diff options
Diffstat (limited to 'fs/ceph')
-rw-r--r-- | fs/ceph/caps.c | 41 | ||||
-rw-r--r-- | fs/ceph/debugfs.c | 13 | ||||
-rw-r--r-- | fs/ceph/file.c | 10 | ||||
-rw-r--r-- | fs/ceph/inode.c | 4 | ||||
-rw-r--r-- | fs/ceph/mds_client.c | 11 | ||||
-rw-r--r-- | fs/ceph/mds_client.h | 14 | ||||
-rw-r--r-- | fs/ceph/snap.c | 48 | ||||
-rw-r--r-- | fs/ceph/super.c | 14 | ||||
-rw-r--r-- | fs/ceph/xattr.c | 10 |
9 files changed, 136 insertions, 29 deletions
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 06e9b26bf277..45b8f6741f8d 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -1558,6 +1558,7 @@ void ceph_flush_snaps(struct ceph_inode_info *ci, struct inode *inode = &ci->vfs_inode; struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; struct ceph_mds_session *session = NULL; + bool need_put = false; int mds; dout("ceph_flush_snaps %p\n", inode); @@ -1609,8 +1610,13 @@ out: } /* we flushed them all; remove this inode from the queue */ spin_lock(&mdsc->snap_flush_lock); + if (!list_empty(&ci->i_snap_flush_item)) + need_put = true; list_del_init(&ci->i_snap_flush_item); spin_unlock(&mdsc->snap_flush_lock); + + if (need_put) + iput(inode); } /* @@ -2784,7 +2790,19 @@ int ceph_get_caps(struct file *filp, int need, int want, if (ret == -EAGAIN) continue; if (!ret) { + struct ceph_mds_client *mdsc = fsc->mdsc; + struct cap_wait cw; DEFINE_WAIT_FUNC(wait, woken_wake_function); + + cw.ino = inode->i_ino; + cw.tgid = current->tgid; + cw.need = need; + cw.want = want; + + spin_lock(&mdsc->caps_list_lock); + list_add(&cw.list, &mdsc->cap_wait_list); + spin_unlock(&mdsc->caps_list_lock); + add_wait_queue(&ci->i_cap_wq, &wait); flags |= NON_BLOCKING; @@ -2798,6 +2816,11 @@ int ceph_get_caps(struct file *filp, int need, int want, } remove_wait_queue(&ci->i_cap_wq, &wait); + + spin_lock(&mdsc->caps_list_lock); + list_del(&cw.list); + spin_unlock(&mdsc->caps_list_lock); + if (ret == -EAGAIN) continue; } @@ -3334,6 +3357,15 @@ static void handle_cap_grant(struct inode *inode, } BUG_ON(cap->issued & ~cap->implemented); + /* don't let check_caps skip sending a response to MDS for revoke msgs */ + if (le32_to_cpu(grant->op) == CEPH_CAP_OP_REVOKE) { + cap->mds_wanted = 0; + if (cap == ci->i_auth_cap) + check_caps = 1; /* check auth cap only */ + else + check_caps = 2; /* check all caps */ + } + if (extra_info->inline_version > 0 && extra_info->inline_version >= ci->i_inline_version) { ci->i_inline_version = extra_info->inline_version; @@ -4272,12 +4304,14 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry, struct inode *dir, int mds, int drop, int unless) { - struct dentry *parent = NULL; struct ceph_mds_request_release *rel = *p; struct ceph_dentry_info *di = ceph_dentry(dentry); int force = 0; int ret; + /* This shouldn't happen */ + BUG_ON(!dir); + /* * force an record for the directory caps if we have a dentry lease. * this is racy (can't take i_ceph_lock and d_lock together), but it @@ -4287,14 +4321,9 @@ int ceph_encode_dentry_release(void **p, struct dentry *dentry, spin_lock(&dentry->d_lock); if (di->lease_session && di->lease_session->s_mds == mds) force = 1; - if (!dir) { - parent = dget(dentry->d_parent); - dir = d_inode(parent); - } spin_unlock(&dentry->d_lock); ret = ceph_encode_inode_release(p, dir, mds, drop, unless, force); - dput(parent); spin_lock(&dentry->d_lock); if (ret && di->lease_session && di->lease_session->s_mds == mds) { diff --git a/fs/ceph/debugfs.c b/fs/ceph/debugfs.c index facb387c2735..c281f32b54f7 100644 --- a/fs/ceph/debugfs.c +++ b/fs/ceph/debugfs.c @@ -139,6 +139,7 @@ static int caps_show(struct seq_file *s, void *p) struct ceph_fs_client *fsc = s->private; struct ceph_mds_client *mdsc = fsc->mdsc; int total, avail, used, reserved, min, i; + struct cap_wait *cw; ceph_reservation_status(fsc, &total, &avail, &used, &reserved, &min); seq_printf(s, "total\t\t%d\n" @@ -166,6 +167,18 @@ static int caps_show(struct seq_file *s, void *p) } mutex_unlock(&mdsc->mutex); + seq_printf(s, "\n\nWaiters:\n--------\n"); + seq_printf(s, "tgid ino need want\n"); + seq_printf(s, "-----------------------------------------------------\n"); + + spin_lock(&mdsc->caps_list_lock); + list_for_each_entry(cw, &mdsc->cap_wait_list, list) { + seq_printf(s, "%-13d0x%-17lx%-17s%-17s\n", cw->tgid, cw->ino, + ceph_cap_string(cw->need), + ceph_cap_string(cw->want)); + } + spin_unlock(&mdsc->caps_list_lock); + return 0; } diff --git a/fs/ceph/file.c b/fs/ceph/file.c index aa1eac6d89f2..83122fc5f813 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -452,6 +452,12 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, if (dentry->d_name.len > NAME_MAX) return -ENAMETOOLONG; + /* + * Do not truncate the file, since atomic_open is called before the + * permission check. The caller will do the truncation afterward. + */ + flags &= ~O_TRUNC; + if (flags & O_CREAT) { if (ceph_quota_is_max_files_exceeded(dir)) return -EDQUOT; @@ -490,9 +496,7 @@ int ceph_atomic_open(struct inode *dir, struct dentry *dentry, req->r_parent = dir; set_bit(CEPH_MDS_R_PARENT_LOCKED, &req->r_req_flags); - err = ceph_mdsc_do_request(mdsc, - (flags & (O_CREAT|O_TRUNC)) ? dir : NULL, - req); + err = ceph_mdsc_do_request(mdsc, (flags & O_CREAT) ? dir : NULL, req); err = ceph_handle_snapdir(req, dentry, err); if (err) goto out_req; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index af85a7237604..a08ddd4e26d9 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -619,9 +619,7 @@ int ceph_fill_file_size(struct inode *inode, int issued, ci->i_truncate_seq = truncate_seq; /* the MDS should have revoked these caps */ - WARN_ON_ONCE(issued & (CEPH_CAP_FILE_EXCL | - CEPH_CAP_FILE_RD | - CEPH_CAP_FILE_WR | + WARN_ON_ONCE(issued & (CEPH_CAP_FILE_RD | CEPH_CAP_FILE_LAZYIO)); /* * If we hold relevant caps, or in the case where we're diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 37fb71797b34..f7acf9680c9b 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -3151,6 +3151,12 @@ static void handle_session(struct ceph_mds_session *session, break; case CEPH_SESSION_FLUSHMSG: + /* flush cap releases */ + spin_lock(&session->s_cap_lock); + if (session->s_num_cap_releases) + ceph_flush_cap_releases(mdsc, session); + spin_unlock(&session->s_cap_lock); + send_flushmsg_ack(mdsc, session, seq); break; @@ -4068,7 +4074,7 @@ static void delayed_work(struct work_struct *work) dout("mdsc delayed_work\n"); - if (mdsc->stopping) + if (mdsc->stopping >= CEPH_MDSC_STOPPING_FLUSHED) return; mutex_lock(&mdsc->mutex); @@ -4168,6 +4174,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) INIT_DELAYED_WORK(&mdsc->delayed_work, delayed_work); mdsc->last_renew_caps = jiffies; INIT_LIST_HEAD(&mdsc->cap_delay_list); + INIT_LIST_HEAD(&mdsc->cap_wait_list); spin_lock_init(&mdsc->cap_delay_lock); INIT_LIST_HEAD(&mdsc->snap_flush_list); spin_lock_init(&mdsc->snap_flush_lock); @@ -4239,7 +4246,7 @@ static void wait_requests(struct ceph_mds_client *mdsc) void ceph_mdsc_pre_umount(struct ceph_mds_client *mdsc) { dout("pre_umount\n"); - mdsc->stopping = 1; + mdsc->stopping = CEPH_MDSC_STOPPING_BEGIN; lock_unlock_sessions(mdsc); ceph_flush_dirty_caps(mdsc); diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 5cd131b41d84..4fbbc33023c9 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -340,6 +340,19 @@ struct ceph_quotarealm_inode { struct inode *inode; }; +struct cap_wait { + struct list_head list; + unsigned long ino; + pid_t tgid; + int need; + int want; +}; + +enum { + CEPH_MDSC_STOPPING_BEGIN = 1, + CEPH_MDSC_STOPPING_FLUSHED = 2, +}; + /* * mds client state */ @@ -416,6 +429,7 @@ struct ceph_mds_client { spinlock_t caps_list_lock; struct list_head caps_list; /* unused (reserved or unreserved) */ + struct list_head cap_wait_list; int caps_total_count; /* total caps allocated */ int caps_use_count; /* in use */ int caps_use_max; /* max used caps */ diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index e1b9b224fcb2..4df068bdba68 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -644,8 +644,10 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci, capsnap->size); spin_lock(&mdsc->snap_flush_lock); - if (list_empty(&ci->i_snap_flush_item)) + if (list_empty(&ci->i_snap_flush_item)) { + ihold(inode); list_add_tail(&ci->i_snap_flush_item, &mdsc->snap_flush_list); + } spin_unlock(&mdsc->snap_flush_lock); return 1; /* caller may want to ceph_flush_snaps */ } @@ -694,9 +696,10 @@ int ceph_update_snap_trace(struct ceph_mds_client *mdsc, struct ceph_mds_snap_realm *ri; /* encoded */ __le64 *snaps; /* encoded */ __le64 *prior_parent_snaps; /* encoded */ - struct ceph_snap_realm *realm = NULL; + struct ceph_snap_realm *realm; struct ceph_snap_realm *first_realm = NULL; - int invalidate = 0; + struct ceph_snap_realm *realm_to_rebuild = NULL; + int rebuild_snapcs; int err = -ENOMEM; LIST_HEAD(dirty_realms); @@ -704,6 +707,8 @@ int ceph_update_snap_trace(struct ceph_mds_client *mdsc, dout("update_snap_trace deletion=%d\n", deletion); more: + realm = NULL; + rebuild_snapcs = 0; ceph_decode_need(&p, e, sizeof(*ri), bad); ri = p; p += sizeof(*ri); @@ -727,7 +732,7 @@ more: err = adjust_snap_realm_parent(mdsc, realm, le64_to_cpu(ri->parent)); if (err < 0) goto fail; - invalidate += err; + rebuild_snapcs += err; if (le64_to_cpu(ri->seq) > realm->seq) { dout("update_snap_trace updating %llx %p %lld -> %lld\n", @@ -752,22 +757,30 @@ more: if (realm->seq > mdsc->last_snap_seq) mdsc->last_snap_seq = realm->seq; - invalidate = 1; + rebuild_snapcs = 1; } else if (!realm->cached_context) { dout("update_snap_trace %llx %p seq %lld new\n", realm->ino, realm, realm->seq); - invalidate = 1; + rebuild_snapcs = 1; } else { dout("update_snap_trace %llx %p seq %lld unchanged\n", realm->ino, realm, realm->seq); } - dout("done with %llx %p, invalidated=%d, %p %p\n", realm->ino, - realm, invalidate, p, e); + dout("done with %llx %p, rebuild_snapcs=%d, %p %p\n", realm->ino, + realm, rebuild_snapcs, p, e); + + /* + * this will always track the uppest parent realm from which + * we need to rebuild the snapshot contexts _downward_ in + * hierarchy. + */ + if (rebuild_snapcs) + realm_to_rebuild = realm; - /* invalidate when we reach the _end_ (root) of the trace */ - if (invalidate && p >= e) - rebuild_snap_realms(realm, &dirty_realms); + /* rebuild_snapcs when we reach the _end_ (root) of the trace */ + if (realm_to_rebuild && p >= e) + rebuild_snap_realms(realm_to_rebuild, &dirty_realms); if (!first_realm) first_realm = realm; @@ -994,6 +1007,19 @@ skip_inode: continue; adjust_snap_realm_parent(mdsc, child, realm->ino); } + } else { + /* + * In the non-split case both 'num_split_inos' and + * 'num_split_realms' should be 0, making this a no-op. + * However the MDS happens to populate 'split_realms' list + * in one of the UPDATE op cases by mistake. + * + * Skip both lists just in case to ensure that 'p' is + * positioned at the start of realm info, as expected by + * ceph_update_snap_trace(). + */ + p += sizeof(u64) * num_split_inos; + p += sizeof(u64) * num_split_realms; } /* diff --git a/fs/ceph/super.c b/fs/ceph/super.c index d40658d5e808..0e38678d5add 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -1174,14 +1174,23 @@ out_final: static void ceph_kill_sb(struct super_block *s) { struct ceph_fs_client *fsc = ceph_sb_to_client(s); - dev_t dev = s->s_dev; dout("kill_sb %p\n", s); ceph_mdsc_pre_umount(fsc->mdsc); flush_fs_workqueues(fsc); - generic_shutdown_super(s); + /* + * Though the kill_anon_super() will finally trigger the + * sync_filesystem() anyway, we still need to do it here + * and then bump the stage of shutdown to stop the work + * queue as earlier as possible. + */ + sync_filesystem(s); + + fsc->mdsc->stopping = CEPH_MDSC_STOPPING_FLUSHED; + + kill_anon_super(s); fsc->client->extra_mon_dispatch = NULL; ceph_fs_debugfs_cleanup(fsc); @@ -1189,7 +1198,6 @@ static void ceph_kill_sb(struct super_block *s) ceph_fscache_unregister_fs(fsc); destroy_fs_client(fsc); - free_anon_bdev(dev); } static struct file_system_type ceph_fs_type = { diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index cb18ee637cb7..4bcf0226818d 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -316,6 +316,14 @@ static ssize_t ceph_vxattrcb_snap_btime(struct ceph_inode_info *ci, char *val, } #define XATTR_RSTAT_FIELD(_type, _name) \ XATTR_NAME_CEPH(_type, _name, VXATTR_FLAG_RSTAT) +#define XATTR_RSTAT_FIELD_UPDATABLE(_type, _name) \ + { \ + .name = CEPH_XATTR_NAME(_type, _name), \ + .name_size = sizeof (CEPH_XATTR_NAME(_type, _name)), \ + .getxattr_cb = ceph_vxattrcb_ ## _type ## _ ## _name, \ + .exists_cb = NULL, \ + .flags = VXATTR_FLAG_RSTAT, \ + } #define XATTR_LAYOUT_FIELD(_type, _name, _field) \ { \ .name = CEPH_XATTR_NAME2(_type, _name, _field), \ @@ -353,7 +361,7 @@ static struct ceph_vxattr ceph_dir_vxattrs[] = { XATTR_RSTAT_FIELD(dir, rfiles), XATTR_RSTAT_FIELD(dir, rsubdirs), XATTR_RSTAT_FIELD(dir, rbytes), - XATTR_RSTAT_FIELD(dir, rctime), + XATTR_RSTAT_FIELD_UPDATABLE(dir, rctime), { .name = "ceph.dir.pin", .name_size = sizeof("ceph.dir.pin"), |