summaryrefslogtreecommitdiffstats
path: root/fs/ceph
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ceph')
-rw-r--r--fs/ceph/addr.c73
-rw-r--r--fs/ceph/auth.c1
-rw-r--r--fs/ceph/auth_none.c1
-rw-r--r--fs/ceph/auth_x.c54
-rw-r--r--fs/ceph/buffer.c3
-rw-r--r--fs/ceph/caps.c116
-rw-r--r--fs/ceph/crypto.c1
-rw-r--r--fs/ceph/debugfs.c1
-rw-r--r--fs/ceph/dir.c12
-rw-r--r--fs/ceph/export.c1
-rw-r--r--fs/ceph/file.c1
-rw-r--r--fs/ceph/inode.c26
-rw-r--r--fs/ceph/mds_client.c44
-rw-r--r--fs/ceph/messenger.c29
-rw-r--r--fs/ceph/messenger.h1
-rw-r--r--fs/ceph/mon_client.c1
-rw-r--r--fs/ceph/osd_client.c29
-rw-r--r--fs/ceph/osd_client.h2
-rw-r--r--fs/ceph/osdmap.c201
-rw-r--r--fs/ceph/osdmap.h1
-rw-r--r--fs/ceph/pagelist.c1
-rw-r--r--fs/ceph/rados.h6
-rw-r--r--fs/ceph/snap.c33
-rw-r--r--fs/ceph/super.c1
-rw-r--r--fs/ceph/super.h4
-rw-r--r--fs/ceph/xattr.c1
26 files changed, 418 insertions, 226 deletions
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index 23bb0ceabe3..412593703d1 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -5,6 +5,7 @@
#include <linux/mm.h>
#include <linux/pagemap.h>
#include <linux/writeback.h> /* generic_writepages */
+#include <linux/slab.h>
#include <linux/pagevec.h>
#include <linux/task_io_accounting_ops.h>
@@ -336,16 +337,15 @@ out:
/*
* Get ref for the oldest snapc for an inode with dirty data... that is, the
* only snap context we are allowed to write back.
- *
- * Caller holds i_lock.
*/
-static struct ceph_snap_context *__get_oldest_context(struct inode *inode,
- u64 *snap_size)
+static struct ceph_snap_context *get_oldest_context(struct inode *inode,
+ u64 *snap_size)
{
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_snap_context *snapc = NULL;
struct ceph_cap_snap *capsnap = NULL;
+ spin_lock(&inode->i_lock);
list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
dout(" cap_snap %p snapc %p has %d dirty pages\n", capsnap,
capsnap->context, capsnap->dirty_pages);
@@ -356,21 +356,11 @@ static struct ceph_snap_context *__get_oldest_context(struct inode *inode,
break;
}
}
- if (!snapc && ci->i_snap_realm) {
- snapc = ceph_get_snap_context(ci->i_snap_realm->cached_context);
+ if (!snapc && ci->i_head_snapc) {
+ snapc = ceph_get_snap_context(ci->i_head_snapc);
dout(" head snapc %p has %d dirty pages\n",
snapc, ci->i_wrbuffer_ref_head);
}
- return snapc;
-}
-
-static struct ceph_snap_context *get_oldest_context(struct inode *inode,
- u64 *snap_size)
-{
- struct ceph_snap_context *snapc = NULL;
-
- spin_lock(&inode->i_lock);
- snapc = __get_oldest_context(inode, snap_size);
spin_unlock(&inode->i_lock);
return snapc;
}
@@ -391,7 +381,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
int len = PAGE_CACHE_SIZE;
loff_t i_size;
int err = 0;
- struct ceph_snap_context *snapc;
+ struct ceph_snap_context *snapc, *oldest;
u64 snap_size = 0;
long writeback_stat;
@@ -412,13 +402,16 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
dout("writepage %p page %p not dirty?\n", inode, page);
goto out;
}
- if (snapc != get_oldest_context(inode, &snap_size)) {
+ oldest = get_oldest_context(inode, &snap_size);
+ if (snapc->seq > oldest->seq) {
dout("writepage %p page %p snapc %p not writeable - noop\n",
inode, page, (void *)page->private);
/* we should only noop if called by kswapd */
WARN_ON((current->flags & PF_MEMALLOC) == 0);
+ ceph_put_snap_context(oldest);
goto out;
}
+ ceph_put_snap_context(oldest);
/* is this a partial page at end of file? */
if (snap_size)
@@ -457,7 +450,7 @@ static int writepage_nounlock(struct page *page, struct writeback_control *wbc)
ClearPagePrivate(page);
end_page_writeback(page);
ceph_put_wrbuffer_cap_refs(ci, 1, snapc);
- ceph_put_snap_context(snapc);
+ ceph_put_snap_context(snapc); /* page's reference */
out:
return err;
}
@@ -557,9 +550,9 @@ static void writepages_finish(struct ceph_osd_request *req,
dout("inode %p skipping page %p\n", inode, page);
wbc->pages_skipped++;
}
+ ceph_put_snap_context((void *)page->private);
page->private = 0;
ClearPagePrivate(page);
- ceph_put_snap_context(snapc);
dout("unlocking %d %p\n", i, page);
end_page_writeback(page);
@@ -617,7 +610,7 @@ static int ceph_writepages_start(struct address_space *mapping,
int range_whole = 0;
int should_loop = 1;
pgoff_t max_pages = 0, max_pages_ever = 0;
- struct ceph_snap_context *snapc = NULL, *last_snapc = NULL;
+ struct ceph_snap_context *snapc = NULL, *last_snapc = NULL, *pgsnapc;
struct pagevec pvec;
int done = 0;
int rc = 0;
@@ -769,9 +762,10 @@ get_more_pages:
}
/* only if matching snap context */
- if (snapc != (void *)page->private) {
- dout("page snapc %p != oldest %p\n",
- (void *)page->private, snapc);
+ pgsnapc = (void *)page->private;
+ if (pgsnapc->seq > snapc->seq) {
+ dout("page snapc %p %lld > oldest %p %lld\n",
+ pgsnapc, pgsnapc->seq, snapc, snapc->seq);
unlock_page(page);
if (!locked_pages)
continue; /* keep looking for snap */
@@ -913,12 +907,19 @@ static int context_is_writeable_or_written(struct inode *inode,
struct ceph_snap_context *snapc)
{
struct ceph_snap_context *oldest = get_oldest_context(inode, NULL);
- return !oldest || snapc->seq <= oldest->seq;
+ int ret = !oldest || snapc->seq <= oldest->seq;
+
+ ceph_put_snap_context(oldest);
+ return ret;
}
/*
* We are only allowed to write into/dirty the page if the page is
* clean, or already dirty within the same snap context.
+ *
+ * called with page locked.
+ * return success with page locked,
+ * or any failure (incl -EAGAIN) with page unlocked.
*/
static int ceph_update_writeable_page(struct file *file,
loff_t pos, unsigned len,
@@ -931,8 +932,8 @@ static int ceph_update_writeable_page(struct file *file,
int pos_in_page = pos & ~PAGE_CACHE_MASK;
int end_in_page = pos_in_page + len;
loff_t i_size;
- struct ceph_snap_context *snapc;
int r;
+ struct ceph_snap_context *snapc, *oldest;
retry_locked:
/* writepages currently holds page lock, but if we change that later, */
@@ -942,30 +943,34 @@ retry_locked:
BUG_ON(!ci->i_snap_realm);
down_read(&mdsc->snap_rwsem);
BUG_ON(!ci->i_snap_realm->cached_context);
- if (page->private &&
- (void *)page->private != ci->i_snap_realm->cached_context) {
+ snapc = (void *)page->private;
+ if (snapc && snapc != ci->i_head_snapc) {
/*
* this page is already dirty in another (older) snap
* context! is it writeable now?
*/
- snapc = get_oldest_context(inode, NULL);
+ oldest = get_oldest_context(inode, NULL);
up_read(&mdsc->snap_rwsem);
- if (snapc != (void *)page->private) {
+ if (snapc->seq > oldest->seq) {
+ ceph_put_snap_context(oldest);
dout(" page %p snapc %p not current or oldest\n",
- page, (void *)page->private);
+ page, snapc);
/*
* queue for writeback, and wait for snapc to
* be writeable or written
*/
- snapc = ceph_get_snap_context((void *)page->private);
+ snapc = ceph_get_snap_context(snapc);
unlock_page(page);
ceph_queue_writeback(inode);
- wait_event_interruptible(ci->i_cap_wq,
+ r = wait_event_interruptible(ci->i_cap_wq,
context_is_writeable_or_written(inode, snapc));
ceph_put_snap_context(snapc);
+ if (r == -ERESTARTSYS)
+ return r;
return -EAGAIN;
}
+ ceph_put_snap_context(oldest);
/* yay, writeable, do it now (without dropping page lock) */
dout(" page %p snapc %p not current, but oldest\n",
@@ -1035,7 +1040,7 @@ static int ceph_write_begin(struct file *file, struct address_space *mapping,
int r;
do {
- /* get a page*/
+ /* get a page */
page = grab_cache_page_write_begin(mapping, index, 0);
if (!page)
return -ENOMEM;
diff --git a/fs/ceph/auth.c b/fs/ceph/auth.c
index abb204fea6c..f6394b94b86 100644
--- a/fs/ceph/auth.c
+++ b/fs/ceph/auth.c
@@ -1,6 +1,7 @@
#include "ceph_debug.h"
#include <linux/module.h>
+#include <linux/slab.h>
#include <linux/err.h>
#include "types.h"
diff --git a/fs/ceph/auth_none.c b/fs/ceph/auth_none.c
index b4ef6f0a6c8..8cd9e3af07f 100644
--- a/fs/ceph/auth_none.c
+++ b/fs/ceph/auth_none.c
@@ -4,6 +4,7 @@
#include <linux/err.h>
#include <linux/module.h>
#include <linux/random.h>
+#include <linux/slab.h>
#include "auth_none.h"
#include "auth.h"
diff --git a/fs/ceph/auth_x.c b/fs/ceph/auth_x.c
index f0318427b6d..d9001a4dc8c 100644
--- a/fs/ceph/auth_x.c
+++ b/fs/ceph/auth_x.c
@@ -4,6 +4,7 @@
#include <linux/err.h>
#include <linux/module.h>
#include <linux/random.h>
+#include <linux/slab.h>
#include "auth_x.h"
#include "auth_x_protocol.h"
@@ -28,6 +29,12 @@ static int ceph_x_is_authenticated(struct ceph_auth_client *ac)
return (ac->want_keys & xi->have_keys) == ac->want_keys;
}
+static int ceph_x_encrypt_buflen(int ilen)
+{
+ return sizeof(struct ceph_x_encrypt_header) + ilen + 16 +
+ sizeof(u32);
+}
+
static int ceph_x_encrypt(struct ceph_crypto_key *secret,
void *ibuf, int ilen, void *obuf, size_t olen)
{
@@ -150,6 +157,11 @@ static int ceph_x_proc_ticket_reply(struct ceph_auth_client *ac,
struct timespec validity;
struct ceph_crypto_key old_key;
void *tp, *tpend;
+ struct ceph_timespec new_validity;
+ struct ceph_crypto_key new_session_key;
+ struct ceph_buffer *new_ticket_blob;
+ unsigned long new_expires, new_renew_after;
+ u64 new_secret_id;
ceph_decode_need(&p, end, sizeof(u32) + 1, bad);
@@ -182,16 +194,16 @@ static int ceph_x_proc_ticket_reply(struct ceph_auth_client *ac,
goto bad;
memcpy(&old_key, &th->session_key, sizeof(old_key));
- ret = ceph_crypto_key_decode(&th->session_key, &dp, dend);
+ ret = ceph_crypto_key_decode(&new_session_key, &dp, dend);
if (ret)
goto out;
- ceph_decode_copy(&dp, &th->validity, sizeof(th->validity));
- ceph_decode_timespec(&validity, &th->validity);
- th->expires = get_seconds() + validity.tv_sec;
- th->renew_after = th->expires - (validity.tv_sec / 4);
- dout(" expires=%lu renew_after=%lu\n", th->expires,
- th->renew_after);
+ ceph_decode_copy(&dp, &new_validity, sizeof(new_validity));
+ ceph_decode_timespec(&validity, &new_validity);
+ new_expires = get_seconds() + validity.tv_sec;
+ new_renew_after = new_expires - (validity.tv_sec / 4);
+ dout(" expires=%lu renew_after=%lu\n", new_expires,
+ new_renew_after);
/* ticket blob for service */
ceph_decode_8_safe(&p, end, is_enc, bad);
@@ -216,10 +228,21 @@ static int ceph_x_proc_ticket_reply(struct ceph_auth_client *ac,
dout(" ticket blob is %d bytes\n", dlen);
ceph_decode_need(&tp, tpend, 1 + sizeof(u64), bad);
struct_v = ceph_decode_8(&tp);
- th->secret_id = ceph_decode_64(&tp);
- ret = ceph_decode_buffer(&th->ticket_blob, &tp, tpend);
+ new_secret_id = ceph_decode_64(&tp);
+ ret = ceph_decode_buffer(&new_ticket_blob, &tp, tpend);
if (ret)
goto out;
+
+ /* all is well, update our ticket */
+ ceph_crypto_key_destroy(&th->session_key);
+ if (th->ticket_blob)
+ ceph_buffer_put(th->ticket_blob);
+ th->session_key = new_session_key;
+ th->ticket_blob = new_ticket_blob;
+ th->validity = new_validity;
+ th->secret_id = new_secret_id;
+ th->expires = new_expires;
+ th->renew_after = new_renew_after;
dout(" got ticket service %d (%s) secret_id %lld len %d\n",
type, ceph_entity_type_name(type), th->secret_id,
(int)th->ticket_blob->vec.iov_len);
@@ -242,7 +265,7 @@ static int ceph_x_build_authorizer(struct ceph_auth_client *ac,
struct ceph_x_ticket_handler *th,
struct ceph_x_authorizer *au)
{
- int len;
+ int maxlen;
struct ceph_x_authorize_a *msg_a;
struct ceph_x_authorize_b msg_b;
void *p, *end;
@@ -253,15 +276,15 @@ static int ceph_x_build_authorizer(struct ceph_auth_client *ac,
dout("build_authorizer for %s %p\n",
ceph_entity_type_name(th->service), au);
- len = sizeof(*msg_a) + sizeof(msg_b) + sizeof(u32) +
- ticket_blob_len + 16;
- dout(" need len %d\n", len);
- if (au->buf && au->buf->alloc_len < len) {
+ maxlen = sizeof(*msg_a) + sizeof(msg_b) +
+ ceph_x_encrypt_buflen(ticket_blob_len);
+ dout(" need len %d\n", maxlen);
+ if (au->buf && au->buf->alloc_len < maxlen) {
ceph_buffer_put(au->buf);
au->buf = NULL;
}
if (!au->buf) {
- au->buf = ceph_buffer_new(len, GFP_NOFS);
+ au->buf = ceph_buffer_new(maxlen, GFP_NOFS);
if (!au->buf)
return -ENOMEM;
}
@@ -296,6 +319,7 @@ static int ceph_x_build_authorizer(struct ceph_auth_client *ac,
au->buf->vec.iov_len = p - au->buf->vec.iov_base;
dout(" built authorizer nonce %llx len %d\n", au->nonce,
(int)au->buf->vec.iov_len);
+ BUG_ON(au->buf->vec.iov_len > maxlen);
return 0;
out_buf:
diff --git a/fs/ceph/buffer.c b/fs/ceph/buffer.c
index b98086c7aeb..c67535d70aa 100644
--- a/fs/ceph/buffer.c
+++ b/fs/ceph/buffer.c
@@ -1,5 +1,8 @@
#include "ceph_debug.h"
+
+#include <linux/slab.h>
+
#include "buffer.h"
#include "decode.h"
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index db122bb357b..aa2239fa9a3 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -3,6 +3,7 @@
#include <linux/fs.h>
#include <linux/kernel.h>
#include <linux/sched.h>
+#include <linux/slab.h>
#include <linux/vmalloc.h>
#include <linux/wait.h>
#include <linux/writeback.h>
@@ -1204,6 +1205,12 @@ retry:
if (capsnap->dirty_pages || capsnap->writing)
continue;
+ /*
+ * if cap writeback already occurred, we should have dropped
+ * the capsnap in ceph_put_wrbuffer_cap_refs.
+ */
+ BUG_ON(capsnap->dirty == 0);
+
/* pick mds, take s_mutex */
mds = __ceph_get_cap_mds(ci, &mseq);
if (session && session->s_mds != mds) {
@@ -1407,6 +1414,7 @@ static int try_nonblocking_invalidate(struct inode *inode)
*/
void ceph_check_caps(struct ceph_inode_info *ci, int flags,
struct ceph_mds_session *session)
+ __releases(session->s_mutex)
{
struct ceph_client *client = ceph_inode_to_client(&ci->vfs_inode);
struct ceph_mds_client *mdsc = &client->mdsc;
@@ -1414,7 +1422,6 @@ void ceph_check_caps(struct ceph_inode_info *ci, int flags,
struct ceph_cap *cap;
int file_wanted, used;
int took_snap_rwsem = 0; /* true if mdsc->snap_rwsem held */
- int drop_session_lock = session ? 0 : 1;
int issued, implemented, want, retain, revoking, flushing = 0;
int mds = -1; /* keep track of how far we've gone through i_caps list
to avoid an infinite loop on retry */
@@ -1639,7 +1646,7 @@ ack:
if (queue_invalidate)
ceph_queue_invalidate(inode);
- if (session && drop_session_lock)
+ if (session)
mutex_unlock(&session->s_mutex);
if (took_snap_rwsem)
up_read(&mdsc->snap_rwsem);
@@ -2117,8 +2124,8 @@ void ceph_put_cap_refs(struct ceph_inode_info *ci, int had)
}
spin_unlock(&inode->i_lock);
- dout("put_cap_refs %p had %s %s\n", inode, ceph_cap_string(had),
- last ? "last" : "");
+ dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had),
+ last ? " last" : "", put ? " put" : "");
if (last && !flushsnaps)
ceph_check_caps(ci, 0, NULL);
@@ -2142,7 +2149,8 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
{
struct inode *inode = &ci->vfs_inode;
int last = 0;
- int last_snap = 0;
+ int complete_capsnap = 0;
+ int drop_capsnap = 0;
int found = 0;
struct ceph_cap_snap *capsnap = NULL;
@@ -2165,19 +2173,32 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
list_for_each_entry(capsnap, &ci->i_cap_snaps, ci_item) {
if (capsnap->context == snapc) {
found = 1;
- capsnap->dirty_pages -= nr;
- last_snap = !capsnap->dirty_pages;
break;
}
}
BUG_ON(!found);
+ capsnap->dirty_pages -= nr;
+ if (capsnap->dirty_pages == 0) {
+ complete_capsnap = 1;
+ if (capsnap->dirty == 0)
+ /* cap writeback completed before we created
+ * the cap_snap; no FLUSHSNAP is needed */
+ drop_capsnap = 1;
+ }
dout("put_wrbuffer_cap_refs on %p cap_snap %p "
- " snap %lld %d/%d -> %d/%d %s%s\n",
+ " snap %lld %d/%d -> %d/%d %s%s%s\n",
inode, capsnap, capsnap->context->seq,
ci->i_wrbuffer_ref+nr, capsnap->dirty_pages + nr,
ci->i_wrbuffer_ref, capsnap->dirty_pages,
last ? " (wrbuffer last)" : "",
- last_snap ? " (capsnap last)" : "");
+ complete_capsnap ? " (complete capsnap)" : "",
+ drop_capsnap ? " (drop capsnap)" : "");
+ if (drop_capsnap) {
+ ceph_put_snap_context(capsnap->context);
+ list_del(&capsnap->ci_item);
+ list_del(&capsnap->flushing_item);
+ ceph_put_cap_snap(capsnap);
+ }
}
spin_unlock(&inode->i_lock);
@@ -2185,28 +2206,31 @@ void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr,
if (last) {
ceph_check_caps(ci, CHECK_CAPS_AUTHONLY, NULL);
iput(inode);
- } else if (last_snap) {
+ } else if (complete_capsnap) {
ceph_flush_snaps(ci);
wake_up(&ci->i_cap_wq);
}
+ if (drop_capsnap)
+ iput(inode);
}
/*
* Handle a cap GRANT message from the MDS. (Note that a GRANT may
* actually be a revocation if it specifies a smaller cap set.)
*
- * caller holds s_mutex.
+ * caller holds s_mutex and i_lock, we drop both.
+ *
* return value:
* 0 - ok
* 1 - check_caps on auth cap only (writeback)
* 2 - check_caps (ack revoke)
*/
-static int handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
- struct ceph_mds_session *session,
- struct ceph_cap *cap,
- struct ceph_buffer *xattr_buf)
+static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
+ struct ceph_mds_session *session,
+ struct ceph_cap *cap,
+ struct ceph_buffer *xattr_buf)
__releases(inode->i_lock)
-
+ __releases(session->s_mutex)
{
struct ceph_inode_info *ci = ceph_inode(inode);
int mds = session->s_mds;
@@ -2216,7 +2240,7 @@ static int handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
u64 size = le64_to_cpu(grant->size);
u64 max_size = le64_to_cpu(grant->max_size);
struct timespec mtime, atime, ctime;
- int reply = 0;
+ int check_caps = 0;
int wake = 0;
int writeback = 0;
int revoked_rdcache = 0;
@@ -2329,11 +2353,12 @@ static int handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
if ((used & ~newcaps) & CEPH_CAP_FILE_BUFFER)
writeback = 1; /* will delay ack */
else if (dirty & ~newcaps)
- reply = 1; /* initiate writeback in check_caps */
+ check_caps = 1; /* initiate writeback in check_caps */
else if (((used & ~newcaps) & CEPH_CAP_FILE_CACHE) == 0 ||
revoked_rdcache)
- reply = 2; /* send revoke ack in check_caps */
+ check_caps = 2; /* send revoke ack in check_caps */
cap->issued = newcaps;
+ cap->implemented |= newcaps;
} else if (cap->issued == newcaps) {
dout("caps unchanged: %s -> %s\n",
ceph_cap_string(cap->issued), ceph_cap_string(newcaps));
@@ -2346,6 +2371,7 @@ static int handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
* pending revocation */
wake = 1;
}
+ BUG_ON(cap->issued & ~cap->implemented);
spin_unlock(&inode->i_lock);
if (writeback)
@@ -2359,7 +2385,14 @@ static int handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant,
ceph_queue_invalidate(inode);
if (wake)
wake_up(&ci->i_cap_wq);
- return reply;
+
+ if (check_caps == 1)
+ ceph_check_caps(ci, CHECK_CAPS_NODELAY|CHECK_CAPS_AUTHONLY,
+ session);
+ else if (check_caps == 2)
+ ceph_check_caps(ci, CHECK_CAPS_NODELAY, session);
+ else
+ mutex_unlock(&session->s_mutex);
}
/*
@@ -2454,8 +2487,8 @@ static void handle_cap_flushsnap_ack(struct inode *inode, u64 flush_tid,
break;
}
WARN_ON(capsnap->dirty_pages || capsnap->writing);
- dout(" removing cap_snap %p follows %lld\n",
- capsnap, follows);
+ dout(" removing %p cap_snap %p follows %lld\n",
+ inode, capsnap, follows);
ceph_put_snap_context(capsnap->context);
list_del(&capsnap->ci_item);
list_del(&capsnap->flushing_item);
@@ -2548,9 +2581,8 @@ static void handle_cap_export(struct inode *inode, struct ceph_mds_caps *ex,
ci->i_cap_exporting_issued = cap->issued;
}
__ceph_remove_cap(cap);
- } else {
- WARN_ON(!cap);
}
+ /* else, we already released it */
spin_unlock(&inode->i_lock);
}
@@ -2621,9 +2653,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
u64 cap_id;
u64 size, max_size;
u64 tid;
- int check_caps = 0;
void *snaptrace;
- int r;
dout("handle_caps from mds%d\n", mds);
@@ -2668,8 +2698,9 @@ void ceph_handle_caps(struct ceph_mds_session *session,
case CEPH_CAP_OP_IMPORT:
handle_cap_import(mdsc, inode, h, session,
snaptrace, le32_to_cpu(h->snap_trace_len));
- check_caps = 1; /* we may have sent a RELEASE to the old auth */
- goto done;
+ ceph_check_caps(ceph_inode(inode), CHECK_CAPS_NODELAY,
+ session);
+ goto done_unlocked;
}
/* the rest require a cap */
@@ -2686,16 +2717,8 @@ void ceph_handle_caps(struct ceph_mds_session *session,
switch (op) {
case CEPH_CAP_OP_REVOKE:
case CEPH_CAP_OP_GRANT:
- r = handle_cap_grant(inode, h, session, cap, msg->middle);
- if (r == 1)
- ceph_check_caps(ceph_inode(inode),
- CHECK_CAPS_NODELAY|CHECK_CAPS_AUTHONLY,
- session);
- else if (r == 2)
- ceph_check_caps(ceph_inode(inode),
- CHECK_CAPS_NODELAY,
- session);
- break;
+ handle_cap_grant(inode, h, session, cap, msg->middle);
+ goto done_unlocked;
case CEPH_CAP_OP_FLUSH_ACK:
handle_cap_flush_ack(inode, tid, h, session, cap);
@@ -2713,9 +2736,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
done:
mutex_unlock(&session->s_mutex);
-
- if (check_caps)
- ceph_check_caps(ceph_inode(inode), CHECK_CAPS_NODELAY, NULL);
+done_unlocked:
if (inode)
iput(inode);
return;
@@ -2838,11 +2859,18 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
struct ceph_cap *cap;
struct ceph_mds_request_release *rel = *p;
int ret = 0;
-
- dout("encode_inode_release %p mds%d drop %s unless %s\n", inode,
- mds, ceph_cap_string(drop), ceph_cap_string(unless));
+ int used = 0;
spin_lock(&inode->i_lock);
+ used = __ceph_caps_used(ci);
+
+ dout("encode_inode_release %p mds%d used %s drop %s unless %s\n", inode,
+ mds, ceph_cap_string(used), ceph_cap_string(drop),
+ ceph_cap_string(unless));
+
+ /* only drop unused caps */
+ drop &= ~used;
+
cap = __get_cap_for_mds(ci, mds);
if (cap && __cap_is_valid(cap)) {
if (force ||
diff --git a/fs/ceph/crypto.c b/fs/ceph/crypto.c
index 291ac288e79..f704b3b6242 100644
--- a/fs/ceph/crypto.c
+++ b/fs/ceph/crypto.c
@@ -3,6 +3,7 @@
#include <linux/err.h>
#include <linux/scatterlist.h>
+#include <linux/slab.h>
#include <crypto/hash.h>
#include "crypto.h"
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index e159f141511..f7048da92ac 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -1,6 +1,7 @@
#include "ceph_debug.h"
#include <linux/device.h>
+#include <linux/slab.h>
#include <linux/module.h>
#include <linux/ctype.h>
#include <linux/debugfs.h>
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 5107384ee02..ea8ee2e526a 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -3,6 +3,7 @@
#include <linux/spinlock.h>
#include <linux/fs_struct.h>
#include <linux/namei.h>
+#include <linux/slab.h>
#include <linux/sched.h>
#include "super.h"
@@ -170,11 +171,11 @@ more:
spin_lock(&inode->i_lock);
spin_lock(&dcache_lock);
+ last = dentry;
+
if (err < 0)
goto out_unlock;
- last = dentry;
-
p = p->prev;
filp->f_pos++;
@@ -288,8 +289,10 @@ more:
CEPH_MDS_OP_LSSNAP : CEPH_MDS_OP_READDIR;
/* discard old result, if any */
- if (fi->last_readdir)
+ if (fi->last_readdir) {
ceph_mdsc_put_request(fi->last_readdir);
+ fi->last_readdir = NULL;
+ }
/* requery frag tree, as the frag topology may have changed */
frag = ceph_choose_frag(ceph_inode(inode), frag, NULL, NULL);
@@ -309,7 +312,7 @@ more:
req->r_readdir_offset = fi->next_offset;
req->r_args.readdir.frag = cpu_to_le32(frag);
req->r_args.readdir.max_entries = cpu_to_le32(max_entries);
- req->r_num_caps = max_entries;
+ req->r_num_caps = max_entries + 1;
err = ceph_mdsc_do_request(mdsc, NULL, req);
if (err < 0) {
ceph_mdsc_put_request(req);
@@ -486,6 +489,7 @@ struct dentry *ceph_finish_lookup(struct ceph_mds_request *req,
struct inode *inode = ceph_get_snapdir(parent);
dout("ENOENT on snapdir %p '%.*s', linking to snapdir %p\n",
dentry, dentry->d_name.len, dentry->d_name.name, inode);
+ BUG_ON(!d_unhashed(dentry));
d_add(dentry, inode);
err = 0;
}
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index fc68e39cbad..9d67572fb32 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -1,6 +1,7 @@
#include "ceph_debug.h"
#include <linux/exportfs.h>
+#include <linux/slab.h>
#include <asm/unaligned.h>
#include "super.h"
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 5d2af8464f6..4add3d5da2c 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1,6 +1,7 @@
#include "ceph_debug.h"
#include <linux/sched.h>
+#include <linux/slab.h>
#include <linux/file.h>
#include <linux/namei.h>
#include <linux/writeback.h>
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 7abe1aed819..26f883c275e 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -378,6 +378,22 @@ void ceph_destroy_inode(struct inode *inode)
ceph_queue_caps_release(inode);
+ /*
+ * we may still have a snap_realm reference if there are stray
+ * caps in i_cap_exporting_issued or i_snap_caps.
+ */
+ if (ci->i_snap_realm) {
+ struct ceph_mds_client *mdsc =
+ &ceph_client(ci->vfs_inode.i_sb)->mdsc;
+ struct ceph_snap_realm *realm = ci->i_snap_realm;
+
+ dout(" dropping residual ref to snap realm %p\n", realm);
+ spin_lock(&realm->inodes_with_caps_lock);
+ list_del_init(&ci->i_snap_realm_item);
+ spin_unlock(&realm->inodes_with_caps_lock);
+ ceph_put_snap_realm(mdsc, realm);
+ }
+
kfree(ci->i_symlink);
while ((n = rb_first(&ci->i_fragtree)) != NULL) {
frag = rb_entry(n, struct ceph_inode_frag, node);
@@ -870,6 +886,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
struct inode *in = NULL;
struct ceph_mds_reply_inode *ininfo;
struct ceph_vino vino;
+ struct ceph_client *client = ceph_sb_to_client(sb);
int i = 0;
int err = 0;
@@ -933,7 +950,14 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req,
return err;
}
- if (rinfo->head->is_dentry && !req->r_aborted) {
+ /*
+ * ignore null lease/binding on snapdir ENOENT, or else we
+ * will have trouble splicing in the virtual snapdir later
+ */
+ if (rinfo->head->is_dentry && !req->r_aborted &&
+ (rinfo->head->is_target || strncmp(req->r_dentry->d_name.name,
+ client->mount_args->snapdir_name,
+ req->r_dentry->d_name.len))) {
/*
* lookup link rename : null -> possibly existing inode
* mknod symlink mkdir : null -> new inode
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index a2600101ec2..60a9a4ae47b 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1,6 +1,7 @@
#include "ceph_debug.h"
#include <linux/wait.h>
+#include <linux/slab.h>
#include <linux/sched.h>
#include "mds_client.h"
@@ -328,6 +329,8 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
struct ceph_mds_session *s;
s = kzalloc(sizeof(*s), GFP_NOFS);
+ if (!s)
+ return ERR_PTR(-ENOMEM);
s->s_mdsc = mdsc;
s->s_mds = mds;
s->s_state = CEPH_MDS_SESSION_NEW;
@@ -529,7 +532,7 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
{
dout("__unregister_request %p tid %lld\n", req, req->r_tid);
rb_erase(&req->r_node, &mdsc->request_tree);
- ceph_mdsc_put_request(req);
+ RB_CLEAR_NODE(&req->r_node);
if (req->r_unsafe_dir) {
struct ceph_inode_info *ci = ceph_inode(req->r_unsafe_dir);
@@ -538,6 +541,8 @@ static void __unregister_request(struct ceph_mds_client *mdsc,
list_del_init(&req->r_unsafe_dir_item);
spin_unlock(&ci->i_unsafe_lock);
}
+
+ ceph_mdsc_put_request(req);
}
/*
@@ -862,6 +867,7 @@ static int send_renew_caps(struct ceph_mds_client *mdsc,
if (time_after_eq(jiffies, session->s_cap_ttl) &&
time_after_eq(session->s_cap_ttl, session->s_renew_requested))
pr_info("mds%d caps stale\n", session->s_mds);
+ session->s_renew_requested = jiffies;
/* do not try to renew caps until a recovering mds has reconnected
* with its clients. */
@@ -874,7 +880,6 @@ static int send_renew_caps(struct ceph_mds_client *mdsc,
dout("send_renew_caps to mds%d (%s)\n", session->s_mds,
ceph_mds_state_name(state));
- session->s_renew_requested = jiffies;
msg = create_session_msg(CEPH_SESSION_REQUEST_RENEWCAPS,
++session->s_renew_seq);
if (IS_ERR(msg))
@@ -1566,8 +1571,13 @@ static int __do_request(struct ceph_mds_client *mdsc,
/* get, open session */
session = __ceph_lookup_mds_session(mdsc, mds);
- if (!session)
+ if (!session) {
session = register_session(mdsc, mds);
+ if (IS_ERR(session)) {
+ err = PTR_ERR(session);
+ goto finish;
+ }
+ }
dout("do_request mds%d session %p state %s\n", mds, session,
session_state_name(session->s_state));
if (session->s_state != CEPH_MDS_SESSION_OPEN &&
@@ -1770,7 +1780,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg)
dout("handle_reply %p\n", req);
/* correct session? */
- if (!req->r_session && req->r_session != session) {
+ if (req->r_session != session) {
pr_err("mdsc_handle_reply got %llu on session mds%d"
" not mds%d\n", tid, session->s_mds,
req->r_session ? req->r_session->s_mds : -1);
@@ -2682,29 +2692,41 @@ void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
*/
static void wait_unsafe_requests(struct ceph_mds_client *mdsc, u64 want_tid)
{
- struct ceph_mds_request *req = NULL;
+ struct ceph_mds_request *req = NULL, *nextreq;
struct rb_node *n;
mutex_lock(&mdsc->mutex);
dout("wait_unsafe_requests want %lld\n", want_tid);
+restart:
req = __get_oldest_req(mdsc);
while (req && req->r_tid <= want_tid) {
+ /* find next request */
+ n = rb_next(&req->r_node);
+ if (n)
+ nextreq = rb_entry(n, struct ceph_mds_request, r_node);
+ else
+ nextreq = NULL;
if ((req->r_op & CEPH_MDS_OP_WRITE)) {
/* write op */
ceph_mdsc_get_request(req);
+ if (nextreq)
+ ceph_mdsc_get_request(nextreq);
mutex_unlock(&mdsc->mutex);
dout("wait_unsafe_requests wait on %llu (want %llu)\n",
req->r_tid, want_tid);
wait_for_completion(&req->r_safe_completion);
mutex_lock(&mdsc->mutex);
- n = rb_next(&req->r_node);
ceph_mdsc_put_request(req);
- } else {
- n = rb_next(&req->r_node);
+ if (!nextreq)
+ break; /* next dne before, so we're done! */
+ if (RB_EMPTY_NODE(&nextreq->r_node)) {
+ /* next request was removed from tree */
+ ceph_mdsc_put_request(nextreq);
+ goto restart;
+ }
+ ceph_mdsc_put_request(nextreq); /* won't go away */
}
- if (!n)
- break;
- req = rb_entry(n, struct ceph_mds_request, r_node);
+ req = nextreq;
}
mutex_unlock(&mdsc->mutex);
dout("wait_unsafe_requests done\n");
diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
index 781656a49bf..cdaaa131add 100644
--- a/fs/ceph/messenger.c
+++ b/fs/ceph/messenger.c
@@ -6,6 +6,7 @@
#include <linux/inet.h>
#include <linux/kthread.h>
#include <linux/net.h>
+#include <linux/slab.h>
#include <linux/socket.h>
#include <linux/string.h>
#include <net/tcp.h>
@@ -29,6 +30,10 @@ static char tag_msg = CEPH_MSGR_TAG_MSG;
static char tag_ack = CEPH_MSGR_TAG_ACK;
static char tag_keepalive = CEPH_MSGR_TAG_KEEPALIVE;
+#ifdef CONFIG_LOCKDEP
+static struct lock_class_key socket_class;
+#endif
+
static void queue_con(struct ceph_connection *con);
static void con_work(struct work_struct *);
@@ -227,6 +232,10 @@ static struct socket *ceph_tcp_connect(struct ceph_connection *con)
con->sock = sock;
sock->sk->sk_allocation = GFP_NOFS;
+#ifdef CONFIG_LOCKDEP
+ lockdep_set_class(&sock->sk->sk_lock, &socket_class);
+#endif
+
set_sock_callbacks(sock, con);
dout("connect %s\n", pr_addr(&con->peer_addr.in_addr));
@@ -332,6 +341,7 @@ static void reset_connection(struct ceph_connection *con)
con->out_msg = NULL;
}
con->in_seq = 0;
+ con->in_seq_acked = 0;
}
/*
@@ -366,6 +376,14 @@ void ceph_con_open(struct ceph_connection *con, struct ceph_entity_addr *addr)
}
/*
+ * return true if this connection ever successfully opened
+ */
+bool ceph_con_opened(struct ceph_connection *con)
+{
+ return con->connect_seq > 0;
+}
+
+/*
* generic get/put
*/
struct ceph_connection *ceph_con_get(struct ceph_connection *con)
@@ -830,13 +848,6 @@ static void prepare_read_connect(struct ceph_connection *con)
con->in_base_pos = 0;
}
-static void prepare_read_connect_retry(struct ceph_connection *con)
-{
- dout("prepare_read_connect_retry %p\n", con);
- con->in_base_pos = strlen(CEPH_BANNER) + sizeof(con->actual_peer_addr)
- + sizeof(con->peer_addr_for_me);
-}
-
static void prepare_read_ack(struct ceph_connection *con)
{
dout("prepare_read_ack %p\n", con);
@@ -1146,7 +1157,7 @@ static int process_connect(struct ceph_connection *con)
}
con->auth_retry = 1;
prepare_write_connect(con->msgr, con, 0);
- prepare_read_connect_retry(con);
+ prepare_read_connect(con);
break;
case CEPH_MSGR_TAG_RESETSESSION:
@@ -1843,8 +1854,6 @@ static void ceph_fault(struct ceph_connection *con)
goto out;
}
- clear_bit(BUSY, &con->state); /* to avoid an improbable race */
-
mutex_lock(&con->mutex);
if (test_bit(CLOSED, &con->state))
goto out_unlock;
diff --git a/fs/ceph/messenger.h b/fs/ceph/messenger.h
index 4caaa591111..a343dae73cd 100644
--- a/fs/ceph/messenger.h
+++ b/fs/ceph/messenger.h
@@ -223,6 +223,7 @@ extern void ceph_con_init(struct ceph_messenger *msgr,
struct ceph_connection *con);
extern void ceph_con_open(struct ceph_connection *con,
struct ceph_entity_addr *addr);
+extern bool ceph_con_opened(struct ceph_connection *con);
extern void ceph_con_close(struct ceph_connection *con);
extern void ceph_con_send(struct ceph_connection *con, struct ceph_msg *msg);
extern void ceph_con_revoke(struct ceph_connection *con, struct ceph_msg *msg);
diff --git a/fs/ceph/mon_client.c b/fs/ceph/mon_client.c
index 890597c09d4..8fdc011ca95 100644
--- a/fs/ceph/mon_client.c
+++ b/fs/ceph/mon_client.c
@@ -1,6 +1,7 @@
#include "ceph_debug.h"
#include <linux/types.h>
+#include <linux/slab.h>
#include <linux/random.h>
#include <linux/sched.h>
diff --git a/fs/ceph/osd_client.c b/fs/ceph/osd_client.c
index dbe63db9762..c7b4dedaace 100644
--- a/fs/ceph/osd_client.c
+++ b/fs/ceph/osd_client.c
@@ -413,11 +413,22 @@ static void remove_old_osds(struct ceph_osd_client *osdc, int remove_all)
*/
static int __reset_osd(struct ceph_osd_client *osdc, struct ceph_osd *osd)
{
+ struct ceph_osd_request *req;
int ret = 0;
dout("__reset_osd %p osd%d\n", osd, osd->o_osd);
if (list_empty(&osd->o_requests)) {
__remove_osd(osdc, osd);
+ } else if (memcmp(&osdc->osdmap->osd_addr[osd->o_osd],
+ &osd->o_con.peer_addr,
+ sizeof(osd->o_con.peer_addr)) == 0 &&
+ !ceph_con_opened(&osd->o_con)) {
+ dout(" osd addr hasn't changed and connection never opened,"
+ " letting msgr retry");
+ /* touch each r_stamp for handle_timeout()'s benfit */
+ list_for_each_entry(req, &osd->o_requests, r_osd_item)
+ req->r_stamp = jiffies;
+ ret = -EAGAIN;
} else {
ceph_con_close(&osd->o_con);
ceph_con_open(&osd->o_con, &osdc->osdmap->osd_addr[osd->o_osd]);
@@ -633,7 +644,7 @@ static int __send_request(struct ceph_osd_client *osdc,
reqhead->flags |= cpu_to_le32(req->r_flags); /* e.g., RETRY */
reqhead->reassert_version = req->r_reassert_version;
- req->r_sent_stamp = jiffies;
+ req->r_stamp = jiffies;
list_move_tail(&osdc->req_lru, &req->r_req_lru_item);
ceph_msg_get(req->r_request); /* send consumes a ref */
@@ -660,7 +671,7 @@ static void handle_timeout(struct work_struct *work)
unsigned long timeout = osdc->client->mount_args->osd_timeout * HZ;
unsigned long keepalive =
osdc->client->mount_args->osd_keepalive_timeout * HZ;
- unsigned long last_sent = 0;
+ unsigned long last_stamp = 0;
struct rb_node *p;
struct list_head slow_osds;
@@ -697,12 +708,12 @@ static void handle_timeout(struct work_struct *work)
req = list_entry(osdc->req_lru.next, struct ceph_osd_request,
r_req_lru_item);
- if (time_before(jiffies, req->r_sent_stamp + timeout))
+ if (time_before(jiffies, req->r_stamp + timeout))
break;
- BUG_ON(req == last_req && req->r_sent_stamp == last_sent);
+ BUG_ON(req == last_req && req->r_stamp == last_stamp);
last_req = req;
- last_sent = req->r_sent_stamp;
+ last_stamp = req->r_stamp;
osd = req->r_osd;
BUG_ON(!osd);
@@ -718,7 +729,7 @@ static void handle_timeout(struct work_struct *work)
*/
INIT_LIST_HEAD(&slow_osds);
list_for_each_entry(req, &osdc->req_lru, r_req_lru_item) {
- if (time_before(jiffies, req->r_sent_stamp + keepalive))
+ if (time_before(jiffies, req->r_stamp + keepalive))
break;
osd = req->r_osd;
@@ -862,7 +873,9 @@ static int __kick_requests(struct ceph_osd_client *osdc,
dout("kick_requests osd%d\n", kickosd ? kickosd->o_osd : -1);
if (kickosd) {
- __reset_osd(osdc, kickosd);
+ err = __reset_osd(osdc, kickosd);
+ if (err == -EAGAIN)
+ return 1;
} else {
for (p = rb_first(&osdc->osds); p; p = n) {
struct ceph_osd *osd =
@@ -913,7 +926,7 @@ static int __kick_requests(struct ceph_osd_client *osdc,
kick:
dout("kicking %p tid %llu osd%d\n", req, req->r_tid,
- req->r_osd->o_osd);
+ req->r_osd ? req->r_osd->o_osd : -1);
req->r_flags |= CEPH_OSD_FLAG_RETRY;
err = __send_request(osdc, req);
if (err) {
diff --git a/fs/ceph/osd_client.h b/fs/ceph/osd_client.h
index 1b1a3ca43af..b0759911e7c 100644
--- a/fs/ceph/osd_client.h
+++ b/fs/ceph/osd_client.h
@@ -70,7 +70,7 @@ struct ceph_osd_request {
char r_oid[40]; /* object name */
int r_oid_len;
- unsigned long r_sent_stamp;
+ unsigned long r_stamp; /* send OR check time */
bool r_resend; /* msg send failed, needs retry */
struct ceph_file_layout r_file_layout;
diff --git a/fs/ceph/osdmap.c b/fs/ceph/osdmap.c
index b83f2692b83..2e2c15eed82 100644
--- a/fs/ceph/osdmap.c
+++ b/fs/ceph/osdmap.c
@@ -1,4 +1,7 @@
+#include "ceph_debug.h"
+
+#include <linux/slab.h>
#include <asm/div64.h>
#include "super.h"
@@ -6,7 +9,6 @@
#include "crush/hash.h"
#include "crush/mapper.h"
#include "decode.h"
-#include "ceph_debug.h"
char *ceph_osdmap_state_str(char *str, int len, int state)
{
@@ -312,71 +314,6 @@ bad:
return ERR_PTR(err);
}
-
-/*
- * osd map
- */
-void ceph_osdmap_destroy(struct ceph_osdmap *map)
-{
- dout("osdmap_destroy %p\n", map);
- if (map->crush)
- crush_destroy(map->crush);
- while (!RB_EMPTY_ROOT(&map->pg_temp)) {
- struct ceph_pg_mapping *pg =
- rb_entry(rb_first(&map->pg_temp),
- struct ceph_pg_mapping, node);
- rb_erase(&pg->node, &map->pg_temp);
- kfree(pg);
- }
- while (!RB_EMPTY_ROOT(&map->pg_pools)) {
- struct ceph_pg_pool_info *pi =
- rb_entry(rb_first(&map->pg_pools),
- struct ceph_pg_pool_info, node);
- rb_erase(&pi->node, &map->pg_pools);
- kfree(pi);
- }
- kfree(map->osd_state);
- kfree(map->osd_weight);
- kfree(map->osd_addr);
- kfree(map);
-}
-
-/*
- * adjust max osd value. reallocate arrays.
- */
-static int osdmap_set_max_osd(struct ceph_osdmap *map, int max)
-{
- u8 *state;
- struct ceph_entity_addr *addr;
- u32 *weight;
-
- state = kcalloc(max, sizeof(*state), GFP_NOFS);
- addr = kcalloc(max, sizeof(*addr), GFP_NOFS);
- weight = kcalloc(max, sizeof(*weight), GFP_NOFS);
- if (state == NULL || addr == NULL || weight == NULL) {
- kfree(state);
- kfree(addr);
- kfree(weight);
- return -ENOMEM;
- }
-
- /* copy old? */
- if (map->osd_state) {
- memcpy(state, map->osd_state, map->max_osd*sizeof(*state));
- memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr));
- memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight));
- kfree(map->osd_state);
- kfree(map->osd_addr);
- kfree(map->osd_weight);
- }
-
- map->osd_state = state;
- map->osd_weight = weight;
- map->osd_addr = addr;
- map->max_osd = max;
- return 0;
-}
-
/*
* rbtree of pg_mapping for handling pg_temp (explicit mapping of pgid
* to a set of osds)
@@ -480,6 +417,113 @@ static struct ceph_pg_pool_info *__lookup_pg_pool(struct rb_root *root, int id)
return NULL;
}
+static void __remove_pg_pool(struct rb_root *root, struct ceph_pg_pool_info *pi)
+{
+ rb_erase(&pi->node, root);
+ kfree(pi->name);
+ kfree(pi);
+}
+
+void __decode_pool(void **p, struct ceph_pg_pool_info *pi)
+{
+ ceph_decode_copy(p, &pi->v, sizeof(pi->v));
+ calc_pg_masks(pi);
+ *p += le32_to_cpu(pi->v.num_snaps) * sizeof(u64);
+ *p += le32_to_cpu(pi->v.num_removed_snap_intervals) * sizeof(u64) * 2;
+}
+
+static int __decode_pool_names(void **p, void *end, struct ceph_osdmap *map)
+{
+ struct ceph_pg_pool_info *pi;
+ u32 num, len, pool;
+
+ ceph_decode_32_safe(p, end, num, bad);
+ dout(" %d pool names\n", num);
+ while (num--) {
+ ceph_decode_32_safe(p, end, pool, bad);
+ ceph_decode_32_safe(p, end, len, bad);
+ dout(" pool %d len %d\n", pool, len);
+ pi = __lookup_pg_pool(&map->pg_pools, pool);
+ if (pi) {
+ kfree(pi->name);
+ pi->name = kmalloc(len + 1, GFP_NOFS);
+ if (pi->name) {
+ memcpy(pi->name, *p, len);
+ pi->name[len] = '\0';
+ dout(" name is %s\n", pi->name);
+ }
+ }
+ *p += len;
+ }
+ return 0;
+
+bad:
+ return -EINVAL;
+}
+
+/*
+ * osd map
+ */
+void ceph_osdmap_destroy(struct ceph_osdmap *map)
+{
+ dout("osdmap_destroy %p\n", map);
+ if (map->crush)
+ crush_destroy(map->crush);
+ while (!RB_EMPTY_ROOT(&map->pg_temp)) {
+ struct ceph_pg_mapping *pg =
+ rb_entry(rb_first(&map->pg_temp),
+ struct ceph_pg_mapping, node);
+ rb_erase(&pg->node, &map->pg_temp);
+ kfree(pg);
+ }
+ while (!RB_EMPTY_ROOT(&map->pg_pools)) {
+ struct ceph_pg_pool_info *pi =
+ rb_entry(rb_first(&map->pg_pools),
+ struct ceph_pg_pool_info, node);
+ __remove_pg_pool(&map->pg_pools, pi);
+ }
+ kfree(map->osd_state);
+ kfree(map->osd_weight);
+ kfree(map->osd_addr);
+ kfree(map);
+}
+
+/*
+ * adjust max osd value. reallocate arrays.
+ */
+static int osdmap_set_max_osd(struct ceph_osdmap *map, int max)
+{
+ u8 *state;
+ struct ceph_entity_addr *addr;
+ u32 *weight;
+
+ state = kcalloc(max, sizeof(*state), GFP_NOFS);
+ addr = kcalloc(max, sizeof(*addr), GFP_NOFS);
+ weight = kcalloc(max, sizeof(*weight), GFP_NOFS);
+ if (state == NULL || addr == NULL || weight == NULL) {
+ kfree(state);
+ kfree(addr);
+ kfree(weight);
+ return -ENOMEM;
+ }
+
+ /* copy old? */
+ if (map->osd_state) {
+ memcpy(state, map->osd_state, map->max_osd*sizeof(*state));
+ memcpy(addr, map->osd_addr, map->max_osd*sizeof(*addr));
+ memcpy(weight, map->osd_weight, map->max_osd*sizeof(*weight));
+ kfree(map->osd_state);
+ kfree(map->osd_addr);
+ kfree(map->osd_weight);
+ }
+
+ map->osd_state = state;
+ map->osd_weight = weight;
+ map->osd_addr = addr;
+ map->max_osd = max;
+ return 0;
+}
+
/*
* decode a full map.
*/
@@ -516,7 +560,7 @@ 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 + 1 + sizeof(pi->v), bad);
- pi = kmalloc(sizeof(*pi), GFP_NOFS);
+ pi = kzalloc(sizeof(*pi), GFP_NOFS);
if (!pi)
goto bad;
pi->id = ceph_decode_32(p);
@@ -526,13 +570,13 @@ struct ceph_osdmap *osdmap_decode(void **p, void *end)
ev, CEPH_PG_POOL_VERSION);
goto bad;
}
- ceph_decode_copy(p, &pi->v, sizeof(pi->v));
+ __decode_pool(p, pi);
__insert_pg_pool(&map->pg_pools, pi);
- calc_pg_masks(pi);
- *p += le32_to_cpu(pi->v.num_snaps) * sizeof(u64);
- *p += le32_to_cpu(pi->v.num_removed_snap_intervals)
- * sizeof(u64) * 2;
}
+
+ if (version >= 5 && __decode_pool_names(p, end, map) < 0)
+ goto bad;
+
ceph_decode_32_safe(p, end, map->pool_max, bad);
ceph_decode_32_safe(p, end, map->flags, bad);
@@ -706,7 +750,7 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
}
pi = __lookup_pg_pool(&map->pg_pools, pool);
if (!pi) {
- pi = kmalloc(sizeof(*pi), GFP_NOFS);
+ pi = kzalloc(sizeof(*pi), GFP_NOFS);
if (!pi) {
err = -ENOMEM;
goto bad;
@@ -714,9 +758,10 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
pi->id = pool;
__insert_pg_pool(&map->pg_pools, pi);
}
- ceph_decode_copy(p, &pi->v, sizeof(pi->v));
- calc_pg_masks(pi);
+ __decode_pool(p, pi);
}
+ if (version >= 5 && __decode_pool_names(p, end, map) < 0)
+ goto bad;
/* old_pool */
ceph_decode_32_safe(p, end, len, bad);
@@ -725,10 +770,8 @@ struct ceph_osdmap *osdmap_apply_incremental(void **p, void *end,
ceph_decode_32_safe(p, end, pool, bad);
pi = __lookup_pg_pool(&map->pg_pools, pool);
- if (pi) {
- rb_erase(&pi->node, &map->pg_pools);
- kfree(pi);
- }
+ if (pi)
+ __remove_pg_pool(&map->pg_pools, pi);
}
/* new_up */
diff --git a/fs/ceph/osdmap.h b/fs/ceph/osdmap.h
index 1fb55afb264..8bc9f1e4f56 100644
--- a/fs/ceph/osdmap.h
+++ b/fs/ceph/osdmap.h
@@ -23,6 +23,7 @@ struct ceph_pg_pool_info {
int id;
struct ceph_pg_pool v;
int pg_num_mask, pgp_num_mask, lpg_num_mask, lpgp_num_mask;
+ char *name;
};
struct ceph_pg_mapping {
diff --git a/fs/ceph/pagelist.c b/fs/ceph/pagelist.c
index 370e9369547..5f8dbf7c745 100644
--- a/fs/ceph/pagelist.c
+++ b/fs/ceph/pagelist.c
@@ -1,4 +1,5 @@
+#include <linux/gfp.h>
#include <linux/pagemap.h>
#include <linux/highmem.h>
diff --git a/fs/ceph/rados.h b/fs/ceph/rados.h
index 26ac8b89a67..a1fc1d017b5 100644
--- a/fs/ceph/rados.h
+++ b/fs/ceph/rados.h
@@ -11,8 +11,10 @@
/*
* osdmap encoding versions
*/
-#define CEPH_OSDMAP_INC_VERSION 4
-#define CEPH_OSDMAP_VERSION 4
+#define CEPH_OSDMAP_INC_VERSION 5
+#define CEPH_OSDMAP_INC_VERSION_EXT 5
+#define CEPH_OSDMAP_VERSION 5
+#define CEPH_OSDMAP_VERSION_EXT 5
/*
* fs id
diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c
index bf2a5f3846a..2b881262ef6 100644
--- a/fs/ceph/snap.c
+++ b/fs/ceph/snap.c
@@ -1,6 +1,7 @@
#include "ceph_debug.h"
#include <linux/sort.h>
+#include <linux/slab.h>
#include "super.h"
#include "decode.h"
@@ -314,9 +315,9 @@ static int build_snap_context(struct ceph_snap_realm *realm)
because we rebuild_snap_realms() works _downward_ in
hierarchy after each update.) */
if (realm->cached_context &&
- realm->cached_context->seq <= realm->seq &&
+ realm->cached_context->seq == realm->seq &&
(!parent ||
- realm->cached_context->seq <= parent->cached_context->seq)) {
+ realm->cached_context->seq >= parent->cached_context->seq)) {
dout("build_snap_context %llx %p: %p seq %lld (%d snaps)"
" (unchanged)\n",
realm->ino, realm, realm->cached_context,
@@ -430,8 +431,7 @@ static int dup_array(u64 **dst, __le64 *src, int num)
* Caller must hold snap_rwsem for read (i.e., the realm topology won't
* change).
*/
-void ceph_queue_cap_snap(struct ceph_inode_info *ci,
- struct ceph_snap_context *snapc)
+void ceph_queue_cap_snap(struct ceph_inode_info *ci)
{
struct inode *inode = &ci->vfs_inode;
struct ceph_cap_snap *capsnap;
@@ -450,10 +450,11 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci,
as no new writes are allowed to start when pending, so any
writes in progress now were started before the previous
cap_snap. lucky us. */
- dout("queue_cap_snap %p snapc %p seq %llu used %d"
- " already pending\n", inode, snapc, snapc->seq, used);
+ dout("queue_cap_snap %p already pending\n", inode);
kfree(capsnap);
} else if (ci->i_wrbuffer_ref_head || (used & CEPH_CAP_FILE_WR)) {
+ struct ceph_snap_context *snapc = ci->i_head_snapc;
+
igrab(inode);
atomic_set(&capsnap->nref, 1);
@@ -462,7 +463,6 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci,
INIT_LIST_HEAD(&capsnap->flushing_item);
capsnap->follows = snapc->seq - 1;
- capsnap->context = ceph_get_snap_context(snapc);
capsnap->issued = __ceph_caps_issued(ci, NULL);
capsnap->dirty = __ceph_caps_dirty(ci);
@@ -479,7 +479,7 @@ void ceph_queue_cap_snap(struct ceph_inode_info *ci,
snapshot. */
capsnap->dirty_pages = ci->i_wrbuffer_ref_head;
ci->i_wrbuffer_ref_head = 0;
- ceph_put_snap_context(ci->i_head_snapc);
+ capsnap->context = snapc;
ci->i_head_snapc = NULL;
list_add_tail(&capsnap->ci_item, &ci->i_cap_snaps);
@@ -521,15 +521,17 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
capsnap->ctime = inode->i_ctime;
capsnap->time_warp_seq = ci->i_time_warp_seq;
if (capsnap->dirty_pages) {
- dout("finish_cap_snap %p cap_snap %p snapc %p %llu s=%llu "
+ dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu "
"still has %d dirty pages\n", inode, capsnap,
capsnap->context, capsnap->context->seq,
- capsnap->size, capsnap->dirty_pages);
+ ceph_cap_string(capsnap->dirty), capsnap->size,
+ capsnap->dirty_pages);
return 0;
}
- dout("finish_cap_snap %p cap_snap %p snapc %p %llu s=%llu clean\n",
+ dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu\n",
inode, capsnap, capsnap->context,
- capsnap->context->seq, capsnap->size);
+ capsnap->context->seq, ceph_cap_string(capsnap->dirty),
+ capsnap->size);
spin_lock(&mdsc->snap_flush_lock);
list_add_tail(&ci->i_snap_flush_item, &mdsc->snap_flush_list);
@@ -601,7 +603,7 @@ more:
if (lastinode)
iput(lastinode);
lastinode = inode;
- ceph_queue_cap_snap(ci, realm->cached_context);
+ ceph_queue_cap_snap(ci);
spin_lock(&realm->inodes_with_caps_lock);
}
spin_unlock(&realm->inodes_with_caps_lock);
@@ -818,11 +820,12 @@ void ceph_handle_snap(struct ceph_mds_client *mdsc,
* queued (again) by ceph_update_snap_trace()
* below. Queue it _now_, under the old context.
*/
+ spin_lock(&realm->inodes_with_caps_lock);
list_del_init(&ci->i_snap_realm_item);
+ spin_unlock(&realm->inodes_with_caps_lock);
spin_unlock(&inode->i_lock);
- ceph_queue_cap_snap(ci,
- ci->i_snap_realm->cached_context);
+ ceph_queue_cap_snap(ci);
iput(inode);
continue;
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 4290a6e860b..75d02eaa127 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -11,6 +11,7 @@
#include <linux/rwsem.h>
#include <linux/sched.h>
#include <linux/seq_file.h>
+#include <linux/slab.h>
#include <linux/statfs.h>
#include <linux/string.h>
#include <linux/version.h>
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 65d12036b67..e30dfbb056c 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -12,6 +12,7 @@
#include <linux/pagemap.h>
#include <linux/wait.h>
#include <linux/writeback.h>
+#include <linux/slab.h>
#include "types.h"
#include "messenger.h"
@@ -714,8 +715,7 @@ extern int ceph_update_snap_trace(struct ceph_mds_client *m,
extern void ceph_handle_snap(struct ceph_mds_client *mdsc,
struct ceph_mds_session *session,
struct ceph_msg *msg);
-extern void ceph_queue_cap_snap(struct ceph_inode_info *ci,
- struct ceph_snap_context *snapc);
+extern void ceph_queue_cap_snap(struct ceph_inode_info *ci);
extern int __ceph_finish_cap_snap(struct ceph_inode_info *ci,
struct ceph_cap_snap *capsnap);
extern void ceph_cleanup_empty_realms(struct ceph_mds_client *mdsc);
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index 37d6ce64569..2845422907f 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -3,6 +3,7 @@
#include "decode.h"
#include <linux/xattr.h>
+#include <linux/slab.h>
static bool ceph_is_valid_xattr(const char *name)
{
generated by cgit v1.2.3-70-g09d2 (git 2.46.0) at 2024-11-21 22:11:55 +0000