diff options
Diffstat (limited to 'fs')
202 files changed, 9146 insertions, 5835 deletions
diff --git a/fs/afs/addr_list.c b/fs/afs/addr_list.c index df415c05939e..de1ae0bead3b 100644 --- a/fs/afs/addr_list.c +++ b/fs/afs/addr_list.c @@ -19,7 +19,7 @@ void afs_put_addrlist(struct afs_addr_list *alist) { if (alist && refcount_dec_and_test(&alist->usage)) - call_rcu(&alist->rcu, (rcu_callback_t)kfree); + kfree_rcu(alist, rcu); } /* diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c index ff3994a6be23..6765949b3aab 100644 --- a/fs/afs/cmservice.c +++ b/fs/afs/cmservice.c @@ -244,6 +244,17 @@ static void afs_cm_destructor(struct afs_call *call) } /* + * Abort a service call from within an action function. + */ +static void afs_abort_service_call(struct afs_call *call, u32 abort_code, int error, + const char *why) +{ + rxrpc_kernel_abort_call(call->net->socket, call->rxcall, + abort_code, error, why); + afs_set_call_complete(call, error, 0); +} + +/* * The server supplied a list of callbacks that it wanted to break. */ static void SRXAFSCB_CallBack(struct work_struct *work) @@ -510,8 +521,7 @@ static void SRXAFSCB_ProbeUuid(struct work_struct *work) if (memcmp(r, &call->net->uuid, sizeof(call->net->uuid)) == 0) afs_send_empty_reply(call); else - rxrpc_kernel_abort_call(call->net->socket, call->rxcall, - 1, 1, "K-1"); + afs_abort_service_call(call, 1, 1, "K-1"); afs_put_call(call); _leave(""); diff --git a/fs/afs/fs_probe.c b/fs/afs/fs_probe.c index cfe62b154f68..e1b9ed679045 100644 --- a/fs/afs/fs_probe.c +++ b/fs/afs/fs_probe.c @@ -145,6 +145,7 @@ static int afs_do_probe_fileserver(struct afs_net *net, read_lock(&server->fs_lock); ac.alist = rcu_dereference_protected(server->addresses, lockdep_is_held(&server->fs_lock)); + afs_get_addrlist(ac.alist); read_unlock(&server->fs_lock); atomic_set(&server->probe_outstanding, ac.alist->nr_addrs); @@ -163,6 +164,7 @@ static int afs_do_probe_fileserver(struct afs_net *net, if (!in_progress) afs_fs_probe_done(server); + afs_put_addrlist(ac.alist); return in_progress; } diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 1d81fc4c3058..ef732dd4e7ef 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -81,7 +81,7 @@ enum afs_call_state { * List of server addresses. */ struct afs_addr_list { - struct rcu_head rcu; /* Must be first */ + struct rcu_head rcu; refcount_t usage; u32 version; /* Version */ unsigned char max_addrs; @@ -154,7 +154,7 @@ struct afs_call { }; unsigned char unmarshall; /* unmarshalling phase */ unsigned char addr_ix; /* Address in ->alist */ - bool incoming; /* T if incoming call */ + bool drop_ref; /* T if need to drop ref for incoming call */ bool send_pages; /* T if data from mapping should be sent */ bool need_attention; /* T if RxRPC poked us */ bool async; /* T if asynchronous */ @@ -1209,8 +1209,16 @@ static inline void afs_set_call_complete(struct afs_call *call, ok = true; } spin_unlock_bh(&call->state_lock); - if (ok) + if (ok) { trace_afs_call_done(call); + + /* Asynchronous calls have two refs to release - one from the alloc and + * one queued with the work item - and we can't just deallocate the + * call because the work item may be queued again. + */ + if (call->drop_ref) + afs_put_call(call); + } } /* diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c index 58d396592250..1ecc67da6c1a 100644 --- a/fs/afs/rxrpc.c +++ b/fs/afs/rxrpc.c @@ -18,7 +18,6 @@ struct workqueue_struct *afs_async_calls; static void afs_wake_up_call_waiter(struct sock *, struct rxrpc_call *, unsigned long); static void afs_wake_up_async_call(struct sock *, struct rxrpc_call *, unsigned long); -static void afs_delete_async_call(struct work_struct *); static void afs_process_async_call(struct work_struct *); static void afs_rx_new_call(struct sock *, struct rxrpc_call *, unsigned long); static void afs_rx_discard_new_call(struct rxrpc_call *, unsigned long); @@ -169,7 +168,7 @@ void afs_put_call(struct afs_call *call) int n = atomic_dec_return(&call->usage); int o = atomic_read(&net->nr_outstanding_calls); - trace_afs_call(call, afs_call_trace_put, n + 1, o, + trace_afs_call(call, afs_call_trace_put, n, o, __builtin_return_address(0)); ASSERTCMP(n, >=, 0); @@ -402,8 +401,10 @@ void afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, gfp_t gfp) /* If the call is going to be asynchronous, we need an extra ref for * the call to hold itself so the caller need not hang on to its ref. */ - if (call->async) + if (call->async) { afs_get_call(call, afs_call_trace_get); + call->drop_ref = true; + } /* create a call */ rxcall = rxrpc_kernel_begin_call(call->net->socket, srx, call->key, @@ -413,7 +414,8 @@ void afs_make_call(struct afs_addr_cursor *ac, struct afs_call *call, gfp_t gfp) afs_wake_up_async_call : afs_wake_up_call_waiter), call->upgrade, - call->intr, + (call->intr ? RXRPC_PREINTERRUPTIBLE : + RXRPC_UNINTERRUPTIBLE), call->debug_id); if (IS_ERR(rxcall)) { ret = PTR_ERR(rxcall); @@ -584,8 +586,6 @@ static void afs_deliver_to_call(struct afs_call *call) done: if (call->type->done) call->type->done(call); - if (state == AFS_CALL_COMPLETE && call->incoming) - afs_put_call(call); out: _leave(""); return; @@ -604,11 +604,7 @@ call_complete: long afs_wait_for_call_to_complete(struct afs_call *call, struct afs_addr_cursor *ac) { - signed long rtt2, timeout; long ret; - bool stalled = false; - u64 rtt; - u32 life, last_life; bool rxrpc_complete = false; DECLARE_WAITQUEUE(myself, current); @@ -619,14 +615,6 @@ long afs_wait_for_call_to_complete(struct afs_call *call, if (ret < 0) goto out; - rtt = rxrpc_kernel_get_rtt(call->net->socket, call->rxcall); - rtt2 = nsecs_to_jiffies64(rtt) * 2; - if (rtt2 < 2) - rtt2 = 2; - - timeout = rtt2; - rxrpc_kernel_check_life(call->net->socket, call->rxcall, &last_life); - add_wait_queue(&call->waitq, &myself); for (;;) { set_current_state(TASK_UNINTERRUPTIBLE); @@ -637,37 +625,19 @@ long afs_wait_for_call_to_complete(struct afs_call *call, call->need_attention = false; __set_current_state(TASK_RUNNING); afs_deliver_to_call(call); - timeout = rtt2; continue; } if (afs_check_call_state(call, AFS_CALL_COMPLETE)) break; - if (!rxrpc_kernel_check_life(call->net->socket, call->rxcall, &life)) { + if (!rxrpc_kernel_check_life(call->net->socket, call->rxcall)) { /* rxrpc terminated the call. */ rxrpc_complete = true; break; } - if (call->intr && timeout == 0 && - life == last_life && signal_pending(current)) { - if (stalled) - break; - __set_current_state(TASK_RUNNING); - rxrpc_kernel_probe_life(call->net->socket, call->rxcall); - timeout = rtt2; - stalled = true; - continue; - } - - if (life != last_life) { - timeout = rtt2; - last_life = life; - stalled = false; - } - - timeout = schedule_timeout(timeout); + schedule(); } remove_wait_queue(&call->waitq, &myself); @@ -735,7 +705,7 @@ static void afs_wake_up_async_call(struct sock *sk, struct rxrpc_call *rxcall, u = atomic_fetch_add_unless(&call->usage, 1, 0); if (u != 0) { - trace_afs_call(call, afs_call_trace_wake, u, + trace_afs_call(call, afs_call_trace_wake, u + 1, atomic_read(&call->net->nr_outstanding_calls), __builtin_return_address(0)); @@ -745,21 +715,6 @@ static void afs_wake_up_async_call(struct sock *sk, struct rxrpc_call *rxcall, } /* - * Delete an asynchronous call. The work item carries a ref to the call struct - * that we need to release. - */ -static void afs_delete_async_call(struct work_struct *work) -{ - struct afs_call *call = container_of(work, struct afs_call, async_work); - - _enter(""); - - afs_put_call(call); - - _leave(""); -} - -/* * Perform I/O processing on an asynchronous call. The work item carries a ref * to the call struct that we either need to release or to pass on. */ @@ -774,16 +729,6 @@ static void afs_process_async_call(struct work_struct *work) afs_deliver_to_call(call); } - if (call->state == AFS_CALL_COMPLETE) { - /* We have two refs to release - one from the alloc and one - * queued with the work item - and we can't just deallocate the - * call because the work item may be queued again. - */ - call->async_work.func = afs_delete_async_call; - if (!queue_work(afs_async_calls, &call->async_work)) - afs_put_call(call); - } - afs_put_call(call); _leave(""); } @@ -810,6 +755,7 @@ void afs_charge_preallocation(struct work_struct *work) if (!call) break; + call->drop_ref = true; call->async = true; call->state = AFS_CALL_SV_AWAIT_OP_ID; init_waitqueue_head(&call->waitq); diff --git a/fs/block_dev.c b/fs/block_dev.c index 69bf2fb6f7cd..9501880dff5e 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1520,10 +1520,22 @@ rescan: if (ret) return ret; - if (invalidate) - set_capacity(disk, 0); - else if (disk->fops->revalidate_disk) - disk->fops->revalidate_disk(disk); + /* + * Historically we only set the capacity to zero for devices that + * support partitions (independ of actually having partitions created). + * Doing that is rather inconsistent, but changing it broke legacy + * udisks polling for legacy ide-cdrom devices. Use the crude check + * below to get the sane behavior for most device while not breaking + * userspace for this particular setup. + */ + if (invalidate) { + if (disk_part_scan_enabled(disk) || + !(disk->flags & GENHD_FL_REMOVABLE)) + set_capacity(disk, 0); + } else { + if (disk->fops->revalidate_disk) + disk->fops->revalidate_disk(disk); + } check_disk_size_change(disk, bdev, !invalidate); diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile index 9a0ff3384381..e738f6206ea5 100644 --- a/fs/btrfs/Makefile +++ b/fs/btrfs/Makefile @@ -11,7 +11,7 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \ compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o \ reada.o backref.o ulist.o qgroup.o send.o dev-replace.o raid56.o \ uuid-tree.o props.o free-space-tree.o tree-checker.o space-info.o \ - block-rsv.o delalloc-space.o block-group.o discard.o + block-rsv.o delalloc-space.o block-group.o discard.o reflink.o btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o diff --git a/fs/btrfs/async-thread.c b/fs/btrfs/async-thread.c index 1d32a07bb2d1..309516e6a968 100644 --- a/fs/btrfs/async-thread.c +++ b/fs/btrfs/async-thread.c @@ -395,3 +395,11 @@ void btrfs_set_work_high_priority(struct btrfs_work *work) { set_bit(WORK_HIGH_PRIO_BIT, &work->flags); } + +void btrfs_flush_workqueue(struct btrfs_workqueue *wq) +{ + if (wq->high) + flush_workqueue(wq->high->normal_wq); + + flush_workqueue(wq->normal->normal_wq); +} diff --git a/fs/btrfs/async-thread.h b/fs/btrfs/async-thread.h index a4434301d84d..3204daa51b95 100644 --- a/fs/btrfs/async-thread.h +++ b/fs/btrfs/async-thread.h @@ -44,5 +44,6 @@ void btrfs_set_work_high_priority(struct btrfs_work *work); struct btrfs_fs_info * __pure btrfs_work_owner(const struct btrfs_work *work); struct btrfs_fs_info * __pure btrfs_workqueue_owner(const struct __btrfs_workqueue *wq); bool btrfs_workqueue_normal_congested(const struct btrfs_workqueue *wq); +void btrfs_flush_workqueue(struct btrfs_workqueue *wq); #endif diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index e5d85311d5d5..9c380e7edf62 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -347,33 +347,10 @@ static int add_prelim_ref(const struct btrfs_fs_info *fs_info, return -ENOMEM; ref->root_id = root_id; - if (key) { + if (key) ref->key_for_search = *key; - /* - * We can often find data backrefs with an offset that is too - * large (>= LLONG_MAX, maximum allowed file offset) due to - * underflows when subtracting a file's offset with the data - * offset of its corresponding extent data item. This can - * happen for example in the clone ioctl. - * So if we detect such case we set the search key's offset to - * zero to make sure we will find the matching file extent item - * at add_all_parents(), otherwise we will miss it because the - * offset taken form the backref is much larger then the offset - * of the file extent item. This can make us scan a very large - * number of file extent items, but at least it will not make - * us miss any. - * This is an ugly workaround for a behaviour that should have - * never existed, but it does and a fix for the clone ioctl - * would touch a lot of places, cause backwards incompatibility - * and would not fix the problem for extents cloned with older - * kernels. - */ - if (ref->key_for_search.type == BTRFS_EXTENT_DATA_KEY && - ref->key_for_search.offset >= LLONG_MAX) - ref->key_for_search.offset = 0; - } else { + else memset(&ref->key_for_search, 0, sizeof(ref->key_for_search)); - } ref->inode_list = NULL; ref->level = level; @@ -409,10 +386,36 @@ static int add_indirect_ref(const struct btrfs_fs_info *fs_info, wanted_disk_byte, count, sc, gfp_mask); } +static int is_shared_data_backref(struct preftrees *preftrees, u64 bytenr) +{ + struct rb_node **p = &preftrees->direct.root.rb_root.rb_node; + struct rb_node *parent = NULL; + struct prelim_ref *ref = NULL; + struct prelim_ref target = {0}; + int result; + + target.parent = bytenr; + + while (*p) { + parent = *p; + ref = rb_entry(parent, struct prelim_ref, rbnode); + result = prelim_ref_compare(ref, &target); + + if (result < 0) + p = &(*p)->rb_left; + else if (result > 0) + p = &(*p)->rb_right; + else + return 1; + } + return 0; +} + static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path, - struct ulist *parents, struct prelim_ref *ref, + struct ulist *parents, + struct preftrees *preftrees, struct prelim_ref *ref, int level, u64 time_seq, const u64 *extent_item_pos, - u64 total_refs, bool ignore_offset) + bool ignore_offset) { int ret = 0; int slot; @@ -424,6 +427,7 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path, u64 disk_byte; u64 wanted_disk_byte = ref->wanted_disk_byte; u64 count = 0; + u64 data_offset; if (level != 0) { eb = path->nodes[level]; @@ -434,18 +438,26 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path, } /* - * We normally enter this function with the path already pointing to - * the first item to check. But sometimes, we may enter it with - * slot==nritems. In that case, go to the next leaf before we continue. + * 1. We normally enter this function with the path already pointing to + * the first item to check. But sometimes, we may enter it with + * slot == nritems. + * 2. We are searching for normal backref but bytenr of this leaf + * matches shared data backref + * 3. The leaf owner is not equal to the root we are searching + * + * For these cases, go to the next leaf before we continue. */ - if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) { + eb = path->nodes[0]; + if (path->slots[0] >= btrfs_header_nritems(eb) || + is_shared_data_backref(preftrees, eb->start) || + ref->root_id != btrfs_header_owner(eb)) { if (time_seq == SEQ_LAST) ret = btrfs_next_leaf(root, path); else ret = btrfs_next_old_leaf(root, path, time_seq); } - while (!ret && count < total_refs) { + while (!ret && count < ref->count) { eb = path->nodes[0]; slot = path->slots[0]; @@ -455,13 +467,31 @@ static int add_all_parents(struct btrfs_root *root, struct btrfs_path *path, key.type != BTRFS_EXTENT_DATA_KEY) break; + /* + * We are searching for normal backref but bytenr of this leaf + * matches shared data backref, OR + * the leaf owner is not equal to the root we are searching for + */ + if (slot == 0 && + (is_shared_data_backref(preftrees, eb->start) || + ref->root_id != btrfs_header_owner(eb))) { + if (time_seq == SEQ_LAST) + ret = btrfs_next_leaf(root, path); + else + ret = btrfs_next_old_leaf(root, path, time_seq); + continue; + } fi = btrfs_item_ptr(eb, slot, struct btrfs_file_extent_item); disk_byte = btrfs_file_extent_disk_bytenr(eb, fi); + data_offset = btrfs_file_extent_offset(eb, fi); if (disk_byte == wanted_disk_byte) { eie = NULL; old = NULL; - count++; + if (ref->key_for_search.offset == key.offset - data_offset) + count++; + else + goto next; if (extent_item_pos) { ret = check_extent_in_eb(&key, eb, fi, *extent_item_pos, @@ -502,9 +532,9 @@ next: */ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info, struct btrfs_path *path, u64 time_seq, + struct preftrees *preftrees, struct prelim_ref *ref, struct ulist *parents, - const u64 *extent_item_pos, u64 total_refs, - bool ignore_offset) + const u64 *extent_item_pos, bool ignore_offset) { struct btrfs_root *root; struct btrfs_key root_key; @@ -512,23 +542,25 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info, int ret = 0; int root_level; int level = ref->level; - int index; + struct btrfs_key search_key = ref->key_for_search; root_key.objectid = ref->root_id; root_key.type = BTRFS_ROOT_ITEM_KEY; root_key.offset = (u64)-1; - index = srcu_read_lock(&fs_info->subvol_srcu); - root = btrfs_get_fs_root(fs_info, &root_key, false); if (IS_ERR(root)) { - srcu_read_unlock(&fs_info->subvol_srcu, index); ret = PTR_ERR(root); + goto out_free; + } + + if (!path->search_commit_root && + test_bit(BTRFS_ROOT_DELETING, &root->state)) { + ret = -ENOENT; goto out; } if (btrfs_is_testing(fs_info)) { - srcu_read_unlock(&fs_info->subvol_srcu, index); ret = -ENOENT; goto out; } @@ -540,21 +572,36 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info, else root_level = btrfs_old_root_level(root, time_seq); - if (root_level + 1 == level) { - srcu_read_unlock(&fs_info->subvol_srcu, index); + if (root_level + 1 == level) goto out; - } + /* + * We can often find data backrefs with an offset that is too large + * (>= LLONG_MAX, maximum allowed file offset) due to underflows when + * subtracting a file's offset with the data offset of its + * corresponding extent data item. This can happen for example in the + * clone ioctl. + * + * So if we detect such case we set the search key's offset to zero to + * make sure we will find the matching file extent item at + * add_all_parents(), otherwise we will miss it because the offset + * taken form the backref is much larger then the offset of the file + * extent item. This can make us scan a very large number of file + * extent items, but at least it will not make us miss any. + * + * This is an ugly workaround for a behaviour that should have never + * existed, but it does and a fix for the clone ioctl would touch a lot + * of places, cause backwards incompatibility and would not fix the + * problem for extents cloned with older kernels. + */ + if (search_key.type == BTRFS_EXTENT_DATA_KEY && + search_key.offset >= LLONG_MAX) + search_key.offset = 0; path->lowest_level = level; if (time_seq == SEQ_LAST) - ret = btrfs_search_slot(NULL, root, &ref->key_for_search, path, - 0, 0); + ret = btrfs_search_slot(NULL, root, &search_key, path, 0, 0); else - ret = btrfs_search_old_slot(root, &ref->key_for_search, path, - time_seq); - - /* root node has been locked, we can release @subvol_srcu safely here */ - srcu_read_unlock(&fs_info->subvol_srcu, index); + ret = btrfs_search_old_slot(root, &search_key, path, time_seq); btrfs_debug(fs_info, "search slot in root %llu (level %d, ref count %d) returned %d for key (%llu %u %llu)", @@ -574,9 +621,11 @@ static int resolve_indirect_ref(struct btrfs_fs_info *fs_info, eb = path->nodes[level]; } - ret = add_all_parents(root, path, parents, ref, level, time_seq, - extent_item_pos, total_refs, ignore_offset); + ret = add_all_parents(root, path, parents, preftrees, ref, level, + time_seq, extent_item_pos, ignore_offset); out: + btrfs_put_root(root); +out_free: path->lowest_level = 0; btrfs_release_path(path); return ret; @@ -609,7 +658,7 @@ unode_aux_to_inode_list(struct ulist_node *node) static int resolve_indirect_refs(struct btrfs_fs_info *fs_info, struct btrfs_path *path, u64 time_seq, struct preftrees *preftrees, - const u64 *extent_item_pos, u64 total_refs, + const u64 *extent_item_pos, struct share_check *sc, bool ignore_offset) { int err; @@ -653,9 +702,9 @@ static int resolve_indirect_refs(struct btrfs_fs_info *fs_info, ret = BACKREF_FOUND_SHARED; goto out; } - err = resolve_indirect_ref(fs_info, path, time_seq, ref, - parents, extent_item_pos, - total_refs, ignore_offset); + err = resolve_indirect_ref(fs_info, path, time_seq, preftrees, + ref, parents, extent_item_pos, + ignore_offset); /* * we can only tolerate ENOENT,otherwise,we should catch error * and return directly. @@ -758,8 +807,7 @@ static int add_missing_keys(struct btrfs_fs_info *fs_info, */ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, struct btrfs_delayed_ref_head *head, u64 seq, - struct preftrees *preftrees, u64 *total_refs, - struct share_check *sc) + struct preftrees *preftrees, struct share_check *sc) { struct btrfs_delayed_ref_node *node; struct btrfs_delayed_extent_op *extent_op = head->extent_op; @@ -793,7 +841,6 @@ static int add_delayed_refs(const struct btrfs_fs_info *fs_info, default: BUG(); } - *total_refs += count; switch (node->type) { case BTRFS_TREE_BLOCK_REF_KEY: { /* NORMAL INDIRECT METADATA backref */ @@ -876,7 +923,7 @@ out: static int add_inline_refs(const struct btrfs_fs_info *fs_info, struct btrfs_path *path, u64 bytenr, int *info_level, struct preftrees *preftrees, - u64 *total_refs, struct share_check *sc) + struct share_check *sc) { int ret = 0; int slot; @@ -900,7 +947,6 @@ static int add_inline_refs(const struct btrfs_fs_info *fs_info, ei = btrfs_item_ptr(leaf, slot, struct btrfs_extent_item); flags = btrfs_extent_flags(leaf, ei); - *total_refs += btrfs_extent_refs(leaf, ei); btrfs_item_key_to_cpu(leaf, &found_key, slot); ptr = (unsigned long)(ei + 1); @@ -1125,8 +1171,6 @@ static int find_parent_nodes(struct btrfs_trans_handle *trans, struct prelim_ref *ref; struct rb_node *node; struct extent_inode_elem *eie = NULL; - /* total of both direct AND indirect refs! */ - u64 total_refs = 0; struct preftrees preftrees = { .direct = PREFTREE_INIT, .indirect = PREFTREE_INIT, @@ -1195,7 +1239,7 @@ again: } spin_unlock(&delayed_refs->lock); ret = add_delayed_refs(fs_info, head, time_seq, - &preftrees, &total_refs, sc); + &preftrees, sc); mutex_unlock(&head->mutex); if (ret) goto out; @@ -1216,8 +1260,7 @@ again: (key.type == BTRFS_EXTENT_ITEM_KEY || key.type == BTRFS_METADATA_ITEM_KEY)) { ret = add_inline_refs(fs_info, path, bytenr, - &info_level, &preftrees, - &total_refs, sc); + &info_level, &preftrees, sc); if (ret) goto out; ret = add_keyed_refs(fs_info, path, bytenr, info_level, @@ -1236,7 +1279,7 @@ again: WARN_ON(!RB_EMPTY_ROOT(&preftrees.indirect_missing_keys.root.rb_root)); ret = resolve_indirect_refs(fs_info, path, time_seq, &preftrees, - extent_item_pos, total_refs, sc, ignore_offset); + extent_item_pos, sc, ignore_offset); if (ret) goto out; @@ -1362,10 +1405,10 @@ static void free_leaf_list(struct ulist *blocks) * * returns 0 on success, <0 on error */ -static int btrfs_find_all_leafs(struct btrfs_trans_handle *trans, - struct btrfs_fs_info *fs_info, u64 bytenr, - u64 time_seq, struct ulist **leafs, - const u64 *extent_item_pos, bool ignore_offset) +int btrfs_find_all_leafs(struct btrfs_trans_handle *trans, + struct btrfs_fs_info *fs_info, u64 bytenr, + u64 time_seq, struct ulist **leafs, + const u64 *extent_item_pos, bool ignore_offset) { int ret; diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h index 777f61dc081e..723d6da99114 100644 --- a/fs/btrfs/backref.h +++ b/fs/btrfs/backref.h @@ -40,6 +40,10 @@ int iterate_inodes_from_logical(u64 logical, struct btrfs_fs_info *fs_info, int paths_from_inode(u64 inum, struct inode_fs_paths *ipath); +int btrfs_find_all_leafs(struct btrfs_trans_handle *trans, + struct btrfs_fs_info *fs_info, u64 bytenr, + u64 time_seq, struct ulist **leafs, + const u64 *extent_item_pos, bool ignore_offset); int btrfs_find_all_roots(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, u64 bytenr, u64 time_seq, struct ulist **roots, bool ignore_offset); diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 404e050ce8ee..786849fcc319 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -460,7 +460,7 @@ u64 add_new_free_space(struct btrfs_block_group *block_group, u64 start, u64 end int ret; while (start < end) { - ret = find_first_extent_bit(info->pinned_extents, start, + ret = find_first_extent_bit(&info->excluded_extents, start, &extent_start, &extent_end, EXTENT_DIRTY | EXTENT_UPTODATE, NULL); @@ -856,9 +856,9 @@ static void clear_incompat_bg_bits(struct btrfs_fs_info *fs_info, u64 flags) found_raid1c34 = true; up_read(&sinfo->groups_sem); } - if (found_raid56) + if (!found_raid56) btrfs_clear_fs_incompat(fs_info, RAID56); - if (found_raid1c34) + if (!found_raid1c34) btrfs_clear_fs_incompat(fs_info, RAID1C34); } } @@ -1248,6 +1248,55 @@ out: return ret; } +static bool clean_pinned_extents(struct btrfs_trans_handle *trans, + struct btrfs_block_group *bg) +{ + struct btrfs_fs_info *fs_info = bg->fs_info; + struct btrfs_transaction *prev_trans = NULL; + const u64 start = bg->start; + const u64 end = start + bg->length - 1; + int ret; + + spin_lock(&fs_info->trans_lock); + if (trans->transaction->list.prev != &fs_info->trans_list) { + prev_trans = list_last_entry(&trans->transaction->list, + struct btrfs_transaction, list); + refcount_inc(&prev_trans->use_count); + } + spin_unlock(&fs_info->trans_lock); + + /* + * Hold the unused_bg_unpin_mutex lock to avoid racing with + * btrfs_finish_extent_commit(). If we are at transaction N, another + * task might be running finish_extent_commit() for the previous + * transaction N - 1, and have seen a range belonging to the block + * group in pinned_extents before we were able to clear the whole block + * group range from pinned_extents. This means that task can lookup for + * the block group after we unpinned it from pinned_extents and removed + * it, leading to a BUG_ON() at unpin_extent_range(). + */ + mutex_lock(&fs_info->unused_bg_unpin_mutex); + if (prev_trans) { + ret = clear_extent_bits(&prev_trans->pinned_extents, start, end, + EXTENT_DIRTY); + if (ret) + goto err; + } + + ret = clear_extent_bits(&trans->transaction->pinned_extents, start, end, + EXTENT_DIRTY); + if (ret) + goto err; + mutex_unlock(&fs_info->unused_bg_unpin_mutex); + + return true; + +err: + mutex_unlock(&fs_info->unused_bg_unpin_mutex); + btrfs_dec_block_group_ro(bg); + return false; +} + /* * Process the unused_bgs list and remove any that don't have any allocated * space inside of them. @@ -1265,7 +1314,6 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) spin_lock(&fs_info->unused_bgs_lock); while (!list_empty(&fs_info->unused_bgs)) { - u64 start, end; int trimming; block_group = list_first_entry(&fs_info->unused_bgs, @@ -1344,35 +1392,8 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) * We could have pending pinned extents for this block group, * just delete them, we don't care about them anymore. */ - start = block_group->start; - end = start + block_group->length - 1; - /* - * Hold the unused_bg_unpin_mutex lock to avoid racing with - * btrfs_finish_extent_commit(). If we are at transaction N, - * another task might be running finish_extent_commit() for the - * previous transaction N - 1, and have seen a range belonging - * to the block group in freed_extents[] before we were able to - * clear the whole block group range from freed_extents[]. This - * means that task can lookup for the block group after we - * unpinned it from freed_extents[] and removed it, leading to - * a BUG_ON() at btrfs_unpin_extent_range(). - */ - mutex_lock(&fs_info->unused_bg_unpin_mutex); - ret = clear_extent_bits(&fs_info->freed_extents[0], start, end, - EXTENT_DIRTY); - if (ret) { - mutex_unlock(&fs_info->unused_bg_unpin_mutex); - btrfs_dec_block_group_ro(block_group); + if (!clean_pinned_extents(trans, block_group)) goto end_trans; - } - ret = clear_extent_bits(&fs_info->freed_extents[1], start, end, - EXTENT_DIRTY); - if (ret) { - mutex_unlock(&fs_info->unused_bg_unpin_mutex); - btrfs_dec_block_group_ro(block_group); - goto end_trans; - } - mutex_unlock(&fs_info->unused_bg_unpin_mutex); /* * At this point, the block_group is read only and should fail @@ -1987,6 +2008,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) btrfs_release_path(path); } + rcu_read_lock(); list_for_each_entry_rcu(space_info, &info->space_info, list) { if (!(btrfs_get_alloc_profile(info, space_info->flags) & (BTRFS_BLOCK_GROUP_RAID10 | @@ -2007,6 +2029,7 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) list) inc_block_group_ro(cache, 1); } + rcu_read_unlock(); btrfs_init_global_block_rsv(info); ret = check_chunk_block_group_mappings(info); @@ -2345,7 +2368,7 @@ static int cache_save_setup(struct btrfs_block_group *block_group, return 0; } - if (trans->aborted) + if (TRANS_ABORTED(trans)) return 0; again: inode = lookup_free_space_inode(block_group, path); @@ -2881,7 +2904,7 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans, &cache->space_info->total_bytes_pinned, num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH); - set_extent_dirty(info->pinned_extents, + set_extent_dirty(&trans->transaction->pinned_extents, bytenr, bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL); } diff --git a/fs/btrfs/block-rsv.c b/fs/btrfs/block-rsv.c index d07bd41a7c1e..27efec8f7c5b 100644 --- a/fs/btrfs/block-rsv.c +++ b/fs/btrfs/block-rsv.c @@ -6,6 +6,98 @@ #include "space-info.h" #include "transaction.h" +/* + * HOW DO BLOCK RESERVES WORK + * + * Think of block_rsv's as buckets for logically grouped metadata + * reservations. Each block_rsv has a ->size and a ->reserved. ->size is + * how large we want our block rsv to be, ->reserved is how much space is + * currently reserved for this block reserve. + * + * ->failfast exists for the truncate case, and is described below. + * + * NORMAL OPERATION + * + * -> Reserve + * Entrance: btrfs_block_rsv_add, btrfs_block_rsv_refill + * + * We call into btrfs_reserve_metadata_bytes() with our bytes, which is + * accounted for in space_info->bytes_may_use, and then add the bytes to + * ->reserved, and ->size in the case of btrfs_block_rsv_add. + * + * ->size is an over-estimation of how much we may use for a particular + * operation. + * + * -> Use + * Entrance: btrfs_use_block_rsv + * + * When we do a btrfs_alloc_tree_block() we call into btrfs_use_block_rsv() + * to determine the appropriate block_rsv to use, and then verify that + * ->reserved has enough space for our tree block allocation. Once + * successful we subtract fs_info->nodesize from ->reserved. + * + * -> Finish + * Entrance: btrfs_block_rsv_release + * + * We are finished with our operation, subtract our individual reservation + * from ->size, and then subtract ->size from ->reserved and free up the + * excess if there is any. + * + * There is some logic here to refill the delayed refs rsv or the global rsv + * as needed, otherwise the excess is subtracted from + * space_info->bytes_may_use. + * + * TYPES OF BLOCK RESERVES + * + * BLOCK_RSV_TRANS, BLOCK_RSV_DELOPS, BLOCK_RSV_CHUNK + * These behave normally, as described above, just within the confines of the + * lifetime of their particular operation (transaction for the whole trans + * handle lifetime, for example). + * + * BLOCK_RSV_GLOBAL + * It is impossible to properly account for all the space that may be required + * to make our extent tree updates. This block reserve acts as an overflow + * buffer in case our delayed refs reserve does not reserve enough space to + * update the extent tree. + * + * We can steal from this in some cases as well, notably on evict() or + * truncate() in order to help users recover from ENOSPC conditions. + * + * BLOCK_RSV_DELALLOC + * The individual item sizes are determined by the per-inode size + * calculations, which are described with the delalloc code. This is pretty + * straightforward, it's just the calculation of ->size encodes a lot of + * different items, and thus it gets used when updating inodes, inserting file + * extents, and inserting checksums. + * + * BLOCK_RSV_DELREFS + * We keep a running tally of how many delayed refs we have on the system. + * We assume each one of these delayed refs are going to use a full + * reservation. We use the transaction items and pre-reserve space for every + * operation, and use this reservation to refill any gap between ->size and + * ->reserved that may exist. + * + * From there it's straightforward, removing a delayed ref means we remove its + * count from ->size and free up reservations as necessary. Since this is + * the most dynamic block reserve in the system, we will try to refill this + * block reserve first with any excess returned by any other block reserve. + * + * BLOCK_RSV_EMPTY + * This is the fallback block reserve to make us try to reserve space if we + * don't have a specific bucket for this allocation. It is mostly used for + * updating the device tree and such, since that is a separate pool we're + * content to just reserve space from the space_info on demand. + * + * BLOCK_RSV_TEMP + * This is used by things like truncate and iput. We will temporarily + * allocate a block reserve, set it to some size, and then truncate bytes + * until we have no space left. With ->failfast set we'll simply return + * ENOSPC from btrfs_use_block_rsv() to signal that we need to unwind and try + * to make a new reservation. This is because these operations are + * unbounded, so we want to do as much work as we can, and then back off and + * re-reserve. + */ + static u64 block_rsv_release_bytes(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *block_rsv, struct btrfs_block_rsv *dest, u64 num_bytes, @@ -111,7 +203,7 @@ void btrfs_free_block_rsv(struct btrfs_fs_info *fs_info, { if (!rsv) return; - btrfs_block_rsv_release(fs_info, rsv, (u64)-1); + btrfs_block_rsv_release(fs_info, rsv, (u64)-1, NULL); kfree(rsv); } @@ -178,9 +270,9 @@ int btrfs_block_rsv_refill(struct btrfs_root *root, return ret; } -u64 __btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, - struct btrfs_block_rsv *block_rsv, - u64 num_bytes, u64 *qgroup_to_release) +u64 btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, + struct btrfs_block_rsv *block_rsv, u64 num_bytes, + u64 *qgroup_to_release) { struct btrfs_block_rsv *global_rsv = &fs_info->global_block_rsv; struct btrfs_block_rsv *delayed_rsv = &fs_info->delayed_refs_rsv; @@ -297,9 +389,9 @@ void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info) if (block_rsv->reserved < block_rsv->size) { num_bytes = block_rsv->size - block_rsv->reserved; - block_rsv->reserved += num_bytes; btrfs_space_info_update_bytes_may_use(fs_info, sinfo, num_bytes); + block_rsv->reserved = block_rsv->size; } else if (block_rsv->reserved > block_rsv->size) { num_bytes = block_rsv->reserved - block_rsv->size; btrfs_space_info_update_bytes_may_use(fs_info, sinfo, @@ -344,7 +436,8 @@ void btrfs_init_global_block_rsv(struct btrfs_fs_info *fs_info) void btrfs_release_global_block_rsv(struct btrfs_fs_info *fs_info) { - btrfs_block_rsv_release(fs_info, &fs_info->global_block_rsv, (u64)-1); + btrfs_block_rsv_release(fs_info, &fs_info->global_block_rsv, (u64)-1, + NULL); WARN_ON(fs_info->trans_block_rsv.size > 0); WARN_ON(fs_info->trans_block_rsv.reserved > 0); WARN_ON(fs_info->chunk_block_rsv.size > 0); diff --git a/fs/btrfs/block-rsv.h b/fs/btrfs/block-rsv.h index d1428bb73fc5..0b6ae5302837 100644 --- a/fs/btrfs/block-rsv.h +++ b/fs/btrfs/block-rsv.h @@ -73,7 +73,7 @@ int btrfs_cond_migrate_bytes(struct btrfs_fs_info *fs_info, int min_factor); void btrfs_block_rsv_add_bytes(struct btrfs_block_rsv *block_rsv, u64 num_bytes, bool update_size); -u64 __btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, +u64 btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *block_rsv, u64 num_bytes, u64 *qgroup_to_release); void btrfs_update_global_block_rsv(struct btrfs_fs_info *fs_info); @@ -82,20 +82,12 @@ void btrfs_release_global_block_rsv(struct btrfs_fs_info *fs_info); struct btrfs_block_rsv *btrfs_use_block_rsv(struct btrfs_trans_handle *trans, struct btrfs_root *root, u32 blocksize); - -static inline void btrfs_block_rsv_release(struct btrfs_fs_info *fs_info, - struct btrfs_block_rsv *block_rsv, - u64 num_bytes) -{ - __btrfs_block_rsv_release(fs_info, block_rsv, num_bytes, NULL); -} - static inline void btrfs_unuse_block_rsv(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *block_rsv, u32 blocksize) { btrfs_block_rsv_add_bytes(block_rsv, blocksize, false); - btrfs_block_rsv_release(fs_info, block_rsv, 0); + btrfs_block_rsv_release(fs_info, block_rsv, 0, NULL); } #endif /* BTRFS_BLOCK_RSV_H */ diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index 4e12a477d32e..27a1fefce508 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -60,6 +60,12 @@ struct btrfs_inode { */ struct extent_io_tree io_failure_tree; + /* + * Keep track of where the inode has extent items mapped in order to + * make sure the i_size adjustments are accurate + */ + struct extent_io_tree file_extent_tree; + /* held while logging the inode in tree-log.c */ struct mutex log_mutex; diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index a0ce69f2d27c..32e11a23b47f 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -77,7 +77,6 @@ #include <linux/sched.h> #include <linux/slab.h> -#include <linux/buffer_head.h> #include <linux/mutex.h> #include <linux/genhd.h> #include <linux/blkdev.h> @@ -152,11 +151,8 @@ struct btrfsic_block { struct list_head ref_to_list; /* list */ struct list_head ref_from_list; /* list */ struct btrfsic_block *next_in_same_bio; - void *orig_bio_bh_private; - union { - bio_end_io_t *bio; - bh_end_io_t *bh; - } orig_bio_bh_end_io; + void *orig_bio_private; + bio_end_io_t *orig_bio_end_io; int submit_bio_bh_rw; u64 flush_gen; /* only valid if !never_written */ }; @@ -325,14 +321,12 @@ static void btrfsic_process_written_block(struct btrfsic_dev_state *dev_state, u64 dev_bytenr, char **mapped_datav, unsigned int num_pages, struct bio *bio, int *bio_is_patched, - struct buffer_head *bh, int submit_bio_bh_rw); static int btrfsic_process_written_superblock( struct btrfsic_state *state, struct btrfsic_block *const block, struct btrfs_super_block *const super_hdr); static void btrfsic_bio_end_io(struct bio *bp); -static void btrfsic_bh_end_io(struct buffer_head *bh, int uptodate); static int btrfsic_is_block_ref_by_superblock(const struct btrfsic_state *state, const struct btrfsic_block *block, int recursion_level); @@ -399,8 +393,8 @@ static void btrfsic_block_init(struct btrfsic_block *b) b->never_written = 0; b->mirror_num = 0; b->next_in_same_bio = NULL; - b->orig_bio_bh_private = NULL; - b->orig_bio_bh_end_io.bio = NULL; + b->orig_bio_private = NULL; + b->orig_bio_end_io = NULL; INIT_LIST_HEAD(&b->collision_resolving_node); INIT_LIST_HEAD(&b->all_blocks_node); INIT_LIST_HEAD(&b->ref_to_list); @@ -767,29 +761,31 @@ static int btrfsic_process_superblock_dev_mirror( struct btrfs_fs_info *fs_info = state->fs_info; struct btrfs_super_block *super_tmp; u64 dev_bytenr; - struct buffer_head *bh; struct btrfsic_block *superblock_tmp; int pass; struct block_device *const superblock_bdev = device->bdev; + struct page *page; + struct address_space *mapping = superblock_bdev->bd_inode->i_mapping; + int ret = 0; /* super block bytenr is always the unmapped device bytenr */ dev_bytenr = btrfs_sb_offset(superblock_mirror_num); if (dev_bytenr + BTRFS_SUPER_INFO_SIZE > device->commit_total_bytes) return -1; - bh = __bread(superblock_bdev, dev_bytenr / BTRFS_BDEV_BLOCKSIZE, - BTRFS_SUPER_INFO_SIZE); - if (NULL == bh) + + page = read_cache_page_gfp(mapping, dev_bytenr >> PAGE_SHIFT, GFP_NOFS); + if (IS_ERR(page)) return -1; - super_tmp = (struct btrfs_super_block *) - (bh->b_data + (dev_bytenr & (BTRFS_BDEV_BLOCKSIZE - 1))); + + super_tmp = page_address(page); if (btrfs_super_bytenr(super_tmp) != dev_bytenr || btrfs_super_magic(super_tmp) != BTRFS_MAGIC || memcmp(device->uuid, super_tmp->dev_item.uuid, BTRFS_UUID_SIZE) || btrfs_super_nodesize(super_tmp) != state->metablock_size || btrfs_super_sectorsize(super_tmp) != state->datablock_size) { - brelse(bh); - return 0; + ret = 0; + goto out; } superblock_tmp = @@ -800,8 +796,8 @@ static int btrfsic_process_superblock_dev_mirror( superblock_tmp = btrfsic_block_alloc(); if (NULL == superblock_tmp) { pr_info("btrfsic: error, kmalloc failed!\n"); - brelse(bh); - return -1; + ret = -1; + goto out; } /* for superblock, only the dev_bytenr makes sense */ superblock_tmp->dev_bytenr = dev_bytenr; @@ -885,8 +881,8 @@ static int btrfsic_process_superblock_dev_mirror( mirror_num)) { pr_info("btrfsic: btrfsic_map_block(bytenr @%llu, mirror %d) failed!\n", next_bytenr, mirror_num); - brelse(bh); - return -1; + ret = -1; + goto out; } next_block = btrfsic_block_lookup_or_add( @@ -895,8 +891,8 @@ static int btrfsic_process_superblock_dev_mirror( mirror_num, NULL); if (NULL == next_block) { btrfsic_release_block_ctx(&tmp_next_block_ctx); - brelse(bh); - return -1; + ret = -1; + goto out; } next_block->disk_key = tmp_disk_key; @@ -907,16 +903,17 @@ static int btrfsic_process_superblock_dev_mirror( BTRFSIC_GENERATION_UNKNOWN); btrfsic_release_block_ctx(&tmp_next_block_ctx); if (NULL == l) { - brelse(bh); - return -1; + ret = -1; + goto out; } } } if (state->print_mask & BTRFSIC_PRINT_MASK_INITIAL_ALL_TREES) btrfsic_dump_tree_sub(state, superblock_tmp, 0); - brelse(bh); - return 0; +out: + put_page(page); + return ret; } static struct btrfsic_stack_frame *btrfsic_stack_frame_alloc(void) @@ -1743,7 +1740,6 @@ static void btrfsic_process_written_block(struct btrfsic_dev_state *dev_state, u64 dev_bytenr, char **mapped_datav, unsigned int num_pages, struct bio *bio, int *bio_is_patched, - struct buffer_head *bh, int submit_bio_bh_rw) { int is_metadata; @@ -1902,9 +1898,9 @@ again: block->is_iodone = 0; BUG_ON(NULL == bio_is_patched); if (!*bio_is_patched) { - block->orig_bio_bh_private = + block->orig_bio_private = bio->bi_private; - block->orig_bio_bh_end_io.bio = + block->orig_bio_end_io = bio->bi_end_io; block->next_in_same_bio = NULL; bio->bi_private = block; @@ -1916,25 +1912,17 @@ again: bio->bi_private; BUG_ON(NULL == chained_block); - block->orig_bio_bh_private = - chained_block->orig_bio_bh_private; - block->orig_bio_bh_end_io.bio = - chained_block->orig_bio_bh_end_io. - bio; + block->orig_bio_private = + chained_block->orig_bio_private; + block->orig_bio_end_io = + chained_block->orig_bio_end_io; block->next_in_same_bio = chained_block; bio->bi_private = block; } - } else if (NULL != bh) { - block->is_iodone = 0; - block->orig_bio_bh_private = bh->b_private; - block->orig_bio_bh_end_io.bh = bh->b_end_io; - block->next_in_same_bio = NULL; - bh->b_private = block; - bh->b_end_io = btrfsic_bh_end_io; } else { block->is_iodone = 1; - block->orig_bio_bh_private = NULL; - block->orig_bio_bh_end_io.bio = NULL; + block->orig_bio_private = NULL; + block->orig_bio_end_io = NULL; block->next_in_same_bio = NULL; } } @@ -2042,8 +2030,8 @@ again: block->is_iodone = 0; BUG_ON(NULL == bio_is_patched); if (!*bio_is_patched) { - block->orig_bio_bh_private = bio->bi_private; - block->orig_bio_bh_end_io.bio = bio->bi_end_io; + block->orig_bio_private = bio->bi_private; + block->orig_bio_end_io = bio->bi_end_io; block->next_in_same_bio = NULL; bio->bi_private = block; bio->bi_end_io = btrfsic_bio_end_io; @@ -2054,24 +2042,17 @@ again: bio->bi_private; BUG_ON(NULL == chained_block); - block->orig_bio_bh_private = - chained_block->orig_bio_bh_private; - block->orig_bio_bh_end_io.bio = - chained_block->orig_bio_bh_end_io.bio; + block->orig_bio_private = + chained_block->orig_bio_private; + block->orig_bio_end_io = + chained_block->orig_bio_end_io; block->next_in_same_bio = chained_block; bio->bi_private = block; } - } else if (NULL != bh) { - block->is_iodone = 0; - block->orig_bio_bh_private = bh->b_private; - block->orig_bio_bh_end_io.bh = bh->b_end_io; - block->next_in_same_bio = NULL; - bh->b_private = block; - bh->b_end_io = btrfsic_bh_end_io; } else { block->is_iodone = 1; - block->orig_bio_bh_private = NULL; - block->orig_bio_bh_end_io.bio = NULL; + block->orig_bio_private = NULL; + block->orig_bio_end_io = NULL; block->next_in_same_bio = NULL; } if (state->print_mask & BTRFSIC_PRINT_MASK_VERBOSE) @@ -2112,8 +2093,8 @@ static void btrfsic_bio_end_io(struct bio *bp) iodone_w_error = 1; BUG_ON(NULL == block); - bp->bi_private = block->orig_bio_bh_private; - bp->bi_end_io = block->orig_bio_bh_end_io.bio; + bp->bi_private = block->orig_bio_private; + bp->bi_end_io = block->orig_bio_end_io; do { struct btrfsic_block *next_block; @@ -2146,38 +2127,6 @@ static void btrfsic_bio_end_io(struct bio *bp) bp->bi_end_io(bp); } -static void btrfsic_bh_end_io(struct buffer_head *bh, int uptodate) -{ - struct btrfsic_block *block = (struct btrfsic_block *)bh->b_private; - int iodone_w_error = !uptodate; - struct btrfsic_dev_state *dev_state; - - BUG_ON(NULL == block); - dev_state = block->dev_state; - if ((dev_state->state->print_mask & BTRFSIC_PRINT_MASK_END_IO_BIO_BH)) - pr_info("bh_end_io(error=%d) for %c @%llu (%s/%llu/%d)\n", - iodone_w_error, - btrfsic_get_block_type(dev_state->state, block), - block->logical_bytenr, block->dev_state->name, - block->dev_bytenr, block->mirror_num); - - block->iodone_w_error = iodone_w_error; - if (block->submit_bio_bh_rw & REQ_PREFLUSH) { - dev_state->last_flush_gen++; - if ((dev_state->state->print_mask & - BTRFSIC_PRINT_MASK_END_IO_BIO_BH)) - pr_info("bh_end_io() new %s flush_gen=%llu\n", - dev_state->name, dev_state->last_flush_gen); - } - if (block->submit_bio_bh_rw & REQ_FUA) - block->flush_gen = 0; /* FUA completed means block is on disk */ - - bh->b_private = block->orig_bio_bh_private; - bh->b_end_io = block->orig_bio_bh_end_io.bh; - block->is_iodone = 1; /* for FLUSH, this releases the block */ - bh->b_end_io(bh, uptodate); -} - static int btrfsic_process_written_superblock( struct btrfsic_state *state, struct btrfsic_block *const superblock, @@ -2730,63 +2679,6 @@ static struct btrfsic_dev_state *btrfsic_dev_state_lookup(dev_t dev) &btrfsic_dev_state_hashtable); } -int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh) -{ - struct btrfsic_dev_state *dev_state; - - if (!btrfsic_is_initialized) - return submit_bh(op, op_flags, bh); - - mutex_lock(&btrfsic_mutex); - /* since btrfsic_submit_bh() might also be called before - * btrfsic_mount(), this might return NULL */ - dev_state = btrfsic_dev_state_lookup(bh->b_bdev->bd_dev); - - /* Only called to write the superblock (incl. FLUSH/FUA) */ - if (NULL != dev_state && - (op == REQ_OP_WRITE) && bh->b_size > 0) { - u64 dev_bytenr; - - dev_bytenr = BTRFS_BDEV_BLOCKSIZE * bh->b_blocknr; - if (dev_state->state->print_mask & - BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH) - pr_info("submit_bh(op=0x%x,0x%x, blocknr=%llu (bytenr %llu), size=%zu, data=%p, bdev=%p)\n", - op, op_flags, (unsigned long long)bh->b_blocknr, - dev_bytenr, bh->b_size, bh->b_data, bh->b_bdev); - btrfsic_process_written_block(dev_state, dev_bytenr, - &bh->b_data, 1, NULL, - NULL, bh, op_flags); - } else if (NULL != dev_state && (op_flags & REQ_PREFLUSH)) { - if (dev_state->state->print_mask & - BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH) - pr_info("submit_bh(op=0x%x,0x%x FLUSH, bdev=%p)\n", - op, op_flags, bh->b_bdev); - if (!dev_state->dummy_block_for_bio_bh_flush.is_iodone) { - if ((dev_state->state->print_mask & - (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH | - BTRFSIC_PRINT_MASK_VERBOSE))) - pr_info("btrfsic_submit_bh(%s) with FLUSH but dummy block already in use (ignored)!\n", - dev_state->name); - } else { - struct btrfsic_block *const block = - &dev_state->dummy_block_for_bio_bh_flush; - - block->is_iodone = 0; - block->never_written = 0; - block->iodone_w_error = 0; - block->flush_gen = dev_state->last_flush_gen + 1; - block->submit_bio_bh_rw = op_flags; - block->orig_bio_bh_private = bh->b_private; - block->orig_bio_bh_end_io.bh = bh->b_end_io; - block->next_in_same_bio = NULL; - bh->b_private = block; - bh->b_end_io = btrfsic_bh_end_io; - } - } - mutex_unlock(&btrfsic_mutex); - return submit_bh(op, op_flags, bh); -} - static void __btrfsic_submit_bio(struct bio *bio) { struct btrfsic_dev_state *dev_state; @@ -2838,7 +2730,7 @@ static void __btrfsic_submit_bio(struct bio *bio) btrfsic_process_written_block(dev_state, dev_bytenr, mapped_datav, segs, bio, &bio_is_patched, - NULL, bio->bi_opf); + bio->bi_opf); bio_for_each_segment(bvec, bio, iter) kunmap(bvec.bv_page); kfree(mapped_datav); @@ -2862,8 +2754,8 @@ static void __btrfsic_submit_bio(struct bio *bio) block->iodone_w_error = 0; block->flush_gen = dev_state->last_flush_gen + 1; block->submit_bio_bh_rw = bio->bi_opf; - block->orig_bio_bh_private = bio->bi_private; - block->orig_bio_bh_end_io.bio = bio->bi_end_io; + block->orig_bio_private = bio->bi_private; + block->orig_bio_end_io = bio->bi_end_io; block->next_in_same_bio = NULL; bio->bi_private = block; bio->bi_end_io = btrfsic_bio_end_io; diff --git a/fs/btrfs/check-integrity.h b/fs/btrfs/check-integrity.h index 9bf4359cc44c..bcc730a06cb5 100644 --- a/fs/btrfs/check-integrity.h +++ b/fs/btrfs/check-integrity.h @@ -7,11 +7,9 @@ #define BTRFS_CHECK_INTEGRITY_H #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY -int btrfsic_submit_bh(int op, int op_flags, struct buffer_head *bh); void btrfsic_submit_bio(struct bio *bio); int btrfsic_submit_bio_wait(struct bio *bio); #else -#define btrfsic_submit_bh submit_bh #define btrfsic_submit_bio submit_bio #define btrfsic_submit_bio_wait submit_bio_wait #endif diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index f2ec1a9bae28..bfedbbe2311f 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -31,8 +31,8 @@ static void del_ptr(struct btrfs_root *root, struct btrfs_path *path, static const struct btrfs_csums { u16 size; - const char *name; - const char *driver; + const char name[10]; + const char driver[12]; } btrfs_csums[] = { [BTRFS_CSUM_TYPE_CRC32] = { .size = 4, .name = "crc32c" }, [BTRFS_CSUM_TYPE_XXHASH] = { .size = 8, .name = "xxhash64" }, @@ -63,7 +63,8 @@ const char *btrfs_super_csum_name(u16 csum_type) const char *btrfs_super_csum_driver(u16 csum_type) { /* csum type is validated at mount time */ - return btrfs_csums[csum_type].driver ?: + return btrfs_csums[csum_type].driver[0] ? + btrfs_csums[csum_type].driver : btrfs_csums[csum_type].name; } @@ -143,44 +144,6 @@ struct extent_buffer *btrfs_root_node(struct btrfs_root *root) return eb; } -/* loop around taking references on and locking the root node of the - * tree until you end up with a lock on the root. A locked buffer - * is returned, with a reference held. - */ -struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root) -{ - struct extent_buffer *eb; - - while (1) { - eb = btrfs_root_node(root); - btrfs_tree_lock(eb); - if (eb == root->node) - break; - btrfs_tree_unlock(eb); - free_extent_buffer(eb); - } - return eb; -} - -/* loop around taking references on and locking the root node of the - * tree until you end up with a lock on the root. A locked buffer - * is returned, with a reference held. - */ -struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root) -{ - struct extent_buffer *eb; - - while (1) { - eb = btrfs_root_node(root); - btrfs_tree_read_lock(eb); - if (eb == root->node) - break; - btrfs_tree_read_unlock(eb); - free_extent_buffer(eb); - } - return eb; -} - /* cowonly root (everything not a reference counted cow subvolume), just get * put onto a simple dirty list. transaction.c walks this to make sure they * get properly updated on disk. @@ -341,7 +304,6 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info, struct rb_root *tm_root; struct rb_node *node; struct rb_node *next; - struct seq_list *cur_elem; struct tree_mod_elem *tm; u64 min_seq = (u64)-1; u64 seq_putting = elem->seq; @@ -353,18 +315,20 @@ void btrfs_put_tree_mod_seq(struct btrfs_fs_info *fs_info, list_del(&elem->list); elem->seq = 0; - list_for_each_entry(cur_elem, &fs_info->tree_mod_seq_list, list) { - if (cur_elem->seq < min_seq) { - if (seq_putting > cur_elem->seq) { - /* - * blocker with lower sequence number exists, we - * cannot remove anything from the log - */ - write_unlock(&fs_info->tree_mod_log_lock); - return; - } - min_seq = cur_elem->seq; + if (!list_empty(&fs_info->tree_mod_seq_list)) { + struct seq_list *first; + + first = list_first_entry(&fs_info->tree_mod_seq_list, + struct seq_list, list); + if (seq_putting > first->seq) { + /* + * Blocker with lower sequence number exists, we + * cannot remove anything from the log. + */ + write_unlock(&fs_info->tree_mod_log_lock); + return; } + min_seq = first->seq; } /* @@ -962,9 +926,7 @@ static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans, if (new_flags != 0) { int level = btrfs_header_level(buf); - ret = btrfs_set_disk_extent_flags(trans, - buf->start, - buf->len, + ret = btrfs_set_disk_extent_flags(trans, buf, new_flags, level, 0); if (ret) return ret; diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 36df977b64d9..8aa7b9dac405 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -33,6 +33,7 @@ #include "extent_map.h" #include "async-thread.h" #include "block-rsv.h" +#include "locking.h" struct btrfs_trans_handle; struct btrfs_transaction; @@ -596,8 +597,8 @@ struct btrfs_fs_info { /* keep track of unallocated space */ atomic64_t free_chunk_space; - struct extent_io_tree freed_extents[2]; - struct extent_io_tree *pinned_extents; + /* Track ranges which are used by log trees blocks/logged data extents */ + struct extent_io_tree excluded_extents; /* logical->physical extent mapping */ struct extent_map_tree mapping_tree; @@ -696,7 +697,6 @@ struct btrfs_fs_info { struct rw_semaphore cleanup_work_sem; struct rw_semaphore subvol_sem; - struct srcu_struct subvol_srcu; spinlock_t trans_lock; /* @@ -947,6 +947,10 @@ struct btrfs_fs_info { #ifdef CONFIG_BTRFS_DEBUG struct kobject *debug_kobj; struct kobject *discard_debug_kobj; + struct list_head allocated_roots; + + spinlock_t eb_leak_lock; + struct list_head allocated_ebs; #endif }; @@ -955,11 +959,6 @@ static inline struct btrfs_fs_info *btrfs_sb(struct super_block *sb) return sb->s_fs_info; } -struct btrfs_subvolume_writers { - struct percpu_counter counter; - wait_queue_head_t wait; -}; - /* * The state of btrfs root */ @@ -1131,8 +1130,9 @@ struct btrfs_root { * root_item_lock. */ int dedupe_in_progress; - struct btrfs_subvolume_writers *subv_writers; - atomic_t will_be_snapshotted; + /* For exclusion of snapshot creation and nocow writes */ + struct btrfs_drew_lock snapshot_lock; + atomic_t snapshot_force_cow; /* For qgroup metadata reserved space */ @@ -1149,6 +1149,10 @@ struct btrfs_root { #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS u64 alloc_bytenr; #endif + +#ifdef CONFIG_BTRFS_DEBUG + struct list_head leak_list; +#endif }; struct btrfs_clone_extent_info { @@ -1971,16 +1975,6 @@ static inline void btrfs_set_header_backref_rev(struct extent_buffer *eb, btrfs_set_header_flags(eb, flags); } -static inline unsigned long btrfs_header_fsid(void) -{ - return offsetof(struct btrfs_header, fsid); -} - -static inline unsigned long btrfs_header_chunk_tree_uuid(const struct extent_buffer *eb) -{ - return offsetof(struct btrfs_header, chunk_tree_uuid); -} - static inline int btrfs_is_leaf(const struct extent_buffer *eb) { return btrfs_header_level(eb) == 0; @@ -2458,9 +2452,9 @@ int btrfs_lookup_data_extent(struct btrfs_fs_info *fs_info, u64 start, u64 len); int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info, u64 bytenr, u64 offset, int metadata, u64 *refs, u64 *flags); -int btrfs_pin_extent(struct btrfs_fs_info *fs_info, - u64 bytenr, u64 num, int reserved); -int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info, +int btrfs_pin_extent(struct btrfs_trans_handle *trans, u64 bytenr, u64 num, + int reserved); +int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans, u64 bytenr, u64 num_bytes); int btrfs_exclude_logged_extents(struct extent_buffer *eb); int btrfs_cross_ref_exist(struct btrfs_root *root, @@ -2490,13 +2484,13 @@ int btrfs_inc_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root, int btrfs_dec_ref(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *buf, int full_backref); int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans, - u64 bytenr, u64 num_bytes, u64 flags, + struct extent_buffer *eb, u64 flags, int level, int is_data); int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref); int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info, u64 start, u64 len, int delalloc); -int btrfs_pin_reserved_extent(struct btrfs_fs_info *fs_info, u64 start, +int btrfs_pin_reserved_extent(struct btrfs_trans_handle *trans, u64 start, u64 len); void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info); int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans); @@ -2665,9 +2659,8 @@ static inline int btrfs_next_item(struct btrfs_root *root, struct btrfs_path *p) return btrfs_next_old_item(root, p, 0); } int btrfs_leaf_free_space(struct extent_buffer *leaf); -int __must_check btrfs_drop_snapshot(struct btrfs_root *root, - struct btrfs_block_rsv *block_rsv, - int update_ref, int for_reloc); +int __must_check btrfs_drop_snapshot(struct btrfs_root *root, int update_ref, + int for_reloc); int btrfs_drop_subtree(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *node, @@ -2695,23 +2688,6 @@ static inline int btrfs_need_cleaner_sleep(struct btrfs_fs_info *fs_info) return fs_info->sb->s_flags & SB_RDONLY || btrfs_fs_closing(fs_info); } -static inline void free_fs_info(struct btrfs_fs_info *fs_info) -{ - kfree(fs_info->balance_ctl); - kfree(fs_info->delayed_root); - kfree(fs_info->extent_root); - kfree(fs_info->tree_root); - kfree(fs_info->chunk_root); - kfree(fs_info->dev_root); - kfree(fs_info->csum_root); - kfree(fs_info->quota_root); - kfree(fs_info->uuid_root); - kfree(fs_info->free_space_root); - kfree(fs_info->super_copy); - kfree(fs_info->super_for_commit); - kvfree(fs_info); -} - /* tree mod log functions from ctree.c */ u64 btrfs_get_tree_mod_seq(struct btrfs_fs_info *fs_info, struct seq_list *elem); @@ -2750,9 +2726,7 @@ int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type, u64 subid); int btrfs_uuid_tree_remove(struct btrfs_trans_handle *trans, u8 *uuid, u8 type, u64 subid); -int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info, - int (*check_func)(struct btrfs_fs_info *, u8 *, u8, - u64)); +int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info); /* dir-item.c */ int btrfs_check_dir_item_collision(struct btrfs_root *root, u64 dir, @@ -2859,6 +2833,12 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode, struct btrfs_file_extent_item *fi, const bool new_inline, struct extent_map *em); +int btrfs_inode_clear_file_extent_range(struct btrfs_inode *inode, u64 start, + u64 len); +int btrfs_inode_set_file_extent_range(struct btrfs_inode *inode, u64 start, + u64 len); +void btrfs_inode_safe_disk_i_size_write(struct inode *inode, u64 new_i_size); +u64 btrfs_file_extent_end(const struct btrfs_path *path); /* inode.c */ struct extent_map *btrfs_get_extent_fiemap(struct btrfs_inode *inode, @@ -2996,9 +2976,6 @@ int btrfs_dirty_pages(struct inode *inode, struct page **pages, size_t num_pages, loff_t pos, size_t write_bytes, struct extent_state **cached); int btrfs_fdatawrite_range(struct inode *inode, loff_t start, loff_t end); -loff_t btrfs_remap_file_range(struct file *file_in, loff_t pos_in, - struct file *file_out, loff_t pos_out, - loff_t len, unsigned int remap_flags); /* tree-defrag.c */ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans, @@ -3008,6 +2985,8 @@ int btrfs_defrag_leaves(struct btrfs_trans_handle *trans, int btrfs_parse_options(struct btrfs_fs_info *info, char *options, unsigned long new_flags); int btrfs_sync_fs(struct super_block *sb, int wait); +char *btrfs_get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info, + u64 subvol_objectid); static inline __printf(2, 3) __cold void btrfs_no_printk(const struct btrfs_fs_info *fs_info, const char *fmt, ...) @@ -3401,6 +3380,7 @@ void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending, u64 *bytes_to_reserve); int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans, struct btrfs_pending_snapshot *pending); +int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info); /* scrub.c */ int btrfs_scrub_dev(struct btrfs_fs_info *fs_info, u64 devid, u64 start, diff --git a/fs/btrfs/delalloc-space.c b/fs/btrfs/delalloc-space.c index 4cdac4d834f5..1245739a3a6e 100644 --- a/fs/btrfs/delalloc-space.c +++ b/fs/btrfs/delalloc-space.c @@ -9,6 +9,108 @@ #include "qgroup.h" #include "block-group.h" +/* + * HOW DOES THIS WORK + * + * There are two stages to data reservations, one for data and one for metadata + * to handle the new extents and checksums generated by writing data. + * + * + * DATA RESERVATION + * The general flow of the data reservation is as follows + * + * -> Reserve + * We call into btrfs_reserve_data_bytes() for the user request bytes that + * they wish to write. We make this reservation and add it to + * space_info->bytes_may_use. We set EXTENT_DELALLOC on the inode io_tree + * for the range and carry on if this is buffered, or follow up trying to + * make a real allocation if we are pre-allocating or doing O_DIRECT. + * + * -> Use + * At writepages()/prealloc/O_DIRECT time we will call into + * btrfs_reserve_extent() for some part or all of this range of bytes. We + * will make the allocation and subtract space_info->bytes_may_use by the + * original requested length and increase the space_info->bytes_reserved by + * the allocated length. This distinction is important because compression + * may allocate a smaller on disk extent than we previously reserved. + * + * -> Allocation + * finish_ordered_io() will insert the new file extent item for this range, + * and then add a delayed ref update for the extent tree. Once that delayed + * ref is written the extent size is subtracted from + * space_info->bytes_reserved and added to space_info->bytes_used. + * + * Error handling + * + * -> By the reservation maker + * This is the simplest case, we haven't completed our operation and we know + * how much we reserved, we can simply call + * btrfs_free_reserved_data_space*() and it will be removed from + * space_info->bytes_may_use. + * + * -> After the reservation has been made, but before cow_file_range() + * This is specifically for the delalloc case. You must clear + * EXTENT_DELALLOC with the EXTENT_CLEAR_DATA_RESV bit, and the range will + * be subtracted from space_info->bytes_may_use. + * + * METADATA RESERVATION + * The general metadata reservation lifetimes are discussed elsewhere, this + * will just focus on how it is used for delalloc space. + * + * We keep track of two things on a per inode bases + * + * ->outstanding_extents + * This is the number of file extent items we'll need to handle all of the + * outstanding DELALLOC space we have in this inode. We limit the maximum + * size of an extent, so a large contiguous dirty area may require more than + * one outstanding_extent, which is why count_max_extents() is used to + * determine how many outstanding_extents get added. + * + * ->csum_bytes + * This is essentially how many dirty bytes we have for this inode, so we + * can calculate the number of checksum items we would have to add in order + * to checksum our outstanding data. + * + * We keep a per-inode block_rsv in order to make it easier to keep track of + * our reservation. We use btrfs_calculate_inode_block_rsv_size() to + * calculate the current theoretical maximum reservation we would need for the + * metadata for this inode. We call this and then adjust our reservation as + * necessary, either by attempting to reserve more space, or freeing up excess + * space. + * + * OUTSTANDING_EXTENTS HANDLING + * + * ->outstanding_extents is used for keeping track of how many extents we will + * need to use for this inode, and it will fluctuate depending on where you are + * in the life cycle of the dirty data. Consider the following normal case for + * a completely clean inode, with a num_bytes < our maximum allowed extent size + * + * -> reserve + * ->outstanding_extents += 1 (current value is 1) + * + * -> set_delalloc + * ->outstanding_extents += 1 (currrent value is 2) + * + * -> btrfs_delalloc_release_extents() + * ->outstanding_extents -= 1 (current value is 1) + * + * We must call this once we are done, as we hold our reservation for the + * duration of our operation, and then assume set_delalloc will update the + * counter appropriately. + * + * -> add ordered extent + * ->outstanding_extents += 1 (current value is 2) + * + * -> btrfs_clear_delalloc_extent + * ->outstanding_extents -= 1 (current value is 1) + * + * -> finish_ordered_io/btrfs_remove_ordered_extent + * ->outstanding_extents -= 1 (current value is 0) + * + * Each stage is responsible for their own accounting of the extent, thus + * making error handling and cleanup easier. + */ + int btrfs_alloc_data_chunk_ondemand(struct btrfs_inode *inode, u64 bytes) { struct btrfs_root *root = inode->root; @@ -228,8 +330,8 @@ static void btrfs_inode_rsv_release(struct btrfs_inode *inode, bool qgroup_free) * are releasing 0 bytes, and then we'll just get the reservation over * the size free'd. */ - released = __btrfs_block_rsv_release(fs_info, block_rsv, 0, - &qgroup_to_release); + released = btrfs_block_rsv_release(fs_info, block_rsv, 0, + &qgroup_to_release); if (released > 0) trace_btrfs_space_reservation(fs_info, "delalloc", btrfs_ino(inode), released, 0); diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index d3e15e1d4a91..bf1595a42a98 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -6,6 +6,7 @@ #include <linux/slab.h> #include <linux/iversion.h> +#include <linux/sched/mm.h> #include "misc.h" #include "delayed-inode.h" #include "disk-io.h" @@ -595,8 +596,7 @@ static void btrfs_delayed_item_release_metadata(struct btrfs_root *root, trace_btrfs_space_reservation(fs_info, "delayed_item", item->key.objectid, item->bytes_reserved, 0); - btrfs_block_rsv_release(fs_info, rsv, - item->bytes_reserved); + btrfs_block_rsv_release(fs_info, rsv, item->bytes_reserved, NULL); } static int btrfs_delayed_inode_reserve_metadata( @@ -677,8 +677,7 @@ static void btrfs_delayed_inode_release_metadata(struct btrfs_fs_info *fs_info, rsv = &fs_info->delayed_block_rsv; trace_btrfs_space_reservation(fs_info, "delayed_inode", node->inode_id, node->bytes_reserved, 0); - btrfs_block_rsv_release(fs_info, rsv, - node->bytes_reserved); + btrfs_block_rsv_release(fs_info, rsv, node->bytes_reserved, NULL); if (qgroup_free) btrfs_qgroup_free_meta_prealloc(node->root, node->bytes_reserved); @@ -805,11 +804,14 @@ static int btrfs_insert_delayed_item(struct btrfs_trans_handle *trans, struct btrfs_delayed_item *delayed_item) { struct extent_buffer *leaf; + unsigned int nofs_flag; char *ptr; int ret; + nofs_flag = memalloc_nofs_save(); ret = btrfs_insert_empty_item(trans, root, path, &delayed_item->key, delayed_item->data_len); + memalloc_nofs_restore(nofs_flag); if (ret < 0 && ret != -EEXIST) return ret; @@ -937,6 +939,7 @@ static int btrfs_delete_delayed_items(struct btrfs_trans_handle *trans, struct btrfs_delayed_node *node) { struct btrfs_delayed_item *curr, *prev; + unsigned int nofs_flag; int ret = 0; do_again: @@ -945,7 +948,9 @@ do_again: if (!curr) goto delete_fail; + nofs_flag = memalloc_nofs_save(); ret = btrfs_search_slot(trans, root, &curr->key, path, -1, 1); + memalloc_nofs_restore(nofs_flag); if (ret < 0) goto delete_fail; else if (ret > 0) { @@ -1012,6 +1017,7 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans, struct btrfs_key key; struct btrfs_inode_item *inode_item; struct extent_buffer *leaf; + unsigned int nofs_flag; int mod; int ret; @@ -1024,7 +1030,9 @@ static int __btrfs_update_delayed_inode(struct btrfs_trans_handle *trans, else mod = 1; + nofs_flag = memalloc_nofs_save(); ret = btrfs_lookup_inode(trans, root, path, &key, mod); + memalloc_nofs_restore(nofs_flag); if (ret > 0) { btrfs_release_path(path); return -ENOENT; @@ -1075,7 +1083,10 @@ search: key.type = BTRFS_INODE_EXTREF_KEY; key.offset = -1; + + nofs_flag = memalloc_nofs_save(); ret = btrfs_search_slot(trans, root, &key, path, -1, 1); + memalloc_nofs_restore(nofs_flag); if (ret < 0) goto err_out; ASSERT(ret); @@ -1139,7 +1150,7 @@ static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans, int nr) int ret = 0; bool count = (nr > 0); - if (trans->aborted) + if (TRANS_ABORTED(trans)) return -EIO; path = btrfs_alloc_path(); @@ -1760,6 +1771,7 @@ static void fill_stack_inode_item(struct btrfs_trans_handle *trans, int btrfs_fill_inode(struct inode *inode, u32 *rdev) { + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; struct btrfs_delayed_node *delayed_node; struct btrfs_inode_item *inode_item; @@ -1779,6 +1791,8 @@ int btrfs_fill_inode(struct inode *inode, u32 *rdev) i_uid_write(inode, btrfs_stack_inode_uid(inode_item)); i_gid_write(inode, btrfs_stack_inode_gid(inode_item)); btrfs_i_size_write(BTRFS_I(inode), btrfs_stack_inode_size(inode_item)); + btrfs_inode_set_file_extent_range(BTRFS_I(inode), 0, + round_up(i_size_read(inode), fs_info->sectorsize)); inode->i_mode = btrfs_stack_inode_mode(inode_item); set_nlink(inode, btrfs_stack_inode_nlink(inode_item)); inode_set_bytes(inode, btrfs_stack_inode_nbytes(inode_item)); diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h index 74ae226ffaf0..ca96ef007d8f 100644 --- a/fs/btrfs/delayed-inode.h +++ b/fs/btrfs/delayed-inode.h @@ -70,7 +70,7 @@ struct btrfs_delayed_item { refcount_t refs; int ins_or_del; u32 data_len; - char data[0]; + char data[]; }; static inline void btrfs_init_delayed_root( diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index dfdb7d4f8406..353cc2994d10 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -82,8 +82,7 @@ void btrfs_delayed_refs_rsv_release(struct btrfs_fs_info *fs_info, int nr) u64 num_bytes = btrfs_calc_insert_metadata_size(fs_info, nr); u64 released = 0; - released = __btrfs_block_rsv_release(fs_info, block_rsv, num_bytes, - NULL); + released = btrfs_block_rsv_release(fs_info, block_rsv, num_bytes, NULL); if (released) trace_btrfs_space_reservation(fs_info, "delayed_refs_rsv", 0, released, 0); diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index 2ca2a09d0e23..db93909b25e0 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -22,6 +22,46 @@ #include "dev-replace.h" #include "sysfs.h" +/* + * Device replace overview + * + * [Objective] + * To copy all extents (both new and on-disk) from source device to target + * device, while still keeping the filesystem read-write. + * + * [Method] + * There are two main methods involved: + * + * - Write duplication + * + * All new writes will be written to both target and source devices, so even + * if replace gets canceled, sources device still contans up-to-date data. + * + * Location: handle_ops_on_dev_replace() from __btrfs_map_block() + * Start: btrfs_dev_replace_start() + * End: btrfs_dev_replace_finishing() + * Content: Latest data/metadata + * + * - Copy existing extents + * + * This happens by re-using scrub facility, as scrub also iterates through + * existing extents from commit root. + * + * Location: scrub_write_block_to_dev_replace() from + * scrub_block_complete() + * Content: Data/meta from commit root. + * + * Due to the content difference, we need to avoid nocow write when dev-replace + * is happening. This is done by marking the block group read-only and waiting + * for NOCOW writes. + * + * After replace is done, the finishing part is done by swapping the target and + * source devices. + * + * Location: btrfs_dev_replace_update_device_in_mapping_tree() from + * btrfs_dev_replace_finishing() + */ + static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, int scrub_ret); static void btrfs_dev_replace_update_device_in_mapping_tree( @@ -472,7 +512,7 @@ static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0); up_write(&dev_replace->rwsem); - ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device); + ret = btrfs_sysfs_add_devices_dir(tgt_device->fs_devices, tgt_device); if (ret) btrfs_err(fs_info, "kobj add dev failed %d", ret); @@ -703,7 +743,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, mutex_unlock(&fs_info->fs_devices->device_list_mutex); /* replace the sysfs entry */ - btrfs_sysfs_rm_device_link(fs_info->fs_devices, src_device); + btrfs_sysfs_remove_devices_dir(fs_info->fs_devices, src_device); btrfs_sysfs_update_devid(tgt_device); btrfs_rm_dev_replace_free_srcdev(src_device); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 89422aa8e9d1..a6cb5cbbdb9f 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -7,7 +7,6 @@ #include <linux/blkdev.h> #include <linux/radix-tree.h> #include <linux/writeback.h> -#include <linux/buffer_head.h> #include <linux/workqueue.h> #include <linux/kthread.h> #include <linux/slab.h> @@ -42,6 +41,7 @@ #include "ref-verify.h" #include "block-group.h" #include "discard.h" +#include "space-info.h" #define BTRFS_SUPER_FLAG_SUPP (BTRFS_HEADER_FLAG_WRITTEN |\ BTRFS_HEADER_FLAG_RELOC |\ @@ -98,6 +98,12 @@ void __cold btrfs_end_io_wq_exit(void) kmem_cache_destroy(btrfs_end_io_wq_cache); } +static void btrfs_free_csum_hash(struct btrfs_fs_info *fs_info) +{ + if (fs_info->csum_shash) + crypto_free_shash(fs_info->csum_shash); +} + /* * async submit bios are used to offload expensive checksumming * onto the worker threads. They checksum file and metadata bios @@ -247,47 +253,27 @@ out: /* * Compute the csum of a btree block and store the result to provided buffer. - * - * Returns error if the extent buffer cannot be mapped. */ -static int csum_tree_block(struct extent_buffer *buf, u8 *result) +static void csum_tree_block(struct extent_buffer *buf, u8 *result) { struct btrfs_fs_info *fs_info = buf->fs_info; + const int num_pages = fs_info->nodesize >> PAGE_SHIFT; SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); - unsigned long len; - unsigned long cur_len; - unsigned long offset = BTRFS_CSUM_SIZE; char *kaddr; - unsigned long map_start; - unsigned long map_len; - int err; + int i; shash->tfm = fs_info->csum_shash; crypto_shash_init(shash); + kaddr = page_address(buf->pages[0]); + crypto_shash_update(shash, kaddr + BTRFS_CSUM_SIZE, + PAGE_SIZE - BTRFS_CSUM_SIZE); - len = buf->len - offset; - - while (len > 0) { - /* - * Note: we don't need to check for the err == 1 case here, as - * with the given combination of 'start = BTRFS_CSUM_SIZE (32)' - * and 'min_len = 32' and the currently implemented mapping - * algorithm we cannot cross a page boundary. - */ - err = map_private_extent_buffer(buf, offset, 32, - &kaddr, &map_start, &map_len); - if (WARN_ON(err)) - return err; - cur_len = min(len, map_len - (offset - map_start)); - crypto_shash_update(shash, kaddr + offset - map_start, cur_len); - len -= cur_len; - offset += cur_len; + for (i = 1; i < num_pages; i++) { + kaddr = page_address(buf->pages[i]); + crypto_shash_update(shash, kaddr, PAGE_SIZE); } memset(result, 0, BTRFS_CSUM_SIZE); - crypto_shash_final(shash, result); - - return 0; } /* @@ -535,10 +521,10 @@ static int csum_dirty_buffer(struct btrfs_fs_info *fs_info, struct page *page) return -EUCLEAN; ASSERT(memcmp_extent_buffer(eb, fs_info->fs_devices->metadata_uuid, - btrfs_header_fsid(), BTRFS_FSID_SIZE) == 0); + offsetof(struct btrfs_header, fsid), + BTRFS_FSID_SIZE) == 0); - if (csum_tree_block(eb, result)) - return -EINVAL; + csum_tree_block(eb, result); if (btrfs_header_level(eb)) ret = btrfs_check_node(eb); @@ -565,7 +551,8 @@ static int check_tree_block_fsid(struct extent_buffer *eb) u8 fsid[BTRFS_FSID_SIZE]; int ret = 1; - read_extent_buffer(eb, fsid, btrfs_header_fsid(), BTRFS_FSID_SIZE); + read_extent_buffer(eb, fsid, offsetof(struct btrfs_header, fsid), + BTRFS_FSID_SIZE); while (fs_devices) { u8 *metadata_uuid; @@ -596,9 +583,8 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio, u64 found_start; int found_level; struct extent_buffer *eb; - struct btrfs_root *root = BTRFS_I(page->mapping->host)->root; - struct btrfs_fs_info *fs_info = root->fs_info; - u16 csum_size = btrfs_super_csum_size(fs_info->super_copy); + struct btrfs_fs_info *fs_info; + u16 csum_size; int ret = 0; u8 result[BTRFS_CSUM_SIZE]; int reads_done; @@ -607,6 +593,8 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio, goto out; eb = (struct extent_buffer *)page->private; + fs_info = eb->fs_info; + csum_size = btrfs_super_csum_size(fs_info->super_copy); /* the pending IO might have been the only thing that kept this buffer * in memory. Make sure we have a ref for all this other checks @@ -647,9 +635,7 @@ static int btree_readpage_end_io_hook(struct btrfs_io_bio *io_bio, btrfs_set_buffer_lockdep_class(btrfs_header_owner(eb), eb, found_level); - ret = csum_tree_block(eb, result); - if (ret) - goto err; + csum_tree_block(eb, result); if (memcmp_extent_buffer(eb, result, 0, csum_size)) { u32 val; @@ -972,9 +958,7 @@ static int btree_writepages(struct address_space *mapping, static int btree_readpage(struct file *file, struct page *page) { - struct extent_io_tree *tree; - tree = &BTRFS_I(page->mapping->host)->io_tree; - return extent_read_full_page(tree, page, btree_get_extent, 0); + return extent_read_full_page(page, btree_get_extent, 0); } static int btree_releasepage(struct page *page, gfp_t gfp_flags) @@ -1100,36 +1084,11 @@ void btrfs_clean_tree_block(struct extent_buffer *buf) } } -static struct btrfs_subvolume_writers *btrfs_alloc_subvolume_writers(void) -{ - struct btrfs_subvolume_writers *writers; - int ret; - - writers = kmalloc(sizeof(*writers), GFP_NOFS); - if (!writers) - return ERR_PTR(-ENOMEM); - - ret = percpu_counter_init(&writers->counter, 0, GFP_NOFS); - if (ret < 0) { - kfree(writers); - return ERR_PTR(ret); - } - - init_waitqueue_head(&writers->wait); - return writers; -} - -static void -btrfs_free_subvolume_writers(struct btrfs_subvolume_writers *writers) -{ - percpu_counter_destroy(&writers->counter); - kfree(writers); -} - static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, u64 objectid) { bool dummy = test_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state); + root->fs_info = fs_info; root->node = NULL; root->commit_root = NULL; root->state = 0; @@ -1173,7 +1132,6 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, atomic_set(&root->log_writers, 0); atomic_set(&root->log_batch, 0); refcount_set(&root->refs, 1); - atomic_set(&root->will_be_snapshotted, 0); atomic_set(&root->snapshot_force_cow, 0); atomic_set(&root->nr_swapfiles, 0); root->log_transid = 0; @@ -1195,14 +1153,20 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, spin_lock_init(&root->root_item_lock); btrfs_qgroup_init_swapped_blocks(&root->swapped_blocks); +#ifdef CONFIG_BTRFS_DEBUG + INIT_LIST_HEAD(&root->leak_list); + spin_lock(&fs_info->fs_roots_radix_lock); + list_add_tail(&root->leak_list, &fs_info->allocated_roots); + spin_unlock(&fs_info->fs_roots_radix_lock); +#endif } static struct btrfs_root *btrfs_alloc_root(struct btrfs_fs_info *fs_info, - gfp_t flags) + u64 objectid, gfp_t flags) { struct btrfs_root *root = kzalloc(sizeof(*root), flags); if (root) - root->fs_info = fs_info; + __setup_root(root, fs_info, objectid); return root; } @@ -1215,12 +1179,11 @@ struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info) if (!fs_info) return ERR_PTR(-EINVAL); - root = btrfs_alloc_root(fs_info, GFP_KERNEL); + root = btrfs_alloc_root(fs_info, BTRFS_ROOT_TREE_OBJECTID, GFP_KERNEL); if (!root) return ERR_PTR(-ENOMEM); /* We don't use the stripesize in selftest, set it as sectorsize */ - __setup_root(root, fs_info, BTRFS_ROOT_TREE_OBJECTID); root->alloc_bytenr = 0; return root; @@ -1237,19 +1200,17 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, struct btrfs_key key; unsigned int nofs_flag; int ret = 0; - uuid_le uuid = NULL_UUID_LE; /* * We're holding a transaction handle, so use a NOFS memory allocation * context to avoid deadlock if reclaim happens. */ nofs_flag = memalloc_nofs_save(); - root = btrfs_alloc_root(fs_info, GFP_KERNEL); + root = btrfs_alloc_root(fs_info, objectid, GFP_KERNEL); memalloc_nofs_restore(nofs_flag); if (!root) return ERR_PTR(-ENOMEM); - __setup_root(root, fs_info, objectid); root->root_key.objectid = objectid; root->root_key.type = BTRFS_ROOT_ITEM_KEY; root->root_key.offset = 0; @@ -1277,8 +1238,9 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, btrfs_set_root_last_snapshot(&root->root_item, 0); btrfs_set_root_dirid(&root->root_item, 0); if (is_fstree(objectid)) - uuid_le_gen(&uuid); - memcpy(root->root_item.uuid, uuid.b, BTRFS_UUID_SIZE); + generate_random_guid(root->root_item.uuid); + else + export_guid(root->root_item.uuid, &guid_null); root->root_item.drop_level = 0; key.objectid = objectid; @@ -1293,12 +1255,9 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, return root; fail: - if (leaf) { + if (leaf) btrfs_tree_unlock(leaf); - free_extent_buffer(root->commit_root); - free_extent_buffer(leaf); - } - kfree(root); + btrfs_put_root(root); return ERR_PTR(ret); } @@ -1309,12 +1268,10 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans, struct btrfs_root *root; struct extent_buffer *leaf; - root = btrfs_alloc_root(fs_info, GFP_NOFS); + root = btrfs_alloc_root(fs_info, BTRFS_TREE_LOG_OBJECTID, GFP_NOFS); if (!root) return ERR_PTR(-ENOMEM); - __setup_root(root, fs_info, BTRFS_TREE_LOG_OBJECTID); - root->root_key.objectid = BTRFS_TREE_LOG_OBJECTID; root->root_key.type = BTRFS_ROOT_ITEM_KEY; root->root_key.offset = BTRFS_TREE_LOG_OBJECTID; @@ -1331,7 +1288,7 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans, leaf = btrfs_alloc_tree_block(trans, root, 0, BTRFS_TREE_LOG_OBJECTID, NULL, 0, 0, 0); if (IS_ERR(leaf)) { - kfree(root); + btrfs_put_root(root); return ERR_CAST(leaf); } @@ -1387,8 +1344,8 @@ int btrfs_add_log_tree(struct btrfs_trans_handle *trans, return 0; } -static struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, - struct btrfs_key *key) +struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, + struct btrfs_key *key) { struct btrfs_root *root; struct btrfs_fs_info *fs_info = tree_root->fs_info; @@ -1401,14 +1358,12 @@ static struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, if (!path) return ERR_PTR(-ENOMEM); - root = btrfs_alloc_root(fs_info, GFP_NOFS); + root = btrfs_alloc_root(fs_info, key->objectid, GFP_NOFS); if (!root) { ret = -ENOMEM; goto alloc_fail; } - __setup_root(root, fs_info, key->objectid); - ret = btrfs_find_root(tree_root, key, path, &root->root_item, &root->root_key); if (ret) { @@ -1424,10 +1379,10 @@ static struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, generation, level, NULL); if (IS_ERR(root->node)) { ret = PTR_ERR(root->node); + root->node = NULL; goto find_fail; } else if (!btrfs_buffer_uptodate(root->node, generation, 0)) { ret = -EIO; - free_extent_buffer(root->node); goto find_fail; } root->commit_root = btrfs_root_node(root); @@ -1436,33 +1391,16 @@ out: return root; find_fail: - kfree(root); + btrfs_put_root(root); alloc_fail: root = ERR_PTR(ret); goto out; } -struct btrfs_root *btrfs_read_fs_root(struct btrfs_root *tree_root, - struct btrfs_key *location) -{ - struct btrfs_root *root; - - root = btrfs_read_tree_root(tree_root, location); - if (IS_ERR(root)) - return root; - - if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { - set_bit(BTRFS_ROOT_REF_COWS, &root->state); - btrfs_check_and_init_root_item(&root->root_item); - } - - return root; -} - -int btrfs_init_fs_root(struct btrfs_root *root) +static int btrfs_init_fs_root(struct btrfs_root *root) { int ret; - struct btrfs_subvolume_writers *writers; + unsigned int nofs_flag; root->free_ino_ctl = kzalloc(sizeof(*root->free_ino_ctl), GFP_NOFS); root->free_ino_pinned = kzalloc(sizeof(*root->free_ino_pinned), @@ -1472,12 +1410,20 @@ int btrfs_init_fs_root(struct btrfs_root *root) goto fail; } - writers = btrfs_alloc_subvolume_writers(); - if (IS_ERR(writers)) { - ret = PTR_ERR(writers); + /* + * We might be called under a transaction (e.g. indirect backref + * resolution) which could deadlock if it triggers memory reclaim + */ + nofs_flag = memalloc_nofs_save(); + ret = btrfs_drew_lock_init(&root->snapshot_lock); + memalloc_nofs_restore(nofs_flag); + if (ret) goto fail; + + if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { + set_bit(BTRFS_ROOT_REF_COWS, &root->state); + btrfs_check_and_init_root_item(&root->root_item); } - root->subv_writers = writers; btrfs_init_free_ino_ctl(root); spin_lock_init(&root->ino_cache_lock); @@ -1505,14 +1451,16 @@ fail: return ret; } -struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info, - u64 root_id) +static struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info, + u64 root_id) { struct btrfs_root *root; spin_lock(&fs_info->fs_roots_radix_lock); root = radix_tree_lookup(&fs_info->fs_roots_radix, (unsigned long)root_id); + if (root) + root = btrfs_grab_root(root); spin_unlock(&fs_info->fs_roots_radix_lock); return root; } @@ -1530,14 +1478,62 @@ int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info, ret = radix_tree_insert(&fs_info->fs_roots_radix, (unsigned long)root->root_key.objectid, root); - if (ret == 0) + if (ret == 0) { + btrfs_grab_root(root); set_bit(BTRFS_ROOT_IN_RADIX, &root->state); + } spin_unlock(&fs_info->fs_roots_radix_lock); radix_tree_preload_end(); return ret; } +void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info) +{ +#ifdef CONFIG_BTRFS_DEBUG + struct btrfs_root *root; + + while (!list_empty(&fs_info->allocated_roots)) { + root = list_first_entry(&fs_info->allocated_roots, + struct btrfs_root, leak_list); + btrfs_err(fs_info, "leaked root %llu-%llu refcount %d", + root->root_key.objectid, root->root_key.offset, + refcount_read(&root->refs)); + while (refcount_read(&root->refs) > 1) + btrfs_put_root(root); + btrfs_put_root(root); + } +#endif +} + +void btrfs_free_fs_info(struct btrfs_fs_info *fs_info) +{ + percpu_counter_destroy(&fs_info->dirty_metadata_bytes); + percpu_counter_destroy(&fs_info->delalloc_bytes); + percpu_counter_destroy(&fs_info->dio_bytes); + percpu_counter_destroy(&fs_info->dev_replace.bio_counter); + btrfs_free_csum_hash(fs_info); + btrfs_free_stripe_hash_table(fs_info); + btrfs_free_ref_cache(fs_info); + kfree(fs_info->balance_ctl); + kfree(fs_info->delayed_root); + btrfs_put_root(fs_info->extent_root); + btrfs_put_root(fs_info->tree_root); + btrfs_put_root(fs_info->chunk_root); + btrfs_put_root(fs_info->dev_root); + btrfs_put_root(fs_info->csum_root); + btrfs_put_root(fs_info->quota_root); + btrfs_put_root(fs_info->uuid_root); + btrfs_put_root(fs_info->free_space_root); + btrfs_put_root(fs_info->fs_root); + btrfs_check_leaked_roots(fs_info); + btrfs_extent_buffer_leak_debug_check(fs_info); + kfree(fs_info->super_copy); + kfree(fs_info->super_for_commit); + kvfree(fs_info); +} + + struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_key *location, bool check_ref) @@ -1548,33 +1544,35 @@ struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info, int ret; if (location->objectid == BTRFS_ROOT_TREE_OBJECTID) - return fs_info->tree_root; + return btrfs_grab_root(fs_info->tree_root); if (location->objectid == BTRFS_EXTENT_TREE_OBJECTID) - return fs_info->extent_root; + return btrfs_grab_root(fs_info->extent_root); if (location->objectid == BTRFS_CHUNK_TREE_OBJECTID) - return fs_info->chunk_root; + return btrfs_grab_root(fs_info->chunk_root); if (location->objectid == BTRFS_DEV_TREE_OBJECTID) - return fs_info->dev_root; + return btrfs_grab_root(fs_info->dev_root); if (location->objectid == BTRFS_CSUM_TREE_OBJECTID) - return fs_info->csum_root; + return btrfs_grab_root(fs_info->csum_root); if (location->objectid == BTRFS_QUOTA_TREE_OBJECTID) - return fs_info->quota_root ? fs_info->quota_root : - ERR_PTR(-ENOENT); + return btrfs_grab_root(fs_info->quota_root) ? + fs_info->quota_root : ERR_PTR(-ENOENT); if (location->objectid == BTRFS_UUID_TREE_OBJECTID) - return fs_info->uuid_root ? fs_info->uuid_root : - ERR_PTR(-ENOENT); + return btrfs_grab_root(fs_info->uuid_root) ? + fs_info->uuid_root : ERR_PTR(-ENOENT); if (location->objectid == BTRFS_FREE_SPACE_TREE_OBJECTID) - return fs_info->free_space_root ? fs_info->free_space_root : - ERR_PTR(-ENOENT); + return btrfs_grab_root(fs_info->free_space_root) ? + fs_info->free_space_root : ERR_PTR(-ENOENT); again: root = btrfs_lookup_fs_root(fs_info, location->objectid); if (root) { - if (check_ref && btrfs_root_refs(&root->root_item) == 0) + if (check_ref && btrfs_root_refs(&root->root_item) == 0) { + btrfs_put_root(root); return ERR_PTR(-ENOENT); + } return root; } - root = btrfs_read_fs_root(fs_info->tree_root, location); + root = btrfs_read_tree_root(fs_info->tree_root, location); if (IS_ERR(root)) return root; @@ -1605,15 +1603,14 @@ again: ret = btrfs_insert_fs_root(fs_info, root); if (ret) { - if (ret == -EEXIST) { - btrfs_free_fs_root(root); + btrfs_put_root(root); + if (ret == -EEXIST) goto again; - } goto fail; } return root; fail: - btrfs_free_fs_root(root); + btrfs_put_root(root); return ERR_PTR(ret); } @@ -1985,11 +1982,35 @@ static void free_root_pointers(struct btrfs_fs_info *info, bool free_chunk_root) free_root_extent_buffers(info->csum_root); free_root_extent_buffers(info->quota_root); free_root_extent_buffers(info->uuid_root); + free_root_extent_buffers(info->fs_root); if (free_chunk_root) free_root_extent_buffers(info->chunk_root); free_root_extent_buffers(info->free_space_root); } +void btrfs_put_root(struct btrfs_root *root) +{ + if (!root) + return; + + if (refcount_dec_and_test(&root->refs)) { + WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree)); + if (root->anon_dev) + free_anon_bdev(root->anon_dev); + btrfs_drew_lock_destroy(&root->snapshot_lock); + free_extent_buffer(root->node); + free_extent_buffer(root->commit_root); + kfree(root->free_ino_ctl); + kfree(root->free_ino_pinned); +#ifdef CONFIG_BTRFS_DEBUG + spin_lock(&root->fs_info->fs_roots_radix_lock); + list_del_init(&root->leak_list); + spin_unlock(&root->fs_info->fs_roots_radix_lock); +#endif + kfree(root); + } +} + void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info) { int ret; @@ -2001,13 +2022,9 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info) struct btrfs_root, root_list); list_del(&gang[0]->root_list); - if (test_bit(BTRFS_ROOT_IN_RADIX, &gang[0]->state)) { + if (test_bit(BTRFS_ROOT_IN_RADIX, &gang[0]->state)) btrfs_drop_and_free_fs_root(fs_info, gang[0]); - } else { - free_extent_buffer(gang[0]->node); - free_extent_buffer(gang[0]->commit_root); - btrfs_put_fs_root(gang[0]); - } + btrfs_put_root(gang[0]); } while (1) { @@ -2020,10 +2037,8 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info) btrfs_drop_and_free_fs_root(fs_info, gang[i]); } - if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) { + if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) btrfs_free_log_root_tree(NULL, fs_info); - btrfs_destroy_pinned_extent(fs_info, fs_info->pinned_extents); - } } static void btrfs_init_scrub(struct btrfs_fs_info *fs_info) @@ -2069,7 +2084,7 @@ static void btrfs_init_btree_inode(struct btrfs_fs_info *fs_info) BTRFS_I(inode)->io_tree.ops = &btree_extent_io_ops; - BTRFS_I(inode)->root = fs_info->tree_root; + BTRFS_I(inode)->root = btrfs_grab_root(fs_info->tree_root); memset(&BTRFS_I(inode)->location, 0, sizeof(struct btrfs_key)); set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags); btrfs_insert_inode_hash(inode); @@ -2189,11 +2204,6 @@ static int btrfs_init_csum_hash(struct btrfs_fs_info *fs_info, u16 csum_type) return 0; } -static void btrfs_free_csum_hash(struct btrfs_fs_info *fs_info) -{ - crypto_free_shash(fs_info->csum_shash); -} - static int btrfs_replay_log(struct btrfs_fs_info *fs_info, struct btrfs_fs_devices *fs_devices) { @@ -2208,24 +2218,23 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info, return -EIO; } - log_tree_root = btrfs_alloc_root(fs_info, GFP_KERNEL); + log_tree_root = btrfs_alloc_root(fs_info, BTRFS_TREE_LOG_OBJECTID, + GFP_KERNEL); if (!log_tree_root) return -ENOMEM; - __setup_root(log_tree_root, fs_info, BTRFS_TREE_LOG_OBJECTID); - log_tree_root->node = read_tree_block(fs_info, bytenr, fs_info->generation + 1, level, NULL); if (IS_ERR(log_tree_root->node)) { btrfs_warn(fs_info, "failed to read log tree"); ret = PTR_ERR(log_tree_root->node); - kfree(log_tree_root); + log_tree_root->node = NULL; + btrfs_put_root(log_tree_root); return ret; } else if (!extent_buffer_uptodate(log_tree_root->node)) { btrfs_err(fs_info, "failed to read log tree"); - free_extent_buffer(log_tree_root->node); - kfree(log_tree_root); + btrfs_put_root(log_tree_root); return -EIO; } /* returns with log_tree_root freed on success */ @@ -2233,8 +2242,7 @@ static int btrfs_replay_log(struct btrfs_fs_info *fs_info, if (ret) { btrfs_handle_fs_error(fs_info, ret, "Failed to recover log tree"); - free_extent_buffer(log_tree_root->node); - kfree(log_tree_root); + btrfs_put_root(log_tree_root); return ret; } @@ -2624,67 +2632,8 @@ static int __cold init_tree_roots(struct btrfs_fs_info *fs_info) return ret; } -int __cold open_ctree(struct super_block *sb, - struct btrfs_fs_devices *fs_devices, - char *options) +void btrfs_init_fs_info(struct btrfs_fs_info *fs_info) { - u32 sectorsize; - u32 nodesize; - u32 stripesize; - u64 generation; - u64 features; - u16 csum_type; - struct btrfs_key location; - struct buffer_head *bh; - struct btrfs_super_block *disk_super; - struct btrfs_fs_info *fs_info = btrfs_sb(sb); - struct btrfs_root *tree_root; - struct btrfs_root *chunk_root; - int ret; - int err = -EINVAL; - int clear_free_space_tree = 0; - int level; - - tree_root = fs_info->tree_root = btrfs_alloc_root(fs_info, GFP_KERNEL); - chunk_root = fs_info->chunk_root = btrfs_alloc_root(fs_info, GFP_KERNEL); - if (!tree_root || !chunk_root) { - err = -ENOMEM; - goto fail; - } - - ret = init_srcu_struct(&fs_info->subvol_srcu); - if (ret) { - err = ret; - goto fail; - } - - ret = percpu_counter_init(&fs_info->dio_bytes, 0, GFP_KERNEL); - if (ret) { - err = ret; - goto fail_srcu; - } - - ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0, GFP_KERNEL); - if (ret) { - err = ret; - goto fail_dio_bytes; - } - fs_info->dirty_metadata_batch = PAGE_SIZE * - (1 + ilog2(nr_cpu_ids)); - - ret = percpu_counter_init(&fs_info->delalloc_bytes, 0, GFP_KERNEL); - if (ret) { - err = ret; - goto fail_dirty_metadata_bytes; - } - - ret = percpu_counter_init(&fs_info->dev_replace.bio_counter, 0, - GFP_KERNEL); - if (ret) { - err = ret; - goto fail_delalloc_bytes; - } - INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC); INIT_RADIX_TREE(&fs_info->buffer_radix, GFP_ATOMIC); INIT_LIST_HEAD(&fs_info->trans_list); @@ -2711,6 +2660,11 @@ int __cold open_ctree(struct super_block *sb, INIT_LIST_HEAD(&fs_info->space_info); INIT_LIST_HEAD(&fs_info->tree_mod_seq_list); INIT_LIST_HEAD(&fs_info->unused_bgs); +#ifdef CONFIG_BTRFS_DEBUG + INIT_LIST_HEAD(&fs_info->allocated_roots); + INIT_LIST_HEAD(&fs_info->allocated_ebs); + spin_lock_init(&fs_info->eb_leak_lock); +#endif extent_map_tree_init(&fs_info->mapping_tree); btrfs_init_block_rsv(&fs_info->global_block_rsv, BTRFS_BLOCK_RSV_GLOBAL); @@ -2727,7 +2681,6 @@ int __cold open_ctree(struct super_block *sb, atomic_set(&fs_info->reada_works_cnt, 0); atomic_set(&fs_info->nr_delayed_iputs, 0); atomic64_set(&fs_info->tree_mod_seq, 0); - fs_info->sb = sb; fs_info->max_inline = BTRFS_DEFAULT_MAX_INLINE; fs_info->metadata_ratio = 0; fs_info->defrag_inodes = RB_ROOT; @@ -2746,21 +2699,6 @@ int __cold open_ctree(struct super_block *sb, INIT_LIST_HEAD(&fs_info->ordered_roots); spin_lock_init(&fs_info->ordered_root_lock); - fs_info->btree_inode = new_inode(sb); - if (!fs_info->btree_inode) { - err = -ENOMEM; - goto fail_bio_counter; - } - mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS); - - fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root), - GFP_KERNEL); - if (!fs_info->delayed_root) { - err = -ENOMEM; - goto fail_iput; - } - btrfs_init_delayed_root(fs_info->delayed_root); - btrfs_init_scrub(fs_info); #ifdef CONFIG_BTRFS_FS_CHECK_INTEGRITY fs_info->check_integrity_print_mask = 0; @@ -2768,20 +2706,12 @@ int __cold open_ctree(struct super_block *sb, btrfs_init_balance(fs_info); btrfs_init_async_reclaim_work(&fs_info->async_reclaim_work); - sb->s_blocksize = BTRFS_BDEV_BLOCKSIZE; - sb->s_blocksize_bits = blksize_bits(BTRFS_BDEV_BLOCKSIZE); - - btrfs_init_btree_inode(fs_info); - spin_lock_init(&fs_info->block_group_cache_lock); fs_info->block_group_cache_tree = RB_ROOT; fs_info->first_logical_byte = (u64)-1; - extent_io_tree_init(fs_info, &fs_info->freed_extents[0], - IO_TREE_FS_INFO_FREED_EXTENTS0, NULL); - extent_io_tree_init(fs_info, &fs_info->freed_extents[1], - IO_TREE_FS_INFO_FREED_EXTENTS1, NULL); - fs_info->pinned_extents = &fs_info->freed_extents[0]; + extent_io_tree_init(fs_info, &fs_info->excluded_extents, + IO_TREE_FS_EXCLUDED_EXTENTS, NULL); set_bit(BTRFS_FS_BARRIER, &fs_info->flags); mutex_init(&fs_info->ordered_operations_mutex); @@ -2817,23 +2747,135 @@ int __cold open_ctree(struct super_block *sb, fs_info->swapfile_pins = RB_ROOT; fs_info->send_in_progress = 0; +} + +static int init_mount_fs_info(struct btrfs_fs_info *fs_info, struct super_block *sb) +{ + int ret; + + fs_info->sb = sb; + sb->s_blocksize = BTRFS_BDEV_BLOCKSIZE; + sb->s_blocksize_bits = blksize_bits(BTRFS_BDEV_BLOCKSIZE); + + ret = percpu_counter_init(&fs_info->dio_bytes, 0, GFP_KERNEL); + if (ret) + return ret; + + ret = percpu_counter_init(&fs_info->dirty_metadata_bytes, 0, GFP_KERNEL); + if (ret) + return ret; + + fs_info->dirty_metadata_batch = PAGE_SIZE * + (1 + ilog2(nr_cpu_ids)); + + ret = percpu_counter_init(&fs_info->delalloc_bytes, 0, GFP_KERNEL); + if (ret) + return ret; + + ret = percpu_counter_init(&fs_info->dev_replace.bio_counter, 0, + GFP_KERNEL); + if (ret) + return ret; + + fs_info->delayed_root = kmalloc(sizeof(struct btrfs_delayed_root), + GFP_KERNEL); + if (!fs_info->delayed_root) + return -ENOMEM; + btrfs_init_delayed_root(fs_info->delayed_root); + + return btrfs_alloc_stripe_hash_table(fs_info); +} + +static int btrfs_uuid_rescan_kthread(void *data) +{ + struct btrfs_fs_info *fs_info = (struct btrfs_fs_info *)data; + int ret; - ret = btrfs_alloc_stripe_hash_table(fs_info); + /* + * 1st step is to iterate through the existing UUID tree and + * to delete all entries that contain outdated data. + * 2nd step is to add all missing entries to the UUID tree. + */ + ret = btrfs_uuid_tree_iterate(fs_info); + if (ret < 0) { + if (ret != -EINTR) + btrfs_warn(fs_info, "iterating uuid_tree failed %d", + ret); + up(&fs_info->uuid_tree_rescan_sem); + return ret; + } + return btrfs_uuid_scan_kthread(data); +} + +static int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info) +{ + struct task_struct *task; + + down(&fs_info->uuid_tree_rescan_sem); + task = kthread_run(btrfs_uuid_rescan_kthread, fs_info, "btrfs-uuid"); + if (IS_ERR(task)) { + /* fs_info->update_uuid_tree_gen remains 0 in all error case */ + btrfs_warn(fs_info, "failed to start uuid_rescan task"); + up(&fs_info->uuid_tree_rescan_sem); + return PTR_ERR(task); + } + + return 0; +} + +int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_devices, + char *options) +{ + u32 sectorsize; + u32 nodesize; + u32 stripesize; + u64 generation; + u64 features; + u16 csum_type; + struct btrfs_key location; + struct btrfs_super_block *disk_super; + struct btrfs_fs_info *fs_info = btrfs_sb(sb); + struct btrfs_root *tree_root; + struct btrfs_root *chunk_root; + int ret; + int err = -EINVAL; + int clear_free_space_tree = 0; + int level; + + ret = init_mount_fs_info(fs_info, sb); if (ret) { err = ret; - goto fail_alloc; + goto fail; } - __setup_root(tree_root, fs_info, BTRFS_ROOT_TREE_OBJECTID); + /* These need to be init'ed before we start creating inodes and such. */ + tree_root = btrfs_alloc_root(fs_info, BTRFS_ROOT_TREE_OBJECTID, + GFP_KERNEL); + fs_info->tree_root = tree_root; + chunk_root = btrfs_alloc_root(fs_info, BTRFS_CHUNK_TREE_OBJECTID, + GFP_KERNEL); + fs_info->chunk_root = chunk_root; + if (!tree_root || !chunk_root) { + err = -ENOMEM; + goto fail; + } + + fs_info->btree_inode = new_inode(sb); + if (!fs_info->btree_inode) { + err = -ENOMEM; + goto fail; + } + mapping_set_gfp_mask(fs_info->btree_inode->i_mapping, GFP_NOFS); + btrfs_init_btree_inode(fs_info); invalidate_bdev(fs_devices->latest_bdev); /* * Read super block and check the signature bytes only */ - bh = btrfs_read_dev_super(fs_devices->latest_bdev); - if (IS_ERR(bh)) { - err = PTR_ERR(bh); + disk_super = btrfs_read_dev_super(fs_devices->latest_bdev); + if (IS_ERR(disk_super)) { + err = PTR_ERR(disk_super); goto fail_alloc; } @@ -2841,18 +2883,19 @@ int __cold open_ctree(struct super_block *sb, * Verify the type first, if that or the the checksum value are * corrupted, we'll find out */ - csum_type = btrfs_super_csum_type((struct btrfs_super_block *)bh->b_data); + csum_type = btrfs_super_csum_type(disk_super); if (!btrfs_supported_super_csum(csum_type)) { btrfs_err(fs_info, "unsupported checksum algorithm: %u", csum_type); err = -EINVAL; - brelse(bh); + btrfs_release_disk_super(disk_super); goto fail_alloc; } ret = btrfs_init_csum_hash(fs_info, csum_type); if (ret) { err = ret; + btrfs_release_disk_super(disk_super); goto fail_alloc; } @@ -2860,11 +2903,11 @@ int __cold open_ctree(struct super_block *sb, * We want to check superblock checksum, the type is stored inside. * Pass the whole disk block of size BTRFS_SUPER_INFO_SIZE (4k). */ - if (btrfs_check_super_csum(fs_info, bh->b_data)) { + if (btrfs_check_super_csum(fs_info, (u8 *)disk_super)) { btrfs_err(fs_info, "superblock checksum mismatch"); err = -EINVAL; - brelse(bh); - goto fail_csum; + btrfs_release_disk_super(disk_super); + goto fail_alloc; } /* @@ -2872,8 +2915,8 @@ int __cold open_ctree(struct super_block *sb, * following bytes up to INFO_SIZE, the checksum is calculated from * the whole block of INFO_SIZE */ - memcpy(fs_info->super_copy, bh->b_data, sizeof(*fs_info->super_copy)); - brelse(bh); + memcpy(fs_info->super_copy, disk_super, sizeof(*fs_info->super_copy)); + btrfs_release_disk_super(disk_super); disk_super = fs_info->super_copy; @@ -2901,11 +2944,11 @@ int __cold open_ctree(struct super_block *sb, if (ret) { btrfs_err(fs_info, "superblock contains fatal errors"); err = -EINVAL; - goto fail_csum; + goto fail_alloc; } if (!btrfs_super_root(disk_super)) - goto fail_csum; + goto fail_alloc; /* check FS state, whether FS is broken. */ if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_ERROR) @@ -2920,7 +2963,7 @@ int __cold open_ctree(struct super_block *sb, ret = btrfs_parse_options(fs_info, options, sb->s_flags); if (ret) { err = ret; - goto fail_csum; + goto fail_alloc; } features = btrfs_super_incompat_flags(disk_super) & @@ -2930,7 +2973,7 @@ int __cold open_ctree(struct super_block *sb, "cannot mount because of unsupported optional features (%llx)", features); err = -EINVAL; - goto fail_csum; + goto fail_alloc; } features = btrfs_super_incompat_flags(disk_super); @@ -2974,7 +3017,7 @@ int __cold open_ctree(struct super_block *sb, btrfs_err(fs_info, "unequal nodesize/sectorsize (%u != %u) are not allowed for mixed block groups", nodesize, sectorsize); - goto fail_csum; + goto fail_alloc; } /* @@ -2990,7 +3033,7 @@ int __cold open_ctree(struct super_block *sb, "cannot mount read-write because of unsupported optional features (%llx)", features); err = -EINVAL; - goto fail_csum; + goto fail_alloc; } ret = btrfs_init_workqueues(fs_info, fs_devices); @@ -3021,8 +3064,6 @@ int __cold open_ctree(struct super_block *sb, generation = btrfs_super_chunk_root_generation(disk_super); level = btrfs_super_chunk_root_level(disk_super); - __setup_root(chunk_root, fs_info, BTRFS_CHUNK_TREE_OBJECTID); - chunk_root->node = read_tree_block(fs_info, btrfs_super_chunk_root(disk_super), generation, level, NULL); @@ -3038,7 +3079,8 @@ int __cold open_ctree(struct super_block *sb, chunk_root->commit_root = btrfs_root_node(chunk_root); read_extent_buffer(chunk_root->node, fs_info->chunk_tree_uuid, - btrfs_header_chunk_tree_uuid(chunk_root->node), BTRFS_UUID_SIZE); + offsetof(struct btrfs_header, chunk_tree_uuid), + BTRFS_UUID_SIZE); ret = btrfs_read_chunk_tree(fs_info); if (ret) { @@ -3061,6 +3103,18 @@ int __cold open_ctree(struct super_block *sb, if (ret) goto fail_tree_roots; + /* + * If we have a uuid root and we're not being told to rescan we need to + * check the generation here so we can set the + * BTRFS_FS_UPDATE_UUID_TREE_GEN bit. Otherwise we could commit the + * transaction during a balance or the log replay without updating the + * uuid generation, and then if we crash we would rescan the uuid tree, + * even though it was perfectly fine. + */ + if (fs_info->uuid_root && !btrfs_test_opt(fs_info, RESCAN_UUID_TREE) && + fs_info->generation == btrfs_super_uuid_tree_generation(disk_super)) + set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags); + ret = btrfs_verify_dev_extents(fs_info); if (ret) { btrfs_err(fs_info, @@ -3196,10 +3250,11 @@ int __cold open_ctree(struct super_block *sb, location.type = BTRFS_ROOT_ITEM_KEY; location.offset = 0; - fs_info->fs_root = btrfs_read_fs_root_no_name(fs_info, &location); + fs_info->fs_root = btrfs_get_fs_root(fs_info, &location, true); if (IS_ERR(fs_info->fs_root)) { err = PTR_ERR(fs_info->fs_root); btrfs_warn(fs_info, "failed to read fs tree: %d", err); + fs_info->fs_root = NULL; goto fail_qgroup; } @@ -3284,8 +3339,6 @@ int __cold open_ctree(struct super_block *sb, close_ctree(fs_info); return ret; } - } else { - set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags); } set_bit(BTRFS_FS_OPEN, &fs_info->flags); @@ -3328,90 +3381,78 @@ fail_tree_roots: fail_sb_buffer: btrfs_stop_all_workers(fs_info); btrfs_free_block_groups(fs_info); -fail_csum: - btrfs_free_csum_hash(fs_info); fail_alloc: -fail_iput: btrfs_mapping_tree_free(&fs_info->mapping_tree); iput(fs_info->btree_inode); -fail_bio_counter: - percpu_counter_destroy(&fs_info->dev_replace.bio_counter); -fail_delalloc_bytes: - percpu_counter_destroy(&fs_info->delalloc_bytes); -fail_dirty_metadata_bytes: - percpu_counter_destroy(&fs_info->dirty_metadata_bytes); -fail_dio_bytes: - percpu_counter_destroy(&fs_info->dio_bytes); -fail_srcu: - cleanup_srcu_struct(&fs_info->subvol_srcu); fail: - btrfs_free_stripe_hash_table(fs_info); btrfs_close_devices(fs_info->fs_devices); return err; } ALLOW_ERROR_INJECTION(open_ctree, ERRNO); -static void btrfs_end_buffer_write_sync(struct buffer_head *bh, int uptodate) +static void btrfs_end_super_write(struct bio *bio) { - if (uptodate) { - set_buffer_uptodate(bh); - } else { - struct btrfs_device *device = (struct btrfs_device *) - bh->b_private; - - btrfs_warn_rl_in_rcu(device->fs_info, - "lost page write due to IO error on %s", - rcu_str_deref(device->name)); - /* note, we don't set_buffer_write_io_error because we have - * our own ways of dealing with the IO errors - */ - clear_buffer_uptodate(bh); - btrfs_dev_stat_inc_and_print(device, BTRFS_DEV_STAT_WRITE_ERRS); + struct btrfs_device *device = bio->bi_private; + struct bio_vec *bvec; + struct bvec_iter_all iter_all; + struct page *page; + + bio_for_each_segment_all(bvec, bio, iter_all) { + page = bvec->bv_page; + + if (bio->bi_status) { + btrfs_warn_rl_in_rcu(device->fs_info, + "lost page write due to IO error on %s (%d)", + rcu_str_deref(device->name), + blk_status_to_errno(bio->bi_status)); + ClearPageUptodate(page); + SetPageError(page); + btrfs_dev_stat_inc_and_print(device, + BTRFS_DEV_STAT_WRITE_ERRS); + } else { + SetPageUptodate(page); + } + + put_page(page); + unlock_page(page); } - unlock_buffer(bh); - put_bh(bh); + + bio_put(bio); } -int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num, - struct buffer_head **bh_ret) +struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev, + int copy_num) { - struct buffer_head *bh; struct btrfs_super_block *super; + struct page *page; u64 bytenr; + struct address_space *mapping = bdev->bd_inode->i_mapping; bytenr = btrfs_sb_offset(copy_num); if (bytenr + BTRFS_SUPER_INFO_SIZE >= i_size_read(bdev->bd_inode)) - return -EINVAL; + return ERR_PTR(-EINVAL); - bh = __bread(bdev, bytenr / BTRFS_BDEV_BLOCKSIZE, BTRFS_SUPER_INFO_SIZE); - /* - * If we fail to read from the underlying devices, as of now - * the best option we have is to mark it EIO. - */ - if (!bh) - return -EIO; + page = read_cache_page_gfp(mapping, bytenr >> PAGE_SHIFT, GFP_NOFS); + if (IS_ERR(page)) + return ERR_CAST(page); - super = (struct btrfs_super_block *)bh->b_data; + super = page_address(page); if (btrfs_super_bytenr(super) != bytenr || btrfs_super_magic(super) != BTRFS_MAGIC) { - brelse(bh); - return -EINVAL; + btrfs_release_disk_super(super); + return ERR_PTR(-EINVAL); } - *bh_ret = bh; - return 0; + return super; } -struct buffer_head *btrfs_read_dev_super(struct block_device *bdev) +struct btrfs_super_block *btrfs_read_dev_super(struct block_device *bdev) { - struct buffer_head *bh; - struct buffer_head *latest = NULL; - struct btrfs_super_block *super; + struct btrfs_super_block *super, *latest = NULL; int i; u64 transid = 0; - int ret = -EINVAL; /* we would like to check all the supers, but that would make * a btrfs mount succeed after a mkfs from a different FS. @@ -3419,48 +3460,41 @@ struct buffer_head *btrfs_read_dev_super(struct block_device *bdev) * later supers, using BTRFS_SUPER_MIRROR_MAX instead */ for (i = 0; i < 1; i++) { - ret = btrfs_read_dev_one_super(bdev, i, &bh); - if (ret) + super = btrfs_read_dev_one_super(bdev, i); + if (IS_ERR(super)) continue; - super = (struct btrfs_super_block *)bh->b_data; - if (!latest || btrfs_super_generation(super) > transid) { - brelse(latest); - latest = bh; + if (latest) + btrfs_release_disk_super(super); + + latest = super; transid = btrfs_super_generation(super); - } else { - brelse(bh); } } - if (!latest) - return ERR_PTR(ret); - - return latest; + return super; } /* * Write superblock @sb to the @device. Do not wait for completion, all the - * buffer heads we write are pinned. + * pages we use for writing are locked. * * Write @max_mirrors copies of the superblock, where 0 means default that fit * the expected device size at commit time. Note that max_mirrors must be * same for write and wait phases. * - * Return number of errors when buffer head is not found or submission fails. + * Return number of errors when page is not found or submission fails. */ static int write_dev_supers(struct btrfs_device *device, struct btrfs_super_block *sb, int max_mirrors) { struct btrfs_fs_info *fs_info = device->fs_info; + struct address_space *mapping = device->bdev->bd_inode->i_mapping; SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); - struct buffer_head *bh; int i; - int ret; int errors = 0; u64 bytenr; - int op_flags; if (max_mirrors == 0) max_mirrors = BTRFS_SUPER_MIRROR_MAX; @@ -3468,6 +3502,10 @@ static int write_dev_supers(struct btrfs_device *device, shash->tfm = fs_info->csum_shash; for (i = 0; i < max_mirrors; i++) { + struct page *page; + struct bio *bio; + struct btrfs_super_block *disk_super; + bytenr = btrfs_sb_offset(i); if (bytenr + BTRFS_SUPER_INFO_SIZE >= device->commit_total_bytes) @@ -3480,37 +3518,45 @@ static int write_dev_supers(struct btrfs_device *device, BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE); crypto_shash_final(shash, sb->csum); - /* One reference for us, and we leave it for the caller */ - bh = __getblk(device->bdev, bytenr / BTRFS_BDEV_BLOCKSIZE, - BTRFS_SUPER_INFO_SIZE); - if (!bh) { + page = find_or_create_page(mapping, bytenr >> PAGE_SHIFT, + GFP_NOFS); + if (!page) { btrfs_err(device->fs_info, - "couldn't get super buffer head for bytenr %llu", + "couldn't get super block page for bytenr %llu", bytenr); errors++; continue; } - memcpy(bh->b_data, sb, BTRFS_SUPER_INFO_SIZE); + /* Bump the refcount for wait_dev_supers() */ + get_page(page); - /* one reference for submit_bh */ - get_bh(bh); + disk_super = page_address(page); + memcpy(disk_super, sb, BTRFS_SUPER_INFO_SIZE); - set_buffer_uptodate(bh); - lock_buffer(bh); - bh->b_end_io = btrfs_end_buffer_write_sync; - bh->b_private = device; + /* + * Directly use bios here instead of relying on the page cache + * to do I/O, so we don't lose the ability to do integrity + * checking. + */ + bio = bio_alloc(GFP_NOFS, 1); + bio_set_dev(bio, device->bdev); + bio->bi_iter.bi_sector = bytenr >> SECTOR_SHIFT; + bio->bi_private = device; + bio->bi_end_io = btrfs_end_super_write; + __bio_add_page(bio, page, BTRFS_SUPER_INFO_SIZE, + offset_in_page(bytenr)); /* - * we fua the first super. The others we allow - * to go down lazy. + * We FUA only the first super block. The others we allow to + * go down lazy and there's a short window where the on-disk + * copies might still contain the older version. */ - op_flags = REQ_SYNC | REQ_META | REQ_PRIO; + bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_META | REQ_PRIO; if (i == 0 && !btrfs_test_opt(device->fs_info, NOBARRIER)) - op_flags |= REQ_FUA; - ret = btrfsic_submit_bh(REQ_OP_WRITE, op_flags, bh); - if (ret) - errors++; + bio->bi_opf |= REQ_FUA; + + btrfsic_submit_bio(bio); } return errors < i ? 0 : -1; } @@ -3519,12 +3565,11 @@ static int write_dev_supers(struct btrfs_device *device, * Wait for write completion of superblocks done by write_dev_supers, * @max_mirrors same for write and wait phases. * - * Return number of errors when buffer head is not found or not marked up to + * Return number of errors when page is not found or not marked up to * date. */ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) { - struct buffer_head *bh; int i; int errors = 0; bool primary_failed = false; @@ -3534,32 +3579,34 @@ static int wait_dev_supers(struct btrfs_device *device, int max_mirrors) max_mirrors = BTRFS_SUPER_MIRROR_MAX; for (i = 0; i < max_mirrors; i++) { + struct page *page; + bytenr = btrfs_sb_offset(i); if (bytenr + BTRFS_SUPER_INFO_SIZE >= device->commit_total_bytes) break; - bh = __find_get_block(device->bdev, - bytenr / BTRFS_BDEV_BLOCKSIZE, - BTRFS_SUPER_INFO_SIZE); - if (!bh) { + page = find_get_page(device->bdev->bd_inode->i_mapping, + bytenr >> PAGE_SHIFT); + if (!page) { errors++; if (i == 0) primary_failed = true; continue; } - wait_on_buffer(bh); - if (!buffer_uptodate(bh)) { + /* Page is submitted locked and unlocked once the IO completes */ + wait_on_page_locked(page); + if (PageError(page)) { errors++; if (i == 0) primary_failed = true; } - /* drop our reference */ - brelse(bh); + /* Drop our reference */ + put_page(page); - /* drop the reference from the writing run */ - brelse(bh); + /* Drop the reference from the writing run */ + put_page(page); } /* log error, force error return */ @@ -3831,20 +3878,19 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root) { + bool drop_ref = false; + spin_lock(&fs_info->fs_roots_radix_lock); radix_tree_delete(&fs_info->fs_roots_radix, (unsigned long)root->root_key.objectid); + if (test_and_clear_bit(BTRFS_ROOT_IN_RADIX, &root->state)) + drop_ref = true; spin_unlock(&fs_info->fs_roots_radix_lock); - if (btrfs_root_refs(&root->root_item) == 0) - synchronize_srcu(&fs_info->subvol_srcu); - if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) { btrfs_free_log(NULL, root); if (root->reloc_root) { - free_extent_buffer(root->reloc_root->node); - free_extent_buffer(root->reloc_root->commit_root); - btrfs_put_fs_root(root->reloc_root); + btrfs_put_root(root->reloc_root); root->reloc_root = NULL; } } @@ -3853,22 +3899,12 @@ void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info, __btrfs_remove_free_space_cache(root->free_ino_pinned); if (root->free_ino_ctl) __btrfs_remove_free_space_cache(root->free_ino_ctl); - btrfs_free_fs_root(root); -} - -void btrfs_free_fs_root(struct btrfs_root *root) -{ - iput(root->ino_cache_inode); - WARN_ON(!RB_EMPTY_ROOT(&root->inode_tree)); - if (root->anon_dev) - free_anon_bdev(root->anon_dev); - if (root->subv_writers) - btrfs_free_subvolume_writers(root->subv_writers); - free_extent_buffer(root->node); - free_extent_buffer(root->commit_root); - kfree(root->free_ino_ctl); - kfree(root->free_ino_pinned); - btrfs_put_fs_root(root); + if (root->ino_cache_inode) { + iput(root->ino_cache_inode); + root->ino_cache_inode = NULL; + } + if (drop_ref) + btrfs_put_root(root); } int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info) @@ -3878,15 +3914,14 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info) int i = 0; int err = 0; unsigned int ret = 0; - int index; while (1) { - index = srcu_read_lock(&fs_info->subvol_srcu); + spin_lock(&fs_info->fs_roots_radix_lock); ret = radix_tree_gang_lookup(&fs_info->fs_roots_radix, (void **)gang, root_objectid, ARRAY_SIZE(gang)); if (!ret) { - srcu_read_unlock(&fs_info->subvol_srcu, index); + spin_unlock(&fs_info->fs_roots_radix_lock); break; } root_objectid = gang[ret - 1]->root_key.objectid + 1; @@ -3898,9 +3933,9 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info) continue; } /* grab all the search result for later use */ - gang[i] = btrfs_grab_fs_root(gang[i]); + gang[i] = btrfs_grab_root(gang[i]); } - srcu_read_unlock(&fs_info->subvol_srcu, index); + spin_unlock(&fs_info->fs_roots_radix_lock); for (i = 0; i < ret; i++) { if (!gang[i]) @@ -3909,7 +3944,7 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info) err = btrfs_orphan_cleanup(gang[i]); if (err) break; - btrfs_put_fs_root(gang[i]); + btrfs_put_root(gang[i]); } root_objectid++; } @@ -3917,7 +3952,7 @@ int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info) /* release the uncleaned roots due to error */ for (; i < ret; i++) { if (gang[i]) - btrfs_put_fs_root(gang[i]); + btrfs_put_root(gang[i]); } return err; } @@ -3989,6 +4024,19 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) */ btrfs_delete_unused_bgs(fs_info); + /* + * There might be existing delayed inode workers still running + * and holding an empty delayed inode item. We must wait for + * them to complete first because they can create a transaction. + * This happens when someone calls btrfs_balance_delayed_items() + * and then a transaction commit runs the same delayed nodes + * before any delayed worker has done something with the nodes. + * We must wait for any worker here and not at transaction + * commit time since that could cause a deadlock. + * This is a very rare case. + */ + btrfs_flush_workqueue(fs_info->delayed_workers); + ret = btrfs_commit_super(fs_info); if (ret) btrfs_err(fs_info, "commit super ret %d", ret); @@ -4019,8 +4067,6 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) btrfs_sysfs_remove_mounted(fs_info); btrfs_sysfs_remove_fsid(fs_info->fs_devices); - btrfs_free_fs_roots(fs_info); - btrfs_put_block_group_cache(fs_info); /* @@ -4032,6 +4078,7 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) clear_bit(BTRFS_FS_OPEN, &fs_info->flags); free_root_pointers(fs_info, true); + btrfs_free_fs_roots(fs_info); /* * We must free the block groups after dropping the fs_roots as we could @@ -4051,16 +4098,6 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) btrfs_mapping_tree_free(&fs_info->mapping_tree); btrfs_close_devices(fs_info->fs_devices); - - percpu_counter_destroy(&fs_info->dirty_metadata_bytes); - percpu_counter_destroy(&fs_info->delalloc_bytes); - percpu_counter_destroy(&fs_info->dio_bytes); - percpu_counter_destroy(&fs_info->dev_replace.bio_counter); - cleanup_srcu_struct(&fs_info->subvol_srcu); - - btrfs_free_csum_hash(fs_info); - btrfs_free_stripe_hash_table(fs_info); - btrfs_free_ref_cache(fs_info); } int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid, @@ -4234,7 +4271,7 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, spin_lock(&delayed_refs->lock); if (atomic_read(&delayed_refs->num_entries) == 0) { spin_unlock(&delayed_refs->lock); - btrfs_info(fs_info, "delayed_refs has NO entry"); + btrfs_debug(fs_info, "delayed_refs has NO entry"); return ret; } @@ -4268,14 +4305,36 @@ static int btrfs_destroy_delayed_refs(struct btrfs_transaction *trans, spin_unlock(&delayed_refs->lock); mutex_unlock(&head->mutex); - if (pin_bytes) - btrfs_pin_extent(fs_info, head->bytenr, - head->num_bytes, 1); + if (pin_bytes) { + struct btrfs_block_group *cache; + + cache = btrfs_lookup_block_group(fs_info, head->bytenr); + BUG_ON(!cache); + + spin_lock(&cache->space_info->lock); + spin_lock(&cache->lock); + cache->pinned += head->num_bytes; + btrfs_space_info_update_bytes_pinned(fs_info, + cache->space_info, head->num_bytes); + cache->reserved -= head->num_bytes; + cache->space_info->bytes_reserved -= head->num_bytes; + spin_unlock(&cache->lock); + spin_unlock(&cache->space_info->lock); + percpu_counter_add_batch( + &cache->space_info->total_bytes_pinned, + head->num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH); + + btrfs_put_block_group(cache); + + btrfs_error_unpin_extent_range(fs_info, head->bytenr, + head->bytenr + head->num_bytes - 1); + } btrfs_cleanup_ref_head_accounting(fs_info, delayed_refs, head); btrfs_put_delayed_ref_head(head); cond_resched(); spin_lock(&delayed_refs->lock); } + btrfs_qgroup_destroy_extent_records(trans); spin_unlock(&delayed_refs->lock); @@ -4325,12 +4384,12 @@ static void btrfs_destroy_all_delalloc_inodes(struct btrfs_fs_info *fs_info) while (!list_empty(&splice)) { root = list_first_entry(&splice, struct btrfs_root, delalloc_root); - root = btrfs_grab_fs_root(root); + root = btrfs_grab_root(root); BUG_ON(!root); spin_unlock(&fs_info->delalloc_root_lock); btrfs_destroy_delalloc_inodes(root); - btrfs_put_fs_root(root); + btrfs_put_root(root); spin_lock(&fs_info->delalloc_root_lock); } @@ -4371,16 +4430,12 @@ static int btrfs_destroy_marked_extents(struct btrfs_fs_info *fs_info, } static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info, - struct extent_io_tree *pinned_extents) + struct extent_io_tree *unpin) { - struct extent_io_tree *unpin; u64 start; u64 end; int ret; - bool loop = true; - unpin = pinned_extents; -again: while (1) { struct extent_state *cached_state = NULL; @@ -4405,15 +4460,6 @@ again: cond_resched(); } - if (loop) { - if (unpin == &fs_info->freed_extents[0]) - unpin = &fs_info->freed_extents[1]; - else - unpin = &fs_info->freed_extents[0]; - loop = false; - goto again; - } - return 0; } @@ -4501,12 +4547,10 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans, wake_up(&fs_info->transaction_wait); btrfs_destroy_delayed_inodes(fs_info); - btrfs_assert_delayed_root_empty(fs_info); btrfs_destroy_marked_extents(fs_info, &cur_trans->dirty_pages, EXTENT_DIRTY); - btrfs_destroy_pinned_extent(fs_info, - fs_info->pinned_extents); + btrfs_destroy_pinned_extent(fs_info, &cur_trans->pinned_extents); cur_trans->state =TRANS_STATE_COMPLETED; wake_up(&cur_trans->commit_wait); @@ -4558,7 +4602,6 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info) btrfs_destroy_all_ordered_extents(fs_info); btrfs_destroy_delayed_inodes(fs_info); btrfs_assert_delayed_root_empty(fs_info); - btrfs_destroy_pinned_extent(fs_info, fs_info->pinned_extents); btrfs_destroy_all_delalloc_inodes(fs_info); mutex_unlock(&fs_info->transaction_kthread_mutex); diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index 8c2d6cf1ce59..cd629113f61c 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -39,6 +39,8 @@ static inline u64 btrfs_sb_offset(int mirror) struct btrfs_device; struct btrfs_fs_devices; +void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info); +void btrfs_init_fs_info(struct btrfs_fs_info *fs_info); int btrfs_verify_level_key(struct extent_buffer *eb, int level, struct btrfs_key *first_key, u64 parent_transid); struct extent_buffer *read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr, @@ -54,15 +56,12 @@ int __cold open_ctree(struct super_block *sb, char *options); void __cold close_ctree(struct btrfs_fs_info *fs_info); int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors); -struct buffer_head *btrfs_read_dev_super(struct block_device *bdev); -int btrfs_read_dev_one_super(struct block_device *bdev, int copy_num, - struct buffer_head **bh_ret); +struct btrfs_super_block *btrfs_read_dev_super(struct block_device *bdev); +struct btrfs_super_block *btrfs_read_dev_one_super(struct block_device *bdev, + int copy_num); int btrfs_commit_super(struct btrfs_fs_info *fs_info); -struct btrfs_root *btrfs_read_fs_root(struct btrfs_root *tree_root, - struct btrfs_key *location); -int btrfs_init_fs_root(struct btrfs_root *root); -struct btrfs_root *btrfs_lookup_fs_root(struct btrfs_fs_info *fs_info, - u64 root_id); +struct btrfs_root *btrfs_read_tree_root(struct btrfs_root *tree_root, + struct btrfs_key *key); int btrfs_insert_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root); void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info); @@ -70,19 +69,13 @@ void btrfs_free_fs_roots(struct btrfs_fs_info *fs_info); struct btrfs_root *btrfs_get_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_key *key, bool check_ref); -static inline struct btrfs_root * -btrfs_read_fs_root_no_name(struct btrfs_fs_info *fs_info, - struct btrfs_key *location) -{ - return btrfs_get_fs_root(fs_info, location, true); -} +void btrfs_free_fs_info(struct btrfs_fs_info *fs_info); int btrfs_cleanup_fs_roots(struct btrfs_fs_info *fs_info); void btrfs_btree_balance_dirty(struct btrfs_fs_info *fs_info); void btrfs_btree_balance_dirty_nodelay(struct btrfs_fs_info *fs_info); void btrfs_drop_and_free_fs_root(struct btrfs_fs_info *fs_info, struct btrfs_root *root); -void btrfs_free_fs_root(struct btrfs_root *root); #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info); @@ -95,19 +88,16 @@ struct btrfs_root *btrfs_alloc_dummy_root(struct btrfs_fs_info *fs_info); * If you want to ensure the whole tree is safe, you should use * fs_info->subvol_srcu */ -static inline struct btrfs_root *btrfs_grab_fs_root(struct btrfs_root *root) +static inline struct btrfs_root *btrfs_grab_root(struct btrfs_root *root) { + if (!root) + return NULL; if (refcount_inc_not_zero(&root->refs)) return root; return NULL; } -static inline void btrfs_put_fs_root(struct btrfs_root *root) -{ - if (refcount_dec_and_test(&root->refs)) - kfree(root); -} - +void btrfs_put_root(struct btrfs_root *root); void btrfs_mark_buffer_dirty(struct extent_buffer *buf); int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid, int atomic); diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c index 72e312cae69d..2bb25d2dc44b 100644 --- a/fs/btrfs/export.c +++ b/fs/btrfs/export.c @@ -57,16 +57,14 @@ static int btrfs_encode_fh(struct inode *inode, u32 *fh, int *max_len, return type; } -static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid, - u64 root_objectid, u32 generation, - int check_generation) +struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid, + u64 root_objectid, u32 generation, + int check_generation) { struct btrfs_fs_info *fs_info = btrfs_sb(sb); struct btrfs_root *root; struct inode *inode; struct btrfs_key key; - int index; - int err = 0; if (objectid < BTRFS_FIRST_FREE_OBJECTID) return ERR_PTR(-ESTALE); @@ -75,25 +73,18 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid, key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - index = srcu_read_lock(&fs_info->subvol_srcu); - - root = btrfs_read_fs_root_no_name(fs_info, &key); - if (IS_ERR(root)) { - err = PTR_ERR(root); - goto fail; - } + root = btrfs_get_fs_root(fs_info, &key, true); + if (IS_ERR(root)) + return ERR_CAST(root); key.objectid = objectid; key.type = BTRFS_INODE_ITEM_KEY; key.offset = 0; inode = btrfs_iget(sb, &key, root); - if (IS_ERR(inode)) { - err = PTR_ERR(inode); - goto fail; - } - - srcu_read_unlock(&fs_info->subvol_srcu, index); + btrfs_put_root(root); + if (IS_ERR(inode)) + return ERR_CAST(inode); if (check_generation && generation != inode->i_generation) { iput(inode); @@ -101,9 +92,6 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid, } return d_obtain_alias(inode); -fail: - srcu_read_unlock(&fs_info->subvol_srcu, index); - return ERR_PTR(err); } static struct dentry *btrfs_fh_to_parent(struct super_block *sb, struct fid *fh, @@ -152,7 +140,7 @@ static struct dentry *btrfs_fh_to_dentry(struct super_block *sb, struct fid *fh, return btrfs_get_dentry(sb, objectid, root_objectid, generation, 1); } -static struct dentry *btrfs_get_parent(struct dentry *child) +struct dentry *btrfs_get_parent(struct dentry *child) { struct inode *dir = d_inode(child); struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); diff --git a/fs/btrfs/export.h b/fs/btrfs/export.h index 57488ecd7d4e..f32f4113c976 100644 --- a/fs/btrfs/export.h +++ b/fs/btrfs/export.h @@ -18,4 +18,9 @@ struct btrfs_fid { u64 parent_root_objectid; } __attribute__ ((packed)); +struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid, + u64 root_objectid, u32 generation, + int check_generation); +struct dentry *btrfs_get_parent(struct dentry *child); + #endif diff --git a/fs/btrfs/extent-io-tree.h b/fs/btrfs/extent-io-tree.h index a3febe746c79..b4a7bad3e82e 100644 --- a/fs/btrfs/extent-io-tree.h +++ b/fs/btrfs/extent-io-tree.h @@ -36,13 +36,14 @@ struct io_failure_record; #define CHUNK_TRIMMED EXTENT_DEFRAG enum { - IO_TREE_FS_INFO_FREED_EXTENTS0, - IO_TREE_FS_INFO_FREED_EXTENTS1, + IO_TREE_FS_PINNED_EXTENTS, + IO_TREE_FS_EXCLUDED_EXTENTS, IO_TREE_INODE_IO, IO_TREE_INODE_IO_FAILURE, IO_TREE_RELOC_BLOCKS, IO_TREE_TRANS_DIRTY_PAGES, IO_TREE_ROOT_DIRTY_LOG_PAGES, + IO_TREE_INODE_FILE_EXTENT, IO_TREE_SELFTEST, }; @@ -222,6 +223,8 @@ int find_first_extent_bit(struct extent_io_tree *tree, u64 start, struct extent_state **cached_state); void find_first_clear_extent_bit(struct extent_io_tree *tree, u64 start, u64 *start_ret, u64 *end_ret, unsigned bits); +int find_contiguous_extent_bit(struct extent_io_tree *tree, u64 start, + u64 *start_ret, u64 *end_ret, unsigned bits); int extent_invalidatepage(struct extent_io_tree *tree, struct page *page, unsigned long offset); bool btrfs_find_delalloc_range(struct extent_io_tree *tree, u64 *start, diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 0163fdd59f8f..54a64d1e18c6 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -64,10 +64,8 @@ int btrfs_add_excluded_extent(struct btrfs_fs_info *fs_info, u64 start, u64 num_bytes) { u64 end = start + num_bytes - 1; - set_extent_bits(&fs_info->freed_extents[0], - start, end, EXTENT_UPTODATE); - set_extent_bits(&fs_info->freed_extents[1], - start, end, EXTENT_UPTODATE); + set_extent_bits(&fs_info->excluded_extents, start, end, + EXTENT_UPTODATE); return 0; } @@ -79,10 +77,8 @@ void btrfs_free_excluded_extents(struct btrfs_block_group *cache) start = cache->start; end = start + cache->length - 1; - clear_extent_bits(&fs_info->freed_extents[0], - start, end, EXTENT_UPTODATE); - clear_extent_bits(&fs_info->freed_extents[1], - start, end, EXTENT_UPTODATE); + clear_extent_bits(&fs_info->excluded_extents, start, end, + EXTENT_UPTODATE); } static u64 generic_ref_to_space_flags(struct btrfs_ref *ref) @@ -1193,24 +1189,6 @@ int insert_inline_extent_backref(struct btrfs_trans_handle *trans, return ret; } -static int insert_extent_backref(struct btrfs_trans_handle *trans, - struct btrfs_path *path, - u64 bytenr, u64 parent, u64 root_objectid, - u64 owner, u64 offset, int refs_to_add) -{ - int ret; - if (owner < BTRFS_FIRST_FREE_OBJECTID) { - BUG_ON(refs_to_add != 1); - ret = insert_tree_block_ref(trans, path, bytenr, parent, - root_objectid); - } else { - ret = insert_extent_data_ref(trans, path, bytenr, parent, - root_objectid, owner, offset, - refs_to_add); - } - return ret; -} - static int remove_extent_backref(struct btrfs_trans_handle *trans, struct btrfs_path *path, struct btrfs_extent_inline_ref *iref, @@ -1469,7 +1447,6 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, if (!path) return -ENOMEM; - path->reada = READA_FORWARD; path->leave_spinning = 1; /* this will setup the path even if it fails to insert the back ref */ ret = insert_inline_extent_backref(trans, path, bytenr, num_bytes, @@ -1494,11 +1471,17 @@ static int __btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); - path->reada = READA_FORWARD; path->leave_spinning = 1; /* now insert the actual backref */ - ret = insert_extent_backref(trans, path, bytenr, parent, root_objectid, - owner, offset, refs_to_add); + if (owner < BTRFS_FIRST_FREE_OBJECTID) { + BUG_ON(refs_to_add != 1); + ret = insert_tree_block_ref(trans, path, bytenr, parent, + root_objectid); + } else { + ret = insert_extent_data_ref(trans, path, bytenr, parent, + root_objectid, owner, offset, + refs_to_add); + } if (ret) btrfs_abort_transaction(trans, ret); out: @@ -1583,7 +1566,7 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans, int err = 0; int metadata = !extent_op->is_data; - if (trans->aborted) + if (TRANS_ABORTED(trans)) return 0; if (metadata && !btrfs_fs_incompat(fs_info, SKINNY_METADATA)) @@ -1604,7 +1587,6 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans, } again: - path->reada = READA_FORWARD; path->leave_spinning = 1; ret = btrfs_search_slot(trans, fs_info->extent_root, &key, path, 0, 1); if (ret < 0) { @@ -1703,10 +1685,9 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans, { int ret = 0; - if (trans->aborted) { + if (TRANS_ABORTED(trans)) { if (insert_reserved) - btrfs_pin_extent(trans->fs_info, node->bytenr, - node->num_bytes, 1); + btrfs_pin_extent(trans, node->bytenr, node->num_bytes, 1); return 0; } @@ -1721,8 +1702,7 @@ static int run_one_delayed_ref(struct btrfs_trans_handle *trans, else BUG(); if (ret && insert_reserved) - btrfs_pin_extent(trans->fs_info, node->bytenr, - node->num_bytes, 1); + btrfs_pin_extent(trans, node->bytenr, node->num_bytes, 1); return ret; } @@ -1867,8 +1847,7 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans, spin_unlock(&delayed_refs->lock); if (head->must_insert_reserved) { - btrfs_pin_extent(fs_info, head->bytenr, - head->num_bytes, 1); + btrfs_pin_extent(trans, head->bytenr, head->num_bytes, 1); if (head->is_data) { ret = btrfs_del_csums(trans, fs_info->csum_root, head->bytenr, head->num_bytes); @@ -2191,7 +2170,7 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, int run_all = count == (unsigned long)-1; /* We'll clean this up in btrfs_cleanup_transaction */ - if (trans->aborted) + if (TRANS_ABORTED(trans)) return 0; if (test_bit(BTRFS_FS_CREATING_FREE_SPACE_TREE, &fs_info->flags)) @@ -2238,7 +2217,7 @@ out: } int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans, - u64 bytenr, u64 num_bytes, u64 flags, + struct extent_buffer *eb, u64 flags, int level, int is_data) { struct btrfs_delayed_extent_op *extent_op; @@ -2254,7 +2233,7 @@ int btrfs_set_disk_extent_flags(struct btrfs_trans_handle *trans, extent_op->is_data = is_data ? true : false; extent_op->level = level; - ret = btrfs_add_delayed_extent_op(trans, bytenr, num_bytes, extent_op); + ret = btrfs_add_delayed_extent_op(trans, eb->start, eb->len, extent_op); if (ret) btrfs_free_delayed_extent_op(extent_op); return ret; @@ -2588,7 +2567,8 @@ static u64 first_logical_byte(struct btrfs_fs_info *fs_info, u64 search_start) return bytenr; } -static int pin_down_extent(struct btrfs_block_group *cache, +static int pin_down_extent(struct btrfs_trans_handle *trans, + struct btrfs_block_group *cache, u64 bytenr, u64 num_bytes, int reserved) { struct btrfs_fs_info *fs_info = cache->fs_info; @@ -2607,22 +2587,20 @@ static int pin_down_extent(struct btrfs_block_group *cache, percpu_counter_add_batch(&cache->space_info->total_bytes_pinned, num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH); - set_extent_dirty(fs_info->pinned_extents, bytenr, + set_extent_dirty(&trans->transaction->pinned_extents, bytenr, bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL); return 0; } -int btrfs_pin_extent(struct btrfs_fs_info *fs_info, +int btrfs_pin_extent(struct btrfs_trans_handle *trans, u64 bytenr, u64 num_bytes, int reserved) { struct btrfs_block_group *cache; - ASSERT(fs_info->running_transaction); - - cache = btrfs_lookup_block_group(fs_info, bytenr); + cache = btrfs_lookup_block_group(trans->fs_info, bytenr); BUG_ON(!cache); /* Logic error */ - pin_down_extent(cache, bytenr, num_bytes, reserved); + pin_down_extent(trans, cache, bytenr, num_bytes, reserved); btrfs_put_block_group(cache); return 0; @@ -2631,13 +2609,15 @@ int btrfs_pin_extent(struct btrfs_fs_info *fs_info, /* * this function must be called within transaction */ -int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info, +int btrfs_pin_extent_for_log_replay(struct btrfs_trans_handle *trans, u64 bytenr, u64 num_bytes) { struct btrfs_block_group *cache; int ret; - cache = btrfs_lookup_block_group(fs_info, bytenr); + btrfs_add_excluded_extent(trans->fs_info, bytenr, num_bytes); + + cache = btrfs_lookup_block_group(trans->fs_info, bytenr); if (!cache) return -EINVAL; @@ -2649,7 +2629,7 @@ int btrfs_pin_extent_for_log_replay(struct btrfs_fs_info *fs_info, */ btrfs_cache_block_group(cache, 1); - pin_down_extent(cache, bytenr, num_bytes, 0); + pin_down_extent(trans, cache, bytenr, num_bytes, 0); /* remove us from the free space cache (if we're there at all) */ ret = btrfs_remove_free_space(cache, bytenr, num_bytes); @@ -2763,11 +2743,6 @@ void btrfs_prepare_extent_commit(struct btrfs_fs_info *fs_info) } } - if (fs_info->pinned_extents == &fs_info->freed_extents[0]) - fs_info->pinned_extents = &fs_info->freed_extents[1]; - else - fs_info->pinned_extents = &fs_info->freed_extents[0]; - up_write(&fs_info->commit_root_sem); btrfs_update_global_block_rsv(fs_info); @@ -2908,12 +2883,9 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans) u64 end; int ret; - if (fs_info->pinned_extents == &fs_info->freed_extents[0]) - unpin = &fs_info->freed_extents[1]; - else - unpin = &fs_info->freed_extents[0]; + unpin = &trans->transaction->pinned_extents; - while (!trans->aborted) { + while (!TRANS_ABORTED(trans)) { struct extent_state *cached_state = NULL; mutex_lock(&fs_info->unused_bg_unpin_mutex); @@ -2923,6 +2895,9 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans) mutex_unlock(&fs_info->unused_bg_unpin_mutex); break; } + if (test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags)) + clear_extent_bits(&fs_info->excluded_extents, start, + end, EXTENT_UPTODATE); if (btrfs_test_opt(fs_info, DISCARD_SYNC)) ret = btrfs_discard_extent(fs_info, start, @@ -2950,7 +2925,7 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans) u64 trimmed = 0; ret = -EROFS; - if (!trans->aborted) + if (!TRANS_ABORTED(trans)) ret = btrfs_discard_extent(fs_info, block_group->start, block_group->length, @@ -3000,7 +2975,6 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, if (!path) return -ENOMEM; - path->reada = READA_FORWARD; path->leave_spinning = 1; is_data = owner_objectid >= BTRFS_FIRST_FREE_OBJECTID; @@ -3301,7 +3275,7 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, cache = btrfs_lookup_block_group(fs_info, buf->start); if (btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) { - pin_down_extent(cache, buf->start, buf->len, 1); + pin_down_extent(trans, cache, buf->start, buf->len, 1); btrfs_put_block_group(cache); goto out; } @@ -3345,7 +3319,7 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref) (ref->type == BTRFS_REF_DATA && ref->data_ref.ref_root == BTRFS_TREE_LOG_OBJECTID)) { /* unlocks the pinned mutex */ - btrfs_pin_extent(fs_info, ref->bytenr, ref->len, 1); + btrfs_pin_extent(trans, ref->bytenr, ref->len, 1); old_ref_mod = new_ref_mod = 0; ret = 0; } else if (ref->type == BTRFS_REF_METADATA) { @@ -3438,6 +3412,10 @@ btrfs_release_block_group(struct btrfs_block_group *cache, btrfs_put_block_group(cache); } +enum btrfs_extent_allocation_policy { + BTRFS_EXTENT_ALLOC_CLUSTERED, +}; + /* * Structure used internally for find_free_extent() function. Wraps needed * parameters. @@ -3454,6 +3432,8 @@ struct find_free_extent_ctl { /* For clustered allocation */ u64 empty_cluster; + struct btrfs_free_cluster *last_ptr; + bool use_cluster; bool have_caching_bg; bool orig_have_caching_bg; @@ -3489,6 +3469,12 @@ struct find_free_extent_ctl { /* Found result */ u64 found_offset; + + /* Hint where to start looking for an empty space */ + u64 hint_byte; + + /* Allocation policy */ + enum btrfs_extent_allocation_policy policy; }; @@ -3501,11 +3487,11 @@ struct find_free_extent_ctl { * Return 0 means we have found a location and set ffe_ctl->found_offset. */ static int find_free_extent_clustered(struct btrfs_block_group *bg, - struct btrfs_free_cluster *last_ptr, - struct find_free_extent_ctl *ffe_ctl, - struct btrfs_block_group **cluster_bg_ret) + struct find_free_extent_ctl *ffe_ctl, + struct btrfs_block_group **cluster_bg_ret) { struct btrfs_block_group *cluster_bg; + struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr; u64 aligned_cluster; u64 offset; int ret; @@ -3605,9 +3591,9 @@ refill_cluster: * Return -EAGAIN to inform caller that we need to re-search this block group */ static int find_free_extent_unclustered(struct btrfs_block_group *bg, - struct btrfs_free_cluster *last_ptr, - struct find_free_extent_ctl *ffe_ctl) + struct find_free_extent_ctl *ffe_ctl) { + struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr; u64 offset; /* @@ -3663,16 +3649,101 @@ static int find_free_extent_unclustered(struct btrfs_block_group *bg, return 0; } +static int do_allocation_clustered(struct btrfs_block_group *block_group, + struct find_free_extent_ctl *ffe_ctl, + struct btrfs_block_group **bg_ret) +{ + int ret; + + /* We want to try and use the cluster allocator, so lets look there */ + if (ffe_ctl->last_ptr && ffe_ctl->use_cluster) { + ret = find_free_extent_clustered(block_group, ffe_ctl, bg_ret); + if (ret >= 0 || ret == -EAGAIN) + return ret; + /* ret == -ENOENT case falls through */ + } + + return find_free_extent_unclustered(block_group, ffe_ctl); +} + +static int do_allocation(struct btrfs_block_group *block_group, + struct find_free_extent_ctl *ffe_ctl, + struct btrfs_block_group **bg_ret) +{ + switch (ffe_ctl->policy) { + case BTRFS_EXTENT_ALLOC_CLUSTERED: + return do_allocation_clustered(block_group, ffe_ctl, bg_ret); + default: + BUG(); + } +} + +static void release_block_group(struct btrfs_block_group *block_group, + struct find_free_extent_ctl *ffe_ctl, + int delalloc) +{ + switch (ffe_ctl->policy) { + case BTRFS_EXTENT_ALLOC_CLUSTERED: + ffe_ctl->retry_clustered = false; + ffe_ctl->retry_unclustered = false; + break; + default: + BUG(); + } + + BUG_ON(btrfs_bg_flags_to_raid_index(block_group->flags) != + ffe_ctl->index); + btrfs_release_block_group(block_group, delalloc); +} + +static void found_extent_clustered(struct find_free_extent_ctl *ffe_ctl, + struct btrfs_key *ins) +{ + struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr; + + if (!ffe_ctl->use_cluster && last_ptr) { + spin_lock(&last_ptr->lock); + last_ptr->window_start = ins->objectid; + spin_unlock(&last_ptr->lock); + } +} + +static void found_extent(struct find_free_extent_ctl *ffe_ctl, + struct btrfs_key *ins) +{ + switch (ffe_ctl->policy) { + case BTRFS_EXTENT_ALLOC_CLUSTERED: + found_extent_clustered(ffe_ctl, ins); + break; + default: + BUG(); + } +} + +static int chunk_allocation_failed(struct find_free_extent_ctl *ffe_ctl) +{ + switch (ffe_ctl->policy) { + case BTRFS_EXTENT_ALLOC_CLUSTERED: + /* + * If we can't allocate a new chunk we've already looped through + * at least once, move on to the NO_EMPTY_SIZE case. + */ + ffe_ctl->loop = LOOP_NO_EMPTY_SIZE; + return 0; + default: + BUG(); + } +} + /* * Return >0 means caller needs to re-search for free extent * Return 0 means we have the needed free extent. * Return <0 means we failed to locate any free extent. */ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info, - struct btrfs_free_cluster *last_ptr, struct btrfs_key *ins, struct find_free_extent_ctl *ffe_ctl, - int full_search, bool use_cluster) + bool full_search) { struct btrfs_root *root = fs_info->extent_root; int ret; @@ -3689,11 +3760,7 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info, return 1; if (ins->objectid) { - if (!use_cluster && last_ptr) { - spin_lock(&last_ptr->lock); - last_ptr->window_start = ins->objectid; - spin_unlock(&last_ptr->lock); - } + found_extent(ffe_ctl, ins); return 0; } @@ -3739,16 +3806,10 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info, ret = btrfs_chunk_alloc(trans, ffe_ctl->flags, CHUNK_ALLOC_FORCE); - /* - * If we can't allocate a new chunk we've already looped - * through at least once, move on to the NO_EMPTY_SIZE - * case. - */ - if (ret == -ENOSPC) - ffe_ctl->loop = LOOP_NO_EMPTY_SIZE; - /* Do not bail out on ENOSPC since we can do more. */ - if (ret < 0 && ret != -ENOSPC) + if (ret == -ENOSPC) + ret = chunk_allocation_failed(ffe_ctl); + else if (ret < 0) btrfs_abort_transaction(trans, ret); else ret = 0; @@ -3759,6 +3820,9 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info, } if (ffe_ctl->loop == LOOP_NO_EMPTY_SIZE) { + if (ffe_ctl->policy != BTRFS_EXTENT_ALLOC_CLUSTERED) + return -ENOSPC; + /* * Don't loop again if we already have no empty_size and * no empty_cluster. @@ -3774,6 +3838,71 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info, return -ENOSPC; } +static int prepare_allocation_clustered(struct btrfs_fs_info *fs_info, + struct find_free_extent_ctl *ffe_ctl, + struct btrfs_space_info *space_info, + struct btrfs_key *ins) +{ + /* + * If our free space is heavily fragmented we may not be able to make + * big contiguous allocations, so instead of doing the expensive search + * for free space, simply return ENOSPC with our max_extent_size so we + * can go ahead and search for a more manageable chunk. + * + * If our max_extent_size is large enough for our allocation simply + * disable clustering since we will likely not be able to find enough + * space to create a cluster and induce latency trying. + */ + if (space_info->max_extent_size) { + spin_lock(&space_info->lock); + if (space_info->max_extent_size && + ffe_ctl->num_bytes > space_info->max_extent_size) { + ins->offset = space_info->max_extent_size; + spin_unlock(&space_info->lock); + return -ENOSPC; + } else if (space_info->max_extent_size) { + ffe_ctl->use_cluster = false; + } + spin_unlock(&space_info->lock); + } + + ffe_ctl->last_ptr = fetch_cluster_info(fs_info, space_info, + &ffe_ctl->empty_cluster); + if (ffe_ctl->last_ptr) { + struct btrfs_free_cluster *last_ptr = ffe_ctl->last_ptr; + + spin_lock(&last_ptr->lock); + if (last_ptr->block_group) + ffe_ctl->hint_byte = last_ptr->window_start; + if (last_ptr->fragmented) { + /* + * We still set window_start so we can keep track of the + * last place we found an allocation to try and save + * some time. + */ + ffe_ctl->hint_byte = last_ptr->window_start; + ffe_ctl->use_cluster = false; + } + spin_unlock(&last_ptr->lock); + } + + return 0; +} + +static int prepare_allocation(struct btrfs_fs_info *fs_info, + struct find_free_extent_ctl *ffe_ctl, + struct btrfs_space_info *space_info, + struct btrfs_key *ins) +{ + switch (ffe_ctl->policy) { + case BTRFS_EXTENT_ALLOC_CLUSTERED: + return prepare_allocation_clustered(fs_info, ffe_ctl, + space_info, ins); + default: + BUG(); + } +} + /* * walks the btree of allocated extents and find a hole of a given size. * The key ins is changed to record the hole: @@ -3801,16 +3930,14 @@ static int find_free_extent_update_loop(struct btrfs_fs_info *fs_info, */ static noinline int find_free_extent(struct btrfs_fs_info *fs_info, u64 ram_bytes, u64 num_bytes, u64 empty_size, - u64 hint_byte, struct btrfs_key *ins, + u64 hint_byte_orig, struct btrfs_key *ins, u64 flags, int delalloc) { int ret = 0; int cache_block_group_error = 0; - struct btrfs_free_cluster *last_ptr = NULL; struct btrfs_block_group *block_group = NULL; struct find_free_extent_ctl ffe_ctl = {0}; struct btrfs_space_info *space_info; - bool use_cluster = true; bool full_search = false; WARN_ON(num_bytes < fs_info->sectorsize); @@ -3819,13 +3946,19 @@ static noinline int find_free_extent(struct btrfs_fs_info *fs_info, ffe_ctl.empty_size = empty_size; ffe_ctl.flags = flags; ffe_ctl.search_start = 0; - ffe_ctl.retry_clustered = false; - ffe_ctl.retry_unclustered = false; ffe_ctl.delalloc = delalloc; ffe_ctl.index = btrfs_bg_flags_to_raid_index(flags); ffe_ctl.have_caching_bg = false; ffe_ctl.orig_have_caching_bg = false; ffe_ctl.found_offset = 0; + ffe_ctl.hint_byte = hint_byte_orig; + ffe_ctl.policy = BTRFS_EXTENT_ALLOC_CLUSTERED; + + /* For clustered allocation */ + ffe_ctl.retry_clustered = false; + ffe_ctl.retry_unclustered = false; + ffe_ctl.last_ptr = NULL; + ffe_ctl.use_cluster = true; ins->type = BTRFS_EXTENT_ITEM_KEY; ins->objectid = 0; @@ -3839,51 +3972,14 @@ static noinline int find_free_extent(struct btrfs_fs_info *fs_info, return -ENOSPC; } - /* - * If our free space is heavily fragmented we may not be able to make - * big contiguous allocations, so instead of doing the expensive search - * for free space, simply return ENOSPC with our max_extent_size so we - * can go ahead and search for a more manageable chunk. - * - * If our max_extent_size is large enough for our allocation simply - * disable clustering since we will likely not be able to find enough - * space to create a cluster and induce latency trying. - */ - if (unlikely(space_info->max_extent_size)) { - spin_lock(&space_info->lock); - if (space_info->max_extent_size && - num_bytes > space_info->max_extent_size) { - ins->offset = space_info->max_extent_size; - spin_unlock(&space_info->lock); - return -ENOSPC; - } else if (space_info->max_extent_size) { - use_cluster = false; - } - spin_unlock(&space_info->lock); - } - - last_ptr = fetch_cluster_info(fs_info, space_info, - &ffe_ctl.empty_cluster); - if (last_ptr) { - spin_lock(&last_ptr->lock); - if (last_ptr->block_group) - hint_byte = last_ptr->window_start; - if (last_ptr->fragmented) { - /* - * We still set window_start so we can keep track of the - * last place we found an allocation to try and save - * some time. - */ - hint_byte = last_ptr->window_start; - use_cluster = false; - } - spin_unlock(&last_ptr->lock); - } + ret = prepare_allocation(fs_info, &ffe_ctl, space_info, ins); + if (ret < 0) + return ret; ffe_ctl.search_start = max(ffe_ctl.search_start, first_logical_byte(fs_info, 0)); - ffe_ctl.search_start = max(ffe_ctl.search_start, hint_byte); - if (ffe_ctl.search_start == hint_byte) { + ffe_ctl.search_start = max(ffe_ctl.search_start, ffe_ctl.hint_byte); + if (ffe_ctl.search_start == ffe_ctl.hint_byte) { block_group = btrfs_lookup_block_group(fs_info, ffe_ctl.search_start); /* @@ -3924,6 +4020,8 @@ search: down_read(&space_info->groups_sem); list_for_each_entry(block_group, &space_info->block_groups[ffe_ctl.index], list) { + struct btrfs_block_group *bg_ret; + /* If the block group is read-only, we can skip it entirely. */ if (unlikely(block_group->ro)) continue; @@ -3984,39 +4082,20 @@ have_block_group: if (unlikely(block_group->cached == BTRFS_CACHE_ERROR)) goto loop; - /* - * Ok we want to try and use the cluster allocator, so - * lets look there - */ - if (last_ptr && use_cluster) { - struct btrfs_block_group *cluster_bg = NULL; - - ret = find_free_extent_clustered(block_group, last_ptr, - &ffe_ctl, &cluster_bg); - - if (ret == 0) { - if (cluster_bg && cluster_bg != block_group) { - btrfs_release_block_group(block_group, - delalloc); - block_group = cluster_bg; - } - goto checks; - } else if (ret == -EAGAIN) { - goto have_block_group; - } else if (ret > 0) { - goto loop; + bg_ret = NULL; + ret = do_allocation(block_group, &ffe_ctl, &bg_ret); + if (ret == 0) { + if (bg_ret && bg_ret != block_group) { + btrfs_release_block_group(block_group, delalloc); + block_group = bg_ret; } - /* ret == -ENOENT case falls through */ - } - - ret = find_free_extent_unclustered(block_group, last_ptr, - &ffe_ctl); - if (ret == -EAGAIN) + } else if (ret == -EAGAIN) { goto have_block_group; - else if (ret > 0) + } else if (ret > 0) { goto loop; - /* ret == 0 case falls through */ -checks: + } + + /* Checks */ ffe_ctl.search_start = round_up(ffe_ctl.found_offset, fs_info->stripesize); @@ -4050,17 +4129,12 @@ checks: btrfs_release_block_group(block_group, delalloc); break; loop: - ffe_ctl.retry_clustered = false; - ffe_ctl.retry_unclustered = false; - BUG_ON(btrfs_bg_flags_to_raid_index(block_group->flags) != - ffe_ctl.index); - btrfs_release_block_group(block_group, delalloc); + release_block_group(block_group, &ffe_ctl, delalloc); cond_resched(); } up_read(&space_info->groups_sem); - ret = find_free_extent_update_loop(fs_info, last_ptr, ins, &ffe_ctl, - full_search, use_cluster); + ret = find_free_extent_update_loop(fs_info, ins, &ffe_ctl, full_search); if (ret > 0) goto search; @@ -4189,18 +4263,20 @@ int btrfs_free_reserved_extent(struct btrfs_fs_info *fs_info, return 0; } -int btrfs_pin_reserved_extent(struct btrfs_fs_info *fs_info, u64 start, u64 len) +int btrfs_pin_reserved_extent(struct btrfs_trans_handle *trans, u64 start, + u64 len) { struct btrfs_block_group *cache; int ret = 0; - cache = btrfs_lookup_block_group(fs_info, start); + cache = btrfs_lookup_block_group(trans->fs_info, start); if (!cache) { - btrfs_err(fs_info, "unable to find block group for %llu", start); + btrfs_err(trans->fs_info, "unable to find block group for %llu", + start); return -ENOSPC; } - ret = pin_down_extent(cache, start, len, 1); + ret = pin_down_extent(trans, cache, start, len, 1); btrfs_put_block_group(cache); return ret; } @@ -4430,6 +4506,8 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans, ret = alloc_reserved_file_extent(trans, 0, root_objectid, 0, owner, offset, ins, 1); + if (ret) + btrfs_pin_extent(trans, ins->objectid, ins->offset, 1); btrfs_put_block_group(block_group); return ret; } @@ -4748,8 +4826,7 @@ static noinline int walk_down_proc(struct btrfs_trans_handle *trans, BUG_ON(ret); /* -ENOMEM */ ret = btrfs_dec_ref(trans, root, eb, 0); BUG_ON(ret); /* -ENOMEM */ - ret = btrfs_set_disk_extent_flags(trans, eb->start, - eb->len, flag, + ret = btrfs_set_disk_extent_flags(trans, eb, flag, btrfs_header_level(eb), 0); BUG_ON(ret); /* -ENOMEM */ wc->flags[level] |= flag; @@ -5207,9 +5284,7 @@ static noinline int walk_up_tree(struct btrfs_trans_handle *trans, * * If called with for_reloc == 0, may exit early with -EAGAIN */ -int btrfs_drop_snapshot(struct btrfs_root *root, - struct btrfs_block_rsv *block_rsv, int update_ref, - int for_reloc) +int btrfs_drop_snapshot(struct btrfs_root *root, int update_ref, int for_reloc) { struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_path *path; @@ -5248,9 +5323,6 @@ int btrfs_drop_snapshot(struct btrfs_root *root, if (err) goto out_end_trans; - if (block_rsv) - trans->block_rsv = block_rsv; - /* * This will help us catch people modifying the fs tree while we're * dropping it. It is unsafe to mess with the fs tree while it's being @@ -5378,8 +5450,6 @@ int btrfs_drop_snapshot(struct btrfs_root *root, err = PTR_ERR(trans); goto out_free; } - if (block_rsv) - trans->block_rsv = block_rsv; } } btrfs_release_path(path); @@ -5411,13 +5481,10 @@ int btrfs_drop_snapshot(struct btrfs_root *root, } } - if (test_bit(BTRFS_ROOT_IN_RADIX, &root->state)) { + if (test_bit(BTRFS_ROOT_IN_RADIX, &root->state)) btrfs_add_dropped_root(trans, root); - } else { - free_extent_buffer(root->node); - free_extent_buffer(root->commit_root); - btrfs_put_fs_root(root); - } + else + btrfs_put_root(root); root_dropped = true; out_end_trans: btrfs_end_transaction_throttle(trans); @@ -5747,47 +5814,3 @@ int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range) return bg_ret; return dev_ret; } - -/* - * btrfs_{start,end}_write_no_snapshotting() are similar to - * mnt_{want,drop}_write(), they are used to prevent some tasks from writing - * data into the page cache through nocow before the subvolume is snapshoted, - * but flush the data into disk after the snapshot creation, or to prevent - * operations while snapshotting is ongoing and that cause the snapshot to be - * inconsistent (writes followed by expanding truncates for example). - */ -void btrfs_end_write_no_snapshotting(struct btrfs_root *root) -{ - percpu_counter_dec(&root->subv_writers->counter); - cond_wake_up(&root->subv_writers->wait); -} - -int btrfs_start_write_no_snapshotting(struct btrfs_root *root) -{ - if (atomic_read(&root->will_be_snapshotted)) - return 0; - - percpu_counter_inc(&root->subv_writers->counter); - /* - * Make sure counter is updated before we check for snapshot creation. - */ - smp_mb(); - if (atomic_read(&root->will_be_snapshotted)) { - btrfs_end_write_no_snapshotting(root); - return 0; - } - return 1; -} - -void btrfs_wait_for_snapshot_creation(struct btrfs_root *root) -{ - while (true) { - int ret; - - ret = btrfs_start_write_no_snapshotting(root); - if (ret) - break; - wait_var_event(&root->will_be_snapshotted, - !atomic_read(&root->will_be_snapshotted)); - } -} diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index c0f202741e09..39e45b8a5031 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -35,42 +35,54 @@ static inline bool extent_state_in_tree(const struct extent_state *state) } #ifdef CONFIG_BTRFS_DEBUG -static LIST_HEAD(buffers); static LIST_HEAD(states); - static DEFINE_SPINLOCK(leak_lock); -static inline -void btrfs_leak_debug_add(struct list_head *new, struct list_head *head) +static inline void btrfs_leak_debug_add(spinlock_t *lock, + struct list_head *new, + struct list_head *head) { unsigned long flags; - spin_lock_irqsave(&leak_lock, flags); + spin_lock_irqsave(lock, flags); list_add(new, head); - spin_unlock_irqrestore(&leak_lock, flags); + spin_unlock_irqrestore(lock, flags); } -static inline -void btrfs_leak_debug_del(struct list_head *entry) +static inline void btrfs_leak_debug_del(spinlock_t *lock, + struct list_head *entry) { unsigned long flags; - spin_lock_irqsave(&leak_lock, flags); + spin_lock_irqsave(lock, flags); list_del(entry); - spin_unlock_irqrestore(&leak_lock, flags); + spin_unlock_irqrestore(lock, flags); } -static inline void btrfs_extent_buffer_leak_debug_check(void) +void btrfs_extent_buffer_leak_debug_check(struct btrfs_fs_info *fs_info) { struct extent_buffer *eb; + unsigned long flags; - while (!list_empty(&buffers)) { - eb = list_entry(buffers.next, struct extent_buffer, leak_list); - pr_err("BTRFS: buffer leak start %llu len %lu refs %d bflags %lu\n", - eb->start, eb->len, atomic_read(&eb->refs), eb->bflags); + /* + * If we didn't get into open_ctree our allocated_ebs will not be + * initialized, so just skip this. + */ + if (!fs_info->allocated_ebs.next) + return; + + spin_lock_irqsave(&fs_info->eb_leak_lock, flags); + while (!list_empty(&fs_info->allocated_ebs)) { + eb = list_first_entry(&fs_info->allocated_ebs, + struct extent_buffer, leak_list); + pr_err( + "BTRFS: buffer leak start %llu len %lu refs %d bflags %lu owner %llu\n", + eb->start, eb->len, atomic_read(&eb->refs), eb->bflags, + btrfs_header_owner(eb)); list_del(&eb->leak_list); kmem_cache_free(extent_buffer_cache, eb); } + spin_unlock_irqrestore(&fs_info->eb_leak_lock, flags); } static inline void btrfs_extent_state_leak_debug_check(void) @@ -107,9 +119,8 @@ static inline void __btrfs_debug_check_extent_io_range(const char *caller, } } #else -#define btrfs_leak_debug_add(new, head) do {} while (0) -#define btrfs_leak_debug_del(entry) do {} while (0) -#define btrfs_extent_buffer_leak_debug_check() do {} while (0) +#define btrfs_leak_debug_add(lock, new, head) do {} while (0) +#define btrfs_leak_debug_del(lock, entry) do {} while (0) #define btrfs_extent_state_leak_debug_check() do {} while (0) #define btrfs_debug_check_extent_io_range(c, s, e) do {} while (0) #endif @@ -122,7 +133,6 @@ struct tree_entry { struct extent_page_data { struct bio *bio; - struct extent_io_tree *tree; /* tells writepage not to lock the state bits for this range * it still does the unlocking */ @@ -246,8 +256,6 @@ void __cold extent_state_cache_exit(void) void __cold extent_io_exit(void) { - btrfs_extent_buffer_leak_debug_check(); - /* * Make sure all delayed rcu free are flushed before we * destroy caches. @@ -257,6 +265,15 @@ void __cold extent_io_exit(void) bioset_exit(&btrfs_bioset); } +/* + * For the file_extent_tree, we want to hold the inode lock when we lookup and + * update the disk_i_size, but lockdep will complain because our io_tree we hold + * the tree lock and get the inode lock when setting delalloc. These two things + * are unrelated, so make a class for the file_extent_tree so we don't get the + * two locking patterns mixed up. + */ +static struct lock_class_key file_extent_tree_class; + void extent_io_tree_init(struct btrfs_fs_info *fs_info, struct extent_io_tree *tree, unsigned int owner, void *private_data) @@ -268,6 +285,8 @@ void extent_io_tree_init(struct btrfs_fs_info *fs_info, spin_lock_init(&tree->lock); tree->private_data = private_data; tree->owner = owner; + if (owner == IO_TREE_INODE_FILE_EXTENT) + lockdep_set_class(&tree->lock, &file_extent_tree_class); } void extent_io_tree_release(struct extent_io_tree *tree) @@ -314,7 +333,7 @@ static struct extent_state *alloc_extent_state(gfp_t mask) state->state = 0; state->failrec = NULL; RB_CLEAR_NODE(&state->rb_node); - btrfs_leak_debug_add(&state->leak_list, &states); + btrfs_leak_debug_add(&leak_lock, &state->leak_list, &states); refcount_set(&state->refs, 1); init_waitqueue_head(&state->wq); trace_alloc_extent_state(state, mask, _RET_IP_); @@ -327,7 +346,7 @@ void free_extent_state(struct extent_state *state) return; if (refcount_dec_and_test(&state->refs)) { WARN_ON(extent_state_in_tree(state)); - btrfs_leak_debug_del(&state->leak_list); + btrfs_leak_debug_del(&leak_lock, &state->leak_list); trace_free_extent_state(state, _RET_IP_); kmem_cache_free(extent_state_cache, state); } @@ -1053,6 +1072,16 @@ hit_next: goto out; } + /* + * If this extent already has all the bits we want set, then + * skip it, not necessary to split it or do anything with it. + */ + if ((state->state & bits) == bits) { + start = state->end + 1; + cache_state(state, cached_state); + goto search_again; + } + prealloc = alloc_extent_state_atomic(prealloc); BUG_ON(!prealloc); err = split_state(tree, state, prealloc, start); @@ -1568,6 +1597,43 @@ out: } /** + * find_contiguous_extent_bit: find a contiguous area of bits + * @tree - io tree to check + * @start - offset to start the search from + * @start_ret - the first offset we found with the bits set + * @end_ret - the final contiguous range of the bits that were set + * @bits - bits to look for + * + * set_extent_bit and clear_extent_bit can temporarily split contiguous ranges + * to set bits appropriately, and then merge them again. During this time it + * will drop the tree->lock, so use this helper if you want to find the actual + * contiguous area for given bits. We will search to the first bit we find, and + * then walk down the tree until we find a non-contiguous area. The area + * returned will be the full contiguous area with the bits set. + */ +int find_contiguous_extent_bit(struct extent_io_tree *tree, u64 start, + u64 *start_ret, u64 *end_ret, unsigned bits) +{ + struct extent_state *state; + int ret = 1; + + spin_lock(&tree->lock); + state = find_first_extent_bit_state(tree, start, bits); + if (state) { + *start_ret = state->start; + *end_ret = state->end; + while ((state = next_state(state)) != NULL) { + if (state->start > (*end_ret + 1)) + break; + *end_ret = state->end; + } + ret = 0; + } + spin_unlock(&tree->lock); + return ret; +} + +/** * find_first_clear_extent_bit - find the first range that has @bits not set. * This range could start before @start. * @@ -2926,7 +2992,6 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size) /* * @opf: bio REQ_OP_* and REQ_* flags as one value - * @tree: tree so we can call our merge_bio hook * @wbc: optional writeback control for io accounting * @page: page to add to the bio * @pg_offset: offset of the new bio or to check whether we are adding @@ -2939,7 +3004,7 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size) * @prev_bio_flags: flags of previous bio to see if we can merge the current one * @bio_flags: flags of the current bio to see if we can merge them */ -static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree, +static int submit_extent_page(unsigned int opf, struct writeback_control *wbc, struct page *page, u64 offset, size_t size, unsigned long pg_offset, @@ -2954,6 +3019,7 @@ static int submit_extent_page(unsigned int opf, struct extent_io_tree *tree, struct bio *bio; size_t page_size = min_t(size_t, size, PAGE_SIZE); sector_t sector = offset >> 9; + struct extent_io_tree *tree = &BTRFS_I(page->mapping->host)->io_tree; ASSERT(bio_ret); @@ -3062,8 +3128,7 @@ __get_extent_map(struct inode *inode, struct page *page, size_t pg_offset, * XXX JDM: This needs looking at to ensure proper page locking * return 0 on success, otherwise return error */ -static int __do_readpage(struct extent_io_tree *tree, - struct page *page, +static int __do_readpage(struct page *page, get_extent_t *get_extent, struct extent_map **em_cached, struct bio **bio, int mirror_num, @@ -3086,6 +3151,7 @@ static int __do_readpage(struct extent_io_tree *tree, size_t disk_io_size; size_t blocksize = inode->i_sb->s_blocksize; unsigned long this_bio_flag = 0; + struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; set_page_extent_mapped(page); @@ -3242,7 +3308,7 @@ static int __do_readpage(struct extent_io_tree *tree, continue; } - ret = submit_extent_page(REQ_OP_READ | read_flags, tree, NULL, + ret = submit_extent_page(REQ_OP_READ | read_flags, NULL, page, offset, disk_io_size, pg_offset, bio, end_bio_extent_readpage, mirror_num, @@ -3269,8 +3335,7 @@ out: return ret; } -static inline void contiguous_readpages(struct extent_io_tree *tree, - struct page *pages[], int nr_pages, +static inline void contiguous_readpages(struct page *pages[], int nr_pages, u64 start, u64 end, struct extent_map **em_cached, struct bio **bio, @@ -3280,17 +3345,16 @@ static inline void contiguous_readpages(struct extent_io_tree *tree, struct btrfs_inode *inode = BTRFS_I(pages[0]->mapping->host); int index; - btrfs_lock_and_flush_ordered_range(tree, inode, start, end, NULL); + btrfs_lock_and_flush_ordered_range(inode, start, end, NULL); for (index = 0; index < nr_pages; index++) { - __do_readpage(tree, pages[index], btrfs_get_extent, em_cached, + __do_readpage(pages[index], btrfs_get_extent, em_cached, bio, 0, bio_flags, REQ_RAHEAD, prev_em_start); put_page(pages[index]); } } -static int __extent_read_full_page(struct extent_io_tree *tree, - struct page *page, +static int __extent_read_full_page(struct page *page, get_extent_t *get_extent, struct bio **bio, int mirror_num, unsigned long *bio_flags, @@ -3301,21 +3365,21 @@ static int __extent_read_full_page(struct extent_io_tree *tree, u64 end = start + PAGE_SIZE - 1; int ret; - btrfs_lock_and_flush_ordered_range(tree, inode, start, end, NULL); + btrfs_lock_and_flush_ordered_range(inode, start, end, NULL); - ret = __do_readpage(tree, page, get_extent, NULL, bio, mirror_num, + ret = __do_readpage(page, get_extent, NULL, bio, mirror_num, bio_flags, read_flags, NULL); return ret; } -int extent_read_full_page(struct extent_io_tree *tree, struct page *page, - get_extent_t *get_extent, int mirror_num) +int extent_read_full_page(struct page *page, get_extent_t *get_extent, + int mirror_num) { struct bio *bio = NULL; unsigned long bio_flags = 0; int ret; - ret = __extent_read_full_page(tree, page, get_extent, &bio, mirror_num, + ret = __extent_read_full_page(page, get_extent, &bio, mirror_num, &bio_flags, 0); if (bio) ret = submit_one_bio(bio, mirror_num, bio_flags); @@ -3423,7 +3487,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode, unsigned long nr_written, int *nr_ret) { - struct extent_io_tree *tree = epd->tree; + struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; u64 start = page_offset(page); u64 page_end = start + PAGE_SIZE - 1; u64 end; @@ -3509,7 +3573,7 @@ static noinline_for_stack int __extent_writepage_io(struct inode *inode, page->index, cur, end); } - ret = submit_extent_page(REQ_OP_WRITE | write_flags, tree, wbc, + ret = submit_extent_page(REQ_OP_WRITE | write_flags, wbc, page, offset, iosize, pg_offset, &epd->bio, end_bio_extent_writepage, @@ -3830,8 +3894,6 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, struct writeback_control *wbc, struct extent_page_data *epd) { - struct btrfs_fs_info *fs_info = eb->fs_info; - struct extent_io_tree *tree = &BTRFS_I(fs_info->btree_inode)->io_tree; u64 offset = eb->start; u32 nritems; int i, num_pages; @@ -3864,7 +3926,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, clear_page_dirty_for_io(p); set_page_writeback(p); - ret = submit_extent_page(REQ_OP_WRITE | write_flags, tree, wbc, + ret = submit_extent_page(REQ_OP_WRITE | write_flags, wbc, p, offset, PAGE_SIZE, 0, &epd->bio, end_bio_extent_buffer_writepage, @@ -3897,14 +3959,13 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, int btree_write_cache_pages(struct address_space *mapping, struct writeback_control *wbc) { - struct extent_io_tree *tree = &BTRFS_I(mapping->host)->io_tree; struct extent_buffer *eb, *prev_eb = NULL; struct extent_page_data epd = { .bio = NULL, - .tree = tree, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; + struct btrfs_fs_info *fs_info = BTRFS_I(mapping->host)->root->fs_info; int ret = 0; int done = 0; int nr_to_write_done = 0; @@ -4018,7 +4079,39 @@ retry: end_write_bio(&epd, ret); return ret; } - ret = flush_write_bio(&epd); + /* + * If something went wrong, don't allow any metadata write bio to be + * submitted. + * + * This would prevent use-after-free if we had dirty pages not + * cleaned up, which can still happen by fuzzed images. + * + * - Bad extent tree + * Allowing existing tree block to be allocated for other trees. + * + * - Log tree operations + * Exiting tree blocks get allocated to log tree, bumps its + * generation, then get cleaned in tree re-balance. + * Such tree block will not be written back, since it's clean, + * thus no WRITTEN flag set. + * And after log writes back, this tree block is not traced by + * any dirty extent_io_tree. + * + * - Offending tree block gets re-dirtied from its original owner + * Since it has bumped generation, no WRITTEN flag, it can be + * reused without COWing. This tree block will not be traced + * by btrfs_transaction::dirty_pages. + * + * Now such dirty tree block will not be cleaned by any dirty + * extent io tree. Thus we don't want to submit such wild eb + * if the fs already has error. + */ + if (!test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) { + ret = flush_write_bio(&epd); + } else { + ret = -EUCLEAN; + end_write_bio(&epd, ret); + } return ret; } @@ -4190,7 +4283,6 @@ int extent_write_full_page(struct page *page, struct writeback_control *wbc) int ret; struct extent_page_data epd = { .bio = NULL, - .tree = &BTRFS_I(page->mapping->host)->io_tree, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; @@ -4212,14 +4304,12 @@ int extent_write_locked_range(struct inode *inode, u64 start, u64 end, { int ret = 0; struct address_space *mapping = inode->i_mapping; - struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; struct page *page; unsigned long nr_pages = (end - start + PAGE_SIZE) >> PAGE_SHIFT; struct extent_page_data epd = { .bio = NULL, - .tree = tree, .extent_locked = 1, .sync_io = mode == WB_SYNC_ALL, }; @@ -4263,7 +4353,6 @@ int extent_writepages(struct address_space *mapping, int ret = 0; struct extent_page_data epd = { .bio = NULL, - .tree = &BTRFS_I(mapping->host)->io_tree, .extent_locked = 0, .sync_io = wbc->sync_mode == WB_SYNC_ALL, }; @@ -4285,7 +4374,6 @@ int extent_readpages(struct address_space *mapping, struct list_head *pages, unsigned long bio_flags = 0; struct page *pagepool[16]; struct extent_map *em_cached = NULL; - struct extent_io_tree *tree = &BTRFS_I(mapping->host)->io_tree; int nr = 0; u64 prev_em_start = (u64)-1; @@ -4312,7 +4400,7 @@ int extent_readpages(struct address_space *mapping, struct list_head *pages, ASSERT(contig_start + nr * PAGE_SIZE - 1 == contig_end); - contiguous_readpages(tree, pagepool, nr, contig_start, + contiguous_readpages(pagepool, nr, contig_start, contig_end, &em_cached, &bio, &bio_flags, &prev_em_start); } @@ -4796,7 +4884,6 @@ out_free_ulist: static void __free_extent_buffer(struct extent_buffer *eb) { - btrfs_leak_debug_del(&eb->leak_list); kmem_cache_free(extent_buffer_cache, eb); } @@ -4862,6 +4949,7 @@ static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb) static inline void btrfs_release_extent_buffer(struct extent_buffer *eb) { btrfs_release_extent_buffer_pages(eb); + btrfs_leak_debug_del(&eb->fs_info->eb_leak_lock, &eb->leak_list); __free_extent_buffer(eb); } @@ -4883,7 +4971,8 @@ __alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start, init_waitqueue_head(&eb->write_lock_wq); init_waitqueue_head(&eb->read_lock_wq); - btrfs_leak_debug_add(&eb->leak_list, &buffers); + btrfs_leak_debug_add(&fs_info->eb_leak_lock, &eb->leak_list, + &fs_info->allocated_ebs); spin_lock_init(&eb->refs_lock); atomic_set(&eb->refs, 1); @@ -5230,6 +5319,7 @@ static inline void btrfs_release_extent_buffer_rcu(struct rcu_head *head) } static int release_extent_buffer(struct extent_buffer *eb) + __releases(&eb->refs_lock) { lockdep_assert_held(&eb->refs_lock); @@ -5248,6 +5338,7 @@ static int release_extent_buffer(struct extent_buffer *eb) spin_unlock(&eb->refs_lock); } + btrfs_leak_debug_del(&eb->fs_info->eb_leak_lock, &eb->leak_list); /* Should be safe to release our pages at this point */ btrfs_release_extent_buffer_pages(eb); #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS @@ -5405,7 +5496,6 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num) unsigned long num_reads = 0; struct bio *bio = NULL; unsigned long bio_flags = 0; - struct extent_io_tree *tree = &BTRFS_I(eb->fs_info->btree_inode)->io_tree; if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) return 0; @@ -5453,7 +5543,7 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num) } ClearPageError(page); - err = __extent_read_full_page(tree, page, + err = __extent_read_full_page(page, btree_get_extent, &bio, mirror_num, &bio_flags, REQ_META); diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 5d205bbaafdc..2ed65bd0760e 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -189,8 +189,8 @@ typedef struct extent_map *(get_extent_t)(struct btrfs_inode *inode, int try_release_extent_mapping(struct page *page, gfp_t mask); int try_release_extent_buffer(struct page *page); -int extent_read_full_page(struct extent_io_tree *tree, struct page *page, - get_extent_t *get_extent, int mirror_num); +int extent_read_full_page(struct page *page, get_extent_t *get_extent, + int mirror_num); int extent_write_full_page(struct page *page, struct writeback_control *wbc); int extent_write_locked_range(struct inode *inode, u64 start, u64 end, int mode); @@ -325,4 +325,11 @@ bool find_lock_delalloc_range(struct inode *inode, #endif struct extent_buffer *alloc_test_extent_buffer(struct btrfs_fs_info *fs_info, u64 start); + +#ifdef CONFIG_BTRFS_DEBUG +void btrfs_extent_buffer_leak_debug_check(struct btrfs_fs_info *fs_info); +#else +#define btrfs_extent_buffer_leak_debug_check(fs_info) do {} while (0) +#endif + #endif diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index c2f365662d55..b618ad5339ba 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -23,6 +23,97 @@ #define MAX_CSUM_ITEMS(r, size) (min_t(u32, __MAX_CSUM_ITEMS(r, size), \ PAGE_SIZE)) +/** + * @inode - the inode we want to update the disk_i_size for + * @new_i_size - the i_size we want to set to, 0 if we use i_size + * + * With NO_HOLES set this simply sets the disk_is_size to whatever i_size_read() + * returns as it is perfectly fine with a file that has holes without hole file + * extent items. + * + * However without NO_HOLES we need to only return the area that is contiguous + * from the 0 offset of the file. Otherwise we could end up adjust i_size up + * to an extent that has a gap in between. + * + * Finally new_i_size should only be set in the case of truncate where we're not + * ready to use i_size_read() as the limiter yet. + */ +void btrfs_inode_safe_disk_i_size_write(struct inode *inode, u64 new_i_size) +{ + struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + u64 start, end, i_size; + int ret; + + i_size = new_i_size ?: i_size_read(inode); + if (btrfs_fs_incompat(fs_info, NO_HOLES)) { + BTRFS_I(inode)->disk_i_size = i_size; + return; + } + + spin_lock(&BTRFS_I(inode)->lock); + ret = find_contiguous_extent_bit(&BTRFS_I(inode)->file_extent_tree, 0, + &start, &end, EXTENT_DIRTY); + if (!ret && start == 0) + i_size = min(i_size, end + 1); + else + i_size = 0; + BTRFS_I(inode)->disk_i_size = i_size; + spin_unlock(&BTRFS_I(inode)->lock); +} + +/** + * @inode - the inode we're modifying + * @start - the start file offset of the file extent we've inserted + * @len - the logical length of the file extent item + * + * Call when we are inserting a new file extent where there was none before. + * Does not need to call this in the case where we're replacing an existing file + * extent, however if not sure it's fine to call this multiple times. + * + * The start and len must match the file extent item, so thus must be sectorsize + * aligned. + */ +int btrfs_inode_set_file_extent_range(struct btrfs_inode *inode, u64 start, + u64 len) +{ + if (len == 0) + return 0; + + ASSERT(IS_ALIGNED(start + len, inode->root->fs_info->sectorsize)); + + if (btrfs_fs_incompat(inode->root->fs_info, NO_HOLES)) + return 0; + return set_extent_bits(&inode->file_extent_tree, start, start + len - 1, + EXTENT_DIRTY); +} + +/** + * @inode - the inode we're modifying + * @start - the start file offset of the file extent we've inserted + * @len - the logical length of the file extent item + * + * Called when we drop a file extent, for example when we truncate. Doesn't + * need to be called for cases where we're replacing a file extent, like when + * we've COWed a file extent. + * + * The start and len must match the file extent item, so thus must be sectorsize + * aligned. + */ +int btrfs_inode_clear_file_extent_range(struct btrfs_inode *inode, u64 start, + u64 len) +{ + if (len == 0) + return 0; + + ASSERT(IS_ALIGNED(start + len, inode->root->fs_info->sectorsize) || + len == (u64)-1); + + if (btrfs_fs_incompat(inode->root->fs_info, NO_HOLES)) + return 0; + return clear_extent_bit(&inode->file_extent_tree, start, + start + len - 1, EXTENT_DIRTY, 0, 0, NULL); +} + static inline u32 max_ordered_sum_bytes(struct btrfs_fs_info *fs_info, u16 csum_size) { @@ -949,18 +1040,7 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode, btrfs_item_key_to_cpu(leaf, &key, slot); extent_start = key.offset; - - if (type == BTRFS_FILE_EXTENT_REG || - type == BTRFS_FILE_EXTENT_PREALLOC) { - extent_end = extent_start + - btrfs_file_extent_num_bytes(leaf, fi); - } else if (type == BTRFS_FILE_EXTENT_INLINE) { - size_t size; - size = btrfs_file_extent_ram_bytes(leaf, fi); - extent_end = ALIGN(extent_start + size, - fs_info->sectorsize); - } - + extent_end = btrfs_file_extent_end(path); em->ram_bytes = btrfs_file_extent_ram_bytes(leaf, fi); if (type == BTRFS_FILE_EXTENT_REG || type == BTRFS_FILE_EXTENT_PREALLOC) { @@ -1007,3 +1087,30 @@ void btrfs_extent_item_to_extent_map(struct btrfs_inode *inode, root->root_key.objectid); } } + +/* + * Returns the end offset (non inclusive) of the file extent item the given path + * points to. If it points to an inline extent, the returned offset is rounded + * up to the sector size. + */ +u64 btrfs_file_extent_end(const struct btrfs_path *path) +{ + const struct extent_buffer *leaf = path->nodes[0]; + const int slot = path->slots[0]; + struct btrfs_file_extent_item *fi; + struct btrfs_key key; + u64 end; + + btrfs_item_key_to_cpu(leaf, &key, slot); + ASSERT(key.type == BTRFS_EXTENT_DATA_KEY); + fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); + + if (btrfs_file_extent_type(leaf, fi) == BTRFS_FILE_EXTENT_INLINE) { + end = btrfs_file_extent_ram_bytes(leaf, fi); + end = ALIGN(key.offset + end, leaf->fs_info->sectorsize); + } else { + end = key.offset + btrfs_file_extent_num_bytes(leaf, fi); + } + + return end; +} diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index a16da274c9aa..8a144f9cb7ac 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -27,6 +27,7 @@ #include "qgroup.h" #include "compression.h" #include "delalloc-space.h" +#include "reflink.h" static struct kmem_cache *btrfs_inode_defrag_cachep; /* @@ -277,7 +278,6 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info, struct btrfs_key key; struct btrfs_ioctl_defrag_range_args range; int num_defrag; - int index; int ret; /* get the inode */ @@ -285,9 +285,7 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info, key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - index = srcu_read_lock(&fs_info->subvol_srcu); - - inode_root = btrfs_read_fs_root_no_name(fs_info, &key); + inode_root = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(inode_root)) { ret = PTR_ERR(inode_root); goto cleanup; @@ -297,11 +295,11 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info, key.type = BTRFS_INODE_ITEM_KEY; key.offset = 0; inode = btrfs_iget(fs_info->sb, &key, inode_root); + btrfs_put_root(inode_root); if (IS_ERR(inode)) { ret = PTR_ERR(inode); goto cleanup; } - srcu_read_unlock(&fs_info->subvol_srcu, index); /* do a chunk of defrag */ clear_bit(BTRFS_INODE_IN_DEFRAG, &BTRFS_I(inode)->runtime_flags); @@ -337,7 +335,6 @@ static int __btrfs_run_defrag_inode(struct btrfs_fs_info *fs_info, iput(inode); return 0; cleanup: - srcu_read_unlock(&fs_info->subvol_srcu, index); kmem_cache_free(btrfs_inode_defrag_cachep, defrag); return ret; } @@ -1552,15 +1549,14 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos, u64 num_bytes; int ret; - ret = btrfs_start_write_no_snapshotting(root); - if (!ret) + if (!btrfs_drew_try_write_lock(&root->snapshot_lock)) return -EAGAIN; lockstart = round_down(pos, fs_info->sectorsize); lockend = round_up(pos + *write_bytes, fs_info->sectorsize) - 1; - btrfs_lock_and_flush_ordered_range(&inode->io_tree, inode, lockstart, + btrfs_lock_and_flush_ordered_range(inode, lockstart, lockend, NULL); num_bytes = lockend - lockstart + 1; @@ -1568,7 +1564,7 @@ static noinline int check_can_nocow(struct btrfs_inode *inode, loff_t pos, NULL, NULL, NULL); if (ret <= 0) { ret = 0; - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); } else { *write_bytes = min_t(size_t, *write_bytes , num_bytes - pos + lockstart); @@ -1674,7 +1670,7 @@ static noinline ssize_t btrfs_buffered_write(struct kiocb *iocb, data_reserved, pos, write_bytes); else - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); break; } @@ -1778,7 +1774,7 @@ again: release_bytes = 0; if (only_release_metadata) - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); if (only_release_metadata && copied > 0) { lockstart = round_down(pos, @@ -1807,7 +1803,7 @@ again: if (release_bytes) { if (only_release_metadata) { - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); btrfs_delalloc_release_metadata(BTRFS_I(inode), release_bytes, true); } else { @@ -2071,6 +2067,16 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) btrfs_init_log_ctx(&ctx, inode); /* + * Set the range to full if the NO_HOLES feature is not enabled. + * This is to avoid missing file extent items representing holes after + * replaying the log. + */ + if (!btrfs_fs_incompat(fs_info, NO_HOLES)) { + start = 0; + end = LLONG_MAX; + } + + /* * We write the dirty pages in the range and wait until they complete * out of the ->i_mutex. If so, we can flush the dirty pages by * multi-task, and make the performance up. See @@ -2092,19 +2098,6 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) atomic_inc(&root->log_batch); /* - * If the inode needs a full sync, make sure we use a full range to - * avoid log tree corruption, due to hole detection racing with ordered - * extent completion for adjacent ranges, and assertion failures during - * hole detection. Do this while holding the inode lock, to avoid races - * with other tasks. - */ - if (test_bit(BTRFS_INODE_NEEDS_FULL_SYNC, - &BTRFS_I(inode)->runtime_flags)) { - start = 0; - end = LLONG_MAX; - } - - /* * Before we acquired the inode's lock, someone may have dirtied more * pages in the target range. We need to make sure that writeback for * any such pages does not start while we are logging the inode, because @@ -2124,6 +2117,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) */ ret = start_ordered_ops(inode, start, end); if (ret) { + up_write(&BTRFS_I(inode)->dio_sem); inode_unlock(inode); goto out; } @@ -2486,6 +2480,11 @@ static int btrfs_insert_clone_extent(struct btrfs_trans_handle *trans, btrfs_mark_buffer_dirty(leaf); btrfs_release_path(path); + ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), + clone_info->file_offset, clone_len); + if (ret) + return ret; + /* If it's a hole, nothing more needs to be done. */ if (clone_info->disk_offset == 0) return 0; @@ -2596,6 +2595,24 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path, btrfs_abort_transaction(trans, ret); break; } + } else if (!clone_info && cur_offset < drop_end) { + /* + * We are past the i_size here, but since we didn't + * insert holes we need to clear the mapped area so we + * know to not set disk_i_size in this area until a new + * file extent is inserted here. + */ + ret = btrfs_inode_clear_file_extent_range(BTRFS_I(inode), + cur_offset, drop_end - cur_offset); + if (ret) { + /* + * We couldn't clear our area, so we could + * presumably adjust up and corrupt the fs, so + * we need to abort. + */ + btrfs_abort_transaction(trans, ret); + break; + } } if (clone_info && drop_end > clone_info->file_offset) { @@ -2686,6 +2703,15 @@ int btrfs_punch_hole_range(struct inode *inode, struct btrfs_path *path, btrfs_abort_transaction(trans, ret); goto out_trans; } + } else if (!clone_info && cur_offset < drop_end) { + /* See the comment in the loop above for the reasoning here. */ + ret = btrfs_inode_clear_file_extent_range(BTRFS_I(inode), + cur_offset, drop_end - cur_offset); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out_trans; + } + } if (clone_info) { ret = btrfs_insert_clone_extent(trans, inode, path, clone_info, @@ -2935,7 +2961,7 @@ static int btrfs_fallocate_update_isize(struct inode *inode, inode->i_ctime = current_time(inode); i_size_write(inode, end); - btrfs_ordered_update_i_size(inode, end, NULL); + btrfs_inode_safe_disk_i_size_write(inode, 0); ret = btrfs_update_inode(trans, root, inode); ret2 = btrfs_end_transaction(trans); diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 0598fd3c6e3f..3613da065a73 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -371,10 +371,10 @@ static void io_ctl_drop_pages(struct btrfs_io_ctl *io_ctl) } } -static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, struct inode *inode, - int uptodate) +static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate) { struct page *page; + struct inode *inode = io_ctl->inode; gfp_t mask = btrfs_alloc_write_mask(inode->i_mapping); int i; @@ -732,7 +732,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, readahead_cache(inode); - ret = io_ctl_prepare_pages(&io_ctl, inode, 1); + ret = io_ctl_prepare_pages(&io_ctl, true); if (ret) goto out; @@ -1067,6 +1067,7 @@ fail: } static noinline_for_stack int write_pinned_extent_entries( + struct btrfs_trans_handle *trans, struct btrfs_block_group *block_group, struct btrfs_io_ctl *io_ctl, int *entries) @@ -1085,7 +1086,7 @@ static noinline_for_stack int write_pinned_extent_entries( * We shouldn't have switched the pinned extents yet so this is the * right one */ - unpin = block_group->fs_info->pinned_extents; + unpin = &trans->transaction->pinned_extents; start = block_group->start; @@ -1190,7 +1191,7 @@ out: invalidate_inode_pages2(inode->i_mapping); BTRFS_I(inode)->generation = 0; if (block_group) { -#ifdef DEBUG +#ifdef CONFIG_BTRFS_DEBUG btrfs_err(root->fs_info, "failed to write free space cache for block group %llu", block_group->start); @@ -1291,7 +1292,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, } /* Lock all pages first so we can lock the extent safely. */ - ret = io_ctl_prepare_pages(io_ctl, inode, 0); + ret = io_ctl_prepare_pages(io_ctl, false); if (ret) goto out_unlock; @@ -1317,7 +1318,7 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, * If this changes while we are working we'll get added back to * the dirty list and redo it. No locking needed */ - ret = write_pinned_extent_entries(block_group, io_ctl, &entries); + ret = write_pinned_extent_entries(trans, block_group, io_ctl, &entries); if (ret) goto out_nospc_locked; @@ -1366,18 +1367,6 @@ static int __btrfs_write_out_cache(struct btrfs_root *root, struct inode *inode, return 0; -out: - io_ctl->inode = NULL; - io_ctl_free(io_ctl); - if (ret) { - invalidate_inode_pages2(inode->i_mapping); - BTRFS_I(inode)->generation = 0; - } - btrfs_update_inode(trans, root, inode); - if (must_iput) - iput(inode); - return ret; - out_nospc_locked: cleanup_bitmap_list(&bitmap_list); spin_unlock(&ctl->tree_lock); @@ -1390,7 +1379,17 @@ out_unlock: if (block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA)) up_write(&block_group->data_rwsem); - goto out; +out: + io_ctl->inode = NULL; + io_ctl_free(io_ctl); + if (ret) { + invalidate_inode_pages2(inode->i_mapping); + BTRFS_I(inode)->generation = 0; + } + btrfs_update_inode(trans, root, inode); + if (must_iput) + iput(inode); + return ret; } int btrfs_write_out_cache(struct btrfs_trans_handle *trans, @@ -1416,7 +1415,7 @@ int btrfs_write_out_cache(struct btrfs_trans_handle *trans, ret = __btrfs_write_out_cache(fs_info->tree_root, inode, ctl, block_group, &block_group->io_ctl, trans); if (ret) { -#ifdef DEBUG +#ifdef CONFIG_BTRFS_DEBUG btrfs_err(fs_info, "failed to write free space cache for block group %llu", block_group->start); @@ -4036,7 +4035,7 @@ int btrfs_write_out_ino_cache(struct btrfs_root *root, if (release_metadata) btrfs_delalloc_release_metadata(BTRFS_I(inode), inode->i_size, true); -#ifdef DEBUG +#ifdef CONFIG_BTRFS_DEBUG btrfs_err(fs_info, "failed to write free ino cache for root %llu", root->root_key.objectid); diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c index 258cb3fae17a..8b1f5c8897b7 100644 --- a/fs/btrfs/free-space-tree.c +++ b/fs/btrfs/free-space-tree.c @@ -1251,9 +1251,7 @@ int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info) btrfs_free_tree_block(trans, free_space_root, free_space_root->node, 0, 1); - free_extent_buffer(free_space_root->node); - free_extent_buffer(free_space_root->commit_root); - kfree(free_space_root); + btrfs_put_root(free_space_root); return btrfs_commit_transaction(trans); diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c index d5c9c69d8263..6009e0e939b5 100644 --- a/fs/btrfs/inode-map.c +++ b/fs/btrfs/inode-map.c @@ -515,7 +515,7 @@ out_release: trace_btrfs_space_reservation(fs_info, "ino_cache", trans->transid, trans->bytes_reserved, 0); btrfs_block_rsv_release(fs_info, trans->block_rsv, - trans->bytes_reserved); + trans->bytes_reserved, NULL); out: trans->block_rsv = rsv; trans->bytes_reserved = num_bytes; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 7d26b4bfb2c6..320d1062068d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -28,6 +28,7 @@ #include <linux/magic.h> #include <linux/iversion.h> #include <linux/swap.h> +#include <linux/migrate.h> #include <linux/sched/mm.h> #include <asm/unaligned.h> #include "misc.h" @@ -242,6 +243,15 @@ static int insert_inline_extent(struct btrfs_trans_handle *trans, btrfs_release_path(path); /* + * We align size to sectorsize for inline extents just for simplicity + * sake. + */ + size = ALIGN(size, root->fs_info->sectorsize); + ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), start, size); + if (ret) + goto fail; + + /* * we're an inline extent, so nobody can * extend the file past i_size without locking * a page we already have locked. @@ -2446,6 +2456,11 @@ static int insert_reserved_file_extent(struct btrfs_trans_handle *trans, ins.offset = disk_num_bytes; ins.type = BTRFS_EXTENT_ITEM_KEY; + ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), file_pos, + ram_bytes); + if (ret) + goto out; + /* * Release the reserved range from inode dirty range map, as it is * already moved into delayed_ref_head @@ -2536,7 +2551,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) */ btrfs_qgroup_free_data(inode, NULL, start, ordered_extent->num_bytes); - btrfs_ordered_update_i_size(inode, 0, ordered_extent); + btrfs_inode_safe_disk_i_size_write(inode, 0); if (freespace_inode) trans = btrfs_join_transaction_spacecache(root); else @@ -2607,7 +2622,7 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) goto out; } - btrfs_ordered_update_i_size(inode, 0, ordered_extent); + btrfs_inode_safe_disk_i_size_write(inode, 0); ret = btrfs_update_inode_fallback(trans, root, inode); if (ret) { /* -ENOMEM or corruption */ btrfs_abort_transaction(trans, ret); @@ -3187,6 +3202,8 @@ static int btrfs_read_locked_inode(struct inode *inode, i_uid_write(inode, btrfs_inode_uid(leaf, inode_item)); i_gid_write(inode, btrfs_inode_gid(leaf, inode_item)); btrfs_i_size_write(BTRFS_I(inode), btrfs_inode_size(leaf, inode_item)); + btrfs_inode_set_file_extent_range(BTRFS_I(inode), 0, + round_up(i_size_read(inode), fs_info->sectorsize)); inode->i_atime.tv_sec = btrfs_timespec_sec(leaf, &inode_item->atime); inode->i_atime.tv_nsec = btrfs_timespec_nsec(leaf, &inode_item->atime); @@ -4103,8 +4120,9 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, return -ENOMEM; path->reada = READA_BACK; - lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1, - &cached_state); + if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) + lock_extent_bits(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1, + &cached_state); /* * We want to drop from the next block forward in case this new size is @@ -4142,7 +4160,6 @@ search_again: goto out; } - path->leave_spinning = 1; ret = btrfs_search_slot(trans, root, &key, path, -1, 1); if (ret < 0) goto out; @@ -4158,6 +4175,8 @@ search_again: } while (1) { + u64 clear_start = 0, clear_len = 0; + fi = NULL; leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); @@ -4208,6 +4227,8 @@ search_again: if (extent_type != BTRFS_FILE_EXTENT_INLINE) { u64 num_dec; + + clear_start = found_key.offset; extent_start = btrfs_file_extent_disk_bytenr(leaf, fi); if (!del_item) { u64 orig_num_bytes = @@ -4215,6 +4236,7 @@ search_again: extent_num_bytes = ALIGN(new_size - found_key.offset, fs_info->sectorsize); + clear_start = ALIGN(new_size, fs_info->sectorsize); btrfs_set_file_extent_num_bytes(leaf, fi, extent_num_bytes); num_dec = (orig_num_bytes - @@ -4240,6 +4262,7 @@ search_again: inode_sub_bytes(inode, num_dec); } } + clear_len = num_dec; } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { /* * we can't truncate inline items that have had @@ -4261,12 +4284,33 @@ search_again: */ ret = NEED_TRUNCATE_BLOCK; break; + } else { + /* + * Inline extents are special, we just treat + * them as a full sector worth in the file + * extent tree just for simplicity sake. + */ + clear_len = fs_info->sectorsize; } if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) inode_sub_bytes(inode, item_end + 1 - new_size); } delete: + /* + * We use btrfs_truncate_inode_items() to clean up log trees for + * multiple fsyncs, and in this case we don't want to clear the + * file extent range because it's just the log. + */ + if (root == BTRFS_I(inode)->root) { + ret = btrfs_inode_clear_file_extent_range(BTRFS_I(inode), + clear_start, clear_len); + if (ret) { + btrfs_abort_transaction(trans, ret); + break; + } + } + if (del_item) last_size = found_key.offset; else @@ -4294,7 +4338,6 @@ delete: root == fs_info->tree_root)) { struct btrfs_ref ref = { 0 }; - btrfs_set_path_blocking(path); bytes_deleted += extent_num_bytes; btrfs_init_generic_ref(&ref, BTRFS_DROP_DELAYED_REF, @@ -4369,12 +4412,11 @@ out: ASSERT(last_size >= new_size); if (!ret && last_size > new_size) last_size = new_size; - btrfs_ordered_update_i_size(inode, last_size, NULL); + btrfs_inode_safe_disk_i_size_write(inode, last_size); + unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, + (u64)-1, &cached_state); } - unlock_extent_cached(&BTRFS_I(inode)->io_tree, lock_start, (u64)-1, - &cached_state); - btrfs_free_path(path); return ret; } @@ -4578,7 +4620,7 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) if (size <= hole_start) return 0; - btrfs_lock_and_flush_ordered_range(io_tree, BTRFS_I(inode), hole_start, + btrfs_lock_and_flush_ordered_range(BTRFS_I(inode), hole_start, block_end - 1, &cached_state); cur_offset = hole_start; while (1) { @@ -4591,14 +4633,21 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) } last_byte = min(extent_map_end(em), block_end); last_byte = ALIGN(last_byte, fs_info->sectorsize); + hole_size = last_byte - cur_offset; + if (!test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) { struct extent_map *hole_em; - hole_size = last_byte - cur_offset; err = maybe_insert_hole(root, inode, cur_offset, hole_size); if (err) break; + + err = btrfs_inode_set_file_extent_range(BTRFS_I(inode), + cur_offset, hole_size); + if (err) + break; + btrfs_drop_extent_cache(BTRFS_I(inode), cur_offset, cur_offset + hole_size - 1, 0); hole_em = alloc_extent_map(); @@ -4630,6 +4679,11 @@ int btrfs_cont_expand(struct inode *inode, loff_t oldsize, loff_t size) hole_size - 1, 0); } free_extent_map(hole_em); + } else { + err = btrfs_inode_set_file_extent_range(BTRFS_I(inode), + cur_offset, hole_size); + if (err) + break; } next: free_extent_map(em); @@ -4673,24 +4727,24 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) * truncation, it must capture all writes that happened before * this truncation. */ - btrfs_wait_for_snapshot_creation(root); + btrfs_drew_write_lock(&root->snapshot_lock); ret = btrfs_cont_expand(inode, oldsize, newsize); if (ret) { - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); return ret; } trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); return PTR_ERR(trans); } i_size_write(inode, newsize); - btrfs_ordered_update_i_size(inode, i_size_read(inode), NULL); + btrfs_inode_safe_disk_i_size_write(inode, 0); pagecache_isize_extended(inode, oldsize, newsize); ret = btrfs_update_inode(trans, root, inode); - btrfs_end_write_no_snapshotting(root); + btrfs_drew_write_unlock(&root->snapshot_lock); btrfs_end_transaction(trans); } else { @@ -5100,7 +5154,7 @@ static int fixup_tree_root_location(struct btrfs_fs_info *fs_info, btrfs_release_path(path); - new_root = btrfs_read_fs_root_no_name(fs_info, location); + new_root = btrfs_get_fs_root(fs_info, location, true); if (IS_ERR(new_root)) { err = PTR_ERR(new_root); goto out; @@ -5181,7 +5235,8 @@ static int btrfs_init_locked_inode(struct inode *inode, void *p) inode->i_ino = args->location->objectid; memcpy(&BTRFS_I(inode)->location, args->location, sizeof(*args->location)); - BTRFS_I(inode)->root = args->root; + BTRFS_I(inode)->root = btrfs_grab_root(args->root); + BUG_ON(args->root && !BTRFS_I(inode)->root); return 0; } @@ -5262,7 +5317,7 @@ static struct inode *new_simple_dir(struct super_block *s, if (!inode) return ERR_PTR(-ENOMEM); - BTRFS_I(inode)->root = root; + BTRFS_I(inode)->root = btrfs_grab_root(root); memcpy(&BTRFS_I(inode)->location, key, sizeof(*key)); set_bit(BTRFS_INODE_DUMMY, &BTRFS_I(inode)->runtime_flags); @@ -5309,7 +5364,6 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) struct btrfs_root *sub_root = root; struct btrfs_key location; u8 di_type = 0; - int index; int ret = 0; if (dentry->d_name.len > BTRFS_NAME_LEN) @@ -5336,7 +5390,6 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) return inode; } - index = srcu_read_lock(&fs_info->subvol_srcu); ret = fixup_tree_root_location(fs_info, dir, dentry, &location, &sub_root); if (ret < 0) { @@ -5347,7 +5400,8 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) } else { inode = btrfs_iget(dir->i_sb, &location, sub_root); } - srcu_read_unlock(&fs_info->subvol_srcu, index); + if (root != sub_root) + btrfs_put_root(sub_root); if (!IS_ERR(inode) && root != sub_root) { down_read(&fs_info->cleanup_work_sem); @@ -5828,7 +5882,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, */ BTRFS_I(inode)->index_cnt = 2; BTRFS_I(inode)->dir_index = *index; - BTRFS_I(inode)->root = root; + BTRFS_I(inode)->root = btrfs_grab_root(root); BTRFS_I(inode)->generation = trans->transid; inode->i_generation = BTRFS_I(inode)->generation; @@ -6465,6 +6519,7 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, extent_type = btrfs_file_extent_type(leaf, item); extent_start = found_key.offset; + extent_end = btrfs_file_extent_end(path); if (extent_type == BTRFS_FILE_EXTENT_REG || extent_type == BTRFS_FILE_EXTENT_PREALLOC) { /* Only regular file could have regular/prealloc extent */ @@ -6475,18 +6530,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode, btrfs_ino(inode)); goto out; } - extent_end = extent_start + - btrfs_file_extent_num_bytes(leaf, item); - trace_btrfs_get_extent_show_fi_regular(inode, leaf, item, extent_start); } else if (extent_type == BTRFS_FILE_EXTENT_INLINE) { - size_t size; - - size = btrfs_file_extent_ram_bytes(leaf, item); - extent_end = ALIGN(extent_start + size, - fs_info->sectorsize); - trace_btrfs_get_extent_show_fi_inline(inode, leaf, item, path->slots[0], extent_start); @@ -7785,6 +7831,7 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode, { struct btrfs_io_bio *io_bio = btrfs_io_bio(bio); struct btrfs_io_bio *orig_io_bio = btrfs_io_bio(dip->orig_bio); + u16 csum_size; blk_status_t ret; /* @@ -7804,7 +7851,8 @@ static inline blk_status_t btrfs_lookup_and_bind_dio_csum(struct inode *inode, file_offset -= dip->logical_offset; file_offset >>= inode->i_sb->s_blocksize_bits; - io_bio->csum = (u8 *)(((u32 *)orig_io_bio->csum) + file_offset); + csum_size = btrfs_super_csum_size(btrfs_sb(inode->i_sb)->super_copy); + io_bio->csum = orig_io_bio->csum + csum_size * file_offset; return 0; } @@ -8211,9 +8259,7 @@ static int btrfs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, int btrfs_readpage(struct file *file, struct page *page) { - struct extent_io_tree *tree; - tree = &BTRFS_I(page->mapping->host)->io_tree; - return extent_read_full_page(tree, page, btrfs_get_extent, 0); + return extent_read_full_page(page, btrfs_get_extent, 0); } static int btrfs_writepage(struct page *page, struct writeback_control *wbc) @@ -8272,6 +8318,39 @@ static int btrfs_releasepage(struct page *page, gfp_t gfp_flags) return __btrfs_releasepage(page, gfp_flags); } +#ifdef CONFIG_MIGRATION +static int btrfs_migratepage(struct address_space *mapping, + struct page *newpage, struct page *page, + enum migrate_mode mode) +{ + int ret; + + ret = migrate_page_move_mapping(mapping, newpage, page, 0); + if (ret != MIGRATEPAGE_SUCCESS) + return ret; + + if (page_has_private(page)) { + ClearPagePrivate(page); + get_page(newpage); + set_page_private(newpage, page_private(page)); + set_page_private(page, 0); + put_page(page); + SetPagePrivate(newpage); + } + + if (PagePrivate2(page)) { + ClearPagePrivate2(page); + SetPagePrivate2(newpage); + } + + if (mode != MIGRATE_SYNC_NO_COPY) + migrate_page_copy(newpage, page); + else + migrate_page_states(newpage, page); + return MIGRATEPAGE_SUCCESS; +} +#endif + static void btrfs_invalidatepage(struct page *page, unsigned int offset, unsigned int length) { @@ -8647,7 +8726,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) break; } - btrfs_block_rsv_release(fs_info, rsv, -1); + btrfs_block_rsv_release(fs_info, rsv, -1, NULL); ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, min_size, false); BUG_ON(ret); /* shouldn't happen */ @@ -8672,7 +8751,7 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) ret = PTR_ERR(trans); goto out; } - btrfs_ordered_update_i_size(inode, inode->i_size, NULL); + btrfs_inode_safe_disk_i_size_write(inode, 0); } if (trans) { @@ -8776,6 +8855,8 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) extent_io_tree_init(fs_info, &ei->io_tree, IO_TREE_INODE_IO, inode); extent_io_tree_init(fs_info, &ei->io_failure_tree, IO_TREE_INODE_IO_FAILURE, inode); + extent_io_tree_init(fs_info, &ei->file_extent_tree, + IO_TREE_INODE_FILE_EXTENT, inode); ei->io_tree.track_uptodate = true; ei->io_failure_tree.track_uptodate = true; atomic_set(&ei->sync_writers, 0); @@ -8842,6 +8923,8 @@ void btrfs_destroy_inode(struct inode *inode) btrfs_qgroup_check_reserved_leak(inode); inode_tree_del(inode); btrfs_drop_extent_cache(BTRFS_I(inode), 0, (u64)-1, 0); + btrfs_inode_clear_file_extent_range(BTRFS_I(inode), 0, (u64)-1); + btrfs_put_root(BTRFS_I(inode)->root); } int btrfs_drop_inode(struct inode *inode) @@ -9496,6 +9579,10 @@ out_fail: ret = btrfs_sync_log(trans, BTRFS_I(old_inode)->root, &ctx); if (ret) commit_transaction = true; + } else if (sync_log) { + mutex_lock(&root->log_mutex); + list_del(&ctx.list); + mutex_unlock(&root->log_mutex); } if (commit_transaction) { ret = btrfs_commit_transaction(trans); @@ -9665,14 +9752,14 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, int nr) while (!list_empty(&splice) && nr) { root = list_first_entry(&splice, struct btrfs_root, delalloc_root); - root = btrfs_grab_fs_root(root); + root = btrfs_grab_root(root); BUG_ON(!root); list_move_tail(&root->delalloc_root, &fs_info->delalloc_roots); spin_unlock(&fs_info->delalloc_root_lock); ret = start_delalloc_inodes(root, nr, false); - btrfs_put_fs_root(root); + btrfs_put_root(root); if (ret < 0) goto out; @@ -9826,6 +9913,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; u64 cur_offset = start; + u64 clear_offset = start; u64 i_size; u64 cur_bytes; u64 last_alloc = (u64)-1; @@ -9860,6 +9948,15 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, btrfs_end_transaction(trans); break; } + + /* + * We've reserved this space, and thus converted it from + * ->bytes_may_use to ->bytes_reserved. Any error that happens + * from here on out we will only need to clear our reservation + * for the remaining unreserved area, so advance our + * clear_offset by our extent size. + */ + clear_offset += ins.offset; btrfs_dec_block_group_reservations(fs_info, ins.objectid); last_alloc = ins.offset; @@ -9924,7 +10021,7 @@ next: else i_size = cur_offset; i_size_write(inode, i_size); - btrfs_ordered_update_i_size(inode, i_size, NULL); + btrfs_inode_safe_disk_i_size_write(inode, 0); } ret = btrfs_update_inode(trans, root, inode); @@ -9939,9 +10036,9 @@ next: if (own_trans) btrfs_end_transaction(trans); } - if (cur_offset < end) - btrfs_free_reserved_data_space(inode, NULL, cur_offset, - end - cur_offset + 1); + if (clear_offset < end) + btrfs_free_reserved_data_space(inode, NULL, clear_offset, + end - clear_offset + 1); return ret; } @@ -10460,6 +10557,9 @@ static const struct address_space_operations btrfs_aops = { .direct_IO = btrfs_direct_IO, .invalidatepage = btrfs_invalidatepage, .releasepage = btrfs_releasepage, +#ifdef CONFIG_MIGRATION + .migratepage = btrfs_migratepage, +#endif .set_page_dirty = btrfs_set_page_dirty, .error_remove_page = generic_error_remove_page, .swap_activate = btrfs_swap_activate, diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 4f4b13830b25..40b729dce91c 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -28,6 +28,7 @@ #include <linux/iversion.h> #include "ctree.h" #include "disk-io.h" +#include "export.h" #include "transaction.h" #include "btrfs_inode.h" #include "print-tree.h" @@ -86,10 +87,6 @@ struct btrfs_ioctl_send_args_32 { struct btrfs_ioctl_send_args_32) #endif -static int btrfs_clone(struct inode *src, struct inode *inode, - u64 off, u64 olen, u64 olen_aligned, u64 destoff, - int no_time_update); - /* Mask out flags that are inappropriate for the given type of inode. */ static unsigned int btrfs_mask_fsflags_for_type(struct inode *inode, unsigned int flags) @@ -554,7 +551,6 @@ int __pure btrfs_is_empty_uuid(u8 *uuid) static noinline int create_subvol(struct inode *dir, struct dentry *dentry, const char *name, int namelen, - u64 *async_transid, struct btrfs_qgroup_inherit *inherit) { struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); @@ -573,7 +569,6 @@ static noinline int create_subvol(struct inode *dir, u64 objectid; u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID; u64 index = 0; - uuid_le new_uuid; root_item = kzalloc(sizeof(*root_item), GFP_KERNEL); if (!root_item) @@ -643,8 +638,7 @@ static noinline int create_subvol(struct inode *dir, btrfs_set_root_generation_v2(root_item, btrfs_root_generation(root_item)); - uuid_le_gen(&new_uuid); - memcpy(root_item->uuid, new_uuid.b, BTRFS_UUID_SIZE); + generate_random_guid(root_item->uuid); btrfs_set_stack_timespec_sec(&root_item->otime, cur_time.tv_sec); btrfs_set_stack_timespec_nsec(&root_item->otime, cur_time.tv_nsec); root_item->ctime = root_item->otime; @@ -666,7 +660,7 @@ static noinline int create_subvol(struct inode *dir, goto fail; key.offset = (u64)-1; - new_root = btrfs_read_fs_root_no_name(fs_info, &key); + new_root = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(new_root)) { ret = PTR_ERR(new_root); btrfs_abort_transaction(trans, ret); @@ -676,6 +670,7 @@ static noinline int create_subvol(struct inode *dir, btrfs_record_root_in_trans(trans, new_root); ret = btrfs_create_subvol_root(trans, new_root, root, new_dirid); + btrfs_put_root(new_root); if (ret) { /* We potentially lose an unused inode item here */ btrfs_abort_transaction(trans, ret); @@ -727,14 +722,7 @@ fail: trans->bytes_reserved = 0; btrfs_subvolume_release_metadata(fs_info, &block_rsv); - if (async_transid) { - *async_transid = trans->transid; - err = btrfs_commit_transaction_async(trans, 1); - if (err) - err = btrfs_commit_transaction(trans); - } else { - err = btrfs_commit_transaction(trans); - } + err = btrfs_commit_transaction(trans); if (err && !ret) ret = err; @@ -752,8 +740,7 @@ fail_free: } static int create_snapshot(struct btrfs_root *root, struct inode *dir, - struct dentry *dentry, - u64 *async_transid, bool readonly, + struct dentry *dentry, bool readonly, struct btrfs_qgroup_inherit *inherit) { struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); @@ -789,11 +776,7 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, * possible. This is to avoid later writeback (running dealloc) to * fallback to COW mode and unexpectedly fail with ENOSPC. */ - atomic_inc(&root->will_be_snapshotted); - smp_mb__after_atomic(); - /* wait for no snapshot writes */ - wait_event(root->subv_writers->wait, - percpu_counter_sum(&root->subv_writers->counter) == 0); + btrfs_drew_read_lock(&root->snapshot_lock); ret = btrfs_start_delalloc_snapshot(root); if (ret) @@ -841,14 +824,8 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, list_add(&pending_snapshot->list, &trans->transaction->pending_snapshots); spin_unlock(&fs_info->trans_lock); - if (async_transid) { - *async_transid = trans->transid; - ret = btrfs_commit_transaction_async(trans, 1); - if (ret) - ret = btrfs_commit_transaction(trans); - } else { - ret = btrfs_commit_transaction(trans); - } + + ret = btrfs_commit_transaction(trans); if (ret) goto fail; @@ -869,12 +846,13 @@ static int create_snapshot(struct btrfs_root *root, struct inode *dir, d_instantiate(dentry, inode); ret = 0; fail: + btrfs_put_root(pending_snapshot->snap); btrfs_subvolume_release_metadata(fs_info, &pending_snapshot->block_rsv); dec_and_free: if (snapshot_force_cow) atomic_dec(&root->snapshot_force_cow); - if (atomic_dec_and_test(&root->will_be_snapshotted)) - wake_up_var(&root->will_be_snapshotted); + btrfs_drew_read_unlock(&root->snapshot_lock); + free_pending: kfree(pending_snapshot->root_item); btrfs_free_path(pending_snapshot->path); @@ -953,7 +931,7 @@ static inline int btrfs_may_create(struct inode *dir, struct dentry *child) static noinline int btrfs_mksubvol(const struct path *parent, const char *name, int namelen, struct btrfs_root *snap_src, - u64 *async_transid, bool readonly, + bool readonly, struct btrfs_qgroup_inherit *inherit) { struct inode *dir = d_inode(parent->dentry); @@ -989,13 +967,11 @@ static noinline int btrfs_mksubvol(const struct path *parent, if (btrfs_root_refs(&BTRFS_I(dir)->root->root_item) == 0) goto out_up_read; - if (snap_src) { - error = create_snapshot(snap_src, dir, dentry, - async_transid, readonly, inherit); - } else { - error = create_subvol(dir, dentry, name, namelen, - async_transid, inherit); - } + if (snap_src) + error = create_snapshot(snap_src, dir, dentry, readonly, inherit); + else + error = create_subvol(dir, dentry, name, namelen, inherit); + if (!error) fsnotify_mkdir(dir, dentry); out_up_read: @@ -1711,9 +1687,6 @@ static noinline int btrfs_ioctl_resize(struct file *file, new_size = round_down(new_size, fs_info->sectorsize); - btrfs_info_in_rcu(fs_info, "new size for %s is %llu", - rcu_str_deref(device->name), new_size); - if (new_size > old_size) { trans = btrfs_start_transaction(root, 0); if (IS_ERR(trans)) { @@ -1726,6 +1699,11 @@ static noinline int btrfs_ioctl_resize(struct file *file, ret = btrfs_shrink_device(device, new_size); } /* equal, nothing need to do */ + if (ret == 0 && new_size != old_size) + btrfs_info_in_rcu(fs_info, + "resize device %s (devid %llu) from %llu to %llu", + rcu_str_deref(device->name), device->devid, + old_size, new_size); out_free: kfree(vol_args); out: @@ -1734,9 +1712,9 @@ out: return ret; } -static noinline int btrfs_ioctl_snap_create_transid(struct file *file, +static noinline int __btrfs_ioctl_snap_create(struct file *file, const char *name, unsigned long fd, int subvol, - u64 *transid, bool readonly, + bool readonly, struct btrfs_qgroup_inherit *inherit) { int namelen; @@ -1763,7 +1741,7 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file, if (subvol) { ret = btrfs_mksubvol(&file->f_path, name, namelen, - NULL, transid, readonly, inherit); + NULL, readonly, inherit); } else { struct fd src = fdget(fd); struct inode *src_inode; @@ -1786,7 +1764,7 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file, } else { ret = btrfs_mksubvol(&file->f_path, name, namelen, BTRFS_I(src_inode)->root, - transid, readonly, inherit); + readonly, inherit); } fdput(src); } @@ -1810,9 +1788,8 @@ static noinline int btrfs_ioctl_snap_create(struct file *file, return PTR_ERR(vol_args); vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; - ret = btrfs_ioctl_snap_create_transid(file, vol_args->name, - vol_args->fd, subvol, - NULL, false, NULL); + ret = __btrfs_ioctl_snap_create(file, vol_args->name, vol_args->fd, + subvol, false, NULL); kfree(vol_args); return ret; @@ -1823,8 +1800,6 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file, { struct btrfs_ioctl_vol_args_v2 *vol_args; int ret; - u64 transid = 0; - u64 *ptr = NULL; bool readonly = false; struct btrfs_qgroup_inherit *inherit = NULL; @@ -1836,22 +1811,11 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file, return PTR_ERR(vol_args); vol_args->name[BTRFS_SUBVOL_NAME_MAX] = '\0'; - if (vol_args->flags & - ~(BTRFS_SUBVOL_CREATE_ASYNC | BTRFS_SUBVOL_RDONLY | - BTRFS_SUBVOL_QGROUP_INHERIT)) { + if (vol_args->flags & ~BTRFS_SUBVOL_CREATE_ARGS_MASK) { ret = -EOPNOTSUPP; goto free_args; } - if (vol_args->flags & BTRFS_SUBVOL_CREATE_ASYNC) { - struct inode *inode = file_inode(file); - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - - btrfs_warn(fs_info, -"SNAP_CREATE_V2 ioctl with CREATE_ASYNC is deprecated and will be removed in kernel 5.7"); - - ptr = &transid; - } if (vol_args->flags & BTRFS_SUBVOL_RDONLY) readonly = true; if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) { @@ -1866,18 +1830,10 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file, } } - ret = btrfs_ioctl_snap_create_transid(file, vol_args->name, - vol_args->fd, subvol, ptr, - readonly, inherit); + ret = __btrfs_ioctl_snap_create(file, vol_args->name, vol_args->fd, + subvol, readonly, inherit); if (ret) goto free_inherit; - - if (ptr && copy_to_user(arg + - offsetof(struct btrfs_ioctl_vol_args_v2, - transid), - ptr, sizeof(*ptr))) - ret = -EFAULT; - free_inherit: kfree(inherit); free_args: @@ -1936,11 +1892,6 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file, goto out_drop_write; } - if (flags & BTRFS_SUBVOL_CREATE_ASYNC) { - ret = -EINVAL; - goto out_drop_write; - } - if (flags & ~BTRFS_SUBVOL_RDONLY) { ret = -EOPNOTSUPP; goto out_drop_write; @@ -2174,12 +2125,12 @@ static noinline int search_ioctl(struct inode *inode, if (sk->tree_id == 0) { /* search the root of the inode that was passed */ - root = BTRFS_I(inode)->root; + root = btrfs_grab_root(BTRFS_I(inode)->root); } else { key.objectid = sk->tree_id; key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - root = btrfs_read_fs_root_no_name(info, &key); + root = btrfs_get_fs_root(info, &key, true); if (IS_ERR(root)) { btrfs_free_path(path); return PTR_ERR(root); @@ -2208,6 +2159,7 @@ static noinline int search_ioctl(struct inode *inode, ret = 0; err: sk->nr_items = num_found; + btrfs_put_root(root); btrfs_free_path(path); return ret; } @@ -2314,9 +2266,10 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, key.objectid = tree_id; key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - root = btrfs_read_fs_root_no_name(info, &key); + root = btrfs_get_fs_root(info, &key, true); if (IS_ERR(root)) { ret = PTR_ERR(root); + root = NULL; goto out; } @@ -2367,6 +2320,7 @@ static noinline int btrfs_search_path_in_tree(struct btrfs_fs_info *info, name[total_len] = '\0'; ret = 0; out: + btrfs_put_root(root); btrfs_free_path(path); return ret; } @@ -2383,7 +2337,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, unsigned long item_len; struct btrfs_inode_ref *iref; struct btrfs_root_ref *rref; - struct btrfs_root *root; + struct btrfs_root *root = NULL; struct btrfs_path *path; struct btrfs_key key, key2; struct extent_buffer *leaf; @@ -2408,7 +2362,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, key.objectid = treeid; key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - root = btrfs_read_fs_root_no_name(fs_info, &key); + root = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(root)) { ret = PTR_ERR(root); goto out; @@ -2420,15 +2374,15 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, while (1) { ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) { - goto out; + goto out_put; } else if (ret > 0) { ret = btrfs_previous_item(root, path, dirid, BTRFS_INODE_REF_KEY); if (ret < 0) { - goto out; + goto out_put; } else if (ret > 0) { ret = -ENOENT; - goto out; + goto out_put; } } @@ -2442,7 +2396,7 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, total_len += len + 1; if (ptr < args->path) { ret = -ENAMETOOLONG; - goto out; + goto out_put; } *(ptr + len) = '/'; @@ -2453,10 +2407,10 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, ret = btrfs_previous_item(root, path, dirid, BTRFS_INODE_ITEM_KEY); if (ret < 0) { - goto out; + goto out_put; } else if (ret > 0) { ret = -ENOENT; - goto out; + goto out_put; } leaf = path->nodes[0]; @@ -2464,26 +2418,26 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, btrfs_item_key_to_cpu(leaf, &key2, slot); if (key2.objectid != dirid) { ret = -ENOENT; - goto out; + goto out_put; } temp_inode = btrfs_iget(sb, &key2, root); if (IS_ERR(temp_inode)) { ret = PTR_ERR(temp_inode); - goto out; + goto out_put; } ret = inode_permission(temp_inode, MAY_READ | MAY_EXEC); iput(temp_inode); if (ret) { ret = -EACCES; - goto out; + goto out_put; } if (key.offset == upper_limit.objectid) break; if (key.objectid == BTRFS_FIRST_FREE_OBJECTID) { ret = -EACCES; - goto out; + goto out_put; } btrfs_release_path(path); @@ -2494,15 +2448,16 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, memmove(args->path, ptr, total_len); args->path[total_len] = '\0'; + btrfs_put_root(root); + root = NULL; btrfs_release_path(path); } /* Get the bottom subvolume's name from ROOT_REF */ - root = fs_info->tree_root; key.objectid = treeid; key.type = BTRFS_ROOT_REF_KEY; key.offset = args->treeid; - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0); if (ret < 0) { goto out; } else if (ret > 0) { @@ -2529,6 +2484,8 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, read_extent_buffer(leaf, args->name, item_off, item_len); args->name[item_len] = 0; +out_put: + btrfs_put_root(root); out: btrfs_free_path(path); return ret; @@ -2653,10 +2610,10 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp) key.objectid = BTRFS_I(inode)->root->root_key.objectid; key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - root = btrfs_read_fs_root_no_name(fs_info, &key); + root = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(root)) { ret = PTR_ERR(root); - goto out; + goto out_free; } root_item = &root->root_item; @@ -2689,16 +2646,14 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp) if (key.objectid != BTRFS_FS_TREE_OBJECTID) { /* Search root tree for ROOT_BACKREF of this subvolume */ - root = fs_info->tree_root; - key.type = BTRFS_ROOT_BACKREF_KEY; key.offset = 0; - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + ret = btrfs_search_slot(NULL, fs_info->tree_root, &key, path, 0, 0); if (ret < 0) { goto out; } else if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) { - ret = btrfs_next_leaf(root, path); + ret = btrfs_next_leaf(fs_info->tree_root, path); if (ret < 0) { goto out; } else if (ret > 0) { @@ -2733,6 +2688,8 @@ static int btrfs_ioctl_get_subvol_info(struct file *file, void __user *argp) ret = -EFAULT; out: + btrfs_put_root(root); +out_free: btrfs_free_path(path); kzfree(subvol_info); return ret; @@ -2836,7 +2793,8 @@ out: } static noinline int btrfs_ioctl_snap_destroy(struct file *file, - void __user *arg) + void __user *arg, + bool destroy_v2) { struct dentry *parent = file->f_path.dentry; struct btrfs_fs_info *fs_info = btrfs_sb(parent->d_sb); @@ -2845,34 +2803,120 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, struct inode *inode; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_root *dest = NULL; - struct btrfs_ioctl_vol_args *vol_args; - int namelen; + struct btrfs_ioctl_vol_args *vol_args = NULL; + struct btrfs_ioctl_vol_args_v2 *vol_args2 = NULL; + char *subvol_name, *subvol_name_ptr = NULL; + int subvol_namelen; int err = 0; + bool destroy_parent = false; - if (!S_ISDIR(dir->i_mode)) - return -ENOTDIR; + if (destroy_v2) { + vol_args2 = memdup_user(arg, sizeof(*vol_args2)); + if (IS_ERR(vol_args2)) + return PTR_ERR(vol_args2); - vol_args = memdup_user(arg, sizeof(*vol_args)); - if (IS_ERR(vol_args)) - return PTR_ERR(vol_args); + if (vol_args2->flags & ~BTRFS_SUBVOL_DELETE_ARGS_MASK) { + err = -EOPNOTSUPP; + goto out; + } - vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; - namelen = strlen(vol_args->name); - if (strchr(vol_args->name, '/') || - strncmp(vol_args->name, "..", namelen) == 0) { - err = -EINVAL; - goto out; + /* + * If SPEC_BY_ID is not set, we are looking for the subvolume by + * name, same as v1 currently does. + */ + if (!(vol_args2->flags & BTRFS_SUBVOL_SPEC_BY_ID)) { + vol_args2->name[BTRFS_SUBVOL_NAME_MAX] = 0; + subvol_name = vol_args2->name; + + err = mnt_want_write_file(file); + if (err) + goto out; + } else { + if (vol_args2->subvolid < BTRFS_FIRST_FREE_OBJECTID) { + err = -EINVAL; + goto out; + } + + err = mnt_want_write_file(file); + if (err) + goto out; + + dentry = btrfs_get_dentry(fs_info->sb, + BTRFS_FIRST_FREE_OBJECTID, + vol_args2->subvolid, 0, 0); + if (IS_ERR(dentry)) { + err = PTR_ERR(dentry); + goto out_drop_write; + } + + /* + * Change the default parent since the subvolume being + * deleted can be outside of the current mount point. + */ + parent = btrfs_get_parent(dentry); + + /* + * At this point dentry->d_name can point to '/' if the + * subvolume we want to destroy is outsite of the + * current mount point, so we need to release the + * current dentry and execute the lookup to return a new + * one with ->d_name pointing to the + * <mount point>/subvol_name. + */ + dput(dentry); + if (IS_ERR(parent)) { + err = PTR_ERR(parent); + goto out_drop_write; + } + dir = d_inode(parent); + + /* + * If v2 was used with SPEC_BY_ID, a new parent was + * allocated since the subvolume can be outside of the + * current mount point. Later on we need to release this + * new parent dentry. + */ + destroy_parent = true; + + subvol_name_ptr = btrfs_get_subvol_name_from_objectid( + fs_info, vol_args2->subvolid); + if (IS_ERR(subvol_name_ptr)) { + err = PTR_ERR(subvol_name_ptr); + goto free_parent; + } + /* subvol_name_ptr is already NULL termined */ + subvol_name = (char *)kbasename(subvol_name_ptr); + } + } else { + vol_args = memdup_user(arg, sizeof(*vol_args)); + if (IS_ERR(vol_args)) + return PTR_ERR(vol_args); + + vol_args->name[BTRFS_PATH_NAME_MAX] = 0; + subvol_name = vol_args->name; + + err = mnt_want_write_file(file); + if (err) + goto out; } - err = mnt_want_write_file(file); - if (err) - goto out; + subvol_namelen = strlen(subvol_name); + + if (strchr(subvol_name, '/') || + strncmp(subvol_name, "..", subvol_namelen) == 0) { + err = -EINVAL; + goto free_subvol_name; + } + if (!S_ISDIR(dir->i_mode)) { + err = -ENOTDIR; + goto free_subvol_name; + } err = down_write_killable_nested(&dir->i_rwsem, I_MUTEX_PARENT); if (err == -EINTR) - goto out_drop_write; - dentry = lookup_one_len(vol_args->name, parent, namelen); + goto free_subvol_name; + dentry = lookup_one_len(subvol_name, parent, subvol_namelen); if (IS_ERR(dentry)) { err = PTR_ERR(dentry); goto out_unlock_dir; @@ -2941,9 +2985,15 @@ out_dput: dput(dentry); out_unlock_dir: inode_unlock(dir); +free_subvol_name: + kfree(subvol_name_ptr); +free_parent: + if (destroy_parent) + dput(parent); out_drop_write: mnt_drop_write_file(file); out: + kfree(vol_args2); kfree(vol_args); return err; } @@ -3069,8 +3119,7 @@ static long btrfs_ioctl_rm_dev_v2(struct file *file, void __user *arg) goto err_drop; } - /* Check for compatibility reject unknown flags */ - if (vol_args->flags & ~BTRFS_VOL_ARG_V2_FLAGS_SUPPORTED) { + if (vol_args->flags & ~BTRFS_DEVICE_REMOVE_ARGS_MASK) { ret = -EOPNOTSUPP; goto out; } @@ -3220,733 +3269,6 @@ out: return ret; } -static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1, - struct inode *inode2, u64 loff2, u64 len) -{ - unlock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1); - unlock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1); -} - -static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1, - struct inode *inode2, u64 loff2, u64 len) -{ - if (inode1 < inode2) { - swap(inode1, inode2); - swap(loff1, loff2); - } else if (inode1 == inode2 && loff2 < loff1) { - swap(loff1, loff2); - } - lock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1); - lock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1); -} - -static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 len, - struct inode *dst, u64 dst_loff) -{ - const u64 bs = BTRFS_I(src)->root->fs_info->sb->s_blocksize; - int ret; - - /* - * Lock destination range to serialize with concurrent readpages() and - * source range to serialize with relocation. - */ - btrfs_double_extent_lock(src, loff, dst, dst_loff, len); - ret = btrfs_clone(src, dst, loff, len, ALIGN(len, bs), dst_loff, 1); - btrfs_double_extent_unlock(src, loff, dst, dst_loff, len); - - return ret; -} - -#define BTRFS_MAX_DEDUPE_LEN SZ_16M - -static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen, - struct inode *dst, u64 dst_loff) -{ - int ret; - u64 i, tail_len, chunk_count; - struct btrfs_root *root_dst = BTRFS_I(dst)->root; - - spin_lock(&root_dst->root_item_lock); - if (root_dst->send_in_progress) { - btrfs_warn_rl(root_dst->fs_info, -"cannot deduplicate to root %llu while send operations are using it (%d in progress)", - root_dst->root_key.objectid, - root_dst->send_in_progress); - spin_unlock(&root_dst->root_item_lock); - return -EAGAIN; - } - root_dst->dedupe_in_progress++; - spin_unlock(&root_dst->root_item_lock); - - tail_len = olen % BTRFS_MAX_DEDUPE_LEN; - chunk_count = div_u64(olen, BTRFS_MAX_DEDUPE_LEN); - - for (i = 0; i < chunk_count; i++) { - ret = btrfs_extent_same_range(src, loff, BTRFS_MAX_DEDUPE_LEN, - dst, dst_loff); - if (ret) - goto out; - - loff += BTRFS_MAX_DEDUPE_LEN; - dst_loff += BTRFS_MAX_DEDUPE_LEN; - } - - if (tail_len > 0) - ret = btrfs_extent_same_range(src, loff, tail_len, dst, - dst_loff); -out: - spin_lock(&root_dst->root_item_lock); - root_dst->dedupe_in_progress--; - spin_unlock(&root_dst->root_item_lock); - - return ret; -} - -static int clone_finish_inode_update(struct btrfs_trans_handle *trans, - struct inode *inode, - u64 endoff, - const u64 destoff, - const u64 olen, - int no_time_update) -{ - struct btrfs_root *root = BTRFS_I(inode)->root; - int ret; - - inode_inc_iversion(inode); - if (!no_time_update) - inode->i_mtime = inode->i_ctime = current_time(inode); - /* - * We round up to the block size at eof when determining which - * extents to clone above, but shouldn't round up the file size. - */ - if (endoff > destoff + olen) - endoff = destoff + olen; - if (endoff > inode->i_size) - btrfs_i_size_write(BTRFS_I(inode), endoff); - - ret = btrfs_update_inode(trans, root, inode); - if (ret) { - btrfs_abort_transaction(trans, ret); - btrfs_end_transaction(trans); - goto out; - } - ret = btrfs_end_transaction(trans); -out: - return ret; -} - -/* - * Make sure we do not end up inserting an inline extent into a file that has - * already other (non-inline) extents. If a file has an inline extent it can - * not have any other extents and the (single) inline extent must start at the - * file offset 0. Failing to respect these rules will lead to file corruption, - * resulting in EIO errors on read/write operations, hitting BUG_ON's in mm, etc - * - * We can have extents that have been already written to disk or we can have - * dirty ranges still in delalloc, in which case the extent maps and items are - * created only when we run delalloc, and the delalloc ranges might fall outside - * the range we are currently locking in the inode's io tree. So we check the - * inode's i_size because of that (i_size updates are done while holding the - * i_mutex, which we are holding here). - * We also check to see if the inode has a size not greater than "datal" but has - * extents beyond it, due to an fallocate with FALLOC_FL_KEEP_SIZE (and we are - * protected against such concurrent fallocate calls by the i_mutex). - * - * If the file has no extents but a size greater than datal, do not allow the - * copy because we would need turn the inline extent into a non-inline one (even - * with NO_HOLES enabled). If we find our destination inode only has one inline - * extent, just overwrite it with the source inline extent if its size is less - * than the source extent's size, or we could copy the source inline extent's - * data into the destination inode's inline extent if the later is greater then - * the former. - */ -static int clone_copy_inline_extent(struct inode *dst, - struct btrfs_trans_handle *trans, - struct btrfs_path *path, - struct btrfs_key *new_key, - const u64 drop_start, - const u64 datal, - const u64 skip, - const u64 size, - char *inline_data) -{ - struct btrfs_fs_info *fs_info = btrfs_sb(dst->i_sb); - struct btrfs_root *root = BTRFS_I(dst)->root; - const u64 aligned_end = ALIGN(new_key->offset + datal, - fs_info->sectorsize); - int ret; - struct btrfs_key key; - - if (new_key->offset > 0) - return -EOPNOTSUPP; - - key.objectid = btrfs_ino(BTRFS_I(dst)); - key.type = BTRFS_EXTENT_DATA_KEY; - key.offset = 0; - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); - if (ret < 0) { - return ret; - } else if (ret > 0) { - if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) { - ret = btrfs_next_leaf(root, path); - if (ret < 0) - return ret; - else if (ret > 0) - goto copy_inline_extent; - } - btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); - if (key.objectid == btrfs_ino(BTRFS_I(dst)) && - key.type == BTRFS_EXTENT_DATA_KEY) { - ASSERT(key.offset > 0); - return -EOPNOTSUPP; - } - } else if (i_size_read(dst) <= datal) { - struct btrfs_file_extent_item *ei; - u64 ext_len; - - /* - * If the file size is <= datal, make sure there are no other - * extents following (can happen do to an fallocate call with - * the flag FALLOC_FL_KEEP_SIZE). - */ - ei = btrfs_item_ptr(path->nodes[0], path->slots[0], - struct btrfs_file_extent_item); - /* - * If it's an inline extent, it can not have other extents - * following it. - */ - if (btrfs_file_extent_type(path->nodes[0], ei) == - BTRFS_FILE_EXTENT_INLINE) - goto copy_inline_extent; - - ext_len = btrfs_file_extent_num_bytes(path->nodes[0], ei); - if (ext_len > aligned_end) - return -EOPNOTSUPP; - - ret = btrfs_next_item(root, path); - if (ret < 0) { - return ret; - } else if (ret == 0) { - btrfs_item_key_to_cpu(path->nodes[0], &key, - path->slots[0]); - if (key.objectid == btrfs_ino(BTRFS_I(dst)) && - key.type == BTRFS_EXTENT_DATA_KEY) - return -EOPNOTSUPP; - } - } - -copy_inline_extent: - /* - * We have no extent items, or we have an extent at offset 0 which may - * or may not be inlined. All these cases are dealt the same way. - */ - if (i_size_read(dst) > datal) { - /* - * If the destination inode has an inline extent... - * This would require copying the data from the source inline - * extent into the beginning of the destination's inline extent. - * But this is really complex, both extents can be compressed - * or just one of them, which would require decompressing and - * re-compressing data (which could increase the new compressed - * size, not allowing the compressed data to fit anymore in an - * inline extent). - * So just don't support this case for now (it should be rare, - * we are not really saving space when cloning inline extents). - */ - return -EOPNOTSUPP; - } - - btrfs_release_path(path); - ret = btrfs_drop_extents(trans, root, dst, drop_start, aligned_end, 1); - if (ret) - return ret; - ret = btrfs_insert_empty_item(trans, root, path, new_key, size); - if (ret) - return ret; - - if (skip) { - const u32 start = btrfs_file_extent_calc_inline_size(0); - - memmove(inline_data + start, inline_data + start + skip, datal); - } - - write_extent_buffer(path->nodes[0], inline_data, - btrfs_item_ptr_offset(path->nodes[0], - path->slots[0]), - size); - inode_add_bytes(dst, datal); - set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(dst)->runtime_flags); - - return 0; -} - -/** - * btrfs_clone() - clone a range from inode file to another - * - * @src: Inode to clone from - * @inode: Inode to clone to - * @off: Offset within source to start clone from - * @olen: Original length, passed by user, of range to clone - * @olen_aligned: Block-aligned value of olen - * @destoff: Offset within @inode to start clone - * @no_time_update: Whether to update mtime/ctime on the target inode - */ -static int btrfs_clone(struct inode *src, struct inode *inode, - const u64 off, const u64 olen, const u64 olen_aligned, - const u64 destoff, int no_time_update) -{ - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_root *root = BTRFS_I(inode)->root; - struct btrfs_path *path = NULL; - struct extent_buffer *leaf; - struct btrfs_trans_handle *trans; - char *buf = NULL; - struct btrfs_key key; - u32 nritems; - int slot; - int ret; - const u64 len = olen_aligned; - u64 last_dest_end = destoff; - - ret = -ENOMEM; - buf = kvmalloc(fs_info->nodesize, GFP_KERNEL); - if (!buf) - return ret; - - path = btrfs_alloc_path(); - if (!path) { - kvfree(buf); - return ret; - } - - path->reada = READA_FORWARD; - /* clone data */ - key.objectid = btrfs_ino(BTRFS_I(src)); - key.type = BTRFS_EXTENT_DATA_KEY; - key.offset = off; - - while (1) { - u64 next_key_min_offset = key.offset + 1; - struct btrfs_file_extent_item *extent; - int type; - u32 size; - struct btrfs_key new_key; - u64 disko = 0, diskl = 0; - u64 datao = 0, datal = 0; - u8 comp; - u64 drop_start; - - /* - * note the key will change type as we walk through the - * tree. - */ - path->leave_spinning = 1; - ret = btrfs_search_slot(NULL, BTRFS_I(src)->root, &key, path, - 0, 0); - if (ret < 0) - goto out; - /* - * First search, if no extent item that starts at offset off was - * found but the previous item is an extent item, it's possible - * it might overlap our target range, therefore process it. - */ - if (key.offset == off && ret > 0 && path->slots[0] > 0) { - btrfs_item_key_to_cpu(path->nodes[0], &key, - path->slots[0] - 1); - if (key.type == BTRFS_EXTENT_DATA_KEY) - path->slots[0]--; - } - - nritems = btrfs_header_nritems(path->nodes[0]); -process_slot: - if (path->slots[0] >= nritems) { - ret = btrfs_next_leaf(BTRFS_I(src)->root, path); - if (ret < 0) - goto out; - if (ret > 0) - break; - nritems = btrfs_header_nritems(path->nodes[0]); - } - leaf = path->nodes[0]; - slot = path->slots[0]; - - btrfs_item_key_to_cpu(leaf, &key, slot); - if (key.type > BTRFS_EXTENT_DATA_KEY || - key.objectid != btrfs_ino(BTRFS_I(src))) - break; - - ASSERT(key.type == BTRFS_EXTENT_DATA_KEY); - - extent = btrfs_item_ptr(leaf, slot, - struct btrfs_file_extent_item); - comp = btrfs_file_extent_compression(leaf, extent); - type = btrfs_file_extent_type(leaf, extent); - if (type == BTRFS_FILE_EXTENT_REG || - type == BTRFS_FILE_EXTENT_PREALLOC) { - disko = btrfs_file_extent_disk_bytenr(leaf, extent); - diskl = btrfs_file_extent_disk_num_bytes(leaf, extent); - datao = btrfs_file_extent_offset(leaf, extent); - datal = btrfs_file_extent_num_bytes(leaf, extent); - } else if (type == BTRFS_FILE_EXTENT_INLINE) { - /* Take upper bound, may be compressed */ - datal = btrfs_file_extent_ram_bytes(leaf, extent); - } - - /* - * The first search might have left us at an extent item that - * ends before our target range's start, can happen if we have - * holes and NO_HOLES feature enabled. - */ - if (key.offset + datal <= off) { - path->slots[0]++; - goto process_slot; - } else if (key.offset >= off + len) { - break; - } - next_key_min_offset = key.offset + datal; - size = btrfs_item_size_nr(leaf, slot); - read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, slot), - size); - - btrfs_release_path(path); - path->leave_spinning = 0; - - memcpy(&new_key, &key, sizeof(new_key)); - new_key.objectid = btrfs_ino(BTRFS_I(inode)); - if (off <= key.offset) - new_key.offset = key.offset + destoff - off; - else - new_key.offset = destoff; - - /* - * Deal with a hole that doesn't have an extent item that - * represents it (NO_HOLES feature enabled). - * This hole is either in the middle of the cloning range or at - * the beginning (fully overlaps it or partially overlaps it). - */ - if (new_key.offset != last_dest_end) - drop_start = last_dest_end; - else - drop_start = new_key.offset; - - if (type == BTRFS_FILE_EXTENT_REG || - type == BTRFS_FILE_EXTENT_PREALLOC) { - struct btrfs_clone_extent_info clone_info; - - /* - * a | --- range to clone ---| b - * | ------------- extent ------------- | - */ - - /* Subtract range b */ - if (key.offset + datal > off + len) - datal = off + len - key.offset; - - /* Subtract range a */ - if (off > key.offset) { - datao += off - key.offset; - datal -= off - key.offset; - } - - clone_info.disk_offset = disko; - clone_info.disk_len = diskl; - clone_info.data_offset = datao; - clone_info.data_len = datal; - clone_info.file_offset = new_key.offset; - clone_info.extent_buf = buf; - clone_info.item_size = size; - ret = btrfs_punch_hole_range(inode, path, - drop_start, - new_key.offset + datal - 1, - &clone_info, &trans); - if (ret) - goto out; - } else if (type == BTRFS_FILE_EXTENT_INLINE) { - u64 skip = 0; - u64 trim = 0; - - if (off > key.offset) { - skip = off - key.offset; - new_key.offset += skip; - } - - if (key.offset + datal > off + len) - trim = key.offset + datal - (off + len); - - if (comp && (skip || trim)) { - ret = -EINVAL; - goto out; - } - size -= skip + trim; - datal -= skip + trim; - - /* - * If our extent is inline, we know we will drop or - * adjust at most 1 extent item in the destination root. - * - * 1 - adjusting old extent (we may have to split it) - * 1 - add new extent - * 1 - inode update - */ - trans = btrfs_start_transaction(root, 3); - if (IS_ERR(trans)) { - ret = PTR_ERR(trans); - goto out; - } - - ret = clone_copy_inline_extent(inode, trans, path, - &new_key, drop_start, - datal, skip, size, buf); - if (ret) { - if (ret != -EOPNOTSUPP) - btrfs_abort_transaction(trans, ret); - btrfs_end_transaction(trans); - goto out; - } - } - - btrfs_release_path(path); - - last_dest_end = ALIGN(new_key.offset + datal, - fs_info->sectorsize); - ret = clone_finish_inode_update(trans, inode, last_dest_end, - destoff, olen, no_time_update); - if (ret) - goto out; - if (new_key.offset + datal >= destoff + len) - break; - - btrfs_release_path(path); - key.offset = next_key_min_offset; - - if (fatal_signal_pending(current)) { - ret = -EINTR; - goto out; - } - } - ret = 0; - - if (last_dest_end < destoff + len) { - /* - * We have an implicit hole that fully or partially overlaps our - * cloning range at its end. This means that we either have the - * NO_HOLES feature enabled or the implicit hole happened due to - * mixing buffered and direct IO writes against this file. - */ - btrfs_release_path(path); - path->leave_spinning = 0; - - ret = btrfs_punch_hole_range(inode, path, - last_dest_end, destoff + len - 1, - NULL, &trans); - if (ret) - goto out; - - ret = clone_finish_inode_update(trans, inode, destoff + len, - destoff, olen, no_time_update); - } - -out: - btrfs_free_path(path); - kvfree(buf); - return ret; -} - -static noinline int btrfs_clone_files(struct file *file, struct file *file_src, - u64 off, u64 olen, u64 destoff) -{ - struct inode *inode = file_inode(file); - struct inode *src = file_inode(file_src); - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - int ret; - u64 len = olen; - u64 bs = fs_info->sb->s_blocksize; - - /* - * TODO: - * - split compressed inline extents. annoying: we need to - * decompress into destination's address_space (the file offset - * may change, so source mapping won't do), then recompress (or - * otherwise reinsert) a subrange. - * - * - split destination inode's inline extents. The inline extents can - * be either compressed or non-compressed. - */ - - /* - * VFS's generic_remap_file_range_prep() protects us from cloning the - * eof block into the middle of a file, which would result in corruption - * if the file size is not blocksize aligned. So we don't need to check - * for that case here. - */ - if (off + len == src->i_size) - len = ALIGN(src->i_size, bs) - off; - - if (destoff > inode->i_size) { - const u64 wb_start = ALIGN_DOWN(inode->i_size, bs); - - ret = btrfs_cont_expand(inode, inode->i_size, destoff); - if (ret) - return ret; - /* - * We may have truncated the last block if the inode's size is - * not sector size aligned, so we need to wait for writeback to - * complete before proceeding further, otherwise we can race - * with cloning and attempt to increment a reference to an - * extent that no longer exists (writeback completed right after - * we found the previous extent covering eof and before we - * attempted to increment its reference count). - */ - ret = btrfs_wait_ordered_range(inode, wb_start, - destoff - wb_start); - if (ret) - return ret; - } - - /* - * Lock destination range to serialize with concurrent readpages() and - * source range to serialize with relocation. - */ - btrfs_double_extent_lock(src, off, inode, destoff, len); - ret = btrfs_clone(src, inode, off, olen, len, destoff, 0); - btrfs_double_extent_unlock(src, off, inode, destoff, len); - /* - * Truncate page cache pages so that future reads will see the cloned - * data immediately and not the previous data. - */ - truncate_inode_pages_range(&inode->i_data, - round_down(destoff, PAGE_SIZE), - round_up(destoff + len, PAGE_SIZE) - 1); - - return ret; -} - -static int btrfs_remap_file_range_prep(struct file *file_in, loff_t pos_in, - struct file *file_out, loff_t pos_out, - loff_t *len, unsigned int remap_flags) -{ - struct inode *inode_in = file_inode(file_in); - struct inode *inode_out = file_inode(file_out); - u64 bs = BTRFS_I(inode_out)->root->fs_info->sb->s_blocksize; - bool same_inode = inode_out == inode_in; - u64 wb_len; - int ret; - - if (!(remap_flags & REMAP_FILE_DEDUP)) { - struct btrfs_root *root_out = BTRFS_I(inode_out)->root; - - if (btrfs_root_readonly(root_out)) - return -EROFS; - - if (file_in->f_path.mnt != file_out->f_path.mnt || - inode_in->i_sb != inode_out->i_sb) - return -EXDEV; - } - - /* don't make the dst file partly checksummed */ - if ((BTRFS_I(inode_in)->flags & BTRFS_INODE_NODATASUM) != - (BTRFS_I(inode_out)->flags & BTRFS_INODE_NODATASUM)) { - return -EINVAL; - } - - /* - * Now that the inodes are locked, we need to start writeback ourselves - * and can not rely on the writeback from the VFS's generic helper - * generic_remap_file_range_prep() because: - * - * 1) For compression we must call filemap_fdatawrite_range() range - * twice (btrfs_fdatawrite_range() does it for us), and the generic - * helper only calls it once; - * - * 2) filemap_fdatawrite_range(), called by the generic helper only - * waits for the writeback to complete, i.e. for IO to be done, and - * not for the ordered extents to complete. We need to wait for them - * to complete so that new file extent items are in the fs tree. - */ - if (*len == 0 && !(remap_flags & REMAP_FILE_DEDUP)) - wb_len = ALIGN(inode_in->i_size, bs) - ALIGN_DOWN(pos_in, bs); - else - wb_len = ALIGN(*len, bs); - - /* - * Since we don't lock ranges, wait for ongoing lockless dio writes (as - * any in progress could create its ordered extents after we wait for - * existing ordered extents below). - */ - inode_dio_wait(inode_in); - if (!same_inode) - inode_dio_wait(inode_out); - - /* - * Workaround to make sure NOCOW buffered write reach disk as NOCOW. - * - * Btrfs' back references do not have a block level granularity, they - * work at the whole extent level. - * NOCOW buffered write without data space reserved may not be able - * to fall back to CoW due to lack of data space, thus could cause - * data loss. - * - * Here we take a shortcut by flushing the whole inode, so that all - * nocow write should reach disk as nocow before we increase the - * reference of the extent. We could do better by only flushing NOCOW - * data, but that needs extra accounting. - * - * Also we don't need to check ASYNC_EXTENT, as async extent will be - * CoWed anyway, not affecting nocow part. - */ - ret = filemap_flush(inode_in->i_mapping); - if (ret < 0) - return ret; - - ret = btrfs_wait_ordered_range(inode_in, ALIGN_DOWN(pos_in, bs), - wb_len); - if (ret < 0) - return ret; - ret = btrfs_wait_ordered_range(inode_out, ALIGN_DOWN(pos_out, bs), - wb_len); - if (ret < 0) - return ret; - - return generic_remap_file_range_prep(file_in, pos_in, file_out, pos_out, - len, remap_flags); -} - -loff_t btrfs_remap_file_range(struct file *src_file, loff_t off, - struct file *dst_file, loff_t destoff, loff_t len, - unsigned int remap_flags) -{ - struct inode *src_inode = file_inode(src_file); - struct inode *dst_inode = file_inode(dst_file); - bool same_inode = dst_inode == src_inode; - int ret; - - if (remap_flags & ~(REMAP_FILE_DEDUP | REMAP_FILE_ADVISORY)) - return -EINVAL; - - if (same_inode) - inode_lock(src_inode); - else - lock_two_nondirectories(src_inode, dst_inode); - - ret = btrfs_remap_file_range_prep(src_file, off, dst_file, destoff, - &len, remap_flags); - if (ret < 0 || len == 0) - goto out_unlock; - - if (remap_flags & REMAP_FILE_DEDUP) - ret = btrfs_extent_same(src_inode, off, len, dst_inode, destoff); - else - ret = btrfs_clone_files(dst_file, src_file, off, len, destoff); - -out_unlock: - if (same_inode) - inode_unlock(src_inode); - else - unlock_two_nondirectories(src_inode, dst_inode); - - return ret < 0 ? ret : len; -} - static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) { struct inode *inode = file_inode(file); @@ -3955,7 +3277,7 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) struct btrfs_root *new_root; struct btrfs_dir_item *di; struct btrfs_trans_handle *trans; - struct btrfs_path *path; + struct btrfs_path *path = NULL; struct btrfs_key location; struct btrfs_disk_key disk_key; u64 objectid = 0; @@ -3981,49 +3303,51 @@ static long btrfs_ioctl_default_subvol(struct file *file, void __user *argp) location.type = BTRFS_ROOT_ITEM_KEY; location.offset = (u64)-1; - new_root = btrfs_read_fs_root_no_name(fs_info, &location); + new_root = btrfs_get_fs_root(fs_info, &location, true); if (IS_ERR(new_root)) { ret = PTR_ERR(new_root); goto out; } if (!is_fstree(new_root->root_key.objectid)) { ret = -ENOENT; - goto out; + goto out_free; } path = btrfs_alloc_path(); if (!path) { ret = -ENOMEM; - goto out; + goto out_free; } path->leave_spinning = 1; trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { - btrfs_free_path(path); ret = PTR_ERR(trans); - goto out; + goto out_free; } dir_id = btrfs_super_root_dir(fs_info->super_copy); di = btrfs_lookup_dir_item(trans, fs_info->tree_root, path, dir_id, "default", 7, 1); if (IS_ERR_OR_NULL(di)) { - btrfs_free_path(path); + btrfs_release_path(path); btrfs_end_transaction(trans); btrfs_err(fs_info, "Umm, you don't have the default diritem, this isn't going to work"); ret = -ENOENT; - goto out; + goto out_free; } btrfs_cpu_key_to_disk(&disk_key, &new_root->root_key); btrfs_set_dir_item_key(path->nodes[0], di, &disk_key); btrfs_mark_buffer_dirty(path->nodes[0]); - btrfs_free_path(path); + btrfs_release_path(path); btrfs_set_fs_incompat(fs_info, DEFAULT_SUBVOL); btrfs_end_transaction(trans); +out_free: + btrfs_put_root(new_root); + btrfs_free_path(path); out: mnt_drop_write_file(file); return ret; @@ -5465,7 +4789,9 @@ long btrfs_ioctl(struct file *file, unsigned int case BTRFS_IOC_SUBVOL_CREATE_V2: return btrfs_ioctl_snap_create_v2(file, argp, 1); case BTRFS_IOC_SNAP_DESTROY: - return btrfs_ioctl_snap_destroy(file, argp); + return btrfs_ioctl_snap_destroy(file, argp, false); + case BTRFS_IOC_SNAP_DESTROY_V2: + return btrfs_ioctl_snap_destroy(file, argp, true); case BTRFS_IOC_SUBVOL_GETFLAGS: return btrfs_ioctl_subvol_getflags(file, argp); case BTRFS_IOC_SUBVOL_SETFLAGS: diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 571c4826c428..fb647d8cf527 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -523,3 +523,138 @@ void btrfs_unlock_up_safe(struct btrfs_path *path, int level) path->locks[i] = 0; } } + +/* + * Loop around taking references on and locking the root node of the tree until + * we end up with a lock on the root node. + * + * Return: root extent buffer with write lock held + */ +struct extent_buffer *btrfs_lock_root_node(struct btrfs_root *root) +{ + struct extent_buffer *eb; + + while (1) { + eb = btrfs_root_node(root); + btrfs_tree_lock(eb); + if (eb == root->node) + break; + btrfs_tree_unlock(eb); + free_extent_buffer(eb); + } + return eb; +} + +/* + * Loop around taking references on and locking the root node of the tree until + * we end up with a lock on the root node. + * + * Return: root extent buffer with read lock held + */ +struct extent_buffer *btrfs_read_lock_root_node(struct btrfs_root *root) +{ + struct extent_buffer *eb; + + while (1) { + eb = btrfs_root_node(root); + btrfs_tree_read_lock(eb); + if (eb == root->node) + break; + btrfs_tree_read_unlock(eb); + free_extent_buffer(eb); + } + return eb; +} + +/* + * DREW locks + * ========== + * + * DREW stands for double-reader-writer-exclusion lock. It's used in situation + * where you want to provide A-B exclusion but not AA or BB. + * + * Currently implementation gives more priority to reader. If a reader and a + * writer both race to acquire their respective sides of the lock the writer + * would yield its lock as soon as it detects a concurrent reader. Additionally + * if there are pending readers no new writers would be allowed to come in and + * acquire the lock. + */ + +int btrfs_drew_lock_init(struct btrfs_drew_lock *lock) +{ + int ret; + + ret = percpu_counter_init(&lock->writers, 0, GFP_KERNEL); + if (ret) + return ret; + + atomic_set(&lock->readers, 0); + init_waitqueue_head(&lock->pending_readers); + init_waitqueue_head(&lock->pending_writers); + + return 0; +} + +void btrfs_drew_lock_destroy(struct btrfs_drew_lock *lock) +{ + percpu_counter_destroy(&lock->writers); +} + +/* Return true if acquisition is successful, false otherwise */ +bool btrfs_drew_try_write_lock(struct btrfs_drew_lock *lock) +{ + if (atomic_read(&lock->readers)) + return false; + + percpu_counter_inc(&lock->writers); + + /* Ensure writers count is updated before we check for pending readers */ + smp_mb(); + if (atomic_read(&lock->readers)) { + btrfs_drew_write_unlock(lock); + return false; + } + + return true; +} + +void btrfs_drew_write_lock(struct btrfs_drew_lock *lock) +{ + while (true) { + if (btrfs_drew_try_write_lock(lock)) + return; + wait_event(lock->pending_writers, !atomic_read(&lock->readers)); + } +} + +void btrfs_drew_write_unlock(struct btrfs_drew_lock *lock) +{ + percpu_counter_dec(&lock->writers); + cond_wake_up(&lock->pending_readers); +} + +void btrfs_drew_read_lock(struct btrfs_drew_lock *lock) +{ + atomic_inc(&lock->readers); + + /* + * Ensure the pending reader count is perceieved BEFORE this reader + * goes to sleep in case of active writers. This guarantees new writers + * won't be allowed and that the current reader will be woken up when + * the last active writer finishes its jobs. + */ + smp_mb__after_atomic(); + + wait_event(lock->pending_readers, + percpu_counter_sum(&lock->writers) == 0); +} + +void btrfs_drew_read_unlock(struct btrfs_drew_lock *lock) +{ + /* + * atomic_dec_and_test implies a full barrier, so woken up writers + * are guaranteed to see the decrement + */ + if (atomic_dec_and_test(&lock->readers)) + wake_up(&lock->pending_writers); +} diff --git a/fs/btrfs/locking.h b/fs/btrfs/locking.h index 21a285883e89..d715846c10b8 100644 --- a/fs/btrfs/locking.h +++ b/fs/btrfs/locking.h @@ -6,6 +6,9 @@ #ifndef BTRFS_LOCKING_H #define BTRFS_LOCKING_H +#include <linux/atomic.h> +#include <linux/wait.h> +#include <linux/percpu_counter.h> #include "extent_io.h" #define BTRFS_WRITE_LOCK 1 @@ -13,6 +16,8 @@ #define BTRFS_WRITE_LOCK_BLOCKING 3 #define BTRFS_READ_LOCK_BLOCKING 4 +struct btrfs_path; + void btrfs_tree_lock(struct extent_buffer *eb); void btrfs_tree_unlock(struct extent_buffer *eb); @@ -48,4 +53,19 @@ static inline void btrfs_tree_unlock_rw(struct extent_buffer *eb, int rw) BUG(); } +struct btrfs_drew_lock { + atomic_t readers; + struct percpu_counter writers; + wait_queue_head_t pending_writers; + wait_queue_head_t pending_readers; +}; + +int btrfs_drew_lock_init(struct btrfs_drew_lock *lock); +void btrfs_drew_lock_destroy(struct btrfs_drew_lock *lock); +void btrfs_drew_write_lock(struct btrfs_drew_lock *lock); +bool btrfs_drew_try_write_lock(struct btrfs_drew_lock *lock); +void btrfs_drew_write_unlock(struct btrfs_drew_lock *lock); +void btrfs_drew_read_lock(struct btrfs_drew_lock *lock); +void btrfs_drew_read_unlock(struct btrfs_drew_lock *lock); + #endif diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index ecb9fb6a6fe0..e13b3d28c063 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -580,7 +580,7 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr, while (!list_empty(&splice) && nr) { root = list_first_entry(&splice, struct btrfs_root, ordered_root); - root = btrfs_grab_fs_root(root); + root = btrfs_grab_root(root); BUG_ON(!root); list_move_tail(&root->ordered_root, &fs_info->ordered_roots); @@ -588,7 +588,7 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr, done = btrfs_wait_ordered_extents(root, nr, range_start, range_len); - btrfs_put_fs_root(root); + btrfs_put_root(root); spin_lock(&fs_info->ordered_root_lock); if (nr != U64_MAX) { @@ -679,10 +679,15 @@ int btrfs_wait_ordered_range(struct inode *inode, u64 start, u64 len) } btrfs_start_ordered_extent(inode, ordered, 1); end = ordered->file_offset; + /* + * If the ordered extent had an error save the error but don't + * exit without waiting first for all other ordered extents in + * the range to complete. + */ if (test_bit(BTRFS_ORDERED_IOERR, &ordered->flags)) ret = -EIO; btrfs_put_ordered_extent(ordered); - if (ret || end == 0 || end == start) + if (end == 0 || end == start) break; end--; } @@ -781,134 +786,6 @@ out: } /* - * After an extent is done, call this to conditionally update the on disk - * i_size. i_size is updated to cover any fully written part of the file. - */ -int btrfs_ordered_update_i_size(struct inode *inode, u64 offset, - struct btrfs_ordered_extent *ordered) -{ - struct btrfs_ordered_inode_tree *tree = &BTRFS_I(inode)->ordered_tree; - u64 disk_i_size; - u64 new_i_size; - u64 i_size = i_size_read(inode); - struct rb_node *node; - struct rb_node *prev = NULL; - struct btrfs_ordered_extent *test; - int ret = 1; - u64 orig_offset = offset; - - spin_lock_irq(&tree->lock); - if (ordered) { - offset = entry_end(ordered); - if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered->flags)) - offset = min(offset, - ordered->file_offset + - ordered->truncated_len); - } else { - offset = ALIGN(offset, btrfs_inode_sectorsize(inode)); - } - disk_i_size = BTRFS_I(inode)->disk_i_size; - - /* - * truncate file. - * If ordered is not NULL, then this is called from endio and - * disk_i_size will be updated by either truncate itself or any - * in-flight IOs which are inside the disk_i_size. - * - * Because btrfs_setsize() may set i_size with disk_i_size if truncate - * fails somehow, we need to make sure we have a precise disk_i_size by - * updating it as usual. - * - */ - if (!ordered && disk_i_size > i_size) { - BTRFS_I(inode)->disk_i_size = orig_offset; - ret = 0; - goto out; - } - - /* - * if the disk i_size is already at the inode->i_size, or - * this ordered extent is inside the disk i_size, we're done - */ - if (disk_i_size == i_size) - goto out; - - /* - * We still need to update disk_i_size if outstanding_isize is greater - * than disk_i_size. - */ - if (offset <= disk_i_size && - (!ordered || ordered->outstanding_isize <= disk_i_size)) - goto out; - - /* - * walk backward from this ordered extent to disk_i_size. - * if we find an ordered extent then we can't update disk i_size - * yet - */ - if (ordered) { - node = rb_prev(&ordered->rb_node); - } else { - prev = tree_search(tree, offset); - /* - * we insert file extents without involving ordered struct, - * so there should be no ordered struct cover this offset - */ - if (prev) { - test = rb_entry(prev, struct btrfs_ordered_extent, - rb_node); - BUG_ON(offset_in_entry(test, offset)); - } - node = prev; - } - for (; node; node = rb_prev(node)) { - test = rb_entry(node, struct btrfs_ordered_extent, rb_node); - - /* We treat this entry as if it doesn't exist */ - if (test_bit(BTRFS_ORDERED_UPDATED_ISIZE, &test->flags)) - continue; - - if (entry_end(test) <= disk_i_size) - break; - if (test->file_offset >= i_size) - break; - - /* - * We don't update disk_i_size now, so record this undealt - * i_size. Or we will not know the real i_size. - */ - if (test->outstanding_isize < offset) - test->outstanding_isize = offset; - if (ordered && - ordered->outstanding_isize > test->outstanding_isize) - test->outstanding_isize = ordered->outstanding_isize; - goto out; - } - new_i_size = min_t(u64, offset, i_size); - - /* - * Some ordered extents may completed before the current one, and - * we hold the real i_size in ->outstanding_isize. - */ - if (ordered && ordered->outstanding_isize > new_i_size) - new_i_size = min_t(u64, ordered->outstanding_isize, i_size); - BTRFS_I(inode)->disk_i_size = new_i_size; - ret = 0; -out: - /* - * We need to do this because we can't remove ordered extents until - * after the i_disk_size has been updated and then the inode has been - * updated to reflect the change, so we need to tell anybody who finds - * this ordered extent that we've already done all the real work, we - * just haven't completed all the other work. - */ - if (ordered) - set_bit(BTRFS_ORDERED_UPDATED_ISIZE, &ordered->flags); - spin_unlock_irq(&tree->lock); - return ret; -} - -/* * search the ordered extents for one corresponding to 'offset' and * try to find a checksum. This is used because we allow pages to * be reclaimed before their checksum is actually put into the btree @@ -958,7 +835,6 @@ out: * btrfs_flush_ordered_range - Lock the passed range and ensures all pending * ordered extents in it are run to completion. * - * @tree: IO tree used for locking out other users of the range * @inode: Inode whose ordered tree is to be searched * @start: Beginning of range to flush * @end: Last byte of range to lock @@ -968,8 +844,7 @@ out: * This function always returns with the given range locked, ensuring after it's * called no order extent can be pending. */ -void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree, - struct btrfs_inode *inode, u64 start, +void btrfs_lock_and_flush_ordered_range(struct btrfs_inode *inode, u64 start, u64 end, struct extent_state **cached_state) { @@ -981,7 +856,7 @@ void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree, cachedp = cached_state; while (1) { - lock_extent_bits(tree, start, end, cachedp); + lock_extent_bits(&inode->io_tree, start, end, cachedp); ordered = btrfs_lookup_ordered_range(inode, start, end - start + 1); if (!ordered) { @@ -994,7 +869,7 @@ void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree, refcount_dec(&cache->refs); break; } - unlock_extent_cached(tree, start, end, cachedp); + unlock_extent_cached(&inode->io_tree, start, end, cachedp); btrfs_start_ordered_extent(&inode->vfs_inode, ordered, 1); btrfs_put_ordered_extent(ordered); } diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h index 3beb4da4ab41..c01c9698250b 100644 --- a/fs/btrfs/ordered-data.h +++ b/fs/btrfs/ordered-data.h @@ -52,11 +52,6 @@ enum { BTRFS_ORDERED_DIRECT, /* We had an io error when writing this out */ BTRFS_ORDERED_IOERR, - /* - * indicates whether this ordered extent has done its due diligence in - * updating the isize - */ - BTRFS_ORDERED_UPDATED_ISIZE, /* Set when we have to truncate an extent */ BTRFS_ORDERED_TRUNCATED, /* Regular IO for COW */ @@ -182,16 +177,13 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_range( struct btrfs_inode *inode, u64 file_offset, u64 len); -int btrfs_ordered_update_i_size(struct inode *inode, u64 offset, - struct btrfs_ordered_extent *ordered); int btrfs_find_ordered_sum(struct inode *inode, u64 offset, u64 disk_bytenr, u8 *sum, int len); u64 btrfs_wait_ordered_extents(struct btrfs_root *root, u64 nr, const u64 range_start, const u64 range_len); void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr, const u64 range_start, const u64 range_len); -void btrfs_lock_and_flush_ordered_range(struct extent_io_tree *tree, - struct btrfs_inode *inode, u64 start, +void btrfs_lock_and_flush_ordered_range(struct btrfs_inode *inode, u64 start, u64 end, struct extent_state **cached_state); int __init ordered_data_init(void); diff --git a/fs/btrfs/props.c b/fs/btrfs/props.c index deb59e7cfcac..ff1ff90e48b1 100644 --- a/fs/btrfs/props.c +++ b/fs/btrfs/props.c @@ -383,7 +383,7 @@ static int inherit_props(struct btrfs_trans_handle *trans, if (need_reserve) { btrfs_block_rsv_release(fs_info, trans->block_rsv, - num_bytes); + num_bytes, NULL); if (ret) return ret; } diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 98d9a50352d6..c3888fb367e7 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -1030,6 +1030,7 @@ out_add_root: ret = qgroup_rescan_init(fs_info, 0, 1); if (!ret) { qgroup_rescan_zero_tracking(fs_info); + fs_info->qgroup_rescan_running = true; btrfs_queue_work(fs_info->qgroup_rescan_workers, &fs_info->qgroup_rescan_work); } @@ -1037,11 +1038,8 @@ out_add_root: out_free_path: btrfs_free_path(path); out_free_root: - if (ret) { - free_extent_buffer(quota_root->node); - free_extent_buffer(quota_root->commit_root); - kfree(quota_root); - } + if (ret) + btrfs_put_root(quota_root); out: if (ret) { ulist_free(fs_info->qgroup_ulist); @@ -1104,9 +1102,7 @@ int btrfs_quota_disable(struct btrfs_fs_info *fs_info) btrfs_tree_unlock(quota_root->node); btrfs_free_tree_block(trans, quota_root, quota_root->node, 0, 1); - free_extent_buffer(quota_root->node); - free_extent_buffer(quota_root->commit_root); - kfree(quota_root); + btrfs_put_root(quota_root); end_trans: ret = btrfs_end_transaction(trans); @@ -3237,7 +3233,6 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, } mutex_lock(&fs_info->qgroup_rescan_lock); - spin_lock(&fs_info->qgroup_lock); if (init_flags) { if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) { @@ -3252,7 +3247,6 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, } if (ret) { - spin_unlock(&fs_info->qgroup_lock); mutex_unlock(&fs_info->qgroup_rescan_lock); return ret; } @@ -3263,9 +3257,6 @@ qgroup_rescan_init(struct btrfs_fs_info *fs_info, u64 progress_objectid, sizeof(fs_info->qgroup_rescan_progress)); fs_info->qgroup_rescan_progress.objectid = progress_objectid; init_completion(&fs_info->qgroup_rescan_completion); - fs_info->qgroup_rescan_running = true; - - spin_unlock(&fs_info->qgroup_lock); mutex_unlock(&fs_info->qgroup_rescan_lock); btrfs_init_work(&fs_info->qgroup_rescan_work, @@ -3326,8 +3317,11 @@ btrfs_qgroup_rescan(struct btrfs_fs_info *fs_info) qgroup_rescan_zero_tracking(fs_info); + mutex_lock(&fs_info->qgroup_rescan_lock); + fs_info->qgroup_rescan_running = true; btrfs_queue_work(fs_info->qgroup_rescan_workers, &fs_info->qgroup_rescan_work); + mutex_unlock(&fs_info->qgroup_rescan_lock); return 0; } @@ -3339,9 +3333,7 @@ int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info, int ret = 0; mutex_lock(&fs_info->qgroup_rescan_lock); - spin_lock(&fs_info->qgroup_lock); running = fs_info->qgroup_rescan_running; - spin_unlock(&fs_info->qgroup_lock); mutex_unlock(&fs_info->qgroup_rescan_lock); if (!running) @@ -3363,9 +3355,13 @@ int btrfs_qgroup_wait_for_completion(struct btrfs_fs_info *fs_info, void btrfs_qgroup_rescan_resume(struct btrfs_fs_info *fs_info) { - if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) + if (fs_info->qgroup_flags & BTRFS_QGROUP_STATUS_FLAG_RESCAN) { + mutex_lock(&fs_info->qgroup_rescan_lock); + fs_info->qgroup_rescan_running = true; btrfs_queue_work(fs_info->qgroup_rescan_workers, &fs_info->qgroup_rescan_work); + mutex_unlock(&fs_info->qgroup_rescan_lock); + } } /* @@ -4002,3 +3998,16 @@ out: } return ret; } + +void btrfs_qgroup_destroy_extent_records(struct btrfs_transaction *trans) +{ + struct btrfs_qgroup_extent_record *entry; + struct btrfs_qgroup_extent_record *next; + struct rb_root *root; + + root = &trans->delayed_refs.dirty_extent_root; + rbtree_postorder_for_each_entry_safe(entry, next, root, node) { + ulist_free(entry->old_roots); + kfree(entry); + } +} diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h index 236f12224d52..1bc654459469 100644 --- a/fs/btrfs/qgroup.h +++ b/fs/btrfs/qgroup.h @@ -414,5 +414,6 @@ int btrfs_qgroup_add_swapped_blocks(struct btrfs_trans_handle *trans, u64 last_snapshot); int btrfs_qgroup_trace_subtree_after_cow(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct extent_buffer *eb); +void btrfs_qgroup_destroy_extent_records(struct btrfs_transaction *trans); #endif diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index a8e53c8e7b01..c870ef70f817 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -206,7 +206,6 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info) struct btrfs_stripe_hash *h; int num_entries = 1 << BTRFS_STRIPE_HASH_TABLE_BITS; int i; - int table_size; if (info->stripe_hash_table) return 0; @@ -218,8 +217,7 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info) * Try harder to allocate and fallback to vmalloc to lower the chance * of a failing mount. */ - table_size = sizeof(*table) + sizeof(*h) * num_entries; - table = kvzalloc(table_size, GFP_KERNEL); + table = kvzalloc(struct_size(table, table, num_entries), GFP_KERNEL); if (!table) return -ENOMEM; @@ -1196,22 +1194,19 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio) int nr_data = rbio->nr_data; int stripe; int pagenr; - int p_stripe = -1; - int q_stripe = -1; + bool has_qstripe; struct bio_list bio_list; struct bio *bio; int ret; bio_list_init(&bio_list); - if (rbio->real_stripes - rbio->nr_data == 1) { - p_stripe = rbio->real_stripes - 1; - } else if (rbio->real_stripes - rbio->nr_data == 2) { - p_stripe = rbio->real_stripes - 2; - q_stripe = rbio->real_stripes - 1; - } else { + if (rbio->real_stripes - rbio->nr_data == 1) + has_qstripe = false; + else if (rbio->real_stripes - rbio->nr_data == 2) + has_qstripe = true; + else BUG(); - } /* at this point we either have a full stripe, * or we've read the full stripe from the drive. @@ -1255,7 +1250,7 @@ static noinline void finish_rmw(struct btrfs_raid_bio *rbio) SetPageUptodate(p); pointers[stripe++] = kmap(p); - if (q_stripe != -1) { + if (has_qstripe) { /* * raid6, add the qstripe and call the @@ -2353,8 +2348,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, int nr_data = rbio->nr_data; int stripe; int pagenr; - int p_stripe = -1; - int q_stripe = -1; + bool has_qstripe; struct page *p_page = NULL; struct page *q_page = NULL; struct bio_list bio_list; @@ -2364,14 +2358,12 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, bio_list_init(&bio_list); - if (rbio->real_stripes - rbio->nr_data == 1) { - p_stripe = rbio->real_stripes - 1; - } else if (rbio->real_stripes - rbio->nr_data == 2) { - p_stripe = rbio->real_stripes - 2; - q_stripe = rbio->real_stripes - 1; - } else { + if (rbio->real_stripes - rbio->nr_data == 1) + has_qstripe = false; + else if (rbio->real_stripes - rbio->nr_data == 2) + has_qstripe = true; + else BUG(); - } if (bbio->num_tgtdevs && bbio->tgtdev_map[rbio->scrubp]) { is_replace = 1; @@ -2393,7 +2385,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, goto cleanup; SetPageUptodate(p_page); - if (q_stripe != -1) { + if (has_qstripe) { q_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM); if (!q_page) { __free_page(p_page); @@ -2416,8 +2408,7 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, /* then add the parity stripe */ pointers[stripe++] = kmap(p_page); - if (q_stripe != -1) { - + if (has_qstripe) { /* * raid6, add the qstripe and call the * library function to fill in our p/q diff --git a/fs/btrfs/rcu-string.h b/fs/btrfs/rcu-string.h index a97dc74a4d3d..5c1a617eb25d 100644 --- a/fs/btrfs/rcu-string.h +++ b/fs/btrfs/rcu-string.h @@ -8,7 +8,7 @@ struct rcu_string { struct rcu_head rcu; - char str[0]; + char str[]; }; static inline struct rcu_string *rcu_string_strdup(const char *src, gfp_t mask) diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c index 454a1015d026..7887317033c9 100644 --- a/fs/btrfs/ref-verify.c +++ b/fs/btrfs/ref-verify.c @@ -803,6 +803,15 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info, kfree(ref); kfree(ra); goto out_unlock; + } else if (be->num_refs == 0) { + btrfs_err(fs_info, + "trying to do action %d for a bytenr that has 0 total references", + action); + dump_block_entry(fs_info, be); + dump_ref_action(fs_info, ra); + kfree(ref); + kfree(ra); + goto out_unlock; } if (!parent) { diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c new file mode 100644 index 000000000000..d1973141d3bb --- /dev/null +++ b/fs/btrfs/reflink.c @@ -0,0 +1,804 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <linux/blkdev.h> +#include <linux/iversion.h> +#include "compression.h" +#include "ctree.h" +#include "delalloc-space.h" +#include "reflink.h" +#include "transaction.h" + +#define BTRFS_MAX_DEDUPE_LEN SZ_16M + +static int clone_finish_inode_update(struct btrfs_trans_handle *trans, + struct inode *inode, + u64 endoff, + const u64 destoff, + const u64 olen, + int no_time_update) +{ + struct btrfs_root *root = BTRFS_I(inode)->root; + int ret; + + inode_inc_iversion(inode); + if (!no_time_update) + inode->i_mtime = inode->i_ctime = current_time(inode); + /* + * We round up to the block size at eof when determining which + * extents to clone above, but shouldn't round up the file size. + */ + if (endoff > destoff + olen) + endoff = destoff + olen; + if (endoff > inode->i_size) { + i_size_write(inode, endoff); + btrfs_inode_safe_disk_i_size_write(inode, 0); + } + + ret = btrfs_update_inode(trans, root, inode); + if (ret) { + btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); + goto out; + } + ret = btrfs_end_transaction(trans); +out: + return ret; +} + +static int copy_inline_to_page(struct inode *inode, + const u64 file_offset, + char *inline_data, + const u64 size, + const u64 datal, + const u8 comp_type) +{ + const u64 block_size = btrfs_inode_sectorsize(inode); + const u64 range_end = file_offset + block_size - 1; + const size_t inline_size = size - btrfs_file_extent_calc_inline_size(0); + char *data_start = inline_data + btrfs_file_extent_calc_inline_size(0); + struct extent_changeset *data_reserved = NULL; + struct page *page = NULL; + int ret; + + ASSERT(IS_ALIGNED(file_offset, block_size)); + + /* + * We have flushed and locked the ranges of the source and destination + * inodes, we also have locked the inodes, so we are safe to do a + * reservation here. Also we must not do the reservation while holding + * a transaction open, otherwise we would deadlock. + */ + ret = btrfs_delalloc_reserve_space(inode, &data_reserved, file_offset, + block_size); + if (ret) + goto out; + + page = find_or_create_page(inode->i_mapping, file_offset >> PAGE_SHIFT, + btrfs_alloc_write_mask(inode->i_mapping)); + if (!page) { + ret = -ENOMEM; + goto out_unlock; + } + + set_page_extent_mapped(page); + clear_extent_bit(&BTRFS_I(inode)->io_tree, file_offset, range_end, + EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, + 0, 0, NULL); + ret = btrfs_set_extent_delalloc(inode, file_offset, range_end, 0, NULL); + if (ret) + goto out_unlock; + + if (comp_type == BTRFS_COMPRESS_NONE) { + char *map; + + map = kmap(page); + memcpy(map, data_start, datal); + flush_dcache_page(page); + kunmap(page); + } else { + ret = btrfs_decompress(comp_type, data_start, page, 0, + inline_size, datal); + if (ret) + goto out_unlock; + flush_dcache_page(page); + } + + /* + * If our inline data is smaller then the block/page size, then the + * remaining of the block/page is equivalent to zeroes. We had something + * like the following done: + * + * $ xfs_io -f -c "pwrite -S 0xab 0 500" file + * $ sync # (or fsync) + * $ xfs_io -c "falloc 0 4K" file + * $ xfs_io -c "pwrite -S 0xcd 4K 4K" + * + * So what's in the range [500, 4095] corresponds to zeroes. + */ + if (datal < block_size) { + char *map; + + map = kmap(page); + memset(map + datal, 0, block_size - datal); + flush_dcache_page(page); + kunmap(page); + } + + SetPageUptodate(page); + ClearPageChecked(page); + set_page_dirty(page); +out_unlock: + if (page) { + unlock_page(page); + put_page(page); + } + if (ret) + btrfs_delalloc_release_space(inode, data_reserved, file_offset, + block_size, true); + btrfs_delalloc_release_extents(BTRFS_I(inode), block_size); +out: + extent_changeset_free(data_reserved); + + return ret; +} + +/* + * Deal with cloning of inline extents. We try to copy the inline extent from + * the source inode to destination inode when possible. When not possible we + * copy the inline extent's data into the respective page of the inode. + */ +static int clone_copy_inline_extent(struct inode *dst, + struct btrfs_path *path, + struct btrfs_key *new_key, + const u64 drop_start, + const u64 datal, + const u64 size, + const u8 comp_type, + char *inline_data, + struct btrfs_trans_handle **trans_out) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(dst->i_sb); + struct btrfs_root *root = BTRFS_I(dst)->root; + const u64 aligned_end = ALIGN(new_key->offset + datal, + fs_info->sectorsize); + struct btrfs_trans_handle *trans = NULL; + int ret; + struct btrfs_key key; + + if (new_key->offset > 0) { + ret = copy_inline_to_page(dst, new_key->offset, inline_data, + size, datal, comp_type); + goto out; + } + + key.objectid = btrfs_ino(BTRFS_I(dst)); + key.type = BTRFS_EXTENT_DATA_KEY; + key.offset = 0; + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) { + return ret; + } else if (ret > 0) { + if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) { + ret = btrfs_next_leaf(root, path); + if (ret < 0) + return ret; + else if (ret > 0) + goto copy_inline_extent; + } + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); + if (key.objectid == btrfs_ino(BTRFS_I(dst)) && + key.type == BTRFS_EXTENT_DATA_KEY) { + /* + * There's an implicit hole at file offset 0, copy the + * inline extent's data to the page. + */ + ASSERT(key.offset > 0); + ret = copy_inline_to_page(dst, new_key->offset, + inline_data, size, datal, + comp_type); + goto out; + } + } else if (i_size_read(dst) <= datal) { + struct btrfs_file_extent_item *ei; + + ei = btrfs_item_ptr(path->nodes[0], path->slots[0], + struct btrfs_file_extent_item); + /* + * If it's an inline extent replace it with the source inline + * extent, otherwise copy the source inline extent data into + * the respective page at the destination inode. + */ + if (btrfs_file_extent_type(path->nodes[0], ei) == + BTRFS_FILE_EXTENT_INLINE) + goto copy_inline_extent; + + ret = copy_inline_to_page(dst, new_key->offset, inline_data, + size, datal, comp_type); + goto out; + } + +copy_inline_extent: + ret = 0; + /* + * We have no extent items, or we have an extent at offset 0 which may + * or may not be inlined. All these cases are dealt the same way. + */ + if (i_size_read(dst) > datal) { + /* + * At the destination offset 0 we have either a hole, a regular + * extent or an inline extent larger then the one we want to + * clone. Deal with all these cases by copying the inline extent + * data into the respective page at the destination inode. + */ + ret = copy_inline_to_page(dst, new_key->offset, inline_data, + size, datal, comp_type); + goto out; + } + + btrfs_release_path(path); + /* + * If we end up here it means were copy the inline extent into a leaf + * of the destination inode. We know we will drop or adjust at most one + * extent item in the destination root. + * + * 1 unit - adjusting old extent (we may have to split it) + * 1 unit - add new extent + * 1 unit - inode update + */ + trans = btrfs_start_transaction(root, 3); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + trans = NULL; + goto out; + } + ret = btrfs_drop_extents(trans, root, dst, drop_start, aligned_end, 1); + if (ret) + goto out; + ret = btrfs_insert_empty_item(trans, root, path, new_key, size); + if (ret) + goto out; + + write_extent_buffer(path->nodes[0], inline_data, + btrfs_item_ptr_offset(path->nodes[0], + path->slots[0]), + size); + inode_add_bytes(dst, datal); + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, &BTRFS_I(dst)->runtime_flags); +out: + if (!ret && !trans) { + /* + * No transaction here means we copied the inline extent into a + * page of the destination inode. + * + * 1 unit to update inode item + */ + trans = btrfs_start_transaction(root, 1); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + trans = NULL; + } + } + if (ret && trans) { + btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); + } + if (!ret) + *trans_out = trans; + + return ret; +} + +/** + * btrfs_clone() - clone a range from inode file to another + * + * @src: Inode to clone from + * @inode: Inode to clone to + * @off: Offset within source to start clone from + * @olen: Original length, passed by user, of range to clone + * @olen_aligned: Block-aligned value of olen + * @destoff: Offset within @inode to start clone + * @no_time_update: Whether to update mtime/ctime on the target inode + */ +static int btrfs_clone(struct inode *src, struct inode *inode, + const u64 off, const u64 olen, const u64 olen_aligned, + const u64 destoff, int no_time_update) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + struct btrfs_path *path = NULL; + struct extent_buffer *leaf; + struct btrfs_trans_handle *trans; + char *buf = NULL; + struct btrfs_key key; + u32 nritems; + int slot; + int ret; + const u64 len = olen_aligned; + u64 last_dest_end = destoff; + + ret = -ENOMEM; + buf = kvmalloc(fs_info->nodesize, GFP_KERNEL); + if (!buf) + return ret; + + path = btrfs_alloc_path(); + if (!path) { + kvfree(buf); + return ret; + } + + path->reada = READA_FORWARD; + /* Clone data */ + key.objectid = btrfs_ino(BTRFS_I(src)); + key.type = BTRFS_EXTENT_DATA_KEY; + key.offset = off; + + while (1) { + u64 next_key_min_offset = key.offset + 1; + struct btrfs_file_extent_item *extent; + int type; + u32 size; + struct btrfs_key new_key; + u64 disko = 0, diskl = 0; + u64 datao = 0, datal = 0; + u8 comp; + u64 drop_start; + + /* Note the key will change type as we walk through the tree */ + path->leave_spinning = 1; + ret = btrfs_search_slot(NULL, BTRFS_I(src)->root, &key, path, + 0, 0); + if (ret < 0) + goto out; + /* + * First search, if no extent item that starts at offset off was + * found but the previous item is an extent item, it's possible + * it might overlap our target range, therefore process it. + */ + if (key.offset == off && ret > 0 && path->slots[0] > 0) { + btrfs_item_key_to_cpu(path->nodes[0], &key, + path->slots[0] - 1); + if (key.type == BTRFS_EXTENT_DATA_KEY) + path->slots[0]--; + } + + nritems = btrfs_header_nritems(path->nodes[0]); +process_slot: + if (path->slots[0] >= nritems) { + ret = btrfs_next_leaf(BTRFS_I(src)->root, path); + if (ret < 0) + goto out; + if (ret > 0) + break; + nritems = btrfs_header_nritems(path->nodes[0]); + } + leaf = path->nodes[0]; + slot = path->slots[0]; + + btrfs_item_key_to_cpu(leaf, &key, slot); + if (key.type > BTRFS_EXTENT_DATA_KEY || + key.objectid != btrfs_ino(BTRFS_I(src))) + break; + + ASSERT(key.type == BTRFS_EXTENT_DATA_KEY); + + extent = btrfs_item_ptr(leaf, slot, + struct btrfs_file_extent_item); + comp = btrfs_file_extent_compression(leaf, extent); + type = btrfs_file_extent_type(leaf, extent); + if (type == BTRFS_FILE_EXTENT_REG || + type == BTRFS_FILE_EXTENT_PREALLOC) { + disko = btrfs_file_extent_disk_bytenr(leaf, extent); + diskl = btrfs_file_extent_disk_num_bytes(leaf, extent); + datao = btrfs_file_extent_offset(leaf, extent); + datal = btrfs_file_extent_num_bytes(leaf, extent); + } else if (type == BTRFS_FILE_EXTENT_INLINE) { + /* Take upper bound, may be compressed */ + datal = btrfs_file_extent_ram_bytes(leaf, extent); + } + + /* + * The first search might have left us at an extent item that + * ends before our target range's start, can happen if we have + * holes and NO_HOLES feature enabled. + */ + if (key.offset + datal <= off) { + path->slots[0]++; + goto process_slot; + } else if (key.offset >= off + len) { + break; + } + next_key_min_offset = key.offset + datal; + size = btrfs_item_size_nr(leaf, slot); + read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, slot), + size); + + btrfs_release_path(path); + path->leave_spinning = 0; + + memcpy(&new_key, &key, sizeof(new_key)); + new_key.objectid = btrfs_ino(BTRFS_I(inode)); + if (off <= key.offset) + new_key.offset = key.offset + destoff - off; + else + new_key.offset = destoff; + + /* + * Deal with a hole that doesn't have an extent item that + * represents it (NO_HOLES feature enabled). + * This hole is either in the middle of the cloning range or at + * the beginning (fully overlaps it or partially overlaps it). + */ + if (new_key.offset != last_dest_end) + drop_start = last_dest_end; + else + drop_start = new_key.offset; + + if (type == BTRFS_FILE_EXTENT_REG || + type == BTRFS_FILE_EXTENT_PREALLOC) { + struct btrfs_clone_extent_info clone_info; + + /* + * a | --- range to clone ---| b + * | ------------- extent ------------- | + */ + + /* Subtract range b */ + if (key.offset + datal > off + len) + datal = off + len - key.offset; + + /* Subtract range a */ + if (off > key.offset) { + datao += off - key.offset; + datal -= off - key.offset; + } + + clone_info.disk_offset = disko; + clone_info.disk_len = diskl; + clone_info.data_offset = datao; + clone_info.data_len = datal; + clone_info.file_offset = new_key.offset; + clone_info.extent_buf = buf; + clone_info.item_size = size; + ret = btrfs_punch_hole_range(inode, path, drop_start, + new_key.offset + datal - 1, &clone_info, + &trans); + if (ret) + goto out; + } else if (type == BTRFS_FILE_EXTENT_INLINE) { + /* + * Inline extents always have to start at file offset 0 + * and can never be bigger then the sector size. We can + * never clone only parts of an inline extent, since all + * reflink operations must start at a sector size aligned + * offset, and the length must be aligned too or end at + * the i_size (which implies the whole inlined data). + */ + ASSERT(key.offset == 0); + ASSERT(datal <= fs_info->sectorsize); + if (key.offset != 0 || datal > fs_info->sectorsize) + return -EUCLEAN; + + ret = clone_copy_inline_extent(inode, path, &new_key, + drop_start, datal, size, + comp, buf, &trans); + if (ret) + goto out; + } + + btrfs_release_path(path); + + last_dest_end = ALIGN(new_key.offset + datal, + fs_info->sectorsize); + ret = clone_finish_inode_update(trans, inode, last_dest_end, + destoff, olen, no_time_update); + if (ret) + goto out; + if (new_key.offset + datal >= destoff + len) + break; + + btrfs_release_path(path); + key.offset = next_key_min_offset; + + if (fatal_signal_pending(current)) { + ret = -EINTR; + goto out; + } + } + ret = 0; + + if (last_dest_end < destoff + len) { + /* + * We have an implicit hole that fully or partially overlaps our + * cloning range at its end. This means that we either have the + * NO_HOLES feature enabled or the implicit hole happened due to + * mixing buffered and direct IO writes against this file. + */ + btrfs_release_path(path); + path->leave_spinning = 0; + + ret = btrfs_punch_hole_range(inode, path, last_dest_end, + destoff + len - 1, NULL, &trans); + if (ret) + goto out; + + ret = clone_finish_inode_update(trans, inode, destoff + len, + destoff, olen, no_time_update); + } + +out: + btrfs_free_path(path); + kvfree(buf); + return ret; +} + +static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1, + struct inode *inode2, u64 loff2, u64 len) +{ + unlock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1); + unlock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1); +} + +static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1, + struct inode *inode2, u64 loff2, u64 len) +{ + if (inode1 < inode2) { + swap(inode1, inode2); + swap(loff1, loff2); + } else if (inode1 == inode2 && loff2 < loff1) { + swap(loff1, loff2); + } + lock_extent(&BTRFS_I(inode1)->io_tree, loff1, loff1 + len - 1); + lock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1); +} + +static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 len, + struct inode *dst, u64 dst_loff) +{ + const u64 bs = BTRFS_I(src)->root->fs_info->sb->s_blocksize; + int ret; + + /* + * Lock destination range to serialize with concurrent readpages() and + * source range to serialize with relocation. + */ + btrfs_double_extent_lock(src, loff, dst, dst_loff, len); + ret = btrfs_clone(src, dst, loff, len, ALIGN(len, bs), dst_loff, 1); + btrfs_double_extent_unlock(src, loff, dst, dst_loff, len); + + return ret; +} + +static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen, + struct inode *dst, u64 dst_loff) +{ + int ret; + u64 i, tail_len, chunk_count; + struct btrfs_root *root_dst = BTRFS_I(dst)->root; + + spin_lock(&root_dst->root_item_lock); + if (root_dst->send_in_progress) { + btrfs_warn_rl(root_dst->fs_info, +"cannot deduplicate to root %llu while send operations are using it (%d in progress)", + root_dst->root_key.objectid, + root_dst->send_in_progress); + spin_unlock(&root_dst->root_item_lock); + return -EAGAIN; + } + root_dst->dedupe_in_progress++; + spin_unlock(&root_dst->root_item_lock); + + tail_len = olen % BTRFS_MAX_DEDUPE_LEN; + chunk_count = div_u64(olen, BTRFS_MAX_DEDUPE_LEN); + + for (i = 0; i < chunk_count; i++) { + ret = btrfs_extent_same_range(src, loff, BTRFS_MAX_DEDUPE_LEN, + dst, dst_loff); + if (ret) + goto out; + + loff += BTRFS_MAX_DEDUPE_LEN; + dst_loff += BTRFS_MAX_DEDUPE_LEN; + } + + if (tail_len > 0) + ret = btrfs_extent_same_range(src, loff, tail_len, dst, dst_loff); +out: + spin_lock(&root_dst->root_item_lock); + root_dst->dedupe_in_progress--; + spin_unlock(&root_dst->root_item_lock); + + return ret; +} + +static noinline int btrfs_clone_files(struct file *file, struct file *file_src, + u64 off, u64 olen, u64 destoff) +{ + struct inode *inode = file_inode(file); + struct inode *src = file_inode(file_src); + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + int ret; + int wb_ret; + u64 len = olen; + u64 bs = fs_info->sb->s_blocksize; + + /* + * VFS's generic_remap_file_range_prep() protects us from cloning the + * eof block into the middle of a file, which would result in corruption + * if the file size is not blocksize aligned. So we don't need to check + * for that case here. + */ + if (off + len == src->i_size) + len = ALIGN(src->i_size, bs) - off; + + if (destoff > inode->i_size) { + const u64 wb_start = ALIGN_DOWN(inode->i_size, bs); + + ret = btrfs_cont_expand(inode, inode->i_size, destoff); + if (ret) + return ret; + /* + * We may have truncated the last block if the inode's size is + * not sector size aligned, so we need to wait for writeback to + * complete before proceeding further, otherwise we can race + * with cloning and attempt to increment a reference to an + * extent that no longer exists (writeback completed right after + * we found the previous extent covering eof and before we + * attempted to increment its reference count). + */ + ret = btrfs_wait_ordered_range(inode, wb_start, + destoff - wb_start); + if (ret) + return ret; + } + + /* + * Lock destination range to serialize with concurrent readpages() and + * source range to serialize with relocation. + */ + btrfs_double_extent_lock(src, off, inode, destoff, len); + ret = btrfs_clone(src, inode, off, olen, len, destoff, 0); + btrfs_double_extent_unlock(src, off, inode, destoff, len); + + /* + * We may have copied an inline extent into a page of the destination + * range, so wait for writeback to complete before truncating pages + * from the page cache. This is a rare case. + */ + wb_ret = btrfs_wait_ordered_range(inode, destoff, len); + ret = ret ? ret : wb_ret; + /* + * Truncate page cache pages so that future reads will see the cloned + * data immediately and not the previous data. + */ + truncate_inode_pages_range(&inode->i_data, + round_down(destoff, PAGE_SIZE), + round_up(destoff + len, PAGE_SIZE) - 1); + + return ret; +} + +static int btrfs_remap_file_range_prep(struct file *file_in, loff_t pos_in, + struct file *file_out, loff_t pos_out, + loff_t *len, unsigned int remap_flags) +{ + struct inode *inode_in = file_inode(file_in); + struct inode *inode_out = file_inode(file_out); + u64 bs = BTRFS_I(inode_out)->root->fs_info->sb->s_blocksize; + bool same_inode = inode_out == inode_in; + u64 wb_len; + int ret; + + if (!(remap_flags & REMAP_FILE_DEDUP)) { + struct btrfs_root *root_out = BTRFS_I(inode_out)->root; + + if (btrfs_root_readonly(root_out)) + return -EROFS; + + if (file_in->f_path.mnt != file_out->f_path.mnt || + inode_in->i_sb != inode_out->i_sb) + return -EXDEV; + } + + /* Don't make the dst file partly checksummed */ + if ((BTRFS_I(inode_in)->flags & BTRFS_INODE_NODATASUM) != + (BTRFS_I(inode_out)->flags & BTRFS_INODE_NODATASUM)) { + return -EINVAL; + } + + /* + * Now that the inodes are locked, we need to start writeback ourselves + * and can not rely on the writeback from the VFS's generic helper + * generic_remap_file_range_prep() because: + * + * 1) For compression we must call filemap_fdatawrite_range() range + * twice (btrfs_fdatawrite_range() does it for us), and the generic + * helper only calls it once; + * + * 2) filemap_fdatawrite_range(), called by the generic helper only + * waits for the writeback to complete, i.e. for IO to be done, and + * not for the ordered extents to complete. We need to wait for them + * to complete so that new file extent items are in the fs tree. + */ + if (*len == 0 && !(remap_flags & REMAP_FILE_DEDUP)) + wb_len = ALIGN(inode_in->i_size, bs) - ALIGN_DOWN(pos_in, bs); + else + wb_len = ALIGN(*len, bs); + + /* + * Since we don't lock ranges, wait for ongoing lockless dio writes (as + * any in progress could create its ordered extents after we wait for + * existing ordered extents below). + */ + inode_dio_wait(inode_in); + if (!same_inode) + inode_dio_wait(inode_out); + + /* + * Workaround to make sure NOCOW buffered write reach disk as NOCOW. + * + * Btrfs' back references do not have a block level granularity, they + * work at the whole extent level. + * NOCOW buffered write without data space reserved may not be able + * to fall back to CoW due to lack of data space, thus could cause + * data loss. + * + * Here we take a shortcut by flushing the whole inode, so that all + * nocow write should reach disk as nocow before we increase the + * reference of the extent. We could do better by only flushing NOCOW + * data, but that needs extra accounting. + * + * Also we don't need to check ASYNC_EXTENT, as async extent will be + * CoWed anyway, not affecting nocow part. + */ + ret = filemap_flush(inode_in->i_mapping); + if (ret < 0) + return ret; + + ret = btrfs_wait_ordered_range(inode_in, ALIGN_DOWN(pos_in, bs), + wb_len); + if (ret < 0) + return ret; + ret = btrfs_wait_ordered_range(inode_out, ALIGN_DOWN(pos_out, bs), + wb_len); + if (ret < 0) + return ret; + + return generic_remap_file_range_prep(file_in, pos_in, file_out, pos_out, + len, remap_flags); +} + +loff_t btrfs_remap_file_range(struct file *src_file, loff_t off, + struct file *dst_file, loff_t destoff, loff_t len, + unsigned int remap_flags) +{ + struct inode *src_inode = file_inode(src_file); + struct inode *dst_inode = file_inode(dst_file); + bool same_inode = dst_inode == src_inode; + int ret; + + if (remap_flags & ~(REMAP_FILE_DEDUP | REMAP_FILE_ADVISORY)) + return -EINVAL; + + if (same_inode) + inode_lock(src_inode); + else + lock_two_nondirectories(src_inode, dst_inode); + + ret = btrfs_remap_file_range_prep(src_file, off, dst_file, destoff, + &len, remap_flags); + if (ret < 0 || len == 0) + goto out_unlock; + + if (remap_flags & REMAP_FILE_DEDUP) + ret = btrfs_extent_same(src_inode, off, len, dst_inode, destoff); + else + ret = btrfs_clone_files(dst_file, src_file, off, len, destoff); + +out_unlock: + if (same_inode) + inode_unlock(src_inode); + else + unlock_two_nondirectories(src_inode, dst_inode); + + return ret < 0 ? ret : len; +} diff --git a/fs/btrfs/reflink.h b/fs/btrfs/reflink.h new file mode 100644 index 000000000000..ecb309b4dad0 --- /dev/null +++ b/fs/btrfs/reflink.h @@ -0,0 +1,12 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef BTRFS_REFLINK_H +#define BTRFS_REFLINK_H + +#include <linux/fs.h> + +loff_t btrfs_remap_file_range(struct file *file_in, loff_t pos_in, + struct file *file_out, loff_t pos_out, + loff_t len, unsigned int remap_flags); + +#endif /* BTRFS_REFLINK_H */ diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 995d4b8b1cfd..f65595602aa8 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -9,6 +9,7 @@ #include <linux/blkdev.h> #include <linux/rbtree.h> #include <linux/slab.h> +#include <linux/error-injection.h> #include "ctree.h" #include "disk-io.h" #include "transaction.h" @@ -22,6 +23,54 @@ #include "print-tree.h" #include "delalloc-space.h" #include "block-group.h" +#include "backref.h" + +/* + * Relocation overview + * + * [What does relocation do] + * + * The objective of relocation is to relocate all extents of the target block + * group to other block groups. + * This is utilized by resize (shrink only), profile converting, compacting + * space, or balance routine to spread chunks over devices. + * + * Before | After + * ------------------------------------------------------------------ + * BG A: 10 data extents | BG A: deleted + * BG B: 2 data extents | BG B: 10 data extents (2 old + 8 relocated) + * BG C: 1 extents | BG C: 3 data extents (1 old + 2 relocated) + * + * [How does relocation work] + * + * 1. Mark the target block group read-only + * New extents won't be allocated from the target block group. + * + * 2.1 Record each extent in the target block group + * To build a proper map of extents to be relocated. + * + * 2.2 Build data reloc tree and reloc trees + * Data reloc tree will contain an inode, recording all newly relocated + * data extents. + * There will be only one data reloc tree for one data block group. + * + * Reloc tree will be a special snapshot of its source tree, containing + * relocated tree blocks. + * Each tree referring to a tree block in target block group will get its + * reloc tree built. + * + * 2.3 Swap source tree with its corresponding reloc tree + * Each involved tree only refers to new extents after swap. + * + * 3. Cleanup reloc trees and data reloc tree. + * As old extents in the target block group are still referenced by reloc + * trees, we need to clean them up before really freeing the target block + * group. + * + * The main complexity is in steps 2.2 and 2.3. + * + * The entry point of relocation is relocate_block_group() function. + */ /* * backref_node, mapping_node and tree_block start with this @@ -256,6 +305,7 @@ static void free_backref_node(struct backref_cache *cache, { if (node) { cache->nr_nodes--; + btrfs_put_root(node->root); kfree(node); } } @@ -589,22 +639,7 @@ static struct btrfs_root *find_reloc_root(struct reloc_control *rc, root = (struct btrfs_root *)node->data; } spin_unlock(&rc->reloc_root_tree.lock); - return root; -} - -static int is_cowonly_root(u64 root_objectid) -{ - if (root_objectid == BTRFS_ROOT_TREE_OBJECTID || - root_objectid == BTRFS_EXTENT_TREE_OBJECTID || - root_objectid == BTRFS_CHUNK_TREE_OBJECTID || - root_objectid == BTRFS_DEV_TREE_OBJECTID || - root_objectid == BTRFS_TREE_LOG_OBJECTID || - root_objectid == BTRFS_CSUM_TREE_OBJECTID || - root_objectid == BTRFS_UUID_TREE_OBJECTID || - root_objectid == BTRFS_QUOTA_TREE_OBJECTID || - root_objectid == BTRFS_FREE_SPACE_TREE_OBJECTID) - return 1; - return 0; + return btrfs_grab_root(root); } static struct btrfs_root *read_fs_root(struct btrfs_fs_info *fs_info, @@ -614,10 +649,7 @@ static struct btrfs_root *read_fs_root(struct btrfs_fs_info *fs_info, key.objectid = root_objectid; key.type = BTRFS_ROOT_ITEM_KEY; - if (is_cowonly_root(root_objectid)) - key.offset = 0; - else - key.offset = (u64)-1; + key.offset = (u64)-1; return btrfs_get_fs_root(fs_info, &key, false); } @@ -711,8 +743,6 @@ struct backref_node *build_backref_tree(struct reloc_control *rc, err = -ENOMEM; goto out; } - path1->reada = READA_FORWARD; - path2->reada = READA_FORWARD; node = alloc_backref_node(cache); if (!node) { @@ -899,10 +929,12 @@ again: /* tree root */ ASSERT(btrfs_root_bytenr(&root->root_item) == cur->bytenr); - if (should_ignore_root(root)) + if (should_ignore_root(root)) { + btrfs_put_root(root); list_add(&cur->list, &useless); - else + } else { cur->root = root; + } break; } @@ -915,6 +947,7 @@ again: ret = btrfs_search_slot(NULL, root, node_key, path2, 0, 0); path2->lowest_level = 0; if (ret < 0) { + btrfs_put_root(root); err = ret; goto out; } @@ -930,6 +963,7 @@ again: root->root_key.objectid, node_key->objectid, node_key->type, node_key->offset); + btrfs_put_root(root); err = -ENOENT; goto out; } @@ -941,15 +975,18 @@ again: if (!path2->nodes[level]) { ASSERT(btrfs_root_bytenr(&root->root_item) == lower->bytenr); - if (should_ignore_root(root)) + if (should_ignore_root(root)) { + btrfs_put_root(root); list_add(&lower->list, &useless); - else + } else { lower->root = root; + } break; } edge = alloc_backref_edge(cache); if (!edge) { + btrfs_put_root(root); err = -ENOMEM; goto out; } @@ -959,6 +996,7 @@ again: if (!rb_node) { upper = alloc_backref_node(cache); if (!upper) { + btrfs_put_root(root); free_backref_edge(cache, edge); err = -ENOMEM; goto out; @@ -1006,8 +1044,10 @@ again: edge->node[LOWER] = lower; edge->node[UPPER] = upper; - if (rb_node) + if (rb_node) { + btrfs_put_root(root); break; + } lower = upper; upper = NULL; } @@ -1186,7 +1226,7 @@ out: free_backref_node(cache, lower); } - free_backref_node(cache, node); + remove_backref_node(cache, node); return ERR_PTR(err); } ASSERT(!node || !node->detached); @@ -1244,7 +1284,8 @@ static int clone_backref_node(struct btrfs_trans_handle *trans, new_node->level = node->level; new_node->lowest = node->lowest; new_node->checked = 1; - new_node->root = dest; + new_node->root = btrfs_grab_root(dest); + ASSERT(new_node->root); if (!node->lowest) { list_for_each_entry(edge, &node->lower, list[UPPER]) { @@ -1298,7 +1339,7 @@ static int __must_check __add_reloc_root(struct btrfs_root *root) if (!node) return -ENOMEM; - node->bytenr = root->node->start; + node->bytenr = root->commit_root->start; node->data = root; spin_lock(&rc->reloc_root_tree.lock); @@ -1325,14 +1366,16 @@ static void __del_reloc_root(struct btrfs_root *root) struct rb_node *rb_node; struct mapping_node *node = NULL; struct reloc_control *rc = fs_info->reloc_ctl; + bool put_ref = false; if (rc && root->node) { spin_lock(&rc->reloc_root_tree.lock); rb_node = tree_search(&rc->reloc_root_tree.rb_root, - root->node->start); + root->commit_root->start); if (rb_node) { node = rb_entry(rb_node, struct mapping_node, rb_node); rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root); + RB_CLEAR_NODE(&node->rb_node); } spin_unlock(&rc->reloc_root_tree.lock); if (!node) @@ -1340,9 +1383,22 @@ static void __del_reloc_root(struct btrfs_root *root) BUG_ON((struct btrfs_root *)node->data != root); } + /* + * We only put the reloc root here if it's on the list. There's a lot + * of places where the pattern is to splice the rc->reloc_roots, process + * the reloc roots, and then add the reloc root back onto + * rc->reloc_roots. If we call __del_reloc_root while it's off of the + * list we don't want the reference being dropped, because the guy + * messing with the list is in charge of the reference. + */ spin_lock(&fs_info->trans_lock); - list_del_init(&root->root_list); + if (!list_empty(&root->root_list)) { + put_ref = true; + list_del_init(&root->root_list); + } spin_unlock(&fs_info->trans_lock); + if (put_ref) + btrfs_put_root(root); kfree(node); } @@ -1350,7 +1406,7 @@ static void __del_reloc_root(struct btrfs_root *root) * helper to update the 'address of tree root -> reloc tree' * mapping */ -static int __update_reloc_root(struct btrfs_root *root, u64 new_bytenr) +static int __update_reloc_root(struct btrfs_root *root) { struct btrfs_fs_info *fs_info = root->fs_info; struct rb_node *rb_node; @@ -1359,7 +1415,7 @@ static int __update_reloc_root(struct btrfs_root *root, u64 new_bytenr) spin_lock(&rc->reloc_root_tree.lock); rb_node = tree_search(&rc->reloc_root_tree.rb_root, - root->node->start); + root->commit_root->start); if (rb_node) { node = rb_entry(rb_node, struct mapping_node, rb_node); rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root); @@ -1371,7 +1427,7 @@ static int __update_reloc_root(struct btrfs_root *root, u64 new_bytenr) BUG_ON((struct btrfs_root *)node->data != root); spin_lock(&rc->reloc_root_tree.lock); - node->bytenr = new_bytenr; + node->bytenr = root->node->start; rb_node = tree_insert(&rc->reloc_root_tree.rb_root, node->bytenr, &node->rb_node); spin_unlock(&rc->reloc_root_tree.lock); @@ -1447,8 +1503,9 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans, BUG_ON(ret); kfree(root_item); - reloc_root = btrfs_read_fs_root(fs_info->tree_root, &root_key); + reloc_root = btrfs_read_tree_root(fs_info->tree_root, &root_key); BUG_ON(IS_ERR(reloc_root)); + set_bit(BTRFS_ROOT_REF_COWS, &reloc_root->state); reloc_root->last_trans = trans->transid; return reloc_root; } @@ -1456,6 +1513,9 @@ static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans, /* * create reloc tree for a given fs tree. reloc tree is just a * snapshot of the fs tree with special root objectid. + * + * The reloc_root comes out of here with two references, one for + * root->reloc_root, and another for being on the rc->reloc_roots list. */ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans, struct btrfs_root *root) @@ -1467,6 +1527,10 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans, int clear_rsv = 0; int ret; + if (!rc || !rc->create_reloc_tree || + root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) + return 0; + /* * The subvolume has reloc tree but the swap is finished, no need to * create/update the dead reloc tree @@ -1480,10 +1544,6 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans, return 0; } - if (!rc || !rc->create_reloc_tree || - root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) - return 0; - if (!trans->reloc_reserved) { rsv = trans->block_rsv; trans->block_rsv = rc->block_rsv; @@ -1495,7 +1555,7 @@ int btrfs_init_reloc_root(struct btrfs_trans_handle *trans, ret = __add_reloc_root(reloc_root); BUG_ON(ret < 0); - root->reloc_root = reloc_root; + root->reloc_root = btrfs_grab_root(reloc_root); return 0; } @@ -1516,6 +1576,13 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans, reloc_root = root->reloc_root; root_item = &reloc_root->root_item; + /* + * We are probably ok here, but __del_reloc_root() will drop its ref of + * the root. We have the ref for root->reloc_root, but just in case + * hold it while we update the reloc root. + */ + btrfs_grab_root(reloc_root); + /* root->reloc_root will stay until current relocation finished */ if (fs_info->reloc_ctl->merge_reloc_tree && btrfs_root_refs(root_item) == 0) { @@ -1529,6 +1596,7 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans, } if (reloc_root->commit_root != reloc_root->node) { + __update_reloc_root(reloc_root); btrfs_set_root_node(root_item, reloc_root->node); free_extent_buffer(reloc_root->commit_root); reloc_root->commit_root = btrfs_root_node(reloc_root); @@ -1537,7 +1605,7 @@ int btrfs_update_reloc_root(struct btrfs_trans_handle *trans, ret = btrfs_update_root(trans, fs_info->tree_root, &reloc_root->root_key, root_item); BUG_ON(ret); - + btrfs_put_root(reloc_root); out: return 0; } @@ -2211,7 +2279,7 @@ static void insert_dirty_subvol(struct btrfs_trans_handle *trans, btrfs_update_reloc_root(trans, root); if (list_empty(&root->reloc_dirty_list)) { - btrfs_grab_fs_root(root); + btrfs_grab_root(root); list_add_tail(&root->reloc_dirty_list, &rc->dirty_subvol_roots); } } @@ -2231,24 +2299,34 @@ static int clean_dirty_subvols(struct reloc_control *rc) list_del_init(&root->reloc_dirty_list); root->reloc_root = NULL; - if (reloc_root) { - - ret2 = btrfs_drop_snapshot(reloc_root, NULL, 0, 1); - if (ret2 < 0 && !ret) - ret = ret2; - } /* * Need barrier to ensure clear_bit() only happens after * root->reloc_root = NULL. Pairs with have_reloc_root. */ smp_wmb(); clear_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state); - btrfs_put_fs_root(root); + if (reloc_root) { + /* + * btrfs_drop_snapshot drops our ref we hold for + * ->reloc_root. If it fails however we must + * drop the ref ourselves. + */ + ret2 = btrfs_drop_snapshot(reloc_root, 0, 1); + if (ret2 < 0) { + btrfs_put_root(reloc_root); + if (!ret) + ret = ret2; + } + } + btrfs_put_root(root); } else { /* Orphan reloc tree, just clean it up */ - ret2 = btrfs_drop_snapshot(root, NULL, 0, 1); - if (ret2 < 0 && !ret) - ret = ret2; + ret2 = btrfs_drop_snapshot(root, 0, 1); + if (ret2 < 0) { + btrfs_put_root(root); + if (!ret) + ret = ret2; + } } } return ret; @@ -2325,6 +2403,18 @@ static noinline_for_stack int merge_reloc_root(struct reloc_control *rc, trans = NULL; goto out; } + + /* + * At this point we no longer have a reloc_control, so we can't + * depend on btrfs_init_reloc_root to update our last_trans. + * + * But that's ok, we started the trans handle on our + * corresponding fs_root, which means it's been added to the + * dirty list. At commit time we'll still call + * btrfs_update_reloc_root() and update our root item + * appropriately. + */ + reloc_root->last_trans = trans->transid; trans->block_rsv = rc->block_rsv; replaced = 0; @@ -2435,7 +2525,7 @@ again: if (IS_ERR(trans)) { if (!err) btrfs_block_rsv_release(fs_info, rc->block_rsv, - num_bytes); + num_bytes, NULL); return PTR_ERR(trans); } @@ -2443,7 +2533,7 @@ again: if (num_bytes != rc->merging_rsv_size) { btrfs_end_transaction(trans); btrfs_block_rsv_release(fs_info, rc->block_rsv, - num_bytes); + num_bytes, NULL); goto again; } } @@ -2468,6 +2558,7 @@ again: btrfs_update_reloc_root(trans, root); list_add(&reloc_root->root_list, &reloc_roots); + btrfs_put_root(root); } list_splice(&reloc_roots, &rc->reloc_roots); @@ -2488,10 +2579,6 @@ void free_reloc_roots(struct list_head *list) reloc_root = list_entry(list->next, struct btrfs_root, root_list); __del_reloc_root(reloc_root); - free_extent_buffer(reloc_root->node); - free_extent_buffer(reloc_root->commit_root); - reloc_root->node = NULL; - reloc_root->commit_root = NULL; } } @@ -2529,6 +2616,7 @@ again: BUG_ON(root->reloc_root != reloc_root); ret = merge_reloc_root(rc, root); + btrfs_put_root(root); if (ret) { if (list_empty(&reloc_root->root_list)) list_add_tail(&reloc_root->root_list, @@ -2561,7 +2649,21 @@ out: free_reloc_roots(&reloc_roots); } - BUG_ON(!RB_EMPTY_ROOT(&rc->reloc_root_tree.rb_root)); + /* + * We used to have + * + * BUG_ON(!RB_EMPTY_ROOT(&rc->reloc_root_tree.rb_root)); + * + * here, but it's wrong. If we fail to start the transaction in + * prepare_to_merge() we will have only 0 ref reloc roots, none of which + * have actually been removed from the reloc_root_tree rb tree. This is + * fine because we're bailing here, and we hold a reference on the root + * for the list that holds it, so these roots will be cleaned up when we + * do the reloc_dirty_list afterwards. Meanwhile the root->reloc_root + * will be cleaned up on unmount. + * + * The remaining nodes will be cleaned up by free_reloc_control. + */ } static void free_block_list(struct rb_root *blocks) @@ -2580,6 +2682,7 @@ static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *fs_info = reloc_root->fs_info; struct btrfs_root *root; + int ret; if (reloc_root->last_trans == trans->transid) return 0; @@ -2587,8 +2690,10 @@ static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans, root = read_fs_root(fs_info, reloc_root->root_key.offset); BUG_ON(IS_ERR(root)); BUG_ON(root->reloc_root != reloc_root); + ret = btrfs_record_root_in_trans(trans, root); + btrfs_put_root(root); - return btrfs_record_root_in_trans(trans, root); + return ret; } static noinline_for_stack @@ -2621,7 +2726,9 @@ struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans, BUG_ON(next->new_bytenr); BUG_ON(!list_empty(&next->list)); next->new_bytenr = root->node->start; - next->root = root; + btrfs_put_root(next->root); + next->root = btrfs_grab_root(root); + ASSERT(next->root); list_add_tail(&next->list, &rc->backref_cache.changed); __mark_block_processed(rc, next); @@ -3040,7 +3147,6 @@ static int get_tree_block_key(struct btrfs_fs_info *fs_info, { struct extent_buffer *eb; - BUG_ON(block->key_ready); eb = read_tree_block(fs_info, block->bytenr, block->key.offset, block->level, NULL); if (IS_ERR(eb)) { @@ -3073,6 +3179,14 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, if (!node) return 0; + /* + * If we fail here we want to drop our backref_node because we are going + * to start over and regenerate the tree for it. + */ + ret = reserve_metadata_space(trans, rc, node); + if (ret) + goto out; + BUG_ON(node->processed); root = select_one_root(node); if (root == ERR_PTR(-ENOENT)) { @@ -3080,12 +3194,6 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, goto out; } - if (!root || test_bit(BTRFS_ROOT_REF_COWS, &root->state)) { - ret = reserve_metadata_space(trans, rc, node); - if (ret) - goto out; - } - if (root) { if (test_bit(BTRFS_ROOT_REF_COWS, &root->state)) { BUG_ON(node->new_bytenr); @@ -3093,7 +3201,9 @@ static int relocate_tree_block(struct btrfs_trans_handle *trans, btrfs_record_root_in_trans(trans, root); root = root->reloc_root; node->new_bytenr = root->node->start; - node->root = root; + btrfs_put_root(node->root); + node->root = btrfs_grab_root(root); + ASSERT(node->root); list_add_tail(&node->list, &rc->backref_cache.changed); } else { path->lowest_level = node->level; @@ -3161,9 +3271,8 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans, ret = relocate_tree_block(trans, rc, node, &block->key, path); if (ret < 0) { - if (ret != -EAGAIN || &block->rb_node == rb_first(blocks)) - err = ret; - goto out; + err = ret; + break; } } out: @@ -3264,6 +3373,15 @@ int setup_extent_mapping(struct inode *inode, u64 start, u64 end, return ret; } +/* + * Allow error injection to test balance cancellation + */ +int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info) +{ + return atomic_read(&fs_info->balance_cancel_req); +} +ALLOW_ERROR_INJECTION(btrfs_should_cancel_balance, TRUE); + static int relocate_file_extent_cluster(struct inode *inode, struct file_extent_cluster *cluster) { @@ -3385,6 +3503,10 @@ static int relocate_file_extent_cluster(struct inode *inode, btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); balance_dirty_pages_ratelimited(inode->i_mapping); btrfs_throttle(fs_info); + if (btrfs_should_cancel_balance(fs_info)) { + ret = -ECANCELED; + goto out; + } } WARN_ON(nr != cluster->nr); out: @@ -3556,31 +3678,6 @@ out: return ret; } -/* - * helper to check if the block use full backrefs for pointers in it - */ -static int block_use_full_backref(struct reloc_control *rc, - struct extent_buffer *eb) -{ - u64 flags; - int ret; - - if (btrfs_header_flag(eb, BTRFS_HEADER_FLAG_RELOC) || - btrfs_header_backref_rev(eb) < BTRFS_MIXED_BACKREF_REV) - return 1; - - ret = btrfs_lookup_extent_info(NULL, rc->extent_root->fs_info, - eb->start, btrfs_header_level(eb), 1, - NULL, &flags); - BUG_ON(ret); - - if (flags & BTRFS_BLOCK_FLAG_FULL_BACKREF) - ret = 1; - else - ret = 0; - return ret; -} - static int delete_block_group_cache(struct btrfs_fs_info *fs_info, struct btrfs_block_group *block_group, struct inode *inode, @@ -3624,172 +3721,40 @@ out: } /* - * helper to add tree blocks for backref of type BTRFS_EXTENT_DATA_REF_KEY - * this function scans fs tree to find blocks reference the data extent + * Locate the free space cache EXTENT_DATA in root tree leaf and delete the + * cache inode, to avoid free space cache data extent blocking data relocation. */ -static int find_data_references(struct reloc_control *rc, - struct btrfs_key *extent_key, - struct extent_buffer *leaf, - struct btrfs_extent_data_ref *ref, - struct rb_root *blocks) +static int delete_v1_space_cache(struct extent_buffer *leaf, + struct btrfs_block_group *block_group, + u64 data_bytenr) { - struct btrfs_fs_info *fs_info = rc->extent_root->fs_info; - struct btrfs_path *path; - struct tree_block *block; - struct btrfs_root *root; - struct btrfs_file_extent_item *fi; - struct rb_node *rb_node; + u64 space_cache_ino; + struct btrfs_file_extent_item *ei; struct btrfs_key key; - u64 ref_root; - u64 ref_objectid; - u64 ref_offset; - u32 ref_count; - u32 nritems; - int err = 0; - int added = 0; - int counted; + bool found = false; + int i; int ret; - ref_root = btrfs_extent_data_ref_root(leaf, ref); - ref_objectid = btrfs_extent_data_ref_objectid(leaf, ref); - ref_offset = btrfs_extent_data_ref_offset(leaf, ref); - ref_count = btrfs_extent_data_ref_count(leaf, ref); - - /* - * This is an extent belonging to the free space cache, lets just delete - * it and redo the search. - */ - if (ref_root == BTRFS_ROOT_TREE_OBJECTID) { - ret = delete_block_group_cache(fs_info, rc->block_group, - NULL, ref_objectid); - if (ret != -ENOENT) - return ret; - ret = 0; - } - - path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; - path->reada = READA_FORWARD; - - root = read_fs_root(fs_info, ref_root); - if (IS_ERR(root)) { - err = PTR_ERR(root); - goto out; - } - - key.objectid = ref_objectid; - key.type = BTRFS_EXTENT_DATA_KEY; - if (ref_offset > ((u64)-1 << 32)) - key.offset = 0; - else - key.offset = ref_offset; - - path->search_commit_root = 1; - path->skip_locking = 1; - ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); - if (ret < 0) { - err = ret; - goto out; - } - - leaf = path->nodes[0]; - nritems = btrfs_header_nritems(leaf); - /* - * the references in tree blocks that use full backrefs - * are not counted in - */ - if (block_use_full_backref(rc, leaf)) - counted = 0; - else - counted = 1; - rb_node = tree_search(blocks, leaf->start); - if (rb_node) { - if (counted) - added = 1; - else - path->slots[0] = nritems; - } - - while (ref_count > 0) { - while (path->slots[0] >= nritems) { - ret = btrfs_next_leaf(root, path); - if (ret < 0) { - err = ret; - goto out; - } - if (WARN_ON(ret > 0)) - goto out; - - leaf = path->nodes[0]; - nritems = btrfs_header_nritems(leaf); - added = 0; - - if (block_use_full_backref(rc, leaf)) - counted = 0; - else - counted = 1; - rb_node = tree_search(blocks, leaf->start); - if (rb_node) { - if (counted) - added = 1; - else - path->slots[0] = nritems; - } - } + if (btrfs_header_owner(leaf) != BTRFS_ROOT_TREE_OBJECTID) + return 0; - btrfs_item_key_to_cpu(leaf, &key, path->slots[0]); - if (WARN_ON(key.objectid != ref_objectid || - key.type != BTRFS_EXTENT_DATA_KEY)) + for (i = 0; i < btrfs_header_nritems(leaf); i++) { + btrfs_item_key_to_cpu(leaf, &key, i); + if (key.type != BTRFS_EXTENT_DATA_KEY) + continue; + ei = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item); + if (btrfs_file_extent_type(leaf, ei) == BTRFS_FILE_EXTENT_REG && + btrfs_file_extent_disk_bytenr(leaf, ei) == data_bytenr) { + found = true; + space_cache_ino = key.objectid; break; - - fi = btrfs_item_ptr(leaf, path->slots[0], - struct btrfs_file_extent_item); - - if (btrfs_file_extent_type(leaf, fi) == - BTRFS_FILE_EXTENT_INLINE) - goto next; - - if (btrfs_file_extent_disk_bytenr(leaf, fi) != - extent_key->objectid) - goto next; - - key.offset -= btrfs_file_extent_offset(leaf, fi); - if (key.offset != ref_offset) - goto next; - - if (counted) - ref_count--; - if (added) - goto next; - - if (!tree_block_processed(leaf->start, rc)) { - block = kmalloc(sizeof(*block), GFP_NOFS); - if (!block) { - err = -ENOMEM; - break; - } - block->bytenr = leaf->start; - btrfs_item_key_to_cpu(leaf, &block->key, 0); - block->level = 0; - block->key_ready = 1; - rb_node = tree_insert(blocks, block->bytenr, - &block->rb_node); - if (rb_node) - backref_tree_panic(rb_node, -EEXIST, - block->bytenr); } - if (counted) - added = 1; - else - path->slots[0] = nritems; -next: - path->slots[0]++; - } -out: - btrfs_free_path(path); - return err; + if (!found) + return -ENOENT; + ret = delete_block_group_cache(leaf->fs_info, block_group, NULL, + space_cache_ino); + return ret; } /* @@ -3801,91 +3766,41 @@ int add_data_references(struct reloc_control *rc, struct btrfs_path *path, struct rb_root *blocks) { - struct btrfs_key key; - struct extent_buffer *eb; - struct btrfs_extent_data_ref *dref; - struct btrfs_extent_inline_ref *iref; - unsigned long ptr; - unsigned long end; - u32 blocksize = rc->extent_root->fs_info->nodesize; + struct btrfs_fs_info *fs_info = rc->extent_root->fs_info; + struct ulist *leaves = NULL; + struct ulist_iterator leaf_uiter; + struct ulist_node *ref_node = NULL; + const u32 blocksize = fs_info->nodesize; int ret = 0; - int err = 0; - - eb = path->nodes[0]; - ptr = btrfs_item_ptr_offset(eb, path->slots[0]); - end = ptr + btrfs_item_size_nr(eb, path->slots[0]); - ptr += sizeof(struct btrfs_extent_item); - while (ptr < end) { - iref = (struct btrfs_extent_inline_ref *)ptr; - key.type = btrfs_get_extent_inline_ref_type(eb, iref, - BTRFS_REF_TYPE_DATA); - if (key.type == BTRFS_SHARED_DATA_REF_KEY) { - key.offset = btrfs_extent_inline_ref_offset(eb, iref); - ret = __add_tree_block(rc, key.offset, blocksize, - blocks); - } else if (key.type == BTRFS_EXTENT_DATA_REF_KEY) { - dref = (struct btrfs_extent_data_ref *)(&iref->offset); - ret = find_data_references(rc, extent_key, - eb, dref, blocks); - } else { - ret = -EUCLEAN; - btrfs_err(rc->extent_root->fs_info, - "extent %llu slot %d has an invalid inline ref type", - eb->start, path->slots[0]); - } - if (ret) { - err = ret; - goto out; - } - ptr += btrfs_extent_inline_ref_size(key.type); - } - WARN_ON(ptr > end); + btrfs_release_path(path); + ret = btrfs_find_all_leafs(NULL, fs_info, extent_key->objectid, + 0, &leaves, NULL, true); + if (ret < 0) + return ret; - while (1) { - cond_resched(); - eb = path->nodes[0]; - if (path->slots[0] >= btrfs_header_nritems(eb)) { - ret = btrfs_next_leaf(rc->extent_root, path); - if (ret < 0) { - err = ret; - break; - } - if (ret > 0) - break; - eb = path->nodes[0]; - } + ULIST_ITER_INIT(&leaf_uiter); + while ((ref_node = ulist_next(leaves, &leaf_uiter))) { + struct extent_buffer *eb; - btrfs_item_key_to_cpu(eb, &key, path->slots[0]); - if (key.objectid != extent_key->objectid) + eb = read_tree_block(fs_info, ref_node->val, 0, 0, NULL); + if (IS_ERR(eb)) { + ret = PTR_ERR(eb); break; - - if (key.type == BTRFS_SHARED_DATA_REF_KEY) { - ret = __add_tree_block(rc, key.offset, blocksize, - blocks); - } else if (key.type == BTRFS_EXTENT_DATA_REF_KEY) { - dref = btrfs_item_ptr(eb, path->slots[0], - struct btrfs_extent_data_ref); - ret = find_data_references(rc, extent_key, - eb, dref, blocks); - } else if (unlikely(key.type == BTRFS_EXTENT_REF_V0_KEY)) { - btrfs_print_v0_err(eb->fs_info); - btrfs_handle_fs_error(eb->fs_info, -EINVAL, NULL); - ret = -EINVAL; - } else { - ret = 0; } - if (ret) { - err = ret; + ret = delete_v1_space_cache(eb, rc->block_group, + extent_key->objectid); + free_extent_buffer(eb); + if (ret < 0) + break; + ret = __add_tree_block(rc, ref_node->val, blocksize, blocks); + if (ret < 0) break; - } - path->slots[0]++; } -out: - btrfs_release_path(path); - if (err) + if (ret < 0) free_block_list(blocks); - return err; + ulist_free(leaves); + return ret; } /* @@ -4137,12 +4052,6 @@ restart: if (!RB_EMPTY_ROOT(&blocks)) { ret = relocate_tree_blocks(trans, rc, &blocks); if (ret < 0) { - /* - * if we fail to relocate tree blocks, force to update - * backref cache when committing transaction. - */ - rc->backref_cache.last_trans = trans->transid - 1; - if (ret != -EAGAIN) { err = ret; break; @@ -4166,6 +4075,10 @@ restart: break; } } + if (btrfs_should_cancel_balance(fs_info)) { + err = -ECANCELED; + break; + } } if (trans && progress && err == -ENOSPC) { ret = btrfs_force_chunk_alloc(trans, rc->block_group->flags); @@ -4195,15 +4108,23 @@ restart: set_reloc_control(rc); backref_cache_cleanup(&rc->backref_cache); - btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1); + btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1, NULL); + /* + * Even in the case when the relocation is cancelled, we should all go + * through prepare_to_merge() and merge_reloc_roots(). + * + * For error (including cancelled balance), prepare_to_merge() will + * mark all reloc trees orphan, then queue them for cleanup in + * merge_reloc_roots() + */ err = prepare_to_merge(rc, err); merge_reloc_roots(rc); rc->merge_reloc_tree = 0; unset_reloc_control(rc); - btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1); + btrfs_block_rsv_release(fs_info, rc->block_rsv, (u64)-1, NULL); /* get rid of pinned extents */ trans = btrfs_join_transaction(rc->extent_root); @@ -4212,10 +4133,10 @@ restart: goto out_free; } btrfs_commit_transaction(trans); +out_free: ret = clean_dirty_subvols(rc); if (ret < 0 && !err) err = ret; -out_free: btrfs_free_block_rsv(fs_info, rc->block_rsv); btrfs_free_path(path); return err; @@ -4271,8 +4192,10 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info, return ERR_CAST(root); trans = btrfs_start_transaction(root, 6); - if (IS_ERR(trans)) + if (IS_ERR(trans)) { + btrfs_put_root(root); return ERR_CAST(trans); + } err = btrfs_find_free_objectid(root, &objectid); if (err) @@ -4290,6 +4213,7 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info, err = btrfs_orphan_add(trans, BTRFS_I(inode)); out: + btrfs_put_root(root); btrfs_end_transaction(trans); btrfs_btree_balance_dirty(fs_info); if (err) { @@ -4317,6 +4241,18 @@ static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info) return rc; } +static void free_reloc_control(struct reloc_control *rc) +{ + struct mapping_node *node, *tmp; + + free_reloc_roots(&rc->reloc_roots); + rbtree_postorder_for_each_entry_safe(node, tmp, + &rc->reloc_root_tree.rb_root, rb_node) + kfree(node); + + kfree(rc); +} + /* * Print the block group being relocated */ @@ -4461,7 +4397,7 @@ out: btrfs_dec_block_group_ro(rc->block_group); iput(rc->data_inode); btrfs_put_block_group(rc->block_group); - kfree(rc); + free_reloc_control(rc); return err; } @@ -4537,12 +4473,13 @@ int btrfs_recover_relocation(struct btrfs_root *root) key.type != BTRFS_ROOT_ITEM_KEY) break; - reloc_root = btrfs_read_fs_root(root, &key); + reloc_root = btrfs_read_tree_root(root, &key); if (IS_ERR(reloc_root)) { err = PTR_ERR(reloc_root); goto out; } + set_bit(BTRFS_ROOT_REF_COWS, &reloc_root->state); list_add(&reloc_root->root_list, &reloc_roots); if (btrfs_root_refs(&reloc_root->root_item) > 0) { @@ -4559,6 +4496,8 @@ int btrfs_recover_relocation(struct btrfs_root *root) err = ret; goto out; } + } else { + btrfs_put_root(fs_root); } } @@ -4584,9 +4523,8 @@ int btrfs_recover_relocation(struct btrfs_root *root) trans = btrfs_join_transaction(rc->extent_root); if (IS_ERR(trans)) { - unset_reloc_control(rc); err = PTR_ERR(trans); - goto out_free; + goto out_unset; } rc->merge_reloc_tree = 1; @@ -4606,17 +4544,18 @@ int btrfs_recover_relocation(struct btrfs_root *root) if (IS_ERR(fs_root)) { err = PTR_ERR(fs_root); list_add_tail(&reloc_root->root_list, &reloc_roots); - goto out_free; + goto out_unset; } err = __add_reloc_root(reloc_root); BUG_ON(err < 0); /* -ENOMEM or logic error */ - fs_root->reloc_root = reloc_root; + fs_root->reloc_root = btrfs_grab_root(reloc_root); + btrfs_put_root(fs_root); } err = btrfs_commit_transaction(trans); if (err) - goto out_free; + goto out_unset; merge_reloc_roots(rc); @@ -4625,15 +4564,16 @@ int btrfs_recover_relocation(struct btrfs_root *root) trans = btrfs_join_transaction(rc->extent_root); if (IS_ERR(trans)) { err = PTR_ERR(trans); - goto out_free; + goto out_clean; } err = btrfs_commit_transaction(trans); - +out_clean: ret = clean_dirty_subvols(rc); if (ret < 0 && !err) err = ret; -out_free: - kfree(rc); +out_unset: + unset_reloc_control(rc); + free_reloc_control(rc); out: if (!list_empty(&reloc_roots)) free_reloc_roots(&reloc_roots); @@ -4643,10 +4583,12 @@ out: if (err == 0) { /* cleanup orphan inode in data relocation tree */ fs_root = read_fs_root(fs_info, BTRFS_DATA_RELOC_TREE_OBJECTID); - if (IS_ERR(fs_root)) + if (IS_ERR(fs_root)) { err = PTR_ERR(fs_root); - else + } else { err = btrfs_orphan_cleanup(fs_root); + btrfs_put_root(fs_root); + } } return err; } @@ -4720,11 +4662,6 @@ int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans, BUG_ON(rc->stage == UPDATE_DATA_PTRS && root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID); - if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) { - if (buf == root->node) - __update_reloc_root(root, cow->start); - } - level = btrfs_header_level(buf); if (btrfs_header_generation(buf) <= btrfs_root_last_snapshot(&root->root_item)) @@ -4795,6 +4732,10 @@ void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending, /* * called after snapshot is created. migrate block reservation * and create reloc root for the newly created snapshot + * + * This is similar to btrfs_init_reloc_root(), we come out of here with two + * references held on the reloc_root, one for root->reloc_root and one for + * rc->reloc_roots. */ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans, struct btrfs_pending_snapshot *pending) @@ -4827,7 +4768,7 @@ int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans, ret = __add_reloc_root(reloc_root); BUG_ON(ret < 0); - new_root->reloc_root = reloc_root; + new_root->reloc_root = btrfs_grab_root(reloc_root); if (rc->create_reloc_tree) ret = clone_backref_node(trans, rc, root, reloc_root); diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c index 612411c74550..668f22844017 100644 --- a/fs/btrfs/root-tree.c +++ b/fs/btrfs/root-tree.c @@ -22,7 +22,6 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot, struct btrfs_root_item *item) { - uuid_le uuid; u32 len; int need_reset = 0; @@ -44,8 +43,7 @@ static void btrfs_read_root_item(struct extent_buffer *eb, int slot, sizeof(*item) - offsetof(struct btrfs_root_item, generation_v2)); - uuid_le_gen(&uuid); - memcpy(item->uuid, uuid.b, BTRFS_UUID_SIZE); + generate_random_guid(item->uuid); } } @@ -255,25 +253,7 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info) root_key.objectid = key.offset; key.offset++; - /* - * The root might have been inserted already, as before we look - * for orphan roots, log replay might have happened, which - * triggers a transaction commit and qgroup accounting, which - * in turn reads and inserts fs roots while doing backref - * walking. - */ - root = btrfs_lookup_fs_root(fs_info, root_key.objectid); - if (root) { - WARN_ON(!test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, - &root->state)); - if (btrfs_root_refs(&root->root_item) == 0) { - set_bit(BTRFS_ROOT_DEAD_TREE, &root->state); - btrfs_add_dead_root(root); - } - continue; - } - - root = btrfs_read_fs_root(tree_root, &root_key); + root = btrfs_get_fs_root(fs_info, &root_key, false); err = PTR_ERR_OR_ZERO(root); if (err && err != -ENOENT) { break; @@ -300,25 +280,12 @@ int btrfs_find_orphan_roots(struct btrfs_fs_info *fs_info) continue; } - err = btrfs_init_fs_root(root); - if (err) { - btrfs_free_fs_root(root); - break; - } - - set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state); - - err = btrfs_insert_fs_root(fs_info, root); - if (err) { - BUG_ON(err == -EEXIST); - btrfs_free_fs_root(root); - break; - } - + WARN_ON(!test_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state)); if (btrfs_root_refs(&root->root_item) == 0) { set_bit(BTRFS_ROOT_DEAD_TREE, &root->state); btrfs_add_dead_root(root); } + btrfs_put_root(root); } btrfs_free_path(path); @@ -553,5 +520,5 @@ int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, void btrfs_subvolume_release_metadata(struct btrfs_fs_info *fs_info, struct btrfs_block_rsv *rsv) { - btrfs_block_rsv_release(fs_info, rsv, (u64)-1); + btrfs_block_rsv_release(fs_info, rsv, (u64)-1, NULL); } diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 61b37c56a7fb..adaf8ab694d5 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -149,7 +149,7 @@ struct scrub_parity { */ unsigned long *ebitmap; - unsigned long bitmap[0]; + unsigned long bitmap[]; }; struct scrub_ctx { @@ -653,7 +653,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, root_key.objectid = root; root_key.type = BTRFS_ROOT_ITEM_KEY; root_key.offset = (u64)-1; - local_root = btrfs_read_fs_root_no_name(fs_info, &root_key); + local_root = btrfs_get_fs_root(fs_info, &root_key, true); if (IS_ERR(local_root)) { ret = PTR_ERR(local_root); goto err; @@ -668,6 +668,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, ret = btrfs_search_slot(NULL, local_root, &key, swarn->path, 0, 0); if (ret) { + btrfs_put_root(local_root); btrfs_release_path(swarn->path); goto err; } @@ -688,6 +689,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, ipath = init_ipath(4096, local_root, swarn->path); memalloc_nofs_restore(nofs_flag); if (IS_ERR(ipath)) { + btrfs_put_root(local_root); ret = PTR_ERR(ipath); ipath = NULL; goto err; @@ -711,6 +713,7 @@ static int scrub_print_warning_inode(u64 inum, u64 offset, u64 root, min(isize - offset, (u64)PAGE_SIZE), nlink, (char *)(unsigned long)ipath->fspath->val[i]); + btrfs_put_root(local_root); free_ipath(ipath); return 0; diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index a055b657cb85..c5f41bd86765 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -5586,10 +5586,7 @@ static int get_last_extent(struct send_ctx *sctx, u64 offset) { struct btrfs_path *path; struct btrfs_root *root = sctx->send_root; - struct btrfs_file_extent_item *fi; struct btrfs_key key; - u64 extent_end; - u8 type; int ret; path = alloc_path_for_send(); @@ -5609,18 +5606,7 @@ static int get_last_extent(struct send_ctx *sctx, u64 offset) if (key.objectid != sctx->cur_ino || key.type != BTRFS_EXTENT_DATA_KEY) goto out; - fi = btrfs_item_ptr(path->nodes[0], path->slots[0], - struct btrfs_file_extent_item); - type = btrfs_file_extent_type(path->nodes[0], fi); - if (type == BTRFS_FILE_EXTENT_INLINE) { - u64 size = btrfs_file_extent_ram_bytes(path->nodes[0], fi); - extent_end = ALIGN(key.offset + size, - sctx->send_root->fs_info->sectorsize); - } else { - extent_end = key.offset + - btrfs_file_extent_num_bytes(path->nodes[0], fi); - } - sctx->cur_inode_last_extent = extent_end; + sctx->cur_inode_last_extent = btrfs_file_extent_end(path); out: btrfs_free_path(path); return ret; @@ -5674,16 +5660,7 @@ static int range_is_hole_in_parent(struct send_ctx *sctx, break; fi = btrfs_item_ptr(leaf, slot, struct btrfs_file_extent_item); - if (btrfs_file_extent_type(leaf, fi) == - BTRFS_FILE_EXTENT_INLINE) { - u64 size = btrfs_file_extent_ram_bytes(leaf, fi); - - extent_end = ALIGN(key.offset + size, - root->fs_info->sectorsize); - } else { - extent_end = key.offset + - btrfs_file_extent_num_bytes(leaf, fi); - } + extent_end = btrfs_file_extent_end(path); if (extent_end <= start) goto next; if (btrfs_file_extent_disk_bytenr(leaf, fi) == 0) { @@ -5704,9 +5681,6 @@ out: static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path, struct btrfs_key *key) { - struct btrfs_file_extent_item *fi; - u64 extent_end; - u8 type; int ret = 0; if (sctx->cur_ino != key->objectid || !need_send_hole(sctx)) @@ -5718,18 +5692,6 @@ static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path, return ret; } - fi = btrfs_item_ptr(path->nodes[0], path->slots[0], - struct btrfs_file_extent_item); - type = btrfs_file_extent_type(path->nodes[0], fi); - if (type == BTRFS_FILE_EXTENT_INLINE) { - u64 size = btrfs_file_extent_ram_bytes(path->nodes[0], fi); - extent_end = ALIGN(key->offset + size, - sctx->send_root->fs_info->sectorsize); - } else { - extent_end = key->offset + - btrfs_file_extent_num_bytes(path->nodes[0], fi); - } - if (path->slots[0] == 0 && sctx->cur_inode_last_extent < key->offset) { /* @@ -5755,7 +5717,7 @@ static int maybe_send_hole(struct send_ctx *sctx, struct btrfs_path *path, else ret = 0; } - sctx->cur_inode_last_extent = extent_end; + sctx->cur_inode_last_extent = btrfs_file_extent_end(path); return ret; } @@ -7066,7 +7028,6 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg) int clone_sources_to_rollback = 0; unsigned alloc_size; int sort_clone_roots = 0; - int index; if (!capable(CAP_SYS_ADMIN)) return -EPERM; @@ -7193,11 +7154,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg) key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - index = srcu_read_lock(&fs_info->subvol_srcu); - - clone_root = btrfs_read_fs_root_no_name(fs_info, &key); + clone_root = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(clone_root)) { - srcu_read_unlock(&fs_info->subvol_srcu, index); ret = PTR_ERR(clone_root); goto out; } @@ -7205,20 +7163,19 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg) if (!btrfs_root_readonly(clone_root) || btrfs_root_dead(clone_root)) { spin_unlock(&clone_root->root_item_lock); - srcu_read_unlock(&fs_info->subvol_srcu, index); + btrfs_put_root(clone_root); ret = -EPERM; goto out; } if (clone_root->dedupe_in_progress) { dedupe_in_progress_warn(clone_root); spin_unlock(&clone_root->root_item_lock); - srcu_read_unlock(&fs_info->subvol_srcu, index); + btrfs_put_root(clone_root); ret = -EAGAIN; goto out; } clone_root->send_in_progress++; spin_unlock(&clone_root->root_item_lock); - srcu_read_unlock(&fs_info->subvol_srcu, index); sctx->clone_roots[i].root = clone_root; clone_sources_to_rollback = i + 1; @@ -7232,11 +7189,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg) key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - index = srcu_read_lock(&fs_info->subvol_srcu); - - sctx->parent_root = btrfs_read_fs_root_no_name(fs_info, &key); + sctx->parent_root = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(sctx->parent_root)) { - srcu_read_unlock(&fs_info->subvol_srcu, index); ret = PTR_ERR(sctx->parent_root); goto out; } @@ -7246,20 +7200,16 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg) if (!btrfs_root_readonly(sctx->parent_root) || btrfs_root_dead(sctx->parent_root)) { spin_unlock(&sctx->parent_root->root_item_lock); - srcu_read_unlock(&fs_info->subvol_srcu, index); ret = -EPERM; goto out; } if (sctx->parent_root->dedupe_in_progress) { dedupe_in_progress_warn(sctx->parent_root); spin_unlock(&sctx->parent_root->root_item_lock); - srcu_read_unlock(&fs_info->subvol_srcu, index); ret = -EAGAIN; goto out; } spin_unlock(&sctx->parent_root->root_item_lock); - - srcu_read_unlock(&fs_info->subvol_srcu, index); } /* @@ -7267,7 +7217,8 @@ long btrfs_ioctl_send(struct file *mnt_file, struct btrfs_ioctl_send_args *arg) * is behind the current send position. This is checked while searching * for possible clone sources. */ - sctx->clone_roots[sctx->clone_roots_cnt++].root = sctx->send_root; + sctx->clone_roots[sctx->clone_roots_cnt++].root = + btrfs_grab_root(sctx->send_root); /* We do a bsearch later */ sort(sctx->clone_roots, sctx->clone_roots_cnt, @@ -7352,18 +7303,24 @@ out: } if (sort_clone_roots) { - for (i = 0; i < sctx->clone_roots_cnt; i++) + for (i = 0; i < sctx->clone_roots_cnt; i++) { btrfs_root_dec_send_in_progress( sctx->clone_roots[i].root); + btrfs_put_root(sctx->clone_roots[i].root); + } } else { - for (i = 0; sctx && i < clone_sources_to_rollback; i++) + for (i = 0; sctx && i < clone_sources_to_rollback; i++) { btrfs_root_dec_send_in_progress( sctx->clone_roots[i].root); + btrfs_put_root(sctx->clone_roots[i].root); + } btrfs_root_dec_send_in_progress(send_root); } - if (sctx && !IS_ERR_OR_NULL(sctx->parent_root)) + if (sctx && !IS_ERR_OR_NULL(sctx->parent_root)) { btrfs_root_dec_send_in_progress(sctx->parent_root); + btrfs_put_root(sctx->parent_root); + } kvfree(clone_sources_tmp); diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c index 01297c5b2666..8b0fe053a25d 100644 --- a/fs/btrfs/space-info.c +++ b/fs/btrfs/space-info.c @@ -10,6 +10,153 @@ #include "transaction.h" #include "block-group.h" +/* + * HOW DOES SPACE RESERVATION WORK + * + * If you want to know about delalloc specifically, there is a separate comment + * for that with the delalloc code. This comment is about how the whole system + * works generally. + * + * BASIC CONCEPTS + * + * 1) space_info. This is the ultimate arbiter of how much space we can use. + * There's a description of the bytes_ fields with the struct declaration, + * refer to that for specifics on each field. Suffice it to say that for + * reservations we care about total_bytes - SUM(space_info->bytes_) when + * determining if there is space to make an allocation. There is a space_info + * for METADATA, SYSTEM, and DATA areas. + * + * 2) block_rsv's. These are basically buckets for every different type of + * metadata reservation we have. You can see the comment in the block_rsv + * code on the rules for each type, but generally block_rsv->reserved is how + * much space is accounted for in space_info->bytes_may_use. + * + * 3) btrfs_calc*_size. These are the worst case calculations we used based + * on the number of items we will want to modify. We have one for changing + * items, and one for inserting new items. Generally we use these helpers to + * determine the size of the block reserves, and then use the actual bytes + * values to adjust the space_info counters. + * + * MAKING RESERVATIONS, THE NORMAL CASE + * + * We call into either btrfs_reserve_data_bytes() or + * btrfs_reserve_metadata_bytes(), depending on which we're looking for, with + * num_bytes we want to reserve. + * + * ->reserve + * space_info->bytes_may_reserve += num_bytes + * + * ->extent allocation + * Call btrfs_add_reserved_bytes() which does + * space_info->bytes_may_reserve -= num_bytes + * space_info->bytes_reserved += extent_bytes + * + * ->insert reference + * Call btrfs_update_block_group() which does + * space_info->bytes_reserved -= extent_bytes + * space_info->bytes_used += extent_bytes + * + * MAKING RESERVATIONS, FLUSHING NORMALLY (non-priority) + * + * Assume we are unable to simply make the reservation because we do not have + * enough space + * + * -> __reserve_bytes + * create a reserve_ticket with ->bytes set to our reservation, add it to + * the tail of space_info->tickets, kick async flush thread + * + * ->handle_reserve_ticket + * wait on ticket->wait for ->bytes to be reduced to 0, or ->error to be set + * on the ticket. + * + * -> btrfs_async_reclaim_metadata_space/btrfs_async_reclaim_data_space + * Flushes various things attempting to free up space. + * + * -> btrfs_try_granting_tickets() + * This is called by anything that either subtracts space from + * space_info->bytes_may_use, ->bytes_pinned, etc, or adds to the + * space_info->total_bytes. This loops through the ->priority_tickets and + * then the ->tickets list checking to see if the reservation can be + * completed. If it can the space is added to space_info->bytes_may_use and + * the ticket is woken up. + * + * -> ticket wakeup + * Check if ->bytes == 0, if it does we got our reservation and we can carry + * on, if not return the appropriate error (ENOSPC, but can be EINTR if we + * were interrupted.) + * + * MAKING RESERVATIONS, FLUSHING HIGH PRIORITY + * + * Same as the above, except we add ourselves to the + * space_info->priority_tickets, and we do not use ticket->wait, we simply + * call flush_space() ourselves for the states that are safe for us to call + * without deadlocking and hope for the best. + * + * THE FLUSHING STATES + * + * Generally speaking we will have two cases for each state, a "nice" state + * and a "ALL THE THINGS" state. In btrfs we delay a lot of work in order to + * reduce the locking over head on the various trees, and even to keep from + * doing any work at all in the case of delayed refs. Each of these delayed + * things however hold reservations, and so letting them run allows us to + * reclaim space so we can make new reservations. + * + * FLUSH_DELAYED_ITEMS + * Every inode has a delayed item to update the inode. Take a simple write + * for example, we would update the inode item at write time to update the + * mtime, and then again at finish_ordered_io() time in order to update the + * isize or bytes. We keep these delayed items to coalesce these operations + * into a single operation done on demand. These are an easy way to reclaim + * metadata space. + * + * FLUSH_DELALLOC + * Look at the delalloc comment to get an idea of how much space is reserved + * for delayed allocation. We can reclaim some of this space simply by + * running delalloc, but usually we need to wait for ordered extents to + * reclaim the bulk of this space. + * + * FLUSH_DELAYED_REFS + * We have a block reserve for the outstanding delayed refs space, and every + * delayed ref operation holds a reservation. Running these is a quick way + * to reclaim space, but we want to hold this until the end because COW can + * churn a lot and we can avoid making some extent tree modifications if we + * are able to delay for as long as possible. + * + * ALLOC_CHUNK + * We will skip this the first time through space reservation, because of + * overcommit and we don't want to have a lot of useless metadata space when + * our worst case reservations will likely never come true. + * + * RUN_DELAYED_IPUTS + * If we're freeing inodes we're likely freeing checksums, file extent + * items, and extent tree items. Loads of space could be freed up by these + * operations, however they won't be usable until the transaction commits. + * + * COMMIT_TRANS + * may_commit_transaction() is the ultimate arbiter on whether we commit the + * transaction or not. In order to avoid constantly churning we do all the + * above flushing first and then commit the transaction as the last resort. + * However we need to take into account things like pinned space that would + * be freed, plus any delayed work we may not have gotten rid of in the case + * of metadata. + * + * OVERCOMMIT + * + * Because we hold so many reservations for metadata we will allow you to + * reserve more space than is currently free in the currently allocate + * metadata space. This only happens with metadata, data does not allow + * overcommitting. + * + * You can see the current logic for when we allow overcommit in + * btrfs_can_overcommit(), but it only applies to unallocated space. If there + * is no unallocated space to be had, all reservations are kept within the + * free space in the allocated metadata chunks. + * + * Because of overcommitting, you generally want to use the + * btrfs_can_overcommit() logic for metadata allocations, as it does the right + * thing with or without extra unallocated space. + */ + u64 __pure btrfs_space_info_used(struct btrfs_space_info *s_info, bool may_use_included) { @@ -159,25 +306,19 @@ static inline u64 calc_global_rsv_need_space(struct btrfs_block_rsv *global) return (global->size << 1); } -int btrfs_can_overcommit(struct btrfs_fs_info *fs_info, - struct btrfs_space_info *space_info, u64 bytes, - enum btrfs_reserve_flush_enum flush) +static u64 calc_available_free_space(struct btrfs_fs_info *fs_info, + struct btrfs_space_info *space_info, + enum btrfs_reserve_flush_enum flush) { u64 profile; u64 avail; - u64 used; int factor; - /* Don't overcommit when in mixed mode. */ - if (space_info->flags & BTRFS_BLOCK_GROUP_DATA) - return 0; - if (space_info->flags & BTRFS_BLOCK_GROUP_SYSTEM) profile = btrfs_system_alloc_profile(fs_info); else profile = btrfs_metadata_alloc_profile(fs_info); - used = btrfs_space_info_used(space_info, true); avail = atomic64_read(&fs_info->free_chunk_space); /* @@ -198,6 +339,22 @@ int btrfs_can_overcommit(struct btrfs_fs_info *fs_info, avail >>= 3; else avail >>= 1; + return avail; +} + +int btrfs_can_overcommit(struct btrfs_fs_info *fs_info, + struct btrfs_space_info *space_info, u64 bytes, + enum btrfs_reserve_flush_enum flush) +{ + u64 avail; + u64 used; + + /* Don't overcommit when in mixed mode */ + if (space_info->flags & BTRFS_BLOCK_GROUP_DATA) + return 0; + + used = btrfs_space_info_used(space_info, true); + avail = calc_available_free_space(fs_info, space_info, flush); if (used + bytes < space_info->total_bytes + avail) return 1; @@ -232,6 +389,8 @@ again: space_info, ticket->bytes); list_del_init(&ticket->list); + ASSERT(space_info->reclaim_size >= ticket->bytes); + space_info->reclaim_size -= ticket->bytes; ticket->bytes = 0; space_info->tickets_id++; wake_up(&ticket->wait); @@ -627,15 +786,26 @@ static inline u64 btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info, struct btrfs_space_info *space_info) { - struct reserve_ticket *ticket; u64 used; + u64 avail; u64 expected; - u64 to_reclaim = 0; + u64 to_reclaim = space_info->reclaim_size; + + lockdep_assert_held(&space_info->lock); + + avail = calc_available_free_space(fs_info, space_info, + BTRFS_RESERVE_FLUSH_ALL); + used = btrfs_space_info_used(space_info, true); + + /* + * We may be flushing because suddenly we have less space than we had + * before, and now we're well over-committed based on our current free + * space. If that's the case add in our overage so we make sure to put + * appropriate pressure on the flushing state machine. + */ + if (space_info->total_bytes + avail < used) + to_reclaim += used - (space_info->total_bytes + avail); - list_for_each_entry(ticket, &space_info->tickets, list) - to_reclaim += ticket->bytes; - list_for_each_entry(ticket, &space_info->priority_tickets, list) - to_reclaim += ticket->bytes; if (to_reclaim) return to_reclaim; @@ -1020,8 +1190,10 @@ static int __reserve_metadata_bytes(struct btrfs_fs_info *fs_info, * the list and we will do our own flushing further down. */ if (ret && flush != BTRFS_RESERVE_NO_FLUSH) { + ASSERT(space_info->reclaim_size >= 0); ticket.bytes = orig_bytes; ticket.error = 0; + space_info->reclaim_size += ticket.bytes; init_waitqueue_head(&ticket.wait); if (flush == BTRFS_RESERVE_FLUSH_ALL) { list_add_tail(&ticket.list, &space_info->tickets); diff --git a/fs/btrfs/space-info.h b/fs/btrfs/space-info.h index 24514cd2c6c1..0a5001ef1481 100644 --- a/fs/btrfs/space-info.h +++ b/fs/btrfs/space-info.h @@ -54,6 +54,13 @@ struct btrfs_space_info { struct list_head ro_bgs; struct list_head priority_tickets; struct list_head tickets; + + /* + * Size of space that needs to be reclaimed in order to satisfy pending + * tickets + */ + u64 reclaim_size; + /* * tickets_id just indicates the next ticket will be handled, so note * it's not stored per ticket. diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 67c63858812a..7932d8d07cff 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -244,7 +244,7 @@ void __btrfs_abort_transaction(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *fs_info = trans->fs_info; - trans->aborted = errno; + WRITE_ONCE(trans->aborted, errno); /* Nothing used. The other threads that have joined this * transaction may be able to continue. */ if (!trans->dirty && list_empty(&trans->new_bgs)) { @@ -873,7 +873,7 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, break; #endif case Opt_err: - btrfs_info(info, "unrecognized mount option '%s'", p); + btrfs_err(info, "unrecognized mount option '%s'", p); ret = -EINVAL; goto out; default: @@ -1024,11 +1024,11 @@ out: return error; } -static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info, - u64 subvol_objectid) +char *btrfs_get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info, + u64 subvol_objectid) { struct btrfs_root *root = fs_info->tree_root; - struct btrfs_root *fs_root; + struct btrfs_root *fs_root = NULL; struct btrfs_root_ref *root_ref; struct btrfs_inode_ref *inode_ref; struct btrfs_key key; @@ -1096,9 +1096,10 @@ static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info, key.objectid = subvol_objectid; key.type = BTRFS_ROOT_ITEM_KEY; key.offset = (u64)-1; - fs_root = btrfs_read_fs_root_no_name(fs_info, &key); + fs_root = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(fs_root)) { ret = PTR_ERR(fs_root); + fs_root = NULL; goto err; } @@ -1143,6 +1144,8 @@ static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info, ptr[0] = '/'; btrfs_release_path(path); } + btrfs_put_root(fs_root); + fs_root = NULL; } btrfs_free_path(path); @@ -1155,6 +1158,7 @@ static char *get_subvol_name_from_objectid(struct btrfs_fs_info *fs_info, return name; err: + btrfs_put_root(fs_root); btrfs_free_path(path); kfree(name); return ERR_PTR(ret); @@ -1438,8 +1442,8 @@ static struct dentry *mount_subvol(const char *subvol_name, u64 subvol_objectid, goto out; } } - subvol_name = get_subvol_name_from_objectid(btrfs_sb(mnt->mnt_sb), - subvol_objectid); + subvol_name = btrfs_get_subvol_name_from_objectid( + btrfs_sb(mnt->mnt_sb), subvol_objectid); if (IS_ERR(subvol_name)) { root = ERR_CAST(subvol_name); subvol_name = NULL; @@ -1518,14 +1522,17 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, /* * Setup a dummy root and fs_info for test/set super. This is because * we don't actually fill this stuff out until open_ctree, but we need - * it for searching for existing supers, so this lets us do that and - * then open_ctree will properly initialize everything later. + * then open_ctree will properly initialize the file system specific + * settings later. btrfs_init_fs_info initializes the static elements + * of the fs_info (locks and such) to make cleanup easier if we find a + * superblock with our given fs_devices later on at sget() time. */ fs_info = kvzalloc(sizeof(struct btrfs_fs_info), GFP_KERNEL); if (!fs_info) { error = -ENOMEM; goto error_sec_opts; } + btrfs_init_fs_info(fs_info); fs_info->super_copy = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL); fs_info->super_for_commit = kzalloc(BTRFS_SUPER_INFO_SIZE, GFP_KERNEL); @@ -1571,7 +1578,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, if (s->s_root) { btrfs_close_devices(fs_devices); - free_fs_info(fs_info); + btrfs_free_fs_info(fs_info); if ((flags ^ s->s_flags) & SB_RDONLY) error = -EBUSY; } else { @@ -1594,7 +1601,7 @@ static struct dentry *btrfs_mount_root(struct file_system_type *fs_type, error_close_devices: btrfs_close_devices(fs_devices); error_fs_info: - free_fs_info(fs_info); + btrfs_free_fs_info(fs_info); error_sec_opts: security_free_mnt_opts(&new_sec_opts); return ERR_PTR(error); @@ -2170,7 +2177,7 @@ static void btrfs_kill_super(struct super_block *sb) { struct btrfs_fs_info *fs_info = btrfs_sb(sb); kill_anon_super(sb); - free_fs_info(fs_info); + btrfs_free_fs_info(fs_info); } static struct file_system_type btrfs_fs_type = { @@ -2203,7 +2210,7 @@ static int btrfs_control_open(struct inode *inode, struct file *file) } /* - * used by btrfsctl to scan devices when no FS is mounted + * Used by /dev/btrfs-control for devices ioctls. */ static long btrfs_control_ioctl(struct file *file, unsigned int cmd, unsigned long arg) diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index 3c10e78924d0..a39bff64ff24 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -155,7 +155,7 @@ static ssize_t btrfs_feature_attr_show(struct kobject *kobj, } else val = can_modify_feature(fa); - return snprintf(buf, PAGE_SIZE, "%d\n", val); + return scnprintf(buf, PAGE_SIZE, "%d\n", val); } static ssize_t btrfs_feature_attr_store(struct kobject *kobj, @@ -295,7 +295,7 @@ static const struct attribute_group btrfs_feature_attr_group = { static ssize_t rmdir_subvol_show(struct kobject *kobj, struct kobj_attribute *ka, char *buf) { - return snprintf(buf, PAGE_SIZE, "0\n"); + return scnprintf(buf, PAGE_SIZE, "0\n"); } BTRFS_ATTR(static_feature, rmdir_subvol, rmdir_subvol_show); @@ -310,12 +310,12 @@ static ssize_t supported_checksums_show(struct kobject *kobj, * This "trick" only works as long as 'enum btrfs_csum_type' has * no holes in it */ - ret += snprintf(buf + ret, PAGE_SIZE - ret, "%s%s", + ret += scnprintf(buf + ret, PAGE_SIZE - ret, "%s%s", (i == 0 ? "" : " "), btrfs_super_csum_name(i)); } - ret += snprintf(buf + ret, PAGE_SIZE - ret, "\n"); + ret += scnprintf(buf + ret, PAGE_SIZE - ret, "\n"); return ret; } BTRFS_ATTR(static_feature, supported_checksums, supported_checksums_show); @@ -350,7 +350,7 @@ static ssize_t btrfs_discardable_bytes_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%lld\n", + return scnprintf(buf, PAGE_SIZE, "%lld\n", atomic64_read(&fs_info->discard_ctl.discardable_bytes)); } BTRFS_ATTR(discard, discardable_bytes, btrfs_discardable_bytes_show); @@ -361,7 +361,7 @@ static ssize_t btrfs_discardable_extents_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%d\n", + return scnprintf(buf, PAGE_SIZE, "%d\n", atomic_read(&fs_info->discard_ctl.discardable_extents)); } BTRFS_ATTR(discard, discardable_extents, btrfs_discardable_extents_show); @@ -372,7 +372,7 @@ static ssize_t btrfs_discard_bitmap_bytes_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%lld\n", + return scnprintf(buf, PAGE_SIZE, "%lld\n", fs_info->discard_ctl.discard_bitmap_bytes); } BTRFS_ATTR(discard, discard_bitmap_bytes, btrfs_discard_bitmap_bytes_show); @@ -383,7 +383,7 @@ static ssize_t btrfs_discard_bytes_saved_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%lld\n", + return scnprintf(buf, PAGE_SIZE, "%lld\n", atomic64_read(&fs_info->discard_ctl.discard_bytes_saved)); } BTRFS_ATTR(discard, discard_bytes_saved, btrfs_discard_bytes_saved_show); @@ -394,7 +394,7 @@ static ssize_t btrfs_discard_extent_bytes_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%lld\n", + return scnprintf(buf, PAGE_SIZE, "%lld\n", fs_info->discard_ctl.discard_extent_bytes); } BTRFS_ATTR(discard, discard_extent_bytes, btrfs_discard_extent_bytes_show); @@ -405,7 +405,7 @@ static ssize_t btrfs_discard_iops_limit_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%u\n", + return scnprintf(buf, PAGE_SIZE, "%u\n", READ_ONCE(fs_info->discard_ctl.iops_limit)); } @@ -435,7 +435,7 @@ static ssize_t btrfs_discard_kbps_limit_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%u\n", + return scnprintf(buf, PAGE_SIZE, "%u\n", READ_ONCE(fs_info->discard_ctl.kbps_limit)); } @@ -465,7 +465,7 @@ static ssize_t btrfs_discard_max_discard_size_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = discard_to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%llu\n", + return scnprintf(buf, PAGE_SIZE, "%llu\n", READ_ONCE(fs_info->discard_ctl.max_discard_size)); } @@ -530,7 +530,7 @@ static ssize_t btrfs_show_u64(u64 *value_ptr, spinlock_t *lock, char *buf) val = *value_ptr; if (lock) spin_unlock(lock); - return snprintf(buf, PAGE_SIZE, "%llu\n", val); + return scnprintf(buf, PAGE_SIZE, "%llu\n", val); } static ssize_t global_rsv_size_show(struct kobject *kobj, @@ -576,7 +576,7 @@ static ssize_t raid_bytes_show(struct kobject *kobj, val += block_group->used; } up_read(&sinfo->groups_sem); - return snprintf(buf, PAGE_SIZE, "%llu\n", val); + return scnprintf(buf, PAGE_SIZE, "%llu\n", val); } static struct attribute *raid_attrs[] = { @@ -613,7 +613,7 @@ static ssize_t btrfs_space_info_show_total_bytes_pinned(struct kobject *kobj, { struct btrfs_space_info *sinfo = to_space_info(kobj); s64 val = percpu_counter_sum(&sinfo->total_bytes_pinned); - return snprintf(buf, PAGE_SIZE, "%lld\n", val); + return scnprintf(buf, PAGE_SIZE, "%lld\n", val); } SPACE_INFO_ATTR(flags); @@ -670,7 +670,7 @@ static ssize_t btrfs_label_show(struct kobject *kobj, ssize_t ret; spin_lock(&fs_info->super_lock); - ret = snprintf(buf, PAGE_SIZE, label[0] ? "%s\n" : "%s", label); + ret = scnprintf(buf, PAGE_SIZE, label[0] ? "%s\n" : "%s", label); spin_unlock(&fs_info->super_lock); return ret; @@ -718,7 +718,7 @@ static ssize_t btrfs_nodesize_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->nodesize); + return scnprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->nodesize); } BTRFS_ATTR(, nodesize, btrfs_nodesize_show); @@ -728,8 +728,8 @@ static ssize_t btrfs_sectorsize_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%u\n", - fs_info->super_copy->sectorsize); + return scnprintf(buf, PAGE_SIZE, "%u\n", + fs_info->super_copy->sectorsize); } BTRFS_ATTR(, sectorsize, btrfs_sectorsize_show); @@ -739,8 +739,7 @@ static ssize_t btrfs_clone_alignment_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%u\n", - fs_info->super_copy->sectorsize); + return scnprintf(buf, PAGE_SIZE, "%u\n", fs_info->super_copy->sectorsize); } BTRFS_ATTR(, clone_alignment, btrfs_clone_alignment_show); @@ -752,7 +751,7 @@ static ssize_t quota_override_show(struct kobject *kobj, int quota_override; quota_override = test_bit(BTRFS_FS_QUOTA_OVERRIDE, &fs_info->flags); - return snprintf(buf, PAGE_SIZE, "%d\n", quota_override); + return scnprintf(buf, PAGE_SIZE, "%d\n", quota_override); } static ssize_t quota_override_store(struct kobject *kobj, @@ -790,7 +789,7 @@ static ssize_t btrfs_metadata_uuid_show(struct kobject *kobj, { struct btrfs_fs_info *fs_info = to_fs_info(kobj); - return snprintf(buf, PAGE_SIZE, "%pU\n", + return scnprintf(buf, PAGE_SIZE, "%pU\n", fs_info->fs_devices->metadata_uuid); } @@ -802,7 +801,7 @@ static ssize_t btrfs_checksum_show(struct kobject *kobj, struct btrfs_fs_info *fs_info = to_fs_info(kobj); u16 csum_type = btrfs_super_csum_type(fs_info->super_copy); - return snprintf(buf, PAGE_SIZE, "%s (%s)\n", + return scnprintf(buf, PAGE_SIZE, "%s (%s)\n", btrfs_super_csum_name(csum_type), crypto_shash_driver_name(fs_info->csum_shash)); } @@ -960,7 +959,7 @@ void btrfs_sysfs_remove_mounted(struct btrfs_fs_info *fs_info) addrm_unknown_feature_attrs(fs_info, false); sysfs_remove_group(&fs_info->fs_devices->fsid_kobj, &btrfs_feature_attr_group); sysfs_remove_files(&fs_info->fs_devices->fsid_kobj, btrfs_attrs); - btrfs_sysfs_rm_device_link(fs_info->fs_devices, NULL); + btrfs_sysfs_remove_devices_dir(fs_info->fs_devices, NULL); } static const char * const btrfs_feature_set_names[FEAT_MAX] = { @@ -992,7 +991,7 @@ char *btrfs_printable_features(enum btrfs_feature_set set, u64 flags) continue; name = btrfs_feature_attrs[set][i].kobj_attr.attr.name; - len += snprintf(str + len, bufsize - len, "%s%s", + len += scnprintf(str + len, bufsize - len, "%s%s", len ? "," : "", name); } @@ -1149,7 +1148,7 @@ int btrfs_sysfs_add_space_info_type(struct btrfs_fs_info *fs_info, /* when one_device is NULL, it removes all device links */ -int btrfs_sysfs_rm_device_link(struct btrfs_fs_devices *fs_devices, +int btrfs_sysfs_remove_devices_dir(struct btrfs_fs_devices *fs_devices, struct btrfs_device *one_device) { struct hd_struct *disk; @@ -1201,11 +1200,11 @@ static ssize_t btrfs_devinfo_in_fs_metadata_show(struct kobject *kobj, val = !!test_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); - return snprintf(buf, PAGE_SIZE, "%d\n", val); + return scnprintf(buf, PAGE_SIZE, "%d\n", val); } BTRFS_ATTR(devid, in_fs_metadata, btrfs_devinfo_in_fs_metadata_show); -static ssize_t btrfs_sysfs_missing_show(struct kobject *kobj, +static ssize_t btrfs_devinfo_missing_show(struct kobject *kobj, struct kobj_attribute *a, char *buf) { int val; @@ -1214,9 +1213,9 @@ static ssize_t btrfs_sysfs_missing_show(struct kobject *kobj, val = !!test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state); - return snprintf(buf, PAGE_SIZE, "%d\n", val); + return scnprintf(buf, PAGE_SIZE, "%d\n", val); } -BTRFS_ATTR(devid, missing, btrfs_sysfs_missing_show); +BTRFS_ATTR(devid, missing, btrfs_devinfo_missing_show); static ssize_t btrfs_devinfo_replace_target_show(struct kobject *kobj, struct kobj_attribute *a, @@ -1228,7 +1227,7 @@ static ssize_t btrfs_devinfo_replace_target_show(struct kobject *kobj, val = !!test_bit(BTRFS_DEV_STATE_REPLACE_TGT, &device->dev_state); - return snprintf(buf, PAGE_SIZE, "%d\n", val); + return scnprintf(buf, PAGE_SIZE, "%d\n", val); } BTRFS_ATTR(devid, replace_target, btrfs_devinfo_replace_target_show); @@ -1241,7 +1240,7 @@ static ssize_t btrfs_devinfo_writeable_show(struct kobject *kobj, val = !!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state); - return snprintf(buf, PAGE_SIZE, "%d\n", val); + return scnprintf(buf, PAGE_SIZE, "%d\n", val); } BTRFS_ATTR(devid, writeable, btrfs_devinfo_writeable_show); @@ -1269,7 +1268,7 @@ static struct kobj_type devid_ktype = { .release = btrfs_release_devid_kobj, }; -int btrfs_sysfs_add_device_link(struct btrfs_fs_devices *fs_devices, +int btrfs_sysfs_add_devices_dir(struct btrfs_fs_devices *fs_devices, struct btrfs_device *one_device) { int error = 0; @@ -1371,7 +1370,7 @@ int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs) if (!fs_devs->devices_kobj) { btrfs_err(fs_devs->fs_info, "failed to init sysfs device interface"); - kobject_put(&fs_devs->fsid_kobj); + btrfs_sysfs_remove_fsid(fs_devs); return -ENOMEM; } @@ -1395,13 +1394,13 @@ int btrfs_sysfs_add_mounted(struct btrfs_fs_info *fs_info) btrfs_set_fs_info_ptr(fs_info); - error = btrfs_sysfs_add_device_link(fs_devs, NULL); + error = btrfs_sysfs_add_devices_dir(fs_devs, NULL); if (error) return error; error = sysfs_create_files(fsid_kobj, btrfs_attrs); if (error) { - btrfs_sysfs_rm_device_link(fs_devs, NULL); + btrfs_sysfs_remove_devices_dir(fs_devs, NULL); return error; } diff --git a/fs/btrfs/sysfs.h b/fs/btrfs/sysfs.h index c68582add92e..718a26c97833 100644 --- a/fs/btrfs/sysfs.h +++ b/fs/btrfs/sysfs.h @@ -14,9 +14,9 @@ enum btrfs_feature_set { char *btrfs_printable_features(enum btrfs_feature_set set, u64 flags); const char * const btrfs_feature_set_name(enum btrfs_feature_set set); -int btrfs_sysfs_add_device_link(struct btrfs_fs_devices *fs_devices, +int btrfs_sysfs_add_devices_dir(struct btrfs_fs_devices *fs_devices, struct btrfs_device *one_device); -int btrfs_sysfs_rm_device_link(struct btrfs_fs_devices *fs_devices, +int btrfs_sysfs_remove_devices_dir(struct btrfs_fs_devices *fs_devices, struct btrfs_device *one_device); int btrfs_sysfs_add_fsid(struct btrfs_fs_devices *fs_devs); void btrfs_sysfs_remove_fsid(struct btrfs_fs_devices *fs_devs); diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c index 84fb3fa940a6..999c14e5d0bd 100644 --- a/fs/btrfs/tests/btrfs-tests.c +++ b/fs/btrfs/tests/btrfs-tests.c @@ -120,6 +120,8 @@ struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize) kfree(fs_info); return NULL; } + INIT_LIST_HEAD(&fs_info->fs_devices->devices); + fs_info->super_copy = kzalloc(sizeof(struct btrfs_super_block), GFP_KERNEL); if (!fs_info->super_copy) { @@ -128,39 +130,10 @@ struct btrfs_fs_info *btrfs_alloc_dummy_fs_info(u32 nodesize, u32 sectorsize) return NULL; } + btrfs_init_fs_info(fs_info); + fs_info->nodesize = nodesize; fs_info->sectorsize = sectorsize; - - if (init_srcu_struct(&fs_info->subvol_srcu)) { - kfree(fs_info->fs_devices); - kfree(fs_info->super_copy); - kfree(fs_info); - return NULL; - } - - spin_lock_init(&fs_info->buffer_lock); - spin_lock_init(&fs_info->qgroup_lock); - spin_lock_init(&fs_info->super_lock); - spin_lock_init(&fs_info->fs_roots_radix_lock); - mutex_init(&fs_info->qgroup_ioctl_lock); - mutex_init(&fs_info->qgroup_rescan_lock); - rwlock_init(&fs_info->tree_mod_log_lock); - fs_info->running_transaction = NULL; - fs_info->qgroup_tree = RB_ROOT; - fs_info->qgroup_ulist = NULL; - atomic64_set(&fs_info->tree_mod_seq, 0); - INIT_LIST_HEAD(&fs_info->dirty_qgroups); - INIT_LIST_HEAD(&fs_info->dead_roots); - INIT_LIST_HEAD(&fs_info->tree_mod_seq_list); - INIT_LIST_HEAD(&fs_info->fs_devices->devices); - INIT_RADIX_TREE(&fs_info->buffer_radix, GFP_ATOMIC); - INIT_RADIX_TREE(&fs_info->fs_roots_radix, GFP_ATOMIC); - extent_io_tree_init(fs_info, &fs_info->freed_extents[0], - IO_TREE_FS_INFO_FREED_EXTENTS0, NULL); - extent_io_tree_init(fs_info, &fs_info->freed_extents[1], - IO_TREE_FS_INFO_FREED_EXTENTS1, NULL); - extent_map_tree_init(&fs_info->mapping_tree); - fs_info->pinned_extents = &fs_info->freed_extents[0]; set_bit(BTRFS_FS_STATE_DUMMY_FS_INFO, &fs_info->fs_state); test_mnt->mnt_sb->s_fs_info = fs_info; @@ -210,8 +183,9 @@ void btrfs_free_dummy_fs_info(struct btrfs_fs_info *fs_info) } btrfs_free_qgroup_config(fs_info); btrfs_free_fs_roots(fs_info); - cleanup_srcu_struct(&fs_info->subvol_srcu); kfree(fs_info->super_copy); + btrfs_check_leaked_roots(fs_info); + btrfs_extent_buffer_leak_debug_check(fs_info); kfree(fs_info->fs_devices); kfree(fs_info); } @@ -223,11 +197,7 @@ void btrfs_free_dummy_root(struct btrfs_root *root) /* Will be freed by btrfs_free_fs_roots */ if (WARN_ON(test_bit(BTRFS_ROOT_IN_RADIX, &root->state))) return; - if (root->node) { - /* One for allocate_extent_buffer */ - free_extent_buffer(root->node); - } - kfree(root); + btrfs_put_root(root); } struct btrfs_block_group * diff --git a/fs/btrfs/tests/qgroup-tests.c b/fs/btrfs/tests/qgroup-tests.c index ac035a6fa003..ce1ca8e73c2d 100644 --- a/fs/btrfs/tests/qgroup-tests.c +++ b/fs/btrfs/tests/qgroup-tests.c @@ -507,6 +507,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize) test_err("couldn't insert fs root %d", ret); goto out; } + btrfs_put_root(tmp_root); tmp_root = btrfs_alloc_dummy_root(fs_info); if (IS_ERR(tmp_root)) { @@ -521,6 +522,7 @@ int btrfs_test_qgroups(u32 sectorsize, u32 nodesize) test_err("couldn't insert fs root %d", ret); goto out; } + btrfs_put_root(tmp_root); test_msg("running qgroup tests"); ret = test_no_shared_qgroup(root, sectorsize, nodesize); diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 33dcc88b428a..8cede6eb9843 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -121,6 +121,8 @@ void btrfs_put_transaction(struct btrfs_transaction *transaction) BUG_ON(!list_empty(&transaction->list)); WARN_ON(!RB_EMPTY_ROOT( &transaction->delayed_refs.href_root.rb_root)); + WARN_ON(!RB_EMPTY_ROOT( + &transaction->delayed_refs.dirty_extent_root)); if (transaction->delayed_refs.pending_csums) btrfs_err(transaction->fs_info, "pending csums is %llu", @@ -219,7 +221,7 @@ void btrfs_trans_release_chunk_metadata(struct btrfs_trans_handle *trans) WARN_ON_ONCE(!list_empty(&trans->new_bgs)); btrfs_block_rsv_release(fs_info, &fs_info->chunk_block_rsv, - trans->chunk_bytes_reserved); + trans->chunk_bytes_reserved, NULL); trans->chunk_bytes_reserved = 0; } @@ -241,7 +243,7 @@ loop: cur_trans = fs_info->running_transaction; if (cur_trans) { - if (cur_trans->aborted) { + if (TRANS_ABORTED(cur_trans)) { spin_unlock(&fs_info->trans_lock); return cur_trans->aborted; } @@ -334,6 +336,8 @@ loop: list_add_tail(&cur_trans->list, &fs_info->trans_list); extent_io_tree_init(fs_info, &cur_trans->dirty_pages, IO_TREE_TRANS_DIRTY_PAGES, fs_info->btree_inode); + extent_io_tree_init(fs_info, &cur_trans->pinned_extents, + IO_TREE_FS_PINNED_EXTENTS, NULL); fs_info->generation++; cur_trans->transid = fs_info->generation; fs_info->running_transaction = cur_trans; @@ -457,7 +461,7 @@ static inline int is_transaction_blocked(struct btrfs_transaction *trans) { return (trans->state >= TRANS_STATE_COMMIT_START && trans->state < TRANS_STATE_UNBLOCKED && - !trans->aborted); + !TRANS_ABORTED(trans)); } /* wait for commit against the current transaction to become unblocked @@ -476,7 +480,7 @@ static void wait_current_trans(struct btrfs_fs_info *fs_info) wait_event(fs_info->transaction_wait, cur_trans->state >= TRANS_STATE_UNBLOCKED || - cur_trans->aborted); + TRANS_ABORTED(cur_trans)); btrfs_put_transaction(cur_trans); } else { spin_unlock(&fs_info->trans_lock); @@ -671,7 +675,7 @@ join_fail: alloc_fail: if (num_bytes) btrfs_block_rsv_release(fs_info, &fs_info->trans_block_rsv, - num_bytes); + num_bytes, NULL); reserve_fail: btrfs_qgroup_free_meta_pertrans(root, qgroup_reserved); return ERR_PTR(ret); @@ -894,7 +898,7 @@ static void btrfs_trans_release_metadata(struct btrfs_trans_handle *trans) trace_btrfs_space_reservation(fs_info, "transaction", trans->transid, trans->bytes_reserved, 0); btrfs_block_rsv_release(fs_info, trans->block_rsv, - trans->bytes_reserved); + trans->bytes_reserved, NULL); trans->bytes_reserved = 0; } @@ -935,7 +939,7 @@ static int __btrfs_end_transaction(struct btrfs_trans_handle *trans, if (throttle) btrfs_run_delayed_iputs(info); - if (trans->aborted || + if (TRANS_ABORTED(trans) || test_bit(BTRFS_FS_STATE_ERROR, &info->fs_state)) { wake_up_process(info->transaction_kthread); err = -EIO; @@ -1260,8 +1264,10 @@ void btrfs_add_dead_root(struct btrfs_root *root) struct btrfs_fs_info *fs_info = root->fs_info; spin_lock(&fs_info->trans_lock); - if (list_empty(&root->root_list)) + if (list_empty(&root->root_list)) { + btrfs_grab_root(root); list_add_tail(&root->root_list, &fs_info->dead_roots); + } spin_unlock(&fs_info->trans_lock); } @@ -1475,7 +1481,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, u64 index = 0; u64 objectid; u64 root_flags; - uuid_le new_uuid; ASSERT(pending->path); path = pending->path; @@ -1568,8 +1573,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, btrfs_set_root_generation_v2(new_root_item, trans->transid); - uuid_le_gen(&new_uuid); - memcpy(new_root_item->uuid, new_uuid.b, BTRFS_UUID_SIZE); + generate_random_guid(new_root_item->uuid); memcpy(new_root_item->parent_uuid, root->root_item.uuid, BTRFS_UUID_SIZE); if (!(root_flags & BTRFS_ROOT_SUBVOL_RDONLY)) { @@ -1631,7 +1635,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, } key.offset = (u64)-1; - pending->snap = btrfs_read_fs_root_no_name(fs_info, &key); + pending->snap = btrfs_get_fs_root(fs_info, &key, true); if (IS_ERR(pending->snap)) { ret = PTR_ERR(pending->snap); btrfs_abort_transaction(trans, ret); @@ -1680,7 +1684,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, btrfs_abort_transaction(trans, ret); goto fail; } - ret = btrfs_uuid_tree_add(trans, new_uuid.b, BTRFS_UUID_KEY_SUBVOL, + ret = btrfs_uuid_tree_add(trans, new_root_item->uuid, + BTRFS_UUID_KEY_SUBVOL, objectid); if (ret) { btrfs_abort_transaction(trans, ret); @@ -1792,7 +1797,8 @@ static void wait_current_trans_commit_start(struct btrfs_fs_info *fs_info, struct btrfs_transaction *trans) { wait_event(fs_info->transaction_blocked_wait, - trans->state >= TRANS_STATE_COMMIT_START || trans->aborted); + trans->state >= TRANS_STATE_COMMIT_START || + TRANS_ABORTED(trans)); } /* @@ -1804,7 +1810,8 @@ static void wait_current_trans_commit_start_and_unblock( struct btrfs_transaction *trans) { wait_event(fs_info->transaction_wait, - trans->state >= TRANS_STATE_UNBLOCKED || trans->aborted); + trans->state >= TRANS_STATE_UNBLOCKED || + TRANS_ABORTED(trans)); } /* @@ -2024,7 +2031,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) trans->dirty = true; /* Stop the commit early if ->aborted is set */ - if (unlikely(READ_ONCE(cur_trans->aborted))) { + if (TRANS_ABORTED(cur_trans)) { ret = cur_trans->aborted; btrfs_end_transaction(trans); return ret; @@ -2098,7 +2105,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) wait_for_commit(cur_trans); - if (unlikely(cur_trans->aborted)) + if (TRANS_ABORTED(cur_trans)) ret = cur_trans->aborted; btrfs_put_transaction(cur_trans); @@ -2117,7 +2124,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) spin_unlock(&fs_info->trans_lock); wait_for_commit(prev_trans); - ret = prev_trans->aborted; + ret = READ_ONCE(prev_trans->aborted); btrfs_put_transaction(prev_trans); if (ret) @@ -2171,8 +2178,7 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) wait_event(cur_trans->writer_wait, atomic_read(&cur_trans->num_writers) == 1); - /* ->aborted might be set after the previous check, so check it */ - if (unlikely(READ_ONCE(cur_trans->aborted))) { + if (TRANS_ABORTED(cur_trans)) { ret = cur_trans->aborted; goto scrub_continue; } @@ -2189,10 +2195,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) * core function of the snapshot creation. */ ret = create_pending_snapshots(trans); - if (ret) { - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; - } + if (ret) + goto unlock_reloc; /* * We insert the dir indexes of the snapshots and update the inode @@ -2205,16 +2209,12 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) * the nodes and leaves. */ ret = btrfs_run_delayed_items(trans); - if (ret) { - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; - } + if (ret) + goto unlock_reloc; ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); - if (ret) { - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; - } + if (ret) + goto unlock_reloc; /* * make sure none of the code above managed to slip in a @@ -2240,11 +2240,8 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) mutex_lock(&fs_info->tree_log_mutex); ret = commit_fs_roots(trans); - if (ret) { - mutex_unlock(&fs_info->tree_log_mutex); - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; - } + if (ret) + goto unlock_tree_log; /* * Since the transaction is done, we can apply the pending changes @@ -2262,39 +2259,28 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) * new delayed refs. Must handle them or qgroup can be wrong. */ ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); - if (ret) { - mutex_unlock(&fs_info->tree_log_mutex); - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; - } + if (ret) + goto unlock_tree_log; /* * Since fs roots are all committed, we can get a quite accurate * new_roots. So let's do quota accounting. */ ret = btrfs_qgroup_account_extents(trans); - if (ret < 0) { - mutex_unlock(&fs_info->tree_log_mutex); - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; - } + if (ret < 0) + goto unlock_tree_log; ret = commit_cowonly_roots(trans); - if (ret) { - mutex_unlock(&fs_info->tree_log_mutex); - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; - } + if (ret) + goto unlock_tree_log; /* * The tasks which save the space cache and inode cache may also * update ->aborted, check it. */ - if (unlikely(READ_ONCE(cur_trans->aborted))) { + if (TRANS_ABORTED(cur_trans)) { ret = cur_trans->aborted; - mutex_unlock(&fs_info->tree_log_mutex); - mutex_unlock(&fs_info->reloc_mutex); - goto scrub_continue; + goto unlock_tree_log; } btrfs_prepare_extent_commit(fs_info); @@ -2341,6 +2327,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) if (ret) { btrfs_handle_fs_error(fs_info, ret, "Error while writing out transaction"); + /* + * reloc_mutex has been unlocked, tree_log_mutex is still held + * but we can't jump to unlock_tree_log causing double unlock + */ mutex_unlock(&fs_info->tree_log_mutex); goto scrub_continue; } @@ -2389,6 +2379,10 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) return ret; +unlock_tree_log: + mutex_unlock(&fs_info->tree_log_mutex); +unlock_reloc: + mutex_unlock(&fs_info->reloc_mutex); scrub_continue: btrfs_scrub_continue(fs_info); cleanup_transaction: @@ -2432,13 +2426,18 @@ int btrfs_clean_one_deleted_snapshot(struct btrfs_root *root) btrfs_debug(fs_info, "cleaner removing %llu", root->root_key.objectid); btrfs_kill_all_delayed_nodes(root); + if (root->ino_cache_inode) { + iput(root->ino_cache_inode); + root->ino_cache_inode = NULL; + } if (btrfs_header_backref_rev(root->node) < BTRFS_MIXED_BACKREF_REV) - ret = btrfs_drop_snapshot(root, NULL, 0, 0); + ret = btrfs_drop_snapshot(root, 0, 0); else - ret = btrfs_drop_snapshot(root, NULL, 1, 0); + ret = btrfs_drop_snapshot(root, 1, 0); + btrfs_put_root(root); return (ret < 0) ? 0 : 1; } diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index 49f7196368f5..31ae8d273065 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -71,6 +71,7 @@ struct btrfs_transaction { */ struct list_head io_bgs; struct list_head dropped_roots; + struct extent_io_tree pinned_extents; /* * we need to make sure block group deletion doesn't race with @@ -115,6 +116,10 @@ struct btrfs_trans_handle { struct btrfs_block_rsv *orig_rsv; refcount_t use_count; unsigned int type; + /* + * Error code of transaction abort, set outside of locks and must use + * the READ_ONCE/WRITE_ONCE access + */ short aborted; bool adding_csums; bool allocating_chunk; @@ -126,6 +131,14 @@ struct btrfs_trans_handle { struct list_head new_bgs; }; +/* + * The abort status can be changed between calls and is not protected by locks. + * This accepts btrfs_transaction and btrfs_trans_handle as types. Once it's + * set to a non-zero value it does not change, so the macro should be in checks + * but is not necessary for further reads of the value. + */ +#define TRANS_ABORTED(trans) (unlikely(READ_ONCE((trans)->aborted))) + struct btrfs_pending_snapshot { struct dentry *dentry; struct inode *dir; diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 7dd7552f53a4..58c111474ba5 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -18,6 +18,8 @@ #include "compression.h" #include "qgroup.h" #include "inode-map.h" +#include "block-group.h" +#include "space-info.h" /* magic values for the inode_only field in btrfs_log_inode: * @@ -94,8 +96,8 @@ enum { static int btrfs_log_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_inode *inode, int inode_only, - const loff_t start, - const loff_t end, + u64 start, + u64 end, struct btrfs_log_ctx *ctx); static int link_to_fixup_dir(struct btrfs_trans_handle *trans, struct btrfs_root *root, @@ -311,7 +313,7 @@ static int process_one_buffer(struct btrfs_root *log, } if (wc->pin) - ret = btrfs_pin_extent_for_log_replay(fs_info, eb->start, + ret = btrfs_pin_extent_for_log_replay(wc->trans, eb->start, eb->len); if (!ret && btrfs_buffer_uptodate(eb, gen, 0)) { @@ -830,6 +832,11 @@ static noinline int replay_one_extent(struct btrfs_trans_handle *trans, goto out; } + ret = btrfs_inode_set_file_extent_range(BTRFS_I(inode), start, + extent_end - start); + if (ret) + goto out; + inode_add_bytes(inode, nbytes); update_inode: ret = btrfs_update_inode(trans, root, inode); @@ -2659,18 +2666,39 @@ static int replay_one_buffer(struct btrfs_root *log, struct extent_buffer *eb, return ret; } +/* + * Correctly adjust the reserved bytes occupied by a log tree extent buffer + */ +static void unaccount_log_buffer(struct btrfs_fs_info *fs_info, u64 start) +{ + struct btrfs_block_group *cache; + + cache = btrfs_lookup_block_group(fs_info, start); + if (!cache) { + btrfs_err(fs_info, "unable to find block group for %llu", start); + return; + } + + spin_lock(&cache->space_info->lock); + spin_lock(&cache->lock); + cache->reserved -= fs_info->nodesize; + cache->space_info->bytes_reserved -= fs_info->nodesize; + spin_unlock(&cache->lock); + spin_unlock(&cache->space_info->lock); + + btrfs_put_block_group(cache); +} + static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_path *path, int *level, struct walk_control *wc) { struct btrfs_fs_info *fs_info = root->fs_info; - u64 root_owner; u64 bytenr; u64 ptr_gen; struct extent_buffer *next; struct extent_buffer *cur; - struct extent_buffer *parent; u32 blocksize; int ret = 0; @@ -2690,9 +2718,6 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, btrfs_node_key_to_cpu(cur, &first_key, path->slots[*level]); blocksize = fs_info->nodesize; - parent = path->nodes[*level]; - root_owner = btrfs_header_owner(parent); - next = btrfs_find_create_tree_block(fs_info, bytenr); if (IS_ERR(next)) return PTR_ERR(next); @@ -2720,18 +2745,16 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, btrfs_clean_tree_block(next); btrfs_wait_tree_block_writeback(next); btrfs_tree_unlock(next); + ret = btrfs_pin_reserved_extent(trans, + bytenr, blocksize); + if (ret) { + free_extent_buffer(next); + return ret; + } } else { if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) clear_extent_buffer_dirty(next); - } - - WARN_ON(root_owner != - BTRFS_TREE_LOG_OBJECTID); - ret = btrfs_pin_reserved_extent(fs_info, - bytenr, blocksize); - if (ret) { - free_extent_buffer(next); - return ret; + unaccount_log_buffer(fs_info, bytenr); } } free_extent_buffer(next); @@ -2762,7 +2785,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, struct walk_control *wc) { struct btrfs_fs_info *fs_info = root->fs_info; - u64 root_owner; int i; int slot; int ret; @@ -2775,13 +2797,6 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, WARN_ON(*level == 0); return 0; } else { - struct extent_buffer *parent; - if (path->nodes[*level] == root->node) - parent = path->nodes[*level]; - else - parent = path->nodes[*level + 1]; - - root_owner = btrfs_header_owner(parent); ret = wc->process_func(root, path->nodes[*level], wc, btrfs_header_generation(path->nodes[*level]), *level); @@ -2799,17 +2814,18 @@ static noinline int walk_up_log_tree(struct btrfs_trans_handle *trans, btrfs_clean_tree_block(next); btrfs_wait_tree_block_writeback(next); btrfs_tree_unlock(next); + ret = btrfs_pin_reserved_extent(trans, + path->nodes[*level]->start, + path->nodes[*level]->len); + if (ret) + return ret; } else { if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) clear_extent_buffer_dirty(next); - } - WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID); - ret = btrfs_pin_reserved_extent(fs_info, - path->nodes[*level]->start, - path->nodes[*level]->len); - if (ret) - return ret; + unaccount_log_buffer(fs_info, + path->nodes[*level]->start); + } } free_extent_buffer(path->nodes[*level]); path->nodes[*level] = NULL; @@ -2880,15 +2896,15 @@ static int walk_log_tree(struct btrfs_trans_handle *trans, btrfs_clean_tree_block(next); btrfs_wait_tree_block_writeback(next); btrfs_tree_unlock(next); + ret = btrfs_pin_reserved_extent(trans, + next->start, next->len); + if (ret) + goto out; } else { if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) clear_extent_buffer_dirty(next); + unaccount_log_buffer(fs_info, next->start); } - - ret = btrfs_pin_reserved_extent(fs_info, next->start, - next->len); - if (ret) - goto out; } } @@ -3283,8 +3299,7 @@ static void free_log_tree(struct btrfs_trans_handle *trans, clear_extent_bits(&log->dirty_log_pages, 0, (u64)-1, EXTENT_DIRTY | EXTENT_NEW | EXTENT_NEED_WAIT); - free_extent_buffer(log->node); - kfree(log); + btrfs_put_root(log); } /* @@ -4518,13 +4533,15 @@ static int btrfs_log_all_xattrs(struct btrfs_trans_handle *trans, static int btrfs_log_holes(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_inode *inode, - struct btrfs_path *path) + struct btrfs_path *path, + const u64 start, + const u64 end) { struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_key key; const u64 ino = btrfs_ino(inode); const u64 i_size = i_size_read(&inode->vfs_inode); - u64 prev_extent_end = 0; + u64 prev_extent_end = start; int ret; if (!btrfs_fs_incompat(fs_info, NO_HOLES) || i_size == 0) @@ -4532,16 +4549,21 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans, key.objectid = ino; key.type = BTRFS_EXTENT_DATA_KEY; - key.offset = 0; + key.offset = start; ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); if (ret < 0) return ret; + if (ret > 0 && path->slots[0] > 0) { + btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0] - 1); + if (key.objectid == ino && key.type == BTRFS_EXTENT_DATA_KEY) + path->slots[0]--; + } + while (true) { - struct btrfs_file_extent_item *extent; struct extent_buffer *leaf = path->nodes[0]; - u64 len; + u64 extent_end; if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) { ret = btrfs_next_leaf(root, path); @@ -4558,9 +4580,18 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans, if (key.objectid != ino || key.type != BTRFS_EXTENT_DATA_KEY) break; + extent_end = btrfs_file_extent_end(path); + if (extent_end <= start) + goto next_slot; + /* We have a hole, log it. */ if (prev_extent_end < key.offset) { - const u64 hole_len = key.offset - prev_extent_end; + u64 hole_len; + + if (key.offset >= end) + hole_len = end - prev_extent_end; + else + hole_len = key.offset - prev_extent_end; /* * Release the path to avoid deadlocks with other code @@ -4590,27 +4621,20 @@ static int btrfs_log_holes(struct btrfs_trans_handle *trans, leaf = path->nodes[0]; } - extent = btrfs_item_ptr(leaf, path->slots[0], - struct btrfs_file_extent_item); - if (btrfs_file_extent_type(leaf, extent) == - BTRFS_FILE_EXTENT_INLINE) { - len = btrfs_file_extent_ram_bytes(leaf, extent); - prev_extent_end = ALIGN(key.offset + len, - fs_info->sectorsize); - } else { - len = btrfs_file_extent_num_bytes(leaf, extent); - prev_extent_end = key.offset + len; - } - + prev_extent_end = min(extent_end, end); + if (extent_end >= end) + break; +next_slot: path->slots[0]++; cond_resched(); } - if (prev_extent_end < i_size) { + if (prev_extent_end < end && prev_extent_end < i_size) { u64 hole_len; btrfs_release_path(path); - hole_len = ALIGN(i_size - prev_extent_end, fs_info->sectorsize); + hole_len = min(ALIGN(i_size, fs_info->sectorsize), end); + hole_len -= prev_extent_end; ret = btrfs_insert_file_extent(trans, root->log_root, ino, prev_extent_end, 0, 0, hole_len, 0, hole_len, @@ -4938,6 +4962,178 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans, return ret; } +static int copy_inode_items_to_log(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode, + struct btrfs_key *min_key, + const struct btrfs_key *max_key, + struct btrfs_path *path, + struct btrfs_path *dst_path, + const u64 logged_isize, + const bool recursive_logging, + const int inode_only, + const u64 start, + const u64 end, + struct btrfs_log_ctx *ctx, + bool *need_log_inode_item) +{ + struct btrfs_root *root = inode->root; + int ins_start_slot = 0; + int ins_nr = 0; + int ret; + + /* + * We must make sure we don't copy extent items that are entirely out of + * the range [start, end - 1]. This is not just an optimization to avoid + * copying but also needed to avoid a corruption where we end up with + * file extent items in the log tree that have overlapping ranges - this + * can happen if we race with ordered extent completion for ranges that + * are outside our target range. For example we copy an extent item and + * when we move to the next leaf, that extent was trimmed and a new one + * covering a subrange of it, but with a higher key, was inserted - we + * would then copy this other extent too, resulting in a log tree with + * 2 extent items that represent overlapping ranges. + * + * We can copy the entire extents at the range bondaries however, even + * if they cover an area outside the target range. That's ok. + */ + while (1) { + ret = btrfs_search_forward(root, min_key, path, trans->transid); + if (ret < 0) + return ret; + if (ret > 0) { + ret = 0; + break; + } +again: + /* Note, ins_nr might be > 0 here, cleanup outside the loop */ + if (min_key->objectid != max_key->objectid) + break; + if (min_key->type > max_key->type) + break; + + if (min_key->type == BTRFS_INODE_ITEM_KEY) + *need_log_inode_item = false; + + if ((min_key->type == BTRFS_INODE_REF_KEY || + min_key->type == BTRFS_INODE_EXTREF_KEY) && + inode->generation == trans->transid && + !recursive_logging) { + u64 other_ino = 0; + u64 other_parent = 0; + + ret = btrfs_check_ref_name_override(path->nodes[0], + path->slots[0], min_key, inode, + &other_ino, &other_parent); + if (ret < 0) { + return ret; + } else if (ret > 0 && ctx && + other_ino != btrfs_ino(BTRFS_I(ctx->inode))) { + if (ins_nr > 0) { + ins_nr++; + } else { + ins_nr = 1; + ins_start_slot = path->slots[0]; + } + ret = copy_items(trans, inode, dst_path, path, + ins_start_slot, ins_nr, + inode_only, logged_isize); + if (ret < 0) + return ret; + ins_nr = 0; + + ret = log_conflicting_inodes(trans, root, path, + ctx, other_ino, other_parent); + if (ret) + return ret; + btrfs_release_path(path); + goto next_key; + } + } + + /* Skip xattrs, we log them later with btrfs_log_all_xattrs() */ + if (min_key->type == BTRFS_XATTR_ITEM_KEY) { + if (ins_nr == 0) + goto next_slot; + ret = copy_items(trans, inode, dst_path, path, + ins_start_slot, + ins_nr, inode_only, logged_isize); + if (ret < 0) + return ret; + ins_nr = 0; + goto next_slot; + } + + if (min_key->type == BTRFS_EXTENT_DATA_KEY) { + const u64 extent_end = btrfs_file_extent_end(path); + + if (extent_end <= start) { + if (ins_nr > 0) { + ret = copy_items(trans, inode, dst_path, + path, ins_start_slot, + ins_nr, inode_only, + logged_isize); + if (ret < 0) + return ret; + ins_nr = 0; + } + goto next_slot; + } + if (extent_end >= end) { + ins_nr++; + if (ins_nr == 1) + ins_start_slot = path->slots[0]; + break; + } + } + + if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) { + ins_nr++; + goto next_slot; + } else if (!ins_nr) { + ins_start_slot = path->slots[0]; + ins_nr = 1; + goto next_slot; + } + + ret = copy_items(trans, inode, dst_path, path, ins_start_slot, + ins_nr, inode_only, logged_isize); + if (ret < 0) + return ret; + ins_nr = 1; + ins_start_slot = path->slots[0]; +next_slot: + path->slots[0]++; + if (path->slots[0] < btrfs_header_nritems(path->nodes[0])) { + btrfs_item_key_to_cpu(path->nodes[0], min_key, + path->slots[0]); + goto again; + } + if (ins_nr) { + ret = copy_items(trans, inode, dst_path, path, + ins_start_slot, ins_nr, inode_only, + logged_isize); + if (ret < 0) + return ret; + ins_nr = 0; + } + btrfs_release_path(path); +next_key: + if (min_key->offset < (u64)-1) { + min_key->offset++; + } else if (min_key->type < max_key->type) { + min_key->type++; + min_key->offset = 0; + } else { + break; + } + } + if (ins_nr) + ret = copy_items(trans, inode, dst_path, path, ins_start_slot, + ins_nr, inode_only, logged_isize); + + return ret; +} + /* log a single inode in the tree log. * At least one parent directory for this inode must exist in the tree * or be logged already. @@ -4955,8 +5151,8 @@ static int log_conflicting_inodes(struct btrfs_trans_handle *trans, static int btrfs_log_inode(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct btrfs_inode *inode, int inode_only, - const loff_t start, - const loff_t end, + u64 start, + u64 end, struct btrfs_log_ctx *ctx) { struct btrfs_fs_info *fs_info = root->fs_info; @@ -4967,9 +5163,6 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, struct btrfs_root *log = root->log_root; int err = 0; int ret; - int nritems; - int ins_start_slot = 0; - int ins_nr; bool fast_search = false; u64 ino = btrfs_ino(inode); struct extent_map_tree *em_tree = &inode->extent_tree; @@ -4987,6 +5180,9 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, return -ENOMEM; } + start = ALIGN_DOWN(start, fs_info->sectorsize); + end = ALIGN(end, fs_info->sectorsize); + min_key.objectid = ino; min_key.type = BTRFS_INODE_ITEM_KEY; min_key.offset = 0; @@ -5100,139 +5296,12 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, goto out_unlock; } - while (1) { - ins_nr = 0; - ret = btrfs_search_forward(root, &min_key, - path, trans->transid); - if (ret < 0) { - err = ret; - goto out_unlock; - } - if (ret != 0) - break; -again: - /* note, ins_nr might be > 0 here, cleanup outside the loop */ - if (min_key.objectid != ino) - break; - if (min_key.type > max_key.type) - break; - - if (min_key.type == BTRFS_INODE_ITEM_KEY) - need_log_inode_item = false; - - if ((min_key.type == BTRFS_INODE_REF_KEY || - min_key.type == BTRFS_INODE_EXTREF_KEY) && - inode->generation == trans->transid && - !recursive_logging) { - u64 other_ino = 0; - u64 other_parent = 0; - - ret = btrfs_check_ref_name_override(path->nodes[0], - path->slots[0], &min_key, inode, - &other_ino, &other_parent); - if (ret < 0) { - err = ret; - goto out_unlock; - } else if (ret > 0 && ctx && - other_ino != btrfs_ino(BTRFS_I(ctx->inode))) { - if (ins_nr > 0) { - ins_nr++; - } else { - ins_nr = 1; - ins_start_slot = path->slots[0]; - } - ret = copy_items(trans, inode, dst_path, path, - ins_start_slot, - ins_nr, inode_only, - logged_isize); - if (ret < 0) { - err = ret; - goto out_unlock; - } - ins_nr = 0; - - err = log_conflicting_inodes(trans, root, path, - ctx, other_ino, other_parent); - if (err) - goto out_unlock; - btrfs_release_path(path); - goto next_key; - } - } - - /* Skip xattrs, we log them later with btrfs_log_all_xattrs() */ - if (min_key.type == BTRFS_XATTR_ITEM_KEY) { - if (ins_nr == 0) - goto next_slot; - ret = copy_items(trans, inode, dst_path, path, - ins_start_slot, - ins_nr, inode_only, logged_isize); - if (ret < 0) { - err = ret; - goto out_unlock; - } - ins_nr = 0; - goto next_slot; - } - - if (ins_nr && ins_start_slot + ins_nr == path->slots[0]) { - ins_nr++; - goto next_slot; - } else if (!ins_nr) { - ins_start_slot = path->slots[0]; - ins_nr = 1; - goto next_slot; - } - - ret = copy_items(trans, inode, dst_path, path, - ins_start_slot, ins_nr, inode_only, - logged_isize); - if (ret < 0) { - err = ret; - goto out_unlock; - } - ins_nr = 1; - ins_start_slot = path->slots[0]; -next_slot: - - nritems = btrfs_header_nritems(path->nodes[0]); - path->slots[0]++; - if (path->slots[0] < nritems) { - btrfs_item_key_to_cpu(path->nodes[0], &min_key, - path->slots[0]); - goto again; - } - if (ins_nr) { - ret = copy_items(trans, inode, dst_path, path, - ins_start_slot, - ins_nr, inode_only, logged_isize); - if (ret < 0) { - err = ret; - goto out_unlock; - } - ins_nr = 0; - } - btrfs_release_path(path); -next_key: - if (min_key.offset < (u64)-1) { - min_key.offset++; - } else if (min_key.type < max_key.type) { - min_key.type++; - min_key.offset = 0; - } else { - break; - } - } - if (ins_nr) { - ret = copy_items(trans, inode, dst_path, path, - ins_start_slot, ins_nr, inode_only, - logged_isize); - if (ret < 0) { - err = ret; - goto out_unlock; - } - ins_nr = 0; - } + err = copy_inode_items_to_log(trans, inode, &min_key, &max_key, + path, dst_path, logged_isize, + recursive_logging, inode_only, + start, end, ctx, &need_log_inode_item); + if (err) + goto out_unlock; btrfs_release_path(path); btrfs_release_path(dst_path); @@ -5243,7 +5312,7 @@ next_key: if (max_key.type >= BTRFS_EXTENT_DATA_KEY && !fast_search) { btrfs_release_path(path); btrfs_release_path(dst_path); - err = btrfs_log_holes(trans, root, inode, path); + err = btrfs_log_holes(trans, root, inode, path, start, end); if (err) goto out_unlock; } @@ -6145,7 +6214,7 @@ again: if (found_key.objectid != BTRFS_TREE_LOG_OBJECTID) break; - log = btrfs_read_fs_root(log_root_tree, &found_key); + log = btrfs_read_tree_root(log_root_tree, &found_key); if (IS_ERR(log)) { ret = PTR_ERR(log); btrfs_handle_fs_error(fs_info, ret, @@ -6157,7 +6226,7 @@ again: tmp_key.type = BTRFS_ROOT_ITEM_KEY; tmp_key.offset = (u64)-1; - wc.replay_dest = btrfs_read_fs_root_no_name(fs_info, &tmp_key); + wc.replay_dest = btrfs_get_fs_root(fs_info, &tmp_key, true); if (IS_ERR(wc.replay_dest)) { ret = PTR_ERR(wc.replay_dest); @@ -6173,12 +6242,10 @@ again: * each subsequent pass. */ if (ret == -ENOENT) - ret = btrfs_pin_extent_for_log_replay(fs_info, + ret = btrfs_pin_extent_for_log_replay(trans, log->node->start, log->node->len); - free_extent_buffer(log->node); - free_extent_buffer(log->commit_root); - kfree(log); + btrfs_put_root(log); if (!ret) goto next; @@ -6214,9 +6281,8 @@ again: } wc.replay_dest->log_root = NULL; - free_extent_buffer(log->node); - free_extent_buffer(log->commit_root); - kfree(log); + btrfs_put_root(wc.replay_dest); + btrfs_put_root(log); if (ret) goto error; @@ -6247,10 +6313,9 @@ next: if (ret) return ret; - free_extent_buffer(log_root_tree->node); log_root_tree->log_root = NULL; clear_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags); - kfree(log_root_tree); + btrfs_put_root(log_root_tree); return 0; error: diff --git a/fs/btrfs/uuid-tree.c b/fs/btrfs/uuid-tree.c index 76b84f2397b1..76671a6bcb61 100644 --- a/fs/btrfs/uuid-tree.c +++ b/fs/btrfs/uuid-tree.c @@ -246,9 +246,53 @@ out: return ret; } -int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info, - int (*check_func)(struct btrfs_fs_info *, u8 *, u8, - u64)) +/* + * Check if there's an matching subvolume for given UUID + * + * Return: + * 0 check succeeded, the entry is not outdated + * > 0 if the check failed, the caller should remove the entry + * < 0 if an error occurred + */ +static int btrfs_check_uuid_tree_entry(struct btrfs_fs_info *fs_info, + u8 *uuid, u8 type, u64 subvolid) +{ + struct btrfs_key key; + int ret = 0; + struct btrfs_root *subvol_root; + + if (type != BTRFS_UUID_KEY_SUBVOL && + type != BTRFS_UUID_KEY_RECEIVED_SUBVOL) + goto out; + + key.objectid = subvolid; + key.type = BTRFS_ROOT_ITEM_KEY; + key.offset = (u64)-1; + subvol_root = btrfs_get_fs_root(fs_info, &key, true); + if (IS_ERR(subvol_root)) { + ret = PTR_ERR(subvol_root); + if (ret == -ENOENT) + ret = 1; + goto out; + } + + switch (type) { + case BTRFS_UUID_KEY_SUBVOL: + if (memcmp(uuid, subvol_root->root_item.uuid, BTRFS_UUID_SIZE)) + ret = 1; + break; + case BTRFS_UUID_KEY_RECEIVED_SUBVOL: + if (memcmp(uuid, subvol_root->root_item.received_uuid, + BTRFS_UUID_SIZE)) + ret = 1; + break; + } + btrfs_put_root(subvol_root); +out: + return ret; +} + +int btrfs_uuid_tree_iterate(struct btrfs_fs_info *fs_info) { struct btrfs_root *root = fs_info->uuid_root; struct btrfs_key key; @@ -278,6 +322,10 @@ again_search_slot: } while (1) { + if (btrfs_fs_closing(fs_info)) { + ret = -EINTR; + goto out; + } cond_resched(); leaf = path->nodes[0]; slot = path->slots[0]; @@ -305,7 +353,8 @@ again_search_slot: read_extent_buffer(leaf, &subid_le, offset, sizeof(subid_le)); subid_cpu = le64_to_cpu(subid_le); - ret = check_func(fs_info, uuid, key.type, subid_cpu); + ret = btrfs_check_uuid_tree_entry(fs_info, uuid, + key.type, subid_cpu); if (ret < 0) goto out; if (ret > 0) { diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 9cfc668f91f4..c1909e5f4506 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -6,7 +6,6 @@ #include <linux/sched.h> #include <linux/bio.h> #include <linux/slab.h> -#include <linux/buffer_head.h> #include <linux/blkdev.h> #include <linux/ratelimit.h> #include <linux/kthread.h> @@ -500,7 +499,7 @@ static struct btrfs_fs_devices *find_fsid_with_metadata_uuid( static int btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder, int flush, struct block_device **bdev, - struct buffer_head **bh) + struct btrfs_super_block **disk_super) { int ret; @@ -519,9 +518,9 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder, goto error; } invalidate_bdev(*bdev); - *bh = btrfs_read_dev_super(*bdev); - if (IS_ERR(*bh)) { - ret = PTR_ERR(*bh); + *disk_super = btrfs_read_dev_super(*bdev); + if (IS_ERR(*disk_super)) { + ret = PTR_ERR(*disk_super); blkdev_put(*bdev, flags); goto error; } @@ -530,7 +529,6 @@ btrfs_get_bdev_and_sb(const char *device_path, fmode_t flags, void *holder, error: *bdev = NULL; - *bh = NULL; return ret; } @@ -611,7 +609,6 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices, { struct request_queue *q; struct block_device *bdev; - struct buffer_head *bh; struct btrfs_super_block *disk_super; u64 devid; int ret; @@ -622,17 +619,16 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices, return -EINVAL; ret = btrfs_get_bdev_and_sb(device->name->str, flags, holder, 1, - &bdev, &bh); + &bdev, &disk_super); if (ret) return ret; - disk_super = (struct btrfs_super_block *)bh->b_data; devid = btrfs_stack_device_id(&disk_super->dev_item); if (devid != device->devid) - goto error_brelse; + goto error_free_page; if (memcmp(device->uuid, disk_super->dev_item.uuid, BTRFS_UUID_SIZE)) - goto error_brelse; + goto error_free_page; device->generation = btrfs_super_generation(disk_super); @@ -641,7 +637,7 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices, BTRFS_FEATURE_INCOMPAT_METADATA_UUID) { pr_err( "BTRFS: Invalid seeding and uuid-changed device detected\n"); - goto error_brelse; + goto error_free_page; } clear_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state); @@ -667,12 +663,12 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices, fs_devices->rw_devices++; list_add_tail(&device->dev_alloc_list, &fs_devices->alloc_list); } - brelse(bh); + btrfs_release_disk_super(disk_super); return 0; -error_brelse: - brelse(bh); +error_free_page: + btrfs_release_disk_super(disk_super); blkdev_put(bdev, flags); return -EINVAL; @@ -1209,6 +1205,7 @@ static int open_fs_devices(struct btrfs_fs_devices *fs_devices, fs_devices->opened = 1; fs_devices->latest_bdev = latest_dev->bdev; fs_devices->total_rw_bytes = 0; + fs_devices->chunk_alloc_policy = BTRFS_CHUNK_ALLOC_REGULAR; out: return ret; } @@ -1247,9 +1244,10 @@ int btrfs_open_devices(struct btrfs_fs_devices *fs_devices, return ret; } -static void btrfs_release_disk_super(struct page *page) +void btrfs_release_disk_super(struct btrfs_super_block *super) { - kunmap(page); + struct page *page = virt_to_page(super); + put_page(page); } @@ -1277,17 +1275,17 @@ static int btrfs_read_disk_super(struct block_device *bdev, u64 bytenr, *page = read_cache_page_gfp(bdev->bd_inode->i_mapping, index, GFP_KERNEL); - if (IS_ERR_OR_NULL(*page)) + if (IS_ERR(*page)) return 1; - p = kmap(*page); + p = page_address(*page); /* align our pointer to the offset of the super block */ *disk_super = p + offset_in_page(bytenr); if (btrfs_super_bytenr(*disk_super) != bytenr || btrfs_super_magic(*disk_super) != BTRFS_MAGIC) { - btrfs_release_disk_super(*page); + btrfs_release_disk_super(p); return 1; } @@ -1350,7 +1348,7 @@ struct btrfs_device *btrfs_scan_one_device(const char *path, fmode_t flags, btrfs_free_stale_devices(path, device); } - btrfs_release_disk_super(page); + btrfs_release_disk_super(disk_super); error_bdev_put: blkdev_put(bdev, flags); @@ -1383,6 +1381,59 @@ static bool contains_pending_extent(struct btrfs_device *device, u64 *start, return false; } +static u64 dev_extent_search_start(struct btrfs_device *device, u64 start) +{ + switch (device->fs_devices->chunk_alloc_policy) { + case BTRFS_CHUNK_ALLOC_REGULAR: + /* + * We don't want to overwrite the superblock on the drive nor + * any area used by the boot loader (grub for example), so we + * make sure to start at an offset of at least 1MB. + */ + return max_t(u64, start, SZ_1M); + default: + BUG(); + } +} + +/** + * dev_extent_hole_check - check if specified hole is suitable for allocation + * @device: the device which we have the hole + * @hole_start: starting position of the hole + * @hole_size: the size of the hole + * @num_bytes: the size of the free space that we need + * + * This function may modify @hole_start and @hole_end to reflect the suitable + * position for allocation. Returns 1 if hole position is updated, 0 otherwise. + */ +static bool dev_extent_hole_check(struct btrfs_device *device, u64 *hole_start, + u64 *hole_size, u64 num_bytes) +{ + bool changed = false; + u64 hole_end = *hole_start + *hole_size; + + /* + * Check before we set max_hole_start, otherwise we could end up + * sending back this offset anyway. + */ + if (contains_pending_extent(device, hole_start, *hole_size)) { + if (hole_end >= *hole_start) + *hole_size = hole_end - *hole_start; + else + *hole_size = 0; + changed = true; + } + + switch (device->fs_devices->chunk_alloc_policy) { + case BTRFS_CHUNK_ALLOC_REGULAR: + /* No extra check */ + break; + default: + BUG(); + } + + return changed; +} /* * find_free_dev_extent_start - find free space in the specified device @@ -1429,12 +1480,7 @@ static int find_free_dev_extent_start(struct btrfs_device *device, int slot; struct extent_buffer *l; - /* - * We don't want to overwrite the superblock on the drive nor any area - * used by the boot loader (grub for example), so we make sure to start - * at an offset of at least 1MB. - */ - search_start = max_t(u64, search_start, SZ_1M); + search_start = dev_extent_search_start(device, search_start); path = btrfs_alloc_path(); if (!path) @@ -1492,18 +1538,8 @@ again: if (key.offset > search_start) { hole_size = key.offset - search_start; - - /* - * Have to check before we set max_hole_start, otherwise - * we could end up sending back this offset anyway. - */ - if (contains_pending_extent(device, &search_start, - hole_size)) { - if (key.offset >= search_start) - hole_size = key.offset - search_start; - else - hole_size = 0; - } + dev_extent_hole_check(device, &search_start, &hole_size, + num_bytes); if (hole_size > max_hole_size) { max_hole_start = search_start; @@ -1542,8 +1578,8 @@ next: */ if (search_end > search_start) { hole_size = search_end - search_start; - - if (contains_pending_extent(device, &search_start, hole_size)) { + if (dev_extent_hole_check(device, &search_start, &hole_size, + num_bytes)) { btrfs_release_path(path); goto again; } @@ -1949,6 +1985,46 @@ static u64 btrfs_num_devices(struct btrfs_fs_info *fs_info) return num_devices; } +static void btrfs_scratch_superblocks(struct btrfs_fs_info *fs_info, + struct block_device *bdev, + const char *device_path) +{ + struct btrfs_super_block *disk_super; + int copy_num; + + if (!bdev) + return; + + for (copy_num = 0; copy_num < BTRFS_SUPER_MIRROR_MAX; copy_num++) { + struct page *page; + int ret; + + disk_super = btrfs_read_dev_one_super(bdev, copy_num); + if (IS_ERR(disk_super)) + continue; + + memset(&disk_super->magic, 0, sizeof(disk_super->magic)); + + page = virt_to_page(disk_super); + set_page_dirty(page); + lock_page(page); + /* write_on_page() unlocks the page */ + ret = write_one_page(page); + if (ret) + btrfs_warn(fs_info, + "error clearing superblock number %d (%d)", + copy_num, ret); + btrfs_release_disk_super(disk_super); + + } + + /* Notify udev that device has changed */ + btrfs_kobject_uevent(bdev, KOBJ_CHANGE); + + /* Update ctime/mtime for device path for libblkid */ + update_dev_time(device_path); +} + int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, u64 devid) { @@ -2054,7 +2130,7 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, if (device->bdev) { cur_devices->open_devices--; /* remove sysfs entry */ - btrfs_sysfs_rm_device_link(fs_devices, device); + btrfs_sysfs_remove_devices_dir(fs_devices, device); } num_devices = btrfs_super_num_devices(fs_info->super_copy) - 1; @@ -2067,7 +2143,8 @@ int btrfs_rm_device(struct btrfs_fs_info *fs_info, const char *device_path, * supers and free the device. */ if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) - btrfs_scratch_superblocks(device->bdev, device->name->str); + btrfs_scratch_superblocks(fs_info, device->bdev, + device->name->str); btrfs_close_bdev(device); synchronize_rcu(); @@ -2135,7 +2212,8 @@ void btrfs_rm_dev_replace_free_srcdev(struct btrfs_device *srcdev) if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &srcdev->dev_state)) { /* zero out the old super if it is writable */ - btrfs_scratch_superblocks(srcdev->bdev, srcdev->name->str); + btrfs_scratch_superblocks(fs_info, srcdev->bdev, + srcdev->name->str); } btrfs_close_bdev(srcdev); @@ -2174,7 +2252,7 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev) mutex_lock(&fs_devices->device_list_mutex); - btrfs_sysfs_rm_device_link(fs_devices, tgtdev); + btrfs_sysfs_remove_devices_dir(fs_devices, tgtdev); if (tgtdev->bdev) fs_devices->open_devices--; @@ -2194,7 +2272,8 @@ void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev) * is already out of device list, so we don't have to hold * the device_list_mutex lock. */ - btrfs_scratch_superblocks(tgtdev->bdev, tgtdev->name->str); + btrfs_scratch_superblocks(tgtdev->fs_info, tgtdev->bdev, + tgtdev->name->str); btrfs_close_bdev(tgtdev); synchronize_rcu(); @@ -2209,14 +2288,13 @@ static struct btrfs_device *btrfs_find_device_by_path( u64 devid; u8 *dev_uuid; struct block_device *bdev; - struct buffer_head *bh; struct btrfs_device *device; ret = btrfs_get_bdev_and_sb(device_path, FMODE_READ, - fs_info->bdev_holder, 0, &bdev, &bh); + fs_info->bdev_holder, 0, &bdev, &disk_super); if (ret) return ERR_PTR(ret); - disk_super = (struct btrfs_super_block *)bh->b_data; + devid = btrfs_stack_device_id(&disk_super->dev_item); dev_uuid = disk_super->dev_item.uuid; if (btrfs_fs_incompat(fs_info, METADATA_UUID)) @@ -2226,7 +2304,7 @@ static struct btrfs_device *btrfs_find_device_by_path( device = btrfs_find_device(fs_info->fs_devices, devid, dev_uuid, disk_super->fsid, true); - brelse(bh); + btrfs_release_disk_super(disk_super); if (!device) device = ERR_PTR(-ENOENT); blkdev_put(bdev, FMODE_READ); @@ -2522,7 +2600,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path orig_super_num_devices + 1); /* add sysfs device entry */ - btrfs_sysfs_add_device_link(fs_devices, device); + btrfs_sysfs_add_devices_dir(fs_devices, device); /* * we've got more storage, clear any full flags on the space @@ -2590,7 +2668,7 @@ int btrfs_init_new_device(struct btrfs_fs_info *fs_info, const char *device_path return ret; error_sysfs: - btrfs_sysfs_rm_device_link(fs_devices, device); + btrfs_sysfs_remove_devices_dir(fs_devices, device); mutex_lock(&fs_info->fs_devices->device_list_mutex); mutex_lock(&fs_info->chunk_mutex); list_del_rcu(&device->dev_list); @@ -3723,13 +3801,25 @@ static inline int balance_need_close(struct btrfs_fs_info *fs_info) atomic_read(&fs_info->balance_cancel_req) == 0); } -/* Non-zero return value signifies invalidity */ -static inline int validate_convert_profile(struct btrfs_balance_args *bctl_arg, - u64 allowed) +/* + * Validate target profile against allowed profiles and return true if it's OK. + * Otherwise print the error message and return false. + */ +static inline int validate_convert_profile(struct btrfs_fs_info *fs_info, + const struct btrfs_balance_args *bargs, + u64 allowed, const char *type) { - return ((bctl_arg->flags & BTRFS_BALANCE_ARGS_CONVERT) && - (!alloc_profile_is_valid(bctl_arg->target, 1) || - (bctl_arg->target & ~allowed))); + if (!(bargs->flags & BTRFS_BALANCE_ARGS_CONVERT)) + return true; + + /* Profile is valid and does not have bits outside of the allowed set */ + if (alloc_profile_is_valid(bargs->target, 1) && + (bargs->target & ~allowed) == 0) + return true; + + btrfs_err(fs_info, "balance: invalid convert %s profile %s", + type, btrfs_bg_type_to_raid_name(bargs->target)); + return false; } /* @@ -3904,7 +3994,7 @@ int btrfs_balance(struct btrfs_fs_info *fs_info, if (btrfs_fs_closing(fs_info) || atomic_read(&fs_info->balance_pause_req) || - atomic_read(&fs_info->balance_cancel_req)) { + btrfs_should_cancel_balance(fs_info)) { ret = -EINVAL; goto out; } @@ -3945,24 +4035,9 @@ int btrfs_balance(struct btrfs_fs_info *fs_info, if (num_devices >= btrfs_raid_array[i].devs_min) allowed |= btrfs_raid_array[i].bg_flag; - if (validate_convert_profile(&bctl->data, allowed)) { - btrfs_err(fs_info, - "balance: invalid convert data profile %s", - btrfs_bg_type_to_raid_name(bctl->data.target)); - ret = -EINVAL; - goto out; - } - if (validate_convert_profile(&bctl->meta, allowed)) { - btrfs_err(fs_info, - "balance: invalid convert metadata profile %s", - btrfs_bg_type_to_raid_name(bctl->meta.target)); - ret = -EINVAL; - goto out; - } - if (validate_convert_profile(&bctl->sys, allowed)) { - btrfs_err(fs_info, - "balance: invalid convert system profile %s", - btrfs_bg_type_to_raid_name(bctl->sys.target)); + if (!validate_convert_profile(fs_info, &bctl->data, allowed, "data") || + !validate_convert_profile(fs_info, &bctl->meta, allowed, "metadata") || + !validate_convert_profile(fs_info, &bctl->sys, allowed, "system")) { ret = -EINVAL; goto out; } @@ -4274,7 +4349,7 @@ int btrfs_cancel_balance(struct btrfs_fs_info *fs_info) return 0; } -static int btrfs_uuid_scan_kthread(void *data) +int btrfs_uuid_scan_kthread(void *data) { struct btrfs_fs_info *fs_info = data; struct btrfs_root *root = fs_info->tree_root; @@ -4286,6 +4361,7 @@ static int btrfs_uuid_scan_kthread(void *data) struct btrfs_root_item root_item; u32 item_size; struct btrfs_trans_handle *trans = NULL; + bool closing = false; path = btrfs_alloc_path(); if (!path) { @@ -4298,6 +4374,10 @@ static int btrfs_uuid_scan_kthread(void *data) key.offset = 0; while (1) { + if (btrfs_fs_closing(fs_info)) { + closing = true; + break; + } ret = btrfs_search_forward(root, &key, path, BTRFS_OLDEST_GENERATION); if (ret) { @@ -4397,76 +4477,12 @@ out: btrfs_end_transaction(trans); if (ret) btrfs_warn(fs_info, "btrfs_uuid_scan_kthread failed %d", ret); - else + else if (!closing) set_bit(BTRFS_FS_UPDATE_UUID_TREE_GEN, &fs_info->flags); up(&fs_info->uuid_tree_rescan_sem); return 0; } -/* - * Callback for btrfs_uuid_tree_iterate(). - * returns: - * 0 check succeeded, the entry is not outdated. - * < 0 if an error occurred. - * > 0 if the check failed, which means the caller shall remove the entry. - */ -static int btrfs_check_uuid_tree_entry(struct btrfs_fs_info *fs_info, - u8 *uuid, u8 type, u64 subid) -{ - struct btrfs_key key; - int ret = 0; - struct btrfs_root *subvol_root; - - if (type != BTRFS_UUID_KEY_SUBVOL && - type != BTRFS_UUID_KEY_RECEIVED_SUBVOL) - goto out; - - key.objectid = subid; - key.type = BTRFS_ROOT_ITEM_KEY; - key.offset = (u64)-1; - subvol_root = btrfs_read_fs_root_no_name(fs_info, &key); - if (IS_ERR(subvol_root)) { - ret = PTR_ERR(subvol_root); - if (ret == -ENOENT) - ret = 1; - goto out; - } - - switch (type) { - case BTRFS_UUID_KEY_SUBVOL: - if (memcmp(uuid, subvol_root->root_item.uuid, BTRFS_UUID_SIZE)) - ret = 1; - break; - case BTRFS_UUID_KEY_RECEIVED_SUBVOL: - if (memcmp(uuid, subvol_root->root_item.received_uuid, - BTRFS_UUID_SIZE)) - ret = 1; - break; - } - -out: - return ret; -} - -static int btrfs_uuid_rescan_kthread(void *data) -{ - struct btrfs_fs_info *fs_info = (struct btrfs_fs_info *)data; - int ret; - - /* - * 1st step is to iterate through the existing UUID tree and - * to delete all entries that contain outdated data. - * 2nd step is to add all missing entries to the UUID tree. - */ - ret = btrfs_uuid_tree_iterate(fs_info, btrfs_check_uuid_tree_entry); - if (ret < 0) { - btrfs_warn(fs_info, "iterating uuid_tree failed %d", ret); - up(&fs_info->uuid_tree_rescan_sem); - return ret; - } - return btrfs_uuid_scan_kthread(data); -} - int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info) { struct btrfs_trans_handle *trans; @@ -4509,22 +4525,6 @@ int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info) return 0; } -int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info) -{ - struct task_struct *task; - - down(&fs_info->uuid_tree_rescan_sem); - task = kthread_run(btrfs_uuid_rescan_kthread, fs_info, "btrfs-uuid"); - if (IS_ERR(task)) { - /* fs_info->update_uuid_tree_gen remains 0 in all error case */ - btrfs_warn(fs_info, "failed to start uuid_rescan task"); - up(&fs_info->uuid_tree_rescan_sem); - return PTR_ERR(task); - } - - return 0; -} - /* * shrinking a device means finding all of the device extents past * the new size, and then following the back refs to the chunks. @@ -4777,96 +4777,111 @@ static void check_raid1c34_incompat_flag(struct btrfs_fs_info *info, u64 type) btrfs_set_fs_incompat(info, RAID1C34); } -static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans, - u64 start, u64 type) -{ - struct btrfs_fs_info *info = trans->fs_info; - struct btrfs_fs_devices *fs_devices = info->fs_devices; - struct btrfs_device *device; - struct map_lookup *map = NULL; - struct extent_map_tree *em_tree; - struct extent_map *em; - struct btrfs_device_info *devices_info = NULL; - u64 total_avail; - int num_stripes; /* total number of stripes to allocate */ - int data_stripes; /* number of stripes that count for - block group size */ - int sub_stripes; /* sub_stripes info for map */ - int dev_stripes; /* stripes per dev */ - int devs_max; /* max devs to use */ - int devs_min; /* min devs needed */ - int devs_increment; /* ndevs has to be a multiple of this */ - int ncopies; /* how many copies to data has */ - int nparity; /* number of stripes worth of bytes to - store parity information */ - int ret; +/* + * Structure used internally for __btrfs_alloc_chunk() function. + * Wraps needed parameters. + */ +struct alloc_chunk_ctl { + u64 start; + u64 type; + /* Total number of stripes to allocate */ + int num_stripes; + /* sub_stripes info for map */ + int sub_stripes; + /* Stripes per device */ + int dev_stripes; + /* Maximum number of devices to use */ + int devs_max; + /* Minimum number of devices to use */ + int devs_min; + /* ndevs has to be a multiple of this */ + int devs_increment; + /* Number of copies */ + int ncopies; + /* Number of stripes worth of bytes to store parity information */ + int nparity; u64 max_stripe_size; u64 max_chunk_size; + u64 dev_extent_min; u64 stripe_size; u64 chunk_size; int ndevs; - int i; - int j; - int index; - - BUG_ON(!alloc_profile_is_valid(type, 0)); - - if (list_empty(&fs_devices->alloc_list)) { - if (btrfs_test_opt(info, ENOSPC_DEBUG)) - btrfs_debug(info, "%s: no writable device", __func__); - return -ENOSPC; - } - - index = btrfs_bg_flags_to_raid_index(type); +}; - sub_stripes = btrfs_raid_array[index].sub_stripes; - dev_stripes = btrfs_raid_array[index].dev_stripes; - devs_max = btrfs_raid_array[index].devs_max; - if (!devs_max) - devs_max = BTRFS_MAX_DEVS(info); - devs_min = btrfs_raid_array[index].devs_min; - devs_increment = btrfs_raid_array[index].devs_increment; - ncopies = btrfs_raid_array[index].ncopies; - nparity = btrfs_raid_array[index].nparity; +static void init_alloc_chunk_ctl_policy_regular( + struct btrfs_fs_devices *fs_devices, + struct alloc_chunk_ctl *ctl) +{ + u64 type = ctl->type; if (type & BTRFS_BLOCK_GROUP_DATA) { - max_stripe_size = SZ_1G; - max_chunk_size = BTRFS_MAX_DATA_CHUNK_SIZE; + ctl->max_stripe_size = SZ_1G; + ctl->max_chunk_size = BTRFS_MAX_DATA_CHUNK_SIZE; } else if (type & BTRFS_BLOCK_GROUP_METADATA) { - /* for larger filesystems, use larger metadata chunks */ + /* For larger filesystems, use larger metadata chunks */ if (fs_devices->total_rw_bytes > 50ULL * SZ_1G) - max_stripe_size = SZ_1G; + ctl->max_stripe_size = SZ_1G; else - max_stripe_size = SZ_256M; - max_chunk_size = max_stripe_size; + ctl->max_stripe_size = SZ_256M; + ctl->max_chunk_size = ctl->max_stripe_size; } else if (type & BTRFS_BLOCK_GROUP_SYSTEM) { - max_stripe_size = SZ_32M; - max_chunk_size = 2 * max_stripe_size; - devs_max = min_t(int, devs_max, BTRFS_MAX_DEVS_SYS_CHUNK); + ctl->max_stripe_size = SZ_32M; + ctl->max_chunk_size = 2 * ctl->max_stripe_size; + ctl->devs_max = min_t(int, ctl->devs_max, + BTRFS_MAX_DEVS_SYS_CHUNK); } else { - btrfs_err(info, "invalid chunk type 0x%llx requested", - type); BUG(); } /* We don't want a chunk larger than 10% of writable space */ - max_chunk_size = min(div_factor(fs_devices->total_rw_bytes, 1), - max_chunk_size); + ctl->max_chunk_size = min(div_factor(fs_devices->total_rw_bytes, 1), + ctl->max_chunk_size); + ctl->dev_extent_min = BTRFS_STRIPE_LEN * ctl->dev_stripes; +} + +static void init_alloc_chunk_ctl(struct btrfs_fs_devices *fs_devices, + struct alloc_chunk_ctl *ctl) +{ + int index = btrfs_bg_flags_to_raid_index(ctl->type); + + ctl->sub_stripes = btrfs_raid_array[index].sub_stripes; + ctl->dev_stripes = btrfs_raid_array[index].dev_stripes; + ctl->devs_max = btrfs_raid_array[index].devs_max; + if (!ctl->devs_max) + ctl->devs_max = BTRFS_MAX_DEVS(fs_devices->fs_info); + ctl->devs_min = btrfs_raid_array[index].devs_min; + ctl->devs_increment = btrfs_raid_array[index].devs_increment; + ctl->ncopies = btrfs_raid_array[index].ncopies; + ctl->nparity = btrfs_raid_array[index].nparity; + ctl->ndevs = 0; + + switch (fs_devices->chunk_alloc_policy) { + case BTRFS_CHUNK_ALLOC_REGULAR: + init_alloc_chunk_ctl_policy_regular(fs_devices, ctl); + break; + default: + BUG(); + } +} - devices_info = kcalloc(fs_devices->rw_devices, sizeof(*devices_info), - GFP_NOFS); - if (!devices_info) - return -ENOMEM; +static int gather_device_info(struct btrfs_fs_devices *fs_devices, + struct alloc_chunk_ctl *ctl, + struct btrfs_device_info *devices_info) +{ + struct btrfs_fs_info *info = fs_devices->fs_info; + struct btrfs_device *device; + u64 total_avail; + u64 dev_extent_want = ctl->max_stripe_size * ctl->dev_stripes; + int ret; + int ndevs = 0; + u64 max_avail; + u64 dev_offset; /* * in the first pass through the devices list, we gather information * about the available holes on each device. */ - ndevs = 0; list_for_each_entry(device, &fs_devices->alloc_list, dev_alloc_list) { - u64 max_avail; - u64 dev_offset; - if (!test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state)) { WARN(1, KERN_ERR "BTRFS: read-only device in alloc_list\n"); @@ -4884,24 +4899,23 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans, total_avail = 0; /* If there is no space on this device, skip it. */ - if (total_avail == 0) + if (total_avail < ctl->dev_extent_min) continue; - ret = find_free_dev_extent(device, - max_stripe_size * dev_stripes, - &dev_offset, &max_avail); + ret = find_free_dev_extent(device, dev_extent_want, &dev_offset, + &max_avail); if (ret && ret != -ENOSPC) - goto error; + return ret; if (ret == 0) - max_avail = max_stripe_size * dev_stripes; + max_avail = dev_extent_want; - if (max_avail < BTRFS_STRIPE_LEN * dev_stripes) { + if (max_avail < ctl->dev_extent_min) { if (btrfs_test_opt(info, ENOSPC_DEBUG)) btrfs_debug(info, - "%s: devid %llu has no free space, have=%llu want=%u", + "%s: devid %llu has no free space, have=%llu want=%llu", __func__, device->devid, max_avail, - BTRFS_STRIPE_LEN * dev_stripes); + ctl->dev_extent_min); continue; } @@ -4916,6 +4930,7 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans, devices_info[ndevs].dev = device; ++ndevs; } + ctl->ndevs = ndevs; /* * now sort the devices by hole size / available space @@ -4923,23 +4938,14 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans, sort(devices_info, ndevs, sizeof(struct btrfs_device_info), btrfs_cmp_device_info, NULL); - /* - * Round down to number of usable stripes, devs_increment can be any - * number so we can't use round_down() - */ - ndevs -= ndevs % devs_increment; - - if (ndevs < devs_min) { - ret = -ENOSPC; - if (btrfs_test_opt(info, ENOSPC_DEBUG)) { - btrfs_debug(info, - "%s: not enough devices with free space: have=%d minimum required=%d", - __func__, ndevs, devs_min); - } - goto error; - } + return 0; +} - ndevs = min(ndevs, devs_max); +static int decide_stripe_size_regular(struct alloc_chunk_ctl *ctl, + struct btrfs_device_info *devices_info) +{ + /* Number of stripes that count for block group size */ + int data_stripes; /* * The primary goal is to maximize the number of stripes, so use as @@ -4948,73 +4954,116 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans, * The DUP profile stores more than one stripe per device, the * max_avail is the total size so we have to adjust. */ - stripe_size = div_u64(devices_info[ndevs - 1].max_avail, dev_stripes); - num_stripes = ndevs * dev_stripes; + ctl->stripe_size = div_u64(devices_info[ctl->ndevs - 1].max_avail, + ctl->dev_stripes); + ctl->num_stripes = ctl->ndevs * ctl->dev_stripes; - /* - * this will have to be fixed for RAID1 and RAID10 over - * more drives - */ - data_stripes = (num_stripes - nparity) / ncopies; + /* This will have to be fixed for RAID1 and RAID10 over more drives */ + data_stripes = (ctl->num_stripes - ctl->nparity) / ctl->ncopies; /* - * Use the number of data stripes to figure out how big this chunk - * is really going to be in terms of logical address space, - * and compare that answer with the max chunk size. If it's higher, - * we try to reduce stripe_size. + * Use the number of data stripes to figure out how big this chunk is + * really going to be in terms of logical address space, and compare + * that answer with the max chunk size. If it's higher, we try to + * reduce stripe_size. */ - if (stripe_size * data_stripes > max_chunk_size) { + if (ctl->stripe_size * data_stripes > ctl->max_chunk_size) { /* * Reduce stripe_size, round it up to a 16MB boundary again and * then use it, unless it ends up being even bigger than the * previous value we had already. */ - stripe_size = min(round_up(div_u64(max_chunk_size, - data_stripes), SZ_16M), - stripe_size); + ctl->stripe_size = min(round_up(div_u64(ctl->max_chunk_size, + data_stripes), SZ_16M), + ctl->stripe_size); } - /* align to BTRFS_STRIPE_LEN */ - stripe_size = round_down(stripe_size, BTRFS_STRIPE_LEN); + /* Align to BTRFS_STRIPE_LEN */ + ctl->stripe_size = round_down(ctl->stripe_size, BTRFS_STRIPE_LEN); + ctl->chunk_size = ctl->stripe_size * data_stripes; - map = kmalloc(map_lookup_size(num_stripes), GFP_NOFS); - if (!map) { - ret = -ENOMEM; - goto error; + return 0; +} + +static int decide_stripe_size(struct btrfs_fs_devices *fs_devices, + struct alloc_chunk_ctl *ctl, + struct btrfs_device_info *devices_info) +{ + struct btrfs_fs_info *info = fs_devices->fs_info; + + /* + * Round down to number of usable stripes, devs_increment can be any + * number so we can't use round_down() that requires power of 2, while + * rounddown is safe. + */ + ctl->ndevs = rounddown(ctl->ndevs, ctl->devs_increment); + + if (ctl->ndevs < ctl->devs_min) { + if (btrfs_test_opt(info, ENOSPC_DEBUG)) { + btrfs_debug(info, + "%s: not enough devices with free space: have=%d minimum required=%d", + __func__, ctl->ndevs, ctl->devs_min); + } + return -ENOSPC; } - map->num_stripes = num_stripes; - for (i = 0; i < ndevs; ++i) { - for (j = 0; j < dev_stripes; ++j) { - int s = i * dev_stripes + j; + ctl->ndevs = min(ctl->ndevs, ctl->devs_max); + + switch (fs_devices->chunk_alloc_policy) { + case BTRFS_CHUNK_ALLOC_REGULAR: + return decide_stripe_size_regular(ctl, devices_info); + default: + BUG(); + } +} + +static int create_chunk(struct btrfs_trans_handle *trans, + struct alloc_chunk_ctl *ctl, + struct btrfs_device_info *devices_info) +{ + struct btrfs_fs_info *info = trans->fs_info; + struct map_lookup *map = NULL; + struct extent_map_tree *em_tree; + struct extent_map *em; + u64 start = ctl->start; + u64 type = ctl->type; + int ret; + int i; + int j; + + map = kmalloc(map_lookup_size(ctl->num_stripes), GFP_NOFS); + if (!map) + return -ENOMEM; + map->num_stripes = ctl->num_stripes; + + for (i = 0; i < ctl->ndevs; ++i) { + for (j = 0; j < ctl->dev_stripes; ++j) { + int s = i * ctl->dev_stripes + j; map->stripes[s].dev = devices_info[i].dev; map->stripes[s].physical = devices_info[i].dev_offset + - j * stripe_size; + j * ctl->stripe_size; } } map->stripe_len = BTRFS_STRIPE_LEN; map->io_align = BTRFS_STRIPE_LEN; map->io_width = BTRFS_STRIPE_LEN; map->type = type; - map->sub_stripes = sub_stripes; - - chunk_size = stripe_size * data_stripes; + map->sub_stripes = ctl->sub_stripes; - trace_btrfs_chunk_alloc(info, map, start, chunk_size); + trace_btrfs_chunk_alloc(info, map, start, ctl->chunk_size); em = alloc_extent_map(); if (!em) { kfree(map); - ret = -ENOMEM; - goto error; + return -ENOMEM; } set_bit(EXTENT_FLAG_FS_MAPPING, &em->flags); em->map_lookup = map; em->start = start; - em->len = chunk_size; + em->len = ctl->chunk_size; em->block_start = 0; em->block_len = em->len; - em->orig_block_len = stripe_size; + em->orig_block_len = ctl->stripe_size; em_tree = &info->mapping_tree; write_lock(&em_tree->lock); @@ -5022,30 +5071,31 @@ static int __btrfs_alloc_chunk(struct btrfs_trans_handle *trans, if (ret) { write_unlock(&em_tree->lock); free_extent_map(em); - goto error; + return ret; } write_unlock(&em_tree->lock); - ret = btrfs_make_block_group(trans, 0, type, start, chunk_size); + ret = btrfs_make_block_group(trans, 0, type, start, ctl->chunk_size); if (ret) goto error_del_extent; for (i = 0; i < map->num_stripes; i++) { struct btrfs_device *dev = map->stripes[i].dev; - btrfs_device_set_bytes_used(dev, dev->bytes_used + stripe_size); + btrfs_device_set_bytes_used(dev, + dev->bytes_used + ctl->stripe_size); if (list_empty(&dev->post_commit_list)) list_add_tail(&dev->post_commit_list, &trans->transaction->dev_update_list); } - atomic64_sub(stripe_size * map->num_stripes, &info->free_chunk_space); + atomic64_sub(ctl->stripe_size * map->num_stripes, + &info->free_chunk_space); free_extent_map(em); check_raid56_incompat_flag(info, type); check_raid1c34_incompat_flag(info, type); - kfree(devices_info); return 0; error_del_extent: @@ -5057,11 +5107,68 @@ error_del_extent: free_extent_map(em); /* One for the tree reference */ free_extent_map(em); -error: + + return ret; +} + +int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type) +{ + struct btrfs_fs_info *info = trans->fs_info; + struct btrfs_fs_devices *fs_devices = info->fs_devices; + struct btrfs_device_info *devices_info = NULL; + struct alloc_chunk_ctl ctl; + int ret; + + lockdep_assert_held(&info->chunk_mutex); + + if (!alloc_profile_is_valid(type, 0)) { + ASSERT(0); + return -EINVAL; + } + + if (list_empty(&fs_devices->alloc_list)) { + if (btrfs_test_opt(info, ENOSPC_DEBUG)) + btrfs_debug(info, "%s: no writable device", __func__); + return -ENOSPC; + } + + if (!(type & BTRFS_BLOCK_GROUP_TYPE_MASK)) { + btrfs_err(info, "invalid chunk type 0x%llx requested", type); + ASSERT(0); + return -EINVAL; + } + + ctl.start = find_next_chunk(info); + ctl.type = type; + init_alloc_chunk_ctl(fs_devices, &ctl); + + devices_info = kcalloc(fs_devices->rw_devices, sizeof(*devices_info), + GFP_NOFS); + if (!devices_info) + return -ENOMEM; + + ret = gather_device_info(fs_devices, &ctl, devices_info); + if (ret < 0) + goto out; + + ret = decide_stripe_size(fs_devices, &ctl, devices_info); + if (ret < 0) + goto out; + + ret = create_chunk(trans, &ctl, devices_info); + +out: kfree(devices_info); return ret; } +/* + * Chunk allocation falls into two parts. The first part does work + * that makes the new allocated chunk usable, but does not do any operation + * that modifies the chunk tree. The second part does the work that + * requires modifying the chunk tree. This division is important for the + * bootstrap process of adding storage to a seed btrfs. + */ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans, u64 chunk_offset, u64 chunk_size) { @@ -5160,39 +5267,19 @@ out: return ret; } -/* - * Chunk allocation falls into two parts. The first part does work - * that makes the new allocated chunk usable, but does not do any operation - * that modifies the chunk tree. The second part does the work that - * requires modifying the chunk tree. This division is important for the - * bootstrap process of adding storage to a seed btrfs. - */ -int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type) -{ - u64 chunk_offset; - - lockdep_assert_held(&trans->fs_info->chunk_mutex); - chunk_offset = find_next_chunk(trans->fs_info); - return __btrfs_alloc_chunk(trans, chunk_offset, type); -} - static noinline int init_first_rw_device(struct btrfs_trans_handle *trans) { struct btrfs_fs_info *fs_info = trans->fs_info; - u64 chunk_offset; - u64 sys_chunk_offset; u64 alloc_profile; int ret; - chunk_offset = find_next_chunk(fs_info); alloc_profile = btrfs_metadata_alloc_profile(fs_info); - ret = __btrfs_alloc_chunk(trans, chunk_offset, alloc_profile); + ret = btrfs_alloc_chunk(trans, alloc_profile); if (ret) return ret; - sys_chunk_offset = find_next_chunk(fs_info); alloc_profile = btrfs_system_alloc_profile(fs_info); - ret = __btrfs_alloc_chunk(trans, sys_chunk_offset, alloc_profile); + ret = btrfs_alloc_chunk(trans, alloc_profile); return ret; } @@ -5389,31 +5476,19 @@ static int find_live_mirror(struct btrfs_fs_info *fs_info, return preferred_mirror; } -static inline int parity_smaller(u64 a, u64 b) -{ - return a > b; -} - /* Bubble-sort the stripe set to put the parity/syndrome stripes last */ static void sort_parity_stripes(struct btrfs_bio *bbio, int num_stripes) { - struct btrfs_bio_stripe s; int i; - u64 l; int again = 1; while (again) { again = 0; for (i = 0; i < num_stripes - 1; i++) { - if (parity_smaller(bbio->raid_map[i], - bbio->raid_map[i+1])) { - s = bbio->stripes[i]; - l = bbio->raid_map[i]; - bbio->stripes[i] = bbio->stripes[i+1]; - bbio->raid_map[i] = bbio->raid_map[i+1]; - bbio->stripes[i+1] = s; - bbio->raid_map[i+1] = l; - + /* Swap if parity is on a smaller index */ + if (bbio->raid_map[i] > bbio->raid_map[i + 1]) { + swap(bbio->stripes[i], bbio->stripes[i + 1]); + swap(bbio->raid_map[i], bbio->raid_map[i + 1]); again = 1; } } @@ -5914,10 +5989,7 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, struct btrfs_io_geometry geom; ASSERT(bbio_ret); - - if (op == BTRFS_MAP_DISCARD) - return __btrfs_map_block_for_discard(fs_info, logical, - length, bbio_ret); + ASSERT(op != BTRFS_MAP_DISCARD); ret = btrfs_get_io_geometry(fs_info, op, logical, *length, &geom); if (ret < 0) @@ -6147,6 +6219,10 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, u64 logical, u64 *length, struct btrfs_bio **bbio_ret, int mirror_num) { + if (op == BTRFS_MAP_DISCARD) + return __btrfs_map_block_for_discard(fs_info, logical, + length, bbio_ret); + return __btrfs_map_block(fs_info, op, logical, length, bbio_ret, mirror_num, 0); } @@ -6241,8 +6317,8 @@ static void submit_stripe_bio(struct btrfs_bio *bbio, struct bio *bio, btrfs_debug_in_rcu(fs_info, "btrfs_map_bio: rw %d 0x%x, sector=%llu, dev=%lu (%s id %llu), size=%u", bio_op(bio), bio->bi_opf, (u64)bio->bi_iter.bi_sector, - (u_long)dev->bdev->bd_dev, rcu_str_deref(dev->name), dev->devid, - bio->bi_iter.bi_size); + (unsigned long)dev->bdev->bd_dev, rcu_str_deref(dev->name), + dev->devid, bio->bi_iter.bi_size); bio_set_dev(bio, dev->bdev); btrfs_bio_counter_inc_noblocked(fs_info); @@ -7317,36 +7393,6 @@ int btrfs_get_dev_stats(struct btrfs_fs_info *fs_info, return 0; } -void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path) -{ - struct buffer_head *bh; - struct btrfs_super_block *disk_super; - int copy_num; - - if (!bdev) - return; - - for (copy_num = 0; copy_num < BTRFS_SUPER_MIRROR_MAX; - copy_num++) { - - if (btrfs_read_dev_one_super(bdev, copy_num, &bh)) - continue; - - disk_super = (struct btrfs_super_block *)bh->b_data; - - memset(&disk_super->magic, 0, sizeof(disk_super->magic)); - set_buffer_dirty(bh); - sync_dirty_buffer(bh); - brelse(bh); - } - - /* Notify udev that device has changed */ - btrfs_kobject_uevent(bdev, KOBJ_CHANGE); - - /* Update ctime/mtime for device path for libblkid */ - update_dev_time(device_path); -} - /* * Update the size and bytes used for each device where it changed. This is * delayed since we would otherwise get errors while writing out the diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index f01552a0785e..f067b5934c46 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -17,8 +17,6 @@ extern struct mutex uuid_mutex; #define BTRFS_STRIPE_LEN SZ_64K -struct buffer_head; - struct btrfs_io_geometry { /* remaining bytes before crossing a stripe */ u64 len; @@ -209,6 +207,10 @@ BTRFS_DEVICE_GETSET_FUNCS(total_bytes); BTRFS_DEVICE_GETSET_FUNCS(disk_total_bytes); BTRFS_DEVICE_GETSET_FUNCS(bytes_used); +enum btrfs_chunk_allocation_policy { + BTRFS_CHUNK_ALLOC_REGULAR, +}; + struct btrfs_fs_devices { u8 fsid[BTRFS_FSID_SIZE]; /* FS specific uuid */ u8 metadata_uuid[BTRFS_FSID_SIZE]; @@ -260,6 +262,8 @@ struct btrfs_fs_devices { struct kobject *devices_kobj; struct kobject *devinfo_kobj; struct completion kobj_unregister; + + enum btrfs_chunk_allocation_policy chunk_alloc_policy; }; #define BTRFS_BIO_INLINE_CSUM_SIZE 64 @@ -461,7 +465,7 @@ int btrfs_recover_balance(struct btrfs_fs_info *fs_info); int btrfs_pause_balance(struct btrfs_fs_info *fs_info); int btrfs_cancel_balance(struct btrfs_fs_info *fs_info); int btrfs_create_uuid_tree(struct btrfs_fs_info *fs_info); -int btrfs_check_uuid_tree(struct btrfs_fs_info *fs_info); +int btrfs_uuid_scan_kthread(void *data); int btrfs_chunk_readonly(struct btrfs_fs_info *fs_info, u64 chunk_offset); int find_free_dev_extent(struct btrfs_device *device, u64 num_bytes, u64 *start, u64 *max_avail); @@ -474,7 +478,6 @@ int btrfs_run_dev_stats(struct btrfs_trans_handle *trans); void btrfs_rm_dev_replace_remove_srcdev(struct btrfs_device *srcdev); void btrfs_rm_dev_replace_free_srcdev(struct btrfs_device *srcdev); void btrfs_destroy_dev_replace_tgtdev(struct btrfs_device *tgtdev); -void btrfs_scratch_superblocks(struct block_device *bdev, const char *device_path); int btrfs_is_parity_mirror(struct btrfs_fs_info *fs_info, u64 logical, u64 len); unsigned long btrfs_full_stripe_len(struct btrfs_fs_info *fs_info, @@ -484,6 +487,7 @@ int btrfs_finish_chunk_alloc(struct btrfs_trans_handle *trans, int btrfs_remove_chunk(struct btrfs_trans_handle *trans, u64 chunk_offset); struct extent_map *btrfs_get_chunk_map(struct btrfs_fs_info *fs_info, u64 logical, u64 length); +void btrfs_release_disk_super(struct btrfs_super_block *super); static inline void btrfs_dev_stat_inc(struct btrfs_device *dev, int index) diff --git a/fs/buffer.c b/fs/buffer.c index b8d28370cfd7..f73276d746bb 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -274,8 +274,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate) * decide that the page is now completely done. */ first = page_buffers(page); - local_irq_save(flags); - bit_spin_lock(BH_Uptodate_Lock, &first->b_state); + spin_lock_irqsave(&first->b_uptodate_lock, flags); clear_buffer_async_read(bh); unlock_buffer(bh); tmp = bh; @@ -288,8 +287,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate) } tmp = tmp->b_this_page; } while (tmp != bh); - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); - local_irq_restore(flags); + spin_unlock_irqrestore(&first->b_uptodate_lock, flags); /* * If none of the buffers had errors and they are all @@ -301,8 +299,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate) return; still_busy: - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); - local_irq_restore(flags); + spin_unlock_irqrestore(&first->b_uptodate_lock, flags); return; } @@ -371,8 +368,7 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate) } first = page_buffers(page); - local_irq_save(flags); - bit_spin_lock(BH_Uptodate_Lock, &first->b_state); + spin_lock_irqsave(&first->b_uptodate_lock, flags); clear_buffer_async_write(bh); unlock_buffer(bh); @@ -384,14 +380,12 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate) } tmp = tmp->b_this_page; } - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); - local_irq_restore(flags); + spin_unlock_irqrestore(&first->b_uptodate_lock, flags); end_page_writeback(page); return; still_busy: - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); - local_irq_restore(flags); + spin_unlock_irqrestore(&first->b_uptodate_lock, flags); return; } EXPORT_SYMBOL(end_buffer_async_write); @@ -3019,49 +3013,6 @@ static void end_bio_bh_io_sync(struct bio *bio) bio_put(bio); } -/* - * This allows us to do IO even on the odd last sectors - * of a device, even if the block size is some multiple - * of the physical sector size. - * - * We'll just truncate the bio to the size of the device, - * and clear the end of the buffer head manually. - * - * Truly out-of-range accesses will turn into actual IO - * errors, this only handles the "we need to be able to - * do IO at the final sector" case. - */ -void guard_bio_eod(struct bio *bio) -{ - sector_t maxsector; - struct hd_struct *part; - - rcu_read_lock(); - part = __disk_get_part(bio->bi_disk, bio->bi_partno); - if (part) - maxsector = part_nr_sects_read(part); - else - maxsector = get_capacity(bio->bi_disk); - rcu_read_unlock(); - - if (!maxsector) - return; - - /* - * If the *whole* IO is past the end of the device, - * let it through, and the IO layer will turn it into - * an EIO. - */ - if (unlikely(bio->bi_iter.bi_sector >= maxsector)) - return; - - maxsector -= bio->bi_iter.bi_sector; - if (likely((bio->bi_iter.bi_size >> 9) <= maxsector)) - return; - - bio_truncate(bio, maxsector << 9); -} - static int submit_bh_wbc(int op, int op_flags, struct buffer_head *bh, enum rw_hint write_hint, struct writeback_control *wbc) { @@ -3385,6 +3336,7 @@ struct buffer_head *alloc_buffer_head(gfp_t gfp_flags) struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags); if (ret) { INIT_LIST_HEAD(&ret->b_assoc_buffers); + spin_lock_init(&ret->b_uptodate_lock); preempt_disable(); __this_cpu_inc(bh_accounting.nr); recalc_bh_state(); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 7e0190b1f821..5a478cd06e11 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -1415,10 +1415,13 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) struct inode *inode = file_inode(file); struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); + struct ceph_osd_client *osdc = &fsc->client->osdc; struct ceph_cap_flush *prealloc_cf; ssize_t count, written = 0; int err, want, got; bool direct_lock = false; + u32 map_flags; + u64 pool_flags; loff_t pos; loff_t limit = max(i_size_read(inode), fsc->max_file_size); @@ -1481,8 +1484,12 @@ retry_snap: goto out; } - /* FIXME: not complete since it doesn't account for being at quota */ - if (ceph_osdmap_flag(&fsc->client->osdc, CEPH_OSDMAP_FULL)) { + down_read(&osdc->lock); + map_flags = osdc->osdmap->flags; + pool_flags = ceph_pg_pool_flags(osdc->osdmap, ci->i_layout.pool_id); + up_read(&osdc->lock); + if ((map_flags & CEPH_OSDMAP_FULL) || + (pool_flags & CEPH_POOL_FLAG_FULL)) { err = -ENOSPC; goto out; } @@ -1575,7 +1582,8 @@ retry_snap: } if (written >= 0) { - if (ceph_osdmap_flag(&fsc->client->osdc, CEPH_OSDMAP_NEARFULL)) + if ((map_flags & CEPH_OSDMAP_NEARFULL) || + (pool_flags & CEPH_POOL_FLAG_NEARFULL)) iocb->ki_flags |= IOCB_DSYNC; written = generic_write_sync(iocb, written); } diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index ccfcc66aaf44..923be9399b21 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -1155,5 +1155,6 @@ void ceph_cleanup_snapid_map(struct ceph_mds_client *mdsc) pr_err("snapid map %llx -> %x still in use\n", sm->snap, sm->dev); } + kfree(sm); } } diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index 606f26d862dc..cc3ada12848d 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -324,6 +324,8 @@ static struct vfsmount *cifs_dfs_do_automount(struct dentry *mntpt) if (full_path == NULL) goto cdda_exit; + convert_delimiter(full_path, '\\'); + cifs_dbg(FYI, "%s: full_path: %s\n", __func__, full_path); if (!cifs_sb_master_tlink(cifs_sb)) { diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 716574aab3b6..ae421634aa42 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -342,7 +342,7 @@ static int sid_to_id(struct cifs_sb_info *cifs_sb, struct cifs_sid *psid, struct cifs_fattr *fattr, uint sidtype) { - int rc; + int rc = 0; struct key *sidkey; char *sidstr; const struct cred *saved_cred; @@ -450,11 +450,12 @@ out_revert_creds: * fails then we just fall back to using the mnt_uid/mnt_gid. */ got_valid_id: + rc = 0; if (sidtype == SIDOWNER) fattr->cf_uid = fuid; else fattr->cf_gid = fgid; - return 0; + return rc; } int diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 46ebaf3f0824..94e3ed4850b5 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -530,6 +530,8 @@ cifs_show_options(struct seq_file *s, struct dentry *root) if (tcon->seal) seq_puts(s, ",seal"); + else if (tcon->ses->server->ignore_signature) + seq_puts(s, ",signloosely"); if (tcon->nocase) seq_puts(s, ",nocase"); if (tcon->local_lease) @@ -1016,7 +1018,7 @@ struct file_system_type cifs_fs_type = { .name = "cifs", .mount = cifs_do_mount, .kill_sb = cifs_kill_sb, - /* .fs_flags */ + .fs_flags = FS_RENAME_DOES_D_MOVE, }; MODULE_ALIAS_FS("cifs"); @@ -1025,7 +1027,7 @@ static struct file_system_type smb3_fs_type = { .name = "smb3", .mount = smb3_do_mount, .kill_sb = cifs_kill_sb, - /* .fs_flags */ + .fs_flags = FS_RENAME_DOES_D_MOVE, }; MODULE_ALIAS_FS("smb3"); MODULE_ALIAS("smb3"); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index b87456bae1a1..c9e2e6bbca13 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -156,5 +156,5 @@ extern int cifs_truncate_page(struct address_space *mapping, loff_t from); extern const struct export_operations cifs_export_ops; #endif /* CONFIG_CIFS_NFSD_EXPORT */ -#define CIFS_VERSION "2.25" +#define CIFS_VERSION "2.26" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index de82cfa44b1a..0d956360e984 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -1281,6 +1281,7 @@ struct cifs_fid { __u64 volatile_fid; /* volatile file id for smb2 */ __u8 lease_key[SMB2_LEASE_KEY_SIZE]; /* lease key for smb2 */ __u8 create_guid[16]; + __u32 access; struct cifs_pending_open *pending_open; unsigned int epoch; #ifdef CONFIG_CIFS_DEBUG2 @@ -1741,6 +1742,12 @@ static inline bool is_retryable_error(int error) return false; } + +/* cifs_get_writable_file() flags */ +#define FIND_WR_ANY 0 +#define FIND_WR_FSUID_ONLY 1 +#define FIND_WR_WITH_DELETE 2 + #define MID_FREE 0 #define MID_REQUEST_ALLOCATED 1 #define MID_REQUEST_SUBMITTED 2 diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 79d842e7240c..593d826820c3 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -1021,7 +1021,7 @@ typedef struct smb_com_writex_req { __le16 ByteCount; __u8 Pad; /* BB check for whether padded to DWORD boundary and optimum performance here */ - char Data[0]; + char Data[]; } __attribute__((packed)) WRITEX_REQ; typedef struct smb_com_write_req { @@ -1041,7 +1041,7 @@ typedef struct smb_com_write_req { __le16 ByteCount; __u8 Pad; /* BB check for whether padded to DWORD boundary and optimum performance here */ - char Data[0]; + char Data[]; } __attribute__((packed)) WRITE_REQ; typedef struct smb_com_write_rsp { @@ -1306,7 +1306,7 @@ typedef struct smb_com_ntransact_req { /* SetupCount words follow then */ __le16 ByteCount; __u8 Pad[3]; - __u8 Parms[0]; + __u8 Parms[]; } __attribute__((packed)) NTRANSACT_REQ; typedef struct smb_com_ntransact_rsp { @@ -1523,7 +1523,7 @@ struct file_notify_information { __le32 NextEntryOffset; __le32 Action; __le32 FileNameLength; - __u8 FileName[0]; + __u8 FileName[]; } __attribute__((packed)); /* For IO_REPARSE_TAG_SYMLINK */ @@ -1536,7 +1536,7 @@ struct reparse_symlink_data { __le16 PrintNameOffset; __le16 PrintNameLength; __le32 Flags; - char PathBuffer[0]; + char PathBuffer[]; } __attribute__((packed)); /* Flag above */ @@ -1553,7 +1553,7 @@ struct reparse_posix_data { __le16 ReparseDataLength; __u16 Reserved; __le64 InodeType; /* LNK, FIFO, CHR etc. */ - char PathBuffer[0]; + char PathBuffer[]; } __attribute__((packed)); struct cifs_quota_data { @@ -1691,6 +1691,7 @@ struct smb_t2_rsp { #define SMB_FIND_FILE_ID_FULL_DIR_INFO 0x105 #define SMB_FIND_FILE_ID_BOTH_DIR_INFO 0x106 #define SMB_FIND_FILE_UNIX 0x202 +#define SMB_FIND_FILE_POSIX_INFO 0x064 typedef struct smb_com_transaction2_qpi_req { struct smb_hdr hdr; /* wct = 14+ */ @@ -1761,7 +1762,7 @@ struct set_file_rename { __le32 overwrite; /* 1 = overwrite dest */ __u32 root_fid; /* zero */ __le32 target_name_len; - char target_name[0]; /* Must be unicode */ + char target_name[]; /* Must be unicode */ } __attribute__((packed)); struct smb_com_transaction2_sfi_req { @@ -2450,7 +2451,7 @@ struct cifs_posix_acl { /* access conrol list (ACL) */ __le16 version; __le16 access_entry_count; /* access ACL - count of entries */ __le16 default_entry_count; /* default ACL - count of entries */ - struct cifs_posix_ace ace_array[0]; + struct cifs_posix_ace ace_array[]; /* followed by struct cifs_posix_ace default_ace_arraay[] */ } __attribute__((packed)); /* level 0x204 */ @@ -2756,7 +2757,7 @@ typedef struct file_xattr_info { /* BB do we need another field for flags? BB */ __u32 xattr_name_len; __u32 xattr_value_len; - char xattr_name[0]; + char xattr_name[]; /* followed by xattr_value[xattr_value_len], no pad */ } __attribute__((packed)) FILE_XATTR_INFO; /* extended attribute info level 0x205 */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 89eaaf46d1ca..12a895e02db4 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -134,11 +134,12 @@ extern bool backup_cred(struct cifs_sb_info *); extern bool is_size_safe_to_change(struct cifsInodeInfo *, __u64 eof); extern void cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, unsigned int bytes_written); -extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, bool); +extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, int); extern int cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, - bool fsuid_only, + int flags, struct cifsFileInfo **ret_file); extern int cifs_get_writable_path(struct cifs_tcon *tcon, const char *name, + int flags, struct cifsFileInfo **ret_file); extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *, bool); extern int cifs_get_readable_path(struct cifs_tcon *tcon, const char *name, @@ -601,6 +602,11 @@ int smb2_parse_query_directory(struct cifs_tcon *tcon, struct kvec *rsp_iov, int resp_buftype, struct cifs_search_info *srch_inf); +struct super_block *cifs_get_tcp_super(struct TCP_Server_Info *server); +void cifs_put_tcp_super(struct super_block *sb); +int update_super_prepath(struct cifs_tcon *tcon, const char *prefix, + size_t prefix_len); + #ifdef CONFIG_CIFS_DFS_UPCALL static inline int get_dfs_path(const unsigned int xid, struct cifs_ses *ses, const char *old_path, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 3c89569e7210..140efc1a9374 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -162,9 +162,18 @@ static int __cifs_reconnect_tcon(const struct nls_table *nlsc, for (it = dfs_cache_get_tgt_iterator(&tl); it; it = dfs_cache_get_next_tgt(&tl, it)) { - const char *tgt = dfs_cache_get_tgt_name(it); + const char *share, *prefix; + size_t share_len, prefix_len; - extract_unc_hostname(tgt, &dfs_host, &dfs_host_len); + rc = dfs_cache_get_tgt_share(it, &share, &share_len, &prefix, + &prefix_len); + if (rc) { + cifs_dbg(VFS, "%s: failed to parse target share %d\n", + __func__, rc); + continue; + } + + extract_unc_hostname(share, &dfs_host, &dfs_host_len); if (dfs_host_len != tcp_host_len || strncasecmp(dfs_host, tcp_host, dfs_host_len) != 0) { @@ -175,11 +184,13 @@ static int __cifs_reconnect_tcon(const struct nls_table *nlsc, continue; } - scnprintf(tree, MAX_TREE_SIZE, "\\%s", tgt); + scnprintf(tree, MAX_TREE_SIZE, "\\%.*s", (int)share_len, share); rc = CIFSTCon(0, tcon->ses, tree, tcon, nlsc); - if (!rc) + if (!rc) { + rc = update_super_prepath(tcon, prefix, prefix_len); break; + } if (rc == -EREMOTE) break; } @@ -320,7 +331,7 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) atomic_inc(&tconInfoReconnectCount); /* tell server Unix caps we support */ - if (ses->capabilities & CAP_UNIX) + if (cap_unix(ses)) reset_cifs_unix_caps(0, tcon, NULL, NULL); /* @@ -1492,6 +1503,7 @@ openRetry: *oplock = rsp->OplockLevel; /* cifs fid stays in le */ oparms->fid->netfid = rsp->Fid; + oparms->fid->access = desired_access; /* Let caller know file was created so we can set the mode. */ /* Do we care about the CreateAction in any other cases? */ @@ -1590,7 +1602,6 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) if (server->ops->is_session_expired && server->ops->is_session_expired(buf)) { cifs_reconnect(server); - wake_up(&server->response_q); return -1; } @@ -2115,7 +2126,7 @@ cifs_writev_requeue(struct cifs_writedata *wdata) wdata2->tailsz = tailsz; wdata2->bytes = cur_len; - rc = cifs_get_writable_file(CIFS_I(inode), false, + rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY, &wdata2->cfile); if (!wdata2->cfile) { cifs_dbg(VFS, "No writable handle to retry writepages rc=%d\n", diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 4804d1df8c1c..95b3ab0ca8c0 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -21,6 +21,7 @@ #include <linux/fs.h> #include <linux/net.h> #include <linux/string.h> +#include <linux/sched/mm.h> #include <linux/sched/signal.h> #include <linux/list.h> #include <linux/wait.h> @@ -57,7 +58,6 @@ #include "smb2proto.h" #include "smbdirect.h" #include "dns_resolve.h" -#include "cifsfs.h" #ifdef CONFIG_CIFS_DFS_UPCALL #include "dfs_cache.h" #endif @@ -389,54 +389,7 @@ static inline int reconn_set_ipaddr(struct TCP_Server_Info *server) #endif #ifdef CONFIG_CIFS_DFS_UPCALL -struct super_cb_data { - struct TCP_Server_Info *server; - struct super_block *sb; -}; - /* These functions must be called with server->srv_mutex held */ - -static void super_cb(struct super_block *sb, void *arg) -{ - struct super_cb_data *d = arg; - struct cifs_sb_info *cifs_sb; - struct cifs_tcon *tcon; - - if (d->sb) - return; - - cifs_sb = CIFS_SB(sb); - tcon = cifs_sb_master_tcon(cifs_sb); - if (tcon->ses->server == d->server) - d->sb = sb; -} - -static struct super_block *get_tcp_super(struct TCP_Server_Info *server) -{ - struct super_cb_data d = { - .server = server, - .sb = NULL, - }; - - iterate_supers_type(&cifs_fs_type, super_cb, &d); - - if (unlikely(!d.sb)) - return ERR_PTR(-ENOENT); - /* - * Grab an active reference in order to prevent automounts (DFS links) - * of expiring and then freeing up our cifs superblock pointer while - * we're doing failover. - */ - cifs_sb_active(d.sb); - return d.sb; -} - -static inline void put_tcp_super(struct super_block *sb) -{ - if (!IS_ERR_OR_NULL(sb)) - cifs_sb_deactive(sb); -} - static void reconn_inval_dfs_target(struct TCP_Server_Info *server, struct cifs_sb_info *cifs_sb, struct dfs_cache_tgt_list *tgt_list, @@ -508,7 +461,7 @@ cifs_reconnect(struct TCP_Server_Info *server) server->nr_targets = 1; #ifdef CONFIG_CIFS_DFS_UPCALL spin_unlock(&GlobalMid_Lock); - sb = get_tcp_super(server); + sb = cifs_get_tcp_super(server); if (IS_ERR(sb)) { rc = PTR_ERR(sb); cifs_dbg(FYI, "%s: will not do DFS failover: rc = %d\n", @@ -535,8 +488,9 @@ cifs_reconnect(struct TCP_Server_Info *server) spin_unlock(&GlobalMid_Lock); #ifdef CONFIG_CIFS_DFS_UPCALL dfs_cache_free_tgts(&tgt_list); - put_tcp_super(sb); + cifs_put_tcp_super(sb); #endif + wake_up(&server->response_q); return rc; } else server->tcpStatus = CifsNeedReconnect; @@ -666,11 +620,12 @@ cifs_reconnect(struct TCP_Server_Info *server) } - put_tcp_super(sb); + cifs_put_tcp_super(sb); #endif if (server->tcpStatus == CifsNeedNegotiate) mod_delayed_work(cifsiod_wq, &server->echo, 0); + wake_up(&server->response_q); return rc; } @@ -765,7 +720,6 @@ server_unresponsive(struct TCP_Server_Info *server) cifs_server_dbg(VFS, "has not responded in %lu seconds. Reconnecting...\n", (3 * server->echo_interval) / HZ); cifs_reconnect(server); - wake_up(&server->response_q); return true; } @@ -898,7 +852,6 @@ is_smb_response(struct TCP_Server_Info *server, unsigned char type) */ cifs_set_port((struct sockaddr *)&server->dstaddr, CIFS_PORT); cifs_reconnect(server); - wake_up(&server->response_q); break; default: cifs_server_dbg(VFS, "RFC 1002 unknown response type 0x%x\n", type); @@ -1070,7 +1023,6 @@ standard_receive3(struct TCP_Server_Info *server, struct mid_q_entry *mid) server->vals->header_preamble_size) { cifs_server_dbg(VFS, "SMB response too long (%u bytes)\n", pdu_length); cifs_reconnect(server); - wake_up(&server->response_q); return -ECONNABORTED; } @@ -1118,7 +1070,6 @@ cifs_handle_standard(struct TCP_Server_Info *server, struct mid_q_entry *mid) if (server->ops->is_session_expired && server->ops->is_session_expired(buf)) { cifs_reconnect(server); - wake_up(&server->response_q); return -1; } @@ -1164,8 +1115,9 @@ cifs_demultiplex_thread(void *p) struct task_struct *task_to_wake = NULL; struct mid_q_entry *mids[MAX_COMPOUND]; char *bufs[MAX_COMPOUND]; + unsigned int noreclaim_flag; - current->flags |= PF_MEMALLOC; + noreclaim_flag = memalloc_noreclaim_save(); cifs_dbg(FYI, "Demultiplex PID: %d\n", task_pid_nr(current)); length = atomic_inc_return(&tcpSesAllocCount); @@ -1212,7 +1164,6 @@ next_pdu: cifs_server_dbg(VFS, "SMB response too short (%u bytes)\n", server->pdu_size); cifs_reconnect(server); - wake_up(&server->response_q); continue; } @@ -1320,6 +1271,7 @@ next_pdu: set_current_state(TASK_RUNNING); } + memalloc_noreclaim_restore(noreclaim_flag); module_put_and_exit(0); } @@ -1522,6 +1474,9 @@ cifs_parse_smb_version(char *value, struct smb_vol *vol, bool is_smb3) cifs_dbg(VFS, "vers=1.0 (cifs) not permitted when mounting with smb3\n"); return 1; } + cifs_dbg(VFS, "Use of the less secure dialect vers=1.0 " + "is not recommended unless required for " + "access to very old servers\n"); vol->ops = &smb1_operations; vol->vals = &smb1_values; break; @@ -2517,11 +2472,12 @@ cifs_parse_mount_options(const char *mountdata, const char *devname, pr_notice("CIFS: ignoring forcegid mount option specified with no gid= option.\n"); if (got_version == false) - pr_warn("No dialect specified on mount. Default has changed to " - "a more secure dialect, SMB2.1 or later (e.g. SMB3), from CIFS " - "(SMB1). To use the less secure SMB1 dialect to access " - "old servers which do not support SMB3 (or SMB2.1) specify vers=1.0" - " on mount.\n"); + pr_warn_once("No dialect specified on mount. Default has changed" + " to a more secure dialect, SMB2.1 or later (e.g. " + "SMB3.1.1), from CIFS (SMB1). To use the less secure " + "SMB1 dialect to access old servers which do not " + "support SMB3.1.1 (or even SMB3 or SMB2.1) specify " + "vers=1.0 on mount.\n"); kfree(mountdata_copy); return 0; @@ -4999,6 +4955,15 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb_vol *vol) * dentry revalidation to think the dentry are stale (ESTALE). */ cifs_autodisable_serverino(cifs_sb); + /* + * Force the use of prefix path to support failover on DFS paths that + * resolve to targets that have different prefix paths. + */ + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; + kfree(cifs_sb->prepath); + cifs_sb->prepath = vol->prepath; + vol->prepath = NULL; + out: free_xid(xid); cifs_try_adding_channels(ses); diff --git a/fs/cifs/dfs_cache.c b/fs/cifs/dfs_cache.c index 43c1b43a07ec..a67f88bf7ae1 100644 --- a/fs/cifs/dfs_cache.c +++ b/fs/cifs/dfs_cache.c @@ -1260,6 +1260,44 @@ void dfs_cache_del_vol(const char *fullpath) kref_put(&vi->refcnt, vol_release); } +/** + * dfs_cache_get_tgt_share - parse a DFS target + * + * @it: DFS target iterator. + * @share: tree name. + * @share_len: length of tree name. + * @prefix: prefix path. + * @prefix_len: length of prefix path. + * + * Return zero if target was parsed correctly, otherwise non-zero. + */ +int dfs_cache_get_tgt_share(const struct dfs_cache_tgt_iterator *it, + const char **share, size_t *share_len, + const char **prefix, size_t *prefix_len) +{ + char *s, sep; + + if (!it || !share || !share_len || !prefix || !prefix_len) + return -EINVAL; + + sep = it->it_name[0]; + if (sep != '\\' && sep != '/') + return -EINVAL; + + s = strchr(it->it_name + 1, sep); + if (!s) + return -EINVAL; + + s = strchrnul(s + 1, sep); + + *share = it->it_name; + *share_len = s - it->it_name; + *prefix = *s ? s + 1 : s; + *prefix_len = &it->it_name[strlen(it->it_name)] - *prefix; + + return 0; +} + /* Get all tcons that are within a DFS namespace and can be refreshed */ static void get_tcons(struct TCP_Server_Info *server, struct list_head *head) { diff --git a/fs/cifs/dfs_cache.h b/fs/cifs/dfs_cache.h index 99ee44f8ad07..bf94d08cfb5a 100644 --- a/fs/cifs/dfs_cache.h +++ b/fs/cifs/dfs_cache.h @@ -49,6 +49,10 @@ extern int dfs_cache_update_vol(const char *fullpath, struct TCP_Server_Info *server); extern void dfs_cache_del_vol(const char *fullpath); +extern int dfs_cache_get_tgt_share(const struct dfs_cache_tgt_iterator *it, + const char **share, size_t *share_len, + const char **prefix, size_t *prefix_len); + static inline struct dfs_cache_tgt_iterator * dfs_cache_get_next_tgt(struct dfs_cache_tgt_list *tl, struct dfs_cache_tgt_iterator *it) diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 0ef099442f20..36e7b2fd2190 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -555,7 +555,6 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry, if (server->ops->close) server->ops->close(xid, tcon, &fid); cifs_del_pending_open(&open); - fput(file); rc = -ENOMEM; } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index bc9516ab4b34..5920820bfbd0 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1169,7 +1169,8 @@ try_again: rc = posix_lock_file(file, flock, NULL); up_write(&cinode->lock_sem); if (rc == FILE_LOCK_DEFERRED) { - rc = wait_event_interruptible(flock->fl_wait, !flock->fl_blocker); + rc = wait_event_interruptible(flock->fl_wait, + list_empty(&flock->fl_blocked_member)); if (!rc) goto try_again; locks_delete_block(flock); @@ -1958,7 +1959,7 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, /* Return -EBADF if no handle is found and general rc otherwise */ int -cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only, +cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags, struct cifsFileInfo **ret_file) { struct cifsFileInfo *open_file, *inv_file = NULL; @@ -1966,7 +1967,8 @@ cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only, bool any_available = false; int rc = -EBADF; unsigned int refind = 0; - + bool fsuid_only = flags & FIND_WR_FSUID_ONLY; + bool with_delete = flags & FIND_WR_WITH_DELETE; *ret_file = NULL; /* @@ -1998,6 +2000,8 @@ refind_writable: continue; if (fsuid_only && !uid_eq(open_file->uid, current_fsuid())) continue; + if (with_delete && !(open_file->fid.access & DELETE)) + continue; if (OPEN_FMODE(open_file->f_flags) & FMODE_WRITE) { if (!open_file->invalidHandle) { /* found a good writable file */ @@ -2045,12 +2049,12 @@ refind_writable: } struct cifsFileInfo * -find_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) +find_writable_file(struct cifsInodeInfo *cifs_inode, int flags) { struct cifsFileInfo *cfile; int rc; - rc = cifs_get_writable_file(cifs_inode, fsuid_only, &cfile); + rc = cifs_get_writable_file(cifs_inode, flags, &cfile); if (rc) cifs_dbg(FYI, "couldn't find writable handle rc=%d", rc); @@ -2059,6 +2063,7 @@ find_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) int cifs_get_writable_path(struct cifs_tcon *tcon, const char *name, + int flags, struct cifsFileInfo **ret_file) { struct list_head *tmp; @@ -2085,7 +2090,7 @@ cifs_get_writable_path(struct cifs_tcon *tcon, const char *name, kfree(full_path); cinode = CIFS_I(d_inode(cfile->dentry)); spin_unlock(&tcon->open_file_lock); - return cifs_get_writable_file(cinode, 0, ret_file); + return cifs_get_writable_file(cinode, flags, ret_file); } spin_unlock(&tcon->open_file_lock); @@ -2162,7 +2167,8 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) if (mapping->host->i_size - offset < (loff_t)to) to = (unsigned)(mapping->host->i_size - offset); - rc = cifs_get_writable_file(CIFS_I(mapping->host), false, &open_file); + rc = cifs_get_writable_file(CIFS_I(mapping->host), FIND_WR_ANY, + &open_file); if (!rc) { bytes_written = cifs_write(open_file, open_file->pid, write_data, to - from, &offset); @@ -2355,7 +2361,7 @@ retry: if (cfile) cifsFileInfo_put(cfile); - rc = cifs_get_writable_file(CIFS_I(inode), false, &cfile); + rc = cifs_get_writable_file(CIFS_I(inode), FIND_WR_ANY, &cfile); /* in case of an error store it to return later */ if (rc) @@ -3835,7 +3841,7 @@ again: if (rc == -ENODATA) rc = 0; - ctx->rc = (rc == 0) ? ctx->total_len : rc; + ctx->rc = (rc == 0) ? (ssize_t)ctx->total_len : rc; mutex_unlock(&ctx->aio_mutex); diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index b5e6635c578e..8d01ec2dca66 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -653,8 +653,8 @@ cifs_all_info_to_fattr(struct cifs_fattr *fattr, FILE_ALL_INFO *info, */ if ((fattr->cf_nlink < 1) && !tcon->unix_ext && !info->DeletePending) { - cifs_dbg(1, "bogus file nlink value %u\n", - fattr->cf_nlink); + cifs_dbg(VFS, "bogus file nlink value %u\n", + fattr->cf_nlink); fattr->cf_flags |= CIFS_FATTR_UNKNOWN_NLINK; } } @@ -1835,6 +1835,8 @@ cifs_do_rename(const unsigned int xid, struct dentry *from_dentry, CIFSSMBClose(xid, tcon, fid.netfid); } do_rename_exit: + if (rc == 0) + d_move(from_dentry, to_dentry); cifs_put_tlink(tlink); return rc; } @@ -2073,6 +2075,7 @@ int cifs_revalidate_dentry_attr(struct dentry *dentry) struct inode *inode = d_inode(dentry); struct super_block *sb = dentry->d_sb; char *full_path = NULL; + int count = 0; if (inode == NULL) return -ENOENT; @@ -2094,15 +2097,18 @@ int cifs_revalidate_dentry_attr(struct dentry *dentry) full_path, inode, inode->i_count.counter, dentry, cifs_get_time(dentry), jiffies); +again: if (cifs_sb_master_tcon(CIFS_SB(sb))->unix_ext) rc = cifs_get_inode_info_unix(&inode, full_path, sb, xid); else rc = cifs_get_inode_info(&inode, full_path, NULL, sb, xid, NULL); - + if (rc == -EAGAIN && count++ < 10) + goto again; out: kfree(full_path); free_xid(xid); + return rc; } @@ -2144,8 +2150,9 @@ int cifs_getattr(const struct path *path, struct kstat *stat, * We need to be sure that all dirty pages are written and the server * has actual ctime, mtime and file length. */ - if (!CIFS_CACHE_READ(CIFS_I(inode)) && inode->i_mapping && - inode->i_mapping->nrpages != 0) { + if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_SIZE)) && + !CIFS_CACHE_READ(CIFS_I(inode)) && + inode->i_mapping && inode->i_mapping->nrpages != 0) { rc = filemap_fdatawait(inode->i_mapping); if (rc) { mapping_set_error(inode->i_mapping, rc); @@ -2153,9 +2160,20 @@ int cifs_getattr(const struct path *path, struct kstat *stat, } } - rc = cifs_revalidate_dentry_attr(dentry); - if (rc) - return rc; + if ((flags & AT_STATX_SYNC_TYPE) == AT_STATX_FORCE_SYNC) + CIFS_I(inode)->time = 0; /* force revalidate */ + + /* + * If the caller doesn't require syncing, only sync if + * necessary (e.g. due to earlier truncate or setattr + * invalidating the cached metadata) + */ + if (((flags & AT_STATX_SYNC_TYPE) != AT_STATX_DONT_SYNC) || + (CIFS_I(inode)->time == 0)) { + rc = cifs_revalidate_dentry_attr(dentry); + if (rc) + return rc; + } generic_fillattr(inode, stat); stat->blksize = cifs_sb->bsize; @@ -2187,7 +2205,7 @@ int cifs_getattr(const struct path *path, struct kstat *stat, if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_OVERR_GID)) stat->gid = current_fsgid(); } - return rc; + return 0; } int cifs_fiemap(struct inode *inode, struct fiemap_extent_info *fei, u64 start, @@ -2278,7 +2296,7 @@ cifs_set_file_size(struct inode *inode, struct iattr *attrs, * writebehind data than the SMB timeout for the SetPathInfo * request would allow */ - open_file = find_writable_file(cifsInode, true); + open_file = find_writable_file(cifsInode, FIND_WR_FSUID_ONLY); if (open_file) { tcon = tlink_tcon(open_file->tlink); server = tcon->ses->server; @@ -2428,7 +2446,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) args->ctime = NO_CHANGE_64; args->device = 0; - open_file = find_writable_file(cifsInode, true); + open_file = find_writable_file(cifsInode, FIND_WR_FSUID_ONLY); if (open_file) { u16 nfid = open_file->fid.netfid; u32 npid = open_file->pid; @@ -2512,26 +2530,27 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) /* * Attempt to flush data before changing attributes. We need to do - * this for ATTR_SIZE and ATTR_MTIME for sure, and if we change the - * ownership or mode then we may also need to do this. Here, we take - * the safe way out and just do the flush on all setattr requests. If - * the flush returns error, store it to report later and continue. + * this for ATTR_SIZE and ATTR_MTIME. If the flush of the data + * returns error, store it to report later and continue. * * BB: This should be smarter. Why bother flushing pages that * will be truncated anyway? Also, should we error out here if - * the flush returns error? + * the flush returns error? Do we need to check for ATTR_MTIME_SET flag? */ - rc = filemap_write_and_wait(inode->i_mapping); - if (is_interrupt_error(rc)) { - rc = -ERESTARTSYS; - goto cifs_setattr_exit; + if (attrs->ia_valid & (ATTR_MTIME | ATTR_SIZE | ATTR_CTIME)) { + rc = filemap_write_and_wait(inode->i_mapping); + if (is_interrupt_error(rc)) { + rc = -ERESTARTSYS; + goto cifs_setattr_exit; + } + mapping_set_error(inode->i_mapping, rc); } - mapping_set_error(inode->i_mapping, rc); rc = 0; - if (attrs->ia_valid & ATTR_MTIME) { - rc = cifs_get_writable_file(cifsInode, false, &wfile); + if ((attrs->ia_valid & ATTR_MTIME) && + !(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NOSSYNC)) { + rc = cifs_get_writable_file(cifsInode, FIND_WR_ANY, &wfile); if (!rc) { tcon = tlink_tcon(wfile->tlink); rc = tcon->ses->server->ops->flush(xid, tcon, &wfile->fid); diff --git a/fs/cifs/link.c b/fs/cifs/link.c index 852aa00ec729..a25ef35b023e 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -416,7 +416,7 @@ smb3_query_mf_symlink(unsigned int xid, struct cifs_tcon *tcon, } rc = SMB2_open(xid, &oparms, utf16_path, &oplock, pfile_info, NULL, - NULL); + NULL, NULL); if (rc) goto qmf_out_open_fail; @@ -470,7 +470,7 @@ smb3_create_mf_symlink(unsigned int xid, struct cifs_tcon *tcon, oparms.reconnect = false; rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, - NULL); + NULL, NULL); if (rc) { kfree(utf16_path); return rc; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 40ca394fd5de..a456febd4109 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -31,6 +31,7 @@ #include "nterr.h" #include "cifs_unicode.h" #include "smb2pdu.h" +#include "cifsfs.h" extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; @@ -1022,3 +1023,82 @@ int copy_path_name(char *dst, const char *src) name_len++; return name_len; } + +struct super_cb_data { + struct TCP_Server_Info *server; + struct super_block *sb; +}; + +static void super_cb(struct super_block *sb, void *arg) +{ + struct super_cb_data *d = arg; + struct cifs_sb_info *cifs_sb; + struct cifs_tcon *tcon; + + if (d->sb) + return; + + cifs_sb = CIFS_SB(sb); + tcon = cifs_sb_master_tcon(cifs_sb); + if (tcon->ses->server == d->server) + d->sb = sb; +} + +struct super_block *cifs_get_tcp_super(struct TCP_Server_Info *server) +{ + struct super_cb_data d = { + .server = server, + .sb = NULL, + }; + + iterate_supers_type(&cifs_fs_type, super_cb, &d); + + if (unlikely(!d.sb)) + return ERR_PTR(-ENOENT); + /* + * Grab an active reference in order to prevent automounts (DFS links) + * of expiring and then freeing up our cifs superblock pointer while + * we're doing failover. + */ + cifs_sb_active(d.sb); + return d.sb; +} + +void cifs_put_tcp_super(struct super_block *sb) +{ + if (!IS_ERR_OR_NULL(sb)) + cifs_sb_deactive(sb); +} + +int update_super_prepath(struct cifs_tcon *tcon, const char *prefix, + size_t prefix_len) +{ + struct super_block *sb; + struct cifs_sb_info *cifs_sb; + int rc = 0; + + sb = cifs_get_tcp_super(tcon->ses->server); + if (IS_ERR(sb)) + return PTR_ERR(sb); + + cifs_sb = CIFS_SB(sb); + + kfree(cifs_sb->prepath); + + if (*prefix && prefix_len) { + cifs_sb->prepath = kstrndup(prefix, prefix_len, GFP_ATOMIC); + if (!cifs_sb->prepath) { + rc = -ENOMEM; + goto out; + } + + convert_delimiter(cifs_sb->prepath, CIFS_DIR_SEP(cifs_sb)); + } else + cifs_sb->prepath = NULL; + + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; + +out: + cifs_put_tcp_super(sb); + return rc; +} diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index ba9dadf3be24..19e4a5d3b4ca 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -32,6 +32,7 @@ #include "cifs_debug.h" #include "cifs_fs_sb.h" #include "cifsfs.h" +#include "smb2proto.h" /* * To be safe - for UCS to UTF-8 with strings loaded with the rare long @@ -217,6 +218,60 @@ cifs_fill_common_info(struct cifs_fattr *fattr, struct cifs_sb_info *cifs_sb) } } +/* Fill a cifs_fattr struct with info from SMB_FIND_FILE_POSIX_INFO. */ +static void +cifs_posix_to_fattr(struct cifs_fattr *fattr, struct smb2_posix_info *info, + struct cifs_sb_info *cifs_sb) +{ + struct smb2_posix_info_parsed parsed; + + posix_info_parse(info, NULL, &parsed); + + memset(fattr, 0, sizeof(*fattr)); + fattr->cf_uniqueid = le64_to_cpu(info->Inode); + fattr->cf_bytes = le64_to_cpu(info->AllocationSize); + fattr->cf_eof = le64_to_cpu(info->EndOfFile); + + fattr->cf_atime = cifs_NTtimeToUnix(info->LastAccessTime); + fattr->cf_mtime = cifs_NTtimeToUnix(info->LastWriteTime); + fattr->cf_ctime = cifs_NTtimeToUnix(info->CreationTime); + + fattr->cf_nlink = le32_to_cpu(info->HardLinks); + fattr->cf_cifsattrs = le32_to_cpu(info->DosAttributes); + + /* + * Since we set the inode type below we need to mask off + * to avoid strange results if bits set above. + * XXX: why not make server&client use the type bits? + */ + fattr->cf_mode = le32_to_cpu(info->Mode) & ~S_IFMT; + + cifs_dbg(VFS, "XXX dev %d, reparse %d, mode %o", + le32_to_cpu(info->DeviceId), + le32_to_cpu(info->ReparseTag), + le32_to_cpu(info->Mode)); + + if (fattr->cf_cifsattrs & ATTR_DIRECTORY) { + fattr->cf_mode |= S_IFDIR; + fattr->cf_dtype = DT_DIR; + } else { + /* + * mark anything that is not a dir as regular + * file. special files should have the REPARSE + * attribute and will be marked as needing revaluation + */ + fattr->cf_mode |= S_IFREG; + fattr->cf_dtype = DT_REG; + } + + if (reparse_file_needs_reval(fattr)) + fattr->cf_flags |= CIFS_FATTR_NEED_REVAL; + + /* TODO map SIDs */ + fattr->cf_uid = cifs_sb->mnt_uid; + fattr->cf_gid = cifs_sb->mnt_gid; +} + static void __dir_info_to_fattr(struct cifs_fattr *fattr, const void *info) { const FILE_DIRECTORY_INFO *fi = info; @@ -359,6 +414,8 @@ ffirst_retry: /* if (cap_unix(tcon->ses) { */ if (tcon->unix_ext) cifsFile->srch_inf.info_level = SMB_FIND_FILE_UNIX; + else if (tcon->posix_extensions) + cifsFile->srch_inf.info_level = SMB_FIND_FILE_POSIX_INFO; else if ((tcon->ses->capabilities & tcon->ses->server->vals->cap_nt_find) == 0) { cifsFile->srch_inf.info_level = SMB_FIND_FILE_INFO_STANDARD; @@ -451,6 +508,23 @@ struct cifs_dirent { u64 ino; }; +static void cifs_fill_dirent_posix(struct cifs_dirent *de, + const struct smb2_posix_info *info) +{ + struct smb2_posix_info_parsed parsed; + + /* payload should have already been checked at this point */ + if (posix_info_parse(info, NULL, &parsed) < 0) { + cifs_dbg(VFS, "invalid POSIX info payload"); + return; + } + + de->name = parsed.name; + de->namelen = parsed.name_len; + de->resume_key = info->Ignored; + de->ino = le64_to_cpu(info->Inode); +} + static void cifs_fill_dirent_unix(struct cifs_dirent *de, const FILE_UNIX_INFO *info, bool is_unicode) { @@ -511,6 +585,9 @@ static int cifs_fill_dirent(struct cifs_dirent *de, const void *info, memset(de, 0, sizeof(*de)); switch (level) { + case SMB_FIND_FILE_POSIX_INFO: + cifs_fill_dirent_posix(de, info); + break; case SMB_FIND_FILE_UNIX: cifs_fill_dirent_unix(de, info, is_unicode); break; @@ -786,6 +863,11 @@ static int cifs_filldir(char *find_entry, struct file *file, } switch (file_info->srch_inf.info_level) { + case SMB_FIND_FILE_POSIX_INFO: + cifs_posix_to_fattr(&fattr, + (struct smb2_posix_info *)find_entry, + cifs_sb); + break; case SMB_FIND_FILE_UNIX: cifs_unix_basic_to_fattr(&fattr, &((FILE_UNIX_INFO *)find_entry)->basic, diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c index eb994e313c6a..b130efaf8feb 100644 --- a/fs/cifs/smb1ops.c +++ b/fs/cifs/smb1ops.c @@ -766,7 +766,7 @@ smb_set_file_info(struct inode *inode, const char *full_path, struct cifs_tcon *tcon; /* if the file is already open for write, just use that fileid */ - open_file = find_writable_file(cinode, true); + open_file = find_writable_file(cinode, FIND_WR_FSUID_ONLY); if (open_file) { fid.netfid = open_file->fid.netfid; netpid = open_file->pid; diff --git a/fs/cifs/smb2file.c b/fs/cifs/smb2file.c index afe1f03aabe3..2fa3ba354cc9 100644 --- a/fs/cifs/smb2file.c +++ b/fs/cifs/smb2file.c @@ -62,7 +62,7 @@ smb2_open_file(const unsigned int xid, struct cifs_open_parms *oparms, smb2_oplock = SMB2_OPLOCK_LEVEL_BATCH; rc = SMB2_open(xid, oparms, smb2_path, &smb2_oplock, smb2_data, NULL, - NULL); + NULL, NULL); if (rc) goto out; @@ -152,7 +152,12 @@ smb2_unlock_range(struct cifsFileInfo *cfile, struct file_lock *flock, (li->offset + li->length)) continue; if (current->tgid != li->pid) - continue; + /* + * flock and OFD lock are associated with an open + * file description, not the process. + */ + if (!(flock->fl_flags & (FL_FLOCK | FL_OFDLCK))) + continue; if (cinode->can_cache_brlcks) { /* * We can cache brlock requests - simply remove a lock diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c index 1cf207564ff9..a8c301ae00ed 100644 --- a/fs/cifs/smb2inode.c +++ b/fs/cifs/smb2inode.c @@ -521,7 +521,7 @@ smb2_mkdir_setinfo(struct inode *inode, const char *name, cifs_i = CIFS_I(inode); dosattrs = cifs_i->cifsAttrs | ATTR_READONLY; data.Attributes = cpu_to_le32(dosattrs); - cifs_get_writable_path(tcon, name, &cfile); + cifs_get_writable_path(tcon, name, FIND_WR_ANY, &cfile); tmprc = smb2_compound_op(xid, tcon, cifs_sb, name, FILE_WRITE_ATTRIBUTES, FILE_CREATE, CREATE_NOT_FILE, ACL_NO_MODE, @@ -577,7 +577,7 @@ smb2_rename_path(const unsigned int xid, struct cifs_tcon *tcon, { struct cifsFileInfo *cfile; - cifs_get_writable_path(tcon, from_name, &cfile); + cifs_get_writable_path(tcon, from_name, FIND_WR_WITH_DELETE, &cfile); return smb2_set_path_attr(xid, tcon, from_name, to_name, cifs_sb, DELETE, SMB2_OP_RENAME, cfile); diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index e47190cae163..b36c46f48705 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -328,16 +328,6 @@ smb2_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *volume_info) /* start with specified wsize, or default */ wsize = volume_info->wsize ? volume_info->wsize : CIFS_DEFAULT_IOSIZE; wsize = min_t(unsigned int, wsize, server->max_write); -#ifdef CONFIG_CIFS_SMB_DIRECT - if (server->rdma) { - if (server->sign) - wsize = min_t(unsigned int, - wsize, server->smbd_conn->max_fragmented_send_size); - else - wsize = min_t(unsigned int, - wsize, server->smbd_conn->max_readwrite_size); - } -#endif if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU)) wsize = min_t(unsigned int, wsize, SMB2_MAX_BUFFER_SIZE); @@ -356,8 +346,15 @@ smb3_negotiate_wsize(struct cifs_tcon *tcon, struct smb_vol *volume_info) #ifdef CONFIG_CIFS_SMB_DIRECT if (server->rdma) { if (server->sign) + /* + * Account for SMB2 data transfer packet header and + * possible encryption header + */ wsize = min_t(unsigned int, - wsize, server->smbd_conn->max_fragmented_send_size); + wsize, + server->smbd_conn->max_fragmented_send_size - + SMB2_READWRITE_PDU_HEADER_SIZE - + sizeof(struct smb2_transform_hdr)); else wsize = min_t(unsigned int, wsize, server->smbd_conn->max_readwrite_size); @@ -378,16 +375,6 @@ smb2_negotiate_rsize(struct cifs_tcon *tcon, struct smb_vol *volume_info) /* start with specified rsize, or default */ rsize = volume_info->rsize ? volume_info->rsize : CIFS_DEFAULT_IOSIZE; rsize = min_t(unsigned int, rsize, server->max_read); -#ifdef CONFIG_CIFS_SMB_DIRECT - if (server->rdma) { - if (server->sign) - rsize = min_t(unsigned int, - rsize, server->smbd_conn->max_fragmented_recv_size); - else - rsize = min_t(unsigned int, - rsize, server->smbd_conn->max_readwrite_size); - } -#endif if (!(server->capabilities & SMB2_GLOBAL_CAP_LARGE_MTU)) rsize = min_t(unsigned int, rsize, SMB2_MAX_BUFFER_SIZE); @@ -407,8 +394,15 @@ smb3_negotiate_rsize(struct cifs_tcon *tcon, struct smb_vol *volume_info) #ifdef CONFIG_CIFS_SMB_DIRECT if (server->rdma) { if (server->sign) + /* + * Account for SMB2 data transfer packet header and + * possible encryption header + */ rsize = min_t(unsigned int, - rsize, server->smbd_conn->max_fragmented_recv_size); + rsize, + server->smbd_conn->max_fragmented_recv_size - + SMB2_READWRITE_PDU_HEADER_SIZE - + sizeof(struct smb2_transform_hdr)); else rsize = min_t(unsigned int, rsize, server->smbd_conn->max_readwrite_size); @@ -794,7 +788,8 @@ int open_shroot(unsigned int xid, struct cifs_tcon *tcon, tcon->crfid.has_lease = true; smb2_parse_contexts(server, o_rsp, &oparms.fid->epoch, - oparms.fid->lease_key, &oplock, NULL); + oparms.fid->lease_key, &oplock, + NULL, NULL); } else goto oshr_exit; @@ -838,7 +833,7 @@ smb3_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon, if (no_cached_open) rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL, - NULL); + NULL, NULL); else rc = open_shroot(xid, tcon, cifs_sb, &fid); @@ -878,7 +873,8 @@ smb2_qfs_tcon(const unsigned int xid, struct cifs_tcon *tcon, oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL, NULL); + rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL, + NULL, NULL); if (rc) return; @@ -913,7 +909,8 @@ smb2_is_path_accessible(const unsigned int xid, struct cifs_tcon *tcon, oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL); + rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL, + NULL); if (rc) { kfree(utf16_path); return rc; @@ -1364,6 +1361,7 @@ smb2_set_fid(struct cifsFileInfo *cfile, struct cifs_fid *fid, __u32 oplock) cfile->fid.persistent_fid = fid->persistent_fid; cfile->fid.volatile_fid = fid->volatile_fid; + cfile->fid.access = fid->access; #ifdef CONFIG_CIFS_DEBUG2 cfile->fid.mid = fid->mid; #endif /* CIFS_DEBUG2 */ @@ -2121,7 +2119,8 @@ smb3_notify(const unsigned int xid, struct file *pfile, oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL); + rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL, + NULL); if (rc) goto notify_exit; @@ -2221,6 +2220,8 @@ smb2_query_dir_first(const unsigned int xid, struct cifs_tcon *tcon, goto qdf_free; } + atomic_inc(&tcon->num_remote_opens); + qd_rsp = (struct smb2_query_directory_rsp *)rsp_iov[1].iov_base; if (qd_rsp->sync_hdr.Status == STATUS_NO_MORE_FILES) { trace_smb3_query_dir_done(xid, fid->persistent_fid, @@ -2540,7 +2541,8 @@ smb311_queryfs(const unsigned int xid, struct cifs_tcon *tcon, oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL, NULL); + rc = SMB2_open(xid, &oparms, &srch_path, &oplock, NULL, NULL, + NULL, NULL); if (rc) return rc; @@ -3025,7 +3027,8 @@ get_smb2_acl_by_path(struct cifs_sb_info *cifs_sb, oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL); + rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL, + NULL); kfree(utf16_path); if (!rc) { rc = SMB2_query_acl(xid, tlink_tcon(tlink), fid.persistent_fid, @@ -3083,7 +3086,8 @@ set_smb2_acl(struct cifs_ntsd *pnntsd, __u32 acllen, oparms.fid = &fid; oparms.reconnect = false; - rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, NULL); + rc = SMB2_open(xid, &oparms, utf16_path, &oplock, NULL, NULL, + NULL, NULL); kfree(utf16_path); if (!rc) { rc = SMB2_set_acl(xid, tlink_tcon(tlink), fid.persistent_fid, @@ -3245,6 +3249,10 @@ static long smb3_simple_falloc(struct file *file, struct cifs_tcon *tcon, * Extending the file */ if ((keep_size == false) && i_size_read(inode) < off + len) { + rc = inode_newsize_ok(inode, off + len); + if (rc) + goto out; + if ((cifsi->cifsAttrs & FILE_ATTRIBUTE_SPARSE_FILE) == 0) smb2_set_sparse(xid, tcon, cfile, inode, false); @@ -3327,7 +3335,7 @@ static loff_t smb3_llseek(struct file *file, struct cifs_tcon *tcon, loff_t offs * some servers (Windows2016) will not reflect recent writes in * QUERY_ALLOCATED_RANGES until SMB2_flush is called. */ - wrcfile = find_writable_file(cifsi, false); + wrcfile = find_writable_file(cifsi, FIND_WR_ANY); if (wrcfile) { filemap_write_and_wait(inode->i_mapping); smb2_flush_file(xid, tcon, &wrcfile->fid); @@ -3416,7 +3424,7 @@ static int smb3_fiemap(struct cifs_tcon *tcon, if (rc) goto out; - if (out_data_len < sizeof(struct file_allocated_range_buffer)) { + if (out_data_len && out_data_len < sizeof(struct file_allocated_range_buffer)) { rc = -EINVAL; goto out; } @@ -4148,7 +4156,6 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid, if (server->ops->is_session_expired && server->ops->is_session_expired(buf)) { cifs_reconnect(server); - wake_up(&server->response_q); return -1; } @@ -4512,14 +4519,12 @@ smb3_receive_transform(struct TCP_Server_Info *server, cifs_server_dbg(VFS, "Transform message is too small (%u)\n", pdu_length); cifs_reconnect(server); - wake_up(&server->response_q); return -ECONNABORTED; } if (pdu_length < orig_len + sizeof(struct smb2_transform_hdr)) { cifs_server_dbg(VFS, "Transform message is broken\n"); cifs_reconnect(server); - wake_up(&server->response_q); return -ECONNABORTED; } diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 1234f9ccab03..47d3e382ecaa 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -193,9 +193,18 @@ static int __smb2_reconnect(const struct nls_table *nlsc, for (it = dfs_cache_get_tgt_iterator(&tl); it; it = dfs_cache_get_next_tgt(&tl, it)) { - const char *tgt = dfs_cache_get_tgt_name(it); + const char *share, *prefix; + size_t share_len, prefix_len; - extract_unc_hostname(tgt, &dfs_host, &dfs_host_len); + rc = dfs_cache_get_tgt_share(it, &share, &share_len, &prefix, + &prefix_len); + if (rc) { + cifs_dbg(VFS, "%s: failed to parse target share %d\n", + __func__, rc); + continue; + } + + extract_unc_hostname(share, &dfs_host, &dfs_host_len); if (dfs_host_len != tcp_host_len || strncasecmp(dfs_host, tcp_host, dfs_host_len) != 0) { @@ -206,11 +215,13 @@ static int __smb2_reconnect(const struct nls_table *nlsc, continue; } - scnprintf(tree, MAX_TREE_SIZE, "\\%s", tgt); + scnprintf(tree, MAX_TREE_SIZE, "\\%.*s", (int)share_len, share); rc = SMB2_tcon(0, tcon->ses, tree, tcon, nlsc); - if (!rc) + if (!rc) { + rc = update_super_prepath(tcon, prefix, prefix_len); break; + } if (rc == -EREMOTE) break; } @@ -378,7 +389,7 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon) } if (smb2_command != SMB2_INTERNAL_CMD) - queue_delayed_work(cifsiod_wq, &server->reconnect, 0); + mod_delayed_work(cifsiod_wq, &server->reconnect, 0); atomic_inc(&tconInfoReconnectCount); out: @@ -1940,20 +1951,46 @@ parse_query_id_ctxt(struct create_context *cc, struct smb2_file_all_info *buf) } static void -parse_posix_ctxt(struct create_context *cc, struct smb_posix_info *pposix_inf) +parse_posix_ctxt(struct create_context *cc, struct smb2_file_all_info *info, + struct create_posix_rsp *posix) { - /* struct smb_posix_info *ppinf = (struct smb_posix_info *)cc; */ + int sid_len; + u8 *beg = (u8 *)cc + le16_to_cpu(cc->DataOffset); + u8 *end = beg + le32_to_cpu(cc->DataLength); + u8 *sid; + + memset(posix, 0, sizeof(*posix)); + + posix->nlink = le32_to_cpu(*(__le32 *)(beg + 0)); + posix->reparse_tag = le32_to_cpu(*(__le32 *)(beg + 4)); + posix->mode = le32_to_cpu(*(__le32 *)(beg + 8)); - /* TODO: Need to add parsing for the context and return */ - printk_once(KERN_WARNING - "SMB3 3.11 POSIX response context not completed yet\n"); + sid = beg + 12; + sid_len = posix_info_sid_size(sid, end); + if (sid_len < 0) { + cifs_dbg(VFS, "bad owner sid in posix create response\n"); + return; + } + memcpy(&posix->owner, sid, sid_len); + + sid = sid + sid_len; + sid_len = posix_info_sid_size(sid, end); + if (sid_len < 0) { + cifs_dbg(VFS, "bad group sid in posix create response\n"); + return; + } + memcpy(&posix->group, sid, sid_len); + + cifs_dbg(FYI, "nlink=%d mode=%o reparse_tag=%x\n", + posix->nlink, posix->mode, posix->reparse_tag); } void smb2_parse_contexts(struct TCP_Server_Info *server, - struct smb2_create_rsp *rsp, - unsigned int *epoch, char *lease_key, __u8 *oplock, - struct smb2_file_all_info *buf) + struct smb2_create_rsp *rsp, + unsigned int *epoch, char *lease_key, __u8 *oplock, + struct smb2_file_all_info *buf, + struct create_posix_rsp *posix) { char *data_offset; struct create_context *cc; @@ -1983,8 +2020,9 @@ smb2_parse_contexts(struct TCP_Server_Info *server, strncmp(name, SMB2_CREATE_QUERY_ON_DISK_ID, 4) == 0) parse_query_id_ctxt(cc, buf); else if ((le16_to_cpu(cc->NameLength) == 16)) { - if (memcmp(name, smb3_create_tag_posix, 16) == 0) - parse_posix_ctxt(cc, NULL); + if (posix && + memcmp(name, smb3_create_tag_posix, 16) == 0) + parse_posix_ctxt(cc, buf, posix); } /* else { cifs_dbg(FYI, "Context not matched with len %d\n", @@ -2709,6 +2747,7 @@ SMB2_open_free(struct smb_rqst *rqst) int SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, __u8 *oplock, struct smb2_file_all_info *buf, + struct create_posix_rsp *posix, struct kvec *err_iov, int *buftype) { struct smb_rqst rqst; @@ -2771,6 +2810,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, atomic_inc(&tcon->num_remote_opens); oparms->fid->persistent_fid = rsp->PersistentFileId; oparms->fid->volatile_fid = rsp->VolatileFileId; + oparms->fid->access = oparms->desired_access; #ifdef CONFIG_CIFS_DEBUG2 oparms->fid->mid = le64_to_cpu(rsp->sync_hdr.MessageId); #endif /* CIFS_DEBUG2 */ @@ -2786,7 +2826,7 @@ SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, smb2_parse_contexts(server, rsp, &oparms->fid->epoch, - oparms->fid->lease_key, oplock, buf); + oparms->fid->lease_key, oplock, buf, posix); creat_exit: SMB2_open_free(&rqst); free_rsp_buf(resp_buftype, rsp); @@ -3558,7 +3598,7 @@ SMB2_echo(struct TCP_Server_Info *server) if (server->tcpStatus == CifsNeedNegotiate) { /* No need to send echo on newly established connections */ - queue_delayed_work(cifsiod_wq, &server->reconnect, 0); + mod_delayed_work(cifsiod_wq, &server->reconnect, 0); return rc; } @@ -4285,8 +4325,104 @@ SMB2_write(const unsigned int xid, struct cifs_io_parms *io_parms, return rc; } +int posix_info_sid_size(const void *beg, const void *end) +{ + size_t subauth; + int total; + + if (beg + 1 > end) + return -1; + + subauth = *(u8 *)(beg+1); + if (subauth < 1 || subauth > 15) + return -1; + + total = 1 + 1 + 6 + 4*subauth; + if (beg + total > end) + return -1; + + return total; +} + +int posix_info_parse(const void *beg, const void *end, + struct smb2_posix_info_parsed *out) + +{ + int total_len = 0; + int sid_len; + int name_len; + const void *owner_sid; + const void *group_sid; + const void *name; + + /* if no end bound given, assume payload to be correct */ + if (!end) { + const struct smb2_posix_info *p = beg; + + end = beg + le32_to_cpu(p->NextEntryOffset); + /* last element will have a 0 offset, pick a sensible bound */ + if (end == beg) + end += 0xFFFF; + } + + /* check base buf */ + if (beg + sizeof(struct smb2_posix_info) > end) + return -1; + total_len = sizeof(struct smb2_posix_info); + + /* check owner sid */ + owner_sid = beg + total_len; + sid_len = posix_info_sid_size(owner_sid, end); + if (sid_len < 0) + return -1; + total_len += sid_len; + + /* check group sid */ + group_sid = beg + total_len; + sid_len = posix_info_sid_size(group_sid, end); + if (sid_len < 0) + return -1; + total_len += sid_len; + + /* check name len */ + if (beg + total_len + 4 > end) + return -1; + name_len = le32_to_cpu(*(__le32 *)(beg + total_len)); + if (name_len < 1 || name_len > 0xFFFF) + return -1; + total_len += 4; + + /* check name */ + name = beg + total_len; + if (name + name_len > end) + return -1; + total_len += name_len; + + if (out) { + out->base = beg; + out->size = total_len; + out->name_len = name_len; + out->name = name; + memcpy(&out->owner, owner_sid, + posix_info_sid_size(owner_sid, end)); + memcpy(&out->group, group_sid, + posix_info_sid_size(group_sid, end)); + } + return total_len; +} + +static int posix_info_extra_size(const void *beg, const void *end) +{ + int len = posix_info_parse(beg, end, NULL); + + if (len < 0) + return -1; + return len - sizeof(struct smb2_posix_info); +} + static unsigned int -num_entries(char *bufstart, char *end_of_buf, char **lastentry, size_t size) +num_entries(int infotype, char *bufstart, char *end_of_buf, char **lastentry, + size_t size) { int len; unsigned int entrycount = 0; @@ -4310,8 +4446,13 @@ num_entries(char *bufstart, char *end_of_buf, char **lastentry, size_t size) entryptr = entryptr + next_offset; dir_info = (FILE_DIRECTORY_INFO *)entryptr; - len = le32_to_cpu(dir_info->FileNameLength); - if (entryptr + len < entryptr || + if (infotype == SMB_FIND_FILE_POSIX_INFO) + len = posix_info_extra_size(entryptr, end_of_buf); + else + len = le32_to_cpu(dir_info->FileNameLength); + + if (len < 0 || + entryptr + len < entryptr || entryptr + len > end_of_buf || entryptr + len + size > end_of_buf) { cifs_dbg(VFS, "directory entry name would overflow frame end of buf %p\n", @@ -4361,6 +4502,9 @@ int SMB2_query_directory_init(const unsigned int xid, case SMB_FIND_FILE_ID_FULL_DIR_INFO: req->FileInformationClass = FILEID_FULL_DIRECTORY_INFORMATION; break; + case SMB_FIND_FILE_POSIX_INFO: + req->FileInformationClass = SMB_FIND_FILE_POSIX_INFO; + break; default: cifs_tcon_dbg(VFS, "info level %u isn't supported\n", info_level); @@ -4426,6 +4570,10 @@ smb2_parse_query_directory(struct cifs_tcon *tcon, case SMB_FIND_FILE_ID_FULL_DIR_INFO: info_buf_size = sizeof(SEARCH_ID_FULL_DIR_INFO) - 1; break; + case SMB_FIND_FILE_POSIX_INFO: + /* note that posix payload are variable size */ + info_buf_size = sizeof(struct smb2_posix_info); + break; default: cifs_tcon_dbg(VFS, "info level %u isn't supported\n", srch_inf->info_level); @@ -4435,8 +4583,10 @@ smb2_parse_query_directory(struct cifs_tcon *tcon, rc = smb2_validate_iov(le16_to_cpu(rsp->OutputBufferOffset), le32_to_cpu(rsp->OutputBufferLength), rsp_iov, info_buf_size); - if (rc) + if (rc) { + cifs_tcon_dbg(VFS, "bad info payload"); return rc; + } srch_inf->unicode = true; @@ -4450,9 +4600,14 @@ smb2_parse_query_directory(struct cifs_tcon *tcon, srch_inf->srch_entries_start = srch_inf->last_entry = (char *)rsp + le16_to_cpu(rsp->OutputBufferOffset); end_of_smb = rsp_iov->iov_len + (char *)rsp; - srch_inf->entries_in_buffer = - num_entries(srch_inf->srch_entries_start, end_of_smb, - &srch_inf->last_entry, info_buf_size); + + srch_inf->entries_in_buffer = num_entries( + srch_inf->info_level, + srch_inf->srch_entries_start, + end_of_smb, + &srch_inf->last_entry, + info_buf_size); + srch_inf->index_of_last_entry += srch_inf->entries_in_buffer; cifs_dbg(FYI, "num entries %d last_index %lld srch start %p srch end %p\n", srch_inf->entries_in_buffer, srch_inf->index_of_last_entry, diff --git a/fs/cifs/smb2pdu.h b/fs/cifs/smb2pdu.h index fa03df130f1a..10acf90f858d 100644 --- a/fs/cifs/smb2pdu.h +++ b/fs/cifs/smb2pdu.h @@ -91,6 +91,7 @@ #define SMB2_PROTO_NUMBER cpu_to_le32(0x424d53fe) #define SMB2_TRANSFORM_PROTO_NUM cpu_to_le32(0x424d53fd) +#define SMB2_COMPRESSION_TRANSFORM_ID cpu_to_le32(0x424d53fc) /* * SMB2 Header Definition @@ -119,6 +120,9 @@ struct smb2_sync_hdr { __u8 Signature[16]; } __packed; +/* The total header size for SMB2 read and write */ +#define SMB2_READWRITE_PDU_HEADER_SIZE (48 + sizeof(struct smb2_sync_hdr)) + struct smb2_sync_pdu { struct smb2_sync_hdr sync_hdr; __le16 StructureSize2; /* size of wct area (varies, request specific) */ @@ -127,16 +131,33 @@ struct smb2_sync_pdu { #define SMB3_AES128CCM_NONCE 11 #define SMB3_AES128GCM_NONCE 12 +/* Transform flags (for 3.0 dialect this flag indicates CCM */ +#define TRANSFORM_FLAG_ENCRYPTED 0x0001 struct smb2_transform_hdr { __le32 ProtocolId; /* 0xFD 'S' 'M' 'B' */ __u8 Signature[16]; __u8 Nonce[16]; __le32 OriginalMessageSize; __u16 Reserved1; - __le16 Flags; /* EncryptionAlgorithm */ + __le16 Flags; /* EncryptionAlgorithm for 3.0, enc enabled for 3.1.1 */ __u64 SessionId; } __packed; +/* See MS-SMB2 2.2.42.1 */ +struct compression_playload_header { + __le16 AlgorithmId; + __le16 Reserved; + __le32 Length; +} __packed; + +/* See MS-SMB2 2.2.42.2 */ +struct compression_pattern_payload_v1 { + __le16 Pattern; + __le16 Reserved1; + __le16 Reserved2; + __le32 Repetitions; +} __packed; + /* * SMB2 flag definitions */ @@ -182,7 +203,7 @@ struct smb2_symlink_err_rsp { __le16 PrintNameOffset; __le16 PrintNameLength; __le32 Flags; - __u8 PathBuffer[0]; + __u8 PathBuffer[]; } __packed; /* SMB 3.1.1 and later dialects. See MS-SMB2 section 2.2.2.1 */ @@ -192,6 +213,10 @@ struct smb2_error_context_rsp { __u8 ErrorContextData; /* ErrorDataLength long array */ } __packed; +/* ErrorId values */ +#define SMB2_ERROR_ID_DEFAULT 0x00000000 +#define SMB2_ERROR_ID_SHARE_REDIRECT cpu_to_le32(0x72645253) /* "rdRS" */ + /* Defines for Type field below (see MS-SMB2 2.2.2.2.2.1) */ #define MOVE_DST_IPADDR_V4 cpu_to_le32(0x00000001) #define MOVE_DST_IPADDR_V6 cpu_to_le32(0x00000002) @@ -210,7 +235,7 @@ struct share_redirect_error_context_rsp { __le16 Flags; __le16 TargetType; __le32 IPAddrCount; - struct move_dst_ipaddr IpAddrMoveList[0]; + struct move_dst_ipaddr IpAddrMoveList[]; /* __u8 ResourceName[] */ /* Name of share as counted Unicode string */ } __packed; @@ -307,11 +332,17 @@ struct smb2_encryption_neg_context { #define SMB3_COMPRESS_LZNT1 cpu_to_le16(0x0001) #define SMB3_COMPRESS_LZ77 cpu_to_le16(0x0002) #define SMB3_COMPRESS_LZ77_HUFF cpu_to_le16(0x0003) +/* Pattern scanning algorithm See MS-SMB2 3.1.4.4.1 */ +#define SMB3_COMPRESS_PATTERN cpu_to_le16(0x0004) + +/* Compression Flags */ +#define SMB2_COMPRESSION_CAPABILITIES_FLAG_NONE cpu_to_le32(0x00000000) +#define SMB2_COMPRESSION_CAPABILITIES_FLAG_CHAINED cpu_to_le32(0x00000001) struct smb2_compression_capabilities_context { __le16 ContextType; /* 3 */ __le16 DataLength; - __u32 Reserved; + __u32 Flags; __le16 CompressionAlgorithmCount; __u16 Padding; __u32 Reserved1; @@ -326,7 +357,7 @@ struct smb2_netname_neg_context { __le16 ContextType; /* 0x100 */ __le16 DataLength; __le32 Reserved; - __le16 NetName[0]; /* hostname of target converted to UCS-2 */ + __le16 NetName[]; /* hostname of target converted to UCS-2 */ } __packed; #define POSIX_CTXT_DATA_LEN 16 @@ -406,7 +437,7 @@ struct smb2_logoff_rsp { struct smb2_tree_connect_req { struct smb2_sync_hdr sync_hdr; __le16 StructureSize; /* Must be 9 */ - __le16 Reserved; /* Flags in SMB3.1.1 */ + __le16 Flags; /* Reserved MBZ for dialects prior to SMB3.1.1 */ __le16 PathOffset; __le16 PathLength; __u8 Buffer[1]; /* variable length */ @@ -421,13 +452,13 @@ struct tree_connect_contexts { __le16 ContextType; __le16 DataLength; __le32 Reserved; - __u8 Data[0]; + __u8 Data[]; } __packed; /* Remoted identity tree connect context structures - see MS-SMB2 2.2.9.2.1 */ struct smb3_blob_data { __le16 BlobSize; - __u8 BlobData[0]; + __u8 BlobData[]; } __packed; /* Valid values for Attr */ @@ -477,14 +508,14 @@ struct remoted_identity_tcon_context { __le16 DeviceGroups; /* offset to SID_ARRAY_DATA struct */ __le16 UserClaims; /* offset to BLOB_DATA struct */ __le16 DeviceClaims; /* offset to BLOB_DATA struct */ - __u8 TicketInfo[0]; /* variable length buf - remoted identity data */ + __u8 TicketInfo[]; /* variable length buf - remoted identity data */ } __packed; struct smb2_tree_connect_req_extension { __le32 TreeConnectContextOffset; __le16 TreeConnectContextCount; __u8 Reserved[10]; - __u8 PathName[0]; /* variable sized array */ + __u8 PathName[]; /* variable sized array */ /* followed by array of TreeConnectContexts */ } __packed; @@ -633,7 +664,7 @@ struct smb2_tree_disconnect_rsp { | FILE_WRITE_EA_LE | FILE_WRITE_ATTRIBUTES_LE) #define FILE_EXEC_RIGHTS_LE (FILE_EXECUTE_LE) -/* Impersonation Levels */ +/* Impersonation Levels. See MS-WPO section 9.7 and MSDN-IMPERS */ #define IL_ANONYMOUS cpu_to_le32(0x00000000) #define IL_IDENTIFICATION cpu_to_le32(0x00000001) #define IL_IMPERSONATION cpu_to_le32(0x00000002) @@ -689,7 +720,7 @@ struct smb2_create_req { __le16 NameLength; __le32 CreateContextsOffset; __le32 CreateContextsLength; - __u8 Buffer[0]; + __u8 Buffer[]; } __packed; /* @@ -727,7 +758,7 @@ struct create_context { __le16 Reserved; __le16 DataOffset; __le32 DataLength; - __u8 Buffer[0]; + __u8 Buffer[]; } __packed; #define SMB2_LEASE_READ_CACHING_HE 0x01 @@ -739,7 +770,7 @@ struct create_context { #define SMB2_LEASE_HANDLE_CACHING cpu_to_le32(0x02) #define SMB2_LEASE_WRITE_CACHING cpu_to_le32(0x04) -#define SMB2_LEASE_FLAG_BREAK_IN_PROGRESS cpu_to_le32(0x02) +#define SMB2_LEASE_FLAG_BREAK_IN_PROGRESS cpu_to_le32(0x00000002) #define SMB2_LEASE_FLAG_PARENT_LEASE_KEY_SET cpu_to_le32(0x00000004) #define SMB2_LEASE_KEY_SIZE 16 @@ -869,7 +900,7 @@ struct crt_sd_ctxt { struct resume_key_req { char ResumeKey[COPY_CHUNK_RES_KEY_SIZE]; __le32 ContextLength; /* MBZ */ - char Context[0]; /* ignored, Windows sets to 4 bytes of zero */ + char Context[]; /* ignored, Windows sets to 4 bytes of zero */ } __packed; /* this goes in the ioctl buffer when doing a copychunk request */ @@ -931,7 +962,7 @@ struct reparse_data_buffer { __le32 ReparseTag; __le16 ReparseDataLength; __u16 Reserved; - __u8 DataBuffer[0]; /* Variable Length */ + __u8 DataBuffer[]; /* Variable Length */ } __packed; struct reparse_guid_data_buffer { @@ -939,7 +970,7 @@ struct reparse_guid_data_buffer { __le16 ReparseDataLength; __u16 Reserved; __u8 ReparseGuid[16]; - __u8 DataBuffer[0]; /* Variable Length */ + __u8 DataBuffer[]; /* Variable Length */ } __packed; struct reparse_mount_point_data_buffer { @@ -950,7 +981,7 @@ struct reparse_mount_point_data_buffer { __le16 SubstituteNameLength; __le16 PrintNameOffset; __le16 PrintNameLength; - __u8 PathBuffer[0]; /* Variable Length */ + __u8 PathBuffer[]; /* Variable Length */ } __packed; #define SYMLINK_FLAG_RELATIVE 0x00000001 @@ -964,7 +995,7 @@ struct reparse_symlink_data_buffer { __le16 PrintNameOffset; __le16 PrintNameLength; __le32 Flags; - __u8 PathBuffer[0]; /* Variable Length */ + __u8 PathBuffer[]; /* Variable Length */ } __packed; /* See MS-FSCC 2.1.2.6 and cifspdu.h for struct reparse_posix_data */ @@ -1066,7 +1097,7 @@ struct smb2_ioctl_req { __le32 MaxOutputResponse; __le32 Flags; __u32 Reserved2; - __u8 Buffer[0]; + __u8 Buffer[]; } __packed; struct smb2_ioctl_rsp { @@ -1180,7 +1211,7 @@ struct smb2_write_req { __le64 Offset; __u64 PersistentFileId; /* opaque endianness */ __u64 VolatileFileId; /* opaque endianness */ - __le32 Channel; /* Reserved MBZ */ + __le32 Channel; /* MBZ unless SMB3.02 or later */ __le32 RemainingBytes; __le16 WriteChannelInfoOffset; __le16 WriteChannelInfoLength; @@ -1469,7 +1500,7 @@ struct smb3_fs_vol_info { __le32 VolumeLabelLength; /* includes trailing null */ __u8 SupportsObjects; /* True if eg like NTFS, supports objects */ __u8 Reserved; - __u8 VolumeLabel[0]; /* variable len */ + __u8 VolumeLabel[]; /* variable len */ } __packed; /* partial list of QUERY INFO levels */ @@ -1531,7 +1562,7 @@ struct smb2_file_rename_info { /* encoding of request for level 10 */ __u8 Reserved[7]; __u64 RootDirectory; /* MBZ for network operations (why says spec?) */ __le32 FileNameLength; - char FileName[0]; /* New name to be assigned */ + char FileName[]; /* New name to be assigned */ } __packed; /* level 10 Set */ struct smb2_file_link_info { /* encoding of request for level 11 */ @@ -1540,7 +1571,7 @@ struct smb2_file_link_info { /* encoding of request for level 11 */ __u8 Reserved[7]; __u64 RootDirectory; /* MBZ for network operations (why says spec?) */ __le32 FileNameLength; - char FileName[0]; /* Name to be assigned to new link */ + char FileName[]; /* Name to be assigned to new link */ } __packed; /* level 11 Set */ struct smb2_file_full_ea_info { /* encoding of response for level 15 */ @@ -1548,7 +1579,7 @@ struct smb2_file_full_ea_info { /* encoding of response for level 15 */ __u8 flags; __u8 ea_name_length; __le16 ea_value_length; - char ea_data[0]; /* \0 terminated name plus value */ + char ea_data[]; /* \0 terminated name plus value */ } __packed; /* level 15 Set */ /* @@ -1604,11 +1635,56 @@ struct smb2_file_id_information { extern char smb2_padding[7]; /* equivalent of the contents of SMB3.1.1 POSIX open context response */ -struct smb_posix_info { - __le32 nlink; - __le32 reparse_tag; - __le32 mode; - kuid_t uid; - kuid_t gid; +struct create_posix_rsp { + u32 nlink; + u32 reparse_tag; + u32 mode; + struct cifs_sid owner; /* var-sized on the wire */ + struct cifs_sid group; /* var-sized on the wire */ +} __packed; + +/* + * SMB2-only POSIX info level + * + * See posix_info_sid_size(), posix_info_extra_size() and + * posix_info_parse() to help with the handling of this struct. + */ +struct smb2_posix_info { + __le32 NextEntryOffset; + __u32 Ignored; + __le64 CreationTime; + __le64 LastAccessTime; + __le64 LastWriteTime; + __le64 ChangeTime; + __le64 EndOfFile; + __le64 AllocationSize; + __le32 DosAttributes; + __le64 Inode; + __le32 DeviceId; + __le32 Zero; + /* beginning of POSIX Create Context Response */ + __le32 HardLinks; + __le32 ReparseTag; + __le32 Mode; + /* + * var sized owner SID + * var sized group SID + * le32 filenamelength + * u8 filename[] + */ +} __packed; + +/* + * Parsed version of the above struct. Allows direct access to the + * variable length fields + */ +struct smb2_posix_info_parsed { + const struct smb2_posix_info *base; + size_t size; + struct cifs_sid owner; + struct cifs_sid group; + int name_len; + const u8 *name; }; + #endif /* _SMB2PDU_H */ diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h index de6388ef344f..4d1ff7b66fdc 100644 --- a/fs/cifs/smb2proto.h +++ b/fs/cifs/smb2proto.h @@ -139,6 +139,7 @@ extern int SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon); extern int SMB2_open(const unsigned int xid, struct cifs_open_parms *oparms, __le16 *path, __u8 *oplock, struct smb2_file_all_info *buf, + struct create_posix_rsp *posix, struct kvec *err_iov, int *resp_buftype); extern int SMB2_open_init(struct cifs_tcon *tcon, struct smb_rqst *rqst, __u8 *oplock, struct cifs_open_parms *oparms, @@ -252,7 +253,8 @@ extern enum securityEnum smb2_select_sectype(struct TCP_Server_Info *, extern void smb2_parse_contexts(struct TCP_Server_Info *server, struct smb2_create_rsp *rsp, unsigned int *epoch, char *lease_key, - __u8 *oplock, struct smb2_file_all_info *buf); + __u8 *oplock, struct smb2_file_all_info *buf, + struct create_posix_rsp *posix); extern int smb3_encryption_required(const struct cifs_tcon *tcon); extern int smb2_validate_iov(unsigned int offset, unsigned int buffer_length, struct kvec *iov, unsigned int min_buf_size); @@ -272,4 +274,7 @@ extern int smb2_query_info_compound(const unsigned int xid, u32 class, u32 type, u32 output_len, struct kvec *rsp, int *buftype, struct cifs_sb_info *cifs_sb); +int posix_info_parse(const void *beg, const void *end, + struct smb2_posix_info_parsed *out); +int posix_info_sid_size(const void *beg, const void *end); #endif /* _SMB2PROTO_H */ diff --git a/fs/cifs/smb2transport.c b/fs/cifs/smb2transport.c index 08b703b7a15e..20cc79e5c15d 100644 --- a/fs/cifs/smb2transport.c +++ b/fs/cifs/smb2transport.c @@ -602,7 +602,7 @@ int smb2_verify_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server) { unsigned int rc; - char server_response_sig[16]; + char server_response_sig[SMB2_SIGNATURE_SIZE]; struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)rqst->rq_iov[0].iov_base; @@ -638,9 +638,11 @@ smb2_verify_signature(struct smb_rqst *rqst, struct TCP_Server_Info *server) if (rc) return rc; - if (memcmp(server_response_sig, shdr->Signature, SMB2_SIGNATURE_SIZE)) + if (memcmp(server_response_sig, shdr->Signature, SMB2_SIGNATURE_SIZE)) { + dump_stack(); + cifs_dbg(VFS, "sign fail cmd 0x%x message id 0x%llx\n", shdr->Command, shdr->MessageId); return -EACCES; - else + } else return 0; } diff --git a/fs/cifs/smbdirect.c b/fs/cifs/smbdirect.c index 5b1b97e9e0c9..8da43a500686 100644 --- a/fs/cifs/smbdirect.c +++ b/fs/cifs/smbdirect.c @@ -459,25 +459,6 @@ static void smbd_post_send_credits(struct work_struct *work) check_and_send_immediate(info); } -static void smbd_recv_done_work(struct work_struct *work) -{ - struct smbd_connection *info = - container_of(work, struct smbd_connection, recv_done_work); - - /* - * We may have new send credits granted from remote peer - * If any sender is blcoked on lack of credets, unblock it - */ - if (atomic_read(&info->send_credits)) - wake_up_interruptible(&info->wait_send_queue); - - /* - * Check if we need to send something to remote peer to - * grant more credits or respond to KEEP_ALIVE packet - */ - check_and_send_immediate(info); -} - /* Called from softirq, when recv is done */ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) { @@ -546,8 +527,15 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) atomic_dec(&info->receive_credits); info->receive_credit_target = le16_to_cpu(data_transfer->credits_requested); - atomic_add(le16_to_cpu(data_transfer->credits_granted), - &info->send_credits); + if (le16_to_cpu(data_transfer->credits_granted)) { + atomic_add(le16_to_cpu(data_transfer->credits_granted), + &info->send_credits); + /* + * We have new send credits granted from remote peer + * If any sender is waiting for credits, unblock it + */ + wake_up_interruptible(&info->wait_send_queue); + } log_incoming(INFO, "data flags %d data_offset %d " "data_length %d remaining_data_length %d\n", @@ -563,7 +551,12 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) info->keep_alive_requested = KEEP_ALIVE_PENDING; } - queue_work(info->workqueue, &info->recv_done_work); + /* + * Check if we need to send something to remote peer to + * grant more credits or respond to KEEP_ALIVE packet + */ + check_and_send_immediate(info); + return; default: @@ -1762,7 +1755,6 @@ static struct smbd_connection *_smbd_get_connection( atomic_set(&info->send_payload_pending, 0); INIT_WORK(&info->disconnect_work, smbd_disconnect_rdma_work); - INIT_WORK(&info->recv_done_work, smbd_recv_done_work); INIT_WORK(&info->post_send_credits_work, smbd_post_send_credits); info->new_credits_offered = 0; spin_lock_init(&info->lock_new_credits_offered); @@ -2097,8 +2089,7 @@ int smbd_send(struct TCP_Server_Info *server, for (i = 0; i < num_rqst; i++) remaining_data_length += smb_rqst_len(server, &rqst_array[i]); - if (remaining_data_length + sizeof(struct smbd_data_transfer) > - info->max_fragmented_send_size) { + if (remaining_data_length > info->max_fragmented_send_size) { log_write(ERR, "payload size %d > max size %d\n", remaining_data_length, info->max_fragmented_send_size); rc = -EINVAL; diff --git a/fs/cifs/smbdirect.h b/fs/cifs/smbdirect.h index 6ff880a1e186..8ede915f2b24 100644 --- a/fs/cifs/smbdirect.h +++ b/fs/cifs/smbdirect.h @@ -67,7 +67,6 @@ struct smbd_connection { bool negotiate_done; struct work_struct disconnect_work; - struct work_struct recv_done_work; struct work_struct post_send_credits_work; spinlock_t lock_new_credits_offered; diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index cb3ee916f527..c97570eb2c18 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -466,7 +466,7 @@ smb_send_rqst(struct TCP_Server_Info *server, int num_rqst, struct smb_rqst *rqst, int flags) { struct kvec iov; - struct smb2_transform_hdr tr_hdr; + struct smb2_transform_hdr *tr_hdr; struct smb_rqst cur_rqst[MAX_COMPOUND]; int rc; @@ -476,28 +476,34 @@ smb_send_rqst(struct TCP_Server_Info *server, int num_rqst, if (num_rqst > MAX_COMPOUND - 1) return -ENOMEM; - memset(&cur_rqst[0], 0, sizeof(cur_rqst)); - memset(&iov, 0, sizeof(iov)); - memset(&tr_hdr, 0, sizeof(tr_hdr)); - - iov.iov_base = &tr_hdr; - iov.iov_len = sizeof(tr_hdr); - cur_rqst[0].rq_iov = &iov; - cur_rqst[0].rq_nvec = 1; - if (!server->ops->init_transform_rq) { cifs_server_dbg(VFS, "Encryption requested but transform " "callback is missing\n"); return -EIO; } + tr_hdr = kmalloc(sizeof(*tr_hdr), GFP_NOFS); + if (!tr_hdr) + return -ENOMEM; + + memset(&cur_rqst[0], 0, sizeof(cur_rqst)); + memset(&iov, 0, sizeof(iov)); + memset(tr_hdr, 0, sizeof(*tr_hdr)); + + iov.iov_base = tr_hdr; + iov.iov_len = sizeof(*tr_hdr); + cur_rqst[0].rq_iov = &iov; + cur_rqst[0].rq_nvec = 1; + rc = server->ops->init_transform_rq(server, num_rqst + 1, &cur_rqst[0], rqst); if (rc) - return rc; + goto out; rc = __smb_send_rqst(server, num_rqst + 1, &cur_rqst[0]); smb3_free_compound_rqst(num_rqst, &cur_rqst[1]); +out: + kfree(tr_hdr); return rc; } diff --git a/fs/crypto/fscrypt_private.h b/fs/crypto/fscrypt_private.h index 9aae851409e5..dbced2937ec8 100644 --- a/fs/crypto/fscrypt_private.h +++ b/fs/crypto/fscrypt_private.h @@ -76,6 +76,26 @@ static inline int fscrypt_context_size(const union fscrypt_context *ctx) return 0; } +/* Check whether an fscrypt_context has a recognized version number and size */ +static inline bool fscrypt_context_is_valid(const union fscrypt_context *ctx, + int ctx_size) +{ + return ctx_size >= 1 && ctx_size == fscrypt_context_size(ctx); +} + +/* Retrieve the context's nonce, assuming the context was already validated */ +static inline const u8 *fscrypt_context_nonce(const union fscrypt_context *ctx) +{ + switch (ctx->version) { + case FSCRYPT_CONTEXT_V1: + return ctx->v1.nonce; + case FSCRYPT_CONTEXT_V2: + return ctx->v2.nonce; + } + WARN_ON(1); + return NULL; +} + #undef fscrypt_policy union fscrypt_policy { u8 version; diff --git a/fs/crypto/keysetup.c b/fs/crypto/keysetup.c index 65cb09fa6ead..302375e9f719 100644 --- a/fs/crypto/keysetup.c +++ b/fs/crypto/keysetup.c @@ -425,20 +425,8 @@ int fscrypt_get_encryption_info(struct inode *inode) goto out; } - switch (ctx.version) { - case FSCRYPT_CONTEXT_V1: - memcpy(crypt_info->ci_nonce, ctx.v1.nonce, - FS_KEY_DERIVATION_NONCE_SIZE); - break; - case FSCRYPT_CONTEXT_V2: - memcpy(crypt_info->ci_nonce, ctx.v2.nonce, - FS_KEY_DERIVATION_NONCE_SIZE); - break; - default: - WARN_ON(1); - res = -EINVAL; - goto out; - } + memcpy(crypt_info->ci_nonce, fscrypt_context_nonce(&ctx), + FS_KEY_DERIVATION_NONCE_SIZE); if (!fscrypt_supported_policy(&crypt_info->ci_policy, inode)) { res = -EINVAL; @@ -539,6 +527,15 @@ int fscrypt_drop_inode(struct inode *inode) mk = ci->ci_master_key->payload.data[0]; /* + * With proper, non-racy use of FS_IOC_REMOVE_ENCRYPTION_KEY, all inodes + * protected by the key were cleaned by sync_filesystem(). But if + * userspace is still using the files, inodes can be dirtied between + * then and now. We mustn't lose any writes, so skip dirty inodes here. + */ + if (inode->i_state & I_DIRTY_ALL) + return 0; + + /* * Note: since we aren't holding ->mk_secret_sem, the result here can * immediately become outdated. But there's no correctness problem with * unnecessarily evicting. Nor is there a correctness problem with not diff --git a/fs/crypto/policy.c b/fs/crypto/policy.c index cf2a9d26ef7d..10ccf945020c 100644 --- a/fs/crypto/policy.c +++ b/fs/crypto/policy.c @@ -258,7 +258,7 @@ int fscrypt_policy_from_context(union fscrypt_policy *policy_u, { memset(policy_u, 0, sizeof(*policy_u)); - if (ctx_size <= 0 || ctx_size != fscrypt_context_size(ctx_u)) + if (!fscrypt_context_is_valid(ctx_u, ctx_size)) return -EINVAL; switch (ctx_u->version) { @@ -481,6 +481,25 @@ int fscrypt_ioctl_get_policy_ex(struct file *filp, void __user *uarg) } EXPORT_SYMBOL_GPL(fscrypt_ioctl_get_policy_ex); +/* FS_IOC_GET_ENCRYPTION_NONCE: retrieve file's encryption nonce for testing */ +int fscrypt_ioctl_get_nonce(struct file *filp, void __user *arg) +{ + struct inode *inode = file_inode(filp); + union fscrypt_context ctx; + int ret; + + ret = inode->i_sb->s_cop->get_context(inode, &ctx, sizeof(ctx)); + if (ret < 0) + return ret; + if (!fscrypt_context_is_valid(&ctx, ret)) + return -EINVAL; + if (copy_to_user(arg, fscrypt_context_nonce(&ctx), + FS_KEY_DERIVATION_NONCE_SIZE)) + return -EFAULT; + return 0; +} +EXPORT_SYMBOL_GPL(fscrypt_ioctl_get_nonce); + /** * fscrypt_has_permitted_context() - is a file's encryption policy permitted * within its directory? diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 634b09d18b77..2d357680094c 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -18,6 +18,7 @@ #include <linux/slab.h> #include <linux/atomic.h> #include <linux/device.h> +#include <linux/pm_runtime.h> #include <linux/poll.h> #include <linux/security.h> @@ -175,8 +176,13 @@ static int open_proxy_open(struct inode *inode, struct file *filp) if (r) goto out; - real_fops = fops_get(real_fops); - if (!real_fops) { + if (!fops_get(real_fops)) { +#ifdef MODULE + if (real_fops->owner && + real_fops->owner->state == MODULE_STATE_GOING) + goto out; +#endif + /* Huh? Module did not clean up after itself at exit? */ WARN(1, "debugfs file owner did not clean up at exit: %pd", dentry); @@ -305,8 +311,13 @@ static int full_proxy_open(struct inode *inode, struct file *filp) if (r) goto out; - real_fops = fops_get(real_fops); - if (!real_fops) { + if (!fops_get(real_fops)) { +#ifdef MODULE + if (real_fops->owner && + real_fops->owner->state == MODULE_STATE_GOING) + goto out; +#endif + /* Huh? Module did not cleanup after itself at exit? */ WARN(1, "debugfs file owner did not clean up at exit: %pd", dentry); @@ -1060,7 +1071,14 @@ static int debugfs_show_regset32(struct seq_file *s, void *data) { struct debugfs_regset32 *regset = s->private; + if (regset->dev) + pm_runtime_get_sync(regset->dev); + debugfs_print_regs32(s, regset->regs, regset->nregs, regset->base, ""); + + if (regset->dev) + pm_runtime_put(regset->dev); + return 0; } @@ -1090,21 +1108,12 @@ static const struct file_operations fops_regset32 = { * This function creates a file in debugfs with the given name that reports * the names and values of a set of 32-bit registers. If the @mode variable * is so set it can be read from. Writing is not supported. - * - * This function will return a pointer to a dentry if it succeeds. This - * pointer must be passed to the debugfs_remove() function when the file is - * to be removed (no automatic cleanup happens if your module is unloaded, - * you are responsible here.) If an error occurs, ERR_PTR(-ERROR) will be - * returned. - * - * If debugfs is not enabled in the kernel, the value ERR_PTR(-ENODEV) will - * be returned. */ -struct dentry *debugfs_create_regset32(const char *name, umode_t mode, - struct dentry *parent, - struct debugfs_regset32 *regset) +void debugfs_create_regset32(const char *name, umode_t mode, + struct dentry *parent, + struct debugfs_regset32 *regset) { - return debugfs_create_file(name, mode, parent, regset, &fops_regset32); + debugfs_create_file(name, mode, parent, regset, &fops_regset32); } EXPORT_SYMBOL_GPL(debugfs_create_regset32); diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index e742dfc66933..b7f2e971ecbc 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -501,26 +501,16 @@ EXPORT_SYMBOL_GPL(debugfs_create_file_unsafe); * wide range of flexibility in creating a file, or a directory (if you want * to create a directory, the debugfs_create_dir() function is * recommended to be used instead.) - * - * This function will return a pointer to a dentry if it succeeds. This - * pointer must be passed to the debugfs_remove() function when the file is - * to be removed (no automatic cleanup happens if your module is unloaded, - * you are responsible here.) If an error occurs, ERR_PTR(-ERROR) will be - * returned. - * - * If debugfs is not enabled in the kernel, the value -%ENODEV will be - * returned. */ -struct dentry *debugfs_create_file_size(const char *name, umode_t mode, - struct dentry *parent, void *data, - const struct file_operations *fops, - loff_t file_size) +void debugfs_create_file_size(const char *name, umode_t mode, + struct dentry *parent, void *data, + const struct file_operations *fops, + loff_t file_size) { struct dentry *de = debugfs_create_file(name, mode, parent, data, fops); if (de) d_inode(de)->i_size = file_size; - return de; } EXPORT_SYMBOL_GPL(debugfs_create_file_size); diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index db1ef144c63a..2c449aed1b92 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -311,8 +311,10 @@ static int crypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, struct extent_crypt_result ecr; int rc = 0; - BUG_ON(!crypt_stat || !crypt_stat->tfm - || !(crypt_stat->flags & ECRYPTFS_STRUCT_INITIALIZED)); + if (!crypt_stat || !crypt_stat->tfm + || !(crypt_stat->flags & ECRYPTFS_STRUCT_INITIALIZED)) + return -EINVAL; + if (unlikely(ecryptfs_verbosity > 0)) { ecryptfs_printk(KERN_DEBUG, "Key size [%zd]; key:\n", crypt_stat->key_size); diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 1c1a56be7ea2..e6ac78c62ca4 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -8,7 +8,7 @@ * Copyright (C) 2004-2008 International Business Machines Corp. * Author(s): Michael A. Halcrow <mahalcro@us.ibm.com> * Trevor S. Highland <trevor.highland@gmail.com> - * Tyler Hicks <tyhicks@ou.edu> + * Tyler Hicks <code@tyhicks.com> */ #ifndef ECRYPTFS_KERNEL_H diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index 7d326aa0308e..af3eb02bbca1 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -1304,7 +1304,7 @@ parse_tag_1_packet(struct ecryptfs_crypt_stat *crypt_stat, printk(KERN_WARNING "Tag 1 packet contains key larger " "than ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES\n"); rc = -EINVAL; - goto out; + goto out_free; } memcpy((*new_auth_tok)->session_key.encrypted_key, &data[(*packet_size)], (body_size - (ECRYPTFS_SIG_SIZE + 2))); diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index b8a7ce379ffe..e63259fdef28 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -7,7 +7,7 @@ * Copyright (C) 2004-2007 International Business Machines Corp. * Author(s): Michael A. Halcrow <mahalcro@us.ibm.com> * Michael C. Thompson <mcthomps@us.ibm.com> - * Tyler Hicks <tyhicks@ou.edu> + * Tyler Hicks <code@tyhicks.com> */ #include <linux/dcache.h> diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c index d668e60b85b5..8646ba76def3 100644 --- a/fs/ecryptfs/messaging.c +++ b/fs/ecryptfs/messaging.c @@ -4,7 +4,7 @@ * * Copyright (C) 2004-2008 International Business Machines Corp. * Author(s): Michael A. Halcrow <mhalcrow@us.ibm.com> - * Tyler Hicks <tyhicks@ou.edu> + * Tyler Hicks <code@tyhicks.com> */ #include <linux/sched.h> #include <linux/slab.h> @@ -379,6 +379,7 @@ int __init ecryptfs_init_messaging(void) * ecryptfs_message_buf_len), GFP_KERNEL); if (!ecryptfs_msg_ctx_arr) { + kfree(ecryptfs_daemon_hash); rc = -ENOMEM; goto out; } diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c index fa4f6447ddad..12c66f5d92dd 100644 --- a/fs/efivarfs/super.c +++ b/fs/efivarfs/super.c @@ -252,7 +252,7 @@ static struct file_system_type efivarfs_type = { static __init int efivarfs_init(void) { - if (!efi_enabled(EFI_RUNTIME_SERVICES)) + if (!efi_rt_services_supported(EFI_RT_SUPPORTED_VARIABLE_SERVICES)) return -ENODEV; if (!efivars_kobject()) diff --git a/fs/erofs/decompressor.c b/fs/erofs/decompressor.c index 5779a15c2cd6..5d2d81940679 100644 --- a/fs/erofs/decompressor.c +++ b/fs/erofs/decompressor.c @@ -157,17 +157,27 @@ static int z_erofs_lz4_decompress(struct z_erofs_decompress_req *rq, u8 *out) } } - ret = LZ4_decompress_safe_partial(src + inputmargin, out, - inlen, rq->outputsize, - rq->outputsize); - if (ret < 0) { - erofs_err(rq->sb, "failed to decompress, in[%u, %u] out[%u]", - inlen, inputmargin, rq->outputsize); + /* legacy format could compress extra data in a pcluster. */ + if (rq->partial_decoding || !support_0padding) + ret = LZ4_decompress_safe_partial(src + inputmargin, out, + inlen, rq->outputsize, + rq->outputsize); + else + ret = LZ4_decompress_safe(src + inputmargin, out, + inlen, rq->outputsize); + + if (ret != rq->outputsize) { + erofs_err(rq->sb, "failed to decompress %d in[%u, %u] out[%u]", + ret, inlen, inputmargin, rq->outputsize); + WARN_ON(1); print_hex_dump(KERN_DEBUG, "[ in]: ", DUMP_PREFIX_OFFSET, 16, 1, src + inputmargin, inlen, true); print_hex_dump(KERN_DEBUG, "[out]: ", DUMP_PREFIX_OFFSET, 16, 1, out, rq->outputsize, true); + + if (ret >= 0) + memset(out + ret, 0, rq->outputsize - ret); ret = -EIO; } diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index c4c6dcdc89ad..5eead7fdc7a6 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -52,8 +52,8 @@ struct erofs_sb_info { struct list_head list; struct mutex umount_mutex; - /* the dedicated workstation for compression */ - struct radix_tree_root workstn_tree; + /* managed XArray arranged in physical block number */ + struct xarray managed_pslots; /* threshold for decompression synchronously */ unsigned int max_sync_decompress_pages; @@ -402,8 +402,8 @@ static inline void *erofs_get_pcpubuf(unsigned int pagenr) int erofs_workgroup_put(struct erofs_workgroup *grp); struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb, pgoff_t index); -int erofs_register_workgroup(struct super_block *sb, - struct erofs_workgroup *grp); +struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb, + struct erofs_workgroup *grp); void erofs_workgroup_free_rcu(struct erofs_workgroup *grp); void erofs_shrinker_register(struct super_block *sb); void erofs_shrinker_unregister(struct super_block *sb); diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 057e6d7b5b7f..b514c67e5fc2 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -425,7 +425,7 @@ static int erofs_fill_super(struct super_block *sb, void *data, int silent) sb->s_flags &= ~SB_POSIXACL; #ifdef CONFIG_EROFS_FS_ZIP - INIT_RADIX_TREE(&sbi->workstn_tree, GFP_ATOMIC); + xa_init(&sbi->managed_pslots); #endif /* get the root inode */ diff --git a/fs/erofs/utils.c b/fs/erofs/utils.c index fddc5059c930..52d0be10f1aa 100644 --- a/fs/erofs/utils.c +++ b/fs/erofs/utils.c @@ -37,9 +37,6 @@ void *erofs_get_pcpubuf(unsigned int pagenr) /* global shrink count (for all mounted EROFS instances) */ static atomic_long_t erofs_global_shrink_cnt; -#define __erofs_workgroup_get(grp) atomic_inc(&(grp)->refcount) -#define __erofs_workgroup_put(grp) atomic_dec(&(grp)->refcount) - static int erofs_workgroup_get(struct erofs_workgroup *grp) { int o; @@ -66,7 +63,7 @@ struct erofs_workgroup *erofs_find_workgroup(struct super_block *sb, repeat: rcu_read_lock(); - grp = radix_tree_lookup(&sbi->workstn_tree, index); + grp = xa_load(&sbi->managed_pslots, index); if (grp) { if (erofs_workgroup_get(grp)) { /* prefer to relax rcu read side */ @@ -80,43 +77,37 @@ repeat: return grp; } -int erofs_register_workgroup(struct super_block *sb, - struct erofs_workgroup *grp) +struct erofs_workgroup *erofs_insert_workgroup(struct super_block *sb, + struct erofs_workgroup *grp) { - struct erofs_sb_info *sbi; - int err; - - /* grp shouldn't be broken or used before */ - if (atomic_read(&grp->refcount) != 1) { - DBG_BUGON(1); - return -EINVAL; - } - - err = radix_tree_preload(GFP_NOFS); - if (err) - return err; - - sbi = EROFS_SB(sb); - xa_lock(&sbi->workstn_tree); + struct erofs_sb_info *const sbi = EROFS_SB(sb); + struct erofs_workgroup *pre; /* - * Bump up reference count before making this workgroup - * visible to other users in order to avoid potential UAF - * without serialized by workstn_lock. + * Bump up a reference count before making this visible + * to others for the XArray in order to avoid potential + * UAF without serialized by xa_lock. */ - __erofs_workgroup_get(grp); - - err = radix_tree_insert(&sbi->workstn_tree, grp->index, grp); - if (err) - /* - * it's safe to decrease since the workgroup isn't visible - * and refcount >= 2 (cannot be freezed). - */ - __erofs_workgroup_put(grp); + atomic_inc(&grp->refcount); - xa_unlock(&sbi->workstn_tree); - radix_tree_preload_end(); - return err; +repeat: + xa_lock(&sbi->managed_pslots); + pre = __xa_cmpxchg(&sbi->managed_pslots, grp->index, + NULL, grp, GFP_NOFS); + if (pre) { + if (xa_is_err(pre)) { + pre = ERR_PTR(xa_err(pre)); + } else if (erofs_workgroup_get(pre)) { + /* try to legitimize the current in-tree one */ + xa_unlock(&sbi->managed_pslots); + cond_resched(); + goto repeat; + } + atomic_dec(&grp->refcount); + grp = pre; + } + xa_unlock(&sbi->managed_pslots); + return grp; } static void __erofs_workgroup_free(struct erofs_workgroup *grp) @@ -155,7 +146,7 @@ static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi, /* * Note that all cached pages should be unattached - * before deleted from the radix tree. Otherwise some + * before deleted from the XArray. Otherwise some * cached pages could be still attached to the orphan * old workgroup when the new one is available in the tree. */ @@ -169,7 +160,7 @@ static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi, * however in order to avoid some race conditions, add a * DBG_BUGON to observe this in advance. */ - DBG_BUGON(radix_tree_delete(&sbi->workstn_tree, grp->index) != grp); + DBG_BUGON(xa_erase(&sbi->managed_pslots, grp->index) != grp); /* * If managed cache is on, last refcount should indicate @@ -182,22 +173,11 @@ static bool erofs_try_to_release_workgroup(struct erofs_sb_info *sbi, static unsigned long erofs_shrink_workstation(struct erofs_sb_info *sbi, unsigned long nr_shrink) { - pgoff_t first_index = 0; - void *batch[PAGEVEC_SIZE]; + struct erofs_workgroup *grp; unsigned int freed = 0; + unsigned long index; - int i, found; -repeat: - xa_lock(&sbi->workstn_tree); - - found = radix_tree_gang_lookup(&sbi->workstn_tree, - batch, first_index, PAGEVEC_SIZE); - - for (i = 0; i < found; ++i) { - struct erofs_workgroup *grp = batch[i]; - - first_index = grp->index + 1; - + xa_for_each(&sbi->managed_pslots, index, grp) { /* try to shrink each valid workgroup */ if (!erofs_try_to_release_workgroup(sbi, grp)) continue; @@ -206,10 +186,6 @@ repeat: if (!--nr_shrink) break; } - xa_unlock(&sbi->workstn_tree); - - if (i && nr_shrink) - goto repeat; return freed; } @@ -286,7 +262,7 @@ static unsigned long erofs_shrink_scan(struct shrinker *shrink, spin_unlock(&erofs_sb_list_lock); sbi->shrinker_run_no = run_no; - freed += erofs_shrink_workstation(sbi, nr); + freed += erofs_shrink_workstation(sbi, nr - freed); spin_lock(&erofs_sb_list_lock); /* Get the next list element before we move this one */ diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 80e47f07d946..c4b6c9aa87ec 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -67,16 +67,6 @@ static void z_erofs_pcluster_init_once(void *ptr) pcl->compressed_pages[i] = NULL; } -static void z_erofs_pcluster_init_always(struct z_erofs_pcluster *pcl) -{ - struct z_erofs_collection *cl = z_erofs_primarycollection(pcl); - - atomic_set(&pcl->obj.refcount, 1); - - DBG_BUGON(cl->nr_pages); - DBG_BUGON(cl->vcnt); -} - int __init z_erofs_init_zip_subsystem(void) { pcluster_cachep = kmem_cache_create("erofs_compress", @@ -341,26 +331,19 @@ static int z_erofs_lookup_collection(struct z_erofs_collector *clt, struct inode *inode, struct erofs_map_blocks *map) { - struct erofs_workgroup *grp; - struct z_erofs_pcluster *pcl; + struct z_erofs_pcluster *pcl = clt->pcl; struct z_erofs_collection *cl; unsigned int length; - grp = erofs_find_workgroup(inode->i_sb, map->m_pa >> PAGE_SHIFT); - if (!grp) - return -ENOENT; - - pcl = container_of(grp, struct z_erofs_pcluster, obj); + /* to avoid unexpected loop formed by corrupted images */ if (clt->owned_head == &pcl->next || pcl == clt->tailpcl) { DBG_BUGON(1); - erofs_workgroup_put(grp); return -EFSCORRUPTED; } cl = z_erofs_primarycollection(pcl); if (cl->pageofs != (map->m_la & ~PAGE_MASK)) { DBG_BUGON(1); - erofs_workgroup_put(grp); return -EFSCORRUPTED; } @@ -368,7 +351,6 @@ static int z_erofs_lookup_collection(struct z_erofs_collector *clt, if (length & Z_EROFS_PCLUSTER_FULL_LENGTH) { if ((map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT) > length) { DBG_BUGON(1); - erofs_workgroup_put(grp); return -EFSCORRUPTED; } } else { @@ -391,7 +373,6 @@ static int z_erofs_lookup_collection(struct z_erofs_collector *clt, /* clean tailpcl if the current owned_head is Z_EROFS_PCLUSTER_TAIL */ if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL) clt->tailpcl = NULL; - clt->pcl = pcl; clt->cl = cl; return 0; } @@ -402,6 +383,7 @@ static int z_erofs_register_collection(struct z_erofs_collector *clt, { struct z_erofs_pcluster *pcl; struct z_erofs_collection *cl; + struct erofs_workgroup *grp; int err; /* no available workgroup, let's allocate one */ @@ -409,7 +391,7 @@ static int z_erofs_register_collection(struct z_erofs_collector *clt, if (!pcl) return -ENOMEM; - z_erofs_pcluster_init_always(pcl); + atomic_set(&pcl->obj.refcount, 1); pcl->obj.index = map->m_pa >> PAGE_SHIFT; pcl->length = (map->m_llen << Z_EROFS_PCLUSTER_LENGTH_BIT) | @@ -429,19 +411,29 @@ static int z_erofs_register_collection(struct z_erofs_collector *clt, clt->mode = COLLECT_PRIMARY_FOLLOWED; cl = z_erofs_primarycollection(pcl); + + /* must be cleaned before freeing to slab */ + DBG_BUGON(cl->nr_pages); + DBG_BUGON(cl->vcnt); + cl->pageofs = map->m_la & ~PAGE_MASK; /* * lock all primary followed works before visible to others * and mutex_trylock *never* fails for a new pcluster. */ - mutex_trylock(&cl->lock); + DBG_BUGON(!mutex_trylock(&cl->lock)); - err = erofs_register_workgroup(inode->i_sb, &pcl->obj); - if (err) { - mutex_unlock(&cl->lock); - kmem_cache_free(pcluster_cachep, pcl); - return -EAGAIN; + grp = erofs_insert_workgroup(inode->i_sb, &pcl->obj); + if (IS_ERR(grp)) { + err = PTR_ERR(grp); + goto err_out; + } + + if (grp != &pcl->obj) { + clt->pcl = container_of(grp, struct z_erofs_pcluster, obj); + err = -EEXIST; + goto err_out; } /* used to check tail merging loop due to corrupted images */ if (clt->owned_head == Z_EROFS_PCLUSTER_TAIL) @@ -450,12 +442,18 @@ static int z_erofs_register_collection(struct z_erofs_collector *clt, clt->pcl = pcl; clt->cl = cl; return 0; + +err_out: + mutex_unlock(&cl->lock); + kmem_cache_free(pcluster_cachep, pcl); + return err; } static int z_erofs_collector_begin(struct z_erofs_collector *clt, struct inode *inode, struct erofs_map_blocks *map) { + struct erofs_workgroup *grp; int ret; DBG_BUGON(clt->cl); @@ -469,21 +467,25 @@ static int z_erofs_collector_begin(struct z_erofs_collector *clt, return -EINVAL; } -repeat: - ret = z_erofs_lookup_collection(clt, inode, map); - if (ret == -ENOENT) { + grp = erofs_find_workgroup(inode->i_sb, map->m_pa >> PAGE_SHIFT); + if (grp) { + clt->pcl = container_of(grp, struct z_erofs_pcluster, obj); + } else { ret = z_erofs_register_collection(clt, inode, map); - /* someone registered at the same time, give another try */ - if (ret == -EAGAIN) { - cond_resched(); - goto repeat; - } + if (!ret) + goto out; + if (ret != -EEXIST) + return ret; } - if (ret) + ret = z_erofs_lookup_collection(clt, inode, map); + if (ret) { + erofs_workgroup_put(&clt->pcl->obj); return ret; + } +out: z_erofs_pagevec_ctor_init(&clt->vector, Z_EROFS_NR_INLINE_PAGEVECS, clt->cl->pagevec, clt->cl->vcnt); diff --git a/fs/eventpoll.c b/fs/eventpoll.c index b041b66002db..eee3c92a9ebf 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1854,9 +1854,9 @@ fetch_events: waiter = true; init_waitqueue_entry(&wait, current); - spin_lock_irq(&ep->wq.lock); + write_lock_irq(&ep->lock); __add_wait_queue_exclusive(&ep->wq, &wait); - spin_unlock_irq(&ep->wq.lock); + write_unlock_irq(&ep->lock); } for (;;) { @@ -1904,9 +1904,9 @@ send_events: goto fetch_events; if (waiter) { - spin_lock_irq(&ep->wq.lock); + write_lock_irq(&ep->lock); __remove_wait_queue(&ep->wq, &wait); - spin_unlock_irq(&ep->wq.lock); + write_unlock_irq(&ep->lock); } return res; diff --git a/fs/exec.c b/fs/exec.c index d55710a36056..06b4c550af5d 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -985,6 +985,32 @@ int kernel_read_file_from_path(const char *path, void **buf, loff_t *size, } EXPORT_SYMBOL_GPL(kernel_read_file_from_path); +int kernel_read_file_from_path_initns(const char *path, void **buf, + loff_t *size, loff_t max_size, + enum kernel_read_file_id id) +{ + struct file *file; + struct path root; + int ret; + + if (!path || !*path) + return -EINVAL; + + task_lock(&init_task); + get_fs_root(init_task.fs, &root); + task_unlock(&init_task); + + file = file_open_root(root.dentry, root.mnt, path, O_RDONLY, 0); + path_put(&root); + if (IS_ERR(file)) + return PTR_ERR(file); + + ret = kernel_read_file(file, buf, size, max_size, id); + fput(file); + return ret; +} +EXPORT_SYMBOL_GPL(kernel_read_file_from_path_initns); + int kernel_read_file_from_fd(int fd, void **buf, loff_t *size, loff_t max_size, enum kernel_read_file_id id) { diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index 5f993a411251..8fd0b3cdab4c 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -270,6 +270,7 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb, ext4_group_t ngroups = ext4_get_groups_count(sb); struct ext4_group_desc *desc; struct ext4_sb_info *sbi = EXT4_SB(sb); + struct buffer_head *bh_p; if (block_group >= ngroups) { ext4_error(sb, "block_group >= groups_count - block_group = %u," @@ -280,7 +281,14 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb, group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); - if (!sbi->s_group_desc[group_desc]) { + bh_p = sbi_array_rcu_deref(sbi, s_group_desc, group_desc); + /* + * sbi_array_rcu_deref returns with rcu unlocked, this is ok since + * the pointer being dereferenced won't be dereferenced again. By + * looking at the usage in add_new_gdb() the value isn't modified, + * just the pointer, and so it remains valid. + */ + if (!bh_p) { ext4_error(sb, "Group descriptor not loaded - " "block_group = %u, group_desc = %u, desc = %u", block_group, group_desc, offset); @@ -288,10 +296,10 @@ struct ext4_group_desc * ext4_get_group_desc(struct super_block *sb, } desc = (struct ext4_group_desc *)( - (__u8 *)sbi->s_group_desc[group_desc]->b_data + + (__u8 *)bh_p->b_data + offset * EXT4_DESC_SIZE(sb)); if (bh) - *bh = sbi->s_group_desc[group_desc]; + *bh = bh_p; return desc; } diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 4441331d06cc..61b37a052052 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1400,7 +1400,7 @@ struct ext4_sb_info { loff_t s_bitmap_maxbytes; /* max bytes for bitmap files */ struct buffer_head * s_sbh; /* Buffer containing the super block */ struct ext4_super_block *s_es; /* Pointer to the super block in the buffer */ - struct buffer_head **s_group_desc; + struct buffer_head * __rcu *s_group_desc; unsigned int s_mount_opt; unsigned int s_mount_opt2; unsigned int s_mount_flags; @@ -1462,7 +1462,7 @@ struct ext4_sb_info { #endif /* for buddy allocator */ - struct ext4_group_info ***s_group_info; + struct ext4_group_info ** __rcu *s_group_info; struct inode *s_buddy_cache; spinlock_t s_md_lock; unsigned short *s_mb_offsets; @@ -1512,7 +1512,7 @@ struct ext4_sb_info { unsigned int s_extent_max_zeroout_kb; unsigned int s_log_groups_per_flex; - struct flex_groups *s_flex_groups; + struct flex_groups * __rcu *s_flex_groups; ext4_group_t s_flex_groups_allocated; /* workqueue for reserved extent conversions (buffered io) */ @@ -1552,8 +1552,11 @@ struct ext4_sb_info { struct ratelimit_state s_warning_ratelimit_state; struct ratelimit_state s_msg_ratelimit_state; - /* Barrier between changing inodes' journal flags and writepages ops. */ - struct percpu_rw_semaphore s_journal_flag_rwsem; + /* + * Barrier between writepages ops and changing any inode's JOURNAL_DATA + * or EXTENTS flag. + */ + struct percpu_rw_semaphore s_writepages_rwsem; struct dax_device *s_daxdev; #ifdef CONFIG_EXT4_DEBUG unsigned long s_simulate_fail; @@ -1577,6 +1580,23 @@ static inline int ext4_valid_inum(struct super_block *sb, unsigned long ino) } /* + * Returns: sbi->field[index] + * Used to access an array element from the following sbi fields which require + * rcu protection to avoid dereferencing an invalid pointer due to reassignment + * - s_group_desc + * - s_group_info + * - s_flex_group + */ +#define sbi_array_rcu_deref(sbi, field, index) \ +({ \ + typeof(*((sbi)->field)) _v; \ + rcu_read_lock(); \ + _v = ((typeof(_v)*)rcu_dereference((sbi)->field))[index]; \ + rcu_read_unlock(); \ + _v; \ +}) + +/* * Simulate_fail codes */ #define EXT4_SIM_BBITMAP_EIO 1 @@ -2730,6 +2750,7 @@ extern int ext4_generic_delete_entry(handle_t *handle, extern bool ext4_empty_dir(struct inode *inode); /* resize.c */ +extern void ext4_kvfree_array_rcu(void *to_free); extern int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input); extern int ext4_group_extend(struct super_block *sb, @@ -2976,13 +2997,13 @@ static inline struct ext4_group_info *ext4_get_group_info(struct super_block *sb, ext4_group_t group) { - struct ext4_group_info ***grp_info; + struct ext4_group_info **grp_info; long indexv, indexh; BUG_ON(group >= EXT4_SB(sb)->s_groups_count); - grp_info = EXT4_SB(sb)->s_group_info; indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb)); indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1); - return grp_info[indexv][indexh]; + grp_info = sbi_array_rcu_deref(EXT4_SB(sb), s_group_info, indexv); + return grp_info[indexh]; } /* @@ -3032,7 +3053,7 @@ static inline void ext4_update_i_disksize(struct inode *inode, loff_t newsize) !inode_is_locked(inode)); down_write(&EXT4_I(inode)->i_data_sem); if (newsize > EXT4_I(inode)->i_disksize) - EXT4_I(inode)->i_disksize = newsize; + WRITE_ONCE(EXT4_I(inode)->i_disksize, newsize); up_write(&EXT4_I(inode)->i_data_sem); } diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index c66e8f9451a2..f95ee99091e4 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -328,11 +328,13 @@ void ext4_free_inode(handle_t *handle, struct inode *inode) percpu_counter_inc(&sbi->s_freeinodes_counter); if (sbi->s_log_groups_per_flex) { - ext4_group_t f = ext4_flex_group(sbi, block_group); + struct flex_groups *fg; - atomic_inc(&sbi->s_flex_groups[f].free_inodes); + fg = sbi_array_rcu_deref(sbi, s_flex_groups, + ext4_flex_group(sbi, block_group)); + atomic_inc(&fg->free_inodes); if (is_directory) - atomic_dec(&sbi->s_flex_groups[f].used_dirs); + atomic_dec(&fg->used_dirs); } BUFFER_TRACE(bh2, "call ext4_handle_dirty_metadata"); fatal = ext4_handle_dirty_metadata(handle, NULL, bh2); @@ -368,12 +370,13 @@ static void get_orlov_stats(struct super_block *sb, ext4_group_t g, int flex_size, struct orlov_stats *stats) { struct ext4_group_desc *desc; - struct flex_groups *flex_group = EXT4_SB(sb)->s_flex_groups; if (flex_size > 1) { - stats->free_inodes = atomic_read(&flex_group[g].free_inodes); - stats->free_clusters = atomic64_read(&flex_group[g].free_clusters); - stats->used_dirs = atomic_read(&flex_group[g].used_dirs); + struct flex_groups *fg = sbi_array_rcu_deref(EXT4_SB(sb), + s_flex_groups, g); + stats->free_inodes = atomic_read(&fg->free_inodes); + stats->free_clusters = atomic64_read(&fg->free_clusters); + stats->used_dirs = atomic_read(&fg->used_dirs); return; } @@ -1054,7 +1057,8 @@ got: if (sbi->s_log_groups_per_flex) { ext4_group_t f = ext4_flex_group(sbi, group); - atomic_inc(&sbi->s_flex_groups[f].used_dirs); + atomic_inc(&sbi_array_rcu_deref(sbi, s_flex_groups, + f)->used_dirs); } } if (ext4_has_group_desc_csum(sb)) { @@ -1077,7 +1081,8 @@ got: if (sbi->s_log_groups_per_flex) { flex_group = ext4_flex_group(sbi, group); - atomic_dec(&sbi->s_flex_groups[flex_group].free_inodes); + atomic_dec(&sbi_array_rcu_deref(sbi, s_flex_groups, + flex_group)->free_inodes); } inode->i_ino = ino + group * EXT4_INODES_PER_GROUP(sb); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index e60aca791d3f..fa0ff78dc033 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2465,7 +2465,7 @@ update_disksize: * truncate are avoided by checking i_size under i_data_sem. */ disksize = ((loff_t)mpd->first_page) << PAGE_SHIFT; - if (disksize > EXT4_I(inode)->i_disksize) { + if (disksize > READ_ONCE(EXT4_I(inode)->i_disksize)) { int err2; loff_t i_size; @@ -2628,7 +2628,7 @@ static int ext4_writepages(struct address_space *mapping, if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; - percpu_down_read(&sbi->s_journal_flag_rwsem); + percpu_down_read(&sbi->s_writepages_rwsem); trace_ext4_writepages(inode, wbc); /* @@ -2849,7 +2849,7 @@ unplug: out_writepages: trace_ext4_writepages_result(inode, wbc, ret, nr_to_write - wbc->nr_to_write); - percpu_up_read(&sbi->s_journal_flag_rwsem); + percpu_up_read(&sbi->s_writepages_rwsem); return ret; } @@ -2864,13 +2864,13 @@ static int ext4_dax_writepages(struct address_space *mapping, if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) return -EIO; - percpu_down_read(&sbi->s_journal_flag_rwsem); + percpu_down_read(&sbi->s_writepages_rwsem); trace_ext4_writepages(inode, wbc); ret = dax_writeback_mapping_range(mapping, sbi->s_daxdev, wbc); trace_ext4_writepages_result(inode, wbc, ret, nr_to_write - wbc->nr_to_write); - percpu_up_read(&sbi->s_journal_flag_rwsem); + percpu_up_read(&sbi->s_writepages_rwsem); return ret; } @@ -5861,7 +5861,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) } } - percpu_down_write(&sbi->s_journal_flag_rwsem); + percpu_down_write(&sbi->s_writepages_rwsem); jbd2_journal_lock_updates(journal); /* @@ -5878,7 +5878,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) err = jbd2_journal_flush(journal); if (err < 0) { jbd2_journal_unlock_updates(journal); - percpu_up_write(&sbi->s_journal_flag_rwsem); + percpu_up_write(&sbi->s_writepages_rwsem); return err; } ext4_clear_inode_flag(inode, EXT4_INODE_JOURNAL_DATA); @@ -5886,7 +5886,7 @@ int ext4_change_inode_journal_flag(struct inode *inode, int val) ext4_set_aops(inode); jbd2_journal_unlock_updates(journal); - percpu_up_write(&sbi->s_journal_flag_rwsem); + percpu_up_write(&sbi->s_writepages_rwsem); if (val) up_write(&EXT4_I(inode)->i_mmap_sem); diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index a0ec750018dd..0c1d1720cf1a 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -1210,6 +1210,11 @@ resizefs_out: return -EOPNOTSUPP; return fscrypt_ioctl_get_key_status(filp, (void __user *)arg); + case FS_IOC_GET_ENCRYPTION_NONCE: + if (!ext4_has_feature_encrypt(sb)) + return -EOPNOTSUPP; + return fscrypt_ioctl_get_nonce(filp, (void __user *)arg); + case EXT4_IOC_CLEAR_ES_CACHE: { if (!inode_owner_or_capable(inode)) @@ -1370,6 +1375,7 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case FS_IOC_REMOVE_ENCRYPTION_KEY: case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS: case FS_IOC_GET_ENCRYPTION_KEY_STATUS: + case FS_IOC_GET_ENCRYPTION_NONCE: case EXT4_IOC_SHUTDOWN: case FS_IOC_GETFSMAP: case FS_IOC_ENABLE_VERITY: diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index f64838187559..51a78eb65f3c 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2356,7 +2356,7 @@ int ext4_mb_alloc_groupinfo(struct super_block *sb, ext4_group_t ngroups) { struct ext4_sb_info *sbi = EXT4_SB(sb); unsigned size; - struct ext4_group_info ***new_groupinfo; + struct ext4_group_info ***old_groupinfo, ***new_groupinfo; size = (ngroups + EXT4_DESC_PER_BLOCK(sb) - 1) >> EXT4_DESC_PER_BLOCK_BITS(sb); @@ -2369,13 +2369,16 @@ int ext4_mb_alloc_groupinfo(struct super_block *sb, ext4_group_t ngroups) ext4_msg(sb, KERN_ERR, "can't allocate buddy meta group"); return -ENOMEM; } - if (sbi->s_group_info) { - memcpy(new_groupinfo, sbi->s_group_info, + rcu_read_lock(); + old_groupinfo = rcu_dereference(sbi->s_group_info); + if (old_groupinfo) + memcpy(new_groupinfo, old_groupinfo, sbi->s_group_info_size * sizeof(*sbi->s_group_info)); - kvfree(sbi->s_group_info); - } - sbi->s_group_info = new_groupinfo; + rcu_read_unlock(); + rcu_assign_pointer(sbi->s_group_info, new_groupinfo); sbi->s_group_info_size = size / sizeof(*sbi->s_group_info); + if (old_groupinfo) + ext4_kvfree_array_rcu(old_groupinfo); ext4_debug("allocated s_groupinfo array for %d meta_bg's\n", sbi->s_group_info_size); return 0; @@ -2387,6 +2390,7 @@ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group, { int i; int metalen = 0; + int idx = group >> EXT4_DESC_PER_BLOCK_BITS(sb); struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_group_info **meta_group_info; struct kmem_cache *cachep = get_groupinfo_cache(sb->s_blocksize_bits); @@ -2405,12 +2409,12 @@ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group, "for a buddy group"); goto exit_meta_group_info; } - sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)] = - meta_group_info; + rcu_read_lock(); + rcu_dereference(sbi->s_group_info)[idx] = meta_group_info; + rcu_read_unlock(); } - meta_group_info = - sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)]; + meta_group_info = sbi_array_rcu_deref(sbi, s_group_info, idx); i = group & (EXT4_DESC_PER_BLOCK(sb) - 1); meta_group_info[i] = kmem_cache_zalloc(cachep, GFP_NOFS); @@ -2458,8 +2462,13 @@ int ext4_mb_add_groupinfo(struct super_block *sb, ext4_group_t group, exit_group_info: /* If a meta_group_info table has been allocated, release it now */ if (group % EXT4_DESC_PER_BLOCK(sb) == 0) { - kfree(sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)]); - sbi->s_group_info[group >> EXT4_DESC_PER_BLOCK_BITS(sb)] = NULL; + struct ext4_group_info ***group_info; + + rcu_read_lock(); + group_info = rcu_dereference(sbi->s_group_info); + kfree(group_info[idx]); + group_info[idx] = NULL; + rcu_read_unlock(); } exit_meta_group_info: return -ENOMEM; @@ -2472,6 +2481,7 @@ static int ext4_mb_init_backend(struct super_block *sb) struct ext4_sb_info *sbi = EXT4_SB(sb); int err; struct ext4_group_desc *desc; + struct ext4_group_info ***group_info; struct kmem_cache *cachep; err = ext4_mb_alloc_groupinfo(sb, ngroups); @@ -2507,11 +2517,16 @@ err_freebuddy: while (i-- > 0) kmem_cache_free(cachep, ext4_get_group_info(sb, i)); i = sbi->s_group_info_size; + rcu_read_lock(); + group_info = rcu_dereference(sbi->s_group_info); while (i-- > 0) - kfree(sbi->s_group_info[i]); + kfree(group_info[i]); + rcu_read_unlock(); iput(sbi->s_buddy_cache); err_freesgi: - kvfree(sbi->s_group_info); + rcu_read_lock(); + kvfree(rcu_dereference(sbi->s_group_info)); + rcu_read_unlock(); return -ENOMEM; } @@ -2700,7 +2715,7 @@ int ext4_mb_release(struct super_block *sb) ext4_group_t ngroups = ext4_get_groups_count(sb); ext4_group_t i; int num_meta_group_infos; - struct ext4_group_info *grinfo; + struct ext4_group_info *grinfo, ***group_info; struct ext4_sb_info *sbi = EXT4_SB(sb); struct kmem_cache *cachep = get_groupinfo_cache(sb->s_blocksize_bits); @@ -2719,9 +2734,12 @@ int ext4_mb_release(struct super_block *sb) num_meta_group_infos = (ngroups + EXT4_DESC_PER_BLOCK(sb) - 1) >> EXT4_DESC_PER_BLOCK_BITS(sb); + rcu_read_lock(); + group_info = rcu_dereference(sbi->s_group_info); for (i = 0; i < num_meta_group_infos; i++) - kfree(sbi->s_group_info[i]); - kvfree(sbi->s_group_info); + kfree(group_info[i]); + kvfree(group_info); + rcu_read_unlock(); } kfree(sbi->s_mb_offsets); kfree(sbi->s_mb_maxs); @@ -3020,7 +3038,8 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, ext4_group_t flex_group = ext4_flex_group(sbi, ac->ac_b_ex.fe_group); atomic64_sub(ac->ac_b_ex.fe_len, - &sbi->s_flex_groups[flex_group].free_clusters); + &sbi_array_rcu_deref(sbi, s_flex_groups, + flex_group)->free_clusters); } err = ext4_handle_dirty_metadata(handle, NULL, bitmap_bh); @@ -4918,7 +4937,8 @@ do_more: if (sbi->s_log_groups_per_flex) { ext4_group_t flex_group = ext4_flex_group(sbi, block_group); atomic64_add(count_clusters, - &sbi->s_flex_groups[flex_group].free_clusters); + &sbi_array_rcu_deref(sbi, s_flex_groups, + flex_group)->free_clusters); } /* @@ -5075,7 +5095,8 @@ int ext4_group_add_blocks(handle_t *handle, struct super_block *sb, if (sbi->s_log_groups_per_flex) { ext4_group_t flex_group = ext4_flex_group(sbi, block_group); atomic64_add(clusters_freed, - &sbi->s_flex_groups[flex_group].free_clusters); + &sbi_array_rcu_deref(sbi, s_flex_groups, + flex_group)->free_clusters); } ext4_mb_unload_buddy(&e4b); diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index 89725fa42573..fb6520f37135 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -407,6 +407,7 @@ static int free_ext_block(handle_t *handle, struct inode *inode) int ext4_ext_migrate(struct inode *inode) { + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); handle_t *handle; int retval = 0, i; __le32 *i_data; @@ -431,6 +432,8 @@ int ext4_ext_migrate(struct inode *inode) */ return retval; + percpu_down_write(&sbi->s_writepages_rwsem); + /* * Worst case we can touch the allocation bitmaps, a bgd * block, and a block to link in the orphan list. We do need @@ -441,7 +444,7 @@ int ext4_ext_migrate(struct inode *inode) if (IS_ERR(handle)) { retval = PTR_ERR(handle); - return retval; + goto out_unlock; } goal = (((inode->i_ino - 1) / EXT4_INODES_PER_GROUP(inode->i_sb)) * EXT4_INODES_PER_GROUP(inode->i_sb)) + 1; @@ -452,7 +455,7 @@ int ext4_ext_migrate(struct inode *inode) if (IS_ERR(tmp_inode)) { retval = PTR_ERR(tmp_inode); ext4_journal_stop(handle); - return retval; + goto out_unlock; } i_size_write(tmp_inode, i_size_read(inode)); /* @@ -494,7 +497,7 @@ int ext4_ext_migrate(struct inode *inode) */ ext4_orphan_del(NULL, tmp_inode); retval = PTR_ERR(handle); - goto out; + goto out_tmp_inode; } ei = EXT4_I(inode); @@ -576,10 +579,11 @@ err_out: ext4_ext_tree_init(handle, tmp_inode); out_stop: ext4_journal_stop(handle); -out: +out_tmp_inode: unlock_new_inode(tmp_inode); iput(tmp_inode); - +out_unlock: + percpu_up_write(&sbi->s_writepages_rwsem); return retval; } @@ -589,7 +593,8 @@ out: int ext4_ind_migrate(struct inode *inode) { struct ext4_extent_header *eh; - struct ext4_super_block *es = EXT4_SB(inode->i_sb)->s_es; + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + struct ext4_super_block *es = sbi->s_es; struct ext4_inode_info *ei = EXT4_I(inode); struct ext4_extent *ex; unsigned int i, len; @@ -613,9 +618,13 @@ int ext4_ind_migrate(struct inode *inode) if (test_opt(inode->i_sb, DELALLOC)) ext4_alloc_da_blocks(inode); + percpu_down_write(&sbi->s_writepages_rwsem); + handle = ext4_journal_start(inode, EXT4_HT_MIGRATE, 1); - if (IS_ERR(handle)) - return PTR_ERR(handle); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_unlock; + } down_write(&EXT4_I(inode)->i_data_sem); ret = ext4_ext_check_inode(inode); @@ -650,5 +659,7 @@ int ext4_ind_migrate(struct inode *inode) errout: ext4_journal_stop(handle); up_write(&EXT4_I(inode)->i_data_sem); +out_unlock: + percpu_up_write(&sbi->s_writepages_rwsem); return ret; } diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index ceff4b4b1877..b05ea72f38fd 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1511,6 +1511,7 @@ restart: /* * We deal with the read-ahead logic here. */ + cond_resched(); if (ra_ptr >= ra_max) { /* Refill the readahead buffer */ ra_ptr = 0; diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 68b39e75446a..de6fe969f773 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -125,11 +125,10 @@ static void ext4_finish_bio(struct bio *bio) } bh = head = page_buffers(page); /* - * We check all buffers in the page under BH_Uptodate_Lock + * We check all buffers in the page under b_uptodate_lock * to avoid races with other end io clearing async_write flags */ - local_irq_save(flags); - bit_spin_lock(BH_Uptodate_Lock, &head->b_state); + spin_lock_irqsave(&head->b_uptodate_lock, flags); do { if (bh_offset(bh) < bio_start || bh_offset(bh) + bh->b_size > bio_end) { @@ -141,8 +140,7 @@ static void ext4_finish_bio(struct bio *bio) if (bio->bi_status) buffer_io_error(bh); } while ((bh = bh->b_this_page) != head); - bit_spin_unlock(BH_Uptodate_Lock, &head->b_state); - local_irq_restore(flags); + spin_unlock_irqrestore(&head->b_uptodate_lock, flags); if (!under_io) { fscrypt_free_bounce_page(bounce_page); end_page_writeback(page); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 86a2500ed292..a50b51270ea9 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -17,6 +17,33 @@ #include "ext4_jbd2.h" +struct ext4_rcu_ptr { + struct rcu_head rcu; + void *ptr; +}; + +static void ext4_rcu_ptr_callback(struct rcu_head *head) +{ + struct ext4_rcu_ptr *ptr; + + ptr = container_of(head, struct ext4_rcu_ptr, rcu); + kvfree(ptr->ptr); + kfree(ptr); +} + +void ext4_kvfree_array_rcu(void *to_free) +{ + struct ext4_rcu_ptr *ptr = kzalloc(sizeof(*ptr), GFP_KERNEL); + + if (ptr) { + ptr->ptr = to_free; + call_rcu(&ptr->rcu, ext4_rcu_ptr_callback); + return; + } + synchronize_rcu(); + kvfree(to_free); +} + int ext4_resize_begin(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); @@ -542,8 +569,8 @@ static int setup_new_flex_group_blocks(struct super_block *sb, brelse(gdb); goto out; } - memcpy(gdb->b_data, sbi->s_group_desc[j]->b_data, - gdb->b_size); + memcpy(gdb->b_data, sbi_array_rcu_deref(sbi, + s_group_desc, j)->b_data, gdb->b_size); set_buffer_uptodate(gdb); err = ext4_handle_dirty_metadata(handle, NULL, gdb); @@ -860,13 +887,15 @@ static int add_new_gdb(handle_t *handle, struct inode *inode, } brelse(dind); - o_group_desc = EXT4_SB(sb)->s_group_desc; + rcu_read_lock(); + o_group_desc = rcu_dereference(EXT4_SB(sb)->s_group_desc); memcpy(n_group_desc, o_group_desc, EXT4_SB(sb)->s_gdb_count * sizeof(struct buffer_head *)); + rcu_read_unlock(); n_group_desc[gdb_num] = gdb_bh; - EXT4_SB(sb)->s_group_desc = n_group_desc; + rcu_assign_pointer(EXT4_SB(sb)->s_group_desc, n_group_desc); EXT4_SB(sb)->s_gdb_count++; - kvfree(o_group_desc); + ext4_kvfree_array_rcu(o_group_desc); le16_add_cpu(&es->s_reserved_gdt_blocks, -1); err = ext4_handle_dirty_super(handle, sb); @@ -909,9 +938,11 @@ static int add_new_gdb_meta_bg(struct super_block *sb, return err; } - o_group_desc = EXT4_SB(sb)->s_group_desc; + rcu_read_lock(); + o_group_desc = rcu_dereference(EXT4_SB(sb)->s_group_desc); memcpy(n_group_desc, o_group_desc, EXT4_SB(sb)->s_gdb_count * sizeof(struct buffer_head *)); + rcu_read_unlock(); n_group_desc[gdb_num] = gdb_bh; BUFFER_TRACE(gdb_bh, "get_write_access"); @@ -922,9 +953,9 @@ static int add_new_gdb_meta_bg(struct super_block *sb, return err; } - EXT4_SB(sb)->s_group_desc = n_group_desc; + rcu_assign_pointer(EXT4_SB(sb)->s_group_desc, n_group_desc); EXT4_SB(sb)->s_gdb_count++; - kvfree(o_group_desc); + ext4_kvfree_array_rcu(o_group_desc); return err; } @@ -1188,7 +1219,8 @@ static int ext4_add_new_descs(handle_t *handle, struct super_block *sb, * use non-sparse filesystems anymore. This is already checked above. */ if (gdb_off) { - gdb_bh = sbi->s_group_desc[gdb_num]; + gdb_bh = sbi_array_rcu_deref(sbi, s_group_desc, + gdb_num); BUFFER_TRACE(gdb_bh, "get_write_access"); err = ext4_journal_get_write_access(handle, gdb_bh); @@ -1270,7 +1302,7 @@ static int ext4_setup_new_descs(handle_t *handle, struct super_block *sb, /* * get_write_access() has been called on gdb_bh by ext4_add_new_desc(). */ - gdb_bh = sbi->s_group_desc[gdb_num]; + gdb_bh = sbi_array_rcu_deref(sbi, s_group_desc, gdb_num); /* Update group descriptor block for new group */ gdp = (struct ext4_group_desc *)(gdb_bh->b_data + gdb_off * EXT4_DESC_SIZE(sb)); @@ -1398,11 +1430,14 @@ static void ext4_update_super(struct super_block *sb, percpu_counter_read(&sbi->s_freeclusters_counter)); if (ext4_has_feature_flex_bg(sb) && sbi->s_log_groups_per_flex) { ext4_group_t flex_group; + struct flex_groups *fg; + flex_group = ext4_flex_group(sbi, group_data[0].group); + fg = sbi_array_rcu_deref(sbi, s_flex_groups, flex_group); atomic64_add(EXT4_NUM_B2C(sbi, free_blocks), - &sbi->s_flex_groups[flex_group].free_clusters); + &fg->free_clusters); atomic_add(EXT4_INODES_PER_GROUP(sb) * flex_gd->count, - &sbi->s_flex_groups[flex_group].free_inodes); + &fg->free_inodes); } /* @@ -1497,7 +1532,8 @@ exit_journal: for (; gdb_num <= gdb_num_end; gdb_num++) { struct buffer_head *gdb_bh; - gdb_bh = sbi->s_group_desc[gdb_num]; + gdb_bh = sbi_array_rcu_deref(sbi, s_group_desc, + gdb_num); if (old_gdb == gdb_bh->b_blocknr) continue; update_backups(sb, gdb_bh->b_blocknr, gdb_bh->b_data, diff --git a/fs/ext4/super.c b/fs/ext4/super.c index f464dff09774..c8dff4c68141 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -43,7 +43,7 @@ #include <linux/uaccess.h> #include <linux/iversion.h> #include <linux/unicode.h> - +#include <linux/part_stat.h> #include <linux/kthread.h> #include <linux/freezer.h> @@ -927,7 +927,6 @@ void ext4_update_dynamic_rev(struct super_block *sb) static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb) { struct block_device *bdev; - char b[BDEVNAME_SIZE]; bdev = blkdev_get_by_dev(dev, FMODE_READ|FMODE_WRITE|FMODE_EXCL, sb); if (IS_ERR(bdev)) @@ -935,8 +934,9 @@ static struct block_device *ext4_blkdev_get(dev_t dev, struct super_block *sb) return bdev; fail: - ext4_msg(sb, KERN_ERR, "failed to open journal device %s: %ld", - __bdevname(dev, b), PTR_ERR(bdev)); + ext4_msg(sb, KERN_ERR, + "failed to open journal device unknown-block(%u,%u) %ld", + MAJOR(dev), MINOR(dev), PTR_ERR(bdev)); return NULL; } @@ -1014,6 +1014,8 @@ static void ext4_put_super(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_super_block *es = sbi->s_es; + struct buffer_head **group_desc; + struct flex_groups **flex_groups; int aborted = 0; int i, err; @@ -1046,15 +1048,23 @@ static void ext4_put_super(struct super_block *sb) if (!sb_rdonly(sb)) ext4_commit_super(sb, 1); + rcu_read_lock(); + group_desc = rcu_dereference(sbi->s_group_desc); for (i = 0; i < sbi->s_gdb_count; i++) - brelse(sbi->s_group_desc[i]); - kvfree(sbi->s_group_desc); - kvfree(sbi->s_flex_groups); + brelse(group_desc[i]); + kvfree(group_desc); + flex_groups = rcu_dereference(sbi->s_flex_groups); + if (flex_groups) { + for (i = 0; i < sbi->s_flex_groups_allocated; i++) + kvfree(flex_groups[i]); + kvfree(flex_groups); + } + rcu_read_unlock(); percpu_counter_destroy(&sbi->s_freeclusters_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); percpu_counter_destroy(&sbi->s_dirtyclusters_counter); - percpu_free_rwsem(&sbi->s_journal_flag_rwsem); + percpu_free_rwsem(&sbi->s_writepages_rwsem); #ifdef CONFIG_QUOTA for (i = 0; i < EXT4_MAXQUOTAS; i++) kfree(get_qf_name(sb, sbi, i)); @@ -2380,8 +2390,8 @@ done: int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup) { struct ext4_sb_info *sbi = EXT4_SB(sb); - struct flex_groups *new_groups; - int size; + struct flex_groups **old_groups, **new_groups; + int size, i, j; if (!sbi->s_log_groups_per_flex) return 0; @@ -2390,22 +2400,37 @@ int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup) if (size <= sbi->s_flex_groups_allocated) return 0; - size = roundup_pow_of_two(size * sizeof(struct flex_groups)); - new_groups = kvzalloc(size, GFP_KERNEL); + new_groups = kvzalloc(roundup_pow_of_two(size * + sizeof(*sbi->s_flex_groups)), GFP_KERNEL); if (!new_groups) { - ext4_msg(sb, KERN_ERR, "not enough memory for %d flex groups", - size / (int) sizeof(struct flex_groups)); + ext4_msg(sb, KERN_ERR, + "not enough memory for %d flex group pointers", size); return -ENOMEM; } - - if (sbi->s_flex_groups) { - memcpy(new_groups, sbi->s_flex_groups, - (sbi->s_flex_groups_allocated * - sizeof(struct flex_groups))); - kvfree(sbi->s_flex_groups); + for (i = sbi->s_flex_groups_allocated; i < size; i++) { + new_groups[i] = kvzalloc(roundup_pow_of_two( + sizeof(struct flex_groups)), + GFP_KERNEL); + if (!new_groups[i]) { + for (j = sbi->s_flex_groups_allocated; j < i; j++) + kvfree(new_groups[j]); + kvfree(new_groups); + ext4_msg(sb, KERN_ERR, + "not enough memory for %d flex groups", size); + return -ENOMEM; + } } - sbi->s_flex_groups = new_groups; - sbi->s_flex_groups_allocated = size / sizeof(struct flex_groups); + rcu_read_lock(); + old_groups = rcu_dereference(sbi->s_flex_groups); + if (old_groups) + memcpy(new_groups, old_groups, + (sbi->s_flex_groups_allocated * + sizeof(struct flex_groups *))); + rcu_read_unlock(); + rcu_assign_pointer(sbi->s_flex_groups, new_groups); + sbi->s_flex_groups_allocated = size; + if (old_groups) + ext4_kvfree_array_rcu(old_groups); return 0; } @@ -2413,6 +2438,7 @@ static int ext4_fill_flex_info(struct super_block *sb) { struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_group_desc *gdp = NULL; + struct flex_groups *fg; ext4_group_t flex_group; int i, err; @@ -2430,12 +2456,11 @@ static int ext4_fill_flex_info(struct super_block *sb) gdp = ext4_get_group_desc(sb, i, NULL); flex_group = ext4_flex_group(sbi, i); - atomic_add(ext4_free_inodes_count(sb, gdp), - &sbi->s_flex_groups[flex_group].free_inodes); + fg = sbi_array_rcu_deref(sbi, s_flex_groups, flex_group); + atomic_add(ext4_free_inodes_count(sb, gdp), &fg->free_inodes); atomic64_add(ext4_free_group_clusters(sb, gdp), - &sbi->s_flex_groups[flex_group].free_clusters); - atomic_add(ext4_used_dirs_count(sb, gdp), - &sbi->s_flex_groups[flex_group].used_dirs); + &fg->free_clusters); + atomic_add(ext4_used_dirs_count(sb, gdp), &fg->used_dirs); } return 1; @@ -3009,7 +3034,7 @@ static int ext4_feature_set_ok(struct super_block *sb, int readonly) return 0; } -#if !defined(CONFIG_QUOTA) || !defined(CONFIG_QFMT_V2) +#if !IS_ENABLED(CONFIG_QUOTA) || !IS_ENABLED(CONFIG_QFMT_V2) if (!readonly && (ext4_has_feature_quota(sb) || ext4_has_feature_project(sb))) { ext4_msg(sb, KERN_ERR, @@ -3634,9 +3659,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) { struct dax_device *dax_dev = fs_dax_get_by_bdev(sb->s_bdev); char *orig_data = kstrdup(data, GFP_KERNEL); - struct buffer_head *bh; + struct buffer_head *bh, **group_desc; struct ext4_super_block *es = NULL; struct ext4_sb_info *sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); + struct flex_groups **flex_groups; ext4_fsblk_t block; ext4_fsblk_t sb_block = get_sb_block(&data); ext4_fsblk_t logical_sb_block; @@ -4290,9 +4316,10 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount; } } - sbi->s_group_desc = kvmalloc_array(db_count, - sizeof(struct buffer_head *), - GFP_KERNEL); + rcu_assign_pointer(sbi->s_group_desc, + kvmalloc_array(db_count, + sizeof(struct buffer_head *), + GFP_KERNEL)); if (sbi->s_group_desc == NULL) { ext4_msg(sb, KERN_ERR, "not enough memory"); ret = -ENOMEM; @@ -4308,14 +4335,19 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) } for (i = 0; i < db_count; i++) { + struct buffer_head *bh; + block = descriptor_loc(sb, logical_sb_block, i); - sbi->s_group_desc[i] = sb_bread_unmovable(sb, block); - if (!sbi->s_group_desc[i]) { + bh = sb_bread_unmovable(sb, block); + if (!bh) { ext4_msg(sb, KERN_ERR, "can't read group descriptor %d", i); db_count = i; goto failed_mount2; } + rcu_read_lock(); + rcu_dereference(sbi->s_group_desc)[i] = bh; + rcu_read_unlock(); } sbi->s_gdb_count = db_count; if (!ext4_check_descriptors(sb, logical_sb_block, &first_not_zeroed)) { @@ -4594,7 +4626,7 @@ no_journal: err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0, GFP_KERNEL); if (!err) - err = percpu_init_rwsem(&sbi->s_journal_flag_rwsem); + err = percpu_init_rwsem(&sbi->s_writepages_rwsem); if (err) { ext4_msg(sb, KERN_ERR, "insufficient memory"); @@ -4682,13 +4714,19 @@ failed_mount7: ext4_unregister_li_request(sb); failed_mount6: ext4_mb_release(sb); - if (sbi->s_flex_groups) - kvfree(sbi->s_flex_groups); + rcu_read_lock(); + flex_groups = rcu_dereference(sbi->s_flex_groups); + if (flex_groups) { + for (i = 0; i < sbi->s_flex_groups_allocated; i++) + kvfree(flex_groups[i]); + kvfree(flex_groups); + } + rcu_read_unlock(); percpu_counter_destroy(&sbi->s_freeclusters_counter); percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); percpu_counter_destroy(&sbi->s_dirtyclusters_counter); - percpu_free_rwsem(&sbi->s_journal_flag_rwsem); + percpu_free_rwsem(&sbi->s_writepages_rwsem); failed_mount5: ext4_ext_release(sb); ext4_release_system_zone(sb); @@ -4717,9 +4755,12 @@ failed_mount3: if (sbi->s_mmp_tsk) kthread_stop(sbi->s_mmp_tsk); failed_mount2: + rcu_read_lock(); + group_desc = rcu_dereference(sbi->s_group_desc); for (i = 0; i < db_count; i++) - brelse(sbi->s_group_desc[i]); - kvfree(sbi->s_group_desc); + brelse(group_desc[i]); + kvfree(group_desc); + rcu_read_unlock(); failed_mount: if (sbi->s_chksum_driver) crypto_free_shash(sbi->s_chksum_driver); diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index d218ebdafa4a..04bfaf63752c 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -13,6 +13,7 @@ #include <linux/seq_file.h> #include <linux/slab.h> #include <linux/proc_fs.h> +#include <linux/part_stat.h> #include "ext4.h" #include "ext4_jbd2.h" diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 5355be6b6755..088c3e7a1080 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -22,6 +22,7 @@ #include <linux/bio.h> #include <linux/blkdev.h> #include <linux/quotaops.h> +#include <linux/part_stat.h> #include <crypto/hash.h> #include <linux/fscrypt.h> diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 0d4da644df3b..351762f77840 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -2423,6 +2423,14 @@ static int f2fs_ioc_get_encryption_key_status(struct file *filp, return fscrypt_ioctl_get_key_status(filp, (void __user *)arg); } +static int f2fs_ioc_get_encryption_nonce(struct file *filp, unsigned long arg) +{ + if (!f2fs_sb_has_encrypt(F2FS_I_SB(file_inode(filp)))) + return -EOPNOTSUPP; + + return fscrypt_ioctl_get_nonce(filp, (void __user *)arg); +} + static int f2fs_ioc_gc(struct file *filp, unsigned long arg) { struct inode *inode = file_inode(filp); @@ -3437,6 +3445,8 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return f2fs_ioc_remove_encryption_key_all_users(filp, arg); case FS_IOC_GET_ENCRYPTION_KEY_STATUS: return f2fs_ioc_get_encryption_key_status(filp, arg); + case FS_IOC_GET_ENCRYPTION_NONCE: + return f2fs_ioc_get_encryption_nonce(filp, arg); case F2FS_IOC_GARBAGE_COLLECT: return f2fs_ioc_gc(filp, arg); case F2FS_IOC_GARBAGE_COLLECT_RANGE: @@ -3611,6 +3621,7 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case FS_IOC_REMOVE_ENCRYPTION_KEY: case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS: case FS_IOC_GET_ENCRYPTION_KEY_STATUS: + case FS_IOC_GET_ENCRYPTION_NONCE: case F2FS_IOC_GARBAGE_COLLECT: case F2FS_IOC_GARBAGE_COLLECT_RANGE: case F2FS_IOC_WRITE_CHECKPOINT: diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 65a7a432dfee..d398b2d90c6c 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -24,6 +24,7 @@ #include <linux/sysfs.h> #include <linux/quota.h> #include <linux/unicode.h> +#include <linux/part_stat.h> #include "f2fs.h" #include "node.h" diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 594b05ae16c9..71946da84388 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -750,6 +750,13 @@ static struct inode *fat_alloc_inode(struct super_block *sb) return NULL; init_rwsem(&ei->truncate_lock); + /* Zeroing to allow iput() even if partial initialized inode. */ + ei->mmu_private = 0; + ei->i_start = 0; + ei->i_logstart = 0; + ei->i_attrs = 0; + ei->i_pos = 0; + return &ei->vfs_inode; } @@ -1374,16 +1381,6 @@ out: return 0; } -static void fat_dummy_inode_init(struct inode *inode) -{ - /* Initialize this dummy inode to work as no-op. */ - MSDOS_I(inode)->mmu_private = 0; - MSDOS_I(inode)->i_start = 0; - MSDOS_I(inode)->i_logstart = 0; - MSDOS_I(inode)->i_attrs = 0; - MSDOS_I(inode)->i_pos = 0; -} - static int fat_read_root(struct inode *inode) { struct msdos_sb_info *sbi = MSDOS_SB(inode->i_sb); @@ -1844,13 +1841,11 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, fat_inode = new_inode(sb); if (!fat_inode) goto out_fail; - fat_dummy_inode_init(fat_inode); sbi->fat_inode = fat_inode; fsinfo_inode = new_inode(sb); if (!fsinfo_inode) goto out_fail; - fat_dummy_inode_init(fsinfo_inode); fsinfo_inode->i_ino = MSDOS_FSINFO_INO; sbi->fsinfo_inode = fsinfo_inode; insert_inode_hash(fsinfo_inode); diff --git a/fs/fcntl.c b/fs/fcntl.c index 9bc167562ee8..2e4c0fa2074b 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -735,8 +735,9 @@ static void send_sigio_to_task(struct task_struct *p, return; switch (signum) { - kernel_siginfo_t si; - default: + default: { + kernel_siginfo_t si; + /* Queue a rt signal with the appropriate fd as its value. We use SI_SIGIO as the source, not SI_KERNEL, since kernel signals always get @@ -769,6 +770,7 @@ static void send_sigio_to_task(struct task_struct *p, si.si_fd = fd; if (!do_send_sig_info(signum, &si, p, type)) break; + } /* fall-through - fall back on the old plain SIGIO signal */ case 0: do_send_sig_info(SIGIO, SEND_SIG_PRIV, p, type); diff --git a/fs/file.c b/fs/file.c index a364e1a9b7e8..c8a4e4c86e55 100644 --- a/fs/file.c +++ b/fs/file.c @@ -540,9 +540,14 @@ static int alloc_fd(unsigned start, unsigned flags) return __alloc_fd(current->files, start, rlimit(RLIMIT_NOFILE), flags); } +int __get_unused_fd_flags(unsigned flags, unsigned long nofile) +{ + return __alloc_fd(current->files, 0, nofile, flags); +} + int get_unused_fd_flags(unsigned flags) { - return __alloc_fd(current->files, 0, rlimit(RLIMIT_NOFILE), flags); + return __get_unused_fd_flags(flags, rlimit(RLIMIT_NOFILE)); } EXPORT_SYMBOL(get_unused_fd_flags); diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 8e02d76fe104..97eec7522bf2 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -276,12 +276,10 @@ static void flush_bg_queue(struct fuse_conn *fc) void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req) { struct fuse_iqueue *fiq = &fc->iq; - bool async; if (test_and_set_bit(FR_FINISHED, &req->flags)) goto put_request; - async = req->args->end; /* * test_and_set_bit() implies smp_mb() between bit * changing and below intr_entry check. Pairs with @@ -324,7 +322,7 @@ void fuse_request_end(struct fuse_conn *fc, struct fuse_req *req) wake_up(&req->waitq); } - if (async) + if (test_bit(FR_ASYNC, &req->flags)) req->args->end(fc, req->args, req->out.h.error); put_request: fuse_put_request(fc, req); @@ -471,6 +469,8 @@ static void fuse_args_to_req(struct fuse_req *req, struct fuse_args *args) req->in.h.opcode = args->opcode; req->in.h.nodeid = args->nodeid; req->args = args; + if (args->end) + __set_bit(FR_ASYNC, &req->flags); } ssize_t fuse_simple_request(struct fuse_conn *fc, struct fuse_args *args) diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index aa75e2305b75..ca344bf71404 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -301,6 +301,7 @@ struct fuse_io_priv { * FR_SENT: request is in userspace, waiting for an answer * FR_FINISHED: request is finished * FR_PRIVATE: request is on private list + * FR_ASYNC: request is asynchronous */ enum fuse_req_flag { FR_ISREPLY, @@ -314,6 +315,7 @@ enum fuse_req_flag { FR_SENT, FR_FINISHED, FR_PRIVATE, + FR_ASYNC, }; /** diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 09e6be8aa036..2e939f5fe751 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -21,6 +21,7 @@ #include "glock.h" #include "inode.h" #include "meta_io.h" +#include "quota.h" #include "rgrp.h" #include "trans.h" #include "util.h" @@ -116,14 +117,14 @@ int gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type) if (acl && acl->a_count > GFS2_ACL_MAX_ENTRIES(GFS2_SB(inode))) return -E2BIG; - ret = gfs2_rsqa_alloc(ip); + ret = gfs2_qa_get(ip); if (ret) return ret; if (!gfs2_glock_is_locked_by_me(ip->i_gl)) { ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); if (ret) - return ret; + goto out; need_unlock = true; } @@ -143,5 +144,7 @@ int gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type) unlock: if (need_unlock) gfs2_glock_dq_uninit(&gh); +out: + gfs2_qa_put(ip); return ret; } diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index ba83b49ce18c..786c1ce8f030 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -805,11 +805,16 @@ int gfs2_releasepage(struct page *page, gfp_t gfp_mask) bd = bh->b_private; if (bd) { gfs2_assert_warn(sdp, bd->bd_bh == bh); - if (!list_empty(&bd->bd_list)) - list_del_init(&bd->bd_list); bd->bd_bh = NULL; bh->b_private = NULL; - kmem_cache_free(gfs2_bufdata_cachep, bd); + /* + * The bd may still be queued as a revoke, in which + * case we must not dequeue nor free it. + */ + if (!bd->bd_blkno && !list_empty(&bd->bd_list)) + list_del_init(&bd->bd_list); + if (list_empty(&bd->bd_list)) + kmem_cache_free(gfs2_bufdata_cachep, bd); } bh = bh->b_this_page; diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 08f6fbb3655e..936a8ec6b48e 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -2183,7 +2183,7 @@ int gfs2_setattr_size(struct inode *inode, u64 newsize) inode_dio_wait(inode); - ret = gfs2_rsqa_alloc(ip); + ret = gfs2_qa_get(ip); if (ret) goto out; @@ -2194,7 +2194,8 @@ int gfs2_setattr_size(struct inode *inode, u64 newsize) ret = do_shrink(inode, newsize); out: - gfs2_rsqa_delete(ip, NULL); + gfs2_rs_delete(ip, NULL); + gfs2_qa_put(ip); return ret; } @@ -2223,7 +2224,7 @@ void gfs2_free_journal_extents(struct gfs2_jdesc *jd) struct gfs2_journal_extent *jext; while(!list_empty(&jd->extent_list)) { - jext = list_entry(jd->extent_list.next, struct gfs2_journal_extent, list); + jext = list_first_entry(&jd->extent_list, struct gfs2_journal_extent, list); list_del(&jext->list); kfree(jext); } @@ -2244,7 +2245,7 @@ static int gfs2_add_jextent(struct gfs2_jdesc *jd, u64 lblock, u64 dblock, u64 b struct gfs2_journal_extent *jext; if (!list_empty(&jd->extent_list)) { - jext = list_entry(jd->extent_list.prev, struct gfs2_journal_extent, list); + jext = list_last_entry(&jd->extent_list, struct gfs2_journal_extent, list); if ((jext->dblock + jext->blocks) == dblock) { jext->blocks += blocks; return 0; diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index c8b62577e2f2..c3f7732415be 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -2028,7 +2028,8 @@ static int leaf_dealloc(struct gfs2_inode *dip, u32 index, u32 len, error = gfs2_trans_begin(sdp, rg_blocks + (DIV_ROUND_UP(size, sdp->sd_jbsize) + 1) + - RES_DINODE + RES_STATFS + RES_QUOTA, l_blocks); + RES_DINODE + RES_STATFS + RES_QUOTA, RES_DINODE + + l_blocks); if (error) goto out_rg_gunlock; diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index cb26be6f4351..fe305e4bfd37 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -458,10 +458,6 @@ static vm_fault_t gfs2_page_mkwrite(struct vm_fault *vmf) sb_start_pagefault(inode->i_sb); - ret = gfs2_rsqa_alloc(ip); - if (ret) - goto out; - gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); ret = gfs2_glock_nq(&gh); if (ret) @@ -558,7 +554,6 @@ out_uninit: set_page_dirty(page); wait_for_stable_page(page); } -out: sb_end_pagefault(inode->i_sb); return block_page_mkwrite_return(ret); } @@ -635,7 +630,17 @@ int gfs2_open_common(struct inode *inode, struct file *file) gfs2_assert_warn(GFS2_SB(inode), !file->private_data); file->private_data = fp; + if (file->f_mode & FMODE_WRITE) { + ret = gfs2_qa_get(GFS2_I(inode)); + if (ret) + goto fail; + } return 0; + +fail: + kfree(file->private_data); + file->private_data = NULL; + return ret; } /** @@ -690,10 +695,10 @@ static int gfs2_release(struct inode *inode, struct file *file) kfree(file->private_data); file->private_data = NULL; - if (!(file->f_mode & FMODE_WRITE)) - return 0; - - gfs2_rsqa_delete(ip, &inode->i_writecount); + if (file->f_mode & FMODE_WRITE) { + gfs2_rs_delete(ip, &inode->i_writecount); + gfs2_qa_put(ip); + } return 0; } @@ -849,10 +854,6 @@ static ssize_t gfs2_file_write_iter(struct kiocb *iocb, struct iov_iter *from) struct gfs2_inode *ip = GFS2_I(inode); ssize_t ret; - ret = gfs2_rsqa_alloc(ip); - if (ret) - return ret; - gfs2_size_hint(file, iocb->ki_pos, iov_iter_count(from)); if (iocb->ki_flags & IOCB_APPEND) { @@ -1149,17 +1150,11 @@ static long gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t le if (mode & FALLOC_FL_PUNCH_HOLE) { ret = __gfs2_punch_hole(file, offset, len); } else { - ret = gfs2_rsqa_alloc(ip); - if (ret) - goto out_putw; - ret = __gfs2_fallocate(file, mode, offset, len); - if (ret) gfs2_rs_deltree(&ip->i_res); } -out_putw: put_write_access(inode); out_unlock: gfs2_glock_dq(&gh); @@ -1173,16 +1168,12 @@ static ssize_t gfs2_file_splice_write(struct pipe_inode_info *pipe, struct file *out, loff_t *ppos, size_t len, unsigned int flags) { - int error; - struct gfs2_inode *ip = GFS2_I(out->f_mapping->host); - - error = gfs2_rsqa_alloc(ip); - if (error) - return (ssize_t)error; + ssize_t ret; gfs2_size_hint(out, *ppos, len); - return iter_file_splice_write(pipe, out, ppos, len, flags); + ret = iter_file_splice_write(pipe, out, ppos, len, flags); + return ret; } #ifdef CONFIG_GFS2_FS_LOCKING_DLM diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index d0eceaff3cea..29f9b6684b74 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -133,6 +133,33 @@ static void gfs2_glock_dealloc(struct rcu_head *rcu) } } +/** + * glock_blocked_by_withdraw - determine if we can still use a glock + * @gl: the glock + * + * We need to allow some glocks to be enqueued, dequeued, promoted, and demoted + * when we're withdrawn. For example, to maintain metadata integrity, we should + * disallow the use of inode and rgrp glocks when withdrawn. Other glocks, like + * iopen or the transaction glocks may be safely used because none of their + * metadata goes through the journal. So in general, we should disallow all + * glocks that are journaled, and allow all the others. One exception is: + * we need to allow our active journal to be promoted and demoted so others + * may recover it and we can reacquire it when they're done. + */ +static bool glock_blocked_by_withdraw(struct gfs2_glock *gl) +{ + struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; + + if (likely(!gfs2_withdrawn(sdp))) + return false; + if (gl->gl_ops->go_flags & GLOF_NONDISK) + return false; + if (!sdp->sd_jdesc || + gl->gl_name.ln_number == sdp->sd_jdesc->jd_no_addr) + return false; + return true; +} + void gfs2_glock_free(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; @@ -244,7 +271,7 @@ static void __gfs2_glock_put(struct gfs2_glock *gl) gfs2_glock_remove_from_lru(gl); spin_unlock(&gl->gl_lockref.lock); GLOCK_BUG_ON(gl, !list_empty(&gl->gl_holders)); - GLOCK_BUG_ON(gl, mapping && mapping->nrpages); + GLOCK_BUG_ON(gl, mapping && mapping->nrpages && !gfs2_withdrawn(sdp)); trace_gfs2_glock_put(gl); sdp->sd_lockstruct.ls_ops->lm_put_lock(gl); } @@ -281,7 +308,7 @@ void gfs2_glock_put(struct gfs2_glock *gl) static inline int may_grant(const struct gfs2_glock *gl, const struct gfs2_holder *gh) { - const struct gfs2_holder *gh_head = list_entry(gl->gl_holders.next, const struct gfs2_holder, gh_list); + const struct gfs2_holder *gh_head = list_first_entry(&gl->gl_holders, const struct gfs2_holder, gh_list); if ((gh->gh_state == LM_ST_EXCLUSIVE || gh_head->gh_state == LM_ST_EXCLUSIVE) && gh != gh_head) return 0; @@ -549,8 +576,8 @@ __acquires(&gl->gl_lockref.lock) unsigned int lck_flags = (unsigned int)(gh ? gh->gh_flags : 0); int ret; - if (unlikely(gfs2_withdrawn(sdp)) && - target != LM_ST_UNLOCKED) + if (target != LM_ST_UNLOCKED && glock_blocked_by_withdraw(gl) && + gh && !(gh->gh_flags & LM_FLAG_NOEXP)) return; lck_flags &= (LM_FLAG_TRY | LM_FLAG_TRY_1CB | LM_FLAG_NOEXP | LM_FLAG_PRIORITY); @@ -575,13 +602,64 @@ __acquires(&gl->gl_lockref.lock) (lck_flags & (LM_FLAG_TRY|LM_FLAG_TRY_1CB))) clear_bit(GLF_BLOCKING, &gl->gl_flags); spin_unlock(&gl->gl_lockref.lock); - if (glops->go_sync) - glops->go_sync(gl); - if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags)) + if (glops->go_sync) { + ret = glops->go_sync(gl); + /* If we had a problem syncing (due to io errors or whatever, + * we should not invalidate the metadata or tell dlm to + * release the glock to other nodes. + */ + if (ret) { + if (cmpxchg(&sdp->sd_log_error, 0, ret)) { + fs_err(sdp, "Error %d syncing glock \n", ret); + gfs2_dump_glock(NULL, gl, true); + } + return; + } + } + if (test_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags)) { + /* + * The call to go_sync should have cleared out the ail list. + * If there are still items, we have a problem. We ought to + * withdraw, but we can't because the withdraw code also uses + * glocks. Warn about the error, dump the glock, then fall + * through and wait for logd to do the withdraw for us. + */ + if ((atomic_read(&gl->gl_ail_count) != 0) && + (!cmpxchg(&sdp->sd_log_error, 0, -EIO))) { + gfs2_assert_warn(sdp, !atomic_read(&gl->gl_ail_count)); + gfs2_dump_glock(NULL, gl, true); + } glops->go_inval(gl, target == LM_ST_DEFERRED ? 0 : DIO_METADATA); - clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); + clear_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); + } gfs2_glock_hold(gl); + /* + * Check for an error encountered since we called go_sync and go_inval. + * If so, we can't withdraw from the glock code because the withdraw + * code itself uses glocks (see function signal_our_withdraw) to + * change the mount to read-only. Most importantly, we must not call + * dlm to unlock the glock until the journal is in a known good state + * (after journal replay) otherwise other nodes may use the object + * (rgrp or dinode) and then later, journal replay will corrupt the + * file system. The best we can do here is wait for the logd daemon + * to see sd_log_error and withdraw, and in the meantime, requeue the + * work for later. + * + * However, if we're just unlocking the lock (say, for unmount, when + * gfs2_gl_hash_clear calls clear_glock) and recovery is complete + * then it's okay to tell dlm to unlock it. + */ + if (unlikely(sdp->sd_log_error && !gfs2_withdrawn(sdp))) + gfs2_withdraw_delayed(sdp); + if (glock_blocked_by_withdraw(gl)) { + if (target != LM_ST_UNLOCKED || + test_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags)) { + gfs2_glock_queue_work(gl, GL_GLOCK_DFT_HOLD); + goto out; + } + } + if (sdp->sd_lockstruct.ls_ops->lm_lock) { /* lock_dlm */ ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags); @@ -590,8 +668,7 @@ __acquires(&gl->gl_lockref.lock) test_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags)) { finish_xmote(gl, target); gfs2_glock_queue_work(gl, 0); - } - else if (ret) { + } else if (ret) { fs_err(sdp, "lm_lock ret %d\n", ret); GLOCK_BUG_ON(gl, !gfs2_withdrawn(sdp)); } @@ -599,7 +676,7 @@ __acquires(&gl->gl_lockref.lock) finish_xmote(gl, target); gfs2_glock_queue_work(gl, 0); } - +out: spin_lock(&gl->gl_lockref.lock); } @@ -613,7 +690,7 @@ static inline struct gfs2_holder *find_first_holder(const struct gfs2_glock *gl) struct gfs2_holder *gh; if (!list_empty(&gl->gl_holders)) { - gh = list_entry(gl->gl_holders.next, struct gfs2_holder, gh_list); + gh = list_first_entry(&gl->gl_holders, struct gfs2_holder, gh_list); if (test_bit(HIF_HOLDER, &gh->gh_iflags)) return gh; } @@ -645,6 +722,9 @@ __acquires(&gl->gl_lockref.lock) goto out_unlock; if (nonblock) goto out_sched; + smp_mb(); + if (atomic_read(&gl->gl_revokes) != 0) + goto out_sched; set_bit(GLF_DEMOTE_IN_PROGRESS, &gl->gl_flags); GLOCK_BUG_ON(gl, gl->gl_demote_state == LM_ST_EXCLUSIVE); gl->gl_target = gl->gl_demote_state; @@ -1160,7 +1240,7 @@ fail: } list_add_tail(&gh->gh_list, insert_pt); do_cancel: - gh = list_entry(gl->gl_holders.next, struct gfs2_holder, gh_list); + gh = list_first_entry(&gl->gl_holders, struct gfs2_holder, gh_list); if (!(gh->gh_flags & LM_FLAG_PRIORITY)) { spin_unlock(&gl->gl_lockref.lock); if (sdp->sd_lockstruct.ls_ops->lm_cancel) @@ -1194,10 +1274,9 @@ trap_recursive: int gfs2_glock_nq(struct gfs2_holder *gh) { struct gfs2_glock *gl = gh->gh_gl; - struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; int error = 0; - if (unlikely(gfs2_withdrawn(sdp))) + if (glock_blocked_by_withdraw(gl) && !(gh->gh_flags & LM_FLAG_NOEXP)) return -EIO; if (test_bit(GLF_LRU, &gl->gl_flags)) @@ -1241,24 +1320,32 @@ int gfs2_glock_poll(struct gfs2_holder *gh) void gfs2_glock_dq(struct gfs2_holder *gh) { struct gfs2_glock *gl = gh->gh_gl; - const struct gfs2_glock_operations *glops = gl->gl_ops; + struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; unsigned delay = 0; int fast_path = 0; spin_lock(&gl->gl_lockref.lock); + /* + * If we're in the process of file system withdraw, we cannot just + * dequeue any glocks until our journal is recovered, lest we + * introduce file system corruption. We need two exceptions to this + * rule: We need to allow unlocking of nondisk glocks and the glock + * for our own journal that needs recovery. + */ + if (test_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags) && + glock_blocked_by_withdraw(gl) && + gh->gh_gl != sdp->sd_jinode_gl) { + sdp->sd_glock_dqs_held++; + might_sleep(); + wait_on_bit(&sdp->sd_flags, SDF_WITHDRAW_RECOVERY, + TASK_UNINTERRUPTIBLE); + } if (gh->gh_flags & GL_NOCACHE) handle_callback(gl, LM_ST_UNLOCKED, 0, false); list_del_init(&gh->gh_list); clear_bit(HIF_HOLDER, &gh->gh_iflags); if (find_first_holder(gl) == NULL) { - if (glops->go_unlock) { - GLOCK_BUG_ON(gl, test_and_set_bit(GLF_LOCK, &gl->gl_flags)); - spin_unlock(&gl->gl_lockref.lock); - glops->go_unlock(gh); - spin_lock(&gl->gl_lockref.lock); - clear_bit(GLF_LOCK, &gl->gl_flags); - } if (list_empty(&gl->gl_holders) && !test_bit(GLF_PENDING_DEMOTE, &gl->gl_flags) && !test_bit(GLF_DEMOTE, &gl->gl_flags)) @@ -1555,7 +1642,7 @@ __acquires(&lru_lock) list_sort(NULL, list, glock_cmp); while(!list_empty(list)) { - gl = list_entry(list->next, struct gfs2_glock, gl_lru); + gl = list_first_entry(list, struct gfs2_glock, gl_lru); list_del_init(&gl->gl_lru); if (!spin_trylock(&gl->gl_lockref.lock)) { add_back_to_lru: @@ -1596,7 +1683,7 @@ static long gfs2_scan_glock_lru(int nr) spin_lock(&lru_lock); while ((nr-- >= 0) && !list_empty(&lru_list)) { - gl = list_entry(lru_list.next, struct gfs2_glock, gl_lru); + gl = list_first_entry(&lru_list, struct gfs2_glock, gl_lru); /* Test for being demotable */ if (!test_bit(GLF_LOCK, &gl->gl_flags)) { diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 061d22e1ceb6..9e9c7a4b8c66 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -29,6 +29,8 @@ struct workqueue_struct *gfs2_freeze_wq; +extern struct workqueue_struct *gfs2_control_wq; + static void gfs2_ail_error(struct gfs2_glock *gl, const struct buffer_head *bh) { fs_err(gl->gl_name.ln_sbd, @@ -39,7 +41,8 @@ static void gfs2_ail_error(struct gfs2_glock *gl, const struct buffer_head *bh) fs_err(gl->gl_name.ln_sbd, "AIL glock %u:%llu mapping %p\n", gl->gl_name.ln_type, gl->gl_name.ln_number, gfs2_glock2aspace(gl)); - gfs2_lm_withdraw(gl->gl_name.ln_sbd, "AIL error\n"); + gfs2_lm(gl->gl_name.ln_sbd, "AIL error\n"); + gfs2_withdraw(gl->gl_name.ln_sbd); } /** @@ -79,34 +82,62 @@ static void __gfs2_ail_flush(struct gfs2_glock *gl, bool fsync, } -static void gfs2_ail_empty_gl(struct gfs2_glock *gl) +static int gfs2_ail_empty_gl(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; struct gfs2_trans tr; + int ret; memset(&tr, 0, sizeof(tr)); INIT_LIST_HEAD(&tr.tr_buf); INIT_LIST_HEAD(&tr.tr_databuf); tr.tr_revokes = atomic_read(&gl->gl_ail_count); - if (!tr.tr_revokes) - return; + if (!tr.tr_revokes) { + bool have_revokes; + bool log_in_flight; + + /* + * We have nothing on the ail, but there could be revokes on + * the sdp revoke queue, in which case, we still want to flush + * the log and wait for it to finish. + * + * If the sdp revoke list is empty too, we might still have an + * io outstanding for writing revokes, so we should wait for + * it before returning. + * + * If none of these conditions are true, our revokes are all + * flushed and we can return. + */ + gfs2_log_lock(sdp); + have_revokes = !list_empty(&sdp->sd_log_revokes); + log_in_flight = atomic_read(&sdp->sd_log_in_flight); + gfs2_log_unlock(sdp); + if (have_revokes) + goto flush; + if (log_in_flight) + log_flush_wait(sdp); + return 0; + } /* A shortened, inline version of gfs2_trans_begin() * tr->alloced is not set since the transaction structure is * on the stack */ tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes); tr.tr_ip = _RET_IP_; - if (gfs2_log_reserve(sdp, tr.tr_reserved) < 0) - return; + ret = gfs2_log_reserve(sdp, tr.tr_reserved); + if (ret < 0) + return ret; WARN_ON_ONCE(current->journal_info); current->journal_info = &tr; __gfs2_ail_flush(gl, 0, tr.tr_revokes); gfs2_trans_end(sdp); +flush: gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_AIL_EMPTY_GL); + return 0; } void gfs2_ail_flush(struct gfs2_glock *gl, bool fsync) @@ -140,35 +171,32 @@ void gfs2_ail_flush(struct gfs2_glock *gl, bool fsync) * return to caller to demote/unlock the glock until I/O is complete. */ -static void rgrp_go_sync(struct gfs2_glock *gl) +static int rgrp_go_sync(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; struct address_space *mapping = &sdp->sd_aspace; - struct gfs2_rgrpd *rgd; + struct gfs2_rgrpd *rgd = gfs2_glock2rgrp(gl); int error; - spin_lock(&gl->gl_lockref.lock); - rgd = gl->gl_object; - if (rgd) - gfs2_rgrp_brelse(rgd); - spin_unlock(&gl->gl_lockref.lock); - if (!test_and_clear_bit(GLF_DIRTY, &gl->gl_flags)) - return; + return 0; GLOCK_BUG_ON(gl, gl->gl_state != LM_ST_EXCLUSIVE); gfs2_log_flush(sdp, gl, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_RGRP_GO_SYNC); filemap_fdatawrite_range(mapping, gl->gl_vm.start, gl->gl_vm.end); error = filemap_fdatawait_range(mapping, gl->gl_vm.start, gl->gl_vm.end); + WARN_ON_ONCE(error); mapping_set_error(mapping, error); - gfs2_ail_empty_gl(gl); + if (!error) + error = gfs2_ail_empty_gl(gl); spin_lock(&gl->gl_lockref.lock); rgd = gl->gl_object; if (rgd) gfs2_free_clones(rgd); spin_unlock(&gl->gl_lockref.lock); + return error; } /** @@ -191,7 +219,6 @@ static void rgrp_go_inval(struct gfs2_glock *gl, int flags) gfs2_rgrp_brelse(rgd); WARN_ON_ONCE(!(flags & DIO_METADATA)); - gfs2_assert_withdraw(sdp, !atomic_read(&gl->gl_ail_count)); truncate_inode_pages_range(mapping, gl->gl_vm.start, gl->gl_vm.end); if (rgd) @@ -236,12 +263,12 @@ static void gfs2_clear_glop_pending(struct gfs2_inode *ip) * */ -static void inode_go_sync(struct gfs2_glock *gl) +static int inode_go_sync(struct gfs2_glock *gl) { struct gfs2_inode *ip = gfs2_glock2inode(gl); int isreg = ip && S_ISREG(ip->i_inode.i_mode); struct address_space *metamapping = gfs2_glock2aspace(gl); - int error; + int error = 0; if (isreg) { if (test_and_clear_bit(GIF_SW_PAGED, &ip->i_flags)) @@ -274,6 +301,7 @@ static void inode_go_sync(struct gfs2_glock *gl) out: gfs2_clear_glop_pending(ip); + return error; } /** @@ -291,8 +319,6 @@ static void inode_go_inval(struct gfs2_glock *gl, int flags) { struct gfs2_inode *ip = gfs2_glock2inode(gl); - gfs2_assert_withdraw(gl->gl_name.ln_sbd, !atomic_read(&gl->gl_ail_count)); - if (flags & DIO_METADATA) { struct address_space *mapping = gfs2_glock2aspace(gl); truncate_inode_pages(mapping, 0); @@ -496,24 +522,29 @@ static void inode_go_dump(struct seq_file *seq, struct gfs2_glock *gl, * */ -static void freeze_go_sync(struct gfs2_glock *gl) +static int freeze_go_sync(struct gfs2_glock *gl) { int error = 0; struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; - if (gl->gl_state == LM_ST_SHARED && + if (gl->gl_state == LM_ST_SHARED && !gfs2_withdrawn(sdp) && test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) { atomic_set(&sdp->sd_freeze_state, SFS_STARTING_FREEZE); error = freeze_super(sdp->sd_vfs); if (error) { fs_info(sdp, "GFS2: couldn't freeze filesystem: %d\n", error); + if (gfs2_withdrawn(sdp)) { + atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN); + return 0; + } gfs2_assert_withdraw(sdp, 0); } queue_work(gfs2_freeze_wq, &sdp->sd_freeze_work); gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_FREEZE | GFS2_LFC_FREEZE_GO_SYNC); } + return 0; } /** @@ -582,8 +613,76 @@ static void iopen_go_callback(struct gfs2_glock *gl, bool remote) } } +/** + * inode_go_free - wake up anyone waiting for dlm's unlock ast to free it + * @gl: glock being freed + * + * For now, this is only used for the journal inode glock. In withdraw + * situations, we need to wait for the glock to be freed so that we know + * other nodes may proceed with recovery / journal replay. + */ +static void inode_go_free(struct gfs2_glock *gl) +{ + /* Note that we cannot reference gl_object because it's already set + * to NULL by this point in its lifecycle. */ + if (!test_bit(GLF_FREEING, &gl->gl_flags)) + return; + clear_bit_unlock(GLF_FREEING, &gl->gl_flags); + wake_up_bit(&gl->gl_flags, GLF_FREEING); +} + +/** + * nondisk_go_callback - used to signal when a node did a withdraw + * @gl: the nondisk glock + * @remote: true if this came from a different cluster node + * + */ +static void nondisk_go_callback(struct gfs2_glock *gl, bool remote) +{ + struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; + + /* Ignore the callback unless it's from another node, and it's the + live lock. */ + if (!remote || gl->gl_name.ln_number != GFS2_LIVE_LOCK) + return; + + /* First order of business is to cancel the demote request. We don't + * really want to demote a nondisk glock. At best it's just to inform + * us of another node's withdraw. We'll keep it in SH mode. */ + clear_bit(GLF_DEMOTE, &gl->gl_flags); + clear_bit(GLF_PENDING_DEMOTE, &gl->gl_flags); + + /* Ignore the unlock if we're withdrawn, unmounting, or in recovery. */ + if (test_bit(SDF_NORECOVERY, &sdp->sd_flags) || + test_bit(SDF_WITHDRAWN, &sdp->sd_flags) || + test_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags)) + return; + + /* We only care when a node wants us to unlock, because that means + * they want a journal recovered. */ + if (gl->gl_demote_state != LM_ST_UNLOCKED) + return; + + if (sdp->sd_args.ar_spectator) { + fs_warn(sdp, "Spectator node cannot recover journals.\n"); + return; + } + + fs_warn(sdp, "Some node has withdrawn; checking for recovery.\n"); + set_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags); + /* + * We can't call remote_withdraw directly here or gfs2_recover_journal + * because this is called from the glock unlock function and the + * remote_withdraw needs to enqueue and dequeue the same "live" glock + * we were called from. So we queue it to the control work queue in + * lock_dlm. + */ + queue_delayed_work(gfs2_control_wq, &sdp->sd_control_work, 0); +} + const struct gfs2_glock_operations gfs2_meta_glops = { .go_type = LM_TYPE_META, + .go_flags = GLOF_NONDISK, }; const struct gfs2_glock_operations gfs2_inode_glops = { @@ -594,13 +693,13 @@ const struct gfs2_glock_operations gfs2_inode_glops = { .go_dump = inode_go_dump, .go_type = LM_TYPE_INODE, .go_flags = GLOF_ASPACE | GLOF_LRU, + .go_free = inode_go_free, }; const struct gfs2_glock_operations gfs2_rgrp_glops = { .go_sync = rgrp_go_sync, .go_inval = rgrp_go_inval, .go_lock = gfs2_rgrp_go_lock, - .go_unlock = gfs2_rgrp_go_unlock, .go_dump = gfs2_rgrp_dump, .go_type = LM_TYPE_RGRP, .go_flags = GLOF_LVB, @@ -611,30 +710,34 @@ const struct gfs2_glock_operations gfs2_freeze_glops = { .go_xmote_bh = freeze_go_xmote_bh, .go_demote_ok = freeze_go_demote_ok, .go_type = LM_TYPE_NONDISK, + .go_flags = GLOF_NONDISK, }; const struct gfs2_glock_operations gfs2_iopen_glops = { .go_type = LM_TYPE_IOPEN, .go_callback = iopen_go_callback, - .go_flags = GLOF_LRU, + .go_flags = GLOF_LRU | GLOF_NONDISK, }; const struct gfs2_glock_operations gfs2_flock_glops = { .go_type = LM_TYPE_FLOCK, - .go_flags = GLOF_LRU, + .go_flags = GLOF_LRU | GLOF_NONDISK, }; const struct gfs2_glock_operations gfs2_nondisk_glops = { .go_type = LM_TYPE_NONDISK, + .go_flags = GLOF_NONDISK, + .go_callback = nondisk_go_callback, }; const struct gfs2_glock_operations gfs2_quota_glops = { .go_type = LM_TYPE_QUOTA, - .go_flags = GLOF_LVB | GLOF_LRU, + .go_flags = GLOF_LVB | GLOF_LRU | GLOF_NONDISK, }; const struct gfs2_glock_operations gfs2_journal_glops = { .go_type = LM_TYPE_JOURNAL, + .go_flags = GLOF_NONDISK, }; const struct gfs2_glock_operations *gfs2_glops_list[] = { diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 9fd88ed18807..84a824293a78 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -234,20 +234,21 @@ struct lm_lockname { struct gfs2_glock_operations { - void (*go_sync) (struct gfs2_glock *gl); + int (*go_sync) (struct gfs2_glock *gl); int (*go_xmote_bh) (struct gfs2_glock *gl, struct gfs2_holder *gh); void (*go_inval) (struct gfs2_glock *gl, int flags); int (*go_demote_ok) (const struct gfs2_glock *gl); int (*go_lock) (struct gfs2_holder *gh); - void (*go_unlock) (struct gfs2_holder *gh); void (*go_dump)(struct seq_file *seq, struct gfs2_glock *gl, const char *fs_id_buf); void (*go_callback)(struct gfs2_glock *gl, bool remote); + void (*go_free)(struct gfs2_glock *gl); const int go_type; const unsigned long go_flags; -#define GLOF_ASPACE 1 -#define GLOF_LVB 2 -#define GLOF_LRU 4 +#define GLOF_ASPACE 1 /* address space attached */ +#define GLOF_LVB 2 /* Lock Value Block attached */ +#define GLOF_LRU 4 /* LRU managed */ +#define GLOF_NONDISK 8 /* not I/O related */ }; enum { @@ -294,6 +295,7 @@ struct gfs2_qadata { /* quota allocation data */ struct gfs2_quota_data *qa_qd[2 * GFS2_MAXQUOTAS]; struct gfs2_holder qa_qd_ghs[2 * GFS2_MAXQUOTAS]; unsigned int qa_qd_num; + int qa_ref; }; /* Resource group multi-block reservation, in order of appearance: @@ -343,6 +345,7 @@ enum { GLF_OBJECT = 14, /* Used only for tracing */ GLF_BLOCKING = 15, GLF_INODE_CREATING = 16, /* Inode creation occurring */ + GLF_FREEING = 18, /* Wait for glock to be freed */ }; struct gfs2_glock { @@ -542,6 +545,7 @@ struct gfs2_jdesc { struct list_head jd_revoke_list; unsigned int jd_replay_tail; + u64 jd_no_addr; }; struct gfs2_statfs_change_host { @@ -616,8 +620,12 @@ enum { SDF_RORECOVERY = 7, /* read only recovery */ SDF_SKIP_DLM_UNLOCK = 8, SDF_FORCE_AIL_FLUSH = 9, - SDF_AIL1_IO_ERROR = 10, - SDF_FS_FROZEN = 11, + SDF_FS_FROZEN = 10, + SDF_WITHDRAWING = 11, /* Will withdraw eventually */ + SDF_WITHDRAW_IN_PROG = 12, /* Withdraw is in progress */ + SDF_REMOTE_WITHDRAW = 13, /* Performing remote recovery */ + SDF_WITHDRAW_RECOVERY = 14, /* Wait for journal recovery when we are + withdrawing */ }; enum gfs2_freeze_state { @@ -768,6 +776,7 @@ struct gfs2_sbd { struct gfs2_jdesc *sd_jdesc; struct gfs2_holder sd_journal_gh; struct gfs2_holder sd_jinode_gh; + struct gfs2_glock *sd_jinode_gl; struct gfs2_holder sd_sc_gh; struct gfs2_holder sd_qc_gh; @@ -828,7 +837,8 @@ struct gfs2_sbd { atomic_t sd_log_in_flight; struct bio *sd_log_bio; wait_queue_head_t sd_log_flush_wait; - int sd_log_error; + int sd_log_error; /* First log error */ + wait_queue_head_t sd_withdraw_wait; atomic_t sd_reserving_log; wait_queue_head_t sd_reserving_log_wait; @@ -852,6 +862,7 @@ struct gfs2_sbd { unsigned long sd_last_warning; struct dentry *debugfs_dir; /* debugfs directory */ + unsigned long sd_glock_dqs_held; }; static inline void gfs2_glstats_inc(struct gfs2_glock *gl, int which) diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 2716d56ed0a0..70b2d3a1e866 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -144,7 +144,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type, error = gfs2_glock_get(sdp, no_addr, &gfs2_iopen_glops, CREATE, &io_gl); if (unlikely(error)) - goto fail_put; + goto fail; if (type == DT_UNKNOWN || blktype != GFS2_BLKST_FREE) { /* @@ -155,13 +155,13 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type, error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_SKIP, &i_gh); if (error) - goto fail_put; + goto fail; if (blktype != GFS2_BLKST_FREE) { error = gfs2_check_blk_type(sdp, no_addr, blktype); if (error) - goto fail_put; + goto fail; } } @@ -169,7 +169,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type, set_bit(GIF_INVALID, &ip->i_flags); error = gfs2_glock_nq_init(io_gl, LM_ST_SHARED, GL_EXACT, &ip->i_iopen_gh); if (unlikely(error)) - goto fail_put; + goto fail; glock_set_object(ip->i_iopen_gh.gh_gl, ip); gfs2_glock_put(io_gl); io_gl = NULL; @@ -182,7 +182,7 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type, /* Inode glock must be locked already */ error = gfs2_inode_refresh(GFS2_I(inode)); if (error) - goto fail_refresh; + goto fail; } else { ip->i_no_formal_ino = no_formal_ino; inode->i_mode = DT2IF(type); @@ -197,17 +197,11 @@ struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned int type, gfs2_glock_dq_uninit(&i_gh); return inode; -fail_refresh: - ip->i_iopen_gh.gh_flags |= GL_NOCACHE; - glock_clear_object(ip->i_iopen_gh.gh_gl, ip); - gfs2_glock_dq_uninit(&ip->i_iopen_gh); -fail_put: +fail: if (io_gl) gfs2_glock_put(io_gl); - glock_clear_object(ip->i_gl, ip); if (gfs2_holder_initialized(&i_gh)) gfs2_glock_dq_uninit(&i_gh); -fail: iget_failed(inode); return ERR_PTR(error); } @@ -594,13 +588,13 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, if (!name->len || name->len > GFS2_FNAMESIZE) return -ENAMETOOLONG; - error = gfs2_rsqa_alloc(dip); + error = gfs2_qa_get(dip); if (error) return error; error = gfs2_rindex_update(sdp); if (error) - return error; + goto fail; error = gfs2_glock_nq_init(dip->i_gl, LM_ST_EXCLUSIVE, 0, ghs); if (error) @@ -647,7 +641,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, goto fail_gunlock; ip = GFS2_I(inode); - error = gfs2_rsqa_alloc(ip); + error = gfs2_qa_get(ip); if (error) goto fail_free_acls; @@ -782,11 +776,13 @@ fail_gunlock2: clear_bit(GLF_INODE_CREATING, &io_gl->gl_flags); gfs2_glock_put(io_gl); fail_free_inode: + gfs2_qa_put(ip); if (ip->i_gl) { glock_clear_object(ip->i_gl, ip); gfs2_glock_put(ip->i_gl); } - gfs2_rsqa_delete(ip, NULL); + gfs2_rs_delete(ip, NULL); + gfs2_qa_put(ip); fail_free_acls: posix_acl_release(default_acl); posix_acl_release(acl); @@ -804,6 +800,7 @@ fail_gunlock: if (gfs2_holder_initialized(ghs + 1)) gfs2_glock_dq_uninit(ghs + 1); fail: + gfs2_qa_put(dip); return error; } @@ -905,7 +902,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (S_ISDIR(inode->i_mode)) return -EPERM; - error = gfs2_rsqa_alloc(dip); + error = gfs2_qa_get(dip); if (error) return error; @@ -1008,6 +1005,7 @@ out_gunlock: out_child: gfs2_glock_dq(ghs); out_parent: + gfs2_qa_put(ip); gfs2_holder_uninit(ghs); gfs2_holder_uninit(ghs + 1); return error; @@ -1248,7 +1246,7 @@ static int gfs2_atomic_open(struct inode *dir, struct dentry *dentry, if (!(file->f_mode & FMODE_OPENED)) return finish_no_open(file, d); dput(d); - return 0; + return excl && (flags & O_CREAT) ? -EEXIST : 0; } BUG_ON(d != NULL); @@ -1368,7 +1366,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, if (error) return error; - error = gfs2_rsqa_alloc(ndip); + error = gfs2_qa_get(ndip); if (error) return error; @@ -1568,6 +1566,7 @@ out_gunlock_r: if (gfs2_holder_initialized(&r_gh)) gfs2_glock_dq_uninit(&r_gh); out: + gfs2_qa_put(ndip); return error; } @@ -1879,10 +1878,9 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) ouid = nuid = NO_UID_QUOTA_CHANGE; if (!(attr->ia_valid & ATTR_GID) || gid_eq(ogid, ngid)) ogid = ngid = NO_GID_QUOTA_CHANGE; - - error = gfs2_rsqa_alloc(ip); + error = gfs2_qa_get(ip); if (error) - goto out; + return error; error = gfs2_rindex_update(sdp); if (error) @@ -1920,6 +1918,7 @@ out_end_trans: out_gunlock_q: gfs2_quota_unlock(ip); out: + gfs2_qa_put(ip); return error; } @@ -1941,21 +1940,21 @@ static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) struct gfs2_holder i_gh; int error; - error = gfs2_rsqa_alloc(ip); + error = gfs2_qa_get(ip); if (error) return error; error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); if (error) - return error; + goto out; error = -EPERM; if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) - goto out; + goto error; error = setattr_prepare(dentry, attr); if (error) - goto out; + goto error; if (attr->ia_valid & ATTR_SIZE) error = gfs2_setattr_size(inode, attr->ia_size); @@ -1967,10 +1966,12 @@ static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) error = posix_acl_chmod(inode, inode->i_mode); } -out: +error: if (!error) mark_inode_dirty(inode); gfs2_glock_dq_uninit(&i_gh); +out: + gfs2_qa_put(ip); return error; } diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c index 7c7197343ee2..9f2b5609f225 100644 --- a/fs/gfs2/lock_dlm.c +++ b/fs/gfs2/lock_dlm.c @@ -16,6 +16,8 @@ #include "incore.h" #include "glock.h" +#include "glops.h" +#include "recovery.h" #include "util.h" #include "sys.h" #include "trace_gfs2.h" @@ -124,6 +126,8 @@ static void gdlm_ast(void *arg) switch (gl->gl_lksb.sb_status) { case -DLM_EUNLOCK: /* Unlocked, so glock can be freed */ + if (gl->gl_ops->go_free) + gl->gl_ops->go_free(gl); gfs2_glock_free(gl); return; case -DLM_ECANCEL: /* Cancel while getting lock */ @@ -323,6 +327,7 @@ static void gdlm_cancel(struct gfs2_glock *gl) /* * dlm/gfs2 recovery coordination using dlm_recover callbacks * + * 0. gfs2 checks for another cluster node withdraw, needing journal replay * 1. dlm_controld sees lockspace members change * 2. dlm_controld blocks dlm-kernel locking activity * 3. dlm_controld within dlm-kernel notifies gfs2 (recover_prep) @@ -571,6 +576,28 @@ static int control_lock(struct gfs2_sbd *sdp, int mode, uint32_t flags) &ls->ls_control_lksb, "control_lock"); } +/** + * remote_withdraw - react to a node withdrawing from the file system + * @sdp: The superblock + */ +static void remote_withdraw(struct gfs2_sbd *sdp) +{ + struct gfs2_jdesc *jd; + int ret = 0, count = 0; + + list_for_each_entry(jd, &sdp->sd_jindex_list, jd_list) { + if (jd->jd_jid == sdp->sd_lockstruct.ls_jid) + continue; + ret = gfs2_recover_journal(jd, true); + if (ret) + break; + count++; + } + + /* Now drop the additional reference we acquired */ + fs_err(sdp, "Journals checked: %d, ret = %d.\n", count, ret); +} + static void gfs2_control_func(struct work_struct *work) { struct gfs2_sbd *sdp = container_of(work, struct gfs2_sbd, sd_control_work.work); @@ -581,6 +608,13 @@ static void gfs2_control_func(struct work_struct *work) int recover_size; int i, error; + /* First check for other nodes that may have done a withdraw. */ + if (test_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags)) { + remote_withdraw(sdp); + clear_bit(SDF_REMOTE_WITHDRAW, &sdp->sd_flags); + return; + } + spin_lock(&ls->ls_recover_spin); /* * No MOUNT_DONE means we're still mounting; control_mount() @@ -1079,6 +1113,10 @@ static void gdlm_recover_prep(void *arg) struct gfs2_sbd *sdp = arg; struct lm_lockstruct *ls = &sdp->sd_lockstruct; + if (gfs2_withdrawn(sdp)) { + fs_err(sdp, "recover_prep ignored due to withdraw.\n"); + return; + } spin_lock(&ls->ls_recover_spin); ls->ls_recover_block = ls->ls_recover_start; set_bit(DFL_DLM_RECOVERY, &ls->ls_recover_flags); @@ -1101,6 +1139,11 @@ static void gdlm_recover_slot(void *arg, struct dlm_slot *slot) struct lm_lockstruct *ls = &sdp->sd_lockstruct; int jid = slot->slot - 1; + if (gfs2_withdrawn(sdp)) { + fs_err(sdp, "recover_slot jid %d ignored due to withdraw.\n", + jid); + return; + } spin_lock(&ls->ls_recover_spin); if (ls->ls_recover_size < jid + 1) { fs_err(sdp, "recover_slot jid %d gen %u short size %d\n", @@ -1125,6 +1168,10 @@ static void gdlm_recover_done(void *arg, struct dlm_slot *slots, int num_slots, struct gfs2_sbd *sdp = arg; struct lm_lockstruct *ls = &sdp->sd_lockstruct; + if (gfs2_withdrawn(sdp)) { + fs_err(sdp, "recover_done ignored due to withdraw.\n"); + return; + } /* ensure the ls jid arrays are large enough */ set_recover_size(sdp, slots, num_slots); @@ -1152,6 +1199,11 @@ static void gdlm_recovery_result(struct gfs2_sbd *sdp, unsigned int jid, { struct lm_lockstruct *ls = &sdp->sd_lockstruct; + if (gfs2_withdrawn(sdp)) { + fs_err(sdp, "recovery_result jid %d ignored due to withdraw.\n", + jid); + return; + } if (test_bit(DFL_NO_DLM_OPS, &ls->ls_recover_flags)) return; diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 00a2e721a374..3a75843ae580 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -88,8 +88,7 @@ static void gfs2_remove_from_ail(struct gfs2_bufdata *bd) static int gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct writeback_control *wbc, - struct gfs2_trans *tr, - bool *withdraw) + struct gfs2_trans *tr) __releases(&sdp->sd_ail_lock) __acquires(&sdp->sd_ail_lock) { @@ -97,6 +96,7 @@ __acquires(&sdp->sd_ail_lock) struct address_space *mapping; struct gfs2_bufdata *bd, *s; struct buffer_head *bh; + int ret = 0; list_for_each_entry_safe_reverse(bd, s, &tr->tr_ail1_list, bd_ail_st_list) { bh = bd->bd_bh; @@ -104,16 +104,21 @@ __acquires(&sdp->sd_ail_lock) gfs2_assert(sdp, bd->bd_tr == tr); if (!buffer_busy(bh)) { - if (!buffer_uptodate(bh) && - !test_and_set_bit(SDF_AIL1_IO_ERROR, - &sdp->sd_flags)) { + if (buffer_uptodate(bh)) { + list_move(&bd->bd_ail_st_list, + &tr->tr_ail2_list); + continue; + } + if (!cmpxchg(&sdp->sd_log_error, 0, -EIO)) { gfs2_io_error_bh(sdp, bh); - *withdraw = true; + gfs2_withdraw_delayed(sdp); } - list_move(&bd->bd_ail_st_list, &tr->tr_ail2_list); - continue; } + if (gfs2_withdrawn(sdp)) { + gfs2_remove_from_ail(bd); + continue; + } if (!buffer_dirty(bh)) continue; if (gl == bd->bd_gl) @@ -124,16 +129,50 @@ __acquires(&sdp->sd_ail_lock) if (!mapping) continue; spin_unlock(&sdp->sd_ail_lock); - generic_writepages(mapping, wbc); + ret = generic_writepages(mapping, wbc); spin_lock(&sdp->sd_ail_lock); - if (wbc->nr_to_write <= 0) + if (ret || wbc->nr_to_write <= 0) break; - return 1; + return -EBUSY; } - return 0; + return ret; } +static void dump_ail_list(struct gfs2_sbd *sdp) +{ + struct gfs2_trans *tr; + struct gfs2_bufdata *bd; + struct buffer_head *bh; + + fs_err(sdp, "Error: In gfs2_ail1_flush for ten minutes! t=%d\n", + current->journal_info ? 1 : 0); + + list_for_each_entry_reverse(tr, &sdp->sd_ail1_list, tr_list) { + list_for_each_entry_reverse(bd, &tr->tr_ail1_list, + bd_ail_st_list) { + bh = bd->bd_bh; + fs_err(sdp, "bd %p: blk:0x%llx bh=%p ", bd, + (unsigned long long)bd->bd_blkno, bh); + if (!bh) { + fs_err(sdp, "\n"); + continue; + } + fs_err(sdp, "0x%llx up2:%d dirt:%d lkd:%d req:%d " + "map:%d new:%d ar:%d aw:%d delay:%d " + "io err:%d unwritten:%d dfr:%d pin:%d esc:%d\n", + (unsigned long long)bh->b_blocknr, + buffer_uptodate(bh), buffer_dirty(bh), + buffer_locked(bh), buffer_req(bh), + buffer_mapped(bh), buffer_new(bh), + buffer_async_read(bh), buffer_async_write(bh), + buffer_delay(bh), buffer_write_io_error(bh), + buffer_unwritten(bh), + buffer_defer_completion(bh), + buffer_pinned(bh), buffer_escaped(bh)); + } + } +} /** * gfs2_ail1_flush - start writeback of some ail1 entries @@ -149,23 +188,36 @@ void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc) struct list_head *head = &sdp->sd_ail1_list; struct gfs2_trans *tr; struct blk_plug plug; - bool withdraw = false; + int ret; + unsigned long flush_start = jiffies; trace_gfs2_ail_flush(sdp, wbc, 1); blk_start_plug(&plug); spin_lock(&sdp->sd_ail_lock); restart: + ret = 0; + if (time_after(jiffies, flush_start + (HZ * 600))) { + dump_ail_list(sdp); + goto out; + } list_for_each_entry_reverse(tr, head, tr_list) { if (wbc->nr_to_write <= 0) break; - if (gfs2_ail1_start_one(sdp, wbc, tr, &withdraw) && - !gfs2_withdrawn(sdp)) - goto restart; + ret = gfs2_ail1_start_one(sdp, wbc, tr); + if (ret) { + if (ret == -EBUSY) + goto restart; + break; + } } +out: spin_unlock(&sdp->sd_ail_lock); blk_finish_plug(&plug); - if (withdraw) - gfs2_lm_withdraw(sdp, NULL); + if (ret) { + gfs2_lm(sdp, "gfs2_ail1_start_one (generic_writepages) " + "returned: %d\n", ret); + gfs2_withdraw(sdp); + } trace_gfs2_ail_flush(sdp, wbc, 0); } @@ -189,12 +241,13 @@ static void gfs2_ail1_start(struct gfs2_sbd *sdp) /** * gfs2_ail1_empty_one - Check whether or not a trans in the AIL has been synced * @sdp: the filesystem - * @ai: the AIL entry + * @tr: the transaction + * @max_revokes: If nonzero, issue revokes for the bd items for written buffers * */ static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr, - bool *withdraw) + int *max_revokes) { struct gfs2_bufdata *bd, *s; struct buffer_head *bh; @@ -203,12 +256,32 @@ static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr, bd_ail_st_list) { bh = bd->bd_bh; gfs2_assert(sdp, bd->bd_tr == tr); - if (buffer_busy(bh)) + /* + * If another process flagged an io error, e.g. writing to the + * journal, error all other bhs and move them off the ail1 to + * prevent a tight loop when unmount tries to flush ail1, + * regardless of whether they're still busy. If no outside + * errors were found and the buffer is busy, move to the next. + * If the ail buffer is not busy and caught an error, flag it + * for others. + */ + if (!sdp->sd_log_error && buffer_busy(bh)) continue; if (!buffer_uptodate(bh) && - !test_and_set_bit(SDF_AIL1_IO_ERROR, &sdp->sd_flags)) { + !cmpxchg(&sdp->sd_log_error, 0, -EIO)) { gfs2_io_error_bh(sdp, bh); - *withdraw = true; + gfs2_withdraw_delayed(sdp); + } + /* + * If we have space for revokes and the bd is no longer on any + * buf list, we can just add a revoke for it immediately and + * avoid having to put it on the ail2 list, where it would need + * to be revoked later. + */ + if (*max_revokes && list_empty(&bd->bd_list)) { + gfs2_add_revoke(sdp, bd); + (*max_revokes)--; + continue; } list_move(&bd->bd_ail_st_list, &tr->tr_ail2_list); } @@ -217,20 +290,20 @@ static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr, /** * gfs2_ail1_empty - Try to empty the ail1 lists * @sdp: The superblock + * @max_revokes: If non-zero, add revokes where appropriate * * Tries to empty the ail1 lists, starting with the oldest first */ -static int gfs2_ail1_empty(struct gfs2_sbd *sdp) +static int gfs2_ail1_empty(struct gfs2_sbd *sdp, int max_revokes) { struct gfs2_trans *tr, *s; int oldest_tr = 1; int ret; - bool withdraw = false; spin_lock(&sdp->sd_ail_lock); list_for_each_entry_safe_reverse(tr, s, &sdp->sd_ail1_list, tr_list) { - gfs2_ail1_empty_one(sdp, tr, &withdraw); + gfs2_ail1_empty_one(sdp, tr, &max_revokes); if (list_empty(&tr->tr_ail1_list) && oldest_tr) list_move(&tr->tr_list, &sdp->sd_ail2_list); else @@ -239,8 +312,10 @@ static int gfs2_ail1_empty(struct gfs2_sbd *sdp) ret = list_empty(&sdp->sd_ail1_list); spin_unlock(&sdp->sd_ail_lock); - if (withdraw) - gfs2_lm_withdraw(sdp, "fatal: I/O error(s)\n"); + if (test_bit(SDF_WITHDRAWING, &sdp->sd_flags)) { + gfs2_lm(sdp, "fatal: I/O error(s)\n"); + gfs2_withdraw(sdp); + } return ret; } @@ -268,20 +343,17 @@ static void gfs2_ail1_wait(struct gfs2_sbd *sdp) } /** - * gfs2_ail2_empty_one - Check whether or not a trans in the AIL has been synced - * @sdp: the filesystem - * @ai: the AIL entry - * + * gfs2_ail_empty_tr - empty one of the ail lists for a transaction */ -static void gfs2_ail2_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr) +static void gfs2_ail_empty_tr(struct gfs2_sbd *sdp, struct gfs2_trans *tr, + struct list_head *head) { - struct list_head *head = &tr->tr_ail2_list; struct gfs2_bufdata *bd; while (!list_empty(head)) { - bd = list_entry(head->prev, struct gfs2_bufdata, - bd_ail_st_list); + bd = list_first_entry(head, struct gfs2_bufdata, + bd_ail_st_list); gfs2_assert(sdp, bd->bd_tr == tr); gfs2_remove_from_ail(bd); } @@ -303,7 +375,7 @@ static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail) if (!rm) continue; - gfs2_ail2_empty_one(sdp, tr); + gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list); list_del(&tr->tr_list); gfs2_assert_warn(sdp, list_empty(&tr->tr_ail1_list)); gfs2_assert_warn(sdp, list_empty(&tr->tr_ail2_list)); @@ -487,7 +559,7 @@ static unsigned int current_tail(struct gfs2_sbd *sdp) if (list_empty(&sdp->sd_ail1_list)) { tail = sdp->sd_log_head; } else { - tr = list_entry(sdp->sd_ail1_list.prev, struct gfs2_trans, + tr = list_last_entry(&sdp->sd_ail1_list, struct gfs2_trans, tr_list); tail = tr->tr_first; } @@ -512,7 +584,7 @@ static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail) } -static void log_flush_wait(struct gfs2_sbd *sdp) +void log_flush_wait(struct gfs2_sbd *sdp) { DEFINE_WAIT(wait); @@ -549,7 +621,7 @@ static void gfs2_ordered_write(struct gfs2_sbd *sdp) spin_lock(&sdp->sd_ordered_lock); list_sort(NULL, &sdp->sd_log_ordered, &ip_cmp); while (!list_empty(&sdp->sd_log_ordered)) { - ip = list_entry(sdp->sd_log_ordered.next, struct gfs2_inode, i_ordered); + ip = list_first_entry(&sdp->sd_log_ordered, struct gfs2_inode, i_ordered); if (ip->i_inode.i_mapping->nrpages == 0) { test_and_clear_bit(GIF_ORDERED, &ip->i_flags); list_del(&ip->i_ordered); @@ -570,7 +642,7 @@ static void gfs2_ordered_wait(struct gfs2_sbd *sdp) spin_lock(&sdp->sd_ordered_lock); while (!list_empty(&sdp->sd_log_ordered)) { - ip = list_entry(sdp->sd_log_ordered.next, struct gfs2_inode, i_ordered); + ip = list_first_entry(&sdp->sd_log_ordered, struct gfs2_inode, i_ordered); list_del(&ip->i_ordered); WARN_ON(!test_and_clear_bit(GIF_ORDERED, &ip->i_flags)); if (ip->i_inode.i_mapping->nrpages == 0) @@ -616,27 +688,24 @@ void gfs2_glock_remove_revoke(struct gfs2_glock *gl) } } +/** + * gfs2_write_revokes - Add as many revokes to the system transaction as we can + * @sdp: The GFS2 superblock + * + * Our usual strategy is to defer writing revokes as much as we can in the hope + * that we'll eventually overwrite the journal, which will make those revokes + * go away. This changes when we flush the log: at that point, there will + * likely be some left-over space in the last revoke block of that transaction. + * We can fill that space with additional revokes for blocks that have already + * been written back. This will basically come at no cost now, and will save + * us from having to keep track of those blocks on the AIL2 list later. + */ void gfs2_write_revokes(struct gfs2_sbd *sdp) { - struct gfs2_trans *tr; - struct gfs2_bufdata *bd, *tmp; - int have_revokes = 0; + /* number of revokes we still have room for */ int max_revokes = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / sizeof(u64); - gfs2_ail1_empty(sdp); - spin_lock(&sdp->sd_ail_lock); - list_for_each_entry_reverse(tr, &sdp->sd_ail1_list, tr_list) { - list_for_each_entry(bd, &tr->tr_ail2_list, bd_ail_st_list) { - if (list_empty(&bd->bd_list)) { - have_revokes = 1; - goto done; - } - } - } -done: - spin_unlock(&sdp->sd_ail_lock); - if (have_revokes == 0) - return; + gfs2_log_lock(sdp); while (sdp->sd_log_num_revoke > max_revokes) max_revokes += (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header)) / sizeof(u64); max_revokes -= sdp->sd_log_num_revoke; @@ -647,20 +716,7 @@ done: if (!sdp->sd_log_blks_reserved) atomic_dec(&sdp->sd_log_blks_free); } - gfs2_log_lock(sdp); - spin_lock(&sdp->sd_ail_lock); - list_for_each_entry_reverse(tr, &sdp->sd_ail1_list, tr_list) { - list_for_each_entry_safe(bd, tmp, &tr->tr_ail2_list, bd_ail_st_list) { - if (max_revokes == 0) - goto out_of_blocks; - if (!list_empty(&bd->bd_list)) - continue; - gfs2_add_revoke(sdp, bd); - max_revokes--; - } - } -out_of_blocks: - spin_unlock(&sdp->sd_ail_lock); + gfs2_ail1_empty(sdp, max_revokes); gfs2_log_unlock(sdp); if (!sdp->sd_log_num_revoke) { @@ -787,6 +843,40 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags) } /** + * ail_drain - drain the ail lists after a withdraw + * @sdp: Pointer to GFS2 superblock + */ +static void ail_drain(struct gfs2_sbd *sdp) +{ + struct gfs2_trans *tr; + + spin_lock(&sdp->sd_ail_lock); + /* + * For transactions on the sd_ail1_list we need to drain both the + * ail1 and ail2 lists. That's because function gfs2_ail1_start_one + * (temporarily) moves items from its tr_ail1 list to tr_ail2 list + * before revokes are sent for that block. Items on the sd_ail2_list + * should have already gotten beyond that point, so no need. + */ + while (!list_empty(&sdp->sd_ail1_list)) { + tr = list_first_entry(&sdp->sd_ail1_list, struct gfs2_trans, + tr_list); + gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail1_list); + gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list); + list_del(&tr->tr_list); + kfree(tr); + } + while (!list_empty(&sdp->sd_ail2_list)) { + tr = list_first_entry(&sdp->sd_ail2_list, struct gfs2_trans, + tr_list); + gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list); + list_del(&tr->tr_list); + kfree(tr); + } + spin_unlock(&sdp->sd_ail_lock); +} + +/** * gfs2_log_flush - flush incore transaction(s) * @sdp: the filesystem * @gl: The glock structure to flush. If NULL, flush the whole incore log @@ -796,11 +886,18 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags) void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) { - struct gfs2_trans *tr; + struct gfs2_trans *tr = NULL; enum gfs2_freeze_state state = atomic_read(&sdp->sd_freeze_state); down_write(&sdp->sd_log_flush_lock); + /* + * Do this check while holding the log_flush_lock to prevent new + * buffers from being added to the ail via gfs2_pin() + */ + if (gfs2_withdrawn(sdp)) + goto out; + /* Log might have been flushed while we waited for the flush lock */ if (gl && !test_bit(GLF_LFLUSH, &gl->gl_flags)) { up_write(&sdp->sd_log_flush_lock); @@ -819,17 +916,27 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) INIT_LIST_HEAD(&tr->tr_ail2_list); tr->tr_first = sdp->sd_log_flush_head; if (unlikely (state == SFS_FROZEN)) - gfs2_assert_withdraw(sdp, !tr->tr_num_buf_new && !tr->tr_num_databuf_new); + if (gfs2_assert_withdraw_delayed(sdp, + !tr->tr_num_buf_new && !tr->tr_num_databuf_new)) + goto out; } if (unlikely(state == SFS_FROZEN)) - gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke); - gfs2_assert_withdraw(sdp, - sdp->sd_log_num_revoke == sdp->sd_log_committed_revoke); + if (gfs2_assert_withdraw_delayed(sdp, !sdp->sd_log_num_revoke)) + goto out; + if (gfs2_assert_withdraw_delayed(sdp, + sdp->sd_log_num_revoke == sdp->sd_log_committed_revoke)) + goto out; gfs2_ordered_write(sdp); + if (gfs2_withdrawn(sdp)) + goto out; lops_before_commit(sdp, tr); + if (gfs2_withdrawn(sdp)) + goto out; gfs2_log_submit_bio(&sdp->sd_log_bio, REQ_OP_WRITE); + if (gfs2_withdrawn(sdp)) + goto out; if (sdp->sd_log_head != sdp->sd_log_flush_head) { log_flush_wait(sdp); @@ -839,6 +946,8 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) trace_gfs2_log_blocks(sdp, -1); log_write_header(sdp, flags); } + if (gfs2_withdrawn(sdp)) + goto out; lops_after_commit(sdp, tr); gfs2_log_lock(sdp); @@ -859,9 +968,11 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) for (;;) { gfs2_ail1_start(sdp); gfs2_ail1_wait(sdp); - if (gfs2_ail1_empty(sdp)) + if (gfs2_ail1_empty(sdp, 0)) break; } + if (gfs2_withdrawn(sdp)) + goto out; atomic_dec(&sdp->sd_log_blks_free); /* Adjust for unreserved buffer */ trace_gfs2_log_blocks(sdp, -1); log_write_header(sdp, flags); @@ -874,6 +985,12 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) atomic_set(&sdp->sd_freeze_state, SFS_FROZEN); } +out: + if (gfs2_withdrawn(sdp)) { + ail_drain(sdp); /* frees all transactions */ + tr = NULL; + } + trace_gfs2_log_flush(sdp, 0, flags); up_write(&sdp->sd_log_flush_lock); @@ -1016,16 +1133,17 @@ int gfs2_logd(void *data) /* Check for errors writing to the journal */ if (sdp->sd_log_error) { - gfs2_lm_withdraw(sdp, - "GFS2: fsid=%s: error %d: " - "withdrawing the file system to " - "prevent further damage.\n", - sdp->sd_fsname, sdp->sd_log_error); + gfs2_lm(sdp, + "GFS2: fsid=%s: error %d: " + "withdrawing the file system to " + "prevent further damage.\n", + sdp->sd_fsname, sdp->sd_log_error); + gfs2_withdraw(sdp); } did_flush = false; if (gfs2_jrnl_flush_reqd(sdp) || t == 0) { - gfs2_ail1_empty(sdp); + gfs2_ail1_empty(sdp, 0); gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_LOGD_JFLUSH_REQD); did_flush = true; @@ -1034,7 +1152,7 @@ int gfs2_logd(void *data) if (gfs2_ail_flush_reqd(sdp)) { gfs2_ail1_start(sdp); gfs2_ail1_wait(sdp); - gfs2_ail1_empty(sdp); + gfs2_ail1_empty(sdp, 0); gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_LOGD_AIL_FLUSH_REQD); did_flush = true; diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index c0a65e5a126b..c1cd6ae17659 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -73,6 +73,7 @@ extern void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 type); extern void gfs2_log_commit(struct gfs2_sbd *sdp, struct gfs2_trans *trans); extern void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc); +extern void log_flush_wait(struct gfs2_sbd *sdp); extern int gfs2_logd(void *data); extern void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd); diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index c090d5ad3f22..5ea96757afc4 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -203,8 +203,12 @@ static void gfs2_end_log_write(struct bio *bio) struct bvec_iter_all iter_all; if (bio->bi_status) { - fs_err(sdp, "Error %d writing to journal, jid=%u\n", - bio->bi_status, sdp->sd_jdesc->jd_jid); + if (!cmpxchg(&sdp->sd_log_error, 0, (int)bio->bi_status)) + fs_err(sdp, "Error %d writing to journal, jid=%u\n", + bio->bi_status, sdp->sd_jdesc->jd_jid); + gfs2_withdraw_delayed(sdp); + /* prevent more writes to the journal */ + clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); wake_up(&sdp->sd_logd_waitq); } @@ -730,7 +734,7 @@ static void buf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr) head = &tr->tr_buf; while (!list_empty(head)) { - bd = list_entry(head->next, struct gfs2_bufdata, bd_list); + bd = list_first_entry(head, struct gfs2_bufdata, bd_list); list_del_init(&bd->bd_list); gfs2_unpin(sdp, bd->bd_bh, tr); } @@ -900,7 +904,7 @@ static void revoke_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr) struct gfs2_glock *gl; while (!list_empty(head)) { - bd = list_entry(head->next, struct gfs2_bufdata, bd_list); + bd = list_first_entry(head, struct gfs2_bufdata, bd_list); list_del_init(&bd->bd_list); gl = bd->bd_gl; gfs2_glock_remove_revoke(gl); @@ -1079,7 +1083,7 @@ static void databuf_lo_after_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr) head = &tr->tr_databuf; while (!list_empty(head)) { - bd = list_entry(head->next, struct gfs2_bufdata, bd_list); + bd = list_first_entry(head, struct gfs2_bufdata, bd_list); list_del_init(&bd->bd_list); gfs2_unpin(sdp, bd->bd_bh, tr); } diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 0c3772974030..4b72abcf83b2 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -251,7 +251,8 @@ int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags, struct buffer_head *bh, *bhs[2]; int num = 0; - if (unlikely(gfs2_withdrawn(sdp))) { + if (unlikely(gfs2_withdrawn(sdp)) && + (!sdp->sd_jdesc || (blkno != sdp->sd_jdesc->jd_no_addr))) { *bhp = NULL; return -EIO; } diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index a1a8ef7ed3fd..e2b69ffcc6a8 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -552,6 +552,8 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) mutex_lock(&sdp->sd_jindex_mutex); for (;;) { + struct gfs2_inode *jip; + error = gfs2_glock_nq_init(dip->i_gl, LM_ST_SHARED, 0, ji_gh); if (error) break; @@ -591,6 +593,8 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) spin_lock(&sdp->sd_jindex_spin); jd->jd_jid = sdp->sd_journals++; + jip = GFS2_I(jd->jd_inode); + jd->jd_no_addr = jip->i_no_addr; list_add_tail(&jd->jd_list, &sdp->sd_jindex_list); spin_unlock(&sdp->sd_jindex_spin); } @@ -600,48 +604,6 @@ static int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh) return error; } -/** - * check_journal_clean - Make sure a journal is clean for a spectator mount - * @sdp: The GFS2 superblock - * @jd: The journal descriptor - * - * Returns: 0 if the journal is clean or locked, else an error - */ -static int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd) -{ - int error; - struct gfs2_holder j_gh; - struct gfs2_log_header_host head; - struct gfs2_inode *ip; - - ip = GFS2_I(jd->jd_inode); - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_NOEXP | - GL_EXACT | GL_NOCACHE, &j_gh); - if (error) { - fs_err(sdp, "Error locking journal for spectator mount.\n"); - return -EPERM; - } - error = gfs2_jdesc_check(jd); - if (error) { - fs_err(sdp, "Error checking journal for spectator mount.\n"); - goto out_unlock; - } - error = gfs2_find_jhead(jd, &head, false); - if (error) { - fs_err(sdp, "Error parsing journal for spectator mount.\n"); - goto out_unlock; - } - if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) { - error = -EPERM; - fs_err(sdp, "jid=%u: Journal is dirty, so the first mounter " - "must not be a spectator.\n", jd->jd_jid); - } - -out_unlock: - gfs2_glock_dq_uninit(&j_gh); - return error; -} - static int init_journal(struct gfs2_sbd *sdp, int undo) { struct inode *master = d_inode(sdp->sd_master_dir); @@ -694,7 +656,8 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) error = gfs2_glock_nq_num(sdp, sdp->sd_lockstruct.ls_jid, &gfs2_journal_glops, - LM_ST_EXCLUSIVE, LM_FLAG_NOEXP, + LM_ST_EXCLUSIVE, + LM_FLAG_NOEXP | GL_NOCACHE, &sdp->sd_journal_gh); if (error) { fs_err(sdp, "can't acquire journal glock: %d\n", error); @@ -702,6 +665,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) } ip = GFS2_I(sdp->sd_jdesc->jd_inode); + sdp->sd_jinode_gl = ip->i_gl; error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_NOEXP | GL_EXACT | GL_NOCACHE, &sdp->sd_jinode_gh); @@ -732,7 +696,7 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) struct gfs2_jdesc *jd = gfs2_jdesc_find(sdp, x); if (sdp->sd_args.ar_spectator) { - error = check_journal_clean(sdp, jd); + error = check_journal_clean(sdp, jd, true); if (error) goto fail_jinode_gh; continue; @@ -762,10 +726,13 @@ static int init_journal(struct gfs2_sbd *sdp, int undo) return 0; fail_jinode_gh: - if (!sdp->sd_args.ar_spectator) + /* A withdraw may have done dq/uninit so now we need to check it */ + if (!sdp->sd_args.ar_spectator && + gfs2_holder_initialized(&sdp->sd_jinode_gh)) gfs2_glock_dq_uninit(&sdp->sd_jinode_gh); fail_journal_gh: - if (!sdp->sd_args.ar_spectator) + if (!sdp->sd_args.ar_spectator && + gfs2_holder_initialized(&sdp->sd_journal_gh)) gfs2_glock_dq_uninit(&sdp->sd_journal_gh); fail_jindex: gfs2_jindex_free(sdp); diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index e9f93045eb01..cc0c4b5800be 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -115,7 +115,7 @@ static void gfs2_qd_dispose(struct list_head *list) struct gfs2_sbd *sdp; while (!list_empty(list)) { - qd = list_entry(list->next, struct gfs2_quota_data, qd_lru); + qd = list_first_entry(list, struct gfs2_quota_data, qd_lru); sdp = qd->qd_gl->gl_name.ln_sbd; list_del(&qd->qd_lru); @@ -525,11 +525,11 @@ static void qdsb_put(struct gfs2_quota_data *qd) } /** - * gfs2_qa_alloc - make sure we have a quota allocations data structure, - * if necessary + * gfs2_qa_get - make sure we have a quota allocations data structure, + * if necessary * @ip: the inode for this reservation */ -int gfs2_qa_alloc(struct gfs2_inode *ip) +int gfs2_qa_get(struct gfs2_inode *ip) { int error = 0; struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); @@ -540,17 +540,21 @@ int gfs2_qa_alloc(struct gfs2_inode *ip) down_write(&ip->i_rw_mutex); if (ip->i_qadata == NULL) { ip->i_qadata = kmem_cache_zalloc(gfs2_qadata_cachep, GFP_NOFS); - if (!ip->i_qadata) + if (!ip->i_qadata) { error = -ENOMEM; + goto out; + } } + ip->i_qadata->qa_ref++; +out: up_write(&ip->i_rw_mutex); return error; } -void gfs2_qa_delete(struct gfs2_inode *ip, atomic_t *wcount) +void gfs2_qa_put(struct gfs2_inode *ip) { down_write(&ip->i_rw_mutex); - if (ip->i_qadata && ((wcount == NULL) || (atomic_read(wcount) <= 1))) { + if (ip->i_qadata && --ip->i_qadata->qa_ref == 0) { kmem_cache_free(gfs2_qadata_cachep, ip->i_qadata); ip->i_qadata = NULL; } @@ -566,27 +570,27 @@ int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid) if (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF) return 0; - if (ip->i_qadata == NULL) { - error = gfs2_rsqa_alloc(ip); - if (error) - return error; - } + error = gfs2_qa_get(ip); + if (error) + return error; qd = ip->i_qadata->qa_qd; if (gfs2_assert_warn(sdp, !ip->i_qadata->qa_qd_num) || - gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags))) - return -EIO; + gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags))) { + error = -EIO; + goto out; + } error = qdsb_get(sdp, make_kqid_uid(ip->i_inode.i_uid), qd); if (error) - goto out; + goto out_unhold; ip->i_qadata->qa_qd_num++; qd++; error = qdsb_get(sdp, make_kqid_gid(ip->i_inode.i_gid), qd); if (error) - goto out; + goto out_unhold; ip->i_qadata->qa_qd_num++; qd++; @@ -594,7 +598,7 @@ int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid) !uid_eq(uid, ip->i_inode.i_uid)) { error = qdsb_get(sdp, make_kqid_uid(uid), qd); if (error) - goto out; + goto out_unhold; ip->i_qadata->qa_qd_num++; qd++; } @@ -603,14 +607,15 @@ int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid) !gid_eq(gid, ip->i_inode.i_gid)) { error = qdsb_get(sdp, make_kqid_gid(gid), qd); if (error) - goto out; + goto out_unhold; ip->i_qadata->qa_qd_num++; qd++; } -out: +out_unhold: if (error) gfs2_quota_unhold(ip); +out: return error; } @@ -621,6 +626,7 @@ void gfs2_quota_unhold(struct gfs2_inode *ip) if (ip->i_qadata == NULL) return; + gfs2_assert_warn(sdp, !test_bit(GIF_QD_LOCKED, &ip->i_flags)); for (x = 0; x < ip->i_qadata->qa_qd_num; x++) { @@ -628,6 +634,7 @@ void gfs2_quota_unhold(struct gfs2_inode *ip) ip->i_qadata->qa_qd[x] = NULL; } ip->i_qadata->qa_qd_num = 0; + gfs2_qa_put(ip); } static int sort_qd(const void *a, const void *b) @@ -876,7 +883,7 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda) unsigned int nalloc = 0, blocks; int error; - error = gfs2_rsqa_alloc(ip); + error = gfs2_qa_get(ip); if (error) return error; @@ -884,8 +891,10 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda) &data_blocks, &ind_blocks); ghs = kmalloc_array(num_qd, sizeof(struct gfs2_holder), GFP_NOFS); - if (!ghs) - return -ENOMEM; + if (!ghs) { + error = -ENOMEM; + goto out; + } sort(qda, num_qd, sizeof(struct gfs2_quota_data *), sort_qd, NULL); inode_lock(&ip->i_inode); @@ -893,12 +902,12 @@ static int do_sync(unsigned int num_qd, struct gfs2_quota_data **qda) error = gfs2_glock_nq_init(qda[qx]->qd_gl, LM_ST_EXCLUSIVE, GL_NOCACHE, &ghs[qx]); if (error) - goto out; + goto out_dq; } error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &i_gh); if (error) - goto out; + goto out_dq; for (x = 0; x < num_qd; x++) { offset = qd2offset(qda[x]); @@ -950,13 +959,15 @@ out_ipres: gfs2_inplace_release(ip); out_alloc: gfs2_glock_dq_uninit(&i_gh); -out: +out_dq: while (qx--) gfs2_glock_dq_uninit(&ghs[qx]); inode_unlock(&ip->i_inode); kfree(ghs); gfs2_log_flush(ip->i_gl->gl_name.ln_sbd, ip->i_gl, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_DO_SYNC); +out: + gfs2_qa_put(ip); return error; } @@ -1259,6 +1270,7 @@ void gfs2_quota_change(struct gfs2_inode *ip, s64 change, if (ip->i_diskflags & GFS2_DIF_SYSTEM) return; + BUG_ON(ip->i_qadata->qa_ref <= 0); for (x = 0; x < ip->i_qadata->qa_qd_num; x++) { qd = ip->i_qadata->qa_qd[x]; @@ -1441,7 +1453,7 @@ void gfs2_quota_cleanup(struct gfs2_sbd *sdp) spin_lock(&qd_lock); while (!list_empty(head)) { - qd = list_entry(head->prev, struct gfs2_quota_data, qd_list); + qd = list_last_entry(head, struct gfs2_quota_data, qd_list); list_del(&qd->qd_list); @@ -1476,8 +1488,8 @@ static void quotad_error(struct gfs2_sbd *sdp, const char *msg, int error) if (error == 0 || error == -EROFS) return; if (!gfs2_withdrawn(sdp)) { - fs_err(sdp, "gfs2_quotad: %s error %d\n", msg, error); - sdp->sd_log_error = error; + if (!cmpxchg(&sdp->sd_log_error, 0, error)) + fs_err(sdp, "gfs2_quotad: %s error %d\n", msg, error); wake_up(&sdp->sd_logd_waitq); } } @@ -1504,7 +1516,7 @@ static void quotad_check_trunc_list(struct gfs2_sbd *sdp) ip = NULL; spin_lock(&sdp->sd_trunc_lock); if (!list_empty(&sdp->sd_trunc_list)) { - ip = list_entry(sdp->sd_trunc_list.next, + ip = list_first_entry(&sdp->sd_trunc_list, struct gfs2_inode, i_trunc_list); list_del_init(&ip->i_trunc_list); } @@ -1541,6 +1553,8 @@ int gfs2_quotad(void *data) while (!kthread_should_stop()) { + if (gfs2_withdrawn(sdp)) + goto bypass; /* Update the master statfs file */ if (sdp->sd_statfs_force_sync) { int error = gfs2_statfs_sync(sdp->sd_vfs, 0); @@ -1561,6 +1575,7 @@ int gfs2_quotad(void *data) try_to_freeze(); +bypass: t = min(quotad_timeo, statfs_timeo); prepare_to_wait(&sdp->sd_quota_wait, &wait, TASK_INTERRUPTIBLE); @@ -1674,7 +1689,7 @@ static int gfs2_set_dqblk(struct super_block *sb, struct kqid qid, if (error) return error; - error = gfs2_rsqa_alloc(ip); + error = gfs2_qa_get(ip); if (error) goto out_put; @@ -1743,6 +1758,7 @@ out_i: out_q: gfs2_glock_dq_uninit(&q_gh); out_unlockput: + gfs2_qa_put(ip); inode_unlock(&ip->i_inode); out_put: qd_put(qd); diff --git a/fs/gfs2/quota.h b/fs/gfs2/quota.h index 765627d9a91e..7f9ca8ef40fc 100644 --- a/fs/gfs2/quota.h +++ b/fs/gfs2/quota.h @@ -15,8 +15,8 @@ struct gfs2_sbd; #define NO_UID_QUOTA_CHANGE INVALID_UID #define NO_GID_QUOTA_CHANGE INVALID_GID -extern int gfs2_qa_alloc(struct gfs2_inode *ip); -extern void gfs2_qa_delete(struct gfs2_inode *ip, atomic_t *wcount); +extern int gfs2_qa_get(struct gfs2_inode *ip); +extern void gfs2_qa_put(struct gfs2_inode *ip); extern int gfs2_quota_hold(struct gfs2_inode *ip, kuid_t uid, kgid_t gid); extern void gfs2_quota_unhold(struct gfs2_inode *ip); diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index 85f830e56945..96c345f49273 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -111,7 +111,7 @@ void gfs2_revoke_clean(struct gfs2_jdesc *jd) struct gfs2_revoke_replay *rr; while (!list_empty(head)) { - rr = list_entry(head->next, struct gfs2_revoke_replay, rr_list); + rr = list_first_entry(head, struct gfs2_revoke_replay, rr_list); list_del(&rr->rr_list); kfree(rr); } @@ -305,6 +305,11 @@ void gfs2_recover_func(struct work_struct *work) int error = 0; int jlocked = 0; + if (gfs2_withdrawn(sdp)) { + fs_err(sdp, "jid=%u: Recovery not attempted due to withdraw.\n", + jd->jd_jid); + goto fail; + } t_start = ktime_get(); if (sdp->sd_args.ar_spectator) goto fail; @@ -393,6 +398,10 @@ void gfs2_recover_func(struct work_struct *work) fs_info(sdp, "jid=%u: Replaying journal...0x%x to 0x%x\n", jd->jd_jid, head.lh_tail, head.lh_blkno); + /* We take the sd_log_flush_lock here primarily to prevent log + * flushes and simultaneous journal replays from stomping on + * each other wrt sd_log_bio. */ + down_read(&sdp->sd_log_flush_lock); for (pass = 0; pass < 2; pass++) { lops_before_scan(jd, &head, pass); error = foreach_descriptor(jd, head.lh_tail, @@ -403,6 +412,7 @@ void gfs2_recover_func(struct work_struct *work) } clean_journal(jd, &head); + up_read(&sdp->sd_log_flush_lock); gfs2_glock_dq_uninit(&thaw_gh); t_rep = ktime_get(); diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index e7bf91ec231c..a321c34e3d6e 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -457,24 +457,24 @@ void gfs2_rgrp_verify(struct gfs2_rgrpd *rgd) } if (count[0] != rgd->rd_free) { - if (gfs2_consist_rgrpd(rgd)) - fs_err(sdp, "free data mismatch: %u != %u\n", - count[0], rgd->rd_free); + gfs2_lm(sdp, "free data mismatch: %u != %u\n", + count[0], rgd->rd_free); + gfs2_consist_rgrpd(rgd); return; } tmp = rgd->rd_data - rgd->rd_free - rgd->rd_dinodes; if (count[1] != tmp) { - if (gfs2_consist_rgrpd(rgd)) - fs_err(sdp, "used data mismatch: %u != %u\n", - count[1], tmp); + gfs2_lm(sdp, "used data mismatch: %u != %u\n", + count[1], tmp); + gfs2_consist_rgrpd(rgd); return; } if (count[2] + count[3] != rgd->rd_dinodes) { - if (gfs2_consist_rgrpd(rgd)) - fs_err(sdp, "used metadata mismatch: %u != %u\n", - count[2] + count[3], rgd->rd_dinodes); + gfs2_lm(sdp, "used metadata mismatch: %u != %u\n", + count[2] + count[3], rgd->rd_dinodes); + gfs2_consist_rgrpd(rgd); return; } } @@ -590,16 +590,6 @@ void gfs2_free_clones(struct gfs2_rgrpd *rgd) } } -/** - * gfs2_rsqa_alloc - make sure we have a reservation assigned to the inode - * plus a quota allocations data structure, if necessary - * @ip: the inode for this reservation - */ -int gfs2_rsqa_alloc(struct gfs2_inode *ip) -{ - return gfs2_qa_alloc(ip); -} - static void dump_rs(struct seq_file *seq, const struct gfs2_blkreserv *rs, const char *fs_id_buf) { @@ -672,18 +662,17 @@ void gfs2_rs_deltree(struct gfs2_blkreserv *rs) } /** - * gfs2_rsqa_delete - delete a multi-block reservation and quota allocation + * gfs2_rs_delete - delete a multi-block reservation * @ip: The inode for this reservation * @wcount: The inode's write count, or NULL * */ -void gfs2_rsqa_delete(struct gfs2_inode *ip, atomic_t *wcount) +void gfs2_rs_delete(struct gfs2_inode *ip, atomic_t *wcount) { down_write(&ip->i_rw_mutex); if ((wcount == NULL) || (atomic_read(wcount) <= 1)) gfs2_rs_deltree(&ip->i_res); up_write(&ip->i_rw_mutex); - gfs2_qa_delete(ip, wcount); } /** @@ -720,8 +709,12 @@ void gfs2_clear_rgrpd(struct gfs2_sbd *sdp) rb_erase(n, &sdp->sd_rindex_tree); if (gl) { - glock_clear_object(gl, rgd); + if (gl->gl_state != LM_ST_UNLOCKED) { + gfs2_glock_cb(gl, LM_ST_UNLOCKED); + flush_delayed_work(&gl->gl_work); + } gfs2_rgrp_brelse(rgd); + glock_clear_object(gl, rgd); gfs2_glock_put(gl); } @@ -733,17 +726,6 @@ void gfs2_clear_rgrpd(struct gfs2_sbd *sdp) } } -static void gfs2_rindex_print(const struct gfs2_rgrpd *rgd) -{ - struct gfs2_sbd *sdp = rgd->rd_sbd; - - fs_info(sdp, "ri_addr = %llu\n", (unsigned long long)rgd->rd_addr); - fs_info(sdp, "ri_length = %u\n", rgd->rd_length); - fs_info(sdp, "ri_data0 = %llu\n", (unsigned long long)rgd->rd_data0); - fs_info(sdp, "ri_data = %u\n", rgd->rd_data); - fs_info(sdp, "ri_bitbytes = %u\n", rgd->rd_bitbytes); -} - /** * gfs2_compute_bitstructs - Compute the bitmap sizes * @rgd: The resource group descriptor @@ -814,11 +796,20 @@ static int compute_bitstructs(struct gfs2_rgrpd *rgd) } bi = rgd->rd_bits + (length - 1); if ((bi->bi_start + bi->bi_bytes) * GFS2_NBBY != rgd->rd_data) { - if (gfs2_consist_rgrpd(rgd)) { - gfs2_rindex_print(rgd); - fs_err(sdp, "start=%u len=%u offset=%u\n", - bi->bi_start, bi->bi_bytes, bi->bi_offset); - } + gfs2_lm(sdp, + "ri_addr = %llu\n" + "ri_length = %u\n" + "ri_data0 = %llu\n" + "ri_data = %u\n" + "ri_bitbytes = %u\n" + "start=%u len=%u offset=%u\n", + (unsigned long long)rgd->rd_addr, + rgd->rd_length, + (unsigned long long)rgd->rd_data0, + rgd->rd_data, + rgd->rd_bitbytes, + bi->bi_start, bi->bi_bytes, bi->bi_offset); + gfs2_consist_rgrpd(rgd); return -EIO; } @@ -1286,23 +1277,6 @@ void gfs2_rgrp_brelse(struct gfs2_rgrpd *rgd) bi->bi_bh = NULL; } } - -} - -/** - * gfs2_rgrp_go_unlock - Unlock a rgrp glock - * @gh: The glock holder for the resource group - * - */ - -void gfs2_rgrp_go_unlock(struct gfs2_holder *gh) -{ - struct gfs2_rgrpd *rgd = gh->gh_gl->gl_object; - int demote_requested = test_bit(GLF_DEMOTE, &gh->gh_gl->gl_flags) | - test_bit(GLF_PENDING_DEMOTE, &gh->gh_gl->gl_flags); - - if (rgd && demote_requested) - gfs2_rgrp_brelse(rgd); } int gfs2_rgrp_send_discards(struct gfs2_sbd *sdp, u64 offset, @@ -1832,10 +1806,8 @@ static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip struct gfs2_rbm rbm = { .rgd = rgd, .bii = 0, .offset = 0 }; while (1) { - down_write(&sdp->sd_log_flush_lock); error = gfs2_rbm_find(&rbm, GFS2_BLKST_UNLINKED, NULL, NULL, true); - up_write(&sdp->sd_log_flush_lock); if (error == -ENOSPC) break; if (WARN_ON_ONCE(error)) diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index c14a673ae36f..a1d7e14fc55b 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -33,7 +33,6 @@ extern int gfs2_rindex_update(struct gfs2_sbd *sdp); extern void gfs2_free_clones(struct gfs2_rgrpd *rgd); extern int gfs2_rgrp_go_lock(struct gfs2_holder *gh); extern void gfs2_rgrp_brelse(struct gfs2_rgrpd *rgd); -extern void gfs2_rgrp_go_unlock(struct gfs2_holder *gh); extern struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip); @@ -45,9 +44,8 @@ extern void gfs2_inplace_release(struct gfs2_inode *ip); extern int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *n, bool dinode, u64 *generation); -extern int gfs2_rsqa_alloc(struct gfs2_inode *ip); extern void gfs2_rs_deltree(struct gfs2_blkreserv *rs); -extern void gfs2_rsqa_delete(struct gfs2_inode *ip, atomic_t *wcount); +extern void gfs2_rs_delete(struct gfs2_inode *ip, atomic_t *wcount); extern void __gfs2_free_blocks(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd, u64 bstart, u32 blen, int meta); extern void gfs2_free_meta(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd, diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 68cc7c291a81..37fc41632aa2 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -61,11 +61,13 @@ void gfs2_jindex_free(struct gfs2_sbd *sdp) sdp->sd_journals = 0; spin_unlock(&sdp->sd_jindex_spin); + sdp->sd_jdesc = NULL; while (!list_empty(&list)) { - jd = list_entry(list.next, struct gfs2_jdesc, jd_list); + jd = list_first_entry(&list, struct gfs2_jdesc, jd_list); gfs2_free_journal_extents(jd); list_del(&jd->jd_list); iput(jd->jd_inode); + jd->jd_inode = NULL; kfree(jd); } } @@ -171,9 +173,13 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) goto fail_threads; j_gl->gl_ops->go_inval(j_gl, DIO_METADATA); + if (gfs2_withdrawn(sdp)) { + error = -EIO; + goto fail; + } error = gfs2_find_jhead(sdp->sd_jdesc, &head, false); - if (error) + if (error || gfs2_withdrawn(sdp)) goto fail; if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) { @@ -187,7 +193,7 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) gfs2_log_pointers_init(sdp, head.lh_blkno); error = gfs2_quota_init(sdp); - if (error) + if (error || gfs2_withdrawn(sdp)) goto fail; set_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); @@ -446,7 +452,7 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp) out: while (!list_empty(&list)) { - lfcc = list_entry(list.next, struct lfcc, list); + lfcc = list_first_entry(&list, struct lfcc, list); list_del(&lfcc->list); gfs2_glock_dq_uninit(&lfcc->gh); kfree(lfcc); @@ -599,34 +605,63 @@ out: int gfs2_make_fs_ro(struct gfs2_sbd *sdp) { struct gfs2_holder freeze_gh; - int error; - - error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, GL_NOCACHE, - &freeze_gh); - if (error && !gfs2_withdrawn(sdp)) - return error; + int error = 0; + int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); + + gfs2_holder_mark_uninitialized(&freeze_gh); + if (sdp->sd_freeze_gl && + !gfs2_glock_is_locked_by_me(sdp->sd_freeze_gl)) { + if (!log_write_allowed) { + error = gfs2_glock_nq_init(sdp->sd_freeze_gl, + LM_ST_SHARED, GL_NOCACHE | + LM_FLAG_TRY, &freeze_gh); + if (error == GLR_TRYFAILED) + error = 0; + } else { + error = gfs2_glock_nq_init(sdp->sd_freeze_gl, + LM_ST_SHARED, GL_NOCACHE, + &freeze_gh); + if (error && !gfs2_withdrawn(sdp)) + return error; + } + } flush_workqueue(gfs2_delete_workqueue); - if (sdp->sd_quotad_process) + if (!log_write_allowed && current == sdp->sd_quotad_process) + fs_warn(sdp, "The quotad daemon is withdrawing.\n"); + else if (sdp->sd_quotad_process) kthread_stop(sdp->sd_quotad_process); sdp->sd_quotad_process = NULL; - if (sdp->sd_logd_process) + + if (!log_write_allowed && current == sdp->sd_logd_process) + fs_warn(sdp, "The logd daemon is withdrawing.\n"); + else if (sdp->sd_logd_process) kthread_stop(sdp->sd_logd_process); sdp->sd_logd_process = NULL; - gfs2_quota_sync(sdp->sd_vfs, 0); - gfs2_statfs_sync(sdp->sd_vfs, 0); - - gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_SHUTDOWN | - GFS2_LFC_MAKE_FS_RO); - wait_event(sdp->sd_reserving_log_wait, atomic_read(&sdp->sd_reserving_log) == 0); - gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) == sdp->sd_jdesc->jd_blocks); + if (log_write_allowed) { + gfs2_quota_sync(sdp->sd_vfs, 0); + gfs2_statfs_sync(sdp->sd_vfs, 0); + gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_SHUTDOWN | + GFS2_LFC_MAKE_FS_RO); + wait_event(sdp->sd_reserving_log_wait, + atomic_read(&sdp->sd_reserving_log) == 0); + gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) == + sdp->sd_jdesc->jd_blocks); + } else { + wait_event_timeout(sdp->sd_reserving_log_wait, + atomic_read(&sdp->sd_reserving_log) == 0, + HZ * 5); + } if (gfs2_holder_initialized(&freeze_gh)) gfs2_glock_dq_uninit(&freeze_gh); gfs2_quota_cleanup(sdp); + if (!log_write_allowed) + sdp->sd_vfs->s_flags |= SB_RDONLY; + return error; } @@ -677,8 +712,10 @@ restart: gfs2_glock_put(sdp->sd_freeze_gl); if (!sdp->sd_args.ar_spectator) { - gfs2_glock_dq_uninit(&sdp->sd_journal_gh); - gfs2_glock_dq_uninit(&sdp->sd_jinode_gh); + if (gfs2_holder_initialized(&sdp->sd_journal_gh)) + gfs2_glock_dq_uninit(&sdp->sd_journal_gh); + if (gfs2_holder_initialized(&sdp->sd_jinode_gh)) + gfs2_glock_dq_uninit(&sdp->sd_jinode_gh); gfs2_glock_dq_uninit(&sdp->sd_sc_gh); gfs2_glock_dq_uninit(&sdp->sd_qc_gh); iput(sdp->sd_sc_inode); @@ -1356,14 +1393,6 @@ out_unlock: if (gfs2_rs_active(&ip->i_res)) gfs2_rs_deltree(&ip->i_res); - if (gfs2_holder_initialized(&ip->i_iopen_gh)) { - glock_clear_object(ip->i_iopen_gh.gh_gl, ip); - if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags)) { - ip->i_iopen_gh.gh_flags |= GL_NOCACHE; - gfs2_glock_dq(&ip->i_iopen_gh); - } - gfs2_holder_uninit(&ip->i_iopen_gh); - } if (gfs2_holder_initialized(&gh)) { glock_clear_object(ip->i_gl, ip); gfs2_glock_dq_uninit(&gh); @@ -1372,22 +1401,30 @@ out_unlock: fs_warn(sdp, "gfs2_evict_inode: %d\n", error); out: truncate_inode_pages_final(&inode->i_data); - gfs2_rsqa_delete(ip, NULL); + if (ip->i_qadata) + gfs2_assert_warn(sdp, ip->i_qadata->qa_ref == 0); + gfs2_rs_delete(ip, NULL); + gfs2_qa_put(ip); gfs2_ordered_del_inode(ip); clear_inode(inode); gfs2_dir_hash_inval(ip); - glock_clear_object(ip->i_gl, ip); - wait_on_bit_io(&ip->i_flags, GIF_GLOP_PENDING, TASK_UNINTERRUPTIBLE); - gfs2_glock_add_to_lru(ip->i_gl); - gfs2_glock_put_eventually(ip->i_gl); - ip->i_gl = NULL; + if (ip->i_gl) { + glock_clear_object(ip->i_gl, ip); + wait_on_bit_io(&ip->i_flags, GIF_GLOP_PENDING, TASK_UNINTERRUPTIBLE); + gfs2_glock_add_to_lru(ip->i_gl); + gfs2_glock_put_eventually(ip->i_gl); + ip->i_gl = NULL; + } if (gfs2_holder_initialized(&ip->i_iopen_gh)) { struct gfs2_glock *gl = ip->i_iopen_gh.gh_gl; glock_clear_object(gl, ip); - ip->i_iopen_gh.gh_flags |= GL_NOCACHE; + if (test_bit(HIF_HOLDER, &ip->i_iopen_gh.gh_iflags)) { + ip->i_iopen_gh.gh_flags |= GL_NOCACHE; + gfs2_glock_dq(&ip->i_iopen_gh); + } gfs2_glock_hold(gl); - gfs2_glock_dq_uninit(&ip->i_iopen_gh); + gfs2_holder_uninit(&ip->i_iopen_gh); gfs2_glock_put_eventually(gl); } } @@ -1401,6 +1438,7 @@ static struct inode *gfs2_alloc_inode(struct super_block *sb) return NULL; ip->i_flags = 0; ip->i_gl = NULL; + gfs2_holder_mark_uninitialized(&ip->i_iopen_gh); memset(&ip->i_res, 0, sizeof(ip->i_res)); RB_CLEAR_NODE(&ip->i_res.rs_node); ip->i_rahead = 0; diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index b8bf811a1305..51900554ed81 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -26,7 +26,6 @@ extern void gfs2_jindex_free(struct gfs2_sbd *sdp); extern struct gfs2_jdesc *gfs2_jdesc_find(struct gfs2_sbd *sdp, unsigned int jid); extern int gfs2_jdesc_check(struct gfs2_jdesc *jd); - extern int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename, struct gfs2_inode **ipp); diff --git a/fs/gfs2/sys.c b/fs/gfs2/sys.c index 8ccb68f4ed16..d28c41bd69b0 100644 --- a/fs/gfs2/sys.c +++ b/fs/gfs2/sys.c @@ -136,7 +136,8 @@ static ssize_t withdraw_store(struct gfs2_sbd *sdp, const char *buf, size_t len) if (val != 1) return -EINVAL; - gfs2_lm_withdraw(sdp, "withdrawing from cluster at user's request\n"); + gfs2_lm(sdp, "withdrawing from cluster at user's request\n"); + gfs2_withdraw(sdp); return len; } @@ -434,6 +435,8 @@ int gfs2_recover_set(struct gfs2_sbd *sdp, unsigned jid) * never clear the DFL_BLOCK_LOCKS flag, so all our locks would * permanently stop working. */ + if (!sdp->sd_jdesc) + goto out; if (sdp->sd_jdesc->jd_jid == jid && !sdp->sd_args.ar_spectator) goto out; rv = -ENOENT; diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c index a685637a5b55..ffe840505082 100644 --- a/fs/gfs2/trans.c +++ b/fs/gfs2/trans.c @@ -228,6 +228,10 @@ void gfs2_trans_add_meta(struct gfs2_glock *gl, struct buffer_head *bh) fs_info(sdp, "GFS2:adding buf while frozen\n"); gfs2_assert_withdraw(sdp, 0); } + if (unlikely(gfs2_withdrawn(sdp))) { + fs_info(sdp, "GFS2:adding buf while withdrawn! 0x%llx\n", + (unsigned long long)bd->bd_bh->b_blocknr); + } gfs2_pin(sdp, bd->bd_bh); mh->__pad0 = cpu_to_be64(0); mh->mh_jid = cpu_to_be32(sdp->sd_jdesc->jd_jid); diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index ec600b487498..9b64d40ab379 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -11,12 +11,18 @@ #include <linux/buffer_head.h> #include <linux/crc32.h> #include <linux/gfs2_ondisk.h> +#include <linux/delay.h> #include <linux/uaccess.h> #include "gfs2.h" #include "incore.h" #include "glock.h" +#include "glops.h" +#include "log.h" +#include "lops.h" +#include "recovery.h" #include "rgrp.h" +#include "super.h" #include "util.h" struct kmem_cache *gfs2_glock_cachep __read_mostly; @@ -33,32 +39,257 @@ void gfs2_assert_i(struct gfs2_sbd *sdp) fs_emerg(sdp, "fatal assertion failed\n"); } -int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...) +/** + * check_journal_clean - Make sure a journal is clean for a spectator mount + * @sdp: The GFS2 superblock + * @jd: The journal descriptor + * + * Returns: 0 if the journal is clean or locked, else an error + */ +int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, + bool verbose) +{ + int error; + struct gfs2_holder j_gh; + struct gfs2_log_header_host head; + struct gfs2_inode *ip; + + ip = GFS2_I(jd->jd_inode); + error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_NOEXP | + GL_EXACT | GL_NOCACHE, &j_gh); + if (error) { + if (verbose) + fs_err(sdp, "Error %d locking journal for spectator " + "mount.\n", error); + return -EPERM; + } + error = gfs2_jdesc_check(jd); + if (error) { + if (verbose) + fs_err(sdp, "Error checking journal for spectator " + "mount.\n"); + goto out_unlock; + } + error = gfs2_find_jhead(jd, &head, false); + if (error) { + if (verbose) + fs_err(sdp, "Error parsing journal for spectator " + "mount.\n"); + goto out_unlock; + } + if (!(head.lh_flags & GFS2_LOG_HEAD_UNMOUNT)) { + error = -EPERM; + if (verbose) + fs_err(sdp, "jid=%u: Journal is dirty, so the first " + "mounter must not be a spectator.\n", + jd->jd_jid); + } + +out_unlock: + gfs2_glock_dq_uninit(&j_gh); + return error; +} + +static void signal_our_withdraw(struct gfs2_sbd *sdp) +{ + struct gfs2_glock *gl = sdp->sd_live_gh.gh_gl; + struct inode *inode = sdp->sd_jdesc->jd_inode; + struct gfs2_inode *ip = GFS2_I(inode); + u64 no_formal_ino = ip->i_no_formal_ino; + int ret = 0; + int tries; + + if (test_bit(SDF_NORECOVERY, &sdp->sd_flags)) + return; + + /* Prevent any glock dq until withdraw recovery is complete */ + set_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags); + /* + * Don't tell dlm we're bailing until we have no more buffers in the + * wind. If journal had an IO error, the log code should just purge + * the outstanding buffers rather than submitting new IO. Making the + * file system read-only will flush the journal, etc. + * + * During a normal unmount, gfs2_make_fs_ro calls gfs2_log_shutdown + * which clears SDF_JOURNAL_LIVE. In a withdraw, we must not write + * any UNMOUNT log header, so we can't call gfs2_log_shutdown, and + * therefore we need to clear SDF_JOURNAL_LIVE manually. + */ + clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); + if (!sb_rdonly(sdp->sd_vfs)) + ret = gfs2_make_fs_ro(sdp); + + /* + * Drop the glock for our journal so another node can recover it. + */ + if (gfs2_holder_initialized(&sdp->sd_journal_gh)) { + gfs2_glock_dq_wait(&sdp->sd_journal_gh); + gfs2_holder_uninit(&sdp->sd_journal_gh); + } + sdp->sd_jinode_gh.gh_flags |= GL_NOCACHE; + gfs2_glock_dq(&sdp->sd_jinode_gh); + if (test_bit(SDF_FS_FROZEN, &sdp->sd_flags)) { + /* Make sure gfs2_unfreeze works if partially-frozen */ + flush_workqueue(gfs2_freeze_wq); + atomic_set(&sdp->sd_freeze_state, SFS_FROZEN); + thaw_super(sdp->sd_vfs); + } else { + wait_on_bit(&gl->gl_flags, GLF_DEMOTE, TASK_UNINTERRUPTIBLE); + } + + /* + * holder_uninit to force glock_put, to force dlm to let go + */ + gfs2_holder_uninit(&sdp->sd_jinode_gh); + + /* + * Note: We need to be careful here: + * Our iput of jd_inode will evict it. The evict will dequeue its + * glock, but the glock dq will wait for the withdraw unless we have + * exception code in glock_dq. + */ + iput(inode); + /* + * Wait until the journal inode's glock is freed. This allows try locks + * on other nodes to be successful, otherwise we remain the owner of + * the glock as far as dlm is concerned. + */ + if (gl->gl_ops->go_free) { + set_bit(GLF_FREEING, &gl->gl_flags); + wait_on_bit(&gl->gl_flags, GLF_FREEING, TASK_UNINTERRUPTIBLE); + } + + if (sdp->sd_lockstruct.ls_ops->lm_lock == NULL) { /* lock_nolock */ + clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags); + goto skip_recovery; + } + /* + * Dequeue the "live" glock, but keep a reference so it's never freed. + */ + gfs2_glock_hold(gl); + gfs2_glock_dq_wait(&sdp->sd_live_gh); + /* + * We enqueue the "live" glock in EX so that all other nodes + * get a demote request and act on it. We don't really want the + * lock in EX, so we send a "try" lock with 1CB to produce a callback. + */ + fs_warn(sdp, "Requesting recovery of jid %d.\n", + sdp->sd_lockstruct.ls_jid); + gfs2_holder_reinit(LM_ST_EXCLUSIVE, LM_FLAG_TRY_1CB | LM_FLAG_NOEXP, + &sdp->sd_live_gh); + msleep(GL_GLOCK_MAX_HOLD); + /* + * This will likely fail in a cluster, but succeed standalone: + */ + ret = gfs2_glock_nq(&sdp->sd_live_gh); + + /* + * If we actually got the "live" lock in EX mode, there are no other + * nodes available to replay our journal. So we try to replay it + * ourselves. We hold the "live" glock to prevent other mounters + * during recovery, then just dequeue it and reacquire it in our + * normal SH mode. Just in case the problem that caused us to + * withdraw prevents us from recovering our journal (e.g. io errors + * and such) we still check if the journal is clean before proceeding + * but we may wait forever until another mounter does the recovery. + */ + if (ret == 0) { + fs_warn(sdp, "No other mounters found. Trying to recover our " + "own journal jid %d.\n", sdp->sd_lockstruct.ls_jid); + if (gfs2_recover_journal(sdp->sd_jdesc, 1)) + fs_warn(sdp, "Unable to recover our journal jid %d.\n", + sdp->sd_lockstruct.ls_jid); + gfs2_glock_dq_wait(&sdp->sd_live_gh); + gfs2_holder_reinit(LM_ST_SHARED, LM_FLAG_NOEXP | GL_EXACT, + &sdp->sd_live_gh); + gfs2_glock_nq(&sdp->sd_live_gh); + } + + gfs2_glock_queue_put(gl); /* drop the extra reference we acquired */ + clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags); + + /* + * At this point our journal is evicted, so we need to get a new inode + * for it. Once done, we need to call gfs2_find_jhead which + * calls gfs2_map_journal_extents to map it for us again. + * + * Note that we don't really want it to look up a FREE block. The + * GFS2_BLKST_FREE simply overrides a block check in gfs2_inode_lookup + * which would otherwise fail because it requires grabbing an rgrp + * glock, which would fail with -EIO because we're withdrawing. + */ + inode = gfs2_inode_lookup(sdp->sd_vfs, DT_UNKNOWN, + sdp->sd_jdesc->jd_no_addr, no_formal_ino, + GFS2_BLKST_FREE); + if (IS_ERR(inode)) { + fs_warn(sdp, "Reprocessing of jid %d failed with %ld.\n", + sdp->sd_lockstruct.ls_jid, PTR_ERR(inode)); + goto skip_recovery; + } + sdp->sd_jdesc->jd_inode = inode; + + /* + * Now wait until recovery is complete. + */ + for (tries = 0; tries < 10; tries++) { + ret = check_journal_clean(sdp, sdp->sd_jdesc, false); + if (!ret) + break; + msleep(HZ); + fs_warn(sdp, "Waiting for journal recovery jid %d.\n", + sdp->sd_lockstruct.ls_jid); + } +skip_recovery: + if (!ret) + fs_warn(sdp, "Journal recovery complete for jid %d.\n", + sdp->sd_lockstruct.ls_jid); + else + fs_warn(sdp, "Journal recovery skipped for %d until next " + "mount.\n", sdp->sd_lockstruct.ls_jid); + fs_warn(sdp, "Glock dequeues delayed: %lu\n", sdp->sd_glock_dqs_held); + sdp->sd_glock_dqs_held = 0; + wake_up_bit(&sdp->sd_flags, SDF_WITHDRAW_RECOVERY); +} + +void gfs2_lm(struct gfs2_sbd *sdp, const char *fmt, ...) { - struct lm_lockstruct *ls = &sdp->sd_lockstruct; - const struct lm_lockops *lm = ls->ls_ops; - va_list args; struct va_format vaf; + va_list args; if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW && - test_and_set_bit(SDF_WITHDRAWN, &sdp->sd_flags)) - return 0; - - if (fmt) { - va_start(args, fmt); + test_bit(SDF_WITHDRAWN, &sdp->sd_flags)) + return; + + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + fs_err(sdp, "%pV", &vaf); + va_end(args); +} - vaf.fmt = fmt; - vaf.va = &args; +int gfs2_withdraw(struct gfs2_sbd *sdp) +{ + struct lm_lockstruct *ls = &sdp->sd_lockstruct; + const struct lm_lockops *lm = ls->ls_ops; - fs_err(sdp, "%pV", &vaf); + if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW && + test_and_set_bit(SDF_WITHDRAWN, &sdp->sd_flags)) { + if (!test_bit(SDF_WITHDRAW_IN_PROG, &sdp->sd_flags)) + return -1; - va_end(args); + wait_on_bit(&sdp->sd_flags, SDF_WITHDRAW_IN_PROG, + TASK_UNINTERRUPTIBLE); + return -1; } + set_bit(SDF_WITHDRAW_IN_PROG, &sdp->sd_flags); + if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW) { fs_err(sdp, "about to withdraw this file system\n"); BUG_ON(sdp->sd_args.ar_debug); + signal_our_withdraw(sdp); + kobject_uevent(&sdp->sd_kobj, KOBJ_OFFLINE); if (!strcmp(sdp->sd_lockstruct.ls_ops->lm_proto_name, "lock_dlm")) @@ -69,8 +300,11 @@ int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...) lm->lm_unmount(sdp); } set_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags); - fs_err(sdp, "withdrawn\n"); + fs_err(sdp, "File system withdrawn\n"); dump_stack(); + clear_bit(SDF_WITHDRAW_IN_PROG, &sdp->sd_flags); + smp_mb__after_atomic(); + wake_up_bit(&sdp->sd_flags, SDF_WITHDRAW_IN_PROG); } if (sdp->sd_args.ar_errors == GFS2_ERRORS_PANIC) @@ -81,35 +315,45 @@ int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...) /** * gfs2_assert_withdraw_i - Cause the machine to withdraw if @assertion is false - * Returns: -1 if this call withdrew the machine, - * -2 if it was already withdrawn */ -int gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion, - const char *function, char *file, unsigned int line) +void gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion, + const char *function, char *file, unsigned int line, + bool delayed) { - int me; - me = gfs2_lm_withdraw(sdp, - "fatal: assertion \"%s\" failed\n" - " function = %s, file = %s, line = %u\n", - assertion, function, file, line); + if (gfs2_withdrawn(sdp)) + return; + + fs_err(sdp, + "fatal: assertion \"%s\" failed\n" + " function = %s, file = %s, line = %u\n", + assertion, function, file, line); + + /* + * If errors=panic was specified on mount, it won't help to delay the + * withdraw. + */ + if (sdp->sd_args.ar_errors == GFS2_ERRORS_PANIC) + delayed = false; + + if (delayed) + gfs2_withdraw_delayed(sdp); + else + gfs2_withdraw(sdp); dump_stack(); - return (me) ? -1 : -2; } /** * gfs2_assert_warn_i - Print a message to the console if @assertion is false - * Returns: -1 if we printed something - * -2 if we didn't */ -int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, - const char *function, char *file, unsigned int line) +void gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, + const char *function, char *file, unsigned int line) { if (time_before(jiffies, sdp->sd_last_warning + gfs2_tune_get(sdp, gt_complain_secs) * HZ)) - return -2; + return; if (sdp->sd_args.ar_errors == GFS2_ERRORS_WITHDRAW) fs_warn(sdp, "warning: assertion \"%s\" failed at function = %s, file = %s, line = %u\n", @@ -127,69 +371,59 @@ int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, sdp->sd_fsname, function, file, line); sdp->sd_last_warning = jiffies; - - return -1; } /** * gfs2_consist_i - Flag a filesystem consistency error and withdraw - * Returns: -1 if this call withdrew the machine, - * 0 if it was already withdrawn */ -int gfs2_consist_i(struct gfs2_sbd *sdp, int cluster_wide, const char *function, - char *file, unsigned int line) +void gfs2_consist_i(struct gfs2_sbd *sdp, const char *function, + char *file, unsigned int line) { - int rv; - rv = gfs2_lm_withdraw(sdp, - "fatal: filesystem consistency error - function = %s, file = %s, line = %u\n", - function, file, line); - return rv; + gfs2_lm(sdp, + "fatal: filesystem consistency error - function = %s, file = %s, line = %u\n", + function, file, line); + gfs2_withdraw(sdp); } /** * gfs2_consist_inode_i - Flag an inode consistency error and withdraw - * Returns: -1 if this call withdrew the machine, - * 0 if it was already withdrawn */ -int gfs2_consist_inode_i(struct gfs2_inode *ip, int cluster_wide, - const char *function, char *file, unsigned int line) +void gfs2_consist_inode_i(struct gfs2_inode *ip, + const char *function, char *file, unsigned int line) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - int rv; - rv = gfs2_lm_withdraw(sdp, - "fatal: filesystem consistency error\n" - " inode = %llu %llu\n" - " function = %s, file = %s, line = %u\n", - (unsigned long long)ip->i_no_formal_ino, - (unsigned long long)ip->i_no_addr, - function, file, line); - return rv; + + gfs2_lm(sdp, + "fatal: filesystem consistency error\n" + " inode = %llu %llu\n" + " function = %s, file = %s, line = %u\n", + (unsigned long long)ip->i_no_formal_ino, + (unsigned long long)ip->i_no_addr, + function, file, line); + gfs2_withdraw(sdp); } /** * gfs2_consist_rgrpd_i - Flag a RG consistency error and withdraw - * Returns: -1 if this call withdrew the machine, - * 0 if it was already withdrawn */ -int gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, int cluster_wide, - const char *function, char *file, unsigned int line) +void gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, + const char *function, char *file, unsigned int line) { struct gfs2_sbd *sdp = rgd->rd_sbd; char fs_id_buf[sizeof(sdp->sd_fsname) + 7]; - int rv; sprintf(fs_id_buf, "fsid=%s: ", sdp->sd_fsname); gfs2_rgrp_dump(NULL, rgd->rd_gl, fs_id_buf); - rv = gfs2_lm_withdraw(sdp, - "fatal: filesystem consistency error\n" - " RG = %llu\n" - " function = %s, file = %s, line = %u\n", - (unsigned long long)rgd->rd_addr, - function, file, line); - return rv; + gfs2_lm(sdp, + "fatal: filesystem consistency error\n" + " RG = %llu\n" + " function = %s, file = %s, line = %u\n", + (unsigned long long)rgd->rd_addr, + function, file, line); + gfs2_withdraw(sdp); } /** @@ -203,12 +437,14 @@ int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, unsigned int line) { int me; - me = gfs2_lm_withdraw(sdp, - "fatal: invalid metadata block\n" - " bh = %llu (%s)\n" - " function = %s, file = %s, line = %u\n", - (unsigned long long)bh->b_blocknr, type, - function, file, line); + + gfs2_lm(sdp, + "fatal: invalid metadata block\n" + " bh = %llu (%s)\n" + " function = %s, file = %s, line = %u\n", + (unsigned long long)bh->b_blocknr, type, + function, file, line); + me = gfs2_withdraw(sdp); return (me) ? -1 : -2; } @@ -223,12 +459,14 @@ int gfs2_metatype_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, char *file, unsigned int line) { int me; - me = gfs2_lm_withdraw(sdp, - "fatal: invalid metadata block\n" - " bh = %llu (type: exp=%u, found=%u)\n" - " function = %s, file = %s, line = %u\n", - (unsigned long long)bh->b_blocknr, type, t, - function, file, line); + + gfs2_lm(sdp, + "fatal: invalid metadata block\n" + " bh = %llu (type: exp=%u, found=%u)\n" + " function = %s, file = %s, line = %u\n", + (unsigned long long)bh->b_blocknr, type, t, + function, file, line); + me = gfs2_withdraw(sdp); return (me) ? -1 : -2; } @@ -241,12 +479,11 @@ int gfs2_metatype_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function, char *file, unsigned int line) { - int rv; - rv = gfs2_lm_withdraw(sdp, - "fatal: I/O error\n" - " function = %s, file = %s, line = %u\n", - function, file, line); - return rv; + gfs2_lm(sdp, + "fatal: I/O error\n" + " function = %s, file = %s, line = %u\n", + function, file, line); + return gfs2_withdraw(sdp); } /** @@ -258,14 +495,14 @@ void gfs2_io_error_bh_i(struct gfs2_sbd *sdp, struct buffer_head *bh, const char *function, char *file, unsigned int line, bool withdraw) { - if (!gfs2_withdrawn(sdp)) - fs_err(sdp, - "fatal: I/O error\n" - " block = %llu\n" - " function = %s, file = %s, line = %u\n", - (unsigned long long)bh->b_blocknr, - function, file, line); + if (gfs2_withdrawn(sdp)) + return; + + fs_err(sdp, "fatal: I/O error\n" + " block = %llu\n" + " function = %s, file = %s, line = %u\n", + (unsigned long long)bh->b_blocknr, function, file, line); if (withdraw) - gfs2_lm_withdraw(sdp, NULL); + gfs2_withdraw(sdp); } diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index f2702bc9837c..a3542560da6f 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -36,41 +36,59 @@ do { \ } while (0) -int gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion, - const char *function, char *file, unsigned int line); +void gfs2_assert_withdraw_i(struct gfs2_sbd *sdp, char *assertion, + const char *function, char *file, unsigned int line, + bool delayed); #define gfs2_assert_withdraw(sdp, assertion) \ -((likely(assertion)) ? 0 : gfs2_assert_withdraw_i((sdp), #assertion, \ - __func__, __FILE__, __LINE__)) - - -int gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, - const char *function, char *file, unsigned int line); + ({ \ + bool _bool = (assertion); \ + if (unlikely(!_bool)) \ + gfs2_assert_withdraw_i((sdp), #assertion, \ + __func__, __FILE__, __LINE__, false); \ + !_bool; \ + }) + +#define gfs2_assert_withdraw_delayed(sdp, assertion) \ + ({ \ + bool _bool = (assertion); \ + if (unlikely(!_bool)) \ + gfs2_assert_withdraw_i((sdp), #assertion, \ + __func__, __FILE__, __LINE__, true); \ + !_bool; \ + }) + +void gfs2_assert_warn_i(struct gfs2_sbd *sdp, char *assertion, + const char *function, char *file, unsigned int line); #define gfs2_assert_warn(sdp, assertion) \ -((likely(assertion)) ? 0 : gfs2_assert_warn_i((sdp), #assertion, \ - __func__, __FILE__, __LINE__)) - + ({ \ + bool _bool = (assertion); \ + if (unlikely(!_bool)) \ + gfs2_assert_warn_i((sdp), #assertion, \ + __func__, __FILE__, __LINE__); \ + !_bool; \ + }) -int gfs2_consist_i(struct gfs2_sbd *sdp, int cluster_wide, - const char *function, char *file, unsigned int line); +void gfs2_consist_i(struct gfs2_sbd *sdp, + const char *function, char *file, unsigned int line); #define gfs2_consist(sdp) \ -gfs2_consist_i((sdp), 0, __func__, __FILE__, __LINE__) +gfs2_consist_i((sdp), __func__, __FILE__, __LINE__) -int gfs2_consist_inode_i(struct gfs2_inode *ip, int cluster_wide, - const char *function, char *file, unsigned int line); +void gfs2_consist_inode_i(struct gfs2_inode *ip, + const char *function, char *file, unsigned int line); #define gfs2_consist_inode(ip) \ -gfs2_consist_inode_i((ip), 0, __func__, __FILE__, __LINE__) +gfs2_consist_inode_i((ip), __func__, __FILE__, __LINE__) -int gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, int cluster_wide, - const char *function, char *file, unsigned int line); +void gfs2_consist_rgrpd_i(struct gfs2_rgrpd *rgd, + const char *function, char *file, unsigned int line); #define gfs2_consist_rgrpd(rgd) \ -gfs2_consist_rgrpd_i((rgd), 0, __func__, __FILE__, __LINE__) +gfs2_consist_rgrpd_i((rgd), __func__, __FILE__, __LINE__) int gfs2_meta_check_ii(struct gfs2_sbd *sdp, struct buffer_head *bh, @@ -129,6 +147,9 @@ static inline void gfs2_metatype_set(struct buffer_head *bh, u16 type, int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function, char *file, unsigned int line); +extern int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, + bool verbose); + #define gfs2_io_error(sdp) \ gfs2_io_error_i((sdp), __func__, __FILE__, __LINE__); @@ -165,18 +186,29 @@ static inline unsigned int gfs2_tune_get_i(struct gfs2_tune *gt, } /** + * gfs2_withdraw_delayed - withdraw as soon as possible without deadlocks + * @sdp: the superblock + */ +static inline void gfs2_withdraw_delayed(struct gfs2_sbd *sdp) +{ + set_bit(SDF_WITHDRAWING, &sdp->sd_flags); +} + +/** * gfs2_withdrawn - test whether the file system is withdrawing or withdrawn * @sdp: the superblock */ static inline bool gfs2_withdrawn(struct gfs2_sbd *sdp) { - return test_bit(SDF_WITHDRAWN, &sdp->sd_flags); + return test_bit(SDF_WITHDRAWN, &sdp->sd_flags) || + test_bit(SDF_WITHDRAWING, &sdp->sd_flags); } #define gfs2_tune_get(sdp, field) \ gfs2_tune_get_i(&(sdp)->sd_tune, &(sdp)->sd_tune.field) __printf(2, 3) -int gfs2_lm_withdraw(struct gfs2_sbd *sdp, const char *fmt, ...); +void gfs2_lm(struct gfs2_sbd *sdp, const char *fmt, ...); +int gfs2_withdraw(struct gfs2_sbd *sdp); #endif /* __UTIL_DOT_H__ */ diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c index bbe593d16bea..9d7667bc4292 100644 --- a/fs/gfs2/xattr.c +++ b/fs/gfs2/xattr.c @@ -1222,7 +1222,7 @@ static int gfs2_xattr_set(const struct xattr_handler *handler, struct gfs2_holder gh; int ret; - ret = gfs2_rsqa_alloc(ip); + ret = gfs2_qa_get(ip); if (ret) return ret; @@ -1231,15 +1231,19 @@ static int gfs2_xattr_set(const struct xattr_handler *handler, if (!gfs2_glock_is_locked_by_me(ip->i_gl)) { ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, 0, &gh); if (ret) - return ret; + goto out; } else { - if (WARN_ON_ONCE(ip->i_gl->gl_state != LM_ST_EXCLUSIVE)) - return -EIO; + if (WARN_ON_ONCE(ip->i_gl->gl_state != LM_ST_EXCLUSIVE)) { + ret = -EIO; + goto out; + } gfs2_holder_mark_uninitialized(&gh); } ret = __gfs2_xattr_set(inode, name, value, size, flags, handler->flags); if (gfs2_holder_initialized(&gh)) gfs2_glock_dq_uninit(&gh); +out: + gfs2_qa_put(ip); return ret; } diff --git a/fs/inode.c b/fs/inode.c index 7d57068b6b7a..93d9252a00ab 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -138,6 +138,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) inode->i_sb = sb; inode->i_blkbits = sb->s_blocksize_bits; inode->i_flags = 0; + atomic64_set(&inode->i_sequence, 0); atomic_set(&inode->i_count, 1); inode->i_op = &empty_iops; inode->i_fop = &no_open_fops; diff --git a/fs/internal.h b/fs/internal.h index f3f280b952a3..4d37912a5587 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -38,7 +38,6 @@ static inline int __sync_blockdev(struct block_device *bdev, int wait) /* * buffer.c */ -extern void guard_bio_eod(struct bio *bio); extern int __block_write_begin_int(struct page *page, loff_t pos, unsigned len, get_block_t *get_block, struct iomap *iomap); diff --git a/fs/io-wq.c b/fs/io-wq.c index 0a5ab1a8f69a..cc5cf2209fb0 100644 --- a/fs/io-wq.c +++ b/fs/io-wq.c @@ -69,6 +69,8 @@ struct io_worker { #define IO_WQ_HASH_ORDER 5 #endif +#define IO_WQ_NR_HASH_BUCKETS (1u << IO_WQ_HASH_ORDER) + struct io_wqe_acct { unsigned nr_workers; unsigned max_workers; @@ -98,6 +100,7 @@ struct io_wqe { struct list_head all_list; struct io_wq *wq; + struct io_wq_work *hash_tail[IO_WQ_NR_HASH_BUCKETS]; }; /* @@ -107,8 +110,7 @@ struct io_wq { struct io_wqe **wqes; unsigned long state; - get_work_fn *get_work; - put_work_fn *put_work; + free_work_fn *free_work; struct task_struct *manager; struct user_struct *user; @@ -376,26 +378,35 @@ static bool __io_worker_idle(struct io_wqe *wqe, struct io_worker *worker) return __io_worker_unuse(wqe, worker); } -static struct io_wq_work *io_get_next_work(struct io_wqe *wqe, unsigned *hash) +static inline unsigned int io_get_work_hash(struct io_wq_work *work) +{ + return work->flags >> IO_WQ_HASH_SHIFT; +} + +static struct io_wq_work *io_get_next_work(struct io_wqe *wqe) __must_hold(wqe->lock) { struct io_wq_work_node *node, *prev; - struct io_wq_work *work; + struct io_wq_work *work, *tail; + unsigned int hash; wq_list_for_each(node, prev, &wqe->work_list) { work = container_of(node, struct io_wq_work, list); /* not hashed, can run anytime */ - if (!(work->flags & IO_WQ_WORK_HASHED)) { - wq_node_del(&wqe->work_list, node, prev); + if (!io_wq_is_hashed(work)) { + wq_list_del(&wqe->work_list, node, prev); return work; } /* hashed, can run if not already running */ - *hash = work->flags >> IO_WQ_HASH_SHIFT; - if (!(wqe->hash_map & BIT_ULL(*hash))) { - wqe->hash_map |= BIT_ULL(*hash); - wq_node_del(&wqe->work_list, node, prev); + hash = io_get_work_hash(work); + if (!(wqe->hash_map & BIT(hash))) { + wqe->hash_map |= BIT(hash); + /* all items with this hash lie in [work, tail] */ + tail = wqe->hash_tail[hash]; + wqe->hash_tail[hash] = NULL; + wq_list_cut(&wqe->work_list, &tail->list, prev); return work; } } @@ -440,16 +451,49 @@ static void io_wq_switch_creds(struct io_worker *worker, worker->saved_creds = old_creds; } +static void io_impersonate_work(struct io_worker *worker, + struct io_wq_work *work) +{ + if (work->files && current->files != work->files) { + task_lock(current); + current->files = work->files; + task_unlock(current); + } + if (work->fs && current->fs != work->fs) + current->fs = work->fs; + if (work->mm != worker->mm) + io_wq_switch_mm(worker, work); + if (worker->cur_creds != work->creds) + io_wq_switch_creds(worker, work); +} + +static void io_assign_current_work(struct io_worker *worker, + struct io_wq_work *work) +{ + if (work) { + /* flush pending signals before assigning new work */ + if (signal_pending(current)) + flush_signals(current); + cond_resched(); + } + + spin_lock_irq(&worker->lock); + worker->cur_work = work; + spin_unlock_irq(&worker->lock); +} + +static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work); + static void io_worker_handle_work(struct io_worker *worker) __releases(wqe->lock) { - struct io_wq_work *work, *old_work = NULL, *put_work = NULL; struct io_wqe *wqe = worker->wqe; struct io_wq *wq = wqe->wq; do { - unsigned hash = -1U; - + struct io_wq_work *work; + unsigned int hash; +get_next: /* * If we got some work, mark us as busy. If we didn't, but * the list isn't empty, it means we stalled on hashed work. @@ -457,120 +501,80 @@ static void io_worker_handle_work(struct io_worker *worker) * can't make progress, any work completion or insertion will * clear the stalled flag. */ - work = io_get_next_work(wqe, &hash); + work = io_get_next_work(wqe); if (work) __io_worker_busy(wqe, worker, work); else if (!wq_list_empty(&wqe->work_list)) wqe->flags |= IO_WQE_FLAG_STALLED; spin_unlock_irq(&wqe->lock); - if (put_work && wq->put_work) - wq->put_work(old_work); if (!work) break; -next: - /* flush any pending signals before assigning new work */ - if (signal_pending(current)) - flush_signals(current); - - cond_resched(); - - spin_lock_irq(&worker->lock); - worker->cur_work = work; - spin_unlock_irq(&worker->lock); - - if (work->flags & IO_WQ_WORK_CB) - work->func(&work); - - if (work->files && current->files != work->files) { - task_lock(current); - current->files = work->files; - task_unlock(current); - } - if (work->fs && current->fs != work->fs) - current->fs = work->fs; - if (work->mm != worker->mm) - io_wq_switch_mm(worker, work); - if (worker->cur_creds != work->creds) - io_wq_switch_creds(worker, work); - /* - * OK to set IO_WQ_WORK_CANCEL even for uncancellable work, - * the worker function will do the right thing. - */ - if (test_bit(IO_WQ_BIT_CANCEL, &wq->state)) - work->flags |= IO_WQ_WORK_CANCEL; - if (worker->mm) - work->flags |= IO_WQ_WORK_HAS_MM; - - if (wq->get_work && !(work->flags & IO_WQ_WORK_INTERNAL)) { - put_work = work; - wq->get_work(work); - } - - old_work = work; - work->func(&work); - - spin_lock_irq(&worker->lock); - worker->cur_work = NULL; - spin_unlock_irq(&worker->lock); - - spin_lock_irq(&wqe->lock); - - if (hash != -1U) { - wqe->hash_map &= ~BIT_ULL(hash); - wqe->flags &= ~IO_WQE_FLAG_STALLED; - } - if (work && work != old_work) { - spin_unlock_irq(&wqe->lock); - - if (put_work && wq->put_work) { - wq->put_work(put_work); - put_work = NULL; + io_assign_current_work(worker, work); + + /* handle a whole dependent link */ + do { + struct io_wq_work *old_work, *next_hashed, *linked; + + next_hashed = wq_next_work(work); + io_impersonate_work(worker, work); + /* + * OK to set IO_WQ_WORK_CANCEL even for uncancellable + * work, the worker function will do the right thing. + */ + if (test_bit(IO_WQ_BIT_CANCEL, &wq->state)) + work->flags |= IO_WQ_WORK_CANCEL; + + hash = io_get_work_hash(work); + linked = old_work = work; + linked->func(&linked); + linked = (old_work == linked) ? NULL : linked; + + work = next_hashed; + if (!work && linked && !io_wq_is_hashed(linked)) { + work = linked; + linked = NULL; } + io_assign_current_work(worker, work); + wq->free_work(old_work); + + if (linked) + io_wqe_enqueue(wqe, linked); + + if (hash != -1U && !next_hashed) { + spin_lock_irq(&wqe->lock); + wqe->hash_map &= ~BIT_ULL(hash); + wqe->flags &= ~IO_WQE_FLAG_STALLED; + /* dependent work is not hashed */ + hash = -1U; + /* skip unnecessary unlock-lock wqe->lock */ + if (!work) + goto get_next; + spin_unlock_irq(&wqe->lock); + } + } while (work); - /* dependent work not hashed */ - hash = -1U; - goto next; - } + spin_lock_irq(&wqe->lock); } while (1); } -static inline void io_worker_spin_for_work(struct io_wqe *wqe) -{ - int i = 0; - - while (++i < 1000) { - if (io_wqe_run_queue(wqe)) - break; - if (need_resched()) - break; - cpu_relax(); - } -} - static int io_wqe_worker(void *data) { struct io_worker *worker = data; struct io_wqe *wqe = worker->wqe; struct io_wq *wq = wqe->wq; - bool did_work; io_worker_start(wqe, worker); - did_work = false; while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)) { set_current_state(TASK_INTERRUPTIBLE); loop: - if (did_work) - io_worker_spin_for_work(wqe); spin_lock_irq(&wqe->lock); if (io_wqe_run_queue(wqe)) { __set_current_state(TASK_RUNNING); io_worker_handle_work(worker); - did_work = true; goto loop; } - did_work = false; /* drops the lock on success, retry */ if (__io_worker_idle(wqe, worker)) { __release(&wqe->lock); @@ -766,6 +770,40 @@ static bool io_wq_can_queue(struct io_wqe *wqe, struct io_wqe_acct *acct, return true; } +static void io_run_cancel(struct io_wq_work *work, struct io_wqe *wqe) +{ + struct io_wq *wq = wqe->wq; + + do { + struct io_wq_work *old_work = work; + + work->flags |= IO_WQ_WORK_CANCEL; + work->func(&work); + work = (work == old_work) ? NULL : work; + wq->free_work(old_work); + } while (work); +} + +static void io_wqe_insert_work(struct io_wqe *wqe, struct io_wq_work *work) +{ + unsigned int hash; + struct io_wq_work *tail; + + if (!io_wq_is_hashed(work)) { +append: + wq_list_add_tail(&work->list, &wqe->work_list); + return; + } + + hash = io_get_work_hash(work); + tail = wqe->hash_tail[hash]; + wqe->hash_tail[hash] = work; + if (!tail) + goto append; + + wq_list_add_after(&work->list, &tail->list, &wqe->work_list); +} + static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work) { struct io_wqe_acct *acct = io_work_get_acct(wqe, work); @@ -779,14 +817,13 @@ static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work) * It's close enough to not be an issue, fork() has the same delay. */ if (unlikely(!io_wq_can_queue(wqe, acct, work))) { - work->flags |= IO_WQ_WORK_CANCEL; - work->func(&work); + io_run_cancel(work, wqe); return; } work_flags = work->flags; spin_lock_irqsave(&wqe->lock, flags); - wq_list_add_tail(&work->list, &wqe->work_list); + io_wqe_insert_work(wqe, work); wqe->flags &= ~IO_WQE_FLAG_STALLED; spin_unlock_irqrestore(&wqe->lock, flags); @@ -803,19 +840,15 @@ void io_wq_enqueue(struct io_wq *wq, struct io_wq_work *work) } /* - * Enqueue work, hashed by some key. Work items that hash to the same value - * will not be done in parallel. Used to limit concurrent writes, generally - * hashed by inode. + * Work items that hash to the same value will not be done in parallel. + * Used to limit concurrent writes, generally hashed by inode. */ -void io_wq_enqueue_hashed(struct io_wq *wq, struct io_wq_work *work, void *val) +void io_wq_hash_work(struct io_wq_work *work, void *val) { - struct io_wqe *wqe = wq->wqes[numa_node_id()]; - unsigned bit; - + unsigned int bit; bit = hash_ptr(val, IO_WQ_HASH_ORDER); work->flags |= (IO_WQ_WORK_HASHED | (bit << IO_WQ_HASH_SHIFT)); - io_wqe_enqueue(wqe, work); } static bool io_wqe_worker_send_sig(struct io_worker *worker, void *data) @@ -865,14 +898,13 @@ void io_wq_cancel_all(struct io_wq *wq) } struct io_cb_cancel_data { - struct io_wqe *wqe; - work_cancel_fn *cancel; - void *caller_data; + work_cancel_fn *fn; + void *data; }; -static bool io_work_cancel(struct io_worker *worker, void *cancel_data) +static bool io_wq_worker_cancel(struct io_worker *worker, void *data) { - struct io_cb_cancel_data *data = cancel_data; + struct io_cb_cancel_data *match = data; unsigned long flags; bool ret = false; @@ -883,84 +915,7 @@ static bool io_work_cancel(struct io_worker *worker, void *cancel_data) spin_lock_irqsave(&worker->lock, flags); if (worker->cur_work && !(worker->cur_work->flags & IO_WQ_WORK_NO_CANCEL) && - data->cancel(worker->cur_work, data->caller_data)) { - send_sig(SIGINT, worker->task, 1); - ret = true; - } - spin_unlock_irqrestore(&worker->lock, flags); - - return ret; -} - -static enum io_wq_cancel io_wqe_cancel_cb_work(struct io_wqe *wqe, - work_cancel_fn *cancel, - void *cancel_data) -{ - struct io_cb_cancel_data data = { - .wqe = wqe, - .cancel = cancel, - .caller_data = cancel_data, - }; - struct io_wq_work_node *node, *prev; - struct io_wq_work *work; - unsigned long flags; - bool found = false; - - spin_lock_irqsave(&wqe->lock, flags); - wq_list_for_each(node, prev, &wqe->work_list) { - work = container_of(node, struct io_wq_work, list); - - if (cancel(work, cancel_data)) { - wq_node_del(&wqe->work_list, node, prev); - found = true; - break; - } - } - spin_unlock_irqrestore(&wqe->lock, flags); - - if (found) { - work->flags |= IO_WQ_WORK_CANCEL; - work->func(&work); - return IO_WQ_CANCEL_OK; - } - - rcu_read_lock(); - found = io_wq_for_each_worker(wqe, io_work_cancel, &data); - rcu_read_unlock(); - return found ? IO_WQ_CANCEL_RUNNING : IO_WQ_CANCEL_NOTFOUND; -} - -enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, - void *data) -{ - enum io_wq_cancel ret = IO_WQ_CANCEL_NOTFOUND; - int node; - - for_each_node(node) { - struct io_wqe *wqe = wq->wqes[node]; - - ret = io_wqe_cancel_cb_work(wqe, cancel, data); - if (ret != IO_WQ_CANCEL_NOTFOUND) - break; - } - - return ret; -} - -struct work_match { - bool (*fn)(struct io_wq_work *, void *data); - void *data; -}; - -static bool io_wq_worker_cancel(struct io_worker *worker, void *data) -{ - struct work_match *match = data; - unsigned long flags; - bool ret = false; - - spin_lock_irqsave(&worker->lock, flags); - if (match->fn(worker->cur_work, match->data) && - !(worker->cur_work->flags & IO_WQ_WORK_NO_CANCEL)) { + match->fn(worker->cur_work, match->data)) { send_sig(SIGINT, worker->task, 1); ret = true; } @@ -970,7 +925,7 @@ static bool io_wq_worker_cancel(struct io_worker *worker, void *data) } static enum io_wq_cancel io_wqe_cancel_work(struct io_wqe *wqe, - struct work_match *match) + struct io_cb_cancel_data *match) { struct io_wq_work_node *node, *prev; struct io_wq_work *work; @@ -987,7 +942,7 @@ static enum io_wq_cancel io_wqe_cancel_work(struct io_wqe *wqe, work = container_of(node, struct io_wq_work, list); if (match->fn(work, match->data)) { - wq_node_del(&wqe->work_list, node, prev); + wq_list_del(&wqe->work_list, node, prev); found = true; break; } @@ -995,8 +950,7 @@ static enum io_wq_cancel io_wqe_cancel_work(struct io_wqe *wqe, spin_unlock_irqrestore(&wqe->lock, flags); if (found) { - work->flags |= IO_WQ_WORK_CANCEL; - work->func(&work); + io_run_cancel(work, wqe); return IO_WQ_CANCEL_OK; } @@ -1012,22 +966,16 @@ static enum io_wq_cancel io_wqe_cancel_work(struct io_wqe *wqe, return found ? IO_WQ_CANCEL_RUNNING : IO_WQ_CANCEL_NOTFOUND; } -static bool io_wq_work_match(struct io_wq_work *work, void *data) -{ - return work == data; -} - -enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork) +enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, + void *data) { - struct work_match match = { - .fn = io_wq_work_match, - .data = cwork + struct io_cb_cancel_data match = { + .fn = cancel, + .data = data, }; enum io_wq_cancel ret = IO_WQ_CANCEL_NOTFOUND; int node; - cwork->flags |= IO_WQ_WORK_CANCEL; - for_each_node(node) { struct io_wqe *wqe = wq->wqes[node]; @@ -1039,69 +987,28 @@ enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork) return ret; } -static bool io_wq_pid_match(struct io_wq_work *work, void *data) +static bool io_wq_io_cb_cancel_data(struct io_wq_work *work, void *data) { - pid_t pid = (pid_t) (unsigned long) data; - - if (work) - return work->task_pid == pid; - return false; + return work == data; } -enum io_wq_cancel io_wq_cancel_pid(struct io_wq *wq, pid_t pid) +enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork) { - struct work_match match = { - .fn = io_wq_pid_match, - .data = (void *) (unsigned long) pid - }; - enum io_wq_cancel ret = IO_WQ_CANCEL_NOTFOUND; - int node; - - for_each_node(node) { - struct io_wqe *wqe = wq->wqes[node]; - - ret = io_wqe_cancel_work(wqe, &match); - if (ret != IO_WQ_CANCEL_NOTFOUND) - break; - } - - return ret; + return io_wq_cancel_cb(wq, io_wq_io_cb_cancel_data, (void *)cwork); } -struct io_wq_flush_data { - struct io_wq_work work; - struct completion done; -}; - -static void io_wq_flush_func(struct io_wq_work **workptr) +static bool io_wq_pid_match(struct io_wq_work *work, void *data) { - struct io_wq_work *work = *workptr; - struct io_wq_flush_data *data; + pid_t pid = (pid_t) (unsigned long) data; - data = container_of(work, struct io_wq_flush_data, work); - complete(&data->done); + return work->task_pid == pid; } -/* - * Doesn't wait for previously queued work to finish. When this completes, - * it just means that previously queued work was started. - */ -void io_wq_flush(struct io_wq *wq) +enum io_wq_cancel io_wq_cancel_pid(struct io_wq *wq, pid_t pid) { - struct io_wq_flush_data data; - int node; + void *data = (void *) (unsigned long) pid; - for_each_node(node) { - struct io_wqe *wqe = wq->wqes[node]; - - if (!node_online(node)) - continue; - init_completion(&data.done); - INIT_IO_WORK(&data.work, io_wq_flush_func); - data.work.flags |= IO_WQ_WORK_INTERNAL; - io_wqe_enqueue(wqe, &data.work); - wait_for_completion(&data.done); - } + return io_wq_cancel_cb(wq, io_wq_pid_match, data); } struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) @@ -1109,6 +1016,9 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) int ret = -ENOMEM, node; struct io_wq *wq; + if (WARN_ON_ONCE(!data->free_work)) + return ERR_PTR(-EINVAL); + wq = kzalloc(sizeof(*wq), GFP_KERNEL); if (!wq) return ERR_PTR(-ENOMEM); @@ -1119,8 +1029,7 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) return ERR_PTR(-ENOMEM); } - wq->get_work = data->get_work; - wq->put_work = data->put_work; + wq->free_work = data->free_work; /* caller must already hold a reference to this */ wq->user = data->user; @@ -1177,7 +1086,7 @@ err: bool io_wq_get(struct io_wq *wq, struct io_wq_data *data) { - if (data->get_work != wq->get_work || data->put_work != wq->put_work) + if (data->free_work != wq->free_work) return false; return refcount_inc_not_zero(&wq->use_refs); diff --git a/fs/io-wq.h b/fs/io-wq.h index ccc7d84af57d..3ee7356d6be5 100644 --- a/fs/io-wq.h +++ b/fs/io-wq.h @@ -5,11 +5,8 @@ struct io_wq; enum { IO_WQ_WORK_CANCEL = 1, - IO_WQ_WORK_HAS_MM = 2, IO_WQ_WORK_HASHED = 4, IO_WQ_WORK_UNBOUND = 32, - IO_WQ_WORK_INTERNAL = 64, - IO_WQ_WORK_CB = 128, IO_WQ_WORK_NO_CANCEL = 256, IO_WQ_WORK_CONCURRENT = 512, @@ -31,6 +28,18 @@ struct io_wq_work_list { struct io_wq_work_node *last; }; +static inline void wq_list_add_after(struct io_wq_work_node *node, + struct io_wq_work_node *pos, + struct io_wq_work_list *list) +{ + struct io_wq_work_node *next = pos->next; + + pos->next = node; + node->next = next; + if (!next) + list->last = node; +} + static inline void wq_list_add_tail(struct io_wq_work_node *node, struct io_wq_work_list *list) { @@ -43,17 +52,26 @@ static inline void wq_list_add_tail(struct io_wq_work_node *node, } } -static inline void wq_node_del(struct io_wq_work_list *list, - struct io_wq_work_node *node, +static inline void wq_list_cut(struct io_wq_work_list *list, + struct io_wq_work_node *last, struct io_wq_work_node *prev) { - if (node == list->first) - WRITE_ONCE(list->first, node->next); - if (node == list->last) + /* first in the list, if prev==NULL */ + if (!prev) + WRITE_ONCE(list->first, last->next); + else + prev->next = last->next; + + if (last == list->last) list->last = prev; - if (prev) - prev->next = node->next; - node->next = NULL; + last->next = NULL; +} + +static inline void wq_list_del(struct io_wq_work_list *list, + struct io_wq_work_node *node, + struct io_wq_work_node *prev) +{ + wq_list_cut(list, node, prev); } #define wq_list_for_each(pos, prv, head) \ @@ -66,10 +84,7 @@ static inline void wq_node_del(struct io_wq_work_list *list, } while (0) struct io_wq_work { - union { - struct io_wq_work_node list; - void *data; - }; + struct io_wq_work_node list; void (*func)(struct io_wq_work **); struct files_struct *files; struct mm_struct *mm; @@ -79,25 +94,25 @@ struct io_wq_work { pid_t task_pid; }; -#define INIT_IO_WORK(work, _func) \ - do { \ - (work)->list.next = NULL; \ - (work)->func = _func; \ - (work)->files = NULL; \ - (work)->mm = NULL; \ - (work)->creds = NULL; \ - (work)->fs = NULL; \ - (work)->flags = 0; \ - } while (0) \ +#define INIT_IO_WORK(work, _func) \ + do { \ + *(work) = (struct io_wq_work){ .func = _func }; \ + } while (0) \ -typedef void (get_work_fn)(struct io_wq_work *); -typedef void (put_work_fn)(struct io_wq_work *); +static inline struct io_wq_work *wq_next_work(struct io_wq_work *work) +{ + if (!work->list.next) + return NULL; + + return container_of(work->list.next, struct io_wq_work, list); +} + +typedef void (free_work_fn)(struct io_wq_work *); struct io_wq_data { struct user_struct *user; - get_work_fn *get_work; - put_work_fn *put_work; + free_work_fn *free_work; }; struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data); @@ -105,8 +120,12 @@ bool io_wq_get(struct io_wq *wq, struct io_wq_data *data); void io_wq_destroy(struct io_wq *wq); void io_wq_enqueue(struct io_wq *wq, struct io_wq_work *work); -void io_wq_enqueue_hashed(struct io_wq *wq, struct io_wq_work *work, void *val); -void io_wq_flush(struct io_wq *wq); +void io_wq_hash_work(struct io_wq_work *work, void *val); + +static inline bool io_wq_is_hashed(struct io_wq_work *work) +{ + return work->flags & IO_WQ_WORK_HASHED; +} void io_wq_cancel_all(struct io_wq *wq); enum io_wq_cancel io_wq_cancel_work(struct io_wq *wq, struct io_wq_work *cwork); diff --git a/fs/io_uring.c b/fs/io_uring.c index 5a826017ebb8..358f97be9c7b 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -44,6 +44,7 @@ #include <linux/errno.h> #include <linux/syscalls.h> #include <linux/compat.h> +#include <net/compat.h> #include <linux/refcount.h> #include <linux/uio.h> #include <linux/bits.h> @@ -76,6 +77,8 @@ #include <linux/fadvise.h> #include <linux/eventpoll.h> #include <linux/fs_struct.h> +#include <linux/splice.h> +#include <linux/task_work.h> #define CREATE_TRACE_POINTS #include <trace/events/io_uring.h> @@ -183,21 +186,23 @@ struct fixed_file_table { struct file **files; }; -enum { - FFD_F_ATOMIC, -}; - struct fixed_file_data { struct fixed_file_table *table; struct io_ring_ctx *ctx; struct percpu_ref refs; struct llist_head put_llist; - unsigned long state; struct work_struct ref_work; struct completion done; }; +struct io_buffer { + struct list_head list; + __u64 addr; + __s32 len; + __u16 bid; +}; + struct io_ring_ctx { struct { struct percpu_ref refs; @@ -275,6 +280,8 @@ struct io_ring_ctx { struct socket *ring_sock; #endif + struct idr io_buffer_idr; + struct idr personality_idr; struct { @@ -295,7 +302,6 @@ struct io_ring_ctx { struct { spinlock_t completion_lock; - struct llist_head poll_llist; /* * ->poll_list is protected by the ctx->uring_lock for @@ -348,6 +354,7 @@ struct io_accept { struct sockaddr __user *addr; int __user *addr_len; int flags; + unsigned long nofile; }; struct io_sync { @@ -390,7 +397,9 @@ struct io_sr_msg { void __user *buf; }; int msg_flags; + int bgid; size_t len; + struct io_buffer *kbuf; }; struct io_open { @@ -402,6 +411,7 @@ struct io_open { struct filename *filename; struct statx __user *buffer; struct open_how how; + unsigned long nofile; }; struct io_files_update { @@ -433,6 +443,24 @@ struct io_epoll { struct epoll_event event; }; +struct io_splice { + struct file *file_out; + struct file *file_in; + loff_t off_out; + loff_t off_in; + u64 len; + unsigned int flags; +}; + +struct io_provide_buf { + struct file *file; + __u64 addr; + __s32 len; + __u32 bgid; + __u16 nbufs; + __u16 bid; +}; + struct io_async_connect { struct sockaddr_storage address; }; @@ -467,6 +495,7 @@ enum { REQ_F_LINK_BIT = IOSQE_IO_LINK_BIT, REQ_F_HARDLINK_BIT = IOSQE_IO_HARDLINK_BIT, REQ_F_FORCE_ASYNC_BIT = IOSQE_ASYNC_BIT, + REQ_F_BUFFER_SELECT_BIT = IOSQE_BUFFER_SELECT_BIT, REQ_F_LINK_NEXT_BIT, REQ_F_FAIL_LINK_BIT, @@ -482,6 +511,11 @@ enum { REQ_F_COMP_LOCKED_BIT, REQ_F_NEED_CLEANUP_BIT, REQ_F_OVERFLOW_BIT, + REQ_F_POLLED_BIT, + REQ_F_BUFFER_SELECTED_BIT, + + /* not a real bit, just to check we're not overflowing the space */ + __REQ_F_LAST_BIT, }; enum { @@ -495,6 +529,8 @@ enum { REQ_F_HARDLINK = BIT(REQ_F_HARDLINK_BIT), /* IOSQE_ASYNC */ REQ_F_FORCE_ASYNC = BIT(REQ_F_FORCE_ASYNC_BIT), + /* IOSQE_BUFFER_SELECT */ + REQ_F_BUFFER_SELECT = BIT(REQ_F_BUFFER_SELECT_BIT), /* already grabbed next link */ REQ_F_LINK_NEXT = BIT(REQ_F_LINK_NEXT_BIT), @@ -524,6 +560,15 @@ enum { REQ_F_NEED_CLEANUP = BIT(REQ_F_NEED_CLEANUP_BIT), /* in overflow list */ REQ_F_OVERFLOW = BIT(REQ_F_OVERFLOW_BIT), + /* already went through poll handler */ + REQ_F_POLLED = BIT(REQ_F_POLLED_BIT), + /* buffer already selected */ + REQ_F_BUFFER_SELECTED = BIT(REQ_F_BUFFER_SELECTED_BIT), +}; + +struct async_poll { + struct io_poll_iocb poll; + struct io_wq_work work; }; /* @@ -549,32 +594,45 @@ struct io_kiocb { struct io_fadvise fadvise; struct io_madvise madvise; struct io_epoll epoll; + struct io_splice splice; + struct io_provide_buf pbuf; }; struct io_async_ctx *io; - /* - * llist_node is only used for poll deferred completions - */ - struct llist_node llist_node; - bool in_async; bool needs_fixed_file; u8 opcode; struct io_ring_ctx *ctx; - union { - struct list_head list; - struct hlist_node hash_node; - }; - struct list_head link_list; + struct list_head list; unsigned int flags; refcount_t refs; + union { + struct task_struct *task; + unsigned long fsize; + }; u64 user_data; u32 result; u32 sequence; + struct list_head link_list; + struct list_head inflight_entry; - struct io_wq_work work; + union { + /* + * Only commands that never go async can use the below fields, + * obviously. Right now only IORING_OP_POLL_ADD uses them, and + * async armed poll handlers for regular commands. The latter + * restore the work, if needed. + */ + struct { + struct callback_head task_work; + struct hlist_node hash_node; + struct async_poll *apoll; + int cflags; + }; + struct io_wq_work work; + }; }; #define IO_PLUG_THRESHOLD 2 @@ -618,6 +676,11 @@ struct io_op_def { unsigned file_table : 1; /* needs ->fs */ unsigned needs_fs : 1; + /* set if opcode supports polled "wait" */ + unsigned pollin : 1; + unsigned pollout : 1; + /* op supports buffer selection */ + unsigned buffer_select : 1; }; static const struct io_op_def io_op_defs[] = { @@ -627,6 +690,8 @@ static const struct io_op_def io_op_defs[] = { .needs_mm = 1, .needs_file = 1, .unbound_nonreg_file = 1, + .pollin = 1, + .buffer_select = 1, }, [IORING_OP_WRITEV] = { .async_ctx = 1, @@ -634,6 +699,7 @@ static const struct io_op_def io_op_defs[] = { .needs_file = 1, .hash_reg_file = 1, .unbound_nonreg_file = 1, + .pollout = 1, }, [IORING_OP_FSYNC] = { .needs_file = 1, @@ -641,11 +707,13 @@ static const struct io_op_def io_op_defs[] = { [IORING_OP_READ_FIXED] = { .needs_file = 1, .unbound_nonreg_file = 1, + .pollin = 1, }, [IORING_OP_WRITE_FIXED] = { .needs_file = 1, .hash_reg_file = 1, .unbound_nonreg_file = 1, + .pollout = 1, }, [IORING_OP_POLL_ADD] = { .needs_file = 1, @@ -661,6 +729,7 @@ static const struct io_op_def io_op_defs[] = { .needs_file = 1, .unbound_nonreg_file = 1, .needs_fs = 1, + .pollout = 1, }, [IORING_OP_RECVMSG] = { .async_ctx = 1, @@ -668,6 +737,8 @@ static const struct io_op_def io_op_defs[] = { .needs_file = 1, .unbound_nonreg_file = 1, .needs_fs = 1, + .pollin = 1, + .buffer_select = 1, }, [IORING_OP_TIMEOUT] = { .async_ctx = 1, @@ -679,6 +750,7 @@ static const struct io_op_def io_op_defs[] = { .needs_file = 1, .unbound_nonreg_file = 1, .file_table = 1, + .pollin = 1, }, [IORING_OP_ASYNC_CANCEL] = {}, [IORING_OP_LINK_TIMEOUT] = { @@ -690,6 +762,7 @@ static const struct io_op_def io_op_defs[] = { .needs_mm = 1, .needs_file = 1, .unbound_nonreg_file = 1, + .pollout = 1, }, [IORING_OP_FALLOCATE] = { .needs_file = 1, @@ -718,11 +791,14 @@ static const struct io_op_def io_op_defs[] = { .needs_mm = 1, .needs_file = 1, .unbound_nonreg_file = 1, + .pollin = 1, + .buffer_select = 1, }, [IORING_OP_WRITE] = { .needs_mm = 1, .needs_file = 1, .unbound_nonreg_file = 1, + .pollout = 1, }, [IORING_OP_FADVISE] = { .needs_file = 1, @@ -734,11 +810,14 @@ static const struct io_op_def io_op_defs[] = { .needs_mm = 1, .needs_file = 1, .unbound_nonreg_file = 1, + .pollout = 1, }, [IORING_OP_RECV] = { .needs_mm = 1, .needs_file = 1, .unbound_nonreg_file = 1, + .pollin = 1, + .buffer_select = 1, }, [IORING_OP_OPENAT2] = { .needs_file = 1, @@ -750,6 +829,13 @@ static const struct io_op_def io_op_defs[] = { .unbound_nonreg_file = 1, .file_table = 1, }, + [IORING_OP_SPLICE] = { + .needs_file = 1, + .hash_reg_file = 1, + .unbound_nonreg_file = 1, + }, + [IORING_OP_PROVIDE_BUFFERS] = {}, + [IORING_OP_REMOVE_BUFFERS] = {}, }; static void io_wq_submit_work(struct io_wq_work **workptr); @@ -764,6 +850,10 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, static int io_grab_files(struct io_kiocb *req); static void io_ring_file_ref_flush(struct fixed_file_data *data); static void io_cleanup_req(struct io_kiocb *req); +static int io_file_get(struct io_submit_state *state, struct io_kiocb *req, + int fd, struct file **out_file, bool fixed); +static void __io_queue_sqe(struct io_kiocb *req, + const struct io_uring_sqe *sqe); static struct kmem_cache *req_cachep; @@ -830,11 +920,11 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) INIT_LIST_HEAD(&ctx->cq_overflow_list); init_completion(&ctx->completions[0]); init_completion(&ctx->completions[1]); + idr_init(&ctx->io_buffer_idr); idr_init(&ctx->personality_idr); mutex_init(&ctx->uring_lock); init_waitqueue_head(&ctx->wait); spin_lock_init(&ctx->completion_lock); - init_llist_head(&ctx->poll_llist); INIT_LIST_HEAD(&ctx->poll_list); INIT_LIST_HEAD(&ctx->defer_list); INIT_LIST_HEAD(&ctx->timeout_list); @@ -955,15 +1045,14 @@ static inline void io_req_work_drop_env(struct io_kiocb *req) } } -static inline bool io_prep_async_work(struct io_kiocb *req, +static inline void io_prep_async_work(struct io_kiocb *req, struct io_kiocb **link) { const struct io_op_def *def = &io_op_defs[req->opcode]; - bool do_hashed = false; if (req->flags & REQ_F_ISREG) { if (def->hash_reg_file) - do_hashed = true; + io_wq_hash_work(&req->work, file_inode(req->file)); } else { if (def->unbound_nonreg_file) req->work.flags |= IO_WQ_WORK_UNBOUND; @@ -972,25 +1061,18 @@ static inline bool io_prep_async_work(struct io_kiocb *req, io_req_work_grab_env(req, def); *link = io_prep_linked_timeout(req); - return do_hashed; } static inline void io_queue_async_work(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; struct io_kiocb *link; - bool do_hashed; - do_hashed = io_prep_async_work(req, &link); + io_prep_async_work(req, &link); - trace_io_uring_queue_async_work(ctx, do_hashed, req, &req->work, - req->flags); - if (!do_hashed) { - io_wq_enqueue(ctx->io_wq, &req->work); - } else { - io_wq_enqueue_hashed(ctx->io_wq, &req->work, - file_inode(req->file)); - } + trace_io_uring_queue_async_work(ctx, io_wq_is_hashed(&req->work), req, + &req->work, req->flags); + io_wq_enqueue(ctx->io_wq, &req->work); if (link) io_queue_linked_timeout(link); @@ -1004,6 +1086,7 @@ static void io_kill_timeout(struct io_kiocb *req) if (ret != -1) { atomic_inc(&req->ctx->cq_timeouts); list_del_init(&req->list); + req->flags |= REQ_F_COMP_LOCKED; io_cqring_fill_event(req, 0); io_put_req(req); } @@ -1056,24 +1139,19 @@ static inline bool io_should_trigger_evfd(struct io_ring_ctx *ctx) return false; if (!ctx->eventfd_async) return true; - return io_wq_current_is_worker() || in_interrupt(); + return io_wq_current_is_worker(); } -static void __io_cqring_ev_posted(struct io_ring_ctx *ctx, bool trigger_ev) +static void io_cqring_ev_posted(struct io_ring_ctx *ctx) { if (waitqueue_active(&ctx->wait)) wake_up(&ctx->wait); if (waitqueue_active(&ctx->sqo_wait)) wake_up(&ctx->sqo_wait); - if (trigger_ev) + if (io_should_trigger_evfd(ctx)) eventfd_signal(ctx->cq_ev_fd, 1); } -static void io_cqring_ev_posted(struct io_ring_ctx *ctx) -{ - __io_cqring_ev_posted(ctx, io_should_trigger_evfd(ctx)); -} - /* Returns true if there are no backlogged entries after the flush */ static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force) { @@ -1110,7 +1188,7 @@ static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force) if (cqe) { WRITE_ONCE(cqe->user_data, req->user_data); WRITE_ONCE(cqe->res, req->result); - WRITE_ONCE(cqe->flags, 0); + WRITE_ONCE(cqe->flags, req->cflags); } else { WRITE_ONCE(ctx->rings->cq_overflow, atomic_inc_return(&ctx->cached_cq_overflow)); @@ -1134,7 +1212,7 @@ static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force) return cqe != NULL; } -static void io_cqring_fill_event(struct io_kiocb *req, long res) +static void __io_cqring_fill_event(struct io_kiocb *req, long res, long cflags) { struct io_ring_ctx *ctx = req->ctx; struct io_uring_cqe *cqe; @@ -1150,7 +1228,7 @@ static void io_cqring_fill_event(struct io_kiocb *req, long res) if (likely(cqe)) { WRITE_ONCE(cqe->user_data, req->user_data); WRITE_ONCE(cqe->res, res); - WRITE_ONCE(cqe->flags, 0); + WRITE_ONCE(cqe->flags, cflags); } else if (ctx->cq_overflow_flushed) { WRITE_ONCE(ctx->rings->cq_overflow, atomic_inc_return(&ctx->cached_cq_overflow)); @@ -1162,23 +1240,34 @@ static void io_cqring_fill_event(struct io_kiocb *req, long res) req->flags |= REQ_F_OVERFLOW; refcount_inc(&req->refs); req->result = res; + req->cflags = cflags; list_add_tail(&req->list, &ctx->cq_overflow_list); } } -static void io_cqring_add_event(struct io_kiocb *req, long res) +static void io_cqring_fill_event(struct io_kiocb *req, long res) +{ + __io_cqring_fill_event(req, res, 0); +} + +static void __io_cqring_add_event(struct io_kiocb *req, long res, long cflags) { struct io_ring_ctx *ctx = req->ctx; unsigned long flags; spin_lock_irqsave(&ctx->completion_lock, flags); - io_cqring_fill_event(req, res); + __io_cqring_fill_event(req, res, cflags); io_commit_cqring(ctx); spin_unlock_irqrestore(&ctx->completion_lock, flags); io_cqring_ev_posted(ctx); } +static void io_cqring_add_event(struct io_kiocb *req, long res) +{ + __io_cqring_add_event(req, res, 0); +} + static inline bool io_is_fallback_req(struct io_kiocb *req) { return req == (struct io_kiocb *) @@ -1248,6 +1337,15 @@ fallback: return NULL; } +static inline void io_put_file(struct io_kiocb *req, struct file *file, + bool fixed) +{ + if (fixed) + percpu_ref_put(&req->ctx->file_data->refs); + else + fput(file); +} + static void __io_req_do_free(struct io_kiocb *req) { if (likely(!io_is_fallback_req(req))) @@ -1258,15 +1356,12 @@ static void __io_req_do_free(struct io_kiocb *req) static void __io_req_aux_free(struct io_kiocb *req) { - struct io_ring_ctx *ctx = req->ctx; + if (req->flags & REQ_F_NEED_CLEANUP) + io_cleanup_req(req); kfree(req->io); - if (req->file) { - if (req->flags & REQ_F_FIXED_FILE) - percpu_ref_put(&ctx->file_data->refs); - else - fput(req->file); - } + if (req->file) + io_put_file(req, req->file, (req->flags & REQ_F_FIXED_FILE)); io_req_work_drop_env(req); } @@ -1275,9 +1370,6 @@ static void __io_free_req(struct io_kiocb *req) { __io_req_aux_free(req); - if (req->flags & REQ_F_NEED_CLEANUP) - io_cleanup_req(req); - if (req->flags & REQ_F_INFLIGHT) { struct io_ring_ctx *ctx = req->ctx; unsigned long flags; @@ -1476,6 +1568,30 @@ static void io_free_req(struct io_kiocb *req) io_queue_async_work(nxt); } +static void io_link_work_cb(struct io_wq_work **workptr) +{ + struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); + struct io_kiocb *link; + + link = list_first_entry(&req->link_list, struct io_kiocb, link_list); + io_queue_linked_timeout(link); + io_wq_submit_work(workptr); +} + +static void io_wq_assign_next(struct io_wq_work **workptr, struct io_kiocb *nxt) +{ + struct io_kiocb *link; + const struct io_op_def *def = &io_op_defs[nxt->opcode]; + + if ((nxt->flags & REQ_F_ISREG) && def->hash_reg_file) + io_wq_hash_work(&nxt->work, file_inode(nxt->file)); + + *workptr = &nxt->work; + link = io_prep_linked_timeout(nxt); + if (link) + nxt->work.func = io_link_work_cb; +} + /* * Drop reference to request, return next in chain (if there is one) if this * was the last reference to this request. @@ -1483,10 +1599,10 @@ static void io_free_req(struct io_kiocb *req) __attribute__((nonnull)) static void io_put_req_find_next(struct io_kiocb *req, struct io_kiocb **nxtptr) { - io_req_find_next(req, nxtptr); - - if (refcount_dec_and_test(&req->refs)) + if (refcount_dec_and_test(&req->refs)) { + io_req_find_next(req, nxtptr); __io_free_req(req); + } } static void io_put_req(struct io_kiocb *req) @@ -1495,6 +1611,26 @@ static void io_put_req(struct io_kiocb *req) io_free_req(req); } +static void io_steal_work(struct io_kiocb *req, + struct io_wq_work **workptr) +{ + /* + * It's in an io-wq worker, so there always should be at least + * one reference, which will be dropped in io_put_work() just + * after the current handler returns. + * + * It also means, that if the counter dropped to 1, then there is + * no asynchronous users left, so it's safe to steal the next work. + */ + if (refcount_read(&req->refs) == 1) { + struct io_kiocb *nxt = NULL; + + io_req_find_next(req, &nxt); + if (nxt) + io_wq_assign_next(workptr, nxt); + } +} + /* * Must only be used if we don't need to care about links, usually from * within the completion handling itself. @@ -1556,6 +1692,19 @@ static inline bool io_req_multi_free(struct req_batch *rb, struct io_kiocb *req) return true; } +static int io_put_kbuf(struct io_kiocb *req) +{ + struct io_buffer *kbuf; + int cflags; + + kbuf = (struct io_buffer *) (unsigned long) req->rw.addr; + cflags = kbuf->bid << IORING_CQE_BUFFER_SHIFT; + cflags |= IORING_CQE_F_BUFFER; + req->rw.addr = 0; + kfree(kbuf); + return cflags; +} + /* * Find and free completed poll iocbs */ @@ -1567,10 +1716,15 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events, rb.to_free = rb.need_iter = 0; while (!list_empty(done)) { + int cflags = 0; + req = list_first_entry(done, struct io_kiocb, list); list_del(&req->list); - io_cqring_fill_event(req, req->result); + if (req->flags & REQ_F_BUFFER_SELECTED) + cflags = io_put_kbuf(req); + + __io_cqring_fill_event(req, req->result, cflags); (*nr_events)++; if (refcount_dec_and_test(&req->refs) && @@ -1579,6 +1733,8 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events, } io_commit_cqring(ctx); + if (ctx->flags & IORING_SETUP_SQPOLL) + io_cqring_ev_posted(ctx); io_free_req_many(ctx, &rb); } @@ -1672,11 +1828,17 @@ static void io_iopoll_reap_events(struct io_ring_ctx *ctx) mutex_unlock(&ctx->uring_lock); } -static int __io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events, - long min) +static int io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events, + long min) { int iters = 0, ret = 0; + /* + * We disallow the app entering submit/complete with polling, but we + * still need to lock the ring to prevent racing with polled issue + * that got punted to a workqueue. + */ + mutex_lock(&ctx->uring_lock); do { int tmin = 0; @@ -1712,21 +1874,6 @@ static int __io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events, ret = 0; } while (min && !*nr_events && !need_resched()); - return ret; -} - -static int io_iopoll_check(struct io_ring_ctx *ctx, unsigned *nr_events, - long min) -{ - int ret; - - /* - * We disallow the app entering submit/complete with polling, but we - * still need to lock the ring to prevent racing with polled issue - * that got punted to a workqueue. - */ - mutex_lock(&ctx->uring_lock); - ret = __io_iopoll_check(ctx, nr_events, min); mutex_unlock(&ctx->uring_lock); return ret; } @@ -1754,13 +1901,16 @@ static inline void req_set_fail_links(struct io_kiocb *req) static void io_complete_rw_common(struct kiocb *kiocb, long res) { struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); + int cflags = 0; if (kiocb->ki_flags & IOCB_WRITE) kiocb_end_write(req); if (res != req->result) req_set_fail_links(req); - io_cqring_add_event(req, res); + if (req->flags & REQ_F_BUFFER_SELECTED) + cflags = io_put_kbuf(req); + __io_cqring_add_event(req, res, cflags); } static void io_complete_rw(struct kiocb *kiocb, long res, long res2) @@ -1771,17 +1921,6 @@ static void io_complete_rw(struct kiocb *kiocb, long res, long res2) io_put_req(req); } -static struct io_kiocb *__io_complete_rw(struct kiocb *kiocb, long res) -{ - struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); - struct io_kiocb *nxt = NULL; - - io_complete_rw_common(kiocb, res); - io_put_req_find_next(req, &nxt); - - return nxt; -} - static void io_complete_rw_iopoll(struct kiocb *kiocb, long res, long res2) { struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); @@ -1830,6 +1969,10 @@ static void io_iopoll_req_issued(struct io_kiocb *req) list_add(&req->list, &ctx->poll_list); else list_add_tail(&req->list, &ctx->poll_list); + + if ((ctx->flags & IORING_SETUP_SQPOLL) && + wq_has_sleeper(&ctx->sqo_wait)) + wake_up(&ctx->sqo_wait); } static void io_file_put(struct io_submit_state *state) @@ -1848,7 +1991,7 @@ static void io_file_put(struct io_submit_state *state) * assuming most submissions are for one file, or at least that each file * has more than one submission. */ -static struct file *io_file_get(struct io_submit_state *state, int fd) +static struct file *__io_file_get(struct io_submit_state *state, int fd) { if (!state) return fget(fd); @@ -1945,7 +2088,7 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe, req->rw.addr = READ_ONCE(sqe->addr); req->rw.len = READ_ONCE(sqe->len); - /* we own ->private, reuse it for the buffer index */ + /* we own ->private, reuse it for the buffer index / buffer ID */ req->rw.kiocb.private = (void *) (unsigned long) READ_ONCE(sqe->buf_index); return 0; @@ -1972,15 +2115,14 @@ static inline void io_rw_done(struct kiocb *kiocb, ssize_t ret) } } -static void kiocb_done(struct kiocb *kiocb, ssize_t ret, struct io_kiocb **nxt, - bool in_async) +static void kiocb_done(struct kiocb *kiocb, ssize_t ret) { struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); if (req->flags & REQ_F_CUR_POS) req->file->f_pos = kiocb->ki_pos; - if (in_async && ret >= 0 && kiocb->ki_complete == io_complete_rw) - *nxt = __io_complete_rw(kiocb, ret); + if (ret >= 0 && kiocb->ki_complete == io_complete_rw) + io_complete_rw(kiocb, ret, 0); else io_rw_done(kiocb, ret); } @@ -2059,11 +2201,147 @@ static ssize_t io_import_fixed(struct io_kiocb *req, int rw, return len; } +static void io_ring_submit_unlock(struct io_ring_ctx *ctx, bool needs_lock) +{ + if (needs_lock) + mutex_unlock(&ctx->uring_lock); +} + +static void io_ring_submit_lock(struct io_ring_ctx *ctx, bool needs_lock) +{ + /* + * "Normal" inline submissions always hold the uring_lock, since we + * grab it from the system call. Same is true for the SQPOLL offload. + * The only exception is when we've detached the request and issue it + * from an async worker thread, grab the lock for that case. + */ + if (needs_lock) + mutex_lock(&ctx->uring_lock); +} + +static struct io_buffer *io_buffer_select(struct io_kiocb *req, size_t *len, + int bgid, struct io_buffer *kbuf, + bool needs_lock) +{ + struct io_buffer *head; + + if (req->flags & REQ_F_BUFFER_SELECTED) + return kbuf; + + io_ring_submit_lock(req->ctx, needs_lock); + + lockdep_assert_held(&req->ctx->uring_lock); + + head = idr_find(&req->ctx->io_buffer_idr, bgid); + if (head) { + if (!list_empty(&head->list)) { + kbuf = list_last_entry(&head->list, struct io_buffer, + list); + list_del(&kbuf->list); + } else { + kbuf = head; + idr_remove(&req->ctx->io_buffer_idr, bgid); + } + if (*len > kbuf->len) + *len = kbuf->len; + } else { + kbuf = ERR_PTR(-ENOBUFS); + } + + io_ring_submit_unlock(req->ctx, needs_lock); + + return kbuf; +} + +static void __user *io_rw_buffer_select(struct io_kiocb *req, size_t *len, + bool needs_lock) +{ + struct io_buffer *kbuf; + int bgid; + + kbuf = (struct io_buffer *) (unsigned long) req->rw.addr; + bgid = (int) (unsigned long) req->rw.kiocb.private; + kbuf = io_buffer_select(req, len, bgid, kbuf, needs_lock); + if (IS_ERR(kbuf)) + return kbuf; + req->rw.addr = (u64) (unsigned long) kbuf; + req->flags |= REQ_F_BUFFER_SELECTED; + return u64_to_user_ptr(kbuf->addr); +} + +#ifdef CONFIG_COMPAT +static ssize_t io_compat_import(struct io_kiocb *req, struct iovec *iov, + bool needs_lock) +{ + struct compat_iovec __user *uiov; + compat_ssize_t clen; + void __user *buf; + ssize_t len; + + uiov = u64_to_user_ptr(req->rw.addr); + if (!access_ok(uiov, sizeof(*uiov))) + return -EFAULT; + if (__get_user(clen, &uiov->iov_len)) + return -EFAULT; + if (clen < 0) + return -EINVAL; + + len = clen; + buf = io_rw_buffer_select(req, &len, needs_lock); + if (IS_ERR(buf)) + return PTR_ERR(buf); + iov[0].iov_base = buf; + iov[0].iov_len = (compat_size_t) len; + return 0; +} +#endif + +static ssize_t __io_iov_buffer_select(struct io_kiocb *req, struct iovec *iov, + bool needs_lock) +{ + struct iovec __user *uiov = u64_to_user_ptr(req->rw.addr); + void __user *buf; + ssize_t len; + + if (copy_from_user(iov, uiov, sizeof(*uiov))) + return -EFAULT; + + len = iov[0].iov_len; + if (len < 0) + return -EINVAL; + buf = io_rw_buffer_select(req, &len, needs_lock); + if (IS_ERR(buf)) + return PTR_ERR(buf); + iov[0].iov_base = buf; + iov[0].iov_len = len; + return 0; +} + +static ssize_t io_iov_buffer_select(struct io_kiocb *req, struct iovec *iov, + bool needs_lock) +{ + if (req->flags & REQ_F_BUFFER_SELECTED) + return 0; + if (!req->rw.len) + return 0; + else if (req->rw.len > 1) + return -EINVAL; + +#ifdef CONFIG_COMPAT + if (req->ctx->compat) + return io_compat_import(req, iov, needs_lock); +#endif + + return __io_iov_buffer_select(req, iov, needs_lock); +} + static ssize_t io_import_iovec(int rw, struct io_kiocb *req, - struct iovec **iovec, struct iov_iter *iter) + struct iovec **iovec, struct iov_iter *iter, + bool needs_lock) { void __user *buf = u64_to_user_ptr(req->rw.addr); size_t sqe_len = req->rw.len; + ssize_t ret; u8 opcode; opcode = req->opcode; @@ -2072,15 +2350,23 @@ static ssize_t io_import_iovec(int rw, struct io_kiocb *req, return io_import_fixed(req, rw, iter); } - /* buffer index only valid with fixed read/write */ - if (req->rw.kiocb.private) + /* buffer index only valid with fixed read/write, or buffer select */ + if (req->rw.kiocb.private && !(req->flags & REQ_F_BUFFER_SELECT)) return -EINVAL; if (opcode == IORING_OP_READ || opcode == IORING_OP_WRITE) { - ssize_t ret; + if (req->flags & REQ_F_BUFFER_SELECT) { + buf = io_rw_buffer_select(req, &sqe_len, needs_lock); + if (IS_ERR(buf)) { + *iovec = NULL; + return PTR_ERR(buf); + } + req->rw.len = sqe_len; + } + ret = import_single_range(rw, buf, sqe_len, *iovec, iter); *iovec = NULL; - return ret; + return ret < 0 ? ret : sqe_len; } if (req->io) { @@ -2093,6 +2379,16 @@ static ssize_t io_import_iovec(int rw, struct io_kiocb *req, return iorw->size; } + if (req->flags & REQ_F_BUFFER_SELECT) { + ret = io_iov_buffer_select(req, *iovec, needs_lock); + if (!ret) { + ret = (*iovec)->iov_len; + iov_iter_init(iter, rw, *iovec, 1, ret); + } + *iovec = NULL; + return ret; + } + #ifdef CONFIG_COMPAT if (req->ctx->compat) return compat_import_iovec(rw, buf, sqe_len, UIO_FASTIOV, @@ -2176,12 +2472,18 @@ static void io_req_map_rw(struct io_kiocb *req, ssize_t io_size, } } +static inline int __io_alloc_async_ctx(struct io_kiocb *req) +{ + req->io = kmalloc(sizeof(*req->io), GFP_KERNEL); + return req->io == NULL; +} + static int io_alloc_async_ctx(struct io_kiocb *req) { if (!io_op_defs[req->opcode].async_ctx) return 0; - req->io = kmalloc(sizeof(*req->io), GFP_KERNEL); - return req->io == NULL; + + return __io_alloc_async_ctx(req); } static int io_setup_async_rw(struct io_kiocb *req, ssize_t io_size, @@ -2191,7 +2493,7 @@ static int io_setup_async_rw(struct io_kiocb *req, ssize_t io_size, if (!io_op_defs[req->opcode].async_ctx) return 0; if (!req->io) { - if (io_alloc_async_ctx(req)) + if (__io_alloc_async_ctx(req)) return -ENOMEM; io_req_map_rw(req, io_size, iovec, fast_iov, iter); @@ -2220,7 +2522,7 @@ static int io_read_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe, io = req->io; io->rw.iov = io->rw.fast_iov; req->io = NULL; - ret = io_import_iovec(READ, req, &io->rw.iov, &iter); + ret = io_import_iovec(READ, req, &io->rw.iov, &iter, !force_nonblock); req->io = io; if (ret < 0) return ret; @@ -2229,8 +2531,7 @@ static int io_read_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe, return 0; } -static int io_read(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_read(struct io_kiocb *req, bool force_nonblock) { struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs; struct kiocb *kiocb = &req->rw.kiocb; @@ -2238,13 +2539,13 @@ static int io_read(struct io_kiocb *req, struct io_kiocb **nxt, size_t iov_count; ssize_t io_size, ret; - ret = io_import_iovec(READ, req, &iovec, &iter); + ret = io_import_iovec(READ, req, &iovec, &iter, !force_nonblock); if (ret < 0) return ret; /* Ensure we clear previously set non-block flag */ if (!force_nonblock) - req->rw.kiocb.ki_flags &= ~IOCB_NOWAIT; + kiocb->ki_flags &= ~IOCB_NOWAIT; req->result = 0; io_size = ret; @@ -2255,10 +2556,8 @@ static int io_read(struct io_kiocb *req, struct io_kiocb **nxt, * If the file doesn't support async, mark it as REQ_F_MUST_PUNT so * we know to async punt it even if it was opened O_NONBLOCK */ - if (force_nonblock && !io_file_supports_async(req->file)) { - req->flags |= REQ_F_MUST_PUNT; + if (force_nonblock && !io_file_supports_async(req->file)) goto copy_iov; - } iov_count = iov_iter_count(&iter); ret = rw_verify_area(READ, req->file, &kiocb->ki_pos, iov_count); @@ -2272,13 +2571,16 @@ static int io_read(struct io_kiocb *req, struct io_kiocb **nxt, /* Catch -EAGAIN return for forced non-blocking submission */ if (!force_nonblock || ret2 != -EAGAIN) { - kiocb_done(kiocb, ret2, nxt, req->in_async); + kiocb_done(kiocb, ret2); } else { copy_iov: ret = io_setup_async_rw(req, io_size, iovec, inline_vecs, &iter); if (ret) goto out_free; + /* any defer here is final, must blocking retry */ + if (!(req->flags & REQ_F_NOWAIT)) + req->flags |= REQ_F_MUST_PUNT; return -EAGAIN; } } @@ -2302,6 +2604,8 @@ static int io_write_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (unlikely(!(req->file->f_mode & FMODE_WRITE))) return -EBADF; + req->fsize = rlimit(RLIMIT_FSIZE); + /* either don't need iovec imported or already have it */ if (!req->io || req->flags & REQ_F_NEED_CLEANUP) return 0; @@ -2309,7 +2613,7 @@ static int io_write_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe, io = req->io; io->rw.iov = io->rw.fast_iov; req->io = NULL; - ret = io_import_iovec(WRITE, req, &io->rw.iov, &iter); + ret = io_import_iovec(WRITE, req, &io->rw.iov, &iter, !force_nonblock); req->io = io; if (ret < 0) return ret; @@ -2318,8 +2622,7 @@ static int io_write_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe, return 0; } -static int io_write(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_write(struct io_kiocb *req, bool force_nonblock) { struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs; struct kiocb *kiocb = &req->rw.kiocb; @@ -2327,7 +2630,7 @@ static int io_write(struct io_kiocb *req, struct io_kiocb **nxt, size_t iov_count; ssize_t ret, io_size; - ret = io_import_iovec(WRITE, req, &iovec, &iter); + ret = io_import_iovec(WRITE, req, &iovec, &iter, !force_nonblock); if (ret < 0) return ret; @@ -2344,10 +2647,8 @@ static int io_write(struct io_kiocb *req, struct io_kiocb **nxt, * If the file doesn't support async, mark it as REQ_F_MUST_PUNT so * we know to async punt it even if it was opened O_NONBLOCK */ - if (force_nonblock && !io_file_supports_async(req->file)) { - req->flags |= REQ_F_MUST_PUNT; + if (force_nonblock && !io_file_supports_async(req->file)) goto copy_iov; - } /* file path doesn't support NOWAIT for non-direct_IO */ if (force_nonblock && !(kiocb->ki_flags & IOCB_DIRECT) && @@ -2374,24 +2675,33 @@ static int io_write(struct io_kiocb *req, struct io_kiocb **nxt, } kiocb->ki_flags |= IOCB_WRITE; + if (!force_nonblock) + current->signal->rlim[RLIMIT_FSIZE].rlim_cur = req->fsize; + if (req->file->f_op->write_iter) ret2 = call_write_iter(req->file, kiocb, &iter); else ret2 = loop_rw_iter(WRITE, req->file, kiocb, &iter); + + if (!force_nonblock) + current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; + /* - * Raw bdev writes will -EOPNOTSUPP for IOCB_NOWAIT. Just + * Raw bdev writes will return -EOPNOTSUPP for IOCB_NOWAIT. Just * retry them without IOCB_NOWAIT. */ if (ret2 == -EOPNOTSUPP && (kiocb->ki_flags & IOCB_NOWAIT)) ret2 = -EAGAIN; if (!force_nonblock || ret2 != -EAGAIN) { - kiocb_done(kiocb, ret2, nxt, req->in_async); + kiocb_done(kiocb, ret2); } else { copy_iov: ret = io_setup_async_rw(req, io_size, iovec, inline_vecs, &iter); if (ret) goto out_free; + /* any defer here is final, must blocking retry */ + req->flags |= REQ_F_MUST_PUNT; return -EAGAIN; } } @@ -2401,6 +2711,76 @@ out_free: return ret; } +static int io_splice_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) +{ + struct io_splice* sp = &req->splice; + unsigned int valid_flags = SPLICE_F_FD_IN_FIXED | SPLICE_F_ALL; + int ret; + + if (req->flags & REQ_F_NEED_CLEANUP) + return 0; + + sp->file_in = NULL; + sp->off_in = READ_ONCE(sqe->splice_off_in); + sp->off_out = READ_ONCE(sqe->off); + sp->len = READ_ONCE(sqe->len); + sp->flags = READ_ONCE(sqe->splice_flags); + + if (unlikely(sp->flags & ~valid_flags)) + return -EINVAL; + + ret = io_file_get(NULL, req, READ_ONCE(sqe->splice_fd_in), &sp->file_in, + (sp->flags & SPLICE_F_FD_IN_FIXED)); + if (ret) + return ret; + req->flags |= REQ_F_NEED_CLEANUP; + + if (!S_ISREG(file_inode(sp->file_in)->i_mode)) + req->work.flags |= IO_WQ_WORK_UNBOUND; + + return 0; +} + +static bool io_splice_punt(struct file *file) +{ + if (get_pipe_info(file)) + return false; + if (!io_file_supports_async(file)) + return true; + return !(file->f_mode & O_NONBLOCK); +} + +static int io_splice(struct io_kiocb *req, bool force_nonblock) +{ + struct io_splice *sp = &req->splice; + struct file *in = sp->file_in; + struct file *out = sp->file_out; + unsigned int flags = sp->flags & ~SPLICE_F_FD_IN_FIXED; + loff_t *poff_in, *poff_out; + long ret; + + if (force_nonblock) { + if (io_splice_punt(in) || io_splice_punt(out)) + return -EAGAIN; + flags |= SPLICE_F_NONBLOCK; + } + + poff_in = (sp->off_in == -1) ? NULL : &sp->off_in; + poff_out = (sp->off_out == -1) ? NULL : &sp->off_out; + ret = do_splice(in, poff_in, out, poff_out, sp->len, flags); + if (force_nonblock && ret == -EAGAIN) + return -EAGAIN; + + io_put_file(req, in, (sp->flags & SPLICE_F_FD_IN_FIXED)); + req->flags &= ~REQ_F_NEED_CLEANUP; + + io_cqring_add_event(req, ret); + if (ret != sp->len) + req_set_fail_links(req); + io_put_req(req); + return 0; +} + /* * IORING_OP_NOP just posts a completion event, nothing else. */ @@ -2449,82 +2829,63 @@ static bool io_req_cancelled(struct io_kiocb *req) return false; } -static void io_link_work_cb(struct io_wq_work **workptr) -{ - struct io_wq_work *work = *workptr; - struct io_kiocb *link = work->data; - - io_queue_linked_timeout(link); - work->func = io_wq_submit_work; -} - -static void io_wq_assign_next(struct io_wq_work **workptr, struct io_kiocb *nxt) -{ - struct io_kiocb *link; - - io_prep_async_work(nxt, &link); - *workptr = &nxt->work; - if (link) { - nxt->work.flags |= IO_WQ_WORK_CB; - nxt->work.func = io_link_work_cb; - nxt->work.data = link; - } -} - -static void io_fsync_finish(struct io_wq_work **workptr) +static void __io_fsync(struct io_kiocb *req) { - struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); loff_t end = req->sync.off + req->sync.len; - struct io_kiocb *nxt = NULL; int ret; - if (io_req_cancelled(req)) - return; - ret = vfs_fsync_range(req->file, req->sync.off, end > 0 ? end : LLONG_MAX, req->sync.flags & IORING_FSYNC_DATASYNC); if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, &nxt); - if (nxt) - io_wq_assign_next(workptr, nxt); + io_put_req(req); } -static int io_fsync(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static void io_fsync_finish(struct io_wq_work **workptr) { - struct io_wq_work *work, *old_work; + struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); + + if (io_req_cancelled(req)) + return; + __io_fsync(req); + io_steal_work(req, workptr); +} +static int io_fsync(struct io_kiocb *req, bool force_nonblock) +{ /* fsync always requires a blocking context */ if (force_nonblock) { - io_put_req(req); req->work.func = io_fsync_finish; return -EAGAIN; } - - work = old_work = &req->work; - io_fsync_finish(&work); - if (work && work != old_work) - *nxt = container_of(work, struct io_kiocb, work); + __io_fsync(req); return 0; } -static void io_fallocate_finish(struct io_wq_work **workptr) +static void __io_fallocate(struct io_kiocb *req) { - struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); - struct io_kiocb *nxt = NULL; int ret; + current->signal->rlim[RLIMIT_FSIZE].rlim_cur = req->fsize; ret = vfs_fallocate(req->file, req->sync.mode, req->sync.off, req->sync.len); + current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, &nxt); - if (nxt) - io_wq_assign_next(workptr, nxt); + io_put_req(req); +} + +static void io_fallocate_finish(struct io_wq_work **workptr) +{ + struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); + + if (io_req_cancelled(req)) + return; + __io_fallocate(req); + io_steal_work(req, workptr); } static int io_fallocate_prep(struct io_kiocb *req, @@ -2536,26 +2897,19 @@ static int io_fallocate_prep(struct io_kiocb *req, req->sync.off = READ_ONCE(sqe->off); req->sync.len = READ_ONCE(sqe->addr); req->sync.mode = READ_ONCE(sqe->len); + req->fsize = rlimit(RLIMIT_FSIZE); return 0; } -static int io_fallocate(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_fallocate(struct io_kiocb *req, bool force_nonblock) { - struct io_wq_work *work, *old_work; - /* fallocate always requiring blocking context */ if (force_nonblock) { - io_put_req(req); req->work.func = io_fallocate_finish; return -EAGAIN; } - work = old_work = &req->work; - io_fallocate_finish(&work); - if (work && work != old_work) - *nxt = container_of(work, struct io_kiocb, work); - + __io_fallocate(req); return 0; } @@ -2583,6 +2937,7 @@ static int io_openat_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return ret; } + req->open.nofile = rlimit(RLIMIT_NOFILE); req->flags |= REQ_F_NEED_CLEANUP; return 0; } @@ -2624,12 +2979,12 @@ static int io_openat2_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return ret; } + req->open.nofile = rlimit(RLIMIT_NOFILE); req->flags |= REQ_F_NEED_CLEANUP; return 0; } -static int io_openat2(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_openat2(struct io_kiocb *req, bool force_nonblock) { struct open_flags op; struct file *file; @@ -2642,7 +2997,7 @@ static int io_openat2(struct io_kiocb *req, struct io_kiocb **nxt, if (ret) goto err; - ret = get_unused_fd_flags(req->open.how.flags); + ret = __get_unused_fd_flags(req->open.how.flags, req->open.nofile); if (ret < 0) goto err; @@ -2660,15 +3015,171 @@ err: if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; } -static int io_openat(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_openat(struct io_kiocb *req, bool force_nonblock) { req->open.how = build_open_how(req->open.how.flags, req->open.how.mode); - return io_openat2(req, nxt, force_nonblock); + return io_openat2(req, force_nonblock); +} + +static int io_remove_buffers_prep(struct io_kiocb *req, + const struct io_uring_sqe *sqe) +{ + struct io_provide_buf *p = &req->pbuf; + u64 tmp; + + if (sqe->ioprio || sqe->rw_flags || sqe->addr || sqe->len || sqe->off) + return -EINVAL; + + tmp = READ_ONCE(sqe->fd); + if (!tmp || tmp > USHRT_MAX) + return -EINVAL; + + memset(p, 0, sizeof(*p)); + p->nbufs = tmp; + p->bgid = READ_ONCE(sqe->buf_group); + return 0; +} + +static int __io_remove_buffers(struct io_ring_ctx *ctx, struct io_buffer *buf, + int bgid, unsigned nbufs) +{ + unsigned i = 0; + + /* shouldn't happen */ + if (!nbufs) + return 0; + + /* the head kbuf is the list itself */ + while (!list_empty(&buf->list)) { + struct io_buffer *nxt; + + nxt = list_first_entry(&buf->list, struct io_buffer, list); + list_del(&nxt->list); + kfree(nxt); + if (++i == nbufs) + return i; + } + i++; + kfree(buf); + idr_remove(&ctx->io_buffer_idr, bgid); + + return i; +} + +static int io_remove_buffers(struct io_kiocb *req, bool force_nonblock) +{ + struct io_provide_buf *p = &req->pbuf; + struct io_ring_ctx *ctx = req->ctx; + struct io_buffer *head; + int ret = 0; + + io_ring_submit_lock(ctx, !force_nonblock); + + lockdep_assert_held(&ctx->uring_lock); + + ret = -ENOENT; + head = idr_find(&ctx->io_buffer_idr, p->bgid); + if (head) + ret = __io_remove_buffers(ctx, head, p->bgid, p->nbufs); + + io_ring_submit_lock(ctx, !force_nonblock); + if (ret < 0) + req_set_fail_links(req); + io_cqring_add_event(req, ret); + io_put_req(req); + return 0; +} + +static int io_provide_buffers_prep(struct io_kiocb *req, + const struct io_uring_sqe *sqe) +{ + struct io_provide_buf *p = &req->pbuf; + u64 tmp; + + if (sqe->ioprio || sqe->rw_flags) + return -EINVAL; + + tmp = READ_ONCE(sqe->fd); + if (!tmp || tmp > USHRT_MAX) + return -E2BIG; + p->nbufs = tmp; + p->addr = READ_ONCE(sqe->addr); + p->len = READ_ONCE(sqe->len); + + if (!access_ok(u64_to_user_ptr(p->addr), p->len)) + return -EFAULT; + + p->bgid = READ_ONCE(sqe->buf_group); + tmp = READ_ONCE(sqe->off); + if (tmp > USHRT_MAX) + return -E2BIG; + p->bid = tmp; + return 0; +} + +static int io_add_buffers(struct io_provide_buf *pbuf, struct io_buffer **head) +{ + struct io_buffer *buf; + u64 addr = pbuf->addr; + int i, bid = pbuf->bid; + + for (i = 0; i < pbuf->nbufs; i++) { + buf = kmalloc(sizeof(*buf), GFP_KERNEL); + if (!buf) + break; + + buf->addr = addr; + buf->len = pbuf->len; + buf->bid = bid; + addr += pbuf->len; + bid++; + if (!*head) { + INIT_LIST_HEAD(&buf->list); + *head = buf; + } else { + list_add_tail(&buf->list, &(*head)->list); + } + } + + return i ? i : -ENOMEM; +} + +static int io_provide_buffers(struct io_kiocb *req, bool force_nonblock) +{ + struct io_provide_buf *p = &req->pbuf; + struct io_ring_ctx *ctx = req->ctx; + struct io_buffer *head, *list; + int ret = 0; + + io_ring_submit_lock(ctx, !force_nonblock); + + lockdep_assert_held(&ctx->uring_lock); + + list = head = idr_find(&ctx->io_buffer_idr, p->bgid); + + ret = io_add_buffers(p, &head); + if (ret < 0) + goto out; + + if (!list) { + ret = idr_alloc(&ctx->io_buffer_idr, head, p->bgid, p->bgid + 1, + GFP_KERNEL); + if (ret < 0) { + __io_remove_buffers(ctx, head, p->bgid, -1U); + goto out; + } + } +out: + io_ring_submit_unlock(ctx, !force_nonblock); + if (ret < 0) + req_set_fail_links(req); + io_cqring_add_event(req, ret); + io_put_req(req); + return 0; } static int io_epoll_ctl_prep(struct io_kiocb *req, @@ -2696,8 +3207,7 @@ static int io_epoll_ctl_prep(struct io_kiocb *req, #endif } -static int io_epoll_ctl(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_epoll_ctl(struct io_kiocb *req, bool force_nonblock) { #if defined(CONFIG_EPOLL) struct io_epoll *ie = &req->epoll; @@ -2710,7 +3220,7 @@ static int io_epoll_ctl(struct io_kiocb *req, struct io_kiocb **nxt, if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; #else return -EOPNOTSUPP; @@ -2732,8 +3242,7 @@ static int io_madvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) #endif } -static int io_madvise(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_madvise(struct io_kiocb *req, bool force_nonblock) { #if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU) struct io_madvise *ma = &req->madvise; @@ -2746,7 +3255,7 @@ static int io_madvise(struct io_kiocb *req, struct io_kiocb **nxt, if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; #else return -EOPNOTSUPP; @@ -2764,8 +3273,7 @@ static int io_fadvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static int io_fadvise(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_fadvise(struct io_kiocb *req, bool force_nonblock) { struct io_fadvise *fa = &req->fadvise; int ret; @@ -2785,7 +3293,7 @@ static int io_fadvise(struct io_kiocb *req, struct io_kiocb **nxt, if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; } @@ -2822,8 +3330,7 @@ static int io_statx_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static int io_statx(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_statx(struct io_kiocb *req, bool force_nonblock) { struct io_open *ctx = &req->open; unsigned lookup_flags; @@ -2860,7 +3367,7 @@ err: if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; } @@ -2887,7 +3394,7 @@ static int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) } /* only called when __close_fd_get_file() is done */ -static void __io_close_finish(struct io_kiocb *req, struct io_kiocb **nxt) +static void __io_close_finish(struct io_kiocb *req) { int ret; @@ -2896,21 +3403,19 @@ static void __io_close_finish(struct io_kiocb *req, struct io_kiocb **nxt) req_set_fail_links(req); io_cqring_add_event(req, ret); fput(req->close.put_file); - io_put_req_find_next(req, nxt); + io_put_req(req); } static void io_close_finish(struct io_wq_work **workptr) { struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); - struct io_kiocb *nxt = NULL; - __io_close_finish(req, &nxt); - if (nxt) - io_wq_assign_next(workptr, nxt); + /* not cancellable, don't do io_req_cancelled() */ + __io_close_finish(req); + io_steal_work(req, workptr); } -static int io_close(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_close(struct io_kiocb *req, bool force_nonblock) { int ret; @@ -2920,23 +3425,25 @@ static int io_close(struct io_kiocb *req, struct io_kiocb **nxt, return ret; /* if the file has a flush method, be safe and punt to async */ - if (req->close.put_file->f_op->flush && !io_wq_current_is_worker()) - goto eagain; + if (req->close.put_file->f_op->flush && force_nonblock) { + /* submission ref will be dropped, take it for async */ + refcount_inc(&req->refs); + + req->work.func = io_close_finish; + /* + * Do manual async queue here to avoid grabbing files - we don't + * need the files, and it'll cause io_close_finish() to close + * the file again and cause a double CQE entry for this request + */ + io_queue_async_work(req); + return 0; + } /* * No ->flush(), safely close from here and just punt the * fput() to async context. */ - __io_close_finish(req, nxt); - return 0; -eagain: - req->work.func = io_close_finish; - /* - * Do manual async queue here to avoid grabbing files - we don't - * need the files, and it'll cause io_close_finish() to close - * the file again and cause a double CQE entry for this request - */ - io_queue_async_work(req); + __io_close_finish(req); return 0; } @@ -2958,47 +3465,62 @@ static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static void io_sync_file_range_finish(struct io_wq_work **workptr) +static void __io_sync_file_range(struct io_kiocb *req) { - struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); - struct io_kiocb *nxt = NULL; int ret; - if (io_req_cancelled(req)) - return; - ret = sync_file_range(req->file, req->sync.off, req->sync.len, req->sync.flags); if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, &nxt); + io_put_req(req); +} + + +static void io_sync_file_range_finish(struct io_wq_work **workptr) +{ + struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); + struct io_kiocb *nxt = NULL; + + if (io_req_cancelled(req)) + return; + __io_sync_file_range(req); + io_put_req(req); /* put submission ref */ if (nxt) io_wq_assign_next(workptr, nxt); } -static int io_sync_file_range(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_sync_file_range(struct io_kiocb *req, bool force_nonblock) { - struct io_wq_work *work, *old_work; - /* sync_file_range always requires a blocking context */ if (force_nonblock) { - io_put_req(req); req->work.func = io_sync_file_range_finish; return -EAGAIN; } - work = old_work = &req->work; - io_sync_file_range_finish(&work); - if (work && work != old_work) - *nxt = container_of(work, struct io_kiocb, work); + __io_sync_file_range(req); return 0; } +#if defined(CONFIG_NET) +static int io_setup_async_msg(struct io_kiocb *req, + struct io_async_msghdr *kmsg) +{ + if (req->io) + return -EAGAIN; + if (io_alloc_async_ctx(req)) { + if (kmsg->iov != kmsg->fast_iov) + kfree(kmsg->iov); + return -ENOMEM; + } + req->flags |= REQ_F_NEED_CLEANUP; + memcpy(&req->io->msg, kmsg, sizeof(*kmsg)); + return -EAGAIN; +} + static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { -#if defined(CONFIG_NET) struct io_sr_msg *sr = &req->sr_msg; struct io_async_ctx *io = req->io; int ret; @@ -3007,6 +3529,11 @@ static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr)); sr->len = READ_ONCE(sqe->len); +#ifdef CONFIG_COMPAT + if (req->ctx->compat) + sr->msg_flags |= MSG_CMSG_COMPAT; +#endif + if (!io || req->opcode == IORING_OP_SEND) return 0; /* iovec is already imported */ @@ -3019,15 +3546,10 @@ static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) if (!ret) req->flags |= REQ_F_NEED_CLEANUP; return ret; -#else - return -EOPNOTSUPP; -#endif } -static int io_sendmsg(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_sendmsg(struct io_kiocb *req, bool force_nonblock) { -#if defined(CONFIG_NET) struct io_async_msghdr *kmsg = NULL; struct socket *sock; int ret; @@ -3067,18 +3589,8 @@ static int io_sendmsg(struct io_kiocb *req, struct io_kiocb **nxt, flags |= MSG_DONTWAIT; ret = __sys_sendmsg_sock(sock, &kmsg->msg, flags); - if (force_nonblock && ret == -EAGAIN) { - if (req->io) - return -EAGAIN; - if (io_alloc_async_ctx(req)) { - if (kmsg && kmsg->iov != kmsg->fast_iov) - kfree(kmsg->iov); - return -ENOMEM; - } - req->flags |= REQ_F_NEED_CLEANUP; - memcpy(&req->io->msg, &io.msg, sizeof(io.msg)); - return -EAGAIN; - } + if (force_nonblock && ret == -EAGAIN) + return io_setup_async_msg(req, kmsg); if (ret == -ERESTARTSYS) ret = -EINTR; } @@ -3089,17 +3601,12 @@ static int io_sendmsg(struct io_kiocb *req, struct io_kiocb **nxt, io_cqring_add_event(req, ret); if (ret < 0) req_set_fail_links(req); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; -#else - return -EOPNOTSUPP; -#endif } -static int io_send(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_send(struct io_kiocb *req, bool force_nonblock) { -#if defined(CONFIG_NET) struct socket *sock; int ret; @@ -3140,17 +3647,120 @@ static int io_send(struct io_kiocb *req, struct io_kiocb **nxt, io_cqring_add_event(req, ret); if (ret < 0) req_set_fail_links(req); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; -#else - return -EOPNOTSUPP; +} + +static int __io_recvmsg_copy_hdr(struct io_kiocb *req, struct io_async_ctx *io) +{ + struct io_sr_msg *sr = &req->sr_msg; + struct iovec __user *uiov; + size_t iov_len; + int ret; + + ret = __copy_msghdr_from_user(&io->msg.msg, sr->msg, &io->msg.uaddr, + &uiov, &iov_len); + if (ret) + return ret; + + if (req->flags & REQ_F_BUFFER_SELECT) { + if (iov_len > 1) + return -EINVAL; + if (copy_from_user(io->msg.iov, uiov, sizeof(*uiov))) + return -EFAULT; + sr->len = io->msg.iov[0].iov_len; + iov_iter_init(&io->msg.msg.msg_iter, READ, io->msg.iov, 1, + sr->len); + io->msg.iov = NULL; + } else { + ret = import_iovec(READ, uiov, iov_len, UIO_FASTIOV, + &io->msg.iov, &io->msg.msg.msg_iter); + if (ret > 0) + ret = 0; + } + + return ret; +} + +#ifdef CONFIG_COMPAT +static int __io_compat_recvmsg_copy_hdr(struct io_kiocb *req, + struct io_async_ctx *io) +{ + struct compat_msghdr __user *msg_compat; + struct io_sr_msg *sr = &req->sr_msg; + struct compat_iovec __user *uiov; + compat_uptr_t ptr; + compat_size_t len; + int ret; + + msg_compat = (struct compat_msghdr __user *) sr->msg; + ret = __get_compat_msghdr(&io->msg.msg, msg_compat, &io->msg.uaddr, + &ptr, &len); + if (ret) + return ret; + + uiov = compat_ptr(ptr); + if (req->flags & REQ_F_BUFFER_SELECT) { + compat_ssize_t clen; + + if (len > 1) + return -EINVAL; + if (!access_ok(uiov, sizeof(*uiov))) + return -EFAULT; + if (__get_user(clen, &uiov->iov_len)) + return -EFAULT; + if (clen < 0) + return -EINVAL; + sr->len = io->msg.iov[0].iov_len; + io->msg.iov = NULL; + } else { + ret = compat_import_iovec(READ, uiov, len, UIO_FASTIOV, + &io->msg.iov, + &io->msg.msg.msg_iter); + if (ret < 0) + return ret; + } + + return 0; +} +#endif + +static int io_recvmsg_copy_hdr(struct io_kiocb *req, struct io_async_ctx *io) +{ + io->msg.iov = io->msg.fast_iov; + +#ifdef CONFIG_COMPAT + if (req->ctx->compat) + return __io_compat_recvmsg_copy_hdr(req, io); #endif + + return __io_recvmsg_copy_hdr(req, io); +} + +static struct io_buffer *io_recv_buffer_select(struct io_kiocb *req, + int *cflags, bool needs_lock) +{ + struct io_sr_msg *sr = &req->sr_msg; + struct io_buffer *kbuf; + + if (!(req->flags & REQ_F_BUFFER_SELECT)) + return NULL; + + kbuf = io_buffer_select(req, &sr->len, sr->bgid, sr->kbuf, needs_lock); + if (IS_ERR(kbuf)) + return kbuf; + + sr->kbuf = kbuf; + req->flags |= REQ_F_BUFFER_SELECTED; + + *cflags = kbuf->bid << IORING_CQE_BUFFER_SHIFT; + *cflags |= IORING_CQE_F_BUFFER; + return kbuf; } static int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { -#if defined(CONFIG_NET) struct io_sr_msg *sr = &req->sr_msg; struct io_async_ctx *io = req->io; int ret; @@ -3158,6 +3768,12 @@ static int io_recvmsg_prep(struct io_kiocb *req, sr->msg_flags = READ_ONCE(sqe->msg_flags); sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr)); sr->len = READ_ONCE(sqe->len); + sr->bgid = READ_ONCE(sqe->buf_group); + +#ifdef CONFIG_COMPAT + if (req->ctx->compat) + sr->msg_flags |= MSG_CMSG_COMPAT; +#endif if (!io || req->opcode == IORING_OP_RECV) return 0; @@ -3165,30 +3781,24 @@ static int io_recvmsg_prep(struct io_kiocb *req, if (req->flags & REQ_F_NEED_CLEANUP) return 0; - io->msg.iov = io->msg.fast_iov; - ret = recvmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags, - &io->msg.uaddr, &io->msg.iov); + ret = io_recvmsg_copy_hdr(req, io); if (!ret) req->flags |= REQ_F_NEED_CLEANUP; return ret; -#else - return -EOPNOTSUPP; -#endif } -static int io_recvmsg(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_recvmsg(struct io_kiocb *req, bool force_nonblock) { -#if defined(CONFIG_NET) struct io_async_msghdr *kmsg = NULL; struct socket *sock; - int ret; + int ret, cflags = 0; if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; sock = sock_from_file(req->file, &ret); if (sock) { + struct io_buffer *kbuf; struct io_async_ctx io; unsigned flags; @@ -3200,19 +3810,23 @@ static int io_recvmsg(struct io_kiocb *req, struct io_kiocb **nxt, kmsg->iov = kmsg->fast_iov; kmsg->msg.msg_iter.iov = kmsg->iov; } else { - struct io_sr_msg *sr = &req->sr_msg; - kmsg = &io.msg; kmsg->msg.msg_name = &io.msg.addr; - io.msg.iov = io.msg.fast_iov; - ret = recvmsg_copy_msghdr(&io.msg.msg, sr->msg, - sr->msg_flags, &io.msg.uaddr, - &io.msg.iov); + ret = io_recvmsg_copy_hdr(req, &io); if (ret) return ret; } + kbuf = io_recv_buffer_select(req, &cflags, !force_nonblock); + if (IS_ERR(kbuf)) { + return PTR_ERR(kbuf); + } else if (kbuf) { + kmsg->fast_iov[0].iov_base = u64_to_user_ptr(kbuf->addr); + iov_iter_init(&kmsg->msg.msg_iter, READ, kmsg->iov, + 1, req->sr_msg.len); + } + flags = req->sr_msg.msg_flags; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; @@ -3221,18 +3835,8 @@ static int io_recvmsg(struct io_kiocb *req, struct io_kiocb **nxt, ret = __sys_recvmsg_sock(sock, &kmsg->msg, req->sr_msg.msg, kmsg->uaddr, flags); - if (force_nonblock && ret == -EAGAIN) { - if (req->io) - return -EAGAIN; - if (io_alloc_async_ctx(req)) { - if (kmsg && kmsg->iov != kmsg->fast_iov) - kfree(kmsg->iov); - return -ENOMEM; - } - memcpy(&req->io->msg, &io.msg, sizeof(io.msg)); - req->flags |= REQ_F_NEED_CLEANUP; - return -EAGAIN; - } + if (force_nonblock && ret == -EAGAIN) + return io_setup_async_msg(req, kmsg); if (ret == -ERESTARTSYS) ret = -EINTR; } @@ -3240,22 +3844,18 @@ static int io_recvmsg(struct io_kiocb *req, struct io_kiocb **nxt, if (kmsg && kmsg->iov != kmsg->fast_iov) kfree(kmsg->iov); req->flags &= ~REQ_F_NEED_CLEANUP; - io_cqring_add_event(req, ret); + __io_cqring_add_event(req, ret, cflags); if (ret < 0) req_set_fail_links(req); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; -#else - return -EOPNOTSUPP; -#endif } -static int io_recv(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_recv(struct io_kiocb *req, bool force_nonblock) { -#if defined(CONFIG_NET) + struct io_buffer *kbuf = NULL; struct socket *sock; - int ret; + int ret, cflags = 0; if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; @@ -3263,15 +3863,25 @@ static int io_recv(struct io_kiocb *req, struct io_kiocb **nxt, sock = sock_from_file(req->file, &ret); if (sock) { struct io_sr_msg *sr = &req->sr_msg; + void __user *buf = sr->buf; struct msghdr msg; struct iovec iov; unsigned flags; - ret = import_single_range(READ, sr->buf, sr->len, &iov, + kbuf = io_recv_buffer_select(req, &cflags, !force_nonblock); + if (IS_ERR(kbuf)) + return PTR_ERR(kbuf); + else if (kbuf) + buf = u64_to_user_ptr(kbuf->addr); + + ret = import_single_range(READ, buf, sr->len, &iov, &msg.msg_iter); - if (ret) + if (ret) { + kfree(kbuf); return ret; + } + req->flags |= REQ_F_NEED_CLEANUP; msg.msg_name = NULL; msg.msg_control = NULL; msg.msg_controllen = 0; @@ -3292,20 +3902,17 @@ static int io_recv(struct io_kiocb *req, struct io_kiocb **nxt, ret = -EINTR; } - io_cqring_add_event(req, ret); + kfree(kbuf); + req->flags &= ~REQ_F_NEED_CLEANUP; + __io_cqring_add_event(req, ret, cflags); if (ret < 0) req_set_fail_links(req); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; -#else - return -EOPNOTSUPP; -#endif } - static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { -#if defined(CONFIG_NET) struct io_accept *accept = &req->accept; if (unlikely(req->ctx->flags & (IORING_SETUP_IOPOLL|IORING_SETUP_SQPOLL))) @@ -3316,15 +3923,11 @@ static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) accept->addr = u64_to_user_ptr(READ_ONCE(sqe->addr)); accept->addr_len = u64_to_user_ptr(READ_ONCE(sqe->addr2)); accept->flags = READ_ONCE(sqe->accept_flags); + accept->nofile = rlimit(RLIMIT_NOFILE); return 0; -#else - return -EOPNOTSUPP; -#endif } -#if defined(CONFIG_NET) -static int __io_accept(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int __io_accept(struct io_kiocb *req, bool force_nonblock) { struct io_accept *accept = &req->accept; unsigned file_flags; @@ -3332,7 +3935,8 @@ static int __io_accept(struct io_kiocb *req, struct io_kiocb **nxt, file_flags = force_nonblock ? O_NONBLOCK : 0; ret = __sys_accept4_file(req->file, file_flags, accept->addr, - accept->addr_len, accept->flags); + accept->addr_len, accept->flags, + accept->nofile); if (ret == -EAGAIN && force_nonblock) return -EAGAIN; if (ret == -ERESTARTSYS) @@ -3340,44 +3944,34 @@ static int __io_accept(struct io_kiocb *req, struct io_kiocb **nxt, if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; } static void io_accept_finish(struct io_wq_work **workptr) { struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); - struct io_kiocb *nxt = NULL; if (io_req_cancelled(req)) return; - __io_accept(req, &nxt, false); - if (nxt) - io_wq_assign_next(workptr, nxt); + __io_accept(req, false); + io_steal_work(req, workptr); } -#endif -static int io_accept(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_accept(struct io_kiocb *req, bool force_nonblock) { -#if defined(CONFIG_NET) int ret; - ret = __io_accept(req, nxt, force_nonblock); + ret = __io_accept(req, force_nonblock); if (ret == -EAGAIN && force_nonblock) { req->work.func = io_accept_finish; - io_put_req(req); return -EAGAIN; } return 0; -#else - return -EOPNOTSUPP; -#endif } static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { -#if defined(CONFIG_NET) struct io_connect *conn = &req->connect; struct io_async_ctx *io = req->io; @@ -3394,15 +3988,10 @@ static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return move_addr_to_kernel(conn->addr, conn->addr_len, &io->connect.address); -#else - return -EOPNOTSUPP; -#endif } -static int io_connect(struct io_kiocb *req, struct io_kiocb **nxt, - bool force_nonblock) +static int io_connect(struct io_kiocb *req, bool force_nonblock) { -#if defined(CONFIG_NET) struct io_async_ctx __io, *io; unsigned file_flags; int ret; @@ -3438,25 +4027,301 @@ out: if (ret < 0) req_set_fail_links(req); io_cqring_add_event(req, ret); - io_put_req_find_next(req, nxt); + io_put_req(req); return 0; -#else +} +#else /* !CONFIG_NET */ +static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) +{ return -EOPNOTSUPP; -#endif } -static void io_poll_remove_one(struct io_kiocb *req) +static int io_sendmsg(struct io_kiocb *req, bool force_nonblock) { - struct io_poll_iocb *poll = &req->poll; + return -EOPNOTSUPP; +} + +static int io_send(struct io_kiocb *req, bool force_nonblock) +{ + return -EOPNOTSUPP; +} + +static int io_recvmsg_prep(struct io_kiocb *req, + const struct io_uring_sqe *sqe) +{ + return -EOPNOTSUPP; +} + +static int io_recvmsg(struct io_kiocb *req, bool force_nonblock) +{ + return -EOPNOTSUPP; +} + +static int io_recv(struct io_kiocb *req, bool force_nonblock) +{ + return -EOPNOTSUPP; +} + +static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) +{ + return -EOPNOTSUPP; +} + +static int io_accept(struct io_kiocb *req, bool force_nonblock) +{ + return -EOPNOTSUPP; +} + +static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) +{ + return -EOPNOTSUPP; +} + +static int io_connect(struct io_kiocb *req, bool force_nonblock) +{ + return -EOPNOTSUPP; +} +#endif /* CONFIG_NET */ + +struct io_poll_table { + struct poll_table_struct pt; + struct io_kiocb *req; + int error; +}; + +static void __io_queue_proc(struct io_poll_iocb *poll, struct io_poll_table *pt, + struct wait_queue_head *head) +{ + if (unlikely(poll->head)) { + pt->error = -EINVAL; + return; + } + + pt->error = 0; + poll->head = head; + add_wait_queue(head, &poll->wait); +} + +static void io_async_queue_proc(struct file *file, struct wait_queue_head *head, + struct poll_table_struct *p) +{ + struct io_poll_table *pt = container_of(p, struct io_poll_table, pt); + + __io_queue_proc(&pt->req->apoll->poll, pt, head); +} + +static int __io_async_wake(struct io_kiocb *req, struct io_poll_iocb *poll, + __poll_t mask, task_work_func_t func) +{ + struct task_struct *tsk; + + /* for instances that support it check for an event match first: */ + if (mask && !(mask & poll->events)) + return 0; + + trace_io_uring_task_add(req->ctx, req->opcode, req->user_data, mask); + + list_del_init(&poll->wait.entry); + + tsk = req->task; + req->result = mask; + init_task_work(&req->task_work, func); + /* + * If this fails, then the task is exiting. If that is the case, then + * the exit check will ultimately cancel these work items. Hence we + * don't need to check here and handle it specifically. + */ + task_work_add(tsk, &req->task_work, true); + wake_up_process(tsk); + return 1; +} + +static void io_async_task_func(struct callback_head *cb) +{ + struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work); + struct async_poll *apoll = req->apoll; + struct io_ring_ctx *ctx = req->ctx; + + trace_io_uring_task_run(req->ctx, req->opcode, req->user_data); + + WARN_ON_ONCE(!list_empty(&req->apoll->poll.wait.entry)); + + if (hash_hashed(&req->hash_node)) { + spin_lock_irq(&ctx->completion_lock); + hash_del(&req->hash_node); + spin_unlock_irq(&ctx->completion_lock); + } + + /* restore ->work in case we need to retry again */ + memcpy(&req->work, &apoll->work, sizeof(req->work)); + + __set_current_state(TASK_RUNNING); + mutex_lock(&ctx->uring_lock); + __io_queue_sqe(req, NULL); + mutex_unlock(&ctx->uring_lock); + + kfree(apoll); +} + +static int io_async_wake(struct wait_queue_entry *wait, unsigned mode, int sync, + void *key) +{ + struct io_kiocb *req = wait->private; + struct io_poll_iocb *poll = &req->apoll->poll; + + trace_io_uring_poll_wake(req->ctx, req->opcode, req->user_data, + key_to_poll(key)); + + return __io_async_wake(req, poll, key_to_poll(key), io_async_task_func); +} + +static void io_poll_req_insert(struct io_kiocb *req) +{ + struct io_ring_ctx *ctx = req->ctx; + struct hlist_head *list; + + list = &ctx->cancel_hash[hash_long(req->user_data, ctx->cancel_hash_bits)]; + hlist_add_head(&req->hash_node, list); +} + +static __poll_t __io_arm_poll_handler(struct io_kiocb *req, + struct io_poll_iocb *poll, + struct io_poll_table *ipt, __poll_t mask, + wait_queue_func_t wake_func) + __acquires(&ctx->completion_lock) +{ + struct io_ring_ctx *ctx = req->ctx; + bool cancel = false; + + poll->file = req->file; + poll->head = NULL; + poll->done = poll->canceled = false; + poll->events = mask; + + ipt->pt._key = mask; + ipt->req = req; + ipt->error = -EINVAL; + + INIT_LIST_HEAD(&poll->wait.entry); + init_waitqueue_func_entry(&poll->wait, wake_func); + poll->wait.private = req; + + mask = vfs_poll(req->file, &ipt->pt) & poll->events; + + spin_lock_irq(&ctx->completion_lock); + if (likely(poll->head)) { + spin_lock(&poll->head->lock); + if (unlikely(list_empty(&poll->wait.entry))) { + if (ipt->error) + cancel = true; + ipt->error = 0; + mask = 0; + } + if (mask || ipt->error) + list_del_init(&poll->wait.entry); + else if (cancel) + WRITE_ONCE(poll->canceled, true); + else if (!poll->done) /* actually waiting for an event */ + io_poll_req_insert(req); + spin_unlock(&poll->head->lock); + } + + return mask; +} + +static bool io_arm_poll_handler(struct io_kiocb *req) +{ + const struct io_op_def *def = &io_op_defs[req->opcode]; + struct io_ring_ctx *ctx = req->ctx; + struct async_poll *apoll; + struct io_poll_table ipt; + __poll_t mask, ret; + + if (!req->file || !file_can_poll(req->file)) + return false; + if (req->flags & (REQ_F_MUST_PUNT | REQ_F_POLLED)) + return false; + if (!def->pollin && !def->pollout) + return false; + + apoll = kmalloc(sizeof(*apoll), GFP_ATOMIC); + if (unlikely(!apoll)) + return false; + + req->flags |= REQ_F_POLLED; + memcpy(&apoll->work, &req->work, sizeof(req->work)); + + /* + * Don't need a reference here, as we're adding it to the task + * task_works list. If the task exits, the list is pruned. + */ + req->task = current; + req->apoll = apoll; + INIT_HLIST_NODE(&req->hash_node); + + mask = 0; + if (def->pollin) + mask |= POLLIN | POLLRDNORM; + if (def->pollout) + mask |= POLLOUT | POLLWRNORM; + mask |= POLLERR | POLLPRI; + + ipt.pt._qproc = io_async_queue_proc; + + ret = __io_arm_poll_handler(req, &apoll->poll, &ipt, mask, + io_async_wake); + if (ret) { + ipt.error = 0; + apoll->poll.done = true; + spin_unlock_irq(&ctx->completion_lock); + memcpy(&req->work, &apoll->work, sizeof(req->work)); + kfree(apoll); + return false; + } + spin_unlock_irq(&ctx->completion_lock); + trace_io_uring_poll_arm(ctx, req->opcode, req->user_data, mask, + apoll->poll.events); + return true; +} + +static bool __io_poll_remove_one(struct io_kiocb *req, + struct io_poll_iocb *poll) +{ + bool do_complete = false; spin_lock(&poll->head->lock); WRITE_ONCE(poll->canceled, true); if (!list_empty(&poll->wait.entry)) { list_del_init(&poll->wait.entry); - io_queue_async_work(req); + do_complete = true; } spin_unlock(&poll->head->lock); + return do_complete; +} + +static bool io_poll_remove_one(struct io_kiocb *req) +{ + bool do_complete; + + if (req->opcode == IORING_OP_POLL_ADD) { + do_complete = __io_poll_remove_one(req, &req->poll); + } else { + /* non-poll requests have submit ref still */ + do_complete = __io_poll_remove_one(req, &req->apoll->poll); + if (do_complete) + io_put_req(req); + } + hash_del(&req->hash_node); + + if (do_complete) { + io_cqring_fill_event(req, -ECANCELED); + io_commit_cqring(req->ctx); + req->flags |= REQ_F_COMP_LOCKED; + io_put_req(req); + } + + return do_complete; } static void io_poll_remove_all(struct io_ring_ctx *ctx) @@ -3474,6 +4339,8 @@ static void io_poll_remove_all(struct io_ring_ctx *ctx) io_poll_remove_one(req); } spin_unlock_irq(&ctx->completion_lock); + + io_cqring_ev_posted(ctx); } static int io_poll_cancel(struct io_ring_ctx *ctx, __u64 sqe_addr) @@ -3483,10 +4350,11 @@ static int io_poll_cancel(struct io_ring_ctx *ctx, __u64 sqe_addr) list = &ctx->cancel_hash[hash_long(sqe_addr, ctx->cancel_hash_bits)]; hlist_for_each_entry(req, list, hash_node) { - if (sqe_addr == req->user_data) { - io_poll_remove_one(req); + if (sqe_addr != req->user_data) + continue; + if (io_poll_remove_one(req)) return 0; - } + return -EALREADY; } return -ENOENT; @@ -3532,186 +4400,54 @@ static void io_poll_complete(struct io_kiocb *req, __poll_t mask, int error) struct io_ring_ctx *ctx = req->ctx; req->poll.done = true; - if (error) - io_cqring_fill_event(req, error); - else - io_cqring_fill_event(req, mangle_poll(mask)); + io_cqring_fill_event(req, error ? error : mangle_poll(mask)); io_commit_cqring(ctx); } -static void io_poll_complete_work(struct io_wq_work **workptr) +static void io_poll_task_handler(struct io_kiocb *req, struct io_kiocb **nxt) { - struct io_wq_work *work = *workptr; - struct io_kiocb *req = container_of(work, struct io_kiocb, work); - struct io_poll_iocb *poll = &req->poll; - struct poll_table_struct pt = { ._key = poll->events }; struct io_ring_ctx *ctx = req->ctx; - struct io_kiocb *nxt = NULL; - __poll_t mask = 0; - int ret = 0; - if (work->flags & IO_WQ_WORK_CANCEL) { - WRITE_ONCE(poll->canceled, true); - ret = -ECANCELED; - } else if (READ_ONCE(poll->canceled)) { - ret = -ECANCELED; - } - - if (ret != -ECANCELED) - mask = vfs_poll(poll->file, &pt) & poll->events; - - /* - * Note that ->ki_cancel callers also delete iocb from active_reqs after - * calling ->ki_cancel. We need the ctx_lock roundtrip here to - * synchronize with them. In the cancellation case the list_del_init - * itself is not actually needed, but harmless so we keep it in to - * avoid further branches in the fast path. - */ spin_lock_irq(&ctx->completion_lock); - if (!mask && ret != -ECANCELED) { - add_wait_queue(poll->head, &poll->wait); - spin_unlock_irq(&ctx->completion_lock); - return; - } hash_del(&req->hash_node); - io_poll_complete(req, mask, ret); + io_poll_complete(req, req->result, 0); + req->flags |= REQ_F_COMP_LOCKED; + io_put_req_find_next(req, nxt); spin_unlock_irq(&ctx->completion_lock); io_cqring_ev_posted(ctx); - - if (ret < 0) - req_set_fail_links(req); - io_put_req_find_next(req, &nxt); - if (nxt) - io_wq_assign_next(workptr, nxt); } -static void __io_poll_flush(struct io_ring_ctx *ctx, struct llist_node *nodes) +static void io_poll_task_func(struct callback_head *cb) { - struct io_kiocb *req, *tmp; - struct req_batch rb; + struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work); + struct io_kiocb *nxt = NULL; - rb.to_free = rb.need_iter = 0; - spin_lock_irq(&ctx->completion_lock); - llist_for_each_entry_safe(req, tmp, nodes, llist_node) { - hash_del(&req->hash_node); - io_poll_complete(req, req->result, 0); + io_poll_task_handler(req, &nxt); + if (nxt) { + struct io_ring_ctx *ctx = nxt->ctx; - if (refcount_dec_and_test(&req->refs) && - !io_req_multi_free(&rb, req)) { - req->flags |= REQ_F_COMP_LOCKED; - io_free_req(req); - } + mutex_lock(&ctx->uring_lock); + __io_queue_sqe(nxt, NULL); + mutex_unlock(&ctx->uring_lock); } - spin_unlock_irq(&ctx->completion_lock); - - io_cqring_ev_posted(ctx); - io_free_req_many(ctx, &rb); -} - -static void io_poll_flush(struct io_wq_work **workptr) -{ - struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); - struct llist_node *nodes; - - nodes = llist_del_all(&req->ctx->poll_llist); - if (nodes) - __io_poll_flush(req->ctx, nodes); -} - -static void io_poll_trigger_evfd(struct io_wq_work **workptr) -{ - struct io_kiocb *req = container_of(*workptr, struct io_kiocb, work); - - eventfd_signal(req->ctx->cq_ev_fd, 1); - io_put_req(req); } static int io_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync, void *key) { - struct io_poll_iocb *poll = wait->private; - struct io_kiocb *req = container_of(poll, struct io_kiocb, poll); - struct io_ring_ctx *ctx = req->ctx; - __poll_t mask = key_to_poll(key); - - /* for instances that support it check for an event match first: */ - if (mask && !(mask & poll->events)) - return 0; - - list_del_init(&poll->wait.entry); - - /* - * Run completion inline if we can. We're using trylock here because - * we are violating the completion_lock -> poll wq lock ordering. - * If we have a link timeout we're going to need the completion_lock - * for finalizing the request, mark us as having grabbed that already. - */ - if (mask) { - unsigned long flags; - - if (llist_empty(&ctx->poll_llist) && - spin_trylock_irqsave(&ctx->completion_lock, flags)) { - bool trigger_ev; - - hash_del(&req->hash_node); - io_poll_complete(req, mask, 0); - - trigger_ev = io_should_trigger_evfd(ctx); - if (trigger_ev && eventfd_signal_count()) { - trigger_ev = false; - req->work.func = io_poll_trigger_evfd; - } else { - req->flags |= REQ_F_COMP_LOCKED; - io_put_req(req); - req = NULL; - } - spin_unlock_irqrestore(&ctx->completion_lock, flags); - __io_cqring_ev_posted(ctx, trigger_ev); - } else { - req->result = mask; - req->llist_node.next = NULL; - /* if the list wasn't empty, we're done */ - if (!llist_add(&req->llist_node, &ctx->poll_llist)) - req = NULL; - else - req->work.func = io_poll_flush; - } - } - if (req) - io_queue_async_work(req); + struct io_kiocb *req = wait->private; + struct io_poll_iocb *poll = &req->poll; - return 1; + return __io_async_wake(req, poll, key_to_poll(key), io_poll_task_func); } -struct io_poll_table { - struct poll_table_struct pt; - struct io_kiocb *req; - int error; -}; - static void io_poll_queue_proc(struct file *file, struct wait_queue_head *head, struct poll_table_struct *p) { struct io_poll_table *pt = container_of(p, struct io_poll_table, pt); - if (unlikely(pt->req->poll.head)) { - pt->error = -EINVAL; - return; - } - - pt->error = 0; - pt->req->poll.head = head; - add_wait_queue(head, &pt->req->poll.wait); -} - -static void io_poll_req_insert(struct io_kiocb *req) -{ - struct io_ring_ctx *ctx = req->ctx; - struct hlist_head *list; - - list = &ctx->cancel_hash[hash_long(req->user_data, ctx->cancel_hash_bits)]; - hlist_add_head(&req->hash_node, list); + __io_queue_proc(&pt->req->poll, pt, head); } static int io_poll_add_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) @@ -3728,55 +4464,29 @@ static int io_poll_add_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe events = READ_ONCE(sqe->poll_events); poll->events = demangle_poll(events) | EPOLLERR | EPOLLHUP; + + /* + * Don't need a reference here, as we're adding it to the task + * task_works list. If the task exits, the list is pruned. + */ + req->task = current; return 0; } -static int io_poll_add(struct io_kiocb *req, struct io_kiocb **nxt) +static int io_poll_add(struct io_kiocb *req) { struct io_poll_iocb *poll = &req->poll; struct io_ring_ctx *ctx = req->ctx; struct io_poll_table ipt; - bool cancel = false; __poll_t mask; - INIT_IO_WORK(&req->work, io_poll_complete_work); INIT_HLIST_NODE(&req->hash_node); - - poll->head = NULL; - poll->done = false; - poll->canceled = false; - - ipt.pt._qproc = io_poll_queue_proc; - ipt.pt._key = poll->events; - ipt.req = req; - ipt.error = -EINVAL; /* same as no support for IOCB_CMD_POLL */ - - /* initialized the list so that we can do list_empty checks */ - INIT_LIST_HEAD(&poll->wait.entry); - init_waitqueue_func_entry(&poll->wait, io_poll_wake); - poll->wait.private = poll; - INIT_LIST_HEAD(&req->list); + ipt.pt._qproc = io_poll_queue_proc; - mask = vfs_poll(poll->file, &ipt.pt) & poll->events; + mask = __io_arm_poll_handler(req, &req->poll, &ipt, poll->events, + io_poll_wake); - spin_lock_irq(&ctx->completion_lock); - if (likely(poll->head)) { - spin_lock(&poll->head->lock); - if (unlikely(list_empty(&poll->wait.entry))) { - if (ipt.error) - cancel = true; - ipt.error = 0; - mask = 0; - } - if (mask || ipt.error) - list_del_init(&poll->wait.entry); - else if (cancel) - WRITE_ONCE(poll->canceled, true); - else if (!poll->done) /* actually waiting for an event */ - io_poll_req_insert(req); - spin_unlock(&poll->head->lock); - } if (mask) { /* no async, we'd stolen it */ ipt.error = 0; io_poll_complete(req, mask, 0); @@ -3785,7 +4495,7 @@ static int io_poll_add(struct io_kiocb *req, struct io_kiocb **nxt) if (mask) { io_cqring_ev_posted(ctx); - io_put_req_find_next(req, nxt); + io_put_req(req); } return ipt.error; } @@ -4034,7 +4744,7 @@ static int io_async_cancel_one(struct io_ring_ctx *ctx, void *sqe_addr) static void io_async_find_and_cancel(struct io_ring_ctx *ctx, struct io_kiocb *req, __u64 sqe_addr, - struct io_kiocb **nxt, int success_ret) + int success_ret) { unsigned long flags; int ret; @@ -4060,7 +4770,7 @@ done: if (ret < 0) req_set_fail_links(req); - io_put_req_find_next(req, nxt); + io_put_req(req); } static int io_async_cancel_prep(struct io_kiocb *req, @@ -4076,11 +4786,11 @@ static int io_async_cancel_prep(struct io_kiocb *req, return 0; } -static int io_async_cancel(struct io_kiocb *req, struct io_kiocb **nxt) +static int io_async_cancel(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; - io_async_find_and_cancel(ctx, req, req->cancel.addr, nxt, 0); + io_async_find_and_cancel(ctx, req, req->cancel.addr, 0); return 0; } @@ -4126,6 +4836,9 @@ static int io_req_defer_prep(struct io_kiocb *req, { ssize_t ret = 0; + if (!sqe) + return 0; + if (io_op_defs[req->opcode].file_table) { ret = io_grab_files(req); if (unlikely(ret)) @@ -4212,6 +4925,15 @@ static int io_req_defer_prep(struct io_kiocb *req, case IORING_OP_EPOLL_CTL: ret = io_epoll_ctl_prep(req, sqe); break; + case IORING_OP_SPLICE: + ret = io_splice_prep(req, sqe); + break; + case IORING_OP_PROVIDE_BUFFERS: + ret = io_provide_buffers_prep(req, sqe); + break; + case IORING_OP_REMOVE_BUFFERS: + ret = io_remove_buffers_prep(req, sqe); + break; default: printk_once(KERN_WARNING "io_uring: unhandled opcode %d\n", req->opcode); @@ -4258,29 +4980,43 @@ static void io_cleanup_req(struct io_kiocb *req) case IORING_OP_READV: case IORING_OP_READ_FIXED: case IORING_OP_READ: + if (req->flags & REQ_F_BUFFER_SELECTED) + kfree((void *)(unsigned long)req->rw.addr); + /* fallthrough */ case IORING_OP_WRITEV: case IORING_OP_WRITE_FIXED: case IORING_OP_WRITE: if (io->rw.iov != io->rw.fast_iov) kfree(io->rw.iov); break; - case IORING_OP_SENDMSG: case IORING_OP_RECVMSG: + if (req->flags & REQ_F_BUFFER_SELECTED) + kfree(req->sr_msg.kbuf); + /* fallthrough */ + case IORING_OP_SENDMSG: if (io->msg.iov != io->msg.fast_iov) kfree(io->msg.iov); break; + case IORING_OP_RECV: + if (req->flags & REQ_F_BUFFER_SELECTED) + kfree(req->sr_msg.kbuf); + break; case IORING_OP_OPENAT: case IORING_OP_OPENAT2: case IORING_OP_STATX: putname(req->open.filename); break; + case IORING_OP_SPLICE: + io_put_file(req, req->splice.file_in, + (req->splice.flags & SPLICE_F_FD_IN_FIXED)); + break; } req->flags &= ~REQ_F_NEED_CLEANUP; } static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, - struct io_kiocb **nxt, bool force_nonblock) + bool force_nonblock) { struct io_ring_ctx *ctx = req->ctx; int ret; @@ -4297,7 +5033,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret < 0) break; } - ret = io_read(req, nxt, force_nonblock); + ret = io_read(req, force_nonblock); break; case IORING_OP_WRITEV: case IORING_OP_WRITE_FIXED: @@ -4307,7 +5043,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret < 0) break; } - ret = io_write(req, nxt, force_nonblock); + ret = io_write(req, force_nonblock); break; case IORING_OP_FSYNC: if (sqe) { @@ -4315,7 +5051,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret < 0) break; } - ret = io_fsync(req, nxt, force_nonblock); + ret = io_fsync(req, force_nonblock); break; case IORING_OP_POLL_ADD: if (sqe) { @@ -4323,7 +5059,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_poll_add(req, nxt); + ret = io_poll_add(req); break; case IORING_OP_POLL_REMOVE: if (sqe) { @@ -4339,7 +5075,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret < 0) break; } - ret = io_sync_file_range(req, nxt, force_nonblock); + ret = io_sync_file_range(req, force_nonblock); break; case IORING_OP_SENDMSG: case IORING_OP_SEND: @@ -4349,9 +5085,9 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, break; } if (req->opcode == IORING_OP_SENDMSG) - ret = io_sendmsg(req, nxt, force_nonblock); + ret = io_sendmsg(req, force_nonblock); else - ret = io_send(req, nxt, force_nonblock); + ret = io_send(req, force_nonblock); break; case IORING_OP_RECVMSG: case IORING_OP_RECV: @@ -4361,9 +5097,9 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, break; } if (req->opcode == IORING_OP_RECVMSG) - ret = io_recvmsg(req, nxt, force_nonblock); + ret = io_recvmsg(req, force_nonblock); else - ret = io_recv(req, nxt, force_nonblock); + ret = io_recv(req, force_nonblock); break; case IORING_OP_TIMEOUT: if (sqe) { @@ -4387,7 +5123,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_accept(req, nxt, force_nonblock); + ret = io_accept(req, force_nonblock); break; case IORING_OP_CONNECT: if (sqe) { @@ -4395,7 +5131,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_connect(req, nxt, force_nonblock); + ret = io_connect(req, force_nonblock); break; case IORING_OP_ASYNC_CANCEL: if (sqe) { @@ -4403,7 +5139,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_async_cancel(req, nxt); + ret = io_async_cancel(req); break; case IORING_OP_FALLOCATE: if (sqe) { @@ -4411,7 +5147,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_fallocate(req, nxt, force_nonblock); + ret = io_fallocate(req, force_nonblock); break; case IORING_OP_OPENAT: if (sqe) { @@ -4419,7 +5155,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_openat(req, nxt, force_nonblock); + ret = io_openat(req, force_nonblock); break; case IORING_OP_CLOSE: if (sqe) { @@ -4427,7 +5163,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_close(req, nxt, force_nonblock); + ret = io_close(req, force_nonblock); break; case IORING_OP_FILES_UPDATE: if (sqe) { @@ -4443,7 +5179,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_statx(req, nxt, force_nonblock); + ret = io_statx(req, force_nonblock); break; case IORING_OP_FADVISE: if (sqe) { @@ -4451,7 +5187,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_fadvise(req, nxt, force_nonblock); + ret = io_fadvise(req, force_nonblock); break; case IORING_OP_MADVISE: if (sqe) { @@ -4459,7 +5195,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_madvise(req, nxt, force_nonblock); + ret = io_madvise(req, force_nonblock); break; case IORING_OP_OPENAT2: if (sqe) { @@ -4467,7 +5203,7 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_openat2(req, nxt, force_nonblock); + ret = io_openat2(req, force_nonblock); break; case IORING_OP_EPOLL_CTL: if (sqe) { @@ -4475,7 +5211,31 @@ static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (ret) break; } - ret = io_epoll_ctl(req, nxt, force_nonblock); + ret = io_epoll_ctl(req, force_nonblock); + break; + case IORING_OP_SPLICE: + if (sqe) { + ret = io_splice_prep(req, sqe); + if (ret < 0) + break; + } + ret = io_splice(req, force_nonblock); + break; + case IORING_OP_PROVIDE_BUFFERS: + if (sqe) { + ret = io_provide_buffers_prep(req, sqe); + if (ret) + break; + } + ret = io_provide_buffers(req, force_nonblock); + break; + case IORING_OP_REMOVE_BUFFERS: + if (sqe) { + ret = io_remove_buffers_prep(req, sqe); + if (ret) + break; + } + ret = io_remove_buffers(req, force_nonblock); break; default: ret = -EINVAL; @@ -4508,7 +5268,6 @@ static void io_wq_submit_work(struct io_wq_work **workptr) { struct io_wq_work *work = *workptr; struct io_kiocb *req = container_of(work, struct io_kiocb, work); - struct io_kiocb *nxt = NULL; int ret = 0; /* if NO_CANCEL is set, we must still run the work */ @@ -4518,9 +5277,8 @@ static void io_wq_submit_work(struct io_wq_work **workptr) } if (!ret) { - req->in_async = true; do { - ret = io_issue_sqe(req, NULL, &nxt, false); + ret = io_issue_sqe(req, NULL, false); /* * We can get EAGAIN for polled IO even though we're * forcing a sync submission from here, since we can't @@ -4532,18 +5290,13 @@ static void io_wq_submit_work(struct io_wq_work **workptr) } while (1); } - /* drop submission reference */ - io_put_req(req); - if (ret) { req_set_fail_links(req); io_cqring_add_event(req, ret); io_put_req(req); } - /* if a dependent link is ready, pass it back */ - if (!ret && nxt) - io_wq_assign_next(workptr, nxt); + io_steal_work(req, workptr); } static int io_req_needs_file(struct io_kiocb *req, int fd) @@ -4564,41 +5317,52 @@ static inline struct file *io_file_from_index(struct io_ring_ctx *ctx, return table->files[index & IORING_FILE_TABLE_MASK];; } -static int io_req_set_file(struct io_submit_state *state, struct io_kiocb *req, - const struct io_uring_sqe *sqe) +static int io_file_get(struct io_submit_state *state, struct io_kiocb *req, + int fd, struct file **out_file, bool fixed) { struct io_ring_ctx *ctx = req->ctx; - unsigned flags; - int fd; - - flags = READ_ONCE(sqe->flags); - fd = READ_ONCE(sqe->fd); - - if (!io_req_needs_file(req, fd)) - return 0; + struct file *file; - if (flags & IOSQE_FIXED_FILE) { + if (fixed) { if (unlikely(!ctx->file_data || (unsigned) fd >= ctx->nr_user_files)) return -EBADF; fd = array_index_nospec(fd, ctx->nr_user_files); - req->file = io_file_from_index(ctx, fd); - if (!req->file) + file = io_file_from_index(ctx, fd); + if (!file) return -EBADF; - req->flags |= REQ_F_FIXED_FILE; percpu_ref_get(&ctx->file_data->refs); } else { - if (req->needs_fixed_file) - return -EBADF; trace_io_uring_file_get(ctx, fd); - req->file = io_file_get(state, fd); - if (unlikely(!req->file)) + file = __io_file_get(state, fd); + if (unlikely(!file)) return -EBADF; } + *out_file = file; return 0; } +static int io_req_set_file(struct io_submit_state *state, struct io_kiocb *req, + const struct io_uring_sqe *sqe) +{ + unsigned flags; + int fd; + bool fixed; + + flags = READ_ONCE(sqe->flags); + fd = READ_ONCE(sqe->fd); + + if (!io_req_needs_file(req, fd)) + return 0; + + fixed = (flags & IOSQE_FIXED_FILE); + if (unlikely(!fixed && req->needs_fixed_file)) + return -EBADF; + + return io_file_get(state, req, fd, &req->file, fixed); +} + static int io_grab_files(struct io_kiocb *req) { int ret = -EBADF; @@ -4658,8 +5422,7 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer) if (prev) { req_set_fail_links(prev); - io_async_find_and_cancel(ctx, req, prev->user_data, NULL, - -ETIME); + io_async_find_and_cancel(ctx, req, prev->user_data, -ETIME); io_put_req(prev); } else { io_cqring_add_event(req, -ETIME); @@ -4696,6 +5459,9 @@ static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req) if (!(req->flags & REQ_F_LINK)) return NULL; + /* for polled retry, if flag is set, we already went through here */ + if (req->flags & REQ_F_POLLED) + return NULL; nxt = list_first_entry_or_null(&req->link_list, struct io_kiocb, link_list); @@ -4709,13 +5475,23 @@ static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req) static void __io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_kiocb *linked_timeout; - struct io_kiocb *nxt = NULL; + struct io_kiocb *nxt; + const struct cred *old_creds = NULL; int ret; again: linked_timeout = io_prep_linked_timeout(req); - ret = io_issue_sqe(req, sqe, &nxt, true); + if (req->work.creds && req->work.creds != current_cred()) { + if (old_creds) + revert_creds(old_creds); + if (old_creds == req->work.creds) + old_creds = NULL; /* restored original creds */ + else + old_creds = override_creds(req->work.creds); + } + + ret = io_issue_sqe(req, sqe, true); /* * We async punt it if the file wasn't marked NOWAIT, or if the file @@ -4723,6 +5499,11 @@ again: */ if (ret == -EAGAIN && (!(req->flags & REQ_F_NOWAIT) || (req->flags & REQ_F_MUST_PUNT))) { + if (io_arm_poll_handler(req)) { + if (linked_timeout) + io_queue_linked_timeout(linked_timeout); + goto exit; + } punt: if (io_op_defs[req->opcode].file_table) { ret = io_grab_files(req); @@ -4735,12 +5516,13 @@ punt: * submit reference when the iocb is actually submitted. */ io_queue_async_work(req); - goto done_req; + goto exit; } err: + nxt = NULL; /* drop submission reference */ - io_put_req(req); + io_put_req_find_next(req, &nxt); if (linked_timeout) { if (!ret) @@ -4755,15 +5537,16 @@ err: req_set_fail_links(req); io_put_req(req); } -done_req: if (nxt) { req = nxt; - nxt = NULL; if (req->flags & REQ_F_FORCE_ASYNC) goto punt; goto again; } +exit: + if (old_creds) + revert_creds(old_creds); } static void io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe) @@ -4803,12 +5586,12 @@ static inline void io_queue_link_head(struct io_kiocb *req) } #define SQE_VALID_FLAGS (IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK| \ - IOSQE_IO_HARDLINK | IOSQE_ASYNC) + IOSQE_IO_HARDLINK | IOSQE_ASYNC | \ + IOSQE_BUFFER_SELECT) static bool io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, struct io_submit_state *state, struct io_kiocb **link) { - const struct cred *old_creds = NULL; struct io_ring_ctx *ctx = req->ctx; unsigned int sqe_flags; int ret, id; @@ -4821,29 +5604,32 @@ static bool io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, goto err_req; } + if ((sqe_flags & IOSQE_BUFFER_SELECT) && + !io_op_defs[req->opcode].buffer_select) { + ret = -EOPNOTSUPP; + goto err_req; + } + id = READ_ONCE(sqe->personality); if (id) { - const struct cred *personality_creds; - - personality_creds = idr_find(&ctx->personality_idr, id); - if (unlikely(!personality_creds)) { + req->work.creds = idr_find(&ctx->personality_idr, id); + if (unlikely(!req->work.creds)) { ret = -EINVAL; goto err_req; } - old_creds = override_creds(personality_creds); + get_cred(req->work.creds); } /* same numerical values with corresponding REQ_F_*, safe to copy */ - req->flags |= sqe_flags & (IOSQE_IO_DRAIN|IOSQE_IO_HARDLINK| - IOSQE_ASYNC); + req->flags |= sqe_flags & (IOSQE_IO_DRAIN | IOSQE_IO_HARDLINK | + IOSQE_ASYNC | IOSQE_FIXED_FILE | + IOSQE_BUFFER_SELECT); ret = io_req_set_file(state, req, sqe); if (unlikely(ret)) { err_req: io_cqring_add_event(req, ret); io_double_put_req(req); - if (old_creds) - revert_creds(old_creds); return false; } @@ -4895,6 +5681,11 @@ err_req: if (sqe_flags & (IOSQE_IO_LINK|IOSQE_IO_HARDLINK)) { req->flags |= REQ_F_LINK; INIT_LIST_HEAD(&req->link_list); + + if (io_alloc_async_ctx(req)) { + ret = -EAGAIN; + goto err_req; + } ret = io_req_defer_prep(req, sqe); if (ret) req->flags |= REQ_F_FAIL_LINK; @@ -4904,8 +5695,6 @@ err_req: } } - if (old_creds) - revert_creds(old_creds); return true; } @@ -5055,7 +5844,6 @@ fail_req: *mm = ctx->sqo_mm; } - req->in_async = async; req->needs_fixed_file = async; trace_io_uring_submit_sqe(ctx, req->opcode, req->user_data, true, async); @@ -5086,9 +5874,8 @@ static int io_sq_thread(void *data) const struct cred *old_cred; mm_segment_t old_fs; DEFINE_WAIT(wait); - unsigned inflight; unsigned long timeout; - int ret; + int ret = 0; complete(&ctx->completions[1]); @@ -5096,39 +5883,19 @@ static int io_sq_thread(void *data) set_fs(USER_DS); old_cred = override_creds(ctx->creds); - ret = timeout = inflight = 0; + timeout = jiffies + ctx->sq_thread_idle; while (!kthread_should_park()) { unsigned int to_submit; - if (inflight) { + if (!list_empty(&ctx->poll_list)) { unsigned nr_events = 0; - if (ctx->flags & IORING_SETUP_IOPOLL) { - /* - * inflight is the count of the maximum possible - * entries we submitted, but it can be smaller - * if we dropped some of them. If we don't have - * poll entries available, then we know that we - * have nothing left to poll for. Reset the - * inflight count to zero in that case. - */ - mutex_lock(&ctx->uring_lock); - if (!list_empty(&ctx->poll_list)) - __io_iopoll_check(ctx, &nr_events, 0); - else - inflight = 0; - mutex_unlock(&ctx->uring_lock); - } else { - /* - * Normal IO, just pretend everything completed. - * We don't have to poll completions for that. - */ - nr_events = inflight; - } - - inflight -= nr_events; - if (!inflight) + mutex_lock(&ctx->uring_lock); + if (!list_empty(&ctx->poll_list)) + io_iopoll_getevents(ctx, &nr_events, 0); + else timeout = jiffies + ctx->sq_thread_idle; + mutex_unlock(&ctx->uring_lock); } to_submit = io_sqring_entries(ctx); @@ -5139,34 +5906,49 @@ static int io_sq_thread(void *data) */ if (!to_submit || ret == -EBUSY) { /* + * Drop cur_mm before scheduling, we can't hold it for + * long periods (or over schedule()). Do this before + * adding ourselves to the waitqueue, as the unuse/drop + * may sleep. + */ + if (cur_mm) { + unuse_mm(cur_mm); + mmput(cur_mm); + cur_mm = NULL; + } + + /* * We're polling. If we're within the defined idle * period, then let us spin without work before going * to sleep. The exception is if we got EBUSY doing * more IO, we should wait for the application to * reap events and wake us up. */ - if (inflight || + if (!list_empty(&ctx->poll_list) || (!time_after(jiffies, timeout) && ret != -EBUSY && !percpu_ref_is_dying(&ctx->refs))) { + if (current->task_works) + task_work_run(); cond_resched(); continue; } + prepare_to_wait(&ctx->sqo_wait, &wait, + TASK_INTERRUPTIBLE); + /* - * Drop cur_mm before scheduling, we can't hold it for - * long periods (or over schedule()). Do this before - * adding ourselves to the waitqueue, as the unuse/drop - * may sleep. + * While doing polled IO, before going to sleep, we need + * to check if there are new reqs added to poll_list, it + * is because reqs may have been punted to io worker and + * will be added to poll_list later, hence check the + * poll_list again. */ - if (cur_mm) { - unuse_mm(cur_mm); - mmput(cur_mm); - cur_mm = NULL; + if ((ctx->flags & IORING_SETUP_IOPOLL) && + !list_empty_careful(&ctx->poll_list)) { + finish_wait(&ctx->sqo_wait, &wait); + continue; } - prepare_to_wait(&ctx->sqo_wait, &wait, - TASK_INTERRUPTIBLE); - /* Tell userspace we may need a wakeup call */ ctx->rings->sq_flags |= IORING_SQ_NEED_WAKEUP; /* make sure to read SQ tail after writing flags */ @@ -5178,6 +5960,10 @@ static int io_sq_thread(void *data) finish_wait(&ctx->sqo_wait, &wait); break; } + if (current->task_works) { + task_work_run(); + continue; + } if (signal_pending(current)) flush_signals(current); schedule(); @@ -5194,10 +5980,12 @@ static int io_sq_thread(void *data) mutex_lock(&ctx->uring_lock); ret = io_submit_sqes(ctx, to_submit, NULL, -1, &cur_mm, true); mutex_unlock(&ctx->uring_lock); - if (ret > 0) - inflight += ret; + timeout = jiffies + ctx->sq_thread_idle; } + if (current->task_works) + task_work_run(); + set_fs(old_fs); if (cur_mm) { unuse_mm(cur_mm); @@ -5262,8 +6050,13 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events, struct io_rings *rings = ctx->rings; int ret = 0; - if (io_cqring_events(ctx, false) >= min_events) - return 0; + do { + if (io_cqring_events(ctx, false) >= min_events) + return 0; + if (!current->task_works) + break; + task_work_run(); + } while (1); if (sig) { #ifdef CONFIG_COMPAT @@ -5283,6 +6076,8 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events, do { prepare_to_wait_exclusive(&ctx->wait, &iowq.wq, TASK_INTERRUPTIBLE); + if (current->task_works) + task_work_run(); if (io_should_wake(&iowq, false)) break; schedule(); @@ -5329,6 +6124,23 @@ static void io_file_ref_kill(struct percpu_ref *ref) complete(&data->done); } +static void io_file_ref_exit_and_free(struct work_struct *work) +{ + struct fixed_file_data *data; + + data = container_of(work, struct fixed_file_data, ref_work); + + /* + * Ensure any percpu-ref atomic switch callback has run, it could have + * been in progress when the files were being unregistered. Once + * that's done, we can safely exit and free the ref and containing + * data structure. + */ + rcu_barrier(); + percpu_ref_exit(&data->refs); + kfree(data); +} + static int io_sqe_files_unregister(struct io_ring_ctx *ctx) { struct fixed_file_data *data = ctx->file_data; @@ -5341,14 +6153,14 @@ static int io_sqe_files_unregister(struct io_ring_ctx *ctx) flush_work(&data->ref_work); wait_for_completion(&data->done); io_ring_file_ref_flush(data); - percpu_ref_exit(&data->refs); __io_sqe_files_unregister(ctx); nr_tables = DIV_ROUND_UP(ctx->nr_user_files, IORING_MAX_FILES_TABLE); for (i = 0; i < nr_tables; i++) kfree(data->table[i].files); kfree(data->table); - kfree(data); + INIT_WORK(&data->ref_work, io_file_ref_exit_and_free); + queue_work(system_wq, &data->ref_work); ctx->file_data = NULL; ctx->nr_user_files = 0; return 0; @@ -5575,7 +6387,6 @@ static void io_ring_file_put(struct io_ring_ctx *ctx, struct file *file) struct io_file_put { struct llist_node llist; struct file *file; - struct completion *done; }; static void io_ring_file_ref_flush(struct fixed_file_data *data) @@ -5586,10 +6397,7 @@ static void io_ring_file_ref_flush(struct fixed_file_data *data) while ((node = llist_del_all(&data->put_llist)) != NULL) { llist_for_each_entry_safe(pfile, tmp, node, llist) { io_ring_file_put(data->ctx, pfile->file); - if (pfile->done) - complete(pfile->done); - else - kfree(pfile); + kfree(pfile); } } } @@ -5600,7 +6408,6 @@ static void io_ring_file_ref_switch(struct work_struct *work) data = container_of(work, struct fixed_file_data, ref_work); io_ring_file_ref_flush(data); - percpu_ref_get(&data->refs); percpu_ref_switch_to_percpu(&data->refs); } @@ -5776,41 +6583,27 @@ static void io_atomic_switch(struct percpu_ref *ref) { struct fixed_file_data *data; + /* + * Juggle reference to ensure we hit zero, if needed, so we can + * switch back to percpu mode + */ data = container_of(ref, struct fixed_file_data, refs); - clear_bit(FFD_F_ATOMIC, &data->state); + percpu_ref_put(&data->refs); + percpu_ref_get(&data->refs); } -static bool io_queue_file_removal(struct fixed_file_data *data, +static int io_queue_file_removal(struct fixed_file_data *data, struct file *file) { - struct io_file_put *pfile, pfile_stack; - DECLARE_COMPLETION_ONSTACK(done); + struct io_file_put *pfile; - /* - * If we fail allocating the struct we need for doing async reomval - * of this file, just punt to sync and wait for it. - */ pfile = kzalloc(sizeof(*pfile), GFP_KERNEL); - if (!pfile) { - pfile = &pfile_stack; - pfile->done = &done; - } + if (!pfile) + return -ENOMEM; pfile->file = file; llist_add(&pfile->llist, &data->put_llist); - - if (pfile == &pfile_stack) { - if (!test_and_set_bit(FFD_F_ATOMIC, &data->state)) { - percpu_ref_put(&data->refs); - percpu_ref_switch_to_atomic(&data->refs, - io_atomic_switch); - } - wait_for_completion(&done); - flush_work(&data->ref_work); - return false; - } - - return true; + return 0; } static int __io_sqe_files_update(struct io_ring_ctx *ctx, @@ -5845,9 +6638,11 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, index = i & IORING_FILE_TABLE_MASK; if (table->files[index]) { file = io_file_from_index(ctx, index); + err = io_queue_file_removal(data, file); + if (err) + break; table->files[index] = NULL; - if (io_queue_file_removal(data, file)) - ref_switch = true; + ref_switch = true; } if (fd != -1) { file = fget(fd); @@ -5878,10 +6673,8 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, up->offset++; } - if (ref_switch && !test_and_set_bit(FFD_F_ATOMIC, &data->state)) { - percpu_ref_put(&data->refs); + if (ref_switch) percpu_ref_switch_to_atomic(&data->refs, io_atomic_switch); - } return done ? done : err; } @@ -5902,20 +6695,14 @@ static int io_sqe_files_update(struct io_ring_ctx *ctx, void __user *arg, return __io_sqe_files_update(ctx, &up, nr_args); } -static void io_put_work(struct io_wq_work *work) +static void io_free_work(struct io_wq_work *work) { struct io_kiocb *req = container_of(work, struct io_kiocb, work); + /* Consider that io_steal_work() relies on this ref */ io_put_req(req); } -static void io_get_work(struct io_wq_work *work) -{ - struct io_kiocb *req = container_of(work, struct io_kiocb, work); - - refcount_inc(&req->refs); -} - static int io_init_wq_offload(struct io_ring_ctx *ctx, struct io_uring_params *p) { @@ -5926,8 +6713,7 @@ static int io_init_wq_offload(struct io_ring_ctx *ctx, int ret = 0; data.user = ctx->user; - data.get_work = io_get_work; - data.put_work = io_put_work; + data.free_work = io_free_work; if (!(p->flags & IORING_SETUP_ATTACH_WQ)) { /* Do QD, or 4 * CPUS, whatever is smallest */ @@ -6329,6 +7115,21 @@ static int io_eventfd_unregister(struct io_ring_ctx *ctx) return -ENXIO; } +static int __io_destroy_buffers(int id, void *p, void *data) +{ + struct io_ring_ctx *ctx = data; + struct io_buffer *buf = p; + + __io_remove_buffers(ctx, buf, id, -1U); + return 0; +} + +static void io_destroy_buffers(struct io_ring_ctx *ctx) +{ + idr_for_each(&ctx->io_buffer_idr, __io_destroy_buffers, ctx); + idr_destroy(&ctx->io_buffer_idr); +} + static void io_ring_ctx_free(struct io_ring_ctx *ctx) { io_finish_async(ctx); @@ -6339,6 +7140,8 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx) io_sqe_buffer_unregister(ctx); io_sqe_files_unregister(ctx); io_eventfd_unregister(ctx); + io_destroy_buffers(ctx); + idr_destroy(&ctx->personality_idr); #if defined(CONFIG_UNIX) if (ctx->ring_sock) { @@ -6592,6 +7395,9 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, int submitted = 0; struct fd f; + if (current->task_works) + task_work_run(); + if (flags & ~(IORING_ENTER_GETEVENTS | IORING_ENTER_SQ_WAKEUP)) return -EINVAL; @@ -6638,7 +7444,14 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, min_complete = min(min_complete, ctx->cq_entries); - if (ctx->flags & IORING_SETUP_IOPOLL) { + /* + * When SETUP_IOPOLL and SETUP_SQPOLL are both enabled, user + * space applications don't need to do io completion events + * polling again, they can rely on io_sq_thread to do polling + * work, which can reduce cpu usage and uring_lock contention. + */ + if (ctx->flags & IORING_SETUP_IOPOLL && + !(ctx->flags & IORING_SETUP_SQPOLL)) { ret = io_iopoll_check(ctx, &nr_events, min_complete); } else { ret = io_cqring_wait(ctx, min_complete, sig, sigsz); @@ -6652,6 +7465,7 @@ out_fput: return submitted ? submitted : ret; } +#ifdef CONFIG_PROC_FS static int io_uring_show_cred(int id, void *p, void *data) { const struct cred *cred = p; @@ -6713,6 +7527,17 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) seq_printf(m, "Personalities:\n"); idr_for_each(&ctx->personality_idr, io_uring_show_cred, m); } + seq_printf(m, "PollList:\n"); + spin_lock_irq(&ctx->completion_lock); + for (i = 0; i < (1U << ctx->cancel_hash_bits); i++) { + struct hlist_head *list = &ctx->cancel_hash[i]; + struct io_kiocb *req; + + hlist_for_each_entry(req, list, hash_node) + seq_printf(m, " op=%d, task_works=%d\n", req->opcode, + req->task->task_works != NULL); + } + spin_unlock_irq(&ctx->completion_lock); mutex_unlock(&ctx->uring_lock); } @@ -6725,6 +7550,7 @@ static void io_uring_show_fdinfo(struct seq_file *m, struct file *f) percpu_ref_put(&ctx->refs); } } +#endif static const struct file_operations io_uring_fops = { .release = io_uring_release, @@ -6736,7 +7562,9 @@ static const struct file_operations io_uring_fops = { #endif .poll = io_uring_poll, .fasync = io_uring_fasync, +#ifdef CONFIG_PROC_FS .show_fdinfo = io_uring_show_fdinfo, +#endif }; static int io_allocate_scq_urings(struct io_ring_ctx *ctx, @@ -6926,7 +7754,7 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p) p->features = IORING_FEAT_SINGLE_MMAP | IORING_FEAT_NODROP | IORING_FEAT_SUBMIT_STABLE | IORING_FEAT_RW_CUR_POS | - IORING_FEAT_CUR_PERSONALITY; + IORING_FEAT_CUR_PERSONALITY | IORING_FEAT_FAST_POLL; trace_io_uring_create(ret, ctx, p->sq_entries, p->cq_entries, p->flags); return ret; err: @@ -7204,6 +8032,7 @@ static int __init io_uring_init(void) BUILD_BUG_SQE_ELEM(8, __u64, off); BUILD_BUG_SQE_ELEM(8, __u64, addr2); BUILD_BUG_SQE_ELEM(16, __u64, addr); + BUILD_BUG_SQE_ELEM(16, __u64, splice_off_in); BUILD_BUG_SQE_ELEM(24, __u32, len); BUILD_BUG_SQE_ELEM(28, __kernel_rwf_t, rw_flags); BUILD_BUG_SQE_ELEM(28, /* compat */ int, rw_flags); @@ -7218,11 +8047,14 @@ static int __init io_uring_init(void) BUILD_BUG_SQE_ELEM(28, __u32, open_flags); BUILD_BUG_SQE_ELEM(28, __u32, statx_flags); BUILD_BUG_SQE_ELEM(28, __u32, fadvise_advice); + BUILD_BUG_SQE_ELEM(28, __u32, splice_flags); BUILD_BUG_SQE_ELEM(32, __u64, user_data); BUILD_BUG_SQE_ELEM(40, __u16, buf_index); BUILD_BUG_SQE_ELEM(42, __u16, personality); + BUILD_BUG_SQE_ELEM(44, __s32, splice_fd_in); BUILD_BUG_ON(ARRAY_SIZE(io_op_defs) != IORING_OP_LAST); + BUILD_BUG_ON(__REQ_F_LAST_BIT >= 8 * sizeof(int)); req_cachep = KMEM_CACHE(io_kiocb, SLAB_HWCACHE_ALIGN | SLAB_PANIC); return 0; }; diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 2dd848a743ed..3dccc23cf010 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -936,8 +936,6 @@ do_get_write_access(handle_t *handle, struct journal_head *jh, char *frozen_buffer = NULL; unsigned long start_lock, time_lock; - if (is_handle_aborted(handle)) - return -EROFS; journal = transaction->t_journal; jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy); @@ -1152,8 +1150,8 @@ static bool jbd2_write_access_granted(handle_t *handle, struct buffer_head *bh, /* For undo access buffer must have data copied */ if (undo && !jh->b_committed_data) goto out; - if (jh->b_transaction != handle->h_transaction && - jh->b_next_transaction != handle->h_transaction) + if (READ_ONCE(jh->b_transaction) != handle->h_transaction && + READ_ONCE(jh->b_next_transaction) != handle->h_transaction) goto out; /* * There are two reasons for the barrier here: @@ -1189,6 +1187,9 @@ int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh) struct journal_head *jh; int rc; + if (is_handle_aborted(handle)) + return -EROFS; + if (jbd2_write_access_granted(handle, bh, false)) return 0; @@ -1326,6 +1327,9 @@ int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh) struct journal_head *jh; char *committed_data = NULL; + if (is_handle_aborted(handle)) + return -EROFS; + if (jbd2_write_access_granted(handle, bh, true)) return 0; @@ -2565,8 +2569,8 @@ bool __jbd2_journal_refile_buffer(struct journal_head *jh) * our jh reference and thus __jbd2_journal_file_buffer() must not * take a new one. */ - jh->b_transaction = jh->b_next_transaction; - jh->b_next_transaction = NULL; + WRITE_ONCE(jh->b_transaction, jh->b_next_transaction); + WRITE_ONCE(jh->b_next_transaction, NULL); if (buffer_freed(bh)) jlist = BJ_Forget; else if (jh->b_modified) diff --git a/fs/libfs.c b/fs/libfs.c index c686bd9caac6..3759fbacf522 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -891,7 +891,7 @@ int simple_attr_open(struct inode *inode, struct file *file, { struct simple_attr *attr; - attr = kmalloc(sizeof(*attr), GFP_KERNEL); + attr = kzalloc(sizeof(*attr), GFP_KERNEL); if (!attr) return -ENOMEM; @@ -931,9 +931,11 @@ ssize_t simple_attr_read(struct file *file, char __user *buf, if (ret) return ret; - if (*ppos) { /* continued read */ + if (*ppos && attr->get_buf[0]) { + /* continued read */ size = strlen(attr->get_buf); - } else { /* first read */ + } else { + /* first read */ u64 val; ret = attr->get(attr->data, &val); if (ret) diff --git a/fs/locks.c b/fs/locks.c index 44b6da032842..b8a31c1c4fff 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -725,7 +725,6 @@ static void __locks_delete_block(struct file_lock *waiter) { locks_delete_global_blocked(waiter); list_del_init(&waiter->fl_blocked_member); - waiter->fl_blocker = NULL; } static void __locks_wake_up_blocks(struct file_lock *blocker) @@ -740,6 +739,13 @@ static void __locks_wake_up_blocks(struct file_lock *blocker) waiter->fl_lmops->lm_notify(waiter); else wake_up(&waiter->fl_wait); + + /* + * The setting of fl_blocker to NULL marks the "done" + * point in deleting a block. Paired with acquire at the top + * of locks_delete_block(). + */ + smp_store_release(&waiter->fl_blocker, NULL); } } @@ -754,24 +760,41 @@ int locks_delete_block(struct file_lock *waiter) int status = -ENOENT; /* - * If fl_blocker is NULL, it won't be set again as this thread - * "owns" the lock and is the only one that might try to claim - * the lock. So it is safe to test fl_blocker locklessly. - * Also if fl_blocker is NULL, this waiter is not listed on - * fl_blocked_requests for some lock, so no other request can - * be added to the list of fl_blocked_requests for this - * request. So if fl_blocker is NULL, it is safe to - * locklessly check if fl_blocked_requests is empty. If both - * of these checks succeed, there is no need to take the lock. + * If fl_blocker is NULL, it won't be set again as this thread "owns" + * the lock and is the only one that might try to claim the lock. + * + * We use acquire/release to manage fl_blocker so that we can + * optimize away taking the blocked_lock_lock in many cases. + * + * The smp_load_acquire guarantees two things: + * + * 1/ that fl_blocked_requests can be tested locklessly. If something + * was recently added to that list it must have been in a locked region + * *before* the locked region when fl_blocker was set to NULL. + * + * 2/ that no other thread is accessing 'waiter', so it is safe to free + * it. __locks_wake_up_blocks is careful not to touch waiter after + * fl_blocker is released. + * + * If a lockless check of fl_blocker shows it to be NULL, we know that + * no new locks can be inserted into its fl_blocked_requests list, and + * can avoid doing anything further if the list is empty. */ - if (waiter->fl_blocker == NULL && + if (!smp_load_acquire(&waiter->fl_blocker) && list_empty(&waiter->fl_blocked_requests)) return status; + spin_lock(&blocked_lock_lock); if (waiter->fl_blocker) status = 0; __locks_wake_up_blocks(waiter); __locks_delete_block(waiter); + + /* + * The setting of fl_blocker to NULL marks the "done" point in deleting + * a block. Paired with acquire at the top of this function. + */ + smp_store_release(&waiter->fl_blocker, NULL); spin_unlock(&blocked_lock_lock); return status; } @@ -1364,7 +1387,8 @@ static int posix_lock_inode_wait(struct inode *inode, struct file_lock *fl) error = posix_lock_inode(inode, fl, NULL); if (error != FILE_LOCK_DEFERRED) break; - error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker); + error = wait_event_interruptible(fl->fl_wait, + list_empty(&fl->fl_blocked_member)); if (error) break; } @@ -1449,7 +1473,8 @@ int locks_mandatory_area(struct inode *inode, struct file *filp, loff_t start, error = posix_lock_inode(inode, &fl, NULL); if (error != FILE_LOCK_DEFERRED) break; - error = wait_event_interruptible(fl.fl_wait, !fl.fl_blocker); + error = wait_event_interruptible(fl.fl_wait, + list_empty(&fl.fl_blocked_member)); if (!error) { /* * If we've been sleeping someone might have @@ -1652,7 +1677,8 @@ restart: locks_dispose_list(&dispose); error = wait_event_interruptible_timeout(new_fl->fl_wait, - !new_fl->fl_blocker, break_time); + list_empty(&new_fl->fl_blocked_member), + break_time); percpu_down_read(&file_rwsem); spin_lock(&ctx->flc_lock); @@ -2136,7 +2162,8 @@ static int flock_lock_inode_wait(struct inode *inode, struct file_lock *fl) error = flock_lock_inode(inode, fl); if (error != FILE_LOCK_DEFERRED) break; - error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker); + error = wait_event_interruptible(fl->fl_wait, + list_empty(&fl->fl_blocked_member)); if (error) break; } @@ -2413,7 +2440,8 @@ static int do_lock_file_wait(struct file *filp, unsigned int cmd, error = vfs_lock_file(filp, cmd, fl, NULL); if (error != FILE_LOCK_DEFERRED) break; - error = wait_event_interruptible(fl->fl_wait, !fl->fl_blocker); + error = wait_event_interruptible(fl->fl_wait, + list_empty(&fl->fl_blocked_member)); if (error) break; } diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index 40b6c5ac46c0..88e1763e02f3 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -164,7 +164,7 @@ config ROOT_NFS If you want your system to mount its root file system via NFS, choose Y here. This is common practice for managing systems without local permanent storage. For details, read - <file:Documentation/filesystems/nfs/nfsroot.txt>. + <file:Documentation/admin-guide/nfs/nfsroot.rst>. Most people say N here. diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 989c30c98511..f1ff3076e4a4 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -153,6 +153,7 @@ struct nfs_client *nfs_alloc_client(const struct nfs_client_initdata *cl_init) if ((clp = kzalloc(sizeof(*clp), GFP_KERNEL)) == NULL) goto error_0; + clp->cl_minorversion = cl_init->minorversion; clp->cl_nfs_mod = cl_init->nfs_mod; if (!try_module_get(clp->cl_nfs_mod->owner)) goto error_dealloc; diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 193d6fb363b7..d4b839b6cf89 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2489,7 +2489,7 @@ static int nfs_access_get_cached_rcu(struct inode *inode, const struct cred *cre rcu_read_lock(); if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS) goto out; - lh = rcu_dereference(nfsi->access_cache_entry_lru.prev); + lh = rcu_dereference(list_tail_rcu(&nfsi->access_cache_entry_lru)); cache = list_entry(lh, struct nfs_access_entry, lru); if (lh == &nfsi->access_cache_entry_lru || cred_fscmp(cred, cache->cred) != 0) diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c index e1b938457ab9..e113fcb4bb4c 100644 --- a/fs/nfs/fs_context.c +++ b/fs/nfs/fs_context.c @@ -832,6 +832,8 @@ static int nfs_parse_source(struct fs_context *fc, if (len > maxnamlen) goto out_hostname; + kfree(ctx->nfs_server.hostname); + /* N.B. caller will free nfs_server.hostname in all cases */ ctx->nfs_server.hostname = kmemdup_nul(dev_name, len, GFP_KERNEL); if (!ctx->nfs_server.hostname) @@ -1240,6 +1242,13 @@ static int nfs_fs_context_validate(struct fs_context *fc) } ctx->nfs_mod = nfs_mod; } + + /* Ensure the filesystem context has the correct fs_type */ + if (fc->fs_type != ctx->nfs_mod->nfs_fs) { + module_put(fc->fs_type->owner); + __module_get(ctx->nfs_mod->nfs_fs->owner); + fc->fs_type = ctx->nfs_mod->nfs_fs; + } return 0; out_no_device_name: diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c index 52270bfac120..1abf126c2df4 100644 --- a/fs/nfs/fscache.c +++ b/fs/nfs/fscache.c @@ -31,6 +31,7 @@ static DEFINE_SPINLOCK(nfs_fscache_keys_lock); struct nfs_server_key { struct { uint16_t nfsversion; /* NFS protocol version */ + uint32_t minorversion; /* NFSv4 minor version */ uint16_t family; /* address family */ __be16 port; /* IP port */ } hdr; @@ -55,6 +56,7 @@ void nfs_fscache_get_client_cookie(struct nfs_client *clp) memset(&key, 0, sizeof(key)); key.hdr.nfsversion = clp->rpc_ops->version; + key.hdr.minorversion = clp->cl_minorversion; key.hdr.family = clp->cl_addr.ss_family; switch (clp->cl_addr.ss_family) { diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index b012c2668a1f..aaeeb4659bff 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -73,6 +73,7 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) struct inode *inode; char *name; int error = -ENOMEM; + unsigned long kflags = 0, kflags_out = 0; name = kstrdup(fc->source, GFP_KERNEL); if (!name) @@ -83,11 +84,14 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) if (fsinfo.fattr == NULL) goto out_name; + fsinfo.fattr->label = nfs4_label_alloc(server, GFP_KERNEL); + if (IS_ERR(fsinfo.fattr->label)) + goto out_fattr; error = server->nfs_client->rpc_ops->getroot(server, ctx->mntfh, &fsinfo); if (error < 0) { dprintk("nfs_get_root: getattr error = %d\n", -error); nfs_errorf(fc, "NFS: Couldn't getattr on root"); - goto out_fattr; + goto out_label; } inode = nfs_fhget(s, ctx->mntfh, fsinfo.fattr, NULL); @@ -95,12 +99,12 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) dprintk("nfs_get_root: get root inode failed\n"); error = PTR_ERR(inode); nfs_errorf(fc, "NFS: Couldn't get root inode"); - goto out_fattr; + goto out_label; } error = nfs_superblock_set_dummy_root(s, inode); if (error != 0) - goto out_fattr; + goto out_label; /* root dentries normally start off anonymous and get spliced in later * if the dentry tree reaches them; however if the dentry already @@ -111,7 +115,7 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) dprintk("nfs_get_root: get root dentry failed\n"); error = PTR_ERR(root); nfs_errorf(fc, "NFS: Couldn't get root dentry"); - goto out_fattr; + goto out_label; } security_d_instantiate(root, inode); @@ -123,12 +127,39 @@ int nfs_get_root(struct super_block *s, struct fs_context *fc) } spin_unlock(&root->d_lock); fc->root = root; + if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL) + kflags |= SECURITY_LSM_NATIVE_LABELS; + if (ctx->clone_data.sb) { + if (d_inode(fc->root)->i_fop != &nfs_dir_operations) { + error = -ESTALE; + goto error_splat_root; + } + /* clone lsm security options from the parent to the new sb */ + error = security_sb_clone_mnt_opts(ctx->clone_data.sb, + s, kflags, &kflags_out); + } else { + error = security_sb_set_mnt_opts(s, fc->security, + kflags, &kflags_out); + } + if (error) + goto error_splat_root; + if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL && + !(kflags_out & SECURITY_LSM_NATIVE_LABELS)) + NFS_SB(s)->caps &= ~NFS_CAP_SECURITY_LABEL; + + nfs_setsecurity(inode, fsinfo.fattr, fsinfo.fattr->label); error = 0; +out_label: + nfs4_label_free(fsinfo.fattr->label); out_fattr: nfs_free_fattr(fsinfo.fattr); out_name: kfree(name); out: return error; +error_splat_root: + dput(fc->root); + fc->root = NULL; + goto out_label; } diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index ad6077404947..f3ece8ed3203 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -153,7 +153,7 @@ struct vfsmount *nfs_d_automount(struct path *path) /* Open a new filesystem context, transferring parameters from the * parent superblock, including the network namespace. */ - fc = fs_context_for_submount(&nfs_fs_type, path->dentry); + fc = fs_context_for_submount(path->mnt->mnt_sb->s_type, path->dentry); if (IS_ERR(fc)) return ERR_CAST(fc); diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 0cd767e5c977..0bd77cc1f639 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -216,7 +216,6 @@ struct nfs_client *nfs4_alloc_client(const struct nfs_client_initdata *cl_init) INIT_LIST_HEAD(&clp->cl_ds_clients); rpc_init_wait_queue(&clp->cl_rpcwaitq, "NFS client"); clp->cl_state = 1 << NFS4CLNT_LEASE_EXPIRED; - clp->cl_minorversion = cl_init->minorversion; clp->cl_mvops = nfs_v4_minor_ops[cl_init->minorversion]; clp->cl_mig_gen = 1; #if IS_ENABLED(CONFIG_NFS_V4_1) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 69b7ab7a5815..cb34e840e4fb 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -4002,7 +4002,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh, { int error; struct nfs_fattr *fattr = info->fattr; - struct nfs4_label *label = NULL; + struct nfs4_label *label = fattr->label; error = nfs4_server_capabilities(server, mntfh); if (error < 0) { @@ -4010,23 +4010,17 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh, return error; } - label = nfs4_label_alloc(server, GFP_KERNEL); - if (IS_ERR(label)) - return PTR_ERR(label); - error = nfs4_proc_getattr(server, mntfh, fattr, label, NULL); if (error < 0) { dprintk("nfs4_get_root: getattr error = %d\n", -error); - goto err_free_label; + goto out; } if (fattr->valid & NFS_ATTR_FATTR_FSID && !nfs_fsid_equal(&server->fsid, &fattr->fsid)) memcpy(&server->fsid, &fattr->fsid, sizeof(server->fsid)); -err_free_label: - nfs4_label_free(label); - +out: return error; } diff --git a/fs/nfs/super.c b/fs/nfs/super.c index dada09b391c6..bb14bede6da5 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -1179,7 +1179,6 @@ int nfs_get_tree_common(struct fs_context *fc) struct super_block *s; int (*compare_super)(struct super_block *, struct fs_context *) = nfs_compare_super; struct nfs_server *server = ctx->server; - unsigned long kflags = 0, kflags_out = 0; int error; ctx->server = NULL; @@ -1239,26 +1238,6 @@ int nfs_get_tree_common(struct fs_context *fc) goto error_splat_super; } - if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL) - kflags |= SECURITY_LSM_NATIVE_LABELS; - if (ctx->clone_data.sb) { - if (d_inode(fc->root)->i_fop != &nfs_dir_operations) { - error = -ESTALE; - goto error_splat_root; - } - /* clone any lsm security options from the parent to the new sb */ - error = security_sb_clone_mnt_opts(ctx->clone_data.sb, s, kflags, - &kflags_out); - } else { - error = security_sb_set_mnt_opts(s, fc->security, - kflags, &kflags_out); - } - if (error) - goto error_splat_root; - if (NFS_SB(s)->caps & NFS_CAP_SECURITY_LABEL && - !(kflags_out & SECURITY_LSM_NATIVE_LABELS)) - NFS_SB(s)->caps &= ~NFS_CAP_SECURITY_LABEL; - s->s_flags |= SB_ACTIVE; error = 0; @@ -1268,10 +1247,6 @@ out: out_err_nosb: nfs_free_server(server); goto out; - -error_splat_root: - dput(fc->root); - fc->root = NULL; error_splat_super: deactivate_locked_super(s); goto out; diff --git a/fs/nsfs.c b/fs/nsfs.c index b13bfd406820..4f1205725cfe 100644 --- a/fs/nsfs.c +++ b/fs/nsfs.c @@ -247,6 +247,20 @@ out_invalid: return ERR_PTR(-EINVAL); } +/** + * ns_match() - Returns true if current namespace matches dev/ino provided. + * @ns_common: current ns + * @dev: dev_t from nsfs that will be matched against current nsfs + * @ino: ino_t from nsfs that will be matched against current nsfs + * + * Return: true if dev and ino matches the current nsfs. + */ +bool ns_match(const struct ns_common *ns, dev_t dev, ino_t ino) +{ + return (ns->inum == ino) && (nsfs_mnt->mnt_sb->s_dev == dev); +} + + static int nsfs_show_path(struct seq_file *seq, struct dentry *dentry) { struct inode *inode = d_inode(dentry); diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index 7202a1e39d70..554b744f41bf 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -92,8 +92,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) "0x%llx.", (unsigned long long)bh->b_blocknr); } first = page_buffers(page); - local_irq_save(flags); - bit_spin_lock(BH_Uptodate_Lock, &first->b_state); + spin_lock_irqsave(&first->b_uptodate_lock, flags); clear_buffer_async_read(bh); unlock_buffer(bh); tmp = bh; @@ -108,8 +107,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) } tmp = tmp->b_this_page; } while (tmp != bh); - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); - local_irq_restore(flags); + spin_unlock_irqrestore(&first->b_uptodate_lock, flags); /* * If none of the buffers had errors then we can set the page uptodate, * but we first have to perform the post read mst fixups, if the @@ -142,8 +140,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate) unlock_page(page); return; still_busy: - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state); - local_irq_restore(flags); + spin_unlock_irqrestore(&first->b_uptodate_lock, flags); return; } diff --git a/fs/open.c b/fs/open.c index 0788b3715731..b69d6eed67e6 100644 --- a/fs/open.c +++ b/fs/open.c @@ -860,9 +860,6 @@ cleanup_file: * the return value of d_splice_alias(), then the caller needs to perform dput() * on it after finish_open(). * - * On successful return @file is a fully instantiated open file. After this, if - * an error occurs in ->atomic_open(), it needs to clean up with fput(). - * * Returns zero on success or -errno if the open failed. */ int finish_open(struct file *file, struct dentry *dentry, diff --git a/fs/overlayfs/Kconfig b/fs/overlayfs/Kconfig index 444e2da4f60e..714c14c47ca5 100644 --- a/fs/overlayfs/Kconfig +++ b/fs/overlayfs/Kconfig @@ -93,6 +93,7 @@ config OVERLAY_FS_XINO_AUTO bool "Overlayfs: auto enable inode number mapping" default n depends on OVERLAY_FS + depends on 64BIT help If this config option is enabled then overlay filesystems will use unused high bits in undelying filesystem inode numbers to map all diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c index a5317216de73..87c362f65448 100644 --- a/fs/overlayfs/file.c +++ b/fs/overlayfs/file.c @@ -244,6 +244,9 @@ static void ovl_aio_cleanup_handler(struct ovl_aio_req *aio_req) if (iocb->ki_flags & IOCB_WRITE) { struct inode *inode = file_inode(orig_iocb->ki_filp); + /* Actually acquired in ovl_write_iter() */ + __sb_writers_acquired(file_inode(iocb->ki_filp)->i_sb, + SB_FREEZE_WRITE); file_end_write(iocb->ki_filp); ovl_copyattr(ovl_inode_real(inode), inode); } @@ -346,6 +349,9 @@ static ssize_t ovl_write_iter(struct kiocb *iocb, struct iov_iter *iter) goto out; file_start_write(real.file); + /* Pacify lockdep, same trick as done in aio_write() */ + __sb_writers_release(file_inode(real.file)->i_sb, + SB_FREEZE_WRITE); aio_req->fd = real; real.flags = 0; aio_req->orig_iocb = iocb; diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h index 3623d28aa4fa..3d3f2b8bdae5 100644 --- a/fs/overlayfs/overlayfs.h +++ b/fs/overlayfs/overlayfs.h @@ -318,7 +318,12 @@ static inline unsigned int ovl_xino_bits(struct super_block *sb) return ovl_same_dev(sb) ? OVL_FS(sb)->xino_mode : 0; } -static inline int ovl_inode_lock(struct inode *inode) +static inline void ovl_inode_lock(struct inode *inode) +{ + mutex_lock(&OVL_I(inode)->lock); +} + +static inline int ovl_inode_lock_interruptible(struct inode *inode) { return mutex_lock_interruptible(&OVL_I(inode)->lock); } diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index 319fe0d355b0..ac967f1cb6e5 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -1411,6 +1411,8 @@ static int ovl_get_layers(struct super_block *sb, struct ovl_fs *ofs, if (ofs->config.xino == OVL_XINO_ON) pr_info("\"xino=on\" is useless with all layers on same fs, ignore.\n"); ofs->xino_mode = 0; + } else if (ofs->config.xino == OVL_XINO_OFF) { + ofs->xino_mode = -1; } else if (ofs->config.xino == OVL_XINO_ON && ofs->xino_mode < 0) { /* * This is a roundup of number of bits needed for encoding @@ -1623,8 +1625,13 @@ static int ovl_fill_super(struct super_block *sb, void *data, int silent) sb->s_stack_depth = 0; sb->s_maxbytes = MAX_LFS_FILESIZE; /* Assume underlaying fs uses 32bit inodes unless proven otherwise */ - if (ofs->config.xino != OVL_XINO_OFF) + if (ofs->config.xino != OVL_XINO_OFF) { ofs->xino_mode = BITS_PER_LONG - 32; + if (!ofs->xino_mode) { + pr_warn("xino not supported on 32bit kernel, falling back to xino=off.\n"); + ofs->config.xino = OVL_XINO_OFF; + } + } /* alloc/destroy_inode needed for setting up traps in inode cache */ sb->s_op = &ovl_super_operations; diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c index ea005085803f..042f7eb4f7f4 100644 --- a/fs/overlayfs/util.c +++ b/fs/overlayfs/util.c @@ -509,7 +509,7 @@ int ovl_copy_up_start(struct dentry *dentry, int flags) struct inode *inode = d_inode(dentry); int err; - err = ovl_inode_lock(inode); + err = ovl_inode_lock_interruptible(inode); if (!err && ovl_already_copied_up_locked(dentry, flags)) { err = 1; /* Already copied up */ ovl_inode_unlock(inode); @@ -764,7 +764,7 @@ int ovl_nlink_start(struct dentry *dentry) return err; } - err = ovl_inode_lock(inode); + err = ovl_inode_lock_interruptible(inode); if (err) return err; diff --git a/fs/pipe.c b/fs/pipe.c index 5a34d6c22d4c..2144507447c5 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -722,9 +722,10 @@ pipe_release(struct inode *inode, struct file *file) if (file->f_mode & FMODE_WRITE) pipe->writers--; - if (pipe->readers || pipe->writers) { - wake_up_interruptible_sync_poll(&pipe->rd_wait, EPOLLIN | EPOLLRDNORM | EPOLLERR | EPOLLHUP); - wake_up_interruptible_sync_poll(&pipe->wr_wait, EPOLLOUT | EPOLLWRNORM | EPOLLERR | EPOLLHUP); + /* Was that the last reader or writer, but not the other side? */ + if (!pipe->readers != !pipe->writers) { + wake_up_interruptible_all(&pipe->rd_wait); + wake_up_interruptible_all(&pipe->wr_wait); kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); kill_fasync(&pipe->fasync_writers, SIGIO, POLL_OUT); } @@ -1026,8 +1027,8 @@ static int wait_for_partner(struct pipe_inode_info *pipe, unsigned int *cnt) static void wake_up_partner(struct pipe_inode_info *pipe) { - wake_up_interruptible(&pipe->rd_wait); - wake_up_interruptible(&pipe->wr_wait); + wake_up_interruptible_all(&pipe->rd_wait); + wake_up_interruptible_all(&pipe->wr_wait); } static int fifo_open(struct inode *inode, struct file *filp) @@ -1144,7 +1145,7 @@ err_rd: err_wr: if (!--pipe->writers) - wake_up_interruptible(&pipe->rd_wait); + wake_up_interruptible_all(&pipe->rd_wait); ret = -ERESTARTSYS; goto err; @@ -1271,8 +1272,9 @@ static long pipe_set_size(struct pipe_inode_info *pipe, unsigned long arg) pipe->max_usage = nr_slots; pipe->tail = tail; pipe->head = head; - wake_up_interruptible_all(&pipe->rd_wait); - wake_up_interruptible_all(&pipe->wr_wait); + + /* This might have made more room for writers */ + wake_up_interruptible(&pipe->wr_wait); return pipe->max_usage * PAGE_SIZE; out_revert_acct: diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c index 7fbe8f058220..d99b5d39aa90 100644 --- a/fs/pstore/inode.c +++ b/fs/pstore/inode.c @@ -87,11 +87,11 @@ static void *pstore_ftrace_seq_next(struct seq_file *s, void *v, loff_t *pos) struct pstore_private *ps = s->private; struct pstore_ftrace_seq_data *data = v; + (*pos)++; data->off += REC_SIZE; if (data->off + REC_SIZE > ps->total_size) return NULL; - (*pos)++; return data; } @@ -101,6 +101,9 @@ static int pstore_ftrace_seq_show(struct seq_file *s, void *v) struct pstore_ftrace_seq_data *data = v; struct pstore_ftrace_record *rec; + if (!data) + return 0; + rec = (struct pstore_ftrace_record *)(ps->record->buf + data->off); seq_printf(s, "CPU:%d ts:%llu %08lx %08lx %ps <- %pS\n", diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index d896457e7c11..408277ee3cdb 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -823,9 +823,9 @@ static int __init pstore_init(void) ret = pstore_init_fs(); if (ret) - return ret; + free_buf_for_compression(); - return 0; + return ret; } late_initcall(pstore_init); diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c index 013486b5125e..795622190c01 100644 --- a/fs/pstore/ram.c +++ b/fs/pstore/ram.c @@ -963,7 +963,6 @@ static void __init ramoops_register_dummy(void) pr_info("could not create platform device: %ld\n", PTR_ERR(dummy)); dummy = NULL; - ramoops_unregister_dummy(); } } diff --git a/fs/pstore/ram_core.c b/fs/pstore/ram_core.c index 1f4d8c06f9be..c917c191e78c 100644 --- a/fs/pstore/ram_core.c +++ b/fs/pstore/ram_core.c @@ -34,7 +34,7 @@ struct persistent_ram_buffer { uint32_t sig; atomic_t start; atomic_t size; - uint8_t data[0]; + uint8_t data[]; }; #define PERSISTENT_RAM_SIG (0x43474244) /* DBGC */ diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 072156c4f895..5c766330e493 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2599,7 +2599,6 @@ static int journal_init_dev(struct super_block *super, int result; dev_t jdev; fmode_t blkdev_mode = FMODE_READ | FMODE_WRITE | FMODE_EXCL; - char b[BDEVNAME_SIZE]; result = 0; @@ -2621,8 +2620,8 @@ static int journal_init_dev(struct super_block *super, result = PTR_ERR(journal->j_dev_bd); journal->j_dev_bd = NULL; reiserfs_warning(super, "sh-458", - "cannot init journal device '%s': %i", - __bdevname(jdev, b), result); + "cannot init journal device unknown-block(%u,%u): %i", + MAJOR(jdev), MINOR(jdev), result); return result; } else if (jdev != super->s_dev) set_blocksize(journal->j_dev_bd, super->s_blocksize); diff --git a/fs/splice.c b/fs/splice.c index d671936d0aad..4735defc46ee 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1109,9 +1109,9 @@ static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe, /* * Determine where to splice to/from. */ -static long do_splice(struct file *in, loff_t __user *off_in, - struct file *out, loff_t __user *off_out, - size_t len, unsigned int flags) +long do_splice(struct file *in, loff_t __user *off_in, + struct file *out, loff_t __user *off_out, + size_t len, unsigned int flags) { struct pipe_inode_info *ipipe; struct pipe_inode_info *opipe; diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 130fc6fbcc03..26bbf960e2a2 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -558,3 +558,151 @@ void sysfs_remove_bin_file(struct kobject *kobj, kernfs_remove_by_name(kobj->sd, attr->attr.name); } EXPORT_SYMBOL_GPL(sysfs_remove_bin_file); + +static int internal_change_owner(struct kernfs_node *kn, kuid_t kuid, + kgid_t kgid) +{ + struct iattr newattrs = { + .ia_valid = ATTR_UID | ATTR_GID, + .ia_uid = kuid, + .ia_gid = kgid, + }; + return kernfs_setattr(kn, &newattrs); +} + +/** + * sysfs_link_change_owner - change owner of a sysfs file. + * @kobj: object of the kernfs_node the symlink is located in. + * @targ: object of the kernfs_node the symlink points to. + * @name: name of the link. + * @kuid: new owner's kuid + * @kgid: new owner's kgid + * + * This function looks up the sysfs symlink entry @name under @kobj and changes + * the ownership to @kuid/@kgid. The symlink is looked up in the namespace of + * @targ. + * + * Returns 0 on success or error code on failure. + */ +int sysfs_link_change_owner(struct kobject *kobj, struct kobject *targ, + const char *name, kuid_t kuid, kgid_t kgid) +{ + struct kernfs_node *kn = NULL; + int error; + + if (!name || !kobj->state_in_sysfs || !targ->state_in_sysfs) + return -EINVAL; + + error = -ENOENT; + kn = kernfs_find_and_get_ns(kobj->sd, name, targ->sd->ns); + if (!kn) + goto out; + + error = -EINVAL; + if (kernfs_type(kn) != KERNFS_LINK) + goto out; + if (kn->symlink.target_kn->priv != targ) + goto out; + + error = internal_change_owner(kn, kuid, kgid); + +out: + kernfs_put(kn); + return error; +} + +/** + * sysfs_file_change_owner - change owner of a sysfs file. + * @kobj: object. + * @name: name of the file to change. + * @kuid: new owner's kuid + * @kgid: new owner's kgid + * + * This function looks up the sysfs entry @name under @kobj and changes the + * ownership to @kuid/@kgid. + * + * Returns 0 on success or error code on failure. + */ +int sysfs_file_change_owner(struct kobject *kobj, const char *name, kuid_t kuid, + kgid_t kgid) +{ + struct kernfs_node *kn; + int error; + + if (!name) + return -EINVAL; + + if (!kobj->state_in_sysfs) + return -EINVAL; + + kn = kernfs_find_and_get(kobj->sd, name); + if (!kn) + return -ENOENT; + + error = internal_change_owner(kn, kuid, kgid); + + kernfs_put(kn); + + return error; +} +EXPORT_SYMBOL_GPL(sysfs_file_change_owner); + +/** + * sysfs_change_owner - change owner of the given object. + * @kobj: object. + * @kuid: new owner's kuid + * @kgid: new owner's kgid + * + * Change the owner of the default directory, files, groups, and attributes of + * @kobj to @kuid/@kgid. Note that sysfs_change_owner mirrors how the sysfs + * entries for a kobject are added by driver core. In summary, + * sysfs_change_owner() takes care of the default directory entry for @kobj, + * the default attributes associated with the ktype of @kobj and the default + * attributes associated with the ktype of @kobj. + * Additional properties not added by driver core have to be changed by the + * driver or subsystem which created them. This is similar to how + * driver/subsystem specific entries are removed. + * + * Returns 0 on success or error code on failure. + */ +int sysfs_change_owner(struct kobject *kobj, kuid_t kuid, kgid_t kgid) +{ + int error; + const struct kobj_type *ktype; + + if (!kobj->state_in_sysfs) + return -EINVAL; + + /* Change the owner of the kobject itself. */ + error = internal_change_owner(kobj->sd, kuid, kgid); + if (error) + return error; + + ktype = get_ktype(kobj); + if (ktype) { + struct attribute **kattr; + + /* + * Change owner of the default attributes associated with the + * ktype of @kobj. + */ + for (kattr = ktype->default_attrs; kattr && *kattr; kattr++) { + error = sysfs_file_change_owner(kobj, (*kattr)->name, + kuid, kgid); + if (error) + return error; + } + + /* + * Change owner of the default groups associated with the + * ktype of @kobj. + */ + error = sysfs_groups_change_owner(kobj, ktype->default_groups, + kuid, kgid); + if (error) + return error; + } + + return 0; +} +EXPORT_SYMBOL_GPL(sysfs_change_owner); diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index c4ab045926b7..5afe0e7ff7cd 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -13,6 +13,7 @@ #include <linux/dcache.h> #include <linux/namei.h> #include <linux/err.h> +#include <linux/fs.h> #include "sysfs.h" @@ -457,3 +458,117 @@ int __compat_only_sysfs_link_entry_to_kobj(struct kobject *kobj, return PTR_ERR_OR_ZERO(link); } EXPORT_SYMBOL_GPL(__compat_only_sysfs_link_entry_to_kobj); + +static int sysfs_group_attrs_change_owner(struct kernfs_node *grp_kn, + const struct attribute_group *grp, + struct iattr *newattrs) +{ + struct kernfs_node *kn; + int error; + + if (grp->attrs) { + struct attribute *const *attr; + + for (attr = grp->attrs; *attr; attr++) { + kn = kernfs_find_and_get(grp_kn, (*attr)->name); + if (!kn) + return -ENOENT; + + error = kernfs_setattr(kn, newattrs); + kernfs_put(kn); + if (error) + return error; + } + } + + if (grp->bin_attrs) { + struct bin_attribute *const *bin_attr; + + for (bin_attr = grp->bin_attrs; *bin_attr; bin_attr++) { + kn = kernfs_find_and_get(grp_kn, (*bin_attr)->attr.name); + if (!kn) + return -ENOENT; + + error = kernfs_setattr(kn, newattrs); + kernfs_put(kn); + if (error) + return error; + } + } + + return 0; +} + +/** + * sysfs_group_change_owner - change owner of an attribute group. + * @kobj: The kobject containing the group. + * @grp: The attribute group. + * @kuid: new owner's kuid + * @kgid: new owner's kgid + * + * Returns 0 on success or error code on failure. + */ +int sysfs_group_change_owner(struct kobject *kobj, + const struct attribute_group *grp, kuid_t kuid, + kgid_t kgid) +{ + struct kernfs_node *grp_kn; + int error; + struct iattr newattrs = { + .ia_valid = ATTR_UID | ATTR_GID, + .ia_uid = kuid, + .ia_gid = kgid, + }; + + if (!kobj->state_in_sysfs) + return -EINVAL; + + if (grp->name) { + grp_kn = kernfs_find_and_get(kobj->sd, grp->name); + } else { + kernfs_get(kobj->sd); + grp_kn = kobj->sd; + } + if (!grp_kn) + return -ENOENT; + + error = kernfs_setattr(grp_kn, &newattrs); + if (!error) + error = sysfs_group_attrs_change_owner(grp_kn, grp, &newattrs); + + kernfs_put(grp_kn); + + return error; +} +EXPORT_SYMBOL_GPL(sysfs_group_change_owner); + +/** + * sysfs_groups_change_owner - change owner of a set of attribute groups. + * @kobj: The kobject containing the groups. + * @groups: The attribute groups. + * @kuid: new owner's kuid + * @kgid: new owner's kgid + * + * Returns 0 on success or error code on failure. + */ +int sysfs_groups_change_owner(struct kobject *kobj, + const struct attribute_group **groups, + kuid_t kuid, kgid_t kgid) +{ + int error = 0, i; + + if (!kobj->state_in_sysfs) + return -EINVAL; + + if (!groups) + return 0; + + for (i = 0; groups[i]; i++) { + error = sysfs_group_change_owner(kobj, groups[i], kuid, kgid); + if (error) + break; + } + + return error; +} +EXPORT_SYMBOL_GPL(sysfs_groups_change_owner); diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c index d49fc04f2d7d..3df9be2c684c 100644 --- a/fs/ubifs/ioctl.c +++ b/fs/ubifs/ioctl.c @@ -208,6 +208,9 @@ long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case FS_IOC_GET_ENCRYPTION_KEY_STATUS: return fscrypt_ioctl_get_key_status(file, (void __user *)arg); + case FS_IOC_GET_ENCRYPTION_NONCE: + return fscrypt_ioctl_get_nonce(file, (void __user *)arg); + default: return -ENOTTY; } @@ -230,6 +233,7 @@ long ubifs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case FS_IOC_REMOVE_ENCRYPTION_KEY: case FS_IOC_REMOVE_ENCRYPTION_KEY_ALL_USERS: case FS_IOC_GET_ENCRYPTION_KEY_STATUS: + case FS_IOC_GET_ENCRYPTION_NONCE: break; default: return -ENOIOCTLCMD; diff --git a/fs/zonefs/Kconfig b/fs/zonefs/Kconfig index fb87ad372e29..ef2697b78820 100644 --- a/fs/zonefs/Kconfig +++ b/fs/zonefs/Kconfig @@ -2,6 +2,7 @@ config ZONEFS_FS tristate "zonefs filesystem support" depends on BLOCK depends on BLK_DEV_ZONED + select FS_IOMAP help zonefs is a simple file system which exposes zones of a zoned block device (e.g. host-managed or host-aware SMR disk drives) as files. diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c index 8bc6ef82d693..3ce9829a6936 100644 --- a/fs/zonefs/super.c +++ b/fs/zonefs/super.c @@ -178,7 +178,8 @@ static void zonefs_update_stats(struct inode *inode, loff_t new_isize) * amount of readable data in the zone. */ static loff_t zonefs_check_zone_condition(struct inode *inode, - struct blk_zone *zone, bool warn) + struct blk_zone *zone, bool warn, + bool mount) { struct zonefs_inode_info *zi = ZONEFS_I(inode); @@ -196,13 +197,26 @@ static loff_t zonefs_check_zone_condition(struct inode *inode, zone->wp = zone->start; return 0; case BLK_ZONE_COND_READONLY: - /* Do not allow writes in read-only zones */ + /* + * The write pointer of read-only zones is invalid. If such a + * zone is found during mount, the file size cannot be retrieved + * so we treat the zone as offline (mount == true case). + * Otherwise, keep the file size as it was when last updated + * so that the user can recover data. In both cases, writes are + * always disabled for the zone. + */ if (warn) zonefs_warn(inode->i_sb, "inode %lu: read-only zone\n", inode->i_ino); inode->i_flags |= S_IMMUTABLE; + if (mount) { + zone->cond = BLK_ZONE_COND_OFFLINE; + inode->i_mode &= ~0777; + zone->wp = zone->start; + return 0; + } inode->i_mode &= ~0222; - /* fallthrough */ + return i_size_read(inode); default: if (zi->i_ztype == ZONEFS_ZTYPE_CNV) return zi->i_max_size; @@ -231,7 +245,7 @@ static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx, * as there is no inconsistency between the inode size and the amount of * data writen in the zone (data_size). */ - data_size = zonefs_check_zone_condition(inode, zone, true); + data_size = zonefs_check_zone_condition(inode, zone, true, false); isize = i_size_read(inode); if (zone->cond != BLK_ZONE_COND_OFFLINE && zone->cond != BLK_ZONE_COND_READONLY && @@ -274,7 +288,7 @@ static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx, if (zone->cond != BLK_ZONE_COND_OFFLINE) { zone->cond = BLK_ZONE_COND_OFFLINE; data_size = zonefs_check_zone_condition(inode, zone, - false); + false, false); } } else if (zone->cond == BLK_ZONE_COND_READONLY || sbi->s_mount_opts & ZONEFS_MNTOPT_ERRORS_ZRO) { @@ -283,7 +297,7 @@ static int zonefs_io_error_cb(struct blk_zone *zone, unsigned int idx, if (zone->cond != BLK_ZONE_COND_READONLY) { zone->cond = BLK_ZONE_COND_READONLY; data_size = zonefs_check_zone_condition(inode, zone, - false); + false, false); } } @@ -601,13 +615,13 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from) ssize_t ret; /* - * For async direct IOs to sequential zone files, ignore IOCB_NOWAIT + * For async direct IOs to sequential zone files, refuse IOCB_NOWAIT * as this can cause write reordering (e.g. the first aio gets EAGAIN * on the inode lock but the second goes through but is now unaligned). */ - if (zi->i_ztype == ZONEFS_ZTYPE_SEQ && !is_sync_kiocb(iocb) - && (iocb->ki_flags & IOCB_NOWAIT)) - iocb->ki_flags &= ~IOCB_NOWAIT; + if (zi->i_ztype == ZONEFS_ZTYPE_SEQ && !is_sync_kiocb(iocb) && + (iocb->ki_flags & IOCB_NOWAIT)) + return -EOPNOTSUPP; if (iocb->ki_flags & IOCB_NOWAIT) { if (!inode_trylock(inode)) @@ -975,7 +989,7 @@ static void zonefs_init_file_inode(struct inode *inode, struct blk_zone *zone, zi->i_zsector = zone->start; zi->i_max_size = min_t(loff_t, MAX_LFS_FILESIZE, zone->len << SECTOR_SHIFT); - zi->i_wpoffset = zonefs_check_zone_condition(inode, zone, true); + zi->i_wpoffset = zonefs_check_zone_condition(inode, zone, true, true); inode->i_uid = sbi->s_uid; inode->i_gid = sbi->s_gid; |