diff options
Diffstat (limited to 'fs')
43 files changed, 806 insertions, 421 deletions
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index a4cb4b642987..da519c1b6ad0 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -682,18 +682,30 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans, u64 search_start; int ret; - if (test_bit(BTRFS_ROOT_DELETING, &root->state)) - btrfs_err(fs_info, - "COW'ing blocks on a fs root that's being dropped"); - - if (trans->transaction != fs_info->running_transaction) - WARN(1, KERN_CRIT "trans %llu running %llu\n", - trans->transid, - fs_info->running_transaction->transid); + if (unlikely(test_bit(BTRFS_ROOT_DELETING, &root->state))) { + btrfs_abort_transaction(trans, -EUCLEAN); + btrfs_crit(fs_info, + "attempt to COW block %llu on root %llu that is being deleted", + buf->start, btrfs_root_id(root)); + return -EUCLEAN; + } - if (trans->transid != fs_info->generation) - WARN(1, KERN_CRIT "trans %llu running %llu\n", - trans->transid, fs_info->generation); + /* + * COWing must happen through a running transaction, which always + * matches the current fs generation (it's a transaction with a state + * less than TRANS_STATE_UNBLOCKED). If it doesn't, then turn the fs + * into error state to prevent the commit of any transaction. + */ + if (unlikely(trans->transaction != fs_info->running_transaction || + trans->transid != fs_info->generation)) { + btrfs_abort_transaction(trans, -EUCLEAN); + btrfs_crit(fs_info, +"unexpected transaction when attempting to COW block %llu on root %llu, transaction %llu running transaction %llu fs generation %llu", + buf->start, btrfs_root_id(root), trans->transid, + fs_info->running_transaction->transid, + fs_info->generation); + return -EUCLEAN; + } if (!should_cow_block(trans, root, buf)) { *cow_ret = buf; @@ -805,8 +817,22 @@ int btrfs_realloc_node(struct btrfs_trans_handle *trans, int progress_passed = 0; struct btrfs_disk_key disk_key; - WARN_ON(trans->transaction != fs_info->running_transaction); - WARN_ON(trans->transid != fs_info->generation); + /* + * COWing must happen through a running transaction, which always + * matches the current fs generation (it's a transaction with a state + * less than TRANS_STATE_UNBLOCKED). If it doesn't, then turn the fs + * into error state to prevent the commit of any transaction. + */ + if (unlikely(trans->transaction != fs_info->running_transaction || + trans->transid != fs_info->generation)) { + btrfs_abort_transaction(trans, -EUCLEAN); + btrfs_crit(fs_info, +"unexpected transaction when attempting to reallocate parent %llu for root %llu, transaction %llu running transaction %llu fs generation %llu", + parent->start, btrfs_root_id(root), trans->transid, + fs_info->running_transaction->transid, + fs_info->generation); + return -EUCLEAN; + } parent_nritems = btrfs_header_nritems(parent); blocksize = fs_info->nodesize; diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index caf0bbd028d1..90aaedce1548 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -313,7 +313,7 @@ static struct btrfs_delayed_item *btrfs_alloc_delayed_item(u16 data_len, { struct btrfs_delayed_item *item; - item = kmalloc(sizeof(*item) + data_len, GFP_NOFS); + item = kmalloc(struct_size(item, data, data_len), GFP_NOFS); if (item) { item->data_len = data_len; item->type = type; diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h index dc1085b2a397..1da213197f55 100644 --- a/fs/btrfs/delayed-inode.h +++ b/fs/btrfs/delayed-inode.h @@ -95,7 +95,7 @@ struct btrfs_delayed_item { bool logged; /* The maximum leaf size is 64K, so u16 is more than enough. */ u16 data_len; - char data[]; + char data[] __counted_by(data_len); }; static inline void btrfs_init_delayed_root( diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 75ab766fe156..8e7d03bc1b56 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -2978,7 +2978,7 @@ static void get_block_group_info(struct list_head *groups_list, static long btrfs_ioctl_space_info(struct btrfs_fs_info *fs_info, void __user *arg) { - struct btrfs_ioctl_space_args space_args; + struct btrfs_ioctl_space_args space_args = { 0 }; struct btrfs_ioctl_space_info space; struct btrfs_ioctl_space_info *dest; struct btrfs_ioctl_space_info *dest_orig; @@ -4338,7 +4338,7 @@ static int _btrfs_ioctl_send(struct inode *inode, void __user *argp, bool compat if (compat) { #if defined(CONFIG_64BIT) && defined(CONFIG_COMPAT) - struct btrfs_ioctl_send_args_32 args32; + struct btrfs_ioctl_send_args_32 args32 = { 0 }; ret = copy_from_user(&args32, argp, sizeof(args32)); if (ret) diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index 6b309f8a99a8..93869cda6af9 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -219,8 +219,8 @@ do { \ (errno))) { \ /* Stack trace printed. */ \ } else { \ - btrfs_debug((trans)->fs_info, \ - "Transaction aborted (error %d)", \ + btrfs_err((trans)->fs_info, \ + "Transaction aborted (error %d)", \ (errno)); \ } \ } \ diff --git a/fs/ceph/crypto.c b/fs/ceph/crypto.c index e1f31b86fd48..5b5112c78462 100644 --- a/fs/ceph/crypto.c +++ b/fs/ceph/crypto.c @@ -460,7 +460,7 @@ int ceph_fname_to_usr(const struct ceph_fname *fname, struct fscrypt_str *tname, out: fscrypt_fname_free_buffer(&_tname); out_inode: - if ((dir != fname->dir) && !IS_ERR(dir)) { + if (dir != fname->dir) { if ((dir->i_state & I_NEW)) discard_new_inode(dir); else diff --git a/fs/ceph/file.c b/fs/ceph/file.c index b1da02f5dbe3..b5f8038065d7 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -2969,7 +2969,7 @@ static ssize_t __ceph_copy_file_range(struct file *src_file, loff_t src_off, ret = do_splice_direct(src_file, &src_off, dst_file, &dst_off, src_objlen, flags); /* Abort on short copies or on error */ - if (ret < src_objlen) { + if (ret < (long)src_objlen) { dout("Failed partial copy (%zd)\n", ret); goto out; } diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 800ab7920513..b79100f720b3 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -769,9 +769,7 @@ int ceph_fill_file_size(struct inode *inode, int issued, ci->i_truncate_seq = truncate_seq; /* the MDS should have revoked these caps */ - WARN_ON_ONCE(issued & (CEPH_CAP_FILE_EXCL | - CEPH_CAP_FILE_RD | - CEPH_CAP_FILE_WR | + WARN_ON_ONCE(issued & (CEPH_CAP_FILE_RD | CEPH_CAP_FILE_LAZYIO)); /* * If we hold relevant caps, or in the case where we're diff --git a/fs/erofs/decompressor_lzma.c b/fs/erofs/decompressor_lzma.c index 73091fbe3ea4..dee10d22ada9 100644 --- a/fs/erofs/decompressor_lzma.c +++ b/fs/erofs/decompressor_lzma.c @@ -217,9 +217,12 @@ again: strm->buf.out_size = min_t(u32, outlen, PAGE_SIZE - pageofs); outlen -= strm->buf.out_size; - if (!rq->out[no] && rq->fillgaps) /* deduped */ + if (!rq->out[no] && rq->fillgaps) { /* deduped */ rq->out[no] = erofs_allocpage(pagepool, GFP_KERNEL | __GFP_NOFAIL); + set_page_private(rq->out[no], + Z_EROFS_SHORTLIVED_PAGE); + } if (rq->out[no]) strm->buf.out = kmap(rq->out[no]) + pageofs; pageofs = 0; diff --git a/fs/erofs/super.c b/fs/erofs/super.c index 44a24d573f1f..3700af9ee173 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -235,7 +235,7 @@ static int erofs_init_device(struct erofs_buf *buf, struct super_block *sb, return PTR_ERR(ptr); dis = ptr + erofs_blkoff(sb, *pos); - if (!dif->path) { + if (!sbi->devs->flatdev && !dif->path) { if (!dis->tag[0]) { erofs_err(sb, "empty device tag @ pos %llu", *pos); return -EINVAL; diff --git a/fs/fs_context.c b/fs/fs_context.c index a0ad7a0c4680..98589aae5208 100644 --- a/fs/fs_context.c +++ b/fs/fs_context.c @@ -192,17 +192,19 @@ int vfs_parse_fs_string(struct fs_context *fc, const char *key, EXPORT_SYMBOL(vfs_parse_fs_string); /** - * generic_parse_monolithic - Parse key[=val][,key[=val]]* mount data + * vfs_parse_monolithic_sep - Parse key[=val][,key[=val]]* mount data * @fc: The superblock configuration to fill in. * @data: The data to parse + * @sep: callback for separating next option * - * Parse a blob of data that's in key[=val][,key[=val]]* form. This can be - * called from the ->monolithic_mount_data() fs_context operation. + * Parse a blob of data that's in key[=val][,key[=val]]* form with a custom + * option separator callback. * * Returns 0 on success or the error returned by the ->parse_option() fs_context * operation on failure. */ -int generic_parse_monolithic(struct fs_context *fc, void *data) +int vfs_parse_monolithic_sep(struct fs_context *fc, void *data, + char *(*sep)(char **)) { char *options = data, *key; int ret = 0; @@ -214,7 +216,7 @@ int generic_parse_monolithic(struct fs_context *fc, void *data) if (ret) return ret; - while ((key = strsep(&options, ",")) != NULL) { + while ((key = sep(&options)) != NULL) { if (*key) { size_t v_len = 0; char *value = strchr(key, '='); @@ -233,6 +235,28 @@ int generic_parse_monolithic(struct fs_context *fc, void *data) return ret; } +EXPORT_SYMBOL(vfs_parse_monolithic_sep); + +static char *vfs_parse_comma_sep(char **s) +{ + return strsep(s, ","); +} + +/** + * generic_parse_monolithic - Parse key[=val][,key[=val]]* mount data + * @fc: The superblock configuration to fill in. + * @data: The data to parse + * + * Parse a blob of data that's in key[=val][,key[=val]]* form. This can be + * called from the ->monolithic_mount_data() fs_context operation. + * + * Returns 0 on success or the error returned by the ->parse_option() fs_context + * operation on failure. + */ +int generic_parse_monolithic(struct fs_context *fc, void *data) +{ + return vfs_parse_monolithic_sep(fc, data, vfs_parse_comma_sep); +} EXPORT_SYMBOL(generic_parse_monolithic); /** diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 3508d8238826..7016eaadf555 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -10622,7 +10622,9 @@ static void nfs4_disable_swap(struct inode *inode) */ struct nfs_client *clp = NFS_SERVER(inode)->nfs_client; - nfs4_schedule_state_manager(clp); + set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state); + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); + wake_up_var(&clp->cl_state); } static const struct inode_operations nfs4_dir_inode_operations = { diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index e079987af4a3..9a5d911a7edc 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1209,16 +1209,26 @@ void nfs4_schedule_state_manager(struct nfs_client *clp) { struct task_struct *task; char buf[INET6_ADDRSTRLEN + sizeof("-manager") + 1]; + struct rpc_clnt *clnt = clp->cl_rpcclient; + bool swapon = false; - if (clp->cl_rpcclient->cl_shutdown) + if (clnt->cl_shutdown) return; set_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state); - if (test_and_set_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state) != 0) { - wake_up_var(&clp->cl_state); - return; + + if (atomic_read(&clnt->cl_swapper)) { + swapon = !test_and_set_bit(NFS4CLNT_MANAGER_AVAILABLE, + &clp->cl_state); + if (!swapon) { + wake_up_var(&clp->cl_state); + return; + } } - set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state); + + if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0) + return; + __module_get(THIS_MODULE); refcount_inc(&clp->cl_count); @@ -1235,8 +1245,9 @@ void nfs4_schedule_state_manager(struct nfs_client *clp) __func__, PTR_ERR(task)); if (!nfs_client_init_is_complete(clp)) nfs_mark_client_ready(clp, PTR_ERR(task)); + if (swapon) + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); nfs4_clear_state_manager_bit(clp); - clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); nfs_put_client(clp); module_put(THIS_MODULE); } @@ -2703,6 +2714,13 @@ static void nfs4_state_manager(struct nfs_client *clp) nfs4_end_drain_session(clp); nfs4_clear_state_manager_bit(clp); + if (test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state) && + !test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, + &clp->cl_state)) { + memflags = memalloc_nofs_save(); + continue; + } + if (!test_and_set_bit(NFS4CLNT_RECALL_RUNNING, &clp->cl_state)) { if (test_and_clear_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state)) { nfs_client_return_marked_delegations(clp); @@ -2741,22 +2759,25 @@ static int nfs4_run_state_manager(void *ptr) allow_signal(SIGKILL); again: - set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state); nfs4_state_manager(clp); - if (atomic_read(&cl->cl_swapper)) { + + if (test_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state) && + !test_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state)) { wait_var_event_interruptible(&clp->cl_state, test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state)); - if (atomic_read(&cl->cl_swapper) && - test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state)) + if (!atomic_read(&cl->cl_swapper)) + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); + if (refcount_read(&clp->cl_count) > 1 && !signalled() && + !test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state)) goto again; /* Either no longer a swapper, or were signalled */ + clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); } - clear_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state); if (refcount_read(&clp->cl_count) > 1 && !signalled() && test_bit(NFS4CLNT_RUN_MANAGER, &clp->cl_state) && - !test_and_set_bit(NFS4CLNT_MANAGER_AVAILABLE, &clp->cl_state)) + !test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state)) goto again; nfs_put_client(clp); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 8c1ee1a1a28f..7720b5e43014 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -802,8 +802,8 @@ static void nfs_inode_remove_request(struct nfs_page *req) } if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) { - nfs_release_request(req); atomic_long_dec(&NFS_I(nfs_page_to_inode(req))->nrequests); + nfs_release_request(req); } } diff --git a/fs/overlayfs/export.c b/fs/overlayfs/export.c index c8c8588bd98c..26b782c53910 100644 --- a/fs/overlayfs/export.c +++ b/fs/overlayfs/export.c @@ -188,7 +188,7 @@ static int ovl_check_encode_origin(struct dentry *dentry) /* Lower file handle for non-upper non-decodable */ if (!ovl_dentry_upper(dentry) && !decodable) - return 0; + return 1; /* Upper file handle for pure upper */ if (!ovl_dentry_lower(dentry)) diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c index 693971d20280..8be4dc050d1e 100644 --- a/fs/overlayfs/file.c +++ b/fs/overlayfs/file.c @@ -341,7 +341,6 @@ static ssize_t ovl_read_iter(struct kiocb *iocb, struct iov_iter *iter) if (!aio_req) goto out; - real.flags = 0; aio_req->orig_iocb = iocb; kiocb_clone(&aio_req->iocb, iocb, get_file(real.file)); aio_req->iocb.ki_complete = ovl_aio_rw_complete; @@ -413,7 +412,6 @@ static ssize_t ovl_write_iter(struct kiocb *iocb, struct iov_iter *iter) if (!aio_req) goto out; - real.flags = 0; aio_req->orig_iocb = iocb; kiocb_clone(&aio_req->iocb, iocb, get_file(real.file)); aio_req->iocb.ki_flags = ifl; diff --git a/fs/overlayfs/ovl_entry.h b/fs/overlayfs/ovl_entry.h index e9539f98e86a..d82d2a043da2 100644 --- a/fs/overlayfs/ovl_entry.h +++ b/fs/overlayfs/ovl_entry.h @@ -8,6 +8,7 @@ struct ovl_config { char *upperdir; char *workdir; + char **lowerdirs; bool default_permissions; int redirect_mode; int verity_mode; @@ -39,17 +40,8 @@ struct ovl_layer { int idx; /* One fsid per unique underlying sb (upper fsid == 0) */ int fsid; - char *name; }; -/* - * ovl_free_fs() relies on @mnt being the first member when unmounting - * the private mounts created for each layer. Let's check both the - * offset and type. - */ -static_assert(offsetof(struct ovl_layer, mnt) == 0); -static_assert(__same_type(typeof_member(struct ovl_layer, mnt), struct vfsmount *)); - struct ovl_path { const struct ovl_layer *layer; struct dentry *dentry; diff --git a/fs/overlayfs/params.c b/fs/overlayfs/params.c index b9355bb6d75a..f6ff23fd101c 100644 --- a/fs/overlayfs/params.c +++ b/fs/overlayfs/params.c @@ -157,6 +157,34 @@ const struct fs_parameter_spec ovl_parameter_spec[] = { {} }; +static char *ovl_next_opt(char **s) +{ + char *sbegin = *s; + char *p; + + if (sbegin == NULL) + return NULL; + + for (p = sbegin; *p; p++) { + if (*p == '\\') { + p++; + if (!*p) + break; + } else if (*p == ',') { + *p = '\0'; + *s = p + 1; + return sbegin; + } + } + *s = NULL; + return sbegin; +} + +static int ovl_parse_monolithic(struct fs_context *fc, void *data) +{ + return vfs_parse_monolithic_sep(fc, data, ovl_next_opt); +} + static ssize_t ovl_parse_param_split_lowerdirs(char *str) { ssize_t nr_layers = 1, nr_colons = 0; @@ -164,7 +192,8 @@ static ssize_t ovl_parse_param_split_lowerdirs(char *str) for (s = d = str;; s++, d++) { if (*s == '\\') { - s++; + /* keep esc chars in split lowerdir */ + *d++ = *s++; } else if (*s == ':') { bool next_colon = (*(s + 1) == ':'); @@ -239,7 +268,7 @@ static void ovl_unescape(char *s) } } -static int ovl_mount_dir(const char *name, struct path *path) +static int ovl_mount_dir(const char *name, struct path *path, bool upper) { int err = -ENOMEM; char *tmp = kstrdup(name, GFP_KERNEL); @@ -248,7 +277,7 @@ static int ovl_mount_dir(const char *name, struct path *path) ovl_unescape(tmp); err = ovl_mount_dir_noesc(tmp, path); - if (!err && path->dentry->d_flags & DCACHE_OP_REAL) { + if (!err && upper && path->dentry->d_flags & DCACHE_OP_REAL) { pr_err("filesystem on '%s' not supported as upperdir\n", tmp); path_put_init(path); @@ -269,7 +298,7 @@ static int ovl_parse_param_upperdir(const char *name, struct fs_context *fc, struct path path; char *dup; - err = ovl_mount_dir(name, &path); + err = ovl_mount_dir(name, &path, true); if (err) return err; @@ -321,12 +350,6 @@ static void ovl_parse_param_drop_lowerdir(struct ovl_fs_context *ctx) * Set "/lower1", "/lower2", and "/lower3" as lower layers and * "/data1" and "/data2" as data lower layers. Any existing lower * layers are replaced. - * (2) lowerdir=:/lower4 - * Append "/lower4" to current stack of lower layers. This requires - * that there already is at least one lower layer configured. - * (3) lowerdir=::/lower5 - * Append data "/lower5" as data lower layer. This requires that - * there's at least one regular lower layer present. */ static int ovl_parse_param_lowerdir(const char *name, struct fs_context *fc) { @@ -348,49 +371,9 @@ static int ovl_parse_param_lowerdir(const char *name, struct fs_context *fc) return 0; } - if (strncmp(name, "::", 2) == 0) { - /* - * This is a data layer. - * There must be at least one regular lower layer - * specified. - */ - if (ctx->nr == 0) { - pr_err("data lower layers without regular lower layers not allowed"); - return -EINVAL; - } - - /* Skip the leading "::". */ - name += 2; - data_layer = true; - /* - * A data layer is automatically an append as there - * must've been at least one regular lower layer. - */ - append = true; - } else if (*name == ':') { - /* - * This is a regular lower layer. - * If users want to append a layer enforce that they - * have already specified a first layer before. It's - * better to be strict. - */ - if (ctx->nr == 0) { - pr_err("cannot append layer if no previous layer has been specified"); - return -EINVAL; - } - - /* - * Once a sequence of data layers has started regular - * lower layers are forbidden. - */ - if (ctx->nr_data > 0) { - pr_err("regular lower layers cannot follow data lower layers"); - return -EINVAL; - } - - /* Skip the leading ":". */ - name++; - append = true; + if (*name == ':') { + pr_err("cannot append lower layer"); + return -EINVAL; } dup = kstrdup(name, GFP_KERNEL); @@ -472,7 +455,7 @@ static int ovl_parse_param_lowerdir(const char *name, struct fs_context *fc) l = &ctx->lower[nr]; memset(l, 0, sizeof(*l)); - err = ovl_mount_dir_noesc(dup_iter, &l->path); + err = ovl_mount_dir(dup_iter, &l->path, false); if (err) goto out_put; @@ -682,6 +665,7 @@ static int ovl_reconfigure(struct fs_context *fc) } static const struct fs_context_operations ovl_context_ops = { + .parse_monolithic = ovl_parse_monolithic, .parse_param = ovl_parse_param, .get_tree = ovl_get_tree, .reconfigure = ovl_reconfigure, @@ -752,12 +736,12 @@ void ovl_free_fs(struct ovl_fs *ofs) if (ofs->upperdir_locked) ovl_inuse_unlock(ovl_upper_mnt(ofs)->mnt_root); - /* Hack! Reuse ofs->layers as a vfsmount array before freeing it */ - mounts = (struct vfsmount **) ofs->layers; + /* Reuse ofs->config.lowerdirs as a vfsmount array before freeing it */ + mounts = (struct vfsmount **) ofs->config.lowerdirs; for (i = 0; i < ofs->numlayer; i++) { iput(ofs->layers[i].trap); + kfree(ofs->config.lowerdirs[i]); mounts[i] = ofs->layers[i].mnt; - kfree(ofs->layers[i].name); } kern_unmount_array(mounts, ofs->numlayer); kfree(ofs->layers); @@ -765,6 +749,7 @@ void ovl_free_fs(struct ovl_fs *ofs) free_anon_bdev(ofs->fs[i].pseudo_dev); kfree(ofs->fs); + kfree(ofs->config.lowerdirs); kfree(ofs->config.upperdir); kfree(ofs->config.workdir); if (ofs->creator_cred) @@ -949,16 +934,23 @@ int ovl_show_options(struct seq_file *m, struct dentry *dentry) struct super_block *sb = dentry->d_sb; struct ovl_fs *ofs = OVL_FS(sb); size_t nr, nr_merged_lower = ofs->numlayer - ofs->numdatalayer; - const struct ovl_layer *data_layers = &ofs->layers[nr_merged_lower]; - - /* ofs->layers[0] is the upper layer */ - seq_printf(m, ",lowerdir=%s", ofs->layers[1].name); - /* dump regular lower layers */ - for (nr = 2; nr < nr_merged_lower; nr++) - seq_printf(m, ":%s", ofs->layers[nr].name); - /* dump data lower layers */ - for (nr = 0; nr < ofs->numdatalayer; nr++) - seq_printf(m, "::%s", data_layers[nr].name); + + /* + * lowerdirs[] starts from offset 1, then + * >= 0 regular lower layers prefixed with : and + * >= 0 data-only lower layers prefixed with :: + * + * we need to escase comma and space like seq_show_option() does and + * we also need to escape the colon separator from lowerdir paths. + */ + seq_puts(m, ",lowerdir="); + for (nr = 1; nr < ofs->numlayer; nr++) { + if (nr > 1) + seq_putc(m, ':'); + if (nr >= nr_merged_lower) + seq_putc(m, ':'); + seq_escape(m, ofs->config.lowerdirs[nr], ":, \t\n\\"); + } if (ofs->config.upperdir) { seq_show_option(m, "upperdir", ofs->config.upperdir); seq_show_option(m, "workdir", ofs->config.workdir); diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index def266b5e2a3..3fa2416264a4 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -104,8 +104,8 @@ static int ovl_revalidate_real(struct dentry *d, unsigned int flags, bool weak) static int ovl_dentry_revalidate_common(struct dentry *dentry, unsigned int flags, bool weak) { - struct ovl_entry *oe = OVL_E(dentry); - struct ovl_path *lowerstack = ovl_lowerstack(oe); + struct ovl_entry *oe; + struct ovl_path *lowerstack; struct inode *inode = d_inode_rcu(dentry); struct dentry *upper; unsigned int i; @@ -115,6 +115,8 @@ static int ovl_dentry_revalidate_common(struct dentry *dentry, if (!inode) return -ECHILD; + oe = OVL_I_E(inode); + lowerstack = ovl_lowerstack(oe); upper = ovl_i_dentry_upper(inode); if (upper) ret = ovl_revalidate_real(upper, flags, weak); @@ -167,6 +169,7 @@ static void ovl_free_inode(struct inode *inode) struct ovl_inode *oi = OVL_I(inode); kfree(oi->redirect); + kfree(oi->oe); mutex_destroy(&oi->lock); kmem_cache_free(ovl_inode_cachep, oi); } @@ -176,7 +179,7 @@ static void ovl_destroy_inode(struct inode *inode) struct ovl_inode *oi = OVL_I(inode); dput(oi->__upperdentry); - ovl_free_entry(oi->oe); + ovl_stack_put(ovl_lowerstack(oi->oe), ovl_numlower(oi->oe)); if (S_ISDIR(inode->i_mode)) ovl_dir_cache_free(inode); else @@ -569,11 +572,6 @@ static int ovl_get_upper(struct super_block *sb, struct ovl_fs *ofs, upper_layer->idx = 0; upper_layer->fsid = 0; - err = -ENOMEM; - upper_layer->name = kstrdup(ofs->config.upperdir, GFP_KERNEL); - if (!upper_layer->name) - goto out; - /* * Inherit SB_NOSEC flag from upperdir. * @@ -1122,7 +1120,8 @@ static int ovl_get_layers(struct super_block *sb, struct ovl_fs *ofs, layers[ofs->numlayer].idx = ofs->numlayer; layers[ofs->numlayer].fsid = fsid; layers[ofs->numlayer].fs = &ofs->fs[fsid]; - layers[ofs->numlayer].name = l->name; + /* Store for printing lowerdir=... in ovl_show_options() */ + ofs->config.lowerdirs[ofs->numlayer] = l->name; l->name = NULL; ofs->numlayer++; ofs->fs[fsid].is_lower = true; @@ -1367,8 +1366,16 @@ int ovl_fill_super(struct super_block *sb, struct fs_context *fc) if (!layers) goto out_err; + ofs->config.lowerdirs = kcalloc(ctx->nr + 1, sizeof(char *), GFP_KERNEL); + if (!ofs->config.lowerdirs) { + kfree(layers); + goto out_err; + } ofs->layers = layers; - /* Layer 0 is reserved for upper even if there's no upper */ + /* + * Layer 0 is reserved for upper even if there's no upper. + * For consistency, config.lowerdirs[0] is NULL. + */ ofs->numlayer = 1; sb->s_stack_depth = 0; diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 9e72bfe8bbad..31e897ad5e6a 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -233,19 +233,18 @@ static void put_quota_format(struct quota_format_type *fmt) * All dquots are placed to the end of inuse_list when first created, and this * list is used for invalidate operation, which must look at every dquot. * - * When the last reference of a dquot will be dropped, the dquot will be - * added to releasing_dquots. We'd then queue work item which would call + * When the last reference of a dquot is dropped, the dquot is added to + * releasing_dquots. We'll then queue work item which will call * synchronize_srcu() and after that perform the final cleanup of all the - * dquots on the list. Both releasing_dquots and free_dquots use the - * dq_free list_head in the dquot struct. When a dquot is removed from - * releasing_dquots, a reference count is always subtracted, and if - * dq_count == 0 at that point, the dquot will be added to the free_dquots. + * dquots on the list. Each cleaned up dquot is moved to free_dquots list. + * Both releasing_dquots and free_dquots use the dq_free list_head in the dquot + * struct. * - * Unused dquots (dq_count == 0) are added to the free_dquots list when freed, - * and this list is searched whenever we need an available dquot. Dquots are - * removed from the list as soon as they are used again, and - * dqstats.free_dquots gives the number of dquots on the list. When - * dquot is invalidated it's completely released from memory. + * Unused and cleaned up dquots are in the free_dquots list and this list is + * searched whenever we need an available dquot. Dquots are removed from the + * list as soon as they are used again and dqstats.free_dquots gives the number + * of dquots on the list. When dquot is invalidated it's completely released + * from memory. * * Dirty dquots are added to the dqi_dirty_list of quota_info when mark * dirtied, and this list is searched when writing dirty dquots back to @@ -321,6 +320,7 @@ static inline void put_dquot_last(struct dquot *dquot) static inline void put_releasing_dquots(struct dquot *dquot) { list_add_tail(&dquot->dq_free, &releasing_dquots); + set_bit(DQ_RELEASING_B, &dquot->dq_flags); } static inline void remove_free_dquot(struct dquot *dquot) @@ -328,8 +328,10 @@ static inline void remove_free_dquot(struct dquot *dquot) if (list_empty(&dquot->dq_free)) return; list_del_init(&dquot->dq_free); - if (!atomic_read(&dquot->dq_count)) + if (!test_bit(DQ_RELEASING_B, &dquot->dq_flags)) dqstats_dec(DQST_FREE_DQUOTS); + else + clear_bit(DQ_RELEASING_B, &dquot->dq_flags); } static inline void put_inuse(struct dquot *dquot) @@ -581,12 +583,6 @@ restart: continue; /* Wait for dquot users */ if (atomic_read(&dquot->dq_count)) { - /* dquot in releasing_dquots, flush and retry */ - if (!list_empty(&dquot->dq_free)) { - spin_unlock(&dq_list_lock); - goto restart; - } - atomic_inc(&dquot->dq_count); spin_unlock(&dq_list_lock); /* @@ -606,6 +602,15 @@ restart: goto restart; } /* + * The last user already dropped its reference but dquot didn't + * get fully cleaned up yet. Restart the scan which flushes the + * work cleaning up released dquots. + */ + if (test_bit(DQ_RELEASING_B, &dquot->dq_flags)) { + spin_unlock(&dq_list_lock); + goto restart; + } + /* * Quota now has no users and it has been written on last * dqput() */ @@ -696,6 +701,13 @@ int dquot_writeback_dquots(struct super_block *sb, int type) dq_dirty); WARN_ON(!dquot_active(dquot)); + /* If the dquot is releasing we should not touch it */ + if (test_bit(DQ_RELEASING_B, &dquot->dq_flags)) { + spin_unlock(&dq_list_lock); + flush_delayed_work("a_release_work); + spin_lock(&dq_list_lock); + continue; + } /* Now we have active dquot from which someone is * holding reference so we can safely just increase @@ -809,18 +821,18 @@ static void quota_release_workfn(struct work_struct *work) /* Exchange the list head to avoid livelock. */ list_replace_init(&releasing_dquots, &rls_head); spin_unlock(&dq_list_lock); + synchronize_srcu(&dquot_srcu); restart: - synchronize_srcu(&dquot_srcu); spin_lock(&dq_list_lock); while (!list_empty(&rls_head)) { dquot = list_first_entry(&rls_head, struct dquot, dq_free); - /* Dquot got used again? */ - if (atomic_read(&dquot->dq_count) > 1) { - remove_free_dquot(dquot); - atomic_dec(&dquot->dq_count); - continue; - } + WARN_ON_ONCE(atomic_read(&dquot->dq_count)); + /* + * Note that DQ_RELEASING_B protects us from racing with + * invalidate_dquots() calls so we are safe to work with the + * dquot even after we drop dq_list_lock. + */ if (dquot_dirty(dquot)) { spin_unlock(&dq_list_lock); /* Commit dquot before releasing */ @@ -834,7 +846,6 @@ restart: } /* Dquot is inactive and clean, now move it to free list */ remove_free_dquot(dquot); - atomic_dec(&dquot->dq_count); put_dquot_last(dquot); } spin_unlock(&dq_list_lock); @@ -875,6 +886,7 @@ void dqput(struct dquot *dquot) BUG_ON(!list_empty(&dquot->dq_free)); #endif put_releasing_dquots(dquot); + atomic_dec(&dquot->dq_count); spin_unlock(&dq_list_lock); queue_delayed_work(system_unbound_wq, "a_release_work, 1); } @@ -963,7 +975,7 @@ we_slept: dqstats_inc(DQST_LOOKUPS); } /* Wait for dq_lock - after this we know that either dquot_release() is - * already finished or it will be canceled due to dq_count > 1 test */ + * already finished or it will be canceled due to dq_count > 0 test */ wait_on_dquot(dquot); /* Read the dquot / allocate space in quota file */ if (!dquot_active(dquot)) { diff --git a/fs/smb/client/cached_dir.c b/fs/smb/client/cached_dir.c index e2be8aedb26e..fe1bf5b6e0cb 100644 --- a/fs/smb/client/cached_dir.c +++ b/fs/smb/client/cached_dir.c @@ -15,6 +15,7 @@ static struct cached_fid *init_cached_dir(const char *path); static void free_cached_dir(struct cached_fid *cfid); static void smb2_close_cached_fid(struct kref *ref); +static void cfids_laundromat_worker(struct work_struct *work); static struct cached_fid *find_or_create_cached_dir(struct cached_fids *cfids, const char *path, @@ -169,15 +170,18 @@ int open_cached_dir(unsigned int xid, struct cifs_tcon *tcon, return -ENOENT; } /* - * At this point we either have a lease already and we can just - * return it. If not we are guaranteed to be the only thread accessing - * this cfid. + * Return cached fid if it has a lease. Otherwise, it is either a new + * entry or laundromat worker removed it from @cfids->entries. Caller + * will put last reference if the latter. */ + spin_lock(&cfids->cfid_list_lock); if (cfid->has_lease) { + spin_unlock(&cfids->cfid_list_lock); *ret_cfid = cfid; kfree(utf16_path); return 0; } + spin_unlock(&cfids->cfid_list_lock); /* * Skip any prefix paths in @path as lookup_positive_unlocked() ends up @@ -294,9 +298,11 @@ int open_cached_dir(unsigned int xid, struct cifs_tcon *tcon, goto oshr_free; } } + spin_lock(&cfids->cfid_list_lock); cfid->dentry = dentry; cfid->time = jiffies; cfid->has_lease = true; + spin_unlock(&cfids->cfid_list_lock); oshr_free: kfree(utf16_path); @@ -305,24 +311,28 @@ oshr_free: free_rsp_buf(resp_buftype[0], rsp_iov[0].iov_base); free_rsp_buf(resp_buftype[1], rsp_iov[1].iov_base); spin_lock(&cfids->cfid_list_lock); - if (rc && !cfid->has_lease) { - if (cfid->on_list) { - list_del(&cfid->entry); - cfid->on_list = false; - cfids->num_entries--; + if (!cfid->has_lease) { + if (rc) { + if (cfid->on_list) { + list_del(&cfid->entry); + cfid->on_list = false; + cfids->num_entries--; + } + rc = -ENOENT; + } else { + /* + * We are guaranteed to have two references at this + * point. One for the caller and one for a potential + * lease. Release the Lease-ref so that the directory + * will be closed when the caller closes the cached + * handle. + */ + spin_unlock(&cfids->cfid_list_lock); + kref_put(&cfid->refcount, smb2_close_cached_fid); + goto out; } - rc = -ENOENT; } spin_unlock(&cfids->cfid_list_lock); - if (!rc && !cfid->has_lease) { - /* - * We are guaranteed to have two references at this point. - * One for the caller and one for a potential lease. - * Release the Lease-ref so that the directory will be closed - * when the caller closes the cached handle. - */ - kref_put(&cfid->refcount, smb2_close_cached_fid); - } if (rc) { if (cfid->is_open) SMB2_close(0, cfid->tcon, cfid->fid.persistent_fid, @@ -330,7 +340,7 @@ oshr_free: free_cached_dir(cfid); cfid = NULL; } - +out: if (rc == 0) { *ret_cfid = cfid; atomic_inc(&tcon->num_remote_opens); @@ -572,53 +582,51 @@ static void free_cached_dir(struct cached_fid *cfid) kfree(cfid); } -static int -cifs_cfids_laundromat_thread(void *p) +static void cfids_laundromat_worker(struct work_struct *work) { - struct cached_fids *cfids = p; + struct cached_fids *cfids; struct cached_fid *cfid, *q; - struct list_head entry; + LIST_HEAD(entry); - while (!kthread_should_stop()) { - ssleep(1); - INIT_LIST_HEAD(&entry); - if (kthread_should_stop()) - return 0; - spin_lock(&cfids->cfid_list_lock); - list_for_each_entry_safe(cfid, q, &cfids->entries, entry) { - if (time_after(jiffies, cfid->time + HZ * dir_cache_timeout)) { - list_del(&cfid->entry); - list_add(&cfid->entry, &entry); - cfids->num_entries--; - } - } - spin_unlock(&cfids->cfid_list_lock); + cfids = container_of(work, struct cached_fids, laundromat_work.work); - list_for_each_entry_safe(cfid, q, &entry, entry) { + spin_lock(&cfids->cfid_list_lock); + list_for_each_entry_safe(cfid, q, &cfids->entries, entry) { + if (cfid->time && + time_after(jiffies, cfid->time + HZ * dir_cache_timeout)) { cfid->on_list = false; - list_del(&cfid->entry); + list_move(&cfid->entry, &entry); + cfids->num_entries--; + /* To prevent race with smb2_cached_lease_break() */ + kref_get(&cfid->refcount); + } + } + spin_unlock(&cfids->cfid_list_lock); + + list_for_each_entry_safe(cfid, q, &entry, entry) { + list_del(&cfid->entry); + /* + * Cancel and wait for the work to finish in case we are racing + * with it. + */ + cancel_work_sync(&cfid->lease_break); + if (cfid->has_lease) { /* - * Cancel, and wait for the work to finish in - * case we are racing with it. + * Our lease has not yet been cancelled from the server + * so we need to drop the reference. */ - cancel_work_sync(&cfid->lease_break); - if (cfid->has_lease) { - /* - * We lease has not yet been cancelled from - * the server so we need to drop the reference. - */ - spin_lock(&cfids->cfid_list_lock); - cfid->has_lease = false; - spin_unlock(&cfids->cfid_list_lock); - kref_put(&cfid->refcount, smb2_close_cached_fid); - } + spin_lock(&cfids->cfid_list_lock); + cfid->has_lease = false; + spin_unlock(&cfids->cfid_list_lock); + kref_put(&cfid->refcount, smb2_close_cached_fid); } + /* Drop the extra reference opened above */ + kref_put(&cfid->refcount, smb2_close_cached_fid); } - - return 0; + queue_delayed_work(cifsiod_wq, &cfids->laundromat_work, + dir_cache_timeout * HZ); } - struct cached_fids *init_cached_dirs(void) { struct cached_fids *cfids; @@ -629,19 +637,10 @@ struct cached_fids *init_cached_dirs(void) spin_lock_init(&cfids->cfid_list_lock); INIT_LIST_HEAD(&cfids->entries); - /* - * since we're in a cifs function already, we know that - * this will succeed. No need for try_module_get(). - */ - __module_get(THIS_MODULE); - cfids->laundromat = kthread_run(cifs_cfids_laundromat_thread, - cfids, "cifsd-cfid-laundromat"); - if (IS_ERR(cfids->laundromat)) { - cifs_dbg(VFS, "Failed to start cfids laundromat thread.\n"); - kfree(cfids); - module_put(THIS_MODULE); - return NULL; - } + INIT_DELAYED_WORK(&cfids->laundromat_work, cfids_laundromat_worker); + queue_delayed_work(cifsiod_wq, &cfids->laundromat_work, + dir_cache_timeout * HZ); + return cfids; } @@ -657,11 +656,7 @@ void free_cached_dirs(struct cached_fids *cfids) if (cfids == NULL) return; - if (cfids->laundromat) { - kthread_stop(cfids->laundromat); - cfids->laundromat = NULL; - module_put(THIS_MODULE); - } + cancel_delayed_work_sync(&cfids->laundromat_work); spin_lock(&cfids->cfid_list_lock); list_for_each_entry_safe(cfid, q, &cfids->entries, entry) { diff --git a/fs/smb/client/cached_dir.h b/fs/smb/client/cached_dir.h index a82ff2cea789..81ba0fd5cc16 100644 --- a/fs/smb/client/cached_dir.h +++ b/fs/smb/client/cached_dir.h @@ -57,7 +57,7 @@ struct cached_fids { spinlock_t cfid_list_lock; int num_entries; struct list_head entries; - struct task_struct *laundromat; + struct delayed_work laundromat_work; }; extern struct cached_fids *init_cached_dirs(void); diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 3902e90dca6b..7b923e36501b 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -2474,8 +2474,9 @@ cifs_put_tcon(struct cifs_tcon *tcon) static struct cifs_tcon * cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) { - int rc, xid; struct cifs_tcon *tcon; + bool nohandlecache; + int rc, xid; tcon = cifs_find_tcon(ses, ctx); if (tcon) { @@ -2493,14 +2494,17 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) goto out_fail; } - if (ses->server->capabilities & SMB2_GLOBAL_CAP_DIRECTORY_LEASING) - tcon = tcon_info_alloc(true); + if (ses->server->dialect >= SMB20_PROT_ID && + (ses->server->capabilities & SMB2_GLOBAL_CAP_DIRECTORY_LEASING)) + nohandlecache = ctx->nohandlecache; else - tcon = tcon_info_alloc(false); + nohandlecache = true; + tcon = tcon_info_alloc(!nohandlecache); if (tcon == NULL) { rc = -ENOMEM; goto out_fail; } + tcon->nohandlecache = nohandlecache; if (ctx->snapshot_time) { if (ses->server->vals->protocol_id == 0) { @@ -2662,10 +2666,6 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) tcon->nocase = ctx->nocase; tcon->broken_sparse_sup = ctx->no_sparse; tcon->max_cached_dirs = ctx->max_cached_dirs; - if (ses->server->capabilities & SMB2_GLOBAL_CAP_DIRECTORY_LEASING) - tcon->nohandlecache = ctx->nohandlecache; - else - tcon->nohandlecache = true; tcon->nodelete = ctx->nodelete; tcon->local_lease = ctx->local_lease; INIT_LIST_HEAD(&tcon->pending_opens); @@ -2895,9 +2895,9 @@ bind_socket(struct TCP_Server_Info *server) if (server->srcaddr.ss_family != AF_UNSPEC) { /* Bind to the specified local IP address */ struct socket *socket = server->ssocket; - rc = socket->ops->bind(socket, - (struct sockaddr *) &server->srcaddr, - sizeof(server->srcaddr)); + rc = kernel_bind(socket, + (struct sockaddr *) &server->srcaddr, + sizeof(server->srcaddr)); if (rc < 0) { struct sockaddr_in *saddr4; struct sockaddr_in6 *saddr6; @@ -3046,8 +3046,8 @@ generic_ip_connect(struct TCP_Server_Info *server) socket->sk->sk_sndbuf, socket->sk->sk_rcvbuf, socket->sk->sk_rcvtimeo); - rc = socket->ops->connect(socket, saddr, slen, - server->noblockcnt ? O_NONBLOCK : 0); + rc = kernel_connect(socket, saddr, slen, + server->noblockcnt ? O_NONBLOCK : 0); /* * When mounting SMB root file systems, we do not want to block in * connect. Otherwise bail out and then let cifs_reconnect() perform diff --git a/fs/smb/server/connection.c b/fs/smb/server/connection.c index db7fa704a3f6..4b38c3a285f6 100644 --- a/fs/smb/server/connection.c +++ b/fs/smb/server/connection.c @@ -84,6 +84,8 @@ struct ksmbd_conn *ksmbd_conn_alloc(void) spin_lock_init(&conn->llist_lock); INIT_LIST_HEAD(&conn->lock_list); + init_rwsem(&conn->session_lock); + down_write(&conn_list_lock); list_add(&conn->conns_list, &conn_list); up_write(&conn_list_lock); diff --git a/fs/smb/server/connection.h b/fs/smb/server/connection.h index ab2583f030ce..3c005246a32e 100644 --- a/fs/smb/server/connection.h +++ b/fs/smb/server/connection.h @@ -50,6 +50,7 @@ struct ksmbd_conn { struct nls_table *local_nls; struct unicode_map *um; struct list_head conns_list; + struct rw_semaphore session_lock; /* smb session 1 per user */ struct xarray sessions; unsigned long last_active; diff --git a/fs/smb/server/mgmt/tree_connect.c b/fs/smb/server/mgmt/tree_connect.c index 408cddf2f094..d2c81a8a11dd 100644 --- a/fs/smb/server/mgmt/tree_connect.c +++ b/fs/smb/server/mgmt/tree_connect.c @@ -73,7 +73,10 @@ ksmbd_tree_conn_connect(struct ksmbd_conn *conn, struct ksmbd_session *sess, tree_conn->user = sess->user; tree_conn->share_conf = sc; + tree_conn->t_state = TREE_NEW; status.tree_conn = tree_conn; + atomic_set(&tree_conn->refcount, 1); + init_waitqueue_head(&tree_conn->refcount_q); ret = xa_err(xa_store(&sess->tree_conns, tree_conn->id, tree_conn, GFP_KERNEL)); @@ -93,14 +96,33 @@ out_error: return status; } +void ksmbd_tree_connect_put(struct ksmbd_tree_connect *tcon) +{ + /* + * Checking waitqueue to releasing tree connect on + * tree disconnect. waitqueue_active is safe because it + * uses atomic operation for condition. + */ + if (!atomic_dec_return(&tcon->refcount) && + waitqueue_active(&tcon->refcount_q)) + wake_up(&tcon->refcount_q); +} + int ksmbd_tree_conn_disconnect(struct ksmbd_session *sess, struct ksmbd_tree_connect *tree_conn) { int ret; + write_lock(&sess->tree_conns_lock); + xa_erase(&sess->tree_conns, tree_conn->id); + write_unlock(&sess->tree_conns_lock); + + if (!atomic_dec_and_test(&tree_conn->refcount)) + wait_event(tree_conn->refcount_q, + atomic_read(&tree_conn->refcount) == 0); + ret = ksmbd_ipc_tree_disconnect_request(sess->id, tree_conn->id); ksmbd_release_tree_conn_id(sess, tree_conn->id); - xa_erase(&sess->tree_conns, tree_conn->id); ksmbd_share_config_put(tree_conn->share_conf); kfree(tree_conn); return ret; @@ -111,11 +133,15 @@ struct ksmbd_tree_connect *ksmbd_tree_conn_lookup(struct ksmbd_session *sess, { struct ksmbd_tree_connect *tcon; + read_lock(&sess->tree_conns_lock); tcon = xa_load(&sess->tree_conns, id); if (tcon) { - if (test_bit(TREE_CONN_EXPIRE, &tcon->status)) + if (tcon->t_state != TREE_CONNECTED) + tcon = NULL; + else if (!atomic_inc_not_zero(&tcon->refcount)) tcon = NULL; } + read_unlock(&sess->tree_conns_lock); return tcon; } @@ -129,8 +155,18 @@ int ksmbd_tree_conn_session_logoff(struct ksmbd_session *sess) if (!sess) return -EINVAL; - xa_for_each(&sess->tree_conns, id, tc) + xa_for_each(&sess->tree_conns, id, tc) { + write_lock(&sess->tree_conns_lock); + if (tc->t_state == TREE_DISCONNECTED) { + write_unlock(&sess->tree_conns_lock); + ret = -ENOENT; + continue; + } + tc->t_state = TREE_DISCONNECTED; + write_unlock(&sess->tree_conns_lock); + ret |= ksmbd_tree_conn_disconnect(sess, tc); + } xa_destroy(&sess->tree_conns); return ret; } diff --git a/fs/smb/server/mgmt/tree_connect.h b/fs/smb/server/mgmt/tree_connect.h index 562d647ad9fa..6377a70b811c 100644 --- a/fs/smb/server/mgmt/tree_connect.h +++ b/fs/smb/server/mgmt/tree_connect.h @@ -14,7 +14,11 @@ struct ksmbd_share_config; struct ksmbd_user; struct ksmbd_conn; -#define TREE_CONN_EXPIRE 1 +enum { + TREE_NEW = 0, + TREE_CONNECTED, + TREE_DISCONNECTED +}; struct ksmbd_tree_connect { int id; @@ -27,7 +31,9 @@ struct ksmbd_tree_connect { int maximal_access; bool posix_extensions; - unsigned long status; + atomic_t refcount; + wait_queue_head_t refcount_q; + unsigned int t_state; }; struct ksmbd_tree_conn_status { @@ -46,6 +52,7 @@ struct ksmbd_session; struct ksmbd_tree_conn_status ksmbd_tree_conn_connect(struct ksmbd_conn *conn, struct ksmbd_session *sess, const char *share_name); +void ksmbd_tree_connect_put(struct ksmbd_tree_connect *tcon); int ksmbd_tree_conn_disconnect(struct ksmbd_session *sess, struct ksmbd_tree_connect *tree_conn); diff --git a/fs/smb/server/mgmt/user_session.c b/fs/smb/server/mgmt/user_session.c index 8a5dcab05614..15f68ee05089 100644 --- a/fs/smb/server/mgmt/user_session.c +++ b/fs/smb/server/mgmt/user_session.c @@ -174,7 +174,7 @@ static void ksmbd_expire_session(struct ksmbd_conn *conn) unsigned long id; struct ksmbd_session *sess; - down_write(&sessions_table_lock); + down_write(&conn->session_lock); xa_for_each(&conn->sessions, id, sess) { if (sess->state != SMB2_SESSION_VALID || time_after(jiffies, @@ -185,7 +185,7 @@ static void ksmbd_expire_session(struct ksmbd_conn *conn) continue; } } - up_write(&sessions_table_lock); + up_write(&conn->session_lock); } int ksmbd_session_register(struct ksmbd_conn *conn, @@ -227,7 +227,9 @@ void ksmbd_sessions_deregister(struct ksmbd_conn *conn) } } } + up_write(&sessions_table_lock); + down_write(&conn->session_lock); xa_for_each(&conn->sessions, id, sess) { unsigned long chann_id; struct channel *chann; @@ -244,7 +246,7 @@ void ksmbd_sessions_deregister(struct ksmbd_conn *conn) ksmbd_session_destroy(sess); } } - up_write(&sessions_table_lock); + up_write(&conn->session_lock); } struct ksmbd_session *ksmbd_session_lookup(struct ksmbd_conn *conn, @@ -252,9 +254,11 @@ struct ksmbd_session *ksmbd_session_lookup(struct ksmbd_conn *conn, { struct ksmbd_session *sess; + down_read(&conn->session_lock); sess = xa_load(&conn->sessions, id); if (sess) sess->last_active = jiffies; + up_read(&conn->session_lock); return sess; } @@ -351,6 +355,7 @@ static struct ksmbd_session *__session_create(int protocol) xa_init(&sess->ksmbd_chann_list); xa_init(&sess->rpc_handle_list); sess->sequence_number = 1; + rwlock_init(&sess->tree_conns_lock); ret = __init_smb2_session(sess); if (ret) diff --git a/fs/smb/server/mgmt/user_session.h b/fs/smb/server/mgmt/user_session.h index f99d475b28db..63cb08fffde8 100644 --- a/fs/smb/server/mgmt/user_session.h +++ b/fs/smb/server/mgmt/user_session.h @@ -60,6 +60,7 @@ struct ksmbd_session { struct ksmbd_file_table file_table; unsigned long last_active; + rwlock_t tree_conns_lock; }; static inline int test_session_flag(struct ksmbd_session *sess, int bit) diff --git a/fs/smb/server/server.c b/fs/smb/server/server.c index 32347fec33c4..3079e607c5fe 100644 --- a/fs/smb/server/server.c +++ b/fs/smb/server/server.c @@ -241,6 +241,8 @@ static void __handle_ksmbd_work(struct ksmbd_work *work, } while (is_chained == true); send: + if (work->tcon) + ksmbd_tree_connect_put(work->tcon); smb3_preauth_hash_rsp(work); if (work->sess && work->sess->enc && work->encrypted && conn->ops->encrypt_resp) { diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 544022dd6d20..93262ca3f58a 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -231,11 +231,12 @@ void set_smb2_rsp_status(struct ksmbd_work *work, __le32 err) { struct smb2_hdr *rsp_hdr; - if (work->next_smb2_rcv_hdr_off) - rsp_hdr = ksmbd_resp_buf_next(work); - else - rsp_hdr = smb2_get_msg(work->response_buf); + rsp_hdr = smb2_get_msg(work->response_buf); rsp_hdr->Status = err; + + work->iov_idx = 0; + work->iov_cnt = 0; + work->next_smb2_rcv_hdr_off = 0; smb2_set_err_rsp(work); } @@ -1993,6 +1994,9 @@ int smb2_tree_connect(struct ksmbd_work *work) if (conn->posix_ext_supported) status.tree_conn->posix_extensions = true; + write_lock(&sess->tree_conns_lock); + status.tree_conn->t_state = TREE_CONNECTED; + write_unlock(&sess->tree_conns_lock); rsp->StructureSize = cpu_to_le16(16); out_err1: rsp->Capabilities = 0; @@ -2122,27 +2126,50 @@ int smb2_tree_disconnect(struct ksmbd_work *work) ksmbd_debug(SMB, "request\n"); + if (!tcon) { + ksmbd_debug(SMB, "Invalid tid %d\n", req->hdr.Id.SyncId.TreeId); + + rsp->hdr.Status = STATUS_NETWORK_NAME_DELETED; + err = -ENOENT; + goto err_out; + } + + ksmbd_close_tree_conn_fds(work); + + write_lock(&sess->tree_conns_lock); + if (tcon->t_state == TREE_DISCONNECTED) { + write_unlock(&sess->tree_conns_lock); + rsp->hdr.Status = STATUS_NETWORK_NAME_DELETED; + err = -ENOENT; + goto err_out; + } + + WARN_ON_ONCE(atomic_dec_and_test(&tcon->refcount)); + tcon->t_state = TREE_DISCONNECTED; + write_unlock(&sess->tree_conns_lock); + + err = ksmbd_tree_conn_disconnect(sess, tcon); + if (err) { + rsp->hdr.Status = STATUS_NETWORK_NAME_DELETED; + goto err_out; + } + + work->tcon = NULL; + rsp->StructureSize = cpu_to_le16(4); err = ksmbd_iov_pin_rsp(work, rsp, sizeof(struct smb2_tree_disconnect_rsp)); if (err) { rsp->hdr.Status = STATUS_INSUFFICIENT_RESOURCES; - smb2_set_err_rsp(work); - return err; + goto err_out; } - if (!tcon || test_and_set_bit(TREE_CONN_EXPIRE, &tcon->status)) { - ksmbd_debug(SMB, "Invalid tid %d\n", req->hdr.Id.SyncId.TreeId); + return 0; - rsp->hdr.Status = STATUS_NETWORK_NAME_DELETED; - smb2_set_err_rsp(work); - return -ENOENT; - } +err_out: + smb2_set_err_rsp(work); + return err; - ksmbd_close_tree_conn_fds(work); - ksmbd_tree_conn_disconnect(sess, tcon); - work->tcon = NULL; - return 0; } /** @@ -2164,17 +2191,17 @@ int smb2_session_logoff(struct ksmbd_work *work) ksmbd_debug(SMB, "request\n"); - sess_id = le64_to_cpu(req->hdr.SessionId); - - rsp->StructureSize = cpu_to_le16(4); - err = ksmbd_iov_pin_rsp(work, rsp, sizeof(struct smb2_logoff_rsp)); - if (err) { - rsp->hdr.Status = STATUS_INSUFFICIENT_RESOURCES; + ksmbd_conn_lock(conn); + if (!ksmbd_conn_good(conn)) { + ksmbd_conn_unlock(conn); + rsp->hdr.Status = STATUS_NETWORK_NAME_DELETED; smb2_set_err_rsp(work); - return err; + return -ENOENT; } - + sess_id = le64_to_cpu(req->hdr.SessionId); ksmbd_all_conn_set_status(sess_id, KSMBD_SESS_NEED_RECONNECT); + ksmbd_conn_unlock(conn); + ksmbd_close_session_fds(work); ksmbd_conn_wait_idle(conn, sess_id); @@ -2196,6 +2223,14 @@ int smb2_session_logoff(struct ksmbd_work *work) ksmbd_free_user(sess->user); sess->user = NULL; ksmbd_all_conn_set_status(sess_id, KSMBD_SESS_NEED_NEGOTIATE); + + rsp->StructureSize = cpu_to_le16(4); + err = ksmbd_iov_pin_rsp(work, rsp, sizeof(struct smb2_logoff_rsp)); + if (err) { + rsp->hdr.Status = STATUS_INSUFFICIENT_RESOURCES; + smb2_set_err_rsp(work); + return err; + } return 0; } @@ -3370,8 +3405,10 @@ err_out: } ksmbd_revert_fsids(work); err_out1: - if (!rc) + if (!rc) { + ksmbd_update_fstate(&work->sess->file_table, fp, FP_INITED); rc = ksmbd_iov_pin_rsp(work, (void *)rsp, iov_len); + } if (rc) { if (rc == -EINVAL) rsp->hdr.Status = STATUS_INVALID_PARAMETER; @@ -6115,12 +6152,12 @@ static noinline int smb2_read_pipe(struct ksmbd_work *work) memcpy(aux_payload_buf, rpc_resp->payload, rpc_resp->payload_sz); nbytes = rpc_resp->payload_sz; - kvfree(rpc_resp); err = ksmbd_iov_pin_rsp_read(work, (void *)rsp, offsetof(struct smb2_read_rsp, Buffer), aux_payload_buf, nbytes); if (err) goto out; + kvfree(rpc_resp); } else { err = ksmbd_iov_pin_rsp(work, (void *)rsp, offsetof(struct smb2_read_rsp, Buffer)); @@ -7028,10 +7065,6 @@ skip: ksmbd_debug(SMB, "would have to wait for getting lock\n"); - spin_lock(&work->conn->llist_lock); - list_add_tail(&smb_lock->clist, - &work->conn->lock_list); - spin_unlock(&work->conn->llist_lock); list_add(&smb_lock->llist, &rollback_list); argv = kmalloc(sizeof(void *), GFP_KERNEL); @@ -7062,9 +7095,6 @@ skip: if (work->state != KSMBD_WORK_ACTIVE) { list_del(&smb_lock->llist); - spin_lock(&work->conn->llist_lock); - list_del(&smb_lock->clist); - spin_unlock(&work->conn->llist_lock); locks_free_lock(flock); if (work->state == KSMBD_WORK_CANCELLED) { @@ -7084,19 +7114,16 @@ skip: } list_del(&smb_lock->llist); - spin_lock(&work->conn->llist_lock); - list_del(&smb_lock->clist); - spin_unlock(&work->conn->llist_lock); release_async_work(work); goto retry; } else if (!rc) { + list_add(&smb_lock->llist, &rollback_list); spin_lock(&work->conn->llist_lock); list_add_tail(&smb_lock->clist, &work->conn->lock_list); list_add_tail(&smb_lock->flist, &fp->lock_list); spin_unlock(&work->conn->llist_lock); - list_add(&smb_lock->llist, &rollback_list); ksmbd_debug(SMB, "successful in taking lock\n"); } else { goto out; @@ -8036,10 +8063,10 @@ static void smb20_oplock_break_ack(struct ksmbd_work *work) goto err_out; } - opinfo_put(opinfo); - ksmbd_fd_put(work, fp); opinfo->op_state = OPLOCK_STATE_NONE; wake_up_interruptible_all(&opinfo->oplock_q); + opinfo_put(opinfo); + ksmbd_fd_put(work, fp); rsp->StructureSize = cpu_to_le16(24); rsp->OplockLevel = rsp_oplevel; diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c index f41f8d6108ce..c91eac6514dd 100644 --- a/fs/smb/server/vfs_cache.c +++ b/fs/smb/server/vfs_cache.c @@ -106,7 +106,7 @@ int ksmbd_query_inode_status(struct inode *inode) ci = __ksmbd_inode_lookup(inode); if (ci) { ret = KSMBD_INODE_STATUS_OK; - if (ci->m_flags & S_DEL_PENDING) + if (ci->m_flags & (S_DEL_PENDING | S_DEL_ON_CLS)) ret = KSMBD_INODE_STATUS_PENDING_DELETE; atomic_dec(&ci->m_count); } @@ -116,7 +116,7 @@ int ksmbd_query_inode_status(struct inode *inode) bool ksmbd_inode_pending_delete(struct ksmbd_file *fp) { - return (fp->f_ci->m_flags & S_DEL_PENDING); + return (fp->f_ci->m_flags & (S_DEL_PENDING | S_DEL_ON_CLS)); } void ksmbd_set_inode_pending_delete(struct ksmbd_file *fp) @@ -333,6 +333,9 @@ static void __ksmbd_close_fd(struct ksmbd_file_table *ft, struct ksmbd_file *fp) static struct ksmbd_file *ksmbd_fp_get(struct ksmbd_file *fp) { + if (fp->f_state != FP_INITED) + return NULL; + if (!atomic_inc_not_zero(&fp->refcount)) return NULL; return fp; @@ -382,15 +385,20 @@ int ksmbd_close_fd(struct ksmbd_work *work, u64 id) return 0; ft = &work->sess->file_table; - read_lock(&ft->lock); + write_lock(&ft->lock); fp = idr_find(ft->idr, id); if (fp) { set_close_state_blocked_works(fp); - if (!atomic_dec_and_test(&fp->refcount)) + if (fp->f_state != FP_INITED) fp = NULL; + else { + fp->f_state = FP_CLOSED; + if (!atomic_dec_and_test(&fp->refcount)) + fp = NULL; + } } - read_unlock(&ft->lock); + write_unlock(&ft->lock); if (!fp) return -EINVAL; @@ -570,6 +578,7 @@ struct ksmbd_file *ksmbd_open_fd(struct ksmbd_work *work, struct file *filp) fp->tcon = work->tcon; fp->volatile_id = KSMBD_NO_FID; fp->persistent_id = KSMBD_NO_FID; + fp->f_state = FP_NEW; fp->f_ci = ksmbd_inode_get(fp); if (!fp->f_ci) { @@ -591,6 +600,17 @@ err_out: return ERR_PTR(ret); } +void ksmbd_update_fstate(struct ksmbd_file_table *ft, struct ksmbd_file *fp, + unsigned int state) +{ + if (!fp) + return; + + write_lock(&ft->lock); + fp->f_state = state; + write_unlock(&ft->lock); +} + static int __close_file_table_ids(struct ksmbd_file_table *ft, struct ksmbd_tree_connect *tcon, diff --git a/fs/smb/server/vfs_cache.h b/fs/smb/server/vfs_cache.h index fcb13413fa8d..03d0bf941216 100644 --- a/fs/smb/server/vfs_cache.h +++ b/fs/smb/server/vfs_cache.h @@ -60,6 +60,12 @@ struct ksmbd_inode { __le32 m_fattr; }; +enum { + FP_NEW = 0, + FP_INITED, + FP_CLOSED +}; + struct ksmbd_file { struct file *filp; u64 persistent_id; @@ -98,6 +104,7 @@ struct ksmbd_file { /* if ls is happening on directory, below is valid*/ struct ksmbd_readdir_data readdir_data; int dot_dotdot[2]; + unsigned int f_state; }; static inline void set_ctx_actor(struct dir_context *ctx, @@ -142,6 +149,8 @@ int ksmbd_close_inode_fds(struct ksmbd_work *work, struct inode *inode); int ksmbd_init_global_file_table(void); void ksmbd_free_global_file_table(void); void ksmbd_set_fd_limit(unsigned long limit); +void ksmbd_update_fstate(struct ksmbd_file_table *ft, struct ksmbd_file *fp, + unsigned int state); /* * INODE hash diff --git a/fs/xfs/libxfs/xfs_ag.c b/fs/xfs/libxfs/xfs_ag.c index e9cc481b4ddf..f9f4d694640d 100644 --- a/fs/xfs/libxfs/xfs_ag.c +++ b/fs/xfs/libxfs/xfs_ag.c @@ -1001,6 +1001,12 @@ xfs_ag_shrink_space( error = -ENOSPC; goto resv_init_out; } + + /* Update perag geometry */ + pag->block_count -= delta; + __xfs_agino_range(pag->pag_mount, pag->block_count, &pag->agino_min, + &pag->agino_max); + xfs_ialloc_log_agi(*tpp, agibp, XFS_AGI_LENGTH); xfs_alloc_log_agf(*tpp, agfbp, XFS_AGF_LENGTH); return 0; diff --git a/fs/xfs/scrub/xfile.c b/fs/xfs/scrub/xfile.c index d98e8e77c684..090c3ead43fd 100644 --- a/fs/xfs/scrub/xfile.c +++ b/fs/xfs/scrub/xfile.c @@ -10,7 +10,6 @@ #include "xfs_log_format.h" #include "xfs_trans_resv.h" #include "xfs_mount.h" -#include "xfs_format.h" #include "scrub/xfile.h" #include "scrub/xfarray.h" #include "scrub/scrub.h" diff --git a/fs/xfs/xfs_discard.c b/fs/xfs/xfs_discard.c index afc4c78b9eed..d5787991bb5b 100644 --- a/fs/xfs/xfs_discard.c +++ b/fs/xfs/xfs_discard.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 /* - * Copyright (C) 2010 Red Hat, Inc. + * Copyright (C) 2010, 2023 Red Hat, Inc. * All Rights Reserved. */ #include "xfs.h" @@ -19,21 +19,147 @@ #include "xfs_log.h" #include "xfs_ag.h" -STATIC int -xfs_trim_extents( +/* + * Notes on an efficient, low latency fstrim algorithm + * + * We need to walk the filesystem free space and issue discards on the free + * space that meet the search criteria (size and location). We cannot issue + * discards on extents that might be in use, or are so recently in use they are + * still marked as busy. To serialise against extent state changes whilst we are + * gathering extents to trim, we must hold the AGF lock to lock out other + * allocations and extent free operations that might change extent state. + * + * However, we cannot just hold the AGF for the entire AG free space walk whilst + * we issue discards on each free space that is found. Storage devices can have + * extremely slow discard implementations (e.g. ceph RBD) and so walking a + * couple of million free extents and issuing synchronous discards on each + * extent can take a *long* time. Whilst we are doing this walk, nothing else + * can access the AGF, and we can stall transactions and hence the log whilst + * modifications wait for the AGF lock to be released. This can lead hung tasks + * kicking the hung task timer and rebooting the system. This is bad. + * + * Hence we need to take a leaf from the bulkstat playbook. It takes the AGI + * lock, gathers a range of inode cluster buffers that are allocated, drops the + * AGI lock and then reads all the inode cluster buffers and processes them. It + * loops doing this, using a cursor to keep track of where it is up to in the AG + * for each iteration to restart the INOBT lookup from. + * + * We can't do this exactly with free space - once we drop the AGF lock, the + * state of the free extent is out of our control and we cannot run a discard + * safely on it in this situation. Unless, of course, we've marked the free + * extent as busy and undergoing a discard operation whilst we held the AGF + * locked. + * + * This is exactly how online discard works - free extents are marked busy when + * they are freed, and once the extent free has been committed to the journal, + * the busy extent record is marked as "undergoing discard" and the discard is + * then issued on the free extent. Once the discard completes, the busy extent + * record is removed and the extent is able to be allocated again. + * + * In the context of fstrim, if we find a free extent we need to discard, we + * don't have to discard it immediately. All we need to do it record that free + * extent as being busy and under discard, and all the allocation routines will + * now avoid trying to allocate it. Hence if we mark the extent as busy under + * the AGF lock, we can safely discard it without holding the AGF lock because + * nothing will attempt to allocate that free space until the discard completes. + * + * This also allows us to issue discards asynchronously like we do with online + * discard, and so for fast devices fstrim will run much faster as we can have + * multiple discard operations in flight at once, as well as pipeline the free + * extent search so that it overlaps in flight discard IO. + */ + +struct workqueue_struct *xfs_discard_wq; + +static void +xfs_discard_endio_work( + struct work_struct *work) +{ + struct xfs_busy_extents *extents = + container_of(work, struct xfs_busy_extents, endio_work); + + xfs_extent_busy_clear(extents->mount, &extents->extent_list, false); + kmem_free(extents->owner); +} + +/* + * Queue up the actual completion to a thread to avoid IRQ-safe locking for + * pagb_lock. + */ +static void +xfs_discard_endio( + struct bio *bio) +{ + struct xfs_busy_extents *extents = bio->bi_private; + + INIT_WORK(&extents->endio_work, xfs_discard_endio_work); + queue_work(xfs_discard_wq, &extents->endio_work); + bio_put(bio); +} + +/* + * Walk the discard list and issue discards on all the busy extents in the + * list. We plug and chain the bios so that we only need a single completion + * call to clear all the busy extents once the discards are complete. + */ +int +xfs_discard_extents( + struct xfs_mount *mp, + struct xfs_busy_extents *extents) +{ + struct xfs_extent_busy *busyp; + struct bio *bio = NULL; + struct blk_plug plug; + int error = 0; + + blk_start_plug(&plug); + list_for_each_entry(busyp, &extents->extent_list, list) { + trace_xfs_discard_extent(mp, busyp->agno, busyp->bno, + busyp->length); + + error = __blkdev_issue_discard(mp->m_ddev_targp->bt_bdev, + XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno), + XFS_FSB_TO_BB(mp, busyp->length), + GFP_NOFS, &bio); + if (error && error != -EOPNOTSUPP) { + xfs_info(mp, + "discard failed for extent [0x%llx,%u], error %d", + (unsigned long long)busyp->bno, + busyp->length, + error); + break; + } + } + + if (bio) { + bio->bi_private = extents; + bio->bi_end_io = xfs_discard_endio; + submit_bio(bio); + } else { + xfs_discard_endio_work(&extents->endio_work); + } + blk_finish_plug(&plug); + + return error; +} + + +static int +xfs_trim_gather_extents( struct xfs_perag *pag, xfs_daddr_t start, xfs_daddr_t end, xfs_daddr_t minlen, + struct xfs_alloc_rec_incore *tcur, + struct xfs_busy_extents *extents, uint64_t *blocks_trimmed) { struct xfs_mount *mp = pag->pag_mount; - struct block_device *bdev = mp->m_ddev_targp->bt_bdev; struct xfs_btree_cur *cur; struct xfs_buf *agbp; - struct xfs_agf *agf; int error; int i; + int batch = 100; /* * Force out the log. This means any transactions that might have freed @@ -45,20 +171,28 @@ xfs_trim_extents( error = xfs_alloc_read_agf(pag, NULL, 0, &agbp); if (error) return error; - agf = agbp->b_addr; cur = xfs_allocbt_init_cursor(mp, NULL, agbp, pag, XFS_BTNUM_CNT); /* - * Look up the longest btree in the AGF and start with it. + * Look up the extent length requested in the AGF and start with it. */ - error = xfs_alloc_lookup_ge(cur, 0, be32_to_cpu(agf->agf_longest), &i); + if (tcur->ar_startblock == NULLAGBLOCK) + error = xfs_alloc_lookup_ge(cur, 0, tcur->ar_blockcount, &i); + else + error = xfs_alloc_lookup_le(cur, tcur->ar_startblock, + tcur->ar_blockcount, &i); if (error) goto out_del_cursor; + if (i == 0) { + /* nothing of that length left in the AG, we are done */ + tcur->ar_blockcount = 0; + goto out_del_cursor; + } /* * Loop until we are done with all extents that are large - * enough to be worth discarding. + * enough to be worth discarding or we hit batch limits. */ while (i) { xfs_agblock_t fbno; @@ -73,7 +207,16 @@ xfs_trim_extents( error = -EFSCORRUPTED; break; } - ASSERT(flen <= be32_to_cpu(agf->agf_longest)); + + if (--batch <= 0) { + /* + * Update the cursor to point at this extent so we + * restart the next batch from this extent. + */ + tcur->ar_startblock = fbno; + tcur->ar_blockcount = flen; + break; + } /* * use daddr format for all range/len calculations as that is @@ -88,6 +231,7 @@ xfs_trim_extents( */ if (dlen < minlen) { trace_xfs_discard_toosmall(mp, pag->pag_agno, fbno, flen); + tcur->ar_blockcount = 0; break; } @@ -110,29 +254,103 @@ xfs_trim_extents( goto next_extent; } - trace_xfs_discard_extent(mp, pag->pag_agno, fbno, flen); - error = blkdev_issue_discard(bdev, dbno, dlen, GFP_NOFS); - if (error) - break; + xfs_extent_busy_insert_discard(pag, fbno, flen, + &extents->extent_list); *blocks_trimmed += flen; - next_extent: error = xfs_btree_decrement(cur, 0, &i); if (error) break; - if (fatal_signal_pending(current)) { - error = -ERESTARTSYS; - break; - } + /* + * If there's no more records in the tree, we are done. Set the + * cursor block count to 0 to indicate to the caller that there + * is no more extents to search. + */ + if (i == 0) + tcur->ar_blockcount = 0; } + /* + * If there was an error, release all the gathered busy extents because + * we aren't going to issue a discard on them any more. + */ + if (error) + xfs_extent_busy_clear(mp, &extents->extent_list, false); out_del_cursor: xfs_btree_del_cursor(cur, error); xfs_buf_relse(agbp); return error; } +static bool +xfs_trim_should_stop(void) +{ + return fatal_signal_pending(current) || freezing(current); +} + +/* + * Iterate the free list gathering extents and discarding them. We need a cursor + * for the repeated iteration of gather/discard loop, so use the longest extent + * we found in the last batch as the key to start the next. + */ +static int +xfs_trim_extents( + struct xfs_perag *pag, + xfs_daddr_t start, + xfs_daddr_t end, + xfs_daddr_t minlen, + uint64_t *blocks_trimmed) +{ + struct xfs_alloc_rec_incore tcur = { + .ar_blockcount = pag->pagf_longest, + .ar_startblock = NULLAGBLOCK, + }; + int error = 0; + + do { + struct xfs_busy_extents *extents; + + extents = kzalloc(sizeof(*extents), GFP_KERNEL); + if (!extents) { + error = -ENOMEM; + break; + } + + extents->mount = pag->pag_mount; + extents->owner = extents; + INIT_LIST_HEAD(&extents->extent_list); + + error = xfs_trim_gather_extents(pag, start, end, minlen, + &tcur, extents, blocks_trimmed); + if (error) { + kfree(extents); + break; + } + + /* + * We hand the extent list to the discard function here so the + * discarded extents can be removed from the busy extent list. + * This allows the discards to run asynchronously with gathering + * the next round of extents to discard. + * + * However, we must ensure that we do not reference the extent + * list after this function call, as it may have been freed by + * the time control returns to us. + */ + error = xfs_discard_extents(pag->pag_mount, extents); + if (error) + break; + + if (xfs_trim_should_stop()) + break; + + } while (tcur.ar_blockcount != 0); + + return error; + +} + /* * trim a range of the filesystem. * @@ -195,12 +413,12 @@ xfs_ioc_trim( for_each_perag_range(mp, agno, xfs_daddr_to_agno(mp, end), pag) { error = xfs_trim_extents(pag, start, end, minlen, &blocks_trimmed); - if (error) { + if (error) last_error = error; - if (error == -ERESTARTSYS) { - xfs_perag_rele(pag); - break; - } + + if (xfs_trim_should_stop()) { + xfs_perag_rele(pag); + break; } } diff --git a/fs/xfs/xfs_discard.h b/fs/xfs/xfs_discard.h index de92d9cc958f..2b1a85223a56 100644 --- a/fs/xfs/xfs_discard.h +++ b/fs/xfs/xfs_discard.h @@ -3,8 +3,10 @@ #define XFS_DISCARD_H 1 struct fstrim_range; -struct list_head; +struct xfs_mount; +struct xfs_busy_extents; -extern int xfs_ioc_trim(struct xfs_mount *, struct fstrim_range __user *); +int xfs_discard_extents(struct xfs_mount *mp, struct xfs_busy_extents *busy); +int xfs_ioc_trim(struct xfs_mount *mp, struct fstrim_range __user *fstrim); #endif /* XFS_DISCARD_H */ diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c index 7c2fdc71e42d..9ecfdcdc752f 100644 --- a/fs/xfs/xfs_extent_busy.c +++ b/fs/xfs/xfs_extent_busy.c @@ -19,13 +19,13 @@ #include "xfs_log.h" #include "xfs_ag.h" -void -xfs_extent_busy_insert( - struct xfs_trans *tp, +static void +xfs_extent_busy_insert_list( struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len, - unsigned int flags) + unsigned int flags, + struct list_head *busy_list) { struct xfs_extent_busy *new; struct xfs_extent_busy *busyp; @@ -40,7 +40,7 @@ xfs_extent_busy_insert( new->flags = flags; /* trace before insert to be able to see failed inserts */ - trace_xfs_extent_busy(tp->t_mountp, pag->pag_agno, bno, len); + trace_xfs_extent_busy(pag->pag_mount, pag->pag_agno, bno, len); spin_lock(&pag->pagb_lock); rbp = &pag->pagb_tree.rb_node; @@ -62,10 +62,33 @@ xfs_extent_busy_insert( rb_link_node(&new->rb_node, parent, rbp); rb_insert_color(&new->rb_node, &pag->pagb_tree); - list_add(&new->list, &tp->t_busy); + /* always process discard lists in fifo order */ + list_add_tail(&new->list, busy_list); spin_unlock(&pag->pagb_lock); } +void +xfs_extent_busy_insert( + struct xfs_trans *tp, + struct xfs_perag *pag, + xfs_agblock_t bno, + xfs_extlen_t len, + unsigned int flags) +{ + xfs_extent_busy_insert_list(pag, bno, len, flags, &tp->t_busy); +} + +void +xfs_extent_busy_insert_discard( + struct xfs_perag *pag, + xfs_agblock_t bno, + xfs_extlen_t len, + struct list_head *busy_list) +{ + xfs_extent_busy_insert_list(pag, bno, len, XFS_EXTENT_BUSY_DISCARDED, + busy_list); +} + /* * Search for a busy extent within the range of the extent we are about to * allocate. You need to be holding the busy extent tree lock when calling diff --git a/fs/xfs/xfs_extent_busy.h b/fs/xfs/xfs_extent_busy.h index c37bf87e6781..0639aab336f3 100644 --- a/fs/xfs/xfs_extent_busy.h +++ b/fs/xfs/xfs_extent_busy.h @@ -16,9 +16,6 @@ struct xfs_alloc_arg; /* * Busy block/extent entry. Indexed by a rbtree in perag to mark blocks that * have been freed but whose transactions aren't committed to disk yet. - * - * Note that we use the transaction ID to record the transaction, not the - * transaction structure itself. See xfs_extent_busy_insert() for details. */ struct xfs_extent_busy { struct rb_node rb_node; /* ag by-bno indexed search tree */ @@ -31,11 +28,32 @@ struct xfs_extent_busy { #define XFS_EXTENT_BUSY_SKIP_DISCARD 0x02 /* do not discard */ }; +/* + * List used to track groups of related busy extents all the way through + * to discard completion. + */ +struct xfs_busy_extents { + struct xfs_mount *mount; + struct list_head extent_list; + struct work_struct endio_work; + + /* + * Owner is the object containing the struct xfs_busy_extents to free + * once the busy extents have been processed. If only the + * xfs_busy_extents object needs freeing, then point this at itself. + */ + void *owner; +}; + void xfs_extent_busy_insert(struct xfs_trans *tp, struct xfs_perag *pag, xfs_agblock_t bno, xfs_extlen_t len, unsigned int flags); void +xfs_extent_busy_insert_discard(struct xfs_perag *pag, xfs_agblock_t bno, + xfs_extlen_t len, struct list_head *busy_list); + +void xfs_extent_busy_clear(struct xfs_mount *mp, struct list_head *list, bool do_discard); diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 1c1e6171209d..2b3b05c28e9e 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -584,6 +584,11 @@ xfs_vn_getattr( } } + if ((request_mask & STATX_CHANGE_COOKIE) && IS_I_VERSION(inode)) { + stat->change_cookie = inode_query_iversion(inode); + stat->result_mask |= STATX_CHANGE_COOKIE; + } + /* * Note: If you add another clause to set an attribute flag, please * update attributes_mask below. diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c index ebc70aaa299c..67a99d94701e 100644 --- a/fs/xfs/xfs_log_cil.c +++ b/fs/xfs/xfs_log_cil.c @@ -16,8 +16,7 @@ #include "xfs_log.h" #include "xfs_log_priv.h" #include "xfs_trace.h" - -struct workqueue_struct *xfs_discard_wq; +#include "xfs_discard.h" /* * Allocate a new ticket. Failing to get a new ticket makes it really hard to @@ -103,7 +102,7 @@ xlog_cil_ctx_alloc(void) ctx = kmem_zalloc(sizeof(*ctx), KM_NOFS); INIT_LIST_HEAD(&ctx->committing); - INIT_LIST_HEAD(&ctx->busy_extents); + INIT_LIST_HEAD(&ctx->busy_extents.extent_list); INIT_LIST_HEAD(&ctx->log_items); INIT_LIST_HEAD(&ctx->lv_chain); INIT_WORK(&ctx->push_work, xlog_cil_push_work); @@ -132,7 +131,7 @@ xlog_cil_push_pcp_aggregate( if (!list_empty(&cilpcp->busy_extents)) { list_splice_init(&cilpcp->busy_extents, - &ctx->busy_extents); + &ctx->busy_extents.extent_list); } if (!list_empty(&cilpcp->log_items)) list_splice_init(&cilpcp->log_items, &ctx->log_items); @@ -708,76 +707,6 @@ xlog_cil_free_logvec( } } -static void -xlog_discard_endio_work( - struct work_struct *work) -{ - struct xfs_cil_ctx *ctx = - container_of(work, struct xfs_cil_ctx, discard_endio_work); - struct xfs_mount *mp = ctx->cil->xc_log->l_mp; - - xfs_extent_busy_clear(mp, &ctx->busy_extents, false); - kmem_free(ctx); -} - -/* - * Queue up the actual completion to a thread to avoid IRQ-safe locking for - * pagb_lock. Note that we need a unbounded workqueue, otherwise we might - * get the execution delayed up to 30 seconds for weird reasons. - */ -static void -xlog_discard_endio( - struct bio *bio) -{ - struct xfs_cil_ctx *ctx = bio->bi_private; - - INIT_WORK(&ctx->discard_endio_work, xlog_discard_endio_work); - queue_work(xfs_discard_wq, &ctx->discard_endio_work); - bio_put(bio); -} - -static void -xlog_discard_busy_extents( - struct xfs_mount *mp, - struct xfs_cil_ctx *ctx) -{ - struct list_head *list = &ctx->busy_extents; - struct xfs_extent_busy *busyp; - struct bio *bio = NULL; - struct blk_plug plug; - int error = 0; - - ASSERT(xfs_has_discard(mp)); - - blk_start_plug(&plug); - list_for_each_entry(busyp, list, list) { - trace_xfs_discard_extent(mp, busyp->agno, busyp->bno, - busyp->length); - - error = __blkdev_issue_discard(mp->m_ddev_targp->bt_bdev, - XFS_AGB_TO_DADDR(mp, busyp->agno, busyp->bno), - XFS_FSB_TO_BB(mp, busyp->length), - GFP_NOFS, &bio); - if (error && error != -EOPNOTSUPP) { - xfs_info(mp, - "discard failed for extent [0x%llx,%u], error %d", - (unsigned long long)busyp->bno, - busyp->length, - error); - break; - } - } - - if (bio) { - bio->bi_private = ctx; - bio->bi_end_io = xlog_discard_endio; - submit_bio(bio); - } else { - xlog_discard_endio_work(&ctx->discard_endio_work); - } - blk_finish_plug(&plug); -} - /* * Mark all items committed and clear busy extents. We free the log vector * chains in a separate pass so that we unpin the log items as quickly as @@ -807,8 +736,8 @@ xlog_cil_committed( xfs_trans_committed_bulk(ctx->cil->xc_log->l_ailp, &ctx->lv_chain, ctx->start_lsn, abort); - xfs_extent_busy_sort(&ctx->busy_extents); - xfs_extent_busy_clear(mp, &ctx->busy_extents, + xfs_extent_busy_sort(&ctx->busy_extents.extent_list); + xfs_extent_busy_clear(mp, &ctx->busy_extents.extent_list, xfs_has_discard(mp) && !abort); spin_lock(&ctx->cil->xc_push_lock); @@ -817,10 +746,14 @@ xlog_cil_committed( xlog_cil_free_logvec(&ctx->lv_chain); - if (!list_empty(&ctx->busy_extents)) - xlog_discard_busy_extents(mp, ctx); - else - kmem_free(ctx); + if (!list_empty(&ctx->busy_extents.extent_list)) { + ctx->busy_extents.mount = mp; + ctx->busy_extents.owner = ctx; + xfs_discard_extents(mp, &ctx->busy_extents); + return; + } + + kmem_free(ctx); } void diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index af87648331d5..fa3ad1d7b31c 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -6,6 +6,8 @@ #ifndef __XFS_LOG_PRIV_H__ #define __XFS_LOG_PRIV_H__ +#include "xfs_extent_busy.h" /* for struct xfs_busy_extents */ + struct xfs_buf; struct xlog; struct xlog_ticket; @@ -223,12 +225,11 @@ struct xfs_cil_ctx { struct xlog_in_core *commit_iclog; struct xlog_ticket *ticket; /* chkpt ticket */ atomic_t space_used; /* aggregate size of regions */ - struct list_head busy_extents; /* busy extents in chkpt */ + struct xfs_busy_extents busy_extents; struct list_head log_items; /* log items in chkpt */ struct list_head lv_chain; /* logvecs being pushed */ struct list_head iclog_entry; struct list_head committing; /* ctx committing list */ - struct work_struct discard_endio_work; struct work_struct push_work; atomic_t order_id; diff --git a/fs/xfs/xfs_notify_failure.c b/fs/xfs/xfs_notify_failure.c index 4a9bbd3fe120..a7daa522e00f 100644 --- a/fs/xfs/xfs_notify_failure.c +++ b/fs/xfs/xfs_notify_failure.c @@ -126,8 +126,8 @@ xfs_dax_notify_ddev_failure( struct xfs_rmap_irec ri_low = { }; struct xfs_rmap_irec ri_high; struct xfs_agf *agf; - xfs_agblock_t agend; struct xfs_perag *pag; + xfs_agblock_t range_agend; pag = xfs_perag_get(mp, agno); error = xfs_alloc_read_agf(pag, tp, 0, &agf_bp); @@ -148,10 +148,10 @@ xfs_dax_notify_ddev_failure( ri_high.rm_startblock = XFS_FSB_TO_AGBNO(mp, end_fsbno); agf = agf_bp->b_addr; - agend = min(be32_to_cpu(agf->agf_length), + range_agend = min(be32_to_cpu(agf->agf_length) - 1, ri_high.rm_startblock); notify.startblock = ri_low.rm_startblock; - notify.blockcount = agend - ri_low.rm_startblock; + notify.blockcount = range_agend + 1 - ri_low.rm_startblock; error = xfs_rmap_query_range(cur, &ri_low, &ri_high, xfs_dax_failure_fn, ¬ify); |