From nobody Sun Feb 8 18:31:57 2026 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id CF2B926FD9B; Wed, 22 Oct 2025 16:07:09 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1761149229; cv=none; b=SNEekEIEg+2Duw6Khk8yQ7o40yiWTjdytsGhvSNb1ZGsn6GpBlomtnMPbRPuolveCOmXH6SLNItb5L7gcxQtZMyMa+xu/UHCT4ZBSmzAzdZQfcMZ8MbhySUGSdQszbwCKbdqWcTlR0NibKp7g7mWSt4V7CKOi+umKS90Y+cHi+g= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1761149229; c=relaxed/simple; bh=I9Gr6zapSqyfvp5VjlbXhfKK8aPWlTON9tgSCFZUh0w=; h=From:Date:Subject:MIME-Version:Content-Type:Message-Id:References: In-Reply-To:To:Cc; b=RIyhpPfDNFITODEvbeG3z7Amg9JoGdqWOh1tsO8eTfNv15BzQoM24wOgOTDU3hDy+D1feUH4VTsey1utDk/1Jx8b4reE3clxfIIt1psjh9B4w2HtOHcBixXEv94Oz6+30rEOSvYSGrCTVIdqieX7Qf6jnMfxwGmYHVNuBpm8+FM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b=vAvNeKsc; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel.org header.i=@kernel.org header.b="vAvNeKsc" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 7CA5CC4CEE7; Wed, 22 Oct 2025 16:07:04 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1761149229; bh=I9Gr6zapSqyfvp5VjlbXhfKK8aPWlTON9tgSCFZUh0w=; h=From:Date:Subject:References:In-Reply-To:To:Cc:From; b=vAvNeKscG15zFM3N/MnNRBtSKqa6Wq26wDv0VhEIVmlz/hpdrrnTnyq9YYj0PFMwn Uem8UzNH8KcxxeogQo7BHbjyLj9DHZv5YOUvWdMrOTJyxd/A6P3yLrmZm3p6qkb3/q cD7sjNNQTVf9X3A+bgscKWUZZVy5dPtLFIxkBzAyRXmBwrEI5qAq8ZDB5X/9ua8y6b 2achFZxUw+50H2Pq5gUy6K7jksgmBC3rr8fTk6EFLfX1QGaCmbLYyEY+HO1W/kawDO 6RPL99KMMq0NYYySjLx2fOWl11Th4tq1aQpVljivKtzjd2Hw7LBZgfrrihAXW6iDEd D7eAP/4/xpjKg== From: Christian Brauner Date: Wed, 22 Oct 2025 18:05:50 +0200 Subject: [PATCH v2 12/63] nstree: introduce a unified tree Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20251022-work-namespace-nstree-listns-v2-12-71a588572371@kernel.org> References: <20251022-work-namespace-nstree-listns-v2-0-71a588572371@kernel.org> In-Reply-To: <20251022-work-namespace-nstree-listns-v2-0-71a588572371@kernel.org> To: linux-fsdevel@vger.kernel.org, Josef Bacik , Jeff Layton Cc: Jann Horn , Mike Yuan , =?utf-8?q?Zbigniew_J=C4=99drzejewski-Szmek?= , Lennart Poettering , Daan De Meyer , Aleksa Sarai , Amir Goldstein , Tejun Heo , Johannes Weiner , Thomas Gleixner , Alexander Viro , Jan Kara , linux-kernel@vger.kernel.org, cgroups@vger.kernel.org, bpf@vger.kernel.org, Eric Dumazet , Jakub Kicinski , netdev@vger.kernel.org, Arnd Bergmann , Christian Brauner X-Mailer: b4 0.15-dev-96507 X-Developer-Signature: v=1; a=openpgp-sha256; l=8126; i=brauner@kernel.org; h=from:subject:message-id; bh=I9Gr6zapSqyfvp5VjlbXhfKK8aPWlTON9tgSCFZUh0w=; b=owGbwMvMwCU28Zj0gdSKO4sYT6slMWT8ZHjy69kGScXY78zPQq9Hd6ws1+5vuCR/xH7znn2FF uKmGgqPO0pZGMS4GGTFFFkc2k3C5ZbzVGw2ytSAmcPKBDKEgYtTACYyV5qRYXWh6JdV/YfUNlzO +JwvLX5r8pbzG+sEjvgU/uZ7v+hPzlSGv9I2p1k45PZn8D3k+LVyTsLDs71Gp86FfV4eJSNl2ud izQ8A X-Developer-Key: i=brauner@kernel.org; a=openpgp; fpr=4880B8C9BD0E5106FC070F4F7B3C391EFEA93624 This will allow userspace to lookup and stat a namespace simply by its identifier without having to know what type of namespace it is. Signed-off-by: Christian Brauner --- include/linux/ns_common.h | 4 ++ kernel/nscommon.c | 1 + kernel/nstree.c | 94 ++++++++++++++++++++++++++++++++++++-------= ---- 3 files changed, 77 insertions(+), 22 deletions(-) diff --git a/include/linux/ns_common.h b/include/linux/ns_common.h index f340a279acc2..b7d2123a9d19 100644 --- a/include/linux/ns_common.h +++ b/include/linux/ns_common.h @@ -115,6 +115,10 @@ struct ns_common { union { struct { u64 ns_id; + struct /* global namespace rbtree and list */ { + struct rb_node ns_unified_tree_node; + struct list_head ns_unified_list_node; + }; struct /* per type rbtree and list */ { struct rb_node ns_tree_node; struct list_head ns_list_node; diff --git a/kernel/nscommon.c b/kernel/nscommon.c index d750e554d50b..bdd32e14a587 100644 --- a/kernel/nscommon.c +++ b/kernel/nscommon.c @@ -61,6 +61,7 @@ int __ns_common_init(struct ns_common *ns, u32 ns_type, c= onst struct proc_ns_ope ns->ns_id =3D 0; ns->ns_type =3D ns_type; RB_CLEAR_NODE(&ns->ns_tree_node); + RB_CLEAR_NODE(&ns->ns_unified_tree_node); INIT_LIST_HEAD(&ns->ns_list_node); =20 #ifdef CONFIG_DEBUG_VFS diff --git a/kernel/nstree.c b/kernel/nstree.c index 369fd1675c6a..d21df06b6747 100644 --- a/kernel/nstree.c +++ b/kernel/nstree.c @@ -4,31 +4,30 @@ #include #include =20 +__cacheline_aligned_in_smp DEFINE_SEQLOCK(ns_tree_lock); +static struct rb_root ns_unified_tree =3D RB_ROOT; /* protected by ns_tree= _lock */ + /** * struct ns_tree - Namespace tree * @ns_tree: Rbtree of namespaces of a particular type * @ns_list: Sequentially walkable list of all namespaces of this type - * @ns_tree_lock: Seqlock to protect the tree and list * @type: type of namespaces in this tree */ struct ns_tree { - struct rb_root ns_tree; - struct list_head ns_list; - seqlock_t ns_tree_lock; - int type; + struct rb_root ns_tree; + struct list_head ns_list; + int type; }; =20 struct ns_tree mnt_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(mnt_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(mnt_ns_tree.ns_tree_lock), .type =3D CLONE_NEWNS, }; =20 struct ns_tree net_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(net_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(net_ns_tree.ns_tree_lock), .type =3D CLONE_NEWNET, }; EXPORT_SYMBOL_GPL(net_ns_tree); @@ -36,42 +35,36 @@ EXPORT_SYMBOL_GPL(net_ns_tree); struct ns_tree uts_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(uts_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(uts_ns_tree.ns_tree_lock), .type =3D CLONE_NEWUTS, }; =20 struct ns_tree user_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(user_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(user_ns_tree.ns_tree_lock), .type =3D CLONE_NEWUSER, }; =20 struct ns_tree ipc_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(ipc_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(ipc_ns_tree.ns_tree_lock), .type =3D CLONE_NEWIPC, }; =20 struct ns_tree pid_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(pid_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(pid_ns_tree.ns_tree_lock), .type =3D CLONE_NEWPID, }; =20 struct ns_tree cgroup_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(cgroup_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(cgroup_ns_tree.ns_tree_lock), .type =3D CLONE_NEWCGROUP, }; =20 struct ns_tree time_ns_tree =3D { .ns_tree =3D RB_ROOT, .ns_list =3D LIST_HEAD_INIT(time_ns_tree.ns_list), - .ns_tree_lock =3D __SEQLOCK_UNLOCKED(time_ns_tree.ns_tree_lock), .type =3D CLONE_NEWTIME, }; =20 @@ -84,6 +77,13 @@ static inline struct ns_common *node_to_ns(const struct = rb_node *node) return rb_entry(node, struct ns_common, ns_tree_node); } =20 +static inline struct ns_common *node_to_ns_unified(const struct rb_node *n= ode) +{ + if (!node) + return NULL; + return rb_entry(node, struct ns_common, ns_unified_tree_node); +} + static inline int ns_cmp(struct rb_node *a, const struct rb_node *b) { struct ns_common *ns_a =3D node_to_ns(a); @@ -98,13 +98,27 @@ static inline int ns_cmp(struct rb_node *a, const struc= t rb_node *b) return 0; } =20 +static inline int ns_cmp_unified(struct rb_node *a, const struct rb_node *= b) +{ + struct ns_common *ns_a =3D node_to_ns_unified(a); + struct ns_common *ns_b =3D node_to_ns_unified(b); + u64 ns_id_a =3D ns_a->ns_id; + u64 ns_id_b =3D ns_b->ns_id; + + if (ns_id_a < ns_id_b) + return -1; + if (ns_id_a > ns_id_b) + return 1; + return 0; +} + void __ns_tree_add_raw(struct ns_common *ns, struct ns_tree *ns_tree) { struct rb_node *node, *prev; =20 VFS_WARN_ON_ONCE(!ns->ns_id); =20 - write_seqlock(&ns_tree->ns_tree_lock); + write_seqlock(&ns_tree_lock); =20 VFS_WARN_ON_ONCE(ns->ns_type !=3D ns_tree->type); =20 @@ -119,7 +133,8 @@ void __ns_tree_add_raw(struct ns_common *ns, struct ns_= tree *ns_tree) else list_add_rcu(&ns->ns_list_node, &node_to_ns(prev)->ns_list_node); =20 - write_sequnlock(&ns_tree->ns_tree_lock); + rb_find_add_rcu(&ns->ns_unified_tree_node, &ns_unified_tree, ns_cmp_unifi= ed); + write_sequnlock(&ns_tree_lock); =20 VFS_WARN_ON_ONCE(node); } @@ -130,11 +145,12 @@ void __ns_tree_remove(struct ns_common *ns, struct ns= _tree *ns_tree) VFS_WARN_ON_ONCE(list_empty(&ns->ns_list_node)); VFS_WARN_ON_ONCE(ns->ns_type !=3D ns_tree->type); =20 - write_seqlock(&ns_tree->ns_tree_lock); + write_seqlock(&ns_tree_lock); rb_erase(&ns->ns_tree_node, &ns_tree->ns_tree); + rb_erase(&ns->ns_unified_tree_node, &ns_unified_tree); list_bidir_del_rcu(&ns->ns_list_node); RB_CLEAR_NODE(&ns->ns_tree_node); - write_sequnlock(&ns_tree->ns_tree_lock); + write_sequnlock(&ns_tree_lock); } EXPORT_SYMBOL_GPL(__ns_tree_remove); =20 @@ -150,6 +166,17 @@ static int ns_find(const void *key, const struct rb_no= de *node) return 0; } =20 +static int ns_find_unified(const void *key, const struct rb_node *node) +{ + const u64 ns_id =3D *(u64 *)key; + const struct ns_common *ns =3D node_to_ns_unified(node); + + if (ns_id < ns->ns_id) + return -1; + if (ns_id > ns->ns_id) + return 1; + return 0; +} =20 static struct ns_tree *ns_tree_from_type(int ns_type) { @@ -175,28 +202,51 @@ static struct ns_tree *ns_tree_from_type(int ns_type) return NULL; } =20 -struct ns_common *ns_tree_lookup_rcu(u64 ns_id, int ns_type) +static struct ns_common *__ns_unified_tree_lookup_rcu(u64 ns_id) { - struct ns_tree *ns_tree; struct rb_node *node; unsigned int seq; =20 - RCU_LOCKDEP_WARN(!rcu_read_lock_held(), "suspicious ns_tree_lookup_rcu() = usage"); + do { + seq =3D read_seqbegin(&ns_tree_lock); + node =3D rb_find_rcu(&ns_id, &ns_unified_tree, ns_find_unified); + if (node) + break; + } while (read_seqretry(&ns_tree_lock, seq)); + + return node_to_ns_unified(node); +} + +static struct ns_common *__ns_tree_lookup_rcu(u64 ns_id, int ns_type) +{ + struct ns_tree *ns_tree; + struct rb_node *node; + unsigned int seq; =20 ns_tree =3D ns_tree_from_type(ns_type); if (!ns_tree) return NULL; =20 do { - seq =3D read_seqbegin(&ns_tree->ns_tree_lock); + seq =3D read_seqbegin(&ns_tree_lock); node =3D rb_find_rcu(&ns_id, &ns_tree->ns_tree, ns_find); if (node) break; - } while (read_seqretry(&ns_tree->ns_tree_lock, seq)); + } while (read_seqretry(&ns_tree_lock, seq)); =20 return node_to_ns(node); } =20 +struct ns_common *ns_tree_lookup_rcu(u64 ns_id, int ns_type) +{ + RCU_LOCKDEP_WARN(!rcu_read_lock_held(), "suspicious ns_tree_lookup_rcu() = usage"); + + if (ns_type) + return __ns_tree_lookup_rcu(ns_id, ns_type); + + return __ns_unified_tree_lookup_rcu(ns_id); +} + /** * ns_tree_adjoined_rcu - find the next/previous namespace in the same * tree --=20 2.47.3