aboutsummaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/afs/write.c1
-rw-r--r--fs/btrfs/ctree.c129
-rw-r--r--fs/btrfs/extent_io.c2
-rw-r--r--fs/btrfs/ioctl.c20
-rw-r--r--fs/ceph/auth_x.c5
-rw-r--r--fs/ceph/caps.c21
-rw-r--r--fs/ceph/crush/mapper.c41
-rw-r--r--fs/ceph/debugfs.c2
-rw-r--r--fs/ceph/inode.c19
-rw-r--r--fs/ceph/mds_client.c41
-rw-r--r--fs/ceph/mds_client.h1
-rw-r--r--fs/ceph/messenger.c75
-rw-r--r--fs/ceph/mon_client.c3
-rw-r--r--fs/ceph/osd_client.c3
-rw-r--r--fs/ceph/osdmap.c2
-rw-r--r--fs/dcache.c2
-rw-r--r--fs/fs-writeback.c335
-rw-r--r--fs/gfs2/bmap.c1
-rw-r--r--fs/gfs2/dir.c2
-rw-r--r--fs/gfs2/glock.c12
-rw-r--r--fs/gfs2/inode.c12
-rw-r--r--fs/gfs2/quota.c10
-rw-r--r--fs/gfs2/quota.h2
-rw-r--r--fs/inode.c2
-rw-r--r--fs/jbd2/journal.c15
-rw-r--r--fs/jbd2/transaction.c9
-rw-r--r--fs/jffs2/xattr.c2
-rw-r--r--fs/mbcache.c5
-rw-r--r--fs/nfs/dir.c2
-rw-r--r--fs/nfs/internal.h3
-rw-r--r--fs/ocfs2/aops.c94
-rw-r--r--fs/ocfs2/dlm/dlmdomain.c3
-rw-r--r--fs/ocfs2/dlm/dlmmaster.c22
-rw-r--r--fs/ocfs2/dlm/dlmrecovery.c2
-rw-r--r--fs/ocfs2/file.c309
-rw-r--r--fs/ocfs2/file.h6
-rw-r--r--fs/ocfs2/journal.c30
-rw-r--r--fs/ocfs2/localalloc.c7
-rw-r--r--fs/ocfs2/quota_global.c2
-rw-r--r--fs/ocfs2/quota_local.c4
-rw-r--r--fs/ocfs2/refcounttree.c12
-rw-r--r--fs/ocfs2/suballoc.c2
-rw-r--r--fs/ocfs2/xattr.c200
-rw-r--r--fs/partitions/ibm.c14
-rw-r--r--fs/quota/dquot.c2
-rw-r--r--fs/splice.c9
-rw-r--r--fs/ubifs/shrinker.c2
-rw-r--r--fs/ubifs/ubifs.h2
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.c5
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c2
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c130
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.h2
-rw-r--r--fs/xfs/linux-2.6/xfs_trace.h3
-rw-r--r--fs/xfs/quota/xfs_qm.c7
-rw-r--r--fs/xfs/xfs_mount.h2
55 files changed, 1019 insertions, 633 deletions
diff --git a/fs/afs/write.c b/fs/afs/write.c
index 3dab9e9948d..722743b152d 100644
--- a/fs/afs/write.c
+++ b/fs/afs/write.c
@@ -680,7 +680,6 @@ int afs_writeback_all(struct afs_vnode *vnode)
{
struct address_space *mapping = vnode->vfs_inode.i_mapping;
struct writeback_control wbc = {
- .bdi = mapping->backing_dev_info,
.sync_mode = WB_SYNC_ALL,
.nr_to_write = LONG_MAX,
.range_cyclic = 1,
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 0d1d966b0fe..c3df14ce2cc 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -2304,12 +2304,17 @@ noinline int btrfs_leaf_free_space(struct btrfs_root *root,
return ret;
}
+/*
+ * min slot controls the lowest index we're willing to push to the
+ * right. We'll push up to and including min_slot, but no lower
+ */
static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
int data_size, int empty,
struct extent_buffer *right,
- int free_space, u32 left_nritems)
+ int free_space, u32 left_nritems,
+ u32 min_slot)
{
struct extent_buffer *left = path->nodes[0];
struct extent_buffer *upper = path->nodes[1];
@@ -2327,7 +2332,7 @@ static noinline int __push_leaf_right(struct btrfs_trans_handle *trans,
if (empty)
nr = 0;
else
- nr = 1;
+ nr = max_t(u32, 1, min_slot);
if (path->slots[0] >= left_nritems)
push_space += data_size;
@@ -2469,10 +2474,14 @@ out_unlock:
*
* returns 1 if the push failed because the other node didn't have enough
* room, 0 if everything worked out and < 0 if there were major errors.
+ *
+ * this will push starting from min_slot to the end of the leaf. It won't
+ * push any slot lower than min_slot
*/
static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_path *path, int data_size,
- int empty)
+ *root, struct btrfs_path *path,
+ int min_data_size, int data_size,
+ int empty, u32 min_slot)
{
struct extent_buffer *left = path->nodes[0];
struct extent_buffer *right;
@@ -2514,8 +2523,8 @@ static int push_leaf_right(struct btrfs_trans_handle *trans, struct btrfs_root
if (left_nritems == 0)
goto out_unlock;
- return __push_leaf_right(trans, root, path, data_size, empty,
- right, free_space, left_nritems);
+ return __push_leaf_right(trans, root, path, min_data_size, empty,
+ right, free_space, left_nritems, min_slot);
out_unlock:
btrfs_tree_unlock(right);
free_extent_buffer(right);
@@ -2525,12 +2534,17 @@ out_unlock:
/*
* push some data in the path leaf to the left, trying to free up at
* least data_size bytes. returns zero if the push worked, nonzero otherwise
+ *
+ * max_slot can put a limit on how far into the leaf we'll push items. The
+ * item at 'max_slot' won't be touched. Use (u32)-1 to make us do all the
+ * items
*/
static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path, int data_size,
int empty, struct extent_buffer *left,
- int free_space, int right_nritems)
+ int free_space, u32 right_nritems,
+ u32 max_slot)
{
struct btrfs_disk_key disk_key;
struct extent_buffer *right = path->nodes[0];
@@ -2549,9 +2563,9 @@ static noinline int __push_leaf_left(struct btrfs_trans_handle *trans,
slot = path->slots[1];
if (empty)
- nr = right_nritems;
+ nr = min(right_nritems, max_slot);
else
- nr = right_nritems - 1;
+ nr = min(right_nritems - 1, max_slot);
for (i = 0; i < nr; i++) {
item = btrfs_item_nr(right, i);
@@ -2712,10 +2726,14 @@ out:
/*
* push some data in the path leaf to the left, trying to free up at
* least data_size bytes. returns zero if the push worked, nonzero otherwise
+ *
+ * max_slot can put a limit on how far into the leaf we'll push items. The
+ * item at 'max_slot' won't be touched. Use (u32)-1 to make us push all the
+ * items
*/
static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
- *root, struct btrfs_path *path, int data_size,
- int empty)
+ *root, struct btrfs_path *path, int min_data_size,
+ int data_size, int empty, u32 max_slot)
{
struct extent_buffer *right = path->nodes[0];
struct extent_buffer *left;
@@ -2761,8 +2779,9 @@ static int push_leaf_left(struct btrfs_trans_handle *trans, struct btrfs_root
goto out;
}
- return __push_leaf_left(trans, root, path, data_size,
- empty, left, free_space, right_nritems);
+ return __push_leaf_left(trans, root, path, min_data_size,
+ empty, left, free_space, right_nritems,
+ max_slot);
out:
btrfs_tree_unlock(left);
free_extent_buffer(left);
@@ -2855,6 +2874,64 @@ static noinline int copy_for_split(struct btrfs_trans_handle *trans,
}
/*
+ * double splits happen when we need to insert a big item in the middle
+ * of a leaf. A double split can leave us with 3 mostly empty leaves:
+ * leaf: [ slots 0 - N] [ our target ] [ N + 1 - total in leaf ]
+ * A B C
+ *
+ * We avoid this by trying to push the items on either side of our target
+ * into the adjacent leaves. If all goes well we can avoid the double split
+ * completely.
+ */
+static noinline int push_for_double_split(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root,
+ struct btrfs_path *path,
+ int data_size)
+{
+ int ret;
+ int progress = 0;
+ int slot;
+ u32 nritems;
+
+ slot = path->slots[0];
+
+ /*
+ * try to push all the items after our slot into the
+ * right leaf
+ */
+ ret = push_leaf_right(trans, root, path, 1, data_size, 0, slot);
+ if (ret < 0)
+ return ret;
+
+ if (ret == 0)
+ progress++;
+
+ nritems = btrfs_header_nritems(path->nodes[0]);
+ /*
+ * our goal is to get our slot at the start or end of a leaf. If
+ * we've done so we're done
+ */
+ if (path->slots[0] == 0 || path->slots[0] == nritems)
+ return 0;
+
+ if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size)
+ return 0;
+
+ /* try to push all the items before our slot into the next leaf */
+ slot = path->slots[0];
+ ret = push_leaf_left(trans, root, path, 1, data_size, 0, slot);
+ if (ret < 0)
+ return ret;
+
+ if (ret == 0)
+ progress++;
+
+ if (progress)
+ return 0;
+ return 1;
+}
+
+/*
* split the path's leaf in two, making sure there is at least data_size
* available for the resulting leaf level of the path.
*
@@ -2876,6 +2953,7 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
int wret;
int split;
int num_doubles = 0;
+ int tried_avoid_double = 0;
l = path->nodes[0];
slot = path->slots[0];
@@ -2884,12 +2962,14 @@ static noinline int split_leaf(struct btrfs_trans_handle *trans,
return -EOVERFLOW;
/* first try to make some room by pushing left and right */
- if (data_size && ins_key->type != BTRFS_DIR_ITEM_KEY) {
- wret = push_leaf_right(trans, root, path, data_size, 0);
+ if (data_size) {
+ wret = push_leaf_right(trans, root, path, data_size,
+ data_size, 0, 0);
if (wret < 0)
return wret;
if (wret) {
- wret = push_leaf_left(trans, root, path, data_size, 0);
+ wret = push_leaf_left(trans, root, path, data_size,
+ data_size, 0, (u32)-1);
if (wret < 0)
return wret;
}
@@ -2923,6 +3003,8 @@ again:
if (mid != nritems &&
leaf_space_used(l, mid, nritems - mid) +
data_size > BTRFS_LEAF_DATA_SIZE(root)) {
+ if (data_size && !tried_avoid_double)
+ goto push_for_double;
split = 2;
}
}
@@ -2939,6 +3021,8 @@ again:
if (mid != nritems &&
leaf_space_used(l, mid, nritems - mid) +
data_size > BTRFS_LEAF_DATA_SIZE(root)) {
+ if (data_size && !tried_avoid_double)
+ goto push_for_double;
split = 2 ;
}
}
@@ -3019,6 +3103,13 @@ again:
}
return ret;
+
+push_for_double:
+ push_for_double_split(trans, root, path, data_size);
+ tried_avoid_double = 1;
+ if (btrfs_leaf_free_space(root, path->nodes[0]) >= data_size)
+ return 0;
+ goto again;
}
static noinline int setup_leaf_for_split(struct btrfs_trans_handle *trans,
@@ -3915,13 +4006,15 @@ int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
extent_buffer_get(leaf);
btrfs_set_path_blocking(path);
- wret = push_leaf_left(trans, root, path, 1, 1);
+ wret = push_leaf_left(trans, root, path, 1, 1,
+ 1, (u32)-1);
if (wret < 0 && wret != -ENOSPC)
ret = wret;
if (path->nodes[0] == leaf &&
btrfs_header_nritems(leaf)) {
- wret = push_leaf_right(trans, root, path, 1, 1);
+ wret = push_leaf_right(trans, root, path, 1,
+ 1, 1, 0);
if (wret < 0 && wret != -ENOSPC)
ret = wret;
}
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index a4080c21ec5..d74e6af9b53 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2594,7 +2594,6 @@ int extent_write_full_page(struct extent_io_tree *tree, struct page *page,
.sync_io = wbc->sync_mode == WB_SYNC_ALL,
};
struct writeback_control wbc_writepages = {
- .bdi = wbc->bdi,
.sync_mode = wbc->sync_mode,
.older_than_this = NULL,
.nr_to_write = 64,
@@ -2628,7 +2627,6 @@ int extent_write_locked_range(struct extent_io_tree *tree, struct inode *inode,
.sync_io = mode == WB_SYNC_ALL,
};
struct writeback_control wbc_writepages = {
- .bdi = inode->i_mapping->backing_dev_info,
.sync_mode = mode,
.older_than_this = NULL,
.nr_to_write = nr_pages * 2,
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 4dbaf89b133..9254b3d58db 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -1458,7 +1458,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
*/
/* the destination must be opened for writing */
- if (!(file->f_mode & FMODE_WRITE))
+ if (!(file->f_mode & FMODE_WRITE) || (file->f_flags & O_APPEND))
return -EINVAL;
ret = mnt_want_write(file->f_path.mnt);
@@ -1511,7 +1511,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
/* determine range to clone */
ret = -EINVAL;
- if (off >= src->i_size || off + len > src->i_size)
+ if (off + len > src->i_size || off + len < off)
goto out_unlock;
if (len == 0)
olen = len = src->i_size - off;
@@ -1578,6 +1578,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
u64 disko = 0, diskl = 0;
u64 datao = 0, datal = 0;
u8 comp;
+ u64 endoff;
size = btrfs_item_size_nr(leaf, slot);
read_extent_buffer(leaf, buf,
@@ -1712,9 +1713,18 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd,
btrfs_release_path(root, path);
inode->i_mtime = inode->i_ctime = CURRENT_TIME;
- if (new_key.offset + datal > inode->i_size)
- btrfs_i_size_write(inode,
- new_key.offset + datal);
+
+ /*
+ * we round up to the block size at eof when
+ * determining which extents to clone above,
+ * but shouldn't round up the file size
+ */
+ endoff = new_key.offset + datal;
+ if (endoff > off+olen)
+ endoff = off+olen;
+ if (endoff > inode->i_size)
+ btrfs_i_size_write(inode, endoff);
+
BTRFS_I(inode)->flags = BTRFS_I(src)->flags;
ret = btrfs_update_inode(trans, root, inode);
BUG_ON(ret);
diff --git a/fs/ceph/auth_x.c b/fs/ceph/auth_x.c
index 83d4d2785ff..6d44053ecff 100644
--- a/fs/ceph/auth_x.c
+++ b/fs/ceph/auth_x.c
@@ -493,7 +493,7 @@ static int ceph_x_handle_reply(struct ceph_auth_client *ac, int result,
return -EAGAIN;
}
- op = le32_to_cpu(head->op);
+ op = le16_to_cpu(head->op);
result = le32_to_cpu(head->result);
dout("handle_reply op %d result %d\n", op, result);
switch (op) {
@@ -613,6 +613,9 @@ static void ceph_x_destroy(struct ceph_auth_client *ac)
remove_ticket_handler(ac, th);
}
+ if (xi->auth_authorizer.buf)
+ ceph_buffer_put(xi->auth_authorizer.buf);
+
kfree(ac->private);
ac->private = NULL;
}
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index 619b61655ee..74144d6389f 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -244,8 +244,14 @@ static struct ceph_cap *get_cap(struct ceph_cap_reservation *ctx)
struct ceph_cap *cap = NULL;
/* temporary, until we do something about cap import/export */
- if (!ctx)
- return kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS);
+ if (!ctx) {
+ cap = kmem_cache_alloc(ceph_cap_cachep, GFP_NOFS);
+ if (cap) {
+ caps_use_count++;
+ caps_total_count++;
+ }
+ return cap;
+ }
spin_lock(&caps_list_lock);
dout("get_cap ctx=%p (%d) %d = %d used + %d resv + %d avail\n",
@@ -2886,18 +2892,19 @@ int ceph_encode_inode_release(void **p, struct inode *inode,
struct ceph_inode_info *ci = ceph_inode(inode);
struct ceph_cap *cap;
struct ceph_mds_request_release *rel = *p;
+ int used, dirty;
int ret = 0;
- int used = 0;
spin_lock(&inode->i_lock);
used = __ceph_caps_used(ci);
+ dirty = __ceph_caps_dirty(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),
+ dout("encode_inode_release %p mds%d used|dirty %s drop %s unless %s\n",
+ inode, mds, ceph_cap_string(used|dirty), ceph_cap_string(drop),
ceph_cap_string(unless));
- /* only drop unused caps */
- drop &= ~used;
+ /* only drop unused, clean caps */
+ drop &= ~(used | dirty);
cap = __get_cap_for_mds(ci, mds);
if (cap && __cap_is_valid(cap)) {
diff --git a/fs/ceph/crush/mapper.c b/fs/ceph/crush/mapper.c
index 9ba54efb654..a4eec133258 100644
--- a/fs/ceph/crush/mapper.c
+++ b/fs/ceph/crush/mapper.c
@@ -238,7 +238,7 @@ static int bucket_straw_choose(struct crush_bucket_straw *bucket,
static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
{
- dprintk("choose %d x=%d r=%d\n", in->id, x, r);
+ dprintk(" crush_bucket_choose %d x=%d r=%d\n", in->id, x, r);
switch (in->alg) {
case CRUSH_BUCKET_UNIFORM:
return bucket_uniform_choose((struct crush_bucket_uniform *)in,
@@ -264,7 +264,7 @@ static int crush_bucket_choose(struct crush_bucket *in, int x, int r)
*/
static int is_out(struct crush_map *map, __u32 *weight, int item, int x)
{
- if (weight[item] >= 0x1000)
+ if (weight[item] >= 0x10000)
return 0;
if (weight[item] == 0)
return 1;
@@ -305,7 +305,9 @@ static int crush_choose(struct crush_map *map,
int itemtype;
int collide, reject;
const int orig_tries = 5; /* attempts before we fall back to search */
- dprintk("choose bucket %d x %d outpos %d\n", bucket->id, x, outpos);
+
+ dprintk("CHOOSE%s bucket %d x %d outpos %d numrep %d\n", recurse_to_leaf ? "_LEAF" : "",
+ bucket->id, x, outpos, numrep);
for (rep = outpos; rep < numrep; rep++) {
/* keep trying until we get a non-out, non-colliding item */
@@ -366,6 +368,7 @@ static int crush_choose(struct crush_map *map,
BUG_ON(item >= 0 ||
(-1-item) >= map->max_buckets);
in = map->buckets[-1-item];
+ retry_bucket = 1;
continue;
}
@@ -377,15 +380,25 @@ static int crush_choose(struct crush_map *map,
}
}
- if (recurse_to_leaf &&
- item < 0 &&
- crush_choose(map, map->buckets[-1-item],
- weight,
- x, outpos+1, 0,
- out2, outpos,
- firstn, 0, NULL) <= outpos) {
- reject = 1;
- } else {
+ reject = 0;
+ if (recurse_to_leaf) {
+ if (item < 0) {
+ if (crush_choose(map,
+ map->buckets[-1-item],
+ weight,
+ x, outpos+1, 0,
+ out2, outpos,
+ firstn, 0,
+ NULL) <= outpos)
+ /* didn't get leaf */
+ reject = 1;
+ } else {
+ /* we already have a leaf! */
+ out2[outpos] = item;
+ }
+ }
+
+ if (!reject) {
/* out? */
if (itemtype == 0)
reject = is_out(map, weight,
@@ -424,12 +437,12 @@ reject:
continue;
}
- dprintk("choose got %d\n", item);
+ dprintk("CHOOSE got %d\n", item);
out[outpos] = item;
outpos++;
}
- dprintk("choose returns %d\n", outpos);
+ dprintk("CHOOSE returns %d\n", outpos);
return outpos;
}
diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c
index 3be33fb066c..f2f5332ddbb 100644
--- a/fs/ceph/debugfs.c
+++ b/fs/ceph/debugfs.c
@@ -261,7 +261,7 @@ static int osdc_show(struct seq_file *s, void *pp)
static int caps_show(struct seq_file *s, void *p)
{
- struct ceph_client *client = p;
+ struct ceph_client *client = s->private;
int total, avail, used, reserved, min;
ceph_reservation_status(client, &total, &avail, &used, &reserved, &min);
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index ab47f46ca28..8f9b9fe8ef9 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -854,8 +854,8 @@ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in,
d_drop(dn);
realdn = d_materialise_unique(dn, in);
if (IS_ERR(realdn)) {
- pr_err("splice_dentry error %p inode %p ino %llx.%llx\n",
- dn, in, ceph_vinop(in));
+ pr_err("splice_dentry error %ld %p inode %p ino %llx.%llx\n",
+ PTR_ERR(realdn), dn, in, ceph_vinop(in));
if (prehash)
*prehash = false; /* don't rehash on error */
dn = realdn; /* note realdn contains the error */
@@ -1234,18 +1234,23 @@ retry_lookup:
goto out;
}
dn = splice_dentry(dn, in, NULL);
+ if (IS_ERR(dn))
+ dn = NULL;
}
if (fill_inode(in, &rinfo->dir_in[i], NULL, session,
req->r_request_started, -1,
&req->r_caps_reservation) < 0) {
pr_err("fill_inode badness on %p\n", in);
- dput(dn);
- continue;
+ goto next_item;
}
- update_dentry_lease(dn, rinfo->dir_dlease[i],
- req->r_session, req->r_request_started);
- dput(dn);
+ if (dn)
+ update_dentry_lease(dn, rinfo->dir_dlease[i],
+ req->r_session,
+ req->r_request_started);
+next_item:
+ if (dn)
+ dput(dn);
}
req->r_did_prepopulate = true;
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 1766947fc07..416c08d315d 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -1514,6 +1514,9 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc,
ceph_encode_filepath(&p, end, ino1, path1);
ceph_encode_filepath(&p, end, ino2, path2);
+ /* make note of release offset, in case we need to replay */
+ req->r_request_release_offset = p - msg->front.iov_base;
+
/* cap releases */
releases = 0;
if (req->r_inode_drop)
@@ -1580,6 +1583,32 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
dout("prepare_send_request %p tid %lld %s (attempt %d)\n", req,
req->r_tid, ceph_mds_op_name(req->r_op), req->r_attempts);
+ if (req->r_got_unsafe) {
+ /*
+ * Replay. Do not regenerate message (and rebuild
+ * paths, etc.); just use the original message.
+ * Rebuilding paths will break for renames because
+ * d_move mangles the src name.
+ */
+ msg = req->r_request;
+ rhead = msg->front.iov_base;
+
+ flags = le32_to_cpu(rhead->flags);
+ flags |= CEPH_MDS_FLAG_REPLAY;
+ rhead->flags = cpu_to_le32(flags);
+
+ if (req->r_target_inode)
+ rhead->ino = cpu_to_le64(ceph_ino(req->r_target_inode));
+
+ rhead->num_retry = req->r_attempts - 1;
+
+ /* remove cap/dentry releases from message */
+ rhead->num_releases = 0;
+ msg->hdr.front_len = cpu_to_le32(req->r_request_release_offset);
+ msg->front.iov_len = req->r_request_release_offset;
+ return 0;
+ }
+
if (req->r_request) {
ceph_msg_put(req->r_request);
req->r_request = NULL;
@@ -1601,13 +1630,9 @@ static int __prepare_send_request(struct ceph_mds_client *mdsc,
rhead->flags = cpu_to_le32(flags);
rhead->num_fwd = req->r_num_fwd;
rhead->num_retry = req->r_attempts - 1;
+ rhead->ino = 0;
dout(" r_locked_dir = %p\n", req->r_locked_dir);
-
- if (req->r_target_inode && req->r_got_unsafe)
- rhead->ino = cpu_to_le64(ceph_ino(req->r_target_inode));
- else
- rhead->ino = 0;
return 0;
}
@@ -2783,6 +2808,12 @@ void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc)
drop_leases(mdsc);
ceph_flush_dirty_caps(mdsc);
wait_requests(mdsc);
+
+ /*
+ * wait for reply handlers to drop their request refs and
+ * their inode/dcache refs
+ */
+ ceph_msgr_flush();
}
/*
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index b292fa42a66..952410c60d0 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -188,6 +188,7 @@ struct ceph_mds_request {
int r_old_inode_drop, r_old_inode_unless;
struct ceph_msg *r_request; /* original request */
+ int r_request_release_offset;
struct ceph_msg *r_reply;
struct ceph_mds_reply_info_parsed r_reply_info;
int r_err;
diff --git a/fs/ceph/messenger.c b/fs/ceph/messenger.c
index 64b8b1f7863..15167b2daa5 100644
--- a/fs/ceph/messenger.c
+++ b/fs/ceph/messenger.c
@@ -43,7 +43,8 @@ static void ceph_fault(struct ceph_connection *con);
* nicely render a sockaddr as a string.
*/
#define MAX_ADDR_STR 20
-static char addr_str[MAX_ADDR_STR][40];
+#define MAX_ADDR_STR_LEN 60
+static char addr_str[MAX_ADDR_STR][MAX_ADDR_STR_LEN];
static DEFINE_SPINLOCK(addr_str_lock);
static int last_addr_str;
@@ -52,7 +53,6 @@ const char *pr_addr(const struct sockaddr_storage *ss)
int i;
char *s;
struct sockaddr_in *in4 = (void *)ss;
- unsigned char *quad = (void *)&in4->sin_addr.s_addr;
struct sockaddr_in6 *in6 = (void *)ss;
spin_lock(&addr_str_lock);
@@ -64,25 +64,13 @@ const char *pr_addr(const struct sockaddr_storage *ss)
switch (ss->ss_family) {
case AF