summaryrefslogtreecommitdiff
path: root/fs/ceph
diff options
context:
space:
mode:
Diffstat (limited to 'fs/ceph')
-rw-r--r--fs/ceph/caps.c4
-rw-r--r--fs/ceph/dir.c80
-rw-r--r--fs/ceph/export.c6
-rw-r--r--fs/ceph/inode.c3
-rw-r--r--fs/ceph/mds_client.c14
-rw-r--r--fs/ceph/mds_client.h7
-rw-r--r--fs/ceph/super.c31
-rw-r--r--fs/ceph/super.h1
-rw-r--r--fs/ceph/xattr.c26
9 files changed, 118 insertions, 54 deletions
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index b60fc8bfb3e9..620daad201db 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -641,10 +641,10 @@ static int __cap_is_valid(struct ceph_cap *cap)
unsigned long ttl;
u32 gen;
- spin_lock(&cap->session->s_cap_lock);
+ spin_lock(&cap->session->s_gen_ttl_lock);
gen = cap->session->s_cap_gen;
ttl = cap->session->s_cap_ttl;
- spin_unlock(&cap->session->s_cap_lock);
+ spin_unlock(&cap->session->s_gen_ttl_lock);
if (cap->cap_gen < gen || time_after_eq(jiffies, ttl)) {
dout("__cap_is_valid %p cap %p issued %s "
diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c
index 74fd74719dc2..3e8094be4604 100644
--- a/fs/ceph/dir.c
+++ b/fs/ceph/dir.c
@@ -973,12 +973,12 @@ static int dentry_lease_is_valid(struct dentry *dentry)
spin_lock(&dentry->d_lock);
di = ceph_dentry(dentry);
- if (di && di->lease_session) {
+ if (di->lease_session) {
s = di->lease_session;
- spin_lock(&s->s_cap_lock);
+ spin_lock(&s->s_gen_ttl_lock);
gen = s->s_cap_gen;
ttl = s->s_cap_ttl;
- spin_unlock(&s->s_cap_lock);
+ spin_unlock(&s->s_gen_ttl_lock);
if (di->lease_gen == gen &&
time_before(jiffies, dentry->d_time) &&
@@ -1072,13 +1072,11 @@ static void ceph_d_release(struct dentry *dentry)
struct ceph_dentry_info *di = ceph_dentry(dentry);
dout("d_release %p\n", dentry);
- if (di) {
- ceph_dentry_lru_del(dentry);
- if (di->lease_session)
- ceph_put_mds_session(di->lease_session);
- kmem_cache_free(ceph_dentry_cachep, di);
- dentry->d_fsdata = NULL;
- }
+ ceph_dentry_lru_del(dentry);
+ if (di->lease_session)
+ ceph_put_mds_session(di->lease_session);
+ kmem_cache_free(ceph_dentry_cachep, di);
+ dentry->d_fsdata = NULL;
}
static int ceph_snapdir_d_revalidate(struct dentry *dentry,
@@ -1096,17 +1094,36 @@ static int ceph_snapdir_d_revalidate(struct dentry *dentry,
*/
void ceph_dir_set_complete(struct inode *inode)
{
- /* not yet implemented */
+ struct dentry *dentry = d_find_any_alias(inode);
+
+ if (dentry && ceph_dentry(dentry) &&
+ ceph_test_mount_opt(ceph_sb_to_client(dentry->d_sb), DCACHE)) {
+ dout(" marking %p (%p) complete\n", inode, dentry);
+ set_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags);
+ }
+ dput(dentry);
}
void ceph_dir_clear_complete(struct inode *inode)
{
- /* not yet implemented */
+ struct dentry *dentry = d_find_any_alias(inode);
+
+ if (dentry && ceph_dentry(dentry)) {
+ dout(" marking %p (%p) complete\n", inode, dentry);
+ set_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags);
+ }
+ dput(dentry);
}
bool ceph_dir_test_complete(struct inode *inode)
{
- /* not yet implemented */
+ struct dentry *dentry = d_find_any_alias(inode);
+
+ if (dentry && ceph_dentry(dentry)) {
+ dout(" marking %p (%p) NOT complete\n", inode, dentry);
+ clear_bit(CEPH_D_COMPLETE, &ceph_dentry(dentry)->flags);
+ }
+ dput(dentry);
return false;
}
@@ -1220,6 +1237,7 @@ static int ceph_dir_fsync(struct file *file, loff_t start, loff_t end,
do {
ceph_mdsc_get_request(req);
spin_unlock(&ci->i_unsafe_lock);
+
dout("dir_fsync %p wait on tid %llu (until %llu)\n",
inode, req->r_tid, last_tid);
if (req->r_timeout) {
@@ -1232,9 +1250,9 @@ static int ceph_dir_fsync(struct file *file, loff_t start, loff_t end,
} else {
wait_for_completion(&req->r_safe_completion);
}
- spin_lock(&ci->i_unsafe_lock);
ceph_mdsc_put_request(req);
+ spin_lock(&ci->i_unsafe_lock);
if (ret || list_empty(head))
break;
req = list_entry(head->next,
@@ -1259,13 +1277,11 @@ void ceph_dentry_lru_add(struct dentry *dn)
dout("dentry_lru_add %p %p '%.*s'\n", di, dn,
dn->d_name.len, dn->d_name.name);
- if (di) {
- mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
- spin_lock(&mdsc->dentry_lru_lock);
- list_add_tail(&di->lru, &mdsc->dentry_lru);
- mdsc->num_dentry++;
- spin_unlock(&mdsc->dentry_lru_lock);
- }
+ mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
+ spin_lock(&mdsc->dentry_lru_lock);
+ list_add_tail(&di->lru, &mdsc->dentry_lru);
+ mdsc->num_dentry++;
+ spin_unlock(&mdsc->dentry_lru_lock);
}
void ceph_dentry_lru_touch(struct dentry *dn)
@@ -1275,12 +1291,10 @@ void ceph_dentry_lru_touch(struct dentry *dn)
dout("dentry_lru_touch %p %p '%.*s' (offset %lld)\n", di, dn,
dn->d_name.len, dn->d_name.name, di->offset);
- if (di) {
- mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
- spin_lock(&mdsc->dentry_lru_lock);
- list_move_tail(&di->lru, &mdsc->dentry_lru);
- spin_unlock(&mdsc->dentry_lru_lock);
- }
+ mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
+ spin_lock(&mdsc->dentry_lru_lock);
+ list_move_tail(&di->lru, &mdsc->dentry_lru);
+ spin_unlock(&mdsc->dentry_lru_lock);
}
void ceph_dentry_lru_del(struct dentry *dn)
@@ -1290,13 +1304,11 @@ void ceph_dentry_lru_del(struct dentry *dn)
dout("dentry_lru_del %p %p '%.*s'\n", di, dn,
dn->d_name.len, dn->d_name.name);
- if (di) {
- mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
- spin_lock(&mdsc->dentry_lru_lock);
- list_del_init(&di->lru);
- mdsc->num_dentry--;
- spin_unlock(&mdsc->dentry_lru_lock);
- }
+ mdsc = ceph_sb_to_client(dn->d_sb)->mdsc;
+ spin_lock(&mdsc->dentry_lru_lock);
+ list_del_init(&di->lru);
+ mdsc->num_dentry--;
+ spin_unlock(&mdsc->dentry_lru_lock);
}
/*
diff --git a/fs/ceph/export.c b/fs/ceph/export.c
index 9fbcdecaaccd..fbb2a643ef10 100644
--- a/fs/ceph/export.c
+++ b/fs/ceph/export.c
@@ -56,9 +56,7 @@ static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len,
return -EINVAL;
spin_lock(&dentry->d_lock);
- parent = dget(dentry->d_parent);
- spin_unlock(&dentry->d_lock);
-
+ parent = dentry->d_parent;
if (*max_len >= connected_handle_length) {
dout("encode_fh %p connectable\n", dentry);
cfh->ino = ceph_ino(dentry->d_inode);
@@ -81,7 +79,7 @@ static int ceph_encode_fh(struct dentry *dentry, u32 *rawfh, int *max_len,
*max_len = handle_length;
type = 255;
}
- dput(parent);
+ spin_unlock(&dentry->d_lock);
return type;
}
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index 25283e7a37f8..2c489378b4cd 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -850,11 +850,12 @@ static void ceph_set_dentry_offset(struct dentry *dn)
{
struct dentry *dir = dn->d_parent;
struct inode *inode = dir->d_inode;
- struct ceph_inode_info *ci = ceph_inode(inode);
+ struct ceph_inode_info *ci;
struct ceph_dentry_info *di;
BUG_ON(!inode);
+ ci = ceph_inode(inode);
di = ceph_dentry(dn);
spin_lock(&ci->i_ceph_lock);
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 6203d805eb45..866e8d7ca37d 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -262,6 +262,7 @@ static int parse_reply_info(struct ceph_msg *msg,
/* trace */
ceph_decode_32_safe(&p, end, len, bad);
if (len > 0) {
+ ceph_decode_need(&p, end, len, bad);
err = parse_reply_info_trace(&p, p+len, info, features);
if (err < 0)
goto out_bad;
@@ -270,6 +271,7 @@ static int parse_reply_info(struct ceph_msg *msg,
/* extra */
ceph_decode_32_safe(&p, end, len, bad);
if (len > 0) {
+ ceph_decode_need(&p, end, len, bad);
err = parse_reply_info_extra(&p, p+len, info, features);
if (err < 0)
goto out_bad;
@@ -398,9 +400,11 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc,
s->s_con.peer_name.type = CEPH_ENTITY_TYPE_MDS;
s->s_con.peer_name.num = cpu_to_le64(mds);
- spin_lock_init(&s->s_cap_lock);
+ spin_lock_init(&s->s_gen_ttl_lock);
s->s_cap_gen = 0;
s->s_cap_ttl = 0;
+
+ spin_lock_init(&s->s_cap_lock);
s->s_renew_requested = 0;
s->s_renew_seq = 0;
INIT_LIST_HEAD(&s->s_caps);
@@ -2326,10 +2330,10 @@ static void handle_session(struct ceph_mds_session *session,
case CEPH_SESSION_STALE:
pr_info("mds%d caps went stale, renewing\n",
session->s_mds);
- spin_lock(&session->s_cap_lock);
+ spin_lock(&session->s_gen_ttl_lock);
session->s_cap_gen++;
session->s_cap_ttl = 0;
- spin_unlock(&session->s_cap_lock);
+ spin_unlock(&session->s_gen_ttl_lock);
send_renew_caps(mdsc, session);
break;
@@ -2772,7 +2776,7 @@ static void handle_lease(struct ceph_mds_client *mdsc,
di = ceph_dentry(dentry);
switch (h->action) {
case CEPH_MDS_LEASE_REVOKE:
- if (di && di->lease_session == session) {
+ if (di->lease_session == session) {
if (ceph_seq_cmp(di->lease_seq, seq) > 0)
h->seq = cpu_to_le32(di->lease_seq);
__ceph_mdsc_drop_dentry_lease(dentry);
@@ -2781,7 +2785,7 @@ static void handle_lease(struct ceph_mds_client *mdsc,
break;
case CEPH_MDS_LEASE_RENEW:
- if (di && di->lease_session == session &&
+ if (di->lease_session == session &&
di->lease_gen == session->s_cap_gen &&
di->lease_renew_from &&
di->lease_renew_after == 0) {
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index a50ca0e39475..8c7c04ebb595 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -117,10 +117,13 @@ struct ceph_mds_session {
void *s_authorizer_buf, *s_authorizer_reply_buf;
size_t s_authorizer_buf_len, s_authorizer_reply_buf_len;
- /* protected by s_cap_lock */
- spinlock_t s_cap_lock;
+ /* protected by s_gen_ttl_lock */
+ spinlock_t s_gen_ttl_lock;
u32 s_cap_gen; /* inc each time we get mds stale msg */
unsigned long s_cap_ttl; /* when session caps expire */
+
+ /* protected by s_cap_lock */
+ spinlock_t s_cap_lock;
struct list_head s_caps; /* all caps issued by this session */
int s_nr_caps, s_trim_caps;
int s_num_cap_releases;
diff --git a/fs/ceph/super.c b/fs/ceph/super.c
index 11bd0fc4853f..00de2c9568cd 100644
--- a/fs/ceph/super.c
+++ b/fs/ceph/super.c
@@ -131,6 +131,8 @@ enum {
Opt_rbytes,
Opt_norbytes,
Opt_noasyncreaddir,
+ Opt_dcache,
+ Opt_nodcache,
Opt_ino32,
};
@@ -152,6 +154,8 @@ static match_table_t fsopt_tokens = {
{Opt_rbytes, "rbytes"},
{Opt_norbytes, "norbytes"},
{Opt_noasyncreaddir, "noasyncreaddir"},
+ {Opt_dcache, "dcache"},
+ {Opt_nodcache, "nodcache"},
{Opt_ino32, "ino32"},
{-1, NULL}
};
@@ -231,6 +235,12 @@ static int parse_fsopt_token(char *c, void *private)
case Opt_noasyncreaddir:
fsopt->flags |= CEPH_MOUNT_OPT_NOASYNCREADDIR;
break;
+ case Opt_dcache:
+ fsopt->flags |= CEPH_MOUNT_OPT_DCACHE;
+ break;
+ case Opt_nodcache:
+ fsopt->flags &= ~CEPH_MOUNT_OPT_DCACHE;
+ break;
case Opt_ino32:
fsopt->flags |= CEPH_MOUNT_OPT_INO32;
break;
@@ -377,6 +387,10 @@ static int ceph_show_options(struct seq_file *m, struct dentry *root)
seq_puts(m, ",norbytes");
if (fsopt->flags & CEPH_MOUNT_OPT_NOASYNCREADDIR)
seq_puts(m, ",noasyncreaddir");
+ if (fsopt->flags & CEPH_MOUNT_OPT_DCACHE)
+ seq_puts(m, ",dcache");
+ else
+ seq_puts(m, ",nodcache");
if (fsopt->wsize)
seq_printf(m, ",wsize=%d", fsopt->wsize);
@@ -636,19 +650,26 @@ static struct dentry *open_root_dentry(struct ceph_fs_client *fsc,
req->r_num_caps = 2;
err = ceph_mdsc_do_request(mdsc, NULL, req);
if (err == 0) {
+ struct inode *inode = req->r_target_inode;
+ req->r_target_inode = NULL;
dout("open_root_inode success\n");
- if (ceph_ino(req->r_target_inode) == CEPH_INO_ROOT &&
+ if (ceph_ino(inode) == CEPH_INO_ROOT &&
fsc->sb->s_root == NULL) {
- root = d_alloc_root(req->r_target_inode);
- ceph_init_dentry(root);
+ root = d_alloc_root(inode);
+ if (!root) {
+ iput(inode);
+ root = ERR_PTR(-ENOMEM);
+ goto out;
+ }
} else {
- root = d_obtain_alias(req->r_target_inode);
+ root = d_obtain_alias(inode);
}
- req->r_target_inode = NULL;
+ ceph_init_dentry(root);
dout("open_root_inode success, root dentry is %p\n", root);
} else {
root = ERR_PTR(err);
}
+out:
ceph_mdsc_put_request(req);
return root;
}
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index cb3652b37271..1421f3d875a2 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -28,6 +28,7 @@
#define CEPH_MOUNT_OPT_RBYTES (1<<5) /* dir st_bytes = rbytes */
#define CEPH_MOUNT_OPT_NOASYNCREADDIR (1<<7) /* no dcache readdir */
#define CEPH_MOUNT_OPT_INO32 (1<<8) /* 32 bit inos */
+#define CEPH_MOUNT_OPT_DCACHE (1<<9) /* use dcache for readdir etc */
#define CEPH_MOUNT_OPT_DEFAULT (CEPH_MOUNT_OPT_RBYTES)
diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c
index a5e36e4488a7..a76f697303d9 100644
--- a/fs/ceph/xattr.c
+++ b/fs/ceph/xattr.c
@@ -111,8 +111,10 @@ static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val,
}
static struct ceph_vxattr_cb ceph_file_vxattrs[] = {
+ { true, "ceph.file.layout", ceph_vxattrcb_layout},
+ /* The following extended attribute name is deprecated */
{ true, "ceph.layout", ceph_vxattrcb_layout},
- { NULL, NULL }
+ { true, NULL, NULL }
};
static struct ceph_vxattr_cb *ceph_inode_vxattrs(struct inode *inode)
@@ -818,6 +820,7 @@ int ceph_removexattr(struct dentry *dentry, const char *name)
struct ceph_vxattr_cb *vxattrs = ceph_inode_vxattrs(inode);
int issued;
int err;
+ int required_blob_size;
int dirty;
if (ceph_snap(inode) != CEPH_NOSNAP)
@@ -833,14 +836,34 @@ int ceph_removexattr(struct dentry *dentry, const char *name)
return -EOPNOTSUPP;
}
+ err = -ENOMEM;
spin_lock(&ci->i_ceph_lock);
__build_xattrs(inode);
+retry:
issued = __ceph_caps_issued(ci, NULL);
dout("removexattr %p issued %s\n", inode, ceph_cap_string(issued));
if (!(issued & CEPH_CAP_XATTR_EXCL))
goto do_sync;
+ required_blob_size = __get_required_blob_size(ci, 0, 0);
+
+ if (!ci->i_xattrs.prealloc_blob ||
+ required_blob_size > ci->i_xattrs.prealloc_blob->alloc_len) {
+ struct ceph_buffer *blob;
+
+ spin_unlock(&ci->i_ceph_lock);
+ dout(" preaallocating new blob size=%d\n", required_blob_size);
+ blob = ceph_buffer_new(required_blob_size, GFP_NOFS);
+ if (!blob)
+ goto out;
+ spin_lock(&ci->i_ceph_lock);
+ if (ci->i_xattrs.prealloc_blob)
+ ceph_buffer_put(ci->i_xattrs.prealloc_blob);
+ ci->i_xattrs.prealloc_blob = blob;
+ goto retry;
+ }
+
err = __remove_xattr_by_name(ceph_inode(inode), name);
dirty = __ceph_mark_dirty_caps(ci, CEPH_CAP_XATTR_EXCL);
ci->i_xattrs.dirty = true;
@@ -853,6 +876,7 @@ int ceph_removexattr(struct dentry *dentry, const char *name)
do_sync:
spin_unlock(&ci->i_ceph_lock);
err = ceph_send_removexattr(dentry, name);
+out:
return err;
}