From nobody Thu Apr 2 21:38:48 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B0BABC6FA90 for ; Tue, 20 Sep 2022 18:09:13 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231135AbiITSJL (ORCPT ); Tue, 20 Sep 2022 14:09:11 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:40558 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231152AbiITSI7 (ORCPT ); Tue, 20 Sep 2022 14:08:59 -0400 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [205.139.111.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 6FF076747B for ; Tue, 20 Sep 2022 11:08:56 -0700 (PDT) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-168-cS9gUiCyO26z06RYJEcy-A-1; Tue, 20 Sep 2022 14:08:52 -0400 X-MC-Unique: cS9gUiCyO26z06RYJEcy-A-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id E98FF8027EA; Tue, 20 Sep 2022 18:08:51 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (unknown [10.40.208.17]) by smtp.corp.redhat.com (Postfix) with ESMTP id 821B92166B26; Tue, 20 Sep 2022 18:08:50 +0000 (UTC) From: Alexey Gladkov To: LKML , Linux Containers Cc: Andrew Morton , Christian Brauner , "Eric W . Biederman" , Kees Cook , Manfred Spraul Subject: [PATCH v2 1/3] sysctl: Allow change system v ipc sysctls inside ipc namespace Date: Tue, 20 Sep 2022 20:08:20 +0200 Message-Id: <0895bd453013370eb4f9600e26e2a9969ee755de.1663696560.git.legion@kernel.org> In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Rootless containers are not allowed to modify kernel IPC parameters. All default limits are set to such high values that in fact there are no limits at all. All limits are not inherited and are initialized to default values when a new ipc_namespace is created. For new ipc_namespace: size_t ipc_ns.shm_ctlmax =3D SHMMAX; // (ULONG_MAX - (1UL << 24)) size_t ipc_ns.shm_ctlall =3D SHMALL; // (ULONG_MAX - (1UL << 24)) int ipc_ns.shm_ctlmni =3D IPCMNI; // (1 << 15) int ipc_ns.shm_rmid_forced =3D 0; unsigned int ipc_ns.msg_ctlmax =3D MSGMAX; // 8192 unsigned int ipc_ns.msg_ctlmni =3D MSGMNI; // 32000 unsigned int ipc_ns.msg_ctlmnb =3D MSGMNB; // 16384 The shm_tot (total amount of shared pages) has also ceased to be global, it is located in ipc_namespace and is not inherited from anywhere. In such conditions, it cannot be said that these limits limit anything. The real limiter for them is cgroups. If we allow rootless containers to change these parameters, then it can only be reduced. Signed-off-by: Alexey Gladkov --- ipc/ipc_sysctl.c | 34 +++++++++++++++++++++++++++++++--- 1 file changed, 31 insertions(+), 3 deletions(-) diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c index ef313ecfb53a..a6a9d7f680dd 100644 --- a/ipc/ipc_sysctl.c +++ b/ipc/ipc_sysctl.c @@ -190,25 +190,53 @@ static int set_is_seen(struct ctl_table_set *set) return ¤t->nsproxy->ipc_ns->ipc_set =3D=3D set; } =20 +static void ipc_set_ownership(struct ctl_table_header *head, + struct ctl_table *table, + kuid_t *uid, kgid_t *gid) +{ + struct ipc_namespace *ns =3D + container_of(head->set, struct ipc_namespace, ipc_set); + + kuid_t ns_root_uid =3D make_kuid(ns->user_ns, 0); + kgid_t ns_root_gid =3D make_kgid(ns->user_ns, 0); + + *uid =3D uid_valid(ns_root_uid) ? ns_root_uid : GLOBAL_ROOT_UID; + *gid =3D gid_valid(ns_root_gid) ? ns_root_gid : GLOBAL_ROOT_GID; +} + static int ipc_permissions(struct ctl_table_header *head, struct ctl_table= *table) { + struct ipc_namespace *ns =3D + container_of(head->set, struct ipc_namespace, ipc_set); int mode =3D table->mode; + kuid_t ns_root_uid; + kgid_t ns_root_gid; =20 -#ifdef CONFIG_CHECKPOINT_RESTORE - struct ipc_namespace *ns =3D current->nsproxy->ipc_ns; + ipc_set_ownership(head, table, &ns_root_uid, ns_root_gid); =20 +#ifdef CONFIG_CHECKPOINT_RESTORE if (((table->data =3D=3D &ns->ids[IPC_SEM_IDS].next_id) || (table->data =3D=3D &ns->ids[IPC_MSG_IDS].next_id) || (table->data =3D=3D &ns->ids[IPC_SHM_IDS].next_id)) && checkpoint_restore_ns_capable(ns->user_ns)) mode =3D 0666; + else #endif - return mode; + if (uid_eq(current_euid(), ns_root_uid)) + mode >>=3D 6; + + else if (in_egroup_p(ns_root_gid)) + mode >>=3D 3; + + mode &=3D 7; + + return (mode << 6) | (mode << 3) | mode; } =20 static struct ctl_table_root set_root =3D { .lookup =3D set_lookup, .permissions =3D ipc_permissions, + .set_ownership =3D ipc_set_ownership, }; =20 bool setup_ipc_sysctls(struct ipc_namespace *ns) --=20 2.33.4 From nobody Thu Apr 2 21:38:48 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id E5E00C54EE9 for ; Tue, 20 Sep 2022 18:09:29 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231137AbiITSJ2 (ORCPT ); Tue, 20 Sep 2022 14:09:28 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41288 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230506AbiITSJZ (ORCPT ); Tue, 20 Sep 2022 14:09:25 -0400 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [207.211.30.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id C601470E68 for ; Tue, 20 Sep 2022 11:09:22 -0700 (PDT) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-315-DWruCP5jNLe8nJwirNOEOA-1; Tue, 20 Sep 2022 14:09:16 -0400 X-MC-Unique: DWruCP5jNLe8nJwirNOEOA-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id A8DD18001B8; Tue, 20 Sep 2022 18:09:14 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (unknown [10.40.208.17]) by smtp.corp.redhat.com (Postfix) with ESMTP id 384462166B26; Tue, 20 Sep 2022 18:09:13 +0000 (UTC) From: Alexey Gladkov To: LKML , Linux Containers Cc: Andrew Morton , Christian Brauner , "Eric W . Biederman" , Kees Cook , Manfred Spraul Subject: [PATCH v2 2/3] sysctl: Allow to change limits for posix messages queues Date: Tue, 20 Sep 2022 20:08:21 +0200 Message-Id: <207698b81c7e55b3d00d2476b081a8fee5d369f6.1663696560.git.legion@kernel.org> In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" All parameters of posix messages queues (queues_max/msg_max/msgsize_max) end up being limited by RLIMIT_MSGQUEUE. The code in mqueue_get_inode is where that limiting happens. The RLIMIT_MSGQUEUE is bound to the user namespace and is counted hierarchically. We can allow root in the user namespace to modify the posix messages queues parameters. Signed-off-by: Alexey Gladkov --- ipc/mq_sysctl.c | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+) diff --git a/ipc/mq_sysctl.c b/ipc/mq_sysctl.c index fbf6a8b93a26..5e573c657775 100644 --- a/ipc/mq_sysctl.c +++ b/ipc/mq_sysctl.c @@ -12,6 +12,7 @@ #include #include #include +#include =20 static int msg_max_limit_min =3D MIN_MSGMAX; static int msg_max_limit_max =3D HARD_MSGMAX; @@ -76,8 +77,45 @@ static int set_is_seen(struct ctl_table_set *set) return ¤t->nsproxy->ipc_ns->mq_set =3D=3D set; } =20 +static void mq_set_ownership(struct ctl_table_header *head, + struct ctl_table *table, + kuid_t *uid, kgid_t *gid) +{ + struct ipc_namespace *ns =3D + container_of(head->set, struct ipc_namespace, mq_set); + + kuid_t ns_root_uid =3D make_kuid(ns->user_ns, 0); + kgid_t ns_root_gid =3D make_kgid(ns->user_ns, 0); + + *uid =3D uid_valid(ns_root_uid) ? ns_root_uid : GLOBAL_ROOT_UID; + *gid =3D gid_valid(ns_root_gid) ? ns_root_gid : GLOBAL_ROOT_GID; +} + +static int mq_permissions(struct ctl_table_header *head, struct ctl_table = *table) +{ + struct ipc_namespace *ns =3D + container_of(head->set, struct ipc_namespace, mq_set); + int mode =3D table->mode; + kuid_t ns_root_uid; + kgid_t ns_root_gid; + + mq_set_ownership(head, table, &ns_root_uid, &ns_root_gid); + + if (uid_eq(current_euid(), ns_root_uid)) + mode >>=3D 6; + + if (in_egroup_p(ns_root_gid)) + mode >>=3D 3; + + mode &=3D 7; + + return (mode << 6) | (mode << 3) | mode; +} + static struct ctl_table_root set_root =3D { .lookup =3D set_lookup, + .permissions =3D mq_permissions, + .set_ownership =3D mq_set_ownership, }; =20 bool setup_mq_sysctls(struct ipc_namespace *ns) --=20 2.33.4 From nobody Thu Apr 2 21:38:48 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B10B7C54EE9 for ; Tue, 20 Sep 2022 18:10:00 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S230240AbiITSJ6 (ORCPT ); Tue, 20 Sep 2022 14:09:58 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:42924 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231285AbiITSJo (ORCPT ); Tue, 20 Sep 2022 14:09:44 -0400 Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [207.211.30.44]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7C04E60DA for ; Tue, 20 Sep 2022 11:09:35 -0700 (PDT) Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-185-DMmJYzxEOWSPHuBH53reNA-1; Tue, 20 Sep 2022 14:09:23 -0400 X-MC-Unique: DMmJYzxEOWSPHuBH53reNA-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id D5E38802801; Tue, 20 Sep 2022 18:09:22 +0000 (UTC) Received: from comp-core-i7-2640m-0182e6.redhat.com (unknown [10.40.208.17]) by smtp.corp.redhat.com (Postfix) with ESMTP id EC0032166B26; Tue, 20 Sep 2022 18:09:20 +0000 (UTC) From: Alexey Gladkov To: LKML , Linux Containers , linux-doc@vger.kernel.org, linux-man@vger.kernel.org Cc: Andrew Morton , Christian Brauner , "Eric W . Biederman" , Kees Cook , Manfred Spraul , Jonathan Corbet Subject: [PATCH v2 3/3] docs: Add information about ipc sysctls limitations Date: Tue, 20 Sep 2022 20:08:22 +0200 Message-Id: In-Reply-To: References: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-Scanned-By: MIMEDefang 3.1 on 10.11.54.6 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" After 25b21cb2f6d6 ("[PATCH] IPC namespace core") and 4e9823111bdc ("[PATCH] IPC namespace - shm") the shared memory page count stopped being global and started counting per ipc namespace. The documentation and shmget(2) still says that shmall is a global option. shmget(2): SHMALL System-wide limit on the total amount of shared memory, measured in units of the system page size. On Linux, this limit can be read and modified via /proc/sys/kernel/shmall. I think the changes made in 2006 should be documented. Signed-off-by: Alexey Gladkov Acked-by: "Eric W. Biederman" --- Documentation/admin-guide/sysctl/kernel.rst | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/ad= min-guide/sysctl/kernel.rst index ee6572b1edad..c8b89bd8f004 100644 --- a/Documentation/admin-guide/sysctl/kernel.rst +++ b/Documentation/admin-guide/sysctl/kernel.rst @@ -541,6 +541,9 @@ default (``MSGMNB``). ``msgmni`` is the maximum number of IPC queues. 32000 by default (``MSGMNI``). =20 +All of these parameters are set per ipc namespace. The maximum number of b= ytes +in POSIX message queues is limited by ``RLIMIT_MSGQUEUE``. This limit is +respected hierarchically in the each user namespace. =20 msg_next_id, sem_next_id, and shm_next_id (System V IPC) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D @@ -1181,15 +1184,20 @@ are doing anyway :) shmall =3D=3D=3D=3D=3D=3D =20 -This parameter sets the total amount of shared memory pages that -can be used system wide. Hence, ``shmall`` should always be at least -``ceil(shmmax/PAGE_SIZE)``. +This parameter sets the total amount of shared memory pages that can be us= ed +inside ipc namespace. The shared memory pages counting occurs for each ipc +namespace separately and is not inherited. Hence, ``shmall`` should always= be at +least ``ceil(shmmax/PAGE_SIZE)``. =20 If you are not sure what the default ``PAGE_SIZE`` is on your Linux system, you can run the following command:: =20 # getconf PAGE_SIZE =20 +To reduce or disable the ability to allocate shared memory, you must creat= e a +new ipc namespace, set this parameter to the required value and prohibit t= he +creation of a new ipc namespace in the current user namespace or cgroups c= an +be used. =20 shmmax =3D=3D=3D=3D=3D=3D --=20 2.33.4