diff options
Diffstat (limited to 'fs/ceph')
| -rw-r--r-- | fs/ceph/caps.c | 74 | ||||
| -rw-r--r-- | fs/ceph/inode.c | 2 | ||||
| -rw-r--r-- | fs/ceph/locks.c | 74 | ||||
| -rw-r--r-- | fs/ceph/mds_client.c | 57 | ||||
| -rw-r--r-- | fs/ceph/mds_client.h | 7 | ||||
| -rw-r--r-- | fs/ceph/mdsmap.c | 7 | ||||
| -rw-r--r-- | fs/ceph/mdsmap.h | 6 | ||||
| -rw-r--r-- | fs/ceph/super.c | 18 | ||||
| -rw-r--r-- | fs/ceph/super.h | 2 |
9 files changed, 158 insertions, 89 deletions
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 9c02f328c966..7fb4aae97412 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -1452,7 +1452,7 @@ static void __prep_cap(struct cap_msg_args *arg, struct ceph_cap *cap, if (flushing & CEPH_CAP_XATTR_EXCL) { arg->old_xattr_buf = __ceph_build_xattrs_blob(ci); arg->xattr_version = ci->i_xattrs.version; - arg->xattr_buf = ci->i_xattrs.blob; + arg->xattr_buf = ceph_buffer_get(ci->i_xattrs.blob); } else { arg->xattr_buf = NULL; arg->old_xattr_buf = NULL; @@ -1553,6 +1553,7 @@ static void __send_cap(struct cap_msg_args *arg, struct ceph_inode_info *ci) encode_cap_msg(msg, arg); ceph_con_send(&arg->session->s_con, msg); ceph_buffer_put(arg->old_xattr_buf); + ceph_buffer_put(arg->xattr_buf); if (arg->wake) wake_up_all(&ci->i_cap_wq); } @@ -2155,6 +2156,30 @@ retry: ceph_cap_string(cap->implemented), ceph_cap_string(revoking)); + /* completed revocation? going down and there are no caps? */ + if (revoking) { + if ((revoking & cap_used) == 0) { + doutc(cl, "completed revocation of %s\n", + ceph_cap_string(cap->implemented & ~cap->issued)); + goto ack; + } + + /* + * If the "i_wrbuffer_ref" was increased by mmap or generic + * cache write just before the ceph_check_caps() is called, + * the Fb capability revoking will fail this time. Then we + * must wait for the BDI's delayed work to flush the dirty + * pages and to release the "i_wrbuffer_ref", which will cost + * at most 5 seconds. That means the MDS needs to wait at + * most 5 seconds to finished the Fb capability's revocation. + * + * Let's queue a writeback for it. + */ + if (S_ISREG(inode->i_mode) && ci->i_wrbuffer_ref && + (revoking & CEPH_CAP_FILE_BUFFER)) + queue_writeback = true; + } + if (cap == ci->i_auth_cap && (cap->issued & CEPH_CAP_FILE_WR)) { /* request larger max_size from MDS? */ @@ -2182,30 +2207,6 @@ retry: } } - /* completed revocation? going down and there are no caps? */ - if (revoking) { - if ((revoking & cap_used) == 0) { - doutc(cl, "completed revocation of %s\n", - ceph_cap_string(cap->implemented & ~cap->issued)); - goto ack; - } - - /* - * If the "i_wrbuffer_ref" was increased by mmap or generic - * cache write just before the ceph_check_caps() is called, - * the Fb capability revoking will fail this time. Then we - * must wait for the BDI's delayed work to flush the dirty - * pages and to release the "i_wrbuffer_ref", which will cost - * at most 5 seconds. That means the MDS needs to wait at - * most 5 seconds to finished the Fb capability's revocation. - * - * Let's queue a writeback for it. - */ - if (S_ISREG(inode->i_mode) && ci->i_wrbuffer_ref && - (revoking & CEPH_CAP_FILE_BUFFER)) - queue_writeback = true; - } - /* want more caps from mds? */ if (want & ~cap->mds_wanted) { if (want & ~(cap->mds_wanted | cap->issued)) @@ -3215,7 +3216,6 @@ static int ceph_try_drop_cap_snap(struct ceph_inode_info *ci, enum put_cap_refs_mode { PUT_CAP_REFS_SYNC = 0, - PUT_CAP_REFS_NO_CHECK, PUT_CAP_REFS_ASYNC, }; @@ -3331,11 +3331,6 @@ void ceph_put_cap_refs_async(struct ceph_inode_info *ci, int had) __ceph_put_cap_refs(ci, had, PUT_CAP_REFS_ASYNC); } -void ceph_put_cap_refs_no_check_caps(struct ceph_inode_info *ci, int had) -{ - __ceph_put_cap_refs(ci, had, PUT_CAP_REFS_NO_CHECK); -} - /* * Release @nr WRBUFFER refs on dirty pages for the given @snapc snap * context. Adjust per-snap dirty page accounting as appropriate. @@ -4777,7 +4772,22 @@ int ceph_drop_caps_for_unlink(struct inode *inode) if (__ceph_caps_dirty(ci)) { struct ceph_mds_client *mdsc = ceph_inode_to_fs_client(inode)->mdsc; - __cap_delay_requeue_front(mdsc, ci); + + doutc(mdsc->fsc->client, "%p %llx.%llx\n", inode, + ceph_vinop(inode)); + spin_lock(&mdsc->cap_unlink_delay_lock); + ci->i_ceph_flags |= CEPH_I_FLUSH; + if (!list_empty(&ci->i_cap_delay_list)) + list_del_init(&ci->i_cap_delay_list); + list_add_tail(&ci->i_cap_delay_list, + &mdsc->cap_unlink_delay_list); + spin_unlock(&mdsc->cap_unlink_delay_lock); + + /* + * Fire the work immediately, because the MDS maybe + * waiting for caps release. + */ + ceph_queue_cap_unlink_work(mdsc); } } spin_unlock(&ci->i_ceph_lock); diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 0c25d326afc4..7b2e77517f23 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -78,6 +78,8 @@ struct inode *ceph_new_inode(struct inode *dir, struct dentry *dentry, if (!inode) return ERR_PTR(-ENOMEM); + inode->i_blkbits = CEPH_FSCRYPT_BLOCK_SHIFT; + if (!S_ISLNK(*mode)) { err = ceph_pre_init_acls(dir, mode, as_ctx); if (err < 0) diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index e07ad29ff8b9..ebf4ac0055dd 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -33,7 +33,7 @@ void __init ceph_flock_init(void) static void ceph_fl_copy_lock(struct file_lock *dst, struct file_lock *src) { - struct inode *inode = file_inode(dst->fl_file); + struct inode *inode = file_inode(dst->c.flc_file); atomic_inc(&ceph_inode(inode)->i_filelock_ref); dst->fl_u.ceph.inode = igrab(inode); } @@ -110,17 +110,18 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode, else length = fl->fl_end - fl->fl_start + 1; - owner = secure_addr(fl->fl_owner); + owner = secure_addr(fl->c.flc_owner); doutc(cl, "rule: %d, op: %d, owner: %llx, pid: %llu, " "start: %llu, length: %llu, wait: %d, type: %d\n", - (int)lock_type, (int)operation, owner, (u64)fl->fl_pid, - fl->fl_start, length, wait, fl->fl_type); + (int)lock_type, (int)operation, owner, + (u64) fl->c.flc_pid, + fl->fl_start, length, wait, fl->c.flc_type); req->r_args.filelock_change.rule = lock_type; req->r_args.filelock_change.type = cmd; req->r_args.filelock_change.owner = cpu_to_le64(owner); - req->r_args.filelock_change.pid = cpu_to_le64((u64)fl->fl_pid); + req->r_args.filelock_change.pid = cpu_to_le64((u64) fl->c.flc_pid); req->r_args.filelock_change.start = cpu_to_le64(fl->fl_start); req->r_args.filelock_change.length = cpu_to_le64(length); req->r_args.filelock_change.wait = wait; @@ -130,13 +131,13 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode, err = ceph_mdsc_wait_request(mdsc, req, wait ? ceph_lock_wait_for_completion : NULL); if (!err && operation == CEPH_MDS_OP_GETFILELOCK) { - fl->fl_pid = -le64_to_cpu(req->r_reply_info.filelock_reply->pid); + fl->c.flc_pid = -le64_to_cpu(req->r_reply_info.filelock_reply->pid); if (CEPH_LOCK_SHARED == req->r_reply_info.filelock_reply->type) - fl->fl_type = F_RDLCK; + fl->c.flc_type = F_RDLCK; else if (CEPH_LOCK_EXCL == req->r_reply_info.filelock_reply->type) - fl->fl_type = F_WRLCK; + fl->c.flc_type = F_WRLCK; else - fl->fl_type = F_UNLCK; + fl->c.flc_type = F_UNLCK; fl->fl_start = le64_to_cpu(req->r_reply_info.filelock_reply->start); length = le64_to_cpu(req->r_reply_info.filelock_reply->start) + @@ -150,8 +151,8 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct inode *inode, ceph_mdsc_put_request(req); doutc(cl, "rule: %d, op: %d, pid: %llu, start: %llu, " "length: %llu, wait: %d, type: %d, err code %d\n", - (int)lock_type, (int)operation, (u64)fl->fl_pid, - fl->fl_start, length, wait, fl->fl_type, err); + (int)lock_type, (int)operation, (u64) fl->c.flc_pid, + fl->fl_start, length, wait, fl->c.flc_type, err); return err; } @@ -227,10 +228,10 @@ static int ceph_lock_wait_for_completion(struct ceph_mds_client *mdsc, static int try_unlock_file(struct file *file, struct file_lock *fl) { int err; - unsigned int orig_flags = fl->fl_flags; - fl->fl_flags |= FL_EXISTS; + unsigned int orig_flags = fl->c.flc_flags; + fl->c.flc_flags |= FL_EXISTS; err = locks_lock_file_wait(file, fl); - fl->fl_flags = orig_flags; + fl->c.flc_flags = orig_flags; if (err == -ENOENT) { if (!(orig_flags & FL_EXISTS)) err = 0; @@ -253,13 +254,13 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) u8 wait = 0; u8 lock_cmd; - if (!(fl->fl_flags & FL_POSIX)) + if (!(fl->c.flc_flags & FL_POSIX)) return -ENOLCK; if (ceph_inode_is_shutdown(inode)) return -ESTALE; - doutc(cl, "fl_owner: %p\n", fl->fl_owner); + doutc(cl, "fl_owner: %p\n", fl->c.flc_owner); /* set wait bit as appropriate, then make command as Ceph expects it*/ if (IS_GETLK(cmd)) @@ -273,19 +274,19 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) } spin_unlock(&ci->i_ceph_lock); if (err < 0) { - if (op == CEPH_MDS_OP_SETFILELOCK && F_UNLCK == fl->fl_type) + if (op == CEPH_MDS_OP_SETFILELOCK && lock_is_unlock(fl)) posix_lock_file(file, fl, NULL); return err; } - if (F_RDLCK == fl->fl_type) + if (lock_is_read(fl)) lock_cmd = CEPH_LOCK_SHARED; - else if (F_WRLCK == fl->fl_type) + else if (lock_is_write(fl)) lock_cmd = CEPH_LOCK_EXCL; else lock_cmd = CEPH_LOCK_UNLOCK; - if (op == CEPH_MDS_OP_SETFILELOCK && F_UNLCK == fl->fl_type) { + if (op == CEPH_MDS_OP_SETFILELOCK && lock_is_unlock(fl)) { err = try_unlock_file(file, fl); if (err <= 0) return err; @@ -293,7 +294,7 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) err = ceph_lock_message(CEPH_LOCK_FCNTL, op, inode, lock_cmd, wait, fl); if (!err) { - if (op == CEPH_MDS_OP_SETFILELOCK && F_UNLCK != fl->fl_type) { + if (op == CEPH_MDS_OP_SETFILELOCK && F_UNLCK != fl->c.flc_type) { doutc(cl, "locking locally\n"); err = posix_lock_file(file, fl, NULL); if (err) { @@ -319,13 +320,13 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) u8 wait = 0; u8 lock_cmd; - if (!(fl->fl_flags & FL_FLOCK)) + if (!(fl->c.flc_flags & FL_FLOCK)) return -ENOLCK; if (ceph_inode_is_shutdown(inode)) return -ESTALE; - doutc(cl, "fl_file: %p\n", fl->fl_file); + doutc(cl, "fl_file: %p\n", fl->c.flc_file); spin_lock(&ci->i_ceph_lock); if (ci->i_ceph_flags & CEPH_I_ERROR_FILELOCK) { @@ -333,7 +334,7 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) } spin_unlock(&ci->i_ceph_lock); if (err < 0) { - if (F_UNLCK == fl->fl_type) + if (lock_is_unlock(fl)) locks_lock_file_wait(file, fl); return err; } @@ -341,14 +342,14 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) if (IS_SETLKW(cmd)) wait = 1; - if (F_RDLCK == fl->fl_type) + if (lock_is_read(fl)) lock_cmd = CEPH_LOCK_SHARED; - else if (F_WRLCK == fl->fl_type) + else if (lock_is_write(fl)) lock_cmd = CEPH_LOCK_EXCL; else lock_cmd = CEPH_LOCK_UNLOCK; - if (F_UNLCK == fl->fl_type) { + if (lock_is_unlock(fl)) { err = try_unlock_file(file, fl); if (err <= 0) return err; @@ -356,7 +357,7 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) err = ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK, inode, lock_cmd, wait, fl); - if (!err && F_UNLCK != fl->fl_type) { + if (!err && F_UNLCK != fl->c.flc_type) { err = locks_lock_file_wait(file, fl); if (err) { ceph_lock_message(CEPH_LOCK_FLOCK, @@ -385,9 +386,9 @@ void ceph_count_locks(struct inode *inode, int *fcntl_count, int *flock_count) ctx = locks_inode_context(inode); if (ctx) { spin_lock(&ctx->flc_lock); - list_for_each_entry(lock, &ctx->flc_posix, fl_list) + for_each_file_lock(lock, &ctx->flc_posix) ++(*fcntl_count); - list_for_each_entry(lock, &ctx->flc_flock, fl_list) + for_each_file_lock(lock, &ctx->flc_flock) ++(*flock_count); spin_unlock(&ctx->flc_lock); } @@ -408,10 +409,10 @@ static int lock_to_ceph_filelock(struct inode *inode, cephlock->start = cpu_to_le64(lock->fl_start); cephlock->length = cpu_to_le64(lock->fl_end - lock->fl_start + 1); cephlock->client = cpu_to_le64(0); - cephlock->pid = cpu_to_le64((u64)lock->fl_pid); - cephlock->owner = cpu_to_le64(secure_addr(lock->fl_owner)); + cephlock->pid = cpu_to_le64((u64) lock->c.flc_pid); + cephlock->owner = cpu_to_le64(secure_addr(lock->c.flc_owner)); - switch (lock->fl_type) { + switch (lock->c.flc_type) { case F_RDLCK: cephlock->type = CEPH_LOCK_SHARED; break; @@ -422,7 +423,8 @@ static int lock_to_ceph_filelock(struct inode *inode, cephlock->type = CEPH_LOCK_UNLOCK; break; default: - doutc(cl, "Have unknown lock type %d\n", lock->fl_type); + doutc(cl, "Have unknown lock type %d\n", + lock->c.flc_type); err = -EINVAL; } @@ -453,7 +455,7 @@ int ceph_encode_locks_to_buffer(struct inode *inode, return 0; spin_lock(&ctx->flc_lock); - list_for_each_entry(lock, &ctx->flc_posix, fl_list) { + for_each_file_lock(lock, &ctx->flc_posix) { ++seen_fcntl; if (seen_fcntl > num_fcntl_locks) { err = -ENOSPC; @@ -464,7 +466,7 @@ int ceph_encode_locks_to_buffer(struct inode *inode, goto fail; ++l; } - list_for_each_entry(lock, &ctx->flc_flock, fl_list) { + for_each_file_lock(lock, &ctx->flc_flock) { ++seen_flock; if (seen_flock > num_flock_locks) { err = -ENOSPC; diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 548d1de379f3..3ab9c268a8bb 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -1089,7 +1089,7 @@ void ceph_mdsc_release_request(struct kref *kref) struct ceph_mds_request *req = container_of(kref, struct ceph_mds_request, r_kref); - ceph_mdsc_release_dir_caps_no_check(req); + ceph_mdsc_release_dir_caps_async(req); destroy_reply_info(&req->r_reply_info); if (req->r_request) ceph_msg_put(req->r_request); @@ -2484,6 +2484,50 @@ void ceph_reclaim_caps_nr(struct ceph_mds_client *mdsc, int nr) } } +void ceph_queue_cap_unlink_work(struct ceph_mds_client *mdsc) +{ + struct ceph_client *cl = mdsc->fsc->client; + if (mdsc->stopping) + return; + + if (queue_work(mdsc->fsc->cap_wq, &mdsc->cap_unlink_work)) { + doutc(cl, "caps unlink work queued\n"); + } else { + doutc(cl, "failed to queue caps unlink work\n"); + } +} + +static void ceph_cap_unlink_work(struct work_struct *work) +{ + struct ceph_mds_client *mdsc = + container_of(work, struct ceph_mds_client, cap_unlink_work); + struct ceph_client *cl = mdsc->fsc->client; + + doutc(cl, "begin\n"); + spin_lock(&mdsc->cap_unlink_delay_lock); + while (!list_empty(&mdsc->cap_unlink_delay_list)) { + struct ceph_inode_info *ci; + struct inode *inode; + + ci = list_first_entry(&mdsc->cap_unlink_delay_list, + struct ceph_inode_info, + i_cap_delay_list); + list_del_init(&ci->i_cap_delay_list); + + inode = igrab(&ci->netfs.inode); + if (inode) { + spin_unlock(&mdsc->cap_unlink_delay_lock); + doutc(cl, "on %p %llx.%llx\n", inode, + ceph_vinop(inode)); + ceph_check_caps(ci, CHECK_CAPS_FLUSH); + iput(inode); + spin_lock(&mdsc->cap_unlink_delay_lock); + } + } + spin_unlock(&mdsc->cap_unlink_delay_lock); + doutc(cl, "done\n"); +} + /* * requests */ @@ -4261,7 +4305,7 @@ void ceph_mdsc_release_dir_caps(struct ceph_mds_request *req) } } -void ceph_mdsc_release_dir_caps_no_check(struct ceph_mds_request *req) +void ceph_mdsc_release_dir_caps_async(struct ceph_mds_request *req) { struct ceph_client *cl = req->r_mdsc->fsc->client; int dcaps; @@ -4269,8 +4313,7 @@ void ceph_mdsc_release_dir_caps_no_check(struct ceph_mds_request *req) dcaps = xchg(&req->r_dir_caps, 0); if (dcaps) { doutc(cl, "releasing r_dir_caps=%s\n", ceph_cap_string(dcaps)); - ceph_put_cap_refs_no_check_caps(ceph_inode(req->r_parent), - dcaps); + ceph_put_cap_refs_async(ceph_inode(req->r_parent), dcaps); } } @@ -4306,7 +4349,7 @@ static void replay_unsafe_requests(struct ceph_mds_client *mdsc, if (req->r_session->s_mds != session->s_mds) continue; - ceph_mdsc_release_dir_caps_no_check(req); + ceph_mdsc_release_dir_caps_async(req); __send_request(session, req, true); } @@ -5360,6 +5403,8 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) 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->cap_unlink_delay_list); + spin_lock_init(&mdsc->cap_unlink_delay_lock); INIT_LIST_HEAD(&mdsc->snap_flush_list); spin_lock_init(&mdsc->snap_flush_lock); mdsc->last_cap_flush_tid = 1; @@ -5368,6 +5413,7 @@ int ceph_mdsc_init(struct ceph_fs_client *fsc) spin_lock_init(&mdsc->cap_dirty_lock); init_waitqueue_head(&mdsc->cap_flushing_wq); INIT_WORK(&mdsc->cap_reclaim_work, ceph_cap_reclaim_work); + INIT_WORK(&mdsc->cap_unlink_work, ceph_cap_unlink_work); err = ceph_metric_init(&mdsc->metric); if (err) goto err_mdsmap; @@ -5641,6 +5687,7 @@ void ceph_mdsc_close_sessions(struct ceph_mds_client *mdsc) ceph_cleanup_global_and_empty_realms(mdsc); cancel_work_sync(&mdsc->cap_reclaim_work); + cancel_work_sync(&mdsc->cap_unlink_work); cancel_delayed_work_sync(&mdsc->delayed_work); /* cancel timer */ doutc(cl, "done\n"); diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index 2e6ddaa13d72..03f8ff00874f 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -462,6 +462,8 @@ struct ceph_mds_client { unsigned long last_renew_caps; /* last time we renewed our caps */ struct list_head cap_delay_list; /* caps with delayed release */ spinlock_t cap_delay_lock; /* protects cap_delay_list */ + struct list_head cap_unlink_delay_list; /* caps with delayed release for unlink */ + spinlock_t cap_unlink_delay_lock; /* protects cap_unlink_delay_list */ struct list_head snap_flush_list; /* cap_snaps ready to flush */ spinlock_t snap_flush_lock; @@ -475,6 +477,8 @@ struct ceph_mds_client { struct work_struct cap_reclaim_work; atomic_t cap_reclaim_pending; + struct work_struct cap_unlink_work; + /* * Cap reservations * @@ -552,7 +556,7 @@ extern int ceph_mdsc_do_request(struct ceph_mds_client *mdsc, struct inode *dir, struct ceph_mds_request *req); extern void ceph_mdsc_release_dir_caps(struct ceph_mds_request *req); -extern void ceph_mdsc_release_dir_caps_no_check(struct ceph_mds_request *req); +extern void ceph_mdsc_release_dir_caps_async(struct ceph_mds_request *req); static inline void ceph_mdsc_get_request(struct ceph_mds_request *req) { kref_get(&req->r_kref); @@ -574,6 +578,7 @@ extern void ceph_flush_cap_releases(struct ceph_mds_client *mdsc, struct ceph_mds_session *session); extern void ceph_queue_cap_reclaim_work(struct ceph_mds_client *mdsc); extern void ceph_reclaim_caps_nr(struct ceph_mds_client *mdsc, int nr); +extern void ceph_queue_cap_unlink_work(struct ceph_mds_client *mdsc); extern int ceph_iterate_session_caps(struct ceph_mds_session *session, int (*cb)(struct inode *, int mds, void *), void *arg); diff --git a/fs/ceph/mdsmap.c b/fs/ceph/mdsmap.c index fae97c25ce58..8109aba66e02 100644 --- a/fs/ceph/mdsmap.c +++ b/fs/ceph/mdsmap.c @@ -380,10 +380,11 @@ struct ceph_mdsmap *ceph_mdsmap_decode(struct ceph_mds_client *mdsc, void **p, ceph_decode_skip_8(p, end, bad_ext); /* required_client_features */ ceph_decode_skip_set(p, end, 64, bad_ext); + /* bal_rank_mask */ + ceph_decode_skip_string(p, end, bad_ext); + } + if (mdsmap_ev >= 18) { ceph_decode_64_safe(p, end, m->m_max_xattr_size, bad_ext); - } else { - /* This forces the usage of the (sync) SETXATTR Op */ - m->m_max_xattr_size = 0; } bad_ext: doutc(cl, "m_enabled: %d, m_damaged: %d, m_num_laggy: %d\n", diff --git a/fs/ceph/mdsmap.h b/fs/ceph/mdsmap.h index 89f1931f1ba6..1f2171dd01bf 100644 --- a/fs/ceph/mdsmap.h +++ b/fs/ceph/mdsmap.h @@ -27,7 +27,11 @@ struct ceph_mdsmap { u32 m_session_timeout; /* seconds */ u32 m_session_autoclose; /* seconds */ u64 m_max_file_size; - u64 m_max_xattr_size; /* maximum size for xattrs blob */ + /* + * maximum size for xattrs blob. + * Zeroed by default to force the usage of the (sync) SETXATTR Op. + */ + u64 m_max_xattr_size; u32 m_max_mds; /* expected up:active mds number */ u32 m_num_active_mds; /* actual up:active mds number */ u32 possible_max_rank; /* possible max rank index */ diff --git a/fs/ceph/super.c b/fs/ceph/super.c index 5ec102f6b1ac..885cb5d4e771 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -928,36 +928,36 @@ static int __init init_caches(void) ceph_inode_cachep = kmem_cache_create("ceph_inode_info", sizeof(struct ceph_inode_info), __alignof__(struct ceph_inode_info), - SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD| - SLAB_ACCOUNT, ceph_inode_init_once); + SLAB_RECLAIM_ACCOUNT | SLAB_ACCOUNT, + ceph_inode_init_once); if (!ceph_inode_cachep) return -ENOMEM; - ceph_cap_cachep = KMEM_CACHE(ceph_cap, SLAB_MEM_SPREAD); + ceph_cap_cachep = KMEM_CACHE(ceph_cap, 0); if (!ceph_cap_cachep) goto bad_cap; - ceph_cap_snap_cachep = KMEM_CACHE(ceph_cap_snap, SLAB_MEM_SPREAD); + ceph_cap_snap_cachep = KMEM_CACHE(ceph_cap_snap, 0); if (!ceph_cap_snap_cachep) goto bad_cap_snap; ceph_cap_flush_cachep = KMEM_CACHE(ceph_cap_flush, - SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD); + SLAB_RECLAIM_ACCOUNT); if (!ceph_cap_flush_cachep) goto bad_cap_flush; ceph_dentry_cachep = KMEM_CACHE(ceph_dentry_info, - SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD); + SLAB_RECLAIM_ACCOUNT); if (!ceph_dentry_cachep) goto bad_dentry; - ceph_file_cachep = KMEM_CACHE(ceph_file_info, SLAB_MEM_SPREAD); + ceph_file_cachep = KMEM_CACHE(ceph_file_info, 0); if (!ceph_file_cachep) goto bad_file; - ceph_dir_file_cachep = KMEM_CACHE(ceph_dir_file_info, SLAB_MEM_SPREAD); + ceph_dir_file_cachep = KMEM_CACHE(ceph_dir_file_info, 0); if (!ceph_dir_file_cachep) goto bad_dir_file; - ceph_mds_request_cachep = KMEM_CACHE(ceph_mds_request, SLAB_MEM_SPREAD); + ceph_mds_request_cachep = KMEM_CACHE(ceph_mds_request, 0); if (!ceph_mds_request_cachep) goto bad_mds_req; diff --git a/fs/ceph/super.h b/fs/ceph/super.h index b06e2bc86221..b63b4cd9b5b6 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -1255,8 +1255,6 @@ extern void ceph_take_cap_refs(struct ceph_inode_info *ci, int caps, extern void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps); extern void ceph_put_cap_refs(struct ceph_inode_info *ci, int had); extern void ceph_put_cap_refs_async(struct ceph_inode_info *ci, int had); -extern void ceph_put_cap_refs_no_check_caps(struct ceph_inode_info *ci, - int had); extern void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, struct ceph_snap_context *snapc); extern void __ceph_remove_capsnap(struct inode *inode, |