizeof(map->modified));
 
-	ceph_decode_32(p, map->num_pools);
+	map->num_pools = ceph_decode_32(p);
 	map->pg_pool = kcalloc(map->num_pools, sizeof(*map->pg_pool),
 			       GFP_NOFS);
 	if (!map->pg_pool) {
@@ -425,7 +425,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 	ceph_decode_32_safe(p, end, max, bad);
 	while (max--) {
 		ceph_decode_need(p, end, 4+sizeof(map->pg_pool->v), bad);
-		ceph_decode_32(p, i);
+		i = ceph_decode_32(p);
 		if (i >= map->num_pools)
 			goto bad;
 		ceph_decode_copy(p, &map->pg_pool[i].v,
@@ -438,7 +438,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 
 	ceph_decode_32_safe(p, end, map->flags, bad);
 
-	ceph_decode_32(p, max);
+	max = ceph_decode_32(p);
 
 	/* (re)alloc osd arrays */
 	err = osdmap_set_max_osd(map, max);
@@ -456,7 +456,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 
 	*p += 4; /* skip length field (should match max) */
 	for (i = 0; i < map->max_osd; i++)
-		ceph_decode_32(p, map->osd_weight[i]);
+		map->osd_weight[i] = ceph_decode_32(p);
 
 	*p += 4; /* skip length field (should match max) */
 	ceph_decode_copy(p, map->osd_addr, map->max_osd*sizeof(*map->osd_addr));
@@ -469,8 +469,8 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 		struct ceph_pg_mapping *pg;
 
 		ceph_decode_need(p, end, sizeof(u32) + sizeof(u64), bad);
-		ceph_decode_64(p, pgid);
-		ceph_decode_32(p, n);
+		pgid = ceph_decode_64(p);
+		n = ceph_decode_32(p);
 		ceph_decode_need(p, end, n * sizeof(u32), bad);
 		pg = kmalloc(sizeof(*pg) + n*sizeof(u32), GFP_NOFS);
 		if (!pg) {
@@ -480,7 +480,7 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 		pg->pgid = pgid;
 		pg->len = n;
 		for (j = 0; j < n; j++)
-			ceph_decode_32(p, pg->osds[j]);
+			pg->osds[j] = ceph_decode_32(p);
 
 		err = __insert_pg_mapping(pg, &map->pg_temp);
 		if (err)
@@ -537,10 +537,10 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 	ceph_decode_need(p, end, sizeof(fsid)+sizeof(modified)+2*sizeof(u32),
 			 bad);
 	ceph_decode_copy(p, &fsid, sizeof(fsid));
-	ceph_decode_32(p, epoch);
+	epoch = ceph_decode_32(p);
 	BUG_ON(epoch != map->epoch+1);
 	ceph_decode_copy(p, &modified, sizeof(modified));
-	ceph_decode_32(p, new_flags);
+	new_flags = ceph_decode_32(p);
 
 	/* full map? */
 	ceph_decode_32_safe(p, end, len, bad);
@@ -568,7 +568,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 	ceph_decode_need(p, end, 5*sizeof(u32), bad);
 
 	/* new max? */
-	ceph_decode_32(p, max);
+	max = ceph_decode_32(p);
 	if (max >= 0) {
 		err = osdmap_set_max_osd(map, max);
 		if (err < 0)
@@ -641,8 +641,8 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 	while (len--) {
 		u32 osd, off;
 		ceph_decode_need(p, end, sizeof(u32)*2, bad);
-		ceph_decode_32(p, osd);
-		ceph_decode_32(p, off);
+		osd = ceph_decode_32(p);
+		off = ceph_decode_32(p);
 		pr_info("osd%d weight 0x%x %s\n", osd, off,
 		     off == CEPH_OSD_IN ? "(in)" :
 		     (off == CEPH_OSD_OUT ? "(out)" : ""));
@@ -659,8 +659,8 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 		u64 pgid;
 		u32 pglen;
 		ceph_decode_need(p, end, sizeof(u64) + sizeof(u32), bad);
-		ceph_decode_64(p, pgid);
-		ceph_decode_32(p, pglen);
+		pgid = ceph_decode_64(p);
+		pglen = ceph_decode_32(p);
 
 		/* remove any? */
 		while (rbp && rb_entry(rbp, struct ceph_pg_mapping,
@@ -683,7 +683,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 			pg->pgid = pgid;
 			pg->len = pglen;
 			for (j = 0; j < len; j++)
-				ceph_decode_32(p, pg->osds[j]);
+				pg->osds[j] = ceph_decode_32(p);
 			err = __insert_pg_mapping(pg, &map->pg_temp);
 			if (err)
 				goto bad;
-- 
cgit v1.2.3


From 63f2d211954b790fea0a9caeae605c7956535af6 Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@newdream.net>
Date: Tue, 3 Nov 2009 15:17:56 -0800
Subject: ceph: use fixed endian encoding for ceph_entity_addr

We exchange struct ceph_entity_addr over the wire and store it on disk.
The sockaddr_storage.ss_family field, however, is host endianness.  So,
fix ss_family endianness to big endian when sending/receiving over the
wire.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/decode.h     | 16 ++++++++++++++--
 fs/ceph/mdsmap.c     |  1 +
 fs/ceph/messenger.c  | 23 ++++++++++++++++++-----
 fs/ceph/messenger.h  |  1 +
 fs/ceph/mon_client.c |  2 ++
 fs/ceph/msgr.h       |  2 +-
 fs/ceph/osdmap.c     |  3 +++
 7 files changed, 40 insertions(+), 8 deletions(-)

(limited to 'fs/ceph/mdsmap.c')

diff --git a/fs/ceph/decode.h b/fs/ceph/decode.h
index 91179fb2cc3f..a382aecc55bb 100644
--- a/fs/ceph/decode.h
+++ b/fs/ceph/decode.h
@@ -76,18 +76,30 @@ static inline void ceph_decode_copy(void **p, void *pv, size_t n)
  * struct ceph_timespec <-> struct timespec
  */
 static inline void ceph_decode_timespec(struct timespec *ts,
-					struct ceph_timespec *tv)
+					const struct ceph_timespec *tv)
 {
 	ts->tv_sec = le32_to_cpu(tv->tv_sec);
 	ts->tv_nsec = le32_to_cpu(tv->tv_nsec);
 }
 static inline void ceph_encode_timespec(struct ceph_timespec *tv,
-					struct timespec *ts)
+					const struct timespec *ts)
 {
 	tv->tv_sec = cpu_to_le32(ts->tv_sec);
 	tv->tv_nsec = cpu_to_le32(ts->tv_nsec);
 }
 
+/*
+ * sockaddr_storage <-> ceph_sockaddr
+ */
+static inline void ceph_encode_addr(struct ceph_entity_addr *a)
+{
+	a->in_addr.ss_family = htons(a->in_addr.ss_family);
+}
+static inline void ceph_decode_addr(struct ceph_entity_addr *a)
+{
+	a->in_addr.ss_family = ntohs(a->in_addr.ss_family);
+}
+
 /*
  * encoders
  */
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 80daea064470..4226c810ce22 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -86,6 +86,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 
 		ceph_decode_need(p, end, sizeof(addr) + 1 + sizeof(u32), bad);
 		ceph_decode_copy(p, &addr, sizeof(addr));
+		ceph_decode_addr(&addr);
 		infoversion = ceph_decode_8(p);
 		namelen = ceph_decode_32(p);  /* skip mds name */
 		*p += namelen;
diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
index b48abc0b3be7..6ff44bbddf67 100644
--- a/fs/ceph/messenger.c
+++ b/fs/ceph/messenger.c
@@ -12,6 +12,7 @@
 
 #include "super.h"
 #include "messenger.h"
+#include "decode.h"
 
 /*
  * Ceph uses the messenger to exchange ceph_msg messages with other
@@ -97,6 +98,12 @@ const char *pr_addr(const struct sockaddr_storage *ss)
 	return s;
 }
 
+static void encode_my_addr(struct ceph_messenger *msgr)
+{
+	memcpy(&msgr->my_enc_addr, &msgr->inst.addr, sizeof(msgr->my_enc_addr));
+	ceph_encode_addr(&msgr->my_enc_addr);
+}
+
 /*
  * work queue for all reading and writing to/from the socket.
  */
@@ -590,12 +597,12 @@ static void prepare_write_connect(struct ceph_messenger *msgr,
 
 	con->out_kvec[0].iov_base = CEPH_BANNER;
 	con->out_kvec[0].iov_len = len;
-	con->out_kvec[1].iov_base = &msgr->inst.addr;
-	con->out_kvec[1].iov_len = sizeof(msgr->inst.addr);
+	con->out_kvec[1].iov_base = &msgr->my_enc_addr;
+	con->out_kvec[1].iov_len = sizeof(msgr->my_enc_addr);
 	con->out_kvec[2].iov_base = &con->out_connect;
 	con->out_kvec[2].iov_len = sizeof(con->out_connect);
 	con->out_kvec_left = 3;
-	con->out_kvec_bytes = len + sizeof(msgr->inst.addr) +
+	con->out_kvec_bytes = len + sizeof(msgr->my_enc_addr) +
 		sizeof(con->out_connect);
 	con->out_kvec_cur = con->out_kvec;
 	con->out_more = 0;
@@ -976,6 +983,9 @@ static int process_connect(struct ceph_connection *con)
 	if (verify_hello(con) < 0)
 		return -1;
 
+	ceph_decode_addr(&con->actual_peer_addr);
+	ceph_decode_addr(&con->peer_addr_for_me);
+
 	/*
 	 * Make sure the other end is who we wanted.  note that the other
 	 * end may not yet know their ip address, so if it's 0.0.0.0, give
@@ -1005,6 +1015,7 @@ static int process_connect(struct ceph_connection *con)
 		       &con->peer_addr_for_me.in_addr,
 		       sizeof(con->peer_addr_for_me.in_addr));
 		addr_set_port(&con->msgr->inst.addr.in_addr, port);
+		encode_my_addr(con->msgr);
 		dout("process_connect learned my addr is %s\n",
 		     pr_addr(&con->msgr->inst.addr.in_addr));
 	}
@@ -1780,6 +1791,7 @@ struct ceph_messenger *ceph_messenger_create(struct ceph_entity_addr *myaddr)
 	/* select a random nonce */
 	get_random_bytes(&msgr->inst.addr.nonce,
 			 sizeof(msgr->inst.addr.nonce));
+	encode_my_addr(msgr);
 
 	dout("messenger_create %p\n", msgr);
 	return msgr;
@@ -1806,8 +1818,9 @@ void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg)
 	}
 
 	/* set src+dst */
-	msg->hdr.src = con->msgr->inst;
-	msg->hdr.orig_src = con->msgr->inst;
+	msg->hdr.src.name = con->msgr->inst.name;
+	msg->hdr.src.addr = con->msgr->my_enc_addr;
+	msg->hdr.orig_src = msg->hdr.src;
 	msg->hdr.dst_erank = con->peer_addr.erank;
 
 	/* queue */
diff --git a/fs/ceph/messenger.h b/fs/ceph/messenger.h
index dcd98b64dca9..e016fa7cf970 100644
--- a/fs/ceph/messenger.h
+++ b/fs/ceph/messenger.h
@@ -53,6 +53,7 @@ extern const char *ceph_name_type_str(int t);
 
 struct ceph_messenger {
 	struct ceph_entity_inst inst;    /* my name+address */
+	struct ceph_entity_addr my_enc_addr;
 	struct page *zero_page;          /* used in certain error cases */
 
 	bool nocrc;
diff --git a/fs/ceph/mon_client.c b/fs/ceph/mon_client.c
index 61263c99c6a8..95b76e761e18 100644
--- a/fs/ceph/mon_client.c
+++ b/fs/ceph/mon_client.c
@@ -59,6 +59,8 @@ struct ceph_monmap *ceph_monmap_decode(void *p, void *end)
 	m->epoch = epoch;
 	m->num_mon = num_mon;
 	ceph_decode_copy(&p, m->mon_inst, num_mon*sizeof(m->mon_inst[0]));
+	for (i = 0; i < num_mon; i++)
+		ceph_decode_addr(&m->mon_inst[i].addr);
 
 	dout("monmap_decode epoch %d, num_mon %d\n", m->epoch,
 	     m->num_mon);
diff --git a/fs/ceph/msgr.h b/fs/ceph/msgr.h
index 9abc879e25b1..8e3ea2eb1bf5 100644
--- a/fs/ceph/msgr.h
+++ b/fs/ceph/msgr.h
@@ -21,7 +21,7 @@
  * whenever the wire protocol changes.  try to keep this string length
  * constant.
  */
-#define CEPH_BANNER "ceph v022"
+#define CEPH_BANNER "ceph v023"
 #define CEPH_BANNER_MAX_LEN 30
 
 
diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c
index d62e111b8a34..cd7bb265d789 100644
--- a/fs/ceph/osdmap.c
+++ b/fs/ceph/osdmap.c
@@ -460,6 +460,8 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
 
 	*p += 4; /* skip length field (should match max) */
 	ceph_decode_copy(p, map->osd_addr, map->max_osd*sizeof(*map->osd_addr));
+	for (i = 0; i < map->max_osd; i++)
+		ceph_decode_addr(&map->osd_addr[i]);
 
 	/* pg_temp */
 	ceph_decode_32_safe(p, end, len, bad);
@@ -619,6 +621,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 		struct ceph_entity_addr addr;
 		ceph_decode_32_safe(p, end, osd, bad);
 		ceph_decode_copy_safe(p, end, &addr, sizeof(addr), bad);
+		ceph_decode_addr(&addr);
 		pr_info("osd%d up\n", osd);
 		BUG_ON(osd >= map->max_osd);
 		map->osd_state[osd] |= CEPH_OSD_UP;
-- 
cgit v1.2.3


From 94045e115ee72aee3b17295791da07078f2f778c Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@newdream.net>
Date: Thu, 19 Nov 2009 15:31:50 -0800
Subject: ceph: decode updated mdsmap format

The mds map now uses the global_id as the 'key' (instead of the addr,
which was a poor choice).

This is protocol change.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/ceph_fs.h |  2 +-
 fs/ceph/mdsmap.c  | 15 +++++++++------
 fs/ceph/mdsmap.h  |  1 +
 3 files changed, 11 insertions(+), 7 deletions(-)

(limited to 'fs/ceph/mdsmap.c')

diff --git a/fs/ceph/ceph_fs.h b/fs/ceph/ceph_fs.h
index 1e96a9a87d8d..4e5f49c738d8 100644
--- a/fs/ceph/ceph_fs.h
+++ b/fs/ceph/ceph_fs.h
@@ -39,7 +39,7 @@
 #define CEPH_MDS_PROTOCOL     9 /* cluster internal */
 #define CEPH_MON_PROTOCOL     5 /* cluster internal */
 #define CEPH_OSDC_PROTOCOL   22 /* server/client */
-#define CEPH_MDSC_PROTOCOL   29 /* server/client */
+#define CEPH_MDSC_PROTOCOL   30 /* server/client */
 #define CEPH_MONC_PROTOCOL   15 /* server/client */
 
 
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index 4226c810ce22..cad8d25861e5 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -76,6 +76,7 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 	/* pick out active nodes from mds_info (state > 0) */
 	n = ceph_decode_32(p);
 	for (i = 0; i < n; i++) {
+		u64 global_id;
 		u32 namelen;
 		s32 mds, inc, state;
 		u64 state_seq;
@@ -84,10 +85,10 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		u32 num_export_targets;
 		void *pexport_targets = NULL;
 
-		ceph_decode_need(p, end, sizeof(addr) + 1 + sizeof(u32), bad);
-		ceph_decode_copy(p, &addr, sizeof(addr));
-		ceph_decode_addr(&addr);
+		ceph_decode_need(p, end, sizeof(u64)*2 + 1 + sizeof(u32), bad);
+		global_id = ceph_decode_64(p);
 		infoversion = ceph_decode_8(p);
+		*p += sizeof(u64);
 		namelen = ceph_decode_32(p);  /* skip mds name */
 		*p += namelen;
 
@@ -99,7 +100,8 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 		inc = ceph_decode_32(p);
 		state = ceph_decode_32(p);
 		state_seq = ceph_decode_64(p);
-		*p += sizeof(addr);
+		ceph_decode_copy(p, &addr, sizeof(addr));
+		ceph_decode_addr(&addr);
 		*p += sizeof(struct ceph_timespec);
 		*p += sizeof(u32);
 		ceph_decode_32_safe(p, end, namelen, bad);
@@ -112,10 +114,11 @@ struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 			num_export_targets = 0;
 		}
 
-		dout("mdsmap_decode %d/%d mds%d.%d %s %s\n",
-		     i+1, n, mds, inc, pr_addr(&addr.in_addr),
+		dout("mdsmap_decode %d/%d %lld mds%d.%d %s %s\n",
+		     i+1, n, global_id, mds, inc, pr_addr(&addr.in_addr),
 		     ceph_mds_state_name(state));
 		if (mds >= 0 && mds < m->m_max_mds && state > 0) {
+			m->m_info[mds].global_id = global_id;
 			m->m_info[mds].state = state;
 			m->m_info[mds].addr = addr;
 			m->m_info[mds].num_export_targets = num_export_targets;
diff --git a/fs/ceph/mdsmap.h b/fs/ceph/mdsmap.h
index d317308648fb..eacc131aa5cb 100644
--- a/fs/ceph/mdsmap.h
+++ b/fs/ceph/mdsmap.h
@@ -9,6 +9,7 @@
  * we limit fields to those the client actually xcares about
  */
 struct ceph_mds_info {
+	u64 global_id;
 	struct ceph_entity_addr addr;
 	s32 state;
 	int num_export_targets;
-- 
cgit v1.2.3


From 9ec7cab14e6de732d4e7c355fe67c5810c32c758 Mon Sep 17 00:00:00 2001
From: Sage Weil <sage@newdream.net>
Date: Mon, 14 Dec 2009 15:13:47 -0800
Subject: ceph: hex dump corrupt server data to KERN_DEBUG

Also, print fsid using standard format, NOT hex dump.

Signed-off-by: Sage Weil <sage@newdream.net>
---
 fs/ceph/caps.c       |  1 +
 fs/ceph/mds_client.c |  4 ++++
 fs/ceph/mdsmap.c     |  4 ++++
 fs/ceph/messenger.c  | 20 ++++++++++++++++++++
 fs/ceph/messenger.h  |  2 ++
 fs/ceph/mon_client.c |  2 ++
 fs/ceph/osd_client.c |  2 ++
 fs/ceph/osdmap.c     |  3 +++
 fs/ceph/snap.c       |  1 +
 fs/ceph/super.c      |  9 ++-------
 10 files changed, 41 insertions(+), 7 deletions(-)

(limited to 'fs/ceph/mdsmap.c')

diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 9b9ce143ac1f..dfb509f53542 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -2681,6 +2681,7 @@ done:
 
 bad:
 	pr_err("ceph_handle_caps: corrupt message\n");
+	ceph_msg_dump(msg);
 	return;
 }
 
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 739093f281d0..29a93fe35f85 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1650,6 +1650,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 		return;
 	if (msg->front.iov_len < sizeof(*head)) {
 		pr_err("mdsc_handle_reply got corrupt (short) reply\n");
+		ceph_msg_dump(msg);
 		return;
 	}
 
@@ -1740,6 +1741,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
 	mutex_lock(&session->s_mutex);
 	if (err < 0) {
 		pr_err("mdsc_handle_reply got corrupt reply mds%d\n", mds);
+		ceph_msg_dump(msg);
 		goto out_err;
 	}
 
@@ -1929,6 +1931,7 @@ static void handle_session(struct ceph_mds_session *session,
 bad:
 	pr_err("mdsc_handle_session corrupt message mds%d len %d\n", mds,
 	       (int)msg->front.iov_len);
+	ceph_msg_dump(msg);
 	return;
 }
 
@@ -2394,6 +2397,7 @@ out:
 
 bad:
 	pr_err("corrupt lease message\n");
+	ceph_msg_dump(msg);
 }
 
 void ceph_mdsc_lease_send_msg(struct ceph_mds_session *session,
diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c
index cad8d25861e5..c4c498e6dfef 100644
--- a/fs/ceph/mdsmap.c
+++ b/fs/ceph/mdsmap.c
@@ -49,6 +49,7 @@ int ceph_mdsmap_get_random_mds(struct ceph_mdsmap *m)
 struct ceph_mdsmap *ceph_mdsmap_decode(void **p, void *end)
 {
 	struct ceph_mdsmap *m;
+	const void *start = *p;
 	int i, j, n;
 	int err = -EINVAL;
 	u16 version;
@@ -154,6 +155,9 @@ badmem:
 	err = -ENOMEM;
 bad:
 	pr_err("corrupt mdsmap\n");
+	print_hex_dump(KERN_DEBUG, "mdsmap: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       start, end - start, true);
 	ceph_mdsmap_destroy(m);
 	return ERR_PTR(-EINVAL);
 }
diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
index d5eef76a253c..b10f88c56706 100644
--- a/fs/ceph/messenger.c
+++ b/fs/ceph/messenger.c
@@ -2115,3 +2115,23 @@ void ceph_msg_last_put(struct kref *kref)
 	else
 		ceph_msg_kfree(m);
 }
+
+void ceph_msg_dump(struct ceph_msg *msg)
+{
+	pr_debug("msg_dump %p (front_max %d nr_pages %d)\n", msg,
+		 msg->front_max, msg->nr_pages);
+	print_hex_dump(KERN_DEBUG, "header: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       &msg->hdr, sizeof(msg->hdr), true);
+	print_hex_dump(KERN_DEBUG, " front: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       msg->front.iov_base, msg->front.iov_len, true);
+	if (msg->middle)
+		print_hex_dump(KERN_DEBUG, "middle: ",
+			       DUMP_PREFIX_OFFSET, 16, 1,
+			       msg->middle->vec.iov_base,
+			       msg->middle->vec.iov_len, true);
+	print_hex_dump(KERN_DEBUG, "footer: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       &msg->footer, sizeof(msg->footer), true);
+}
diff --git a/fs/ceph/messenger.h b/fs/ceph/messenger.h
index eff5cb5197fc..e04c214b4f6f 100644
--- a/fs/ceph/messenger.h
+++ b/fs/ceph/messenger.h
@@ -254,4 +254,6 @@ static inline void ceph_msg_put(struct ceph_msg *msg)
 	kref_put(&msg->kref, ceph_msg_last_put);
 }
 
+extern void ceph_msg_dump(struct ceph_msg *msg);
+
 #endif
diff --git a/fs/ceph/mon_client.c b/fs/ceph/mon_client.c
index a76da5e6dbdd..775a9c029c51 100644
--- a/fs/ceph/mon_client.c
+++ b/fs/ceph/mon_client.c
@@ -242,6 +242,7 @@ static void handle_subscribe_ack(struct ceph_mon_client *monc,
 	return;
 bad:
 	pr_err("got corrupt subscribe-ack msg\n");
+	ceph_msg_dump(msg);
 }
 
 /*
@@ -364,6 +365,7 @@ static void handle_statfs_reply(struct ceph_mon_client *monc,
 
 bad:
 	pr_err("corrupt statfs reply, no tid\n");
+	ceph_msg_dump(msg);
 }
 
 /*
diff --git a/fs/ceph/osd_client.c b/fs/ceph/osd_client.c
index 63482ef3de01..4bfe880d53c8 100644
--- a/fs/ceph/osd_client.c
+++ b/fs/ceph/osd_client.c
@@ -773,6 +773,7 @@ bad:
 	pr_err("corrupt osd_op_reply got %d %d expected %d\n",
 	       (int)msg->front.iov_len, le32_to_cpu(msg->hdr.front_len),
 	       (int)sizeof(*rhead));
+	ceph_msg_dump(msg);
 }
 
 
@@ -964,6 +965,7 @@ done:
 
 bad:
 	pr_err("osdc handle_map corrupt msg\n");
+	ceph_msg_dump(msg);
 	up_write(&osdc->map_sem);
 	return;
 }
diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c
index be5318aa7714..8c8ffe5ef7d4 100644
--- a/fs/ceph/osdmap.c
+++ b/fs/ceph/osdmap.c
@@ -726,6 +726,9 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
 bad:
 	pr_err("corrupt inc osdmap epoch %d off %d (%p of %p-%p)\n",
 	       epoch, (int)(*p - start), *p, start, end);
+	print_hex_dump(KERN_DEBUG, "osdmap: ",
+		       DUMP_PREFIX_OFFSET, 16, 1,
+		       start, end - start, true);
 	if (newcrush)
 		crush_destroy(newcrush);
 	return ERR_PTR(err);
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index 52f46a1208f5..dcf18d92130a 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -877,6 +877,7 @@ split_skip_inode:
 
 bad:
 	pr_err("corrupt snap message from mds%d\n", mds);
+	ceph_msg_dump(msg);
 out:
 	if (locked_rwsem)
 		up_write(&mdsc->snap_rwsem);
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index a828943296c5..6d02a166f8ff 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -602,13 +602,8 @@ int ceph_check_fsid(struct ceph_client *client, struct ceph_fsid *fsid)
 {
 	if (client->have_fsid) {
 		if (ceph_fsid_compare(&client->fsid, fsid)) {
-			print_hex_dump(KERN_ERR, "this fsid: ",
-				       DUMP_PREFIX_NONE, 16, 1,
-				       (void *)fsid, 16, 0);
-			print_hex_dump(KERN_ERR, " old fsid: ",
-				       DUMP_PREFIX_NONE, 16, 1,
-				       (void *)&client->fsid, 16, 0);
-			pr_err("fsid mismatch\n");
+			pr_err("bad fsid, had " FSID_FORMAT " got " FSID_FORMAT,
+			       PR_FSID(&client->fsid), PR_FSID(fsid));
 			return -1;
 		}
 	} else {
-- 
cgit v1.2.3