[v8,08/12] landlock: Implement TCP network hooks

Message ID	20221021152644.155136-9-konstantin.meskhidze@huawei.com (mailing list archive)
State	Not Applicable
Headers	show Return-Path: <netdev-owner@kernel.org> From: Konstantin Meskhidze <konstantin.meskhidze@huawei.com> To: <mic@digikod.net> CC: <willemdebruijn.kernel@gmail.com>, <gnoack3000@gmail.com>, <linux-security-module@vger.kernel.org>, <netdev@vger.kernel.org>, <netfilter-devel@vger.kernel.org>, <artem.kuzin@huawei.com> Subject: [PATCH v8 08/12] landlock: Implement TCP network hooks Date: Fri, 21 Oct 2022 23:26:40 +0800 Message-ID: <20221021152644.155136-9-konstantin.meskhidze@huawei.com> In-Reply-To: <20221021152644.155136-1-konstantin.meskhidze@huawei.com> References: <20221021152644.155136-1-konstantin.meskhidze@huawei.com> MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 8bit Precedence: bulk
Series	Network support for Landlock \| expand [v8,00/12] Network support for Landlock [v8,01/12] landlock: Make ruleset's access masks more generic [v8,02/12] landlock: Refactor landlock_find_rule/insert_rule [v8,03/12] landlock: Refactor merge/inherit_ruleset functions [v8,04/12] landlock: Move unmask_layers() and init_layer_masks() [v8,05/12] landlock: Refactor unmask_layers() and init_layer_masks() [v8,06/12] landlock: Refactor landlock_add_rule() syscall [v8,07/12] landlock: Add network rules support [v8,08/12] landlock: Implement TCP network hooks [v8,09/12] selftests/landlock: Share enforce_ruleset() [v8,10/12] selftests/landlock: Add 10 new test suites dedicated to network [v8,11/12] samples/landlock: Add network demo [v8,12/12] landlock: Document Landlock's network support

Context	Check	Description
netdev/tree_selection	success	Not a local patch, async

Konstantin Meskhidze (A) Oct. 21, 2022, 3:26 p.m. UTC

This patch adds support of socket_bind() and socket_connect() hooks.
It's possible to restrict binding and connecting of TCP sockets to
particular ports.

Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
---

Changes since v7:
* Minor fixes.
* Refactors commit message.

Changes since v6:
* Updates copyright.
* Refactors landlock_append_net_rule() and check_socket_access()
  functions with landlock_id type.

Changes since v5:
* Fixes some logic errors.
* Formats code with clang-format-14.

Changes since v4:
* Factors out CONFIG_INET into make file.
* Refactors check_socket_access().
* Adds helper get_port().
* Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
functions to support AF_INET6 family.
* Adds AF_UNSPEC family support in hook_socket_bind/connect
functions.
* Refactors add_rule_net_service() and landlock_add_rule
syscall to support network rule inserting.
* Refactors init_layer_masks() to support network rules.

Changes since v3:
* Splits commit.
* Adds SECURITY_NETWORK in config.
* Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
* Adds hook_socket_bind and hook_socket_connect hooks.

---
 security/landlock/Kconfig    |   1 +
 security/landlock/Makefile   |   2 +
 security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
 security/landlock/net.h      |  26 ++++++
 security/landlock/setup.c    |   2 +
 security/landlock/syscalls.c |  59 ++++++++++++-
 6 files changed, 251 insertions(+), 3 deletions(-)
 create mode 100644 security/landlock/net.c
 create mode 100644 security/landlock/net.h

--
2.25.1

Mickaël Salaün Nov. 17, 2022, 6:43 p.m. UTC | #1

On 21/10/2022 17:26, Konstantin Meskhidze wrote:
> This patch adds support of socket_bind() and socket_connect() hooks.
> It's possible to restrict binding and connecting of TCP sockets to
> particular ports.

Implement socket_bind() and socket_connect LSM hooks, which enable to 
restrict TCP socket binding and connection to specific ports.


> 
> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
> ---
> 
> Changes since v7:
> * Minor fixes.
> * Refactors commit message.
> 
> Changes since v6:
> * Updates copyright.
> * Refactors landlock_append_net_rule() and check_socket_access()
>    functions with landlock_id type.
> 
> Changes since v5:
> * Fixes some logic errors.
> * Formats code with clang-format-14.
> 
> Changes since v4:
> * Factors out CONFIG_INET into make file.
> * Refactors check_socket_access().
> * Adds helper get_port().
> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
> functions to support AF_INET6 family.
> * Adds AF_UNSPEC family support in hook_socket_bind/connect
> functions.
> * Refactors add_rule_net_service() and landlock_add_rule
> syscall to support network rule inserting.
> * Refactors init_layer_masks() to support network rules.
> 
> Changes since v3:
> * Splits commit.
> * Adds SECURITY_NETWORK in config.
> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
> * Adds hook_socket_bind and hook_socket_connect hooks.
> 
> ---
>   security/landlock/Kconfig    |   1 +
>   security/landlock/Makefile   |   2 +
>   security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>   security/landlock/net.h      |  26 ++++++
>   security/landlock/setup.c    |   2 +
>   security/landlock/syscalls.c |  59 ++++++++++++-
>   6 files changed, 251 insertions(+), 3 deletions(-)
>   create mode 100644 security/landlock/net.c
>   create mode 100644 security/landlock/net.h
> 
> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
> index 8e33c4e8ffb8..10c099097533 100644
> --- a/security/landlock/Kconfig
> +++ b/security/landlock/Kconfig
> @@ -3,6 +3,7 @@
>   config SECURITY_LANDLOCK
>   	bool "Landlock support"
>   	depends on SECURITY && !ARCH_EPHEMERAL_INODES
> +	select SECURITY_NETWORK
>   	select SECURITY_PATH
>   	help
>   	  Landlock is a sandboxing mechanism that enables processes to restrict
> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
> index 7bbd2f413b3e..53d3c92ae22e 100644
> --- a/security/landlock/Makefile
> +++ b/security/landlock/Makefile
> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
> 
>   landlock-y := setup.o syscalls.o object.o ruleset.o \
>   	cred.o ptrace.o fs.o
> +
> +landlock-$(CONFIG_INET) += net.o
> \ No newline at end of file
> diff --git a/security/landlock/net.c b/security/landlock/net.c
> new file mode 100644
> index 000000000000..39e8a156a1f4
> --- /dev/null
> +++ b/security/landlock/net.c
> @@ -0,0 +1,164 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +/*
> + * Landlock LSM - Network management and hooks
> + *
> + * Copyright © 2022 Huawei Tech. Co., Ltd.
> + * Copyright © 2022 Microsoft Corporation
> + */
> +
> +#include <linux/in.h>
> +#include <linux/net.h>
> +#include <linux/socket.h>
> +#include <net/ipv6.h>
> +
> +#include "common.h"
> +#include "cred.h"
> +#include "limits.h"
> +#include "net.h"
> +#include "ruleset.h"
> +
> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
> +			     const u16 port, access_mask_t access_rights)
> +{
> +	int err;
> +	const struct landlock_id id = {
> +		.key.data = port,
> +		.type = LANDLOCK_KEY_NET_PORT,
> +	};
> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
> +
> +	/* Transforms relative access rights to absolute ones. */
> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
> +			 ~landlock_get_net_access_mask(ruleset, 0);
> +
> +	mutex_lock(&ruleset->lock);
> +	err = landlock_insert_rule(ruleset, id, access_rights);
> +	mutex_unlock(&ruleset->lock);
> +
> +	return err;
> +}
> +
> +static int check_socket_access(const struct landlock_ruleset *const domain,
> +			       u16 port, access_mask_t access_request)
> +{
> +	bool allowed = false;
> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
> +	const struct landlock_rule *rule;
> +	access_mask_t handled_access;
> +	const struct landlock_id id = {
> +		.key.data = port,
> +		.type = LANDLOCK_KEY_NET_PORT,
> +	};
> +
> +	if (WARN_ON_ONCE(!domain))
> +		return 0;
> +	if (WARN_ON_ONCE(domain->num_layers < 1))
> +		return -EACCES;
> +
> +	rule = landlock_find_rule(domain, id);
> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
> +					  LANDLOCK_KEY_NET_PORT);
> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
> +				ARRAY_SIZE(layer_masks));
> +
> +	return allowed ? 0 : -EACCES;
> +}
> +
> +static u16 get_port(const struct sockaddr *const address)

get_port() should return a __be16 type. This enables to avoid converting 
port when checking a rule.

make C=2 security/landlock/ must not print any warning.


> +{
> +	/* Gets port value in host byte order. */
> +	switch (address->sa_family) {
> +	case AF_UNSPEC:
> +	case AF_INET: {
> +		const struct sockaddr_in *const sockaddr =
> +			(struct sockaddr_in *)address;
> +		return ntohs(sockaddr->sin_port);
> +	}
> +#if IS_ENABLED(CONFIG_IPV6)
> +	case AF_INET6: {
> +		const struct sockaddr_in6 *const sockaddr_ip6 =
> +			(struct sockaddr_in6 *)address;
> +		return ntohs(sockaddr_ip6->sin6_port);
> +	}
> +#endif
> +	}
> +	WARN_ON_ONCE(1);
> +	return 0;
> +}
> +
> +static int hook_socket_bind(struct socket *sock, struct sockaddr *address,
> +			    int addrlen)
> +{
> +	const struct landlock_ruleset *const dom =
> +		landlock_get_current_domain();
> +
> +	if (!dom)
> +		return 0;
> +
> +	/* Check if it's a TCP socket. */
> +	if (sock->type != SOCK_STREAM)
> +		return 0;
> +
> +	switch (address->sa_family) {
> +	case AF_UNSPEC:
> +	case AF_INET:
> +#if IS_ENABLED(CONFIG_IPV6)
> +	case AF_INET6:
> +#endif
> +		return check_socket_access(dom, get_port(address),
> +					   LANDLOCK_ACCESS_NET_BIND_TCP);
> +	default:
> +		return 0;

You can remove this default case and move the return 0 at the end of the 
function.


> +	}
> +}
> +
> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
> +			       int addrlen)
> +{
> +	const struct landlock_ruleset *const dom =
> +		landlock_get_current_domain();
> +
> +	if (!dom)
> +		return 0;
> +
> +	/* Check if it's a TCP socket. */
> +	if (sock->type != SOCK_STREAM)
> +		return 0;
> +
> +	/* Check if the hook is AF_INET* socket's action. */
> +	switch (address->sa_family) {
> +	case AF_INET:
> +#if IS_ENABLED(CONFIG_IPV6)
> +	case AF_INET6:
> +#endif
> +		return check_socket_access(dom, get_port(address),
> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
> +	case AF_UNSPEC: {
> +		u16 i;

You can move "i" after the "dom" declaration to remove the extra braces.


> +
> +		/*
> +		 * If just in a layer a mask supports connect access,
> +		 * the socket_connect() hook with AF_UNSPEC family flag
> +		 * must be banned. This prevents from disconnecting already
> +		 * connected sockets.
> +		 */
> +		for (i = 0; i < dom->num_layers; i++) {
> +			if (landlock_get_net_access_mask(dom, i) &
> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
> +				return -EACCES;

I'm wondering if this is the right error code for this case. EPERM may 
be more appropriate.

Thinking more about this case, I don't understand what is the rationale 
to deny such action. What would be the consequence to always allow 
connection with AF_UNSPEC (i.e. to disconnect a socket)?


> +		}
> +	}
> +	}
> +	return 0;
> +}
> +
> +static struct security_hook_list landlock_hooks[] __lsm_ro_after_init = {
> +	LSM_HOOK_INIT(socket_bind, hook_socket_bind),
> +	LSM_HOOK_INIT(socket_connect, hook_socket_connect),
> +};
> +
> +__init void landlock_add_net_hooks(void)
> +{
> +	security_add_hooks(landlock_hooks, ARRAY_SIZE(landlock_hooks),
> +			   LANDLOCK_NAME);
> +}
> diff --git a/security/landlock/net.h b/security/landlock/net.h
> new file mode 100644
> index 000000000000..0da1d9dff5ab
> --- /dev/null
> +++ b/security/landlock/net.h
> @@ -0,0 +1,26 @@
> +/* SPDX-License-Identifier: GPL-2.0-only */
> +/*
> + * Landlock LSM - Network management and hooks
> + *
> + * Copyright © 2022 Huawei Tech. Co., Ltd.
> + */
> +
> +#ifndef _SECURITY_LANDLOCK_NET_H
> +#define _SECURITY_LANDLOCK_NET_H
> +
> +#include "common.h"
> +#include "ruleset.h"
> +#include "setup.h"
> +
> +#if IS_ENABLED(CONFIG_INET)
> +__init void landlock_add_net_hooks(void);
> +
> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
> +			     const u16 port, access_mask_t access_rights);
> +#else /* IS_ENABLED(CONFIG_INET) */
> +static inline void landlock_add_net_hooks(void)
> +{
> +}
> +#endif /* IS_ENABLED(CONFIG_INET) */
> +
> +#endif /* _SECURITY_LANDLOCK_NET_H */
> diff --git a/security/landlock/setup.c b/security/landlock/setup.c
> index 3f196d2ce4f9..7e4a598177b8 100644
> --- a/security/landlock/setup.c
> +++ b/security/landlock/setup.c
> @@ -14,6 +14,7 @@
>   #include "fs.h"
>   #include "ptrace.h"
>   #include "setup.h"
> +#include "net.h"
> 
>   bool landlock_initialized __lsm_ro_after_init = false;
> 
> @@ -29,6 +30,7 @@ static int __init landlock_init(void)
>   	landlock_add_cred_hooks();
>   	landlock_add_ptrace_hooks();
>   	landlock_add_fs_hooks();
> +	landlock_add_net_hooks();
>   	landlock_initialized = true;
>   	pr_info("Up and running.\n");
>   	return 0;
> diff --git a/security/landlock/syscalls.c b/security/landlock/syscalls.c
> index c5a6ad4e2fca..7853f32e8325 100644
> --- a/security/landlock/syscalls.c
> +++ b/security/landlock/syscalls.c
> @@ -29,6 +29,7 @@
>   #include "cred.h"
>   #include "fs.h"
>   #include "limits.h"
> +#include "net.h"
>   #include "ruleset.h"
>   #include "setup.h"
> 
> @@ -74,7 +75,8 @@ static void build_check_abi(void)
>   {
>   	struct landlock_ruleset_attr ruleset_attr;
>   	struct landlock_path_beneath_attr path_beneath_attr;
> -	size_t ruleset_size, path_beneath_size;
> +	struct landlock_net_service_attr net_service_attr;
> +	size_t ruleset_size, path_beneath_size, net_service_size;
> 
>   	/*
>   	 * For each user space ABI structures, first checks that there is no
> @@ -90,6 +92,11 @@ static void build_check_abi(void)
>   	path_beneath_size += sizeof(path_beneath_attr.parent_fd);
>   	BUILD_BUG_ON(sizeof(path_beneath_attr) != path_beneath_size);
>   	BUILD_BUG_ON(sizeof(path_beneath_attr) != 12);
> +
> +	net_service_size = sizeof(net_service_attr.allowed_access);
> +	net_service_size += sizeof(net_service_attr.port);
> +	BUILD_BUG_ON(sizeof(net_service_attr) != net_service_size);
> +	BUILD_BUG_ON(sizeof(net_service_attr) != 10);
>   }
> 
>   /* Ruleset handling */
> @@ -322,13 +329,54 @@ static int add_rule_path_beneath(struct landlock_ruleset *const ruleset,
>   	return err;
>   }
> 
> +static int add_rule_net_service(struct landlock_ruleset *ruleset,
> +				const void __user *const rule_attr)
> +{
> +#if IS_ENABLED(CONFIG_INET)
> +	struct landlock_net_service_attr net_service_attr;
> +	int res;
> +	u32 mask;

access_mask_t mask;


> +
> +	/* Copies raw user space buffer, only one type for now. */
> +	res = copy_from_user(&net_service_attr, rule_attr,
> +			     sizeof(net_service_attr));
> +	if (res)
> +		return -EFAULT;
> +
> +	/*
> +	 * Informs about useless rule: empty allowed_access (i.e. deny rules)
> +	 * are ignored by network actions.
> +	 */
> +	if (!net_service_attr.allowed_access)
> +		return -ENOMSG;
> +
> +	/*
> +	 * Checks that allowed_access matches the @ruleset constraints
> +	 * (ruleset->access_masks[0] is automatically upgraded to 64-bits).
> +	 */
> +	mask = landlock_get_net_access_mask(ruleset, 0);
> +	if ((net_service_attr.allowed_access | mask) != mask)
> +		return -EINVAL;
> +
> +	/* Denies inserting a rule with port 0. */
> +	if (net_service_attr.port == 0)
> +		return -EINVAL;
> +
> +	/* Imports the new rule. */
> +	return landlock_append_net_rule(ruleset, net_service_attr.port,
> +					net_service_attr.allowed_access);
> +#else /* IS_ENABLED(CONFIG_INET) */
> +	return -EAFNOSUPPORT;
> +#endif /* IS_ENABLED(CONFIG_INET) */
> +}
> +
>   /**
>    * sys_landlock_add_rule - Add a new rule to a ruleset
>    *
>    * @ruleset_fd: File descriptor tied to the ruleset that should be extended
>    *		with the new rule.
> - * @rule_type: Identify the structure type pointed to by @rule_attr (only
> - *             %LANDLOCK_RULE_PATH_BENEATH for now).
> + * @rule_type: Identify the structure type pointed to by @rule_attr:
> + *             %LANDLOCK_RULE_PATH_BENEATH or %LANDLOCK_RULE_NET_SERVICE.
>    * @rule_attr: Pointer to a rule (only of type &struct
>    *             landlock_path_beneath_attr for now).
>    * @flags: Must be 0.
> @@ -339,6 +387,8 @@ static int add_rule_path_beneath(struct landlock_ruleset *const ruleset,
>    * Possible returned errors are:
>    *
>    * - %EOPNOTSUPP: Landlock is supported by the kernel but disabled at boot time;
> + * - %EAFNOSUPPORT: @rule_type is LANDLOCK_RULE_NET_SERVICE but TCP/IP is not

%LANDLOCK_RULE_NET_SERVICE


> + *   supported by the running kernel;
>    * - %EINVAL: @flags is not 0, or inconsistent access in the rule (i.e.
>    *   &landlock_path_beneath_attr.allowed_access is not a subset of the
>    *   ruleset handled accesses);
> @@ -373,6 +423,9 @@ SYSCALL_DEFINE4(landlock_add_rule, const int, ruleset_fd,
>   	case LANDLOCK_RULE_PATH_BENEATH:
>   		err = add_rule_path_beneath(ruleset, rule_attr);
>   		break;
> +	case LANDLOCK_RULE_NET_SERVICE:
> +		err = add_rule_net_service(ruleset, rule_attr);
> +		break;
>   	default:
>   		err = -EINVAL;
>   		break;
> --
> 2.25.1
>

Konstantin Meskhidze (A) Nov. 28, 2022, 8:21 a.m. UTC | #2

11/17/2022 9:43 PM, Mickaël Salaün пишет:
> 
> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>> This patch adds support of socket_bind() and socket_connect() hooks.
>> It's possible to restrict binding and connecting of TCP sockets to
>> particular ports.
> 
> Implement socket_bind() and socket_connect LSM hooks, which enable to
> restrict TCP socket binding and connection to specific ports.
> 
   Ok. Thanks.
> 
>> 
>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>> ---
>> 
>> Changes since v7:
>> * Minor fixes.
>> * Refactors commit message.
>> 
>> Changes since v6:
>> * Updates copyright.
>> * Refactors landlock_append_net_rule() and check_socket_access()
>>    functions with landlock_id type.
>> 
>> Changes since v5:
>> * Fixes some logic errors.
>> * Formats code with clang-format-14.
>> 
>> Changes since v4:
>> * Factors out CONFIG_INET into make file.
>> * Refactors check_socket_access().
>> * Adds helper get_port().
>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>> functions to support AF_INET6 family.
>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>> functions.
>> * Refactors add_rule_net_service() and landlock_add_rule
>> syscall to support network rule inserting.
>> * Refactors init_layer_masks() to support network rules.
>> 
>> Changes since v3:
>> * Splits commit.
>> * Adds SECURITY_NETWORK in config.
>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>> * Adds hook_socket_bind and hook_socket_connect hooks.
>> 
>> ---
>>   security/landlock/Kconfig    |   1 +
>>   security/landlock/Makefile   |   2 +
>>   security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>   security/landlock/net.h      |  26 ++++++
>>   security/landlock/setup.c    |   2 +
>>   security/landlock/syscalls.c |  59 ++++++++++++-
>>   6 files changed, 251 insertions(+), 3 deletions(-)
>>   create mode 100644 security/landlock/net.c
>>   create mode 100644 security/landlock/net.h
>> 
>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>> index 8e33c4e8ffb8..10c099097533 100644
>> --- a/security/landlock/Kconfig
>> +++ b/security/landlock/Kconfig
>> @@ -3,6 +3,7 @@
>>   config SECURITY_LANDLOCK
>>   	bool "Landlock support"
>>   	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>> +	select SECURITY_NETWORK
>>   	select SECURITY_PATH
>>   	help
>>   	  Landlock is a sandboxing mechanism that enables processes to restrict
>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>> index 7bbd2f413b3e..53d3c92ae22e 100644
>> --- a/security/landlock/Makefile
>> +++ b/security/landlock/Makefile
>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>> 
>>   landlock-y := setup.o syscalls.o object.o ruleset.o \
>>   	cred.o ptrace.o fs.o
>> +
>> +landlock-$(CONFIG_INET) += net.o
>> \ No newline at end of file
>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>> new file mode 100644
>> index 000000000000..39e8a156a1f4
>> --- /dev/null
>> +++ b/security/landlock/net.c
>> @@ -0,0 +1,164 @@
>> +// SPDX-License-Identifier: GPL-2.0-only
>> +/*
>> + * Landlock LSM - Network management and hooks
>> + *
>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>> + * Copyright © 2022 Microsoft Corporation
>> + */
>> +
>> +#include <linux/in.h>
>> +#include <linux/net.h>
>> +#include <linux/socket.h>
>> +#include <net/ipv6.h>
>> +
>> +#include "common.h"
>> +#include "cred.h"
>> +#include "limits.h"
>> +#include "net.h"
>> +#include "ruleset.h"
>> +
>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>> +			     const u16 port, access_mask_t access_rights)
>> +{
>> +	int err;
>> +	const struct landlock_id id = {
>> +		.key.data = port,
>> +		.type = LANDLOCK_KEY_NET_PORT,
>> +	};
>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>> +
>> +	/* Transforms relative access rights to absolute ones. */
>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>> +
>> +	mutex_lock(&ruleset->lock);
>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>> +	mutex_unlock(&ruleset->lock);
>> +
>> +	return err;
>> +}
>> +
>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>> +			       u16 port, access_mask_t access_request)
>> +{
>> +	bool allowed = false;
>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>> +	const struct landlock_rule *rule;
>> +	access_mask_t handled_access;
>> +	const struct landlock_id id = {
>> +		.key.data = port,
>> +		.type = LANDLOCK_KEY_NET_PORT,
>> +	};
>> +
>> +	if (WARN_ON_ONCE(!domain))
>> +		return 0;
>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>> +		return -EACCES;
>> +
>> +	rule = landlock_find_rule(domain, id);
>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>> +					  LANDLOCK_KEY_NET_PORT);
>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>> +				ARRAY_SIZE(layer_masks));
>> +
>> +	return allowed ? 0 : -EACCES;
>> +}
>> +
>> +static u16 get_port(const struct sockaddr *const address)
> 
> get_port() should return a __be16 type. This enables to avoid converting
> port when checking a rule.
> 
> make C=2 security/landlock/ must not print any warning.

   Got it.
> 
> 
>> +{
>> +	/* Gets port value in host byte order. */
>> +	switch (address->sa_family) {
>> +	case AF_UNSPEC:
>> +	case AF_INET: {
>> +		const struct sockaddr_in *const sockaddr =
>> +			(struct sockaddr_in *)address;
>> +		return ntohs(sockaddr->sin_port);
>> +	}
>> +#if IS_ENABLED(CONFIG_IPV6)
>> +	case AF_INET6: {
>> +		const struct sockaddr_in6 *const sockaddr_ip6 =
>> +			(struct sockaddr_in6 *)address;
>> +		return ntohs(sockaddr_ip6->sin6_port);
>> +	}
>> +#endif
>> +	}
>> +	WARN_ON_ONCE(1);
>> +	return 0;
>> +}
>> +
>> +static int hook_socket_bind(struct socket *sock, struct sockaddr *address,
>> +			    int addrlen)
>> +{
>> +	const struct landlock_ruleset *const dom =
>> +		landlock_get_current_domain();
>> +
>> +	if (!dom)
>> +		return 0;
>> +
>> +	/* Check if it's a TCP socket. */
>> +	if (sock->type != SOCK_STREAM)
>> +		return 0;
>> +
>> +	switch (address->sa_family) {
>> +	case AF_UNSPEC:
>> +	case AF_INET:
>> +#if IS_ENABLED(CONFIG_IPV6)
>> +	case AF_INET6:
>> +#endif
>> +		return check_socket_access(dom, get_port(address),
>> +					   LANDLOCK_ACCESS_NET_BIND_TCP);
>> +	default:
>> +		return 0;
> 
> You can remove this default case and move the return 0 at the end of the
> function.
> 
   Ok. Will be refactored.
> 
>> +	}
>> +}
>> +
>> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
>> +			       int addrlen)
>> +{
>> +	const struct landlock_ruleset *const dom =
>> +		landlock_get_current_domain();
>> +
>> +	if (!dom)
>> +		return 0;
>> +
>> +	/* Check if it's a TCP socket. */
>> +	if (sock->type != SOCK_STREAM)
>> +		return 0;
>> +
>> +	/* Check if the hook is AF_INET* socket's action. */
>> +	switch (address->sa_family) {
>> +	case AF_INET:
>> +#if IS_ENABLED(CONFIG_IPV6)
>> +	case AF_INET6:
>> +#endif
>> +		return check_socket_access(dom, get_port(address),
>> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
>> +	case AF_UNSPEC: {
>> +		u16 i;
> 
> You can move "i" after the "dom" declaration to remove the extra braces.
> 
   Ok. Thanks.
> 
>> +
>> +		/*
>> +		 * If just in a layer a mask supports connect access,
>> +		 * the socket_connect() hook with AF_UNSPEC family flag
>> +		 * must be banned. This prevents from disconnecting already
>> +		 * connected sockets.
>> +		 */
>> +		for (i = 0; i < dom->num_layers; i++) {
>> +			if (landlock_get_net_access_mask(dom, i) &
>> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
>> +				return -EACCES;
> 
> I'm wondering if this is the right error code for this case. EPERM may
> be more appropriate.

   Ok. Will be refactored.
> 
> Thinking more about this case, I don't understand what is the rationale
> to deny such action. What would be the consequence to always allow
> connection with AF_UNSPEC (i.e. to disconnect a socket)?
> 
   I thought we have come to a conclusion about connect(...AF_UNSPEC..) 
  behaviour in the patchset V3:
https://lore.kernel.org/linux-security-module/19ad3a01-d76e-0e73-7833-99acd4afd97e@huawei.com/
> 
>> +		}
>> +	}
>> +	}
>> +	return 0;
>> +}
>> +
>> +static struct security_hook_list landlock_hooks[] __lsm_ro_after_init = {
>> +	LSM_HOOK_INIT(socket_bind, hook_socket_bind),
>> +	LSM_HOOK_INIT(socket_connect, hook_socket_connect),
>> +};
>> +
>> +__init void landlock_add_net_hooks(void)
>> +{
>> +	security_add_hooks(landlock_hooks, ARRAY_SIZE(landlock_hooks),
>> +			   LANDLOCK_NAME);
>> +}
>> diff --git a/security/landlock/net.h b/security/landlock/net.h
>> new file mode 100644
>> index 000000000000..0da1d9dff5ab
>> --- /dev/null
>> +++ b/security/landlock/net.h
>> @@ -0,0 +1,26 @@
>> +/* SPDX-License-Identifier: GPL-2.0-only */
>> +/*
>> + * Landlock LSM - Network management and hooks
>> + *
>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>> + */
>> +
>> +#ifndef _SECURITY_LANDLOCK_NET_H
>> +#define _SECURITY_LANDLOCK_NET_H
>> +
>> +#include "common.h"
>> +#include "ruleset.h"
>> +#include "setup.h"
>> +
>> +#if IS_ENABLED(CONFIG_INET)
>> +__init void landlock_add_net_hooks(void);
>> +
>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>> +			     const u16 port, access_mask_t access_rights);
>> +#else /* IS_ENABLED(CONFIG_INET) */
>> +static inline void landlock_add_net_hooks(void)
>> +{
>> +}
>> +#endif /* IS_ENABLED(CONFIG_INET) */
>> +
>> +#endif /* _SECURITY_LANDLOCK_NET_H */
>> diff --git a/security/landlock/setup.c b/security/landlock/setup.c
>> index 3f196d2ce4f9..7e4a598177b8 100644
>> --- a/security/landlock/setup.c
>> +++ b/security/landlock/setup.c
>> @@ -14,6 +14,7 @@
>>   #include "fs.h"
>>   #include "ptrace.h"
>>   #include "setup.h"
>> +#include "net.h"
>> 
>>   bool landlock_initialized __lsm_ro_after_init = false;
>> 
>> @@ -29,6 +30,7 @@ static int __init landlock_init(void)
>>   	landlock_add_cred_hooks();
>>   	landlock_add_ptrace_hooks();
>>   	landlock_add_fs_hooks();
>> +	landlock_add_net_hooks();
>>   	landlock_initialized = true;
>>   	pr_info("Up and running.\n");
>>   	return 0;
>> diff --git a/security/landlock/syscalls.c b/security/landlock/syscalls.c
>> index c5a6ad4e2fca..7853f32e8325 100644
>> --- a/security/landlock/syscalls.c
>> +++ b/security/landlock/syscalls.c
>> @@ -29,6 +29,7 @@
>>   #include "cred.h"
>>   #include "fs.h"
>>   #include "limits.h"
>> +#include "net.h"
>>   #include "ruleset.h"
>>   #include "setup.h"
>> 
>> @@ -74,7 +75,8 @@ static void build_check_abi(void)
>>   {
>>   	struct landlock_ruleset_attr ruleset_attr;
>>   	struct landlock_path_beneath_attr path_beneath_attr;
>> -	size_t ruleset_size, path_beneath_size;
>> +	struct landlock_net_service_attr net_service_attr;
>> +	size_t ruleset_size, path_beneath_size, net_service_size;
>> 
>>   	/*
>>   	 * For each user space ABI structures, first checks that there is no
>> @@ -90,6 +92,11 @@ static void build_check_abi(void)
>>   	path_beneath_size += sizeof(path_beneath_attr.parent_fd);
>>   	BUILD_BUG_ON(sizeof(path_beneath_attr) != path_beneath_size);
>>   	BUILD_BUG_ON(sizeof(path_beneath_attr) != 12);
>> +
>> +	net_service_size = sizeof(net_service_attr.allowed_access);
>> +	net_service_size += sizeof(net_service_attr.port);
>> +	BUILD_BUG_ON(sizeof(net_service_attr) != net_service_size);
>> +	BUILD_BUG_ON(sizeof(net_service_attr) != 10);
>>   }
>> 
>>   /* Ruleset handling */
>> @@ -322,13 +329,54 @@ static int add_rule_path_beneath(struct landlock_ruleset *const ruleset,
>>   	return err;
>>   }
>> 
>> +static int add_rule_net_service(struct landlock_ruleset *ruleset,
>> +				const void __user *const rule_attr)
>> +{
>> +#if IS_ENABLED(CONFIG_INET)
>> +	struct landlock_net_service_attr net_service_attr;
>> +	int res;
>> +	u32 mask;
> 
> access_mask_t mask;

  Got it. Thanks.
> 
> 
>> +
>> +	/* Copies raw user space buffer, only one type for now. */
>> +	res = copy_from_user(&net_service_attr, rule_attr,
>> +			     sizeof(net_service_attr));
>> +	if (res)
>> +		return -EFAULT;
>> +
>> +	/*
>> +	 * Informs about useless rule: empty allowed_access (i.e. deny rules)
>> +	 * are ignored by network actions.
>> +	 */
>> +	if (!net_service_attr.allowed_access)
>> +		return -ENOMSG;
>> +
>> +	/*
>> +	 * Checks that allowed_access matches the @ruleset constraints
>> +	 * (ruleset->access_masks[0] is automatically upgraded to 64-bits).
>> +	 */
>> +	mask = landlock_get_net_access_mask(ruleset, 0);
>> +	if ((net_service_attr.allowed_access | mask) != mask)
>> +		return -EINVAL;
>> +
>> +	/* Denies inserting a rule with port 0. */
>> +	if (net_service_attr.port == 0)
>> +		return -EINVAL;
>> +
>> +	/* Imports the new rule. */
>> +	return landlock_append_net_rule(ruleset, net_service_attr.port,
>> +					net_service_attr.allowed_access);
>> +#else /* IS_ENABLED(CONFIG_INET) */
>> +	return -EAFNOSUPPORT;
>> +#endif /* IS_ENABLED(CONFIG_INET) */
>> +}
>> +
>>   /**
>>    * sys_landlock_add_rule - Add a new rule to a ruleset
>>    *
>>    * @ruleset_fd: File descriptor tied to the ruleset that should be extended
>>    *		with the new rule.
>> - * @rule_type: Identify the structure type pointed to by @rule_attr (only
>> - *             %LANDLOCK_RULE_PATH_BENEATH for now).
>> + * @rule_type: Identify the structure type pointed to by @rule_attr:
>> + *             %LANDLOCK_RULE_PATH_BENEATH or %LANDLOCK_RULE_NET_SERVICE.
>>    * @rule_attr: Pointer to a rule (only of type &struct
>>    *             landlock_path_beneath_attr for now).
>>    * @flags: Must be 0.
>> @@ -339,6 +387,8 @@ static int add_rule_path_beneath(struct landlock_ruleset *const ruleset,
>>    * Possible returned errors are:
>>    *
>>    * - %EOPNOTSUPP: Landlock is supported by the kernel but disabled at boot time;
>> + * - %EAFNOSUPPORT: @rule_type is LANDLOCK_RULE_NET_SERVICE but TCP/IP is not
> 
> %LANDLOCK_RULE_NET_SERVICE

  Ok.
> 
> 
>> + *   supported by the running kernel;
>>    * - %EINVAL: @flags is not 0, or inconsistent access in the rule (i.e.
>>    *   &landlock_path_beneath_attr.allowed_access is not a subset of the
>>    *   ruleset handled accesses);
>> @@ -373,6 +423,9 @@ SYSCALL_DEFINE4(landlock_add_rule, const int, ruleset_fd,
>>   	case LANDLOCK_RULE_PATH_BENEATH:
>>   		err = add_rule_path_beneath(ruleset, rule_attr);
>>   		break;
>> +	case LANDLOCK_RULE_NET_SERVICE:
>> +		err = add_rule_net_service(ruleset, rule_attr);
>> +		break;
>>   	default:
>>   		err = -EINVAL;
>>   		break;
>> --
>> 2.25.1
>> 
> .

Mickaël Salaün Nov. 28, 2022, 9 p.m. UTC | #3

The previous commit provides an interface to theoretically restrict 
network access (i.e. ruleset handled network accesses), but in fact this 
is not enforced until this commit. I like this split but to avoid any 
inconsistency, please squash this commit into the previous one: "7/12 
landlock: Add network rules support"
You should keep all the commit messages but maybe tweak them a bit.


On 28/11/2022 09:21, Konstantin Meskhidze (A) wrote:
> 
> 
> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>
>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>> It's possible to restrict binding and connecting of TCP sockets to
>>> particular ports.
>>
>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>> restrict TCP socket binding and connection to specific ports.
>>
>     Ok. Thanks.
>>
>>>
>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>> ---

[...]

>>> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
>>> +			       int addrlen)
>>> +{
>>> +	const struct landlock_ruleset *const dom =
>>> +		landlock_get_current_domain();
>>> +
>>> +	if (!dom)
>>> +		return 0;
>>> +
>>> +	/* Check if it's a TCP socket. */
>>> +	if (sock->type != SOCK_STREAM)
>>> +		return 0;
>>> +
>>> +	/* Check if the hook is AF_INET* socket's action. */
>>> +	switch (address->sa_family) {
>>> +	case AF_INET:
>>> +#if IS_ENABLED(CONFIG_IPV6)
>>> +	case AF_INET6:
>>> +#endif
>>> +		return check_socket_access(dom, get_port(address),
>>> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
>>> +	case AF_UNSPEC: {
>>> +		u16 i;
>>
>> You can move "i" after the "dom" declaration to remove the extra braces.
>>
>     Ok. Thanks.
>>
>>> +
>>> +		/*
>>> +		 * If just in a layer a mask supports connect access,
>>> +		 * the socket_connect() hook with AF_UNSPEC family flag
>>> +		 * must be banned. This prevents from disconnecting already
>>> +		 * connected sockets.
>>> +		 */
>>> +		for (i = 0; i < dom->num_layers; i++) {
>>> +			if (landlock_get_net_access_mask(dom, i) &
>>> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
>>> +				return -EACCES;
>>
>> I'm wondering if this is the right error code for this case. EPERM may
>> be more appropriate.
> 
>     Ok. Will be refactored.
>>
>> Thinking more about this case, I don't understand what is the rationale
>> to deny such action. What would be the consequence to always allow
>> connection with AF_UNSPEC (i.e. to disconnect a socket)?
>>
>     I thought we have come to a conclusion about connect(...AF_UNSPEC..)
>    behaviour in the patchset V3:
> https://lore.kernel.org/linux-security-module/19ad3a01-d76e-0e73-7833-99acd4afd97e@huawei.com/

The conclusion was that AF_UNSPEC disconnects a socket, but I'm asking 
if this is a security issue. I don't think it is more dangerous than a 
new (unconnected) socket. Am I missing something? Which kind of rule 
could be bypassed? What are we protecting against by restricting AF_UNSPEC?

We could then reduce the hook codes to just:
return current_check_access_socket(sock, address, LANDLOCK_ACCESS_NET_*);

Konstantin Meskhidze (A) Dec. 2, 2022, 3:13 a.m. UTC | #4

11/29/2022 12:00 AM, Mickaël Salaün пишет:
> The previous commit provides an interface to theoretically restrict
> network access (i.e. ruleset handled network accesses), but in fact this
> is not enforced until this commit. I like this split but to avoid any
> inconsistency, please squash this commit into the previous one: "7/12
> landlock: Add network rules support"
> You should keep all the commit messages but maybe tweak them a bit.
> 
   Ok. Will be squashed.
> 
> On 28/11/2022 09:21, Konstantin Meskhidze (A) wrote:
>> 
>> 
>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>
>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>> particular ports.
>>>
>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>> restrict TCP socket binding and connection to specific ports.
>>>
>>     Ok. Thanks.
>>>
>>>>
>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>> ---
> 
> [...]
> 
>>>> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
>>>> +			       int addrlen)
>>>> +{
>>>> +	const struct landlock_ruleset *const dom =
>>>> +		landlock_get_current_domain();
>>>> +
>>>> +	if (!dom)
>>>> +		return 0;
>>>> +
>>>> +	/* Check if it's a TCP socket. */
>>>> +	if (sock->type != SOCK_STREAM)
>>>> +		return 0;
>>>> +
>>>> +	/* Check if the hook is AF_INET* socket's action. */
>>>> +	switch (address->sa_family) {
>>>> +	case AF_INET:
>>>> +#if IS_ENABLED(CONFIG_IPV6)
>>>> +	case AF_INET6:
>>>> +#endif
>>>> +		return check_socket_access(dom, get_port(address),
>>>> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
>>>> +	case AF_UNSPEC: {
>>>> +		u16 i;
>>>
>>> You can move "i" after the "dom" declaration to remove the extra braces.
>>>
>>     Ok. Thanks.
>>>
>>>> +
>>>> +		/*
>>>> +		 * If just in a layer a mask supports connect access,
>>>> +		 * the socket_connect() hook with AF_UNSPEC family flag
>>>> +		 * must be banned. This prevents from disconnecting already
>>>> +		 * connected sockets.
>>>> +		 */
>>>> +		for (i = 0; i < dom->num_layers; i++) {
>>>> +			if (landlock_get_net_access_mask(dom, i) &
>>>> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
>>>> +				return -EACCES;
>>>
>>> I'm wondering if this is the right error code for this case. EPERM may
>>> be more appropriate.
>> 
>>     Ok. Will be refactored.
>>>
>>> Thinking more about this case, I don't understand what is the rationale
>>> to deny such action. What would be the consequence to always allow
>>> connection with AF_UNSPEC (i.e. to disconnect a socket)?
>>>
>>     I thought we have come to a conclusion about connect(...AF_UNSPEC..)
>>    behaviour in the patchset V3:
>> https://lore.kernel.org/linux-security-module/19ad3a01-d76e-0e73-7833-99acd4afd97e@huawei.com/
> 
> The conclusion was that AF_UNSPEC disconnects a socket, but I'm asking
> if this is a security issue. I don't think it is more dangerous than a
> new (unconnected) socket. Am I missing something? Which kind of rule
> could be bypassed? What are we protecting against by restricting AF_UNSPEC?

I just follow Willem de Bruijn concerns about this issue:

quote: "It is valid to pass an address with AF_UNSPEC to a PF_INET(6) 
socket. And there are legitimate reasons to want to deny this. Such as 
passing a connection to a unprivileged process and disallow it from 
disconnect and opening a different new connection."

https://lore.kernel.org/linux-security-module/CA+FuTSf4EjgjBCCOiu-PHJcTMia41UkTh8QJ0+qdxL_J8445EA@mail.gmail.com/


quote: "The intended use-case is for a privileged process to open a 
connection (i.e., bound and connected socket) and pass that to a 
restricted process. The intent is for that process to only be allowed to
communicate over this pre-established channel.

In practice, it is able to disconnect (while staying bound) and
elevate its privileges to that of a listening server: ..."

https://lore.kernel.org/linux-security-module/CA+FuTScaoby-=xRKf_Dz3koSYHqrMN0cauCg4jMmy_nDxwPADA@mail.gmail.com/

Looks like it's a security issue here.

> 
> We could then reduce the hook codes to just:
> return current_check_access_socket(sock, address, LANDLOCK_ACCESS_NET_*);
> .

Mickaël Salaün Dec. 2, 2022, 1:01 p.m. UTC | #5

On 02/12/2022 04:13, Konstantin Meskhidze (A) wrote:
> 
> 
> 11/29/2022 12:00 AM, Mickaël Salaün пишет:
>> The previous commit provides an interface to theoretically restrict
>> network access (i.e. ruleset handled network accesses), but in fact this
>> is not enforced until this commit. I like this split but to avoid any
>> inconsistency, please squash this commit into the previous one: "7/12
>> landlock: Add network rules support"
>> You should keep all the commit messages but maybe tweak them a bit.
>>
>     Ok. Will be squashed.
>>
>> On 28/11/2022 09:21, Konstantin Meskhidze (A) wrote:
>>>
>>>
>>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>>
>>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>>> particular ports.
>>>>
>>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>>> restrict TCP socket binding and connection to specific ports.
>>>>
>>>      Ok. Thanks.
>>>>
>>>>>
>>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>>> ---
>>
>> [...]
>>
>>>>> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
>>>>> +			       int addrlen)
>>>>> +{
>>>>> +	const struct landlock_ruleset *const dom =
>>>>> +		landlock_get_current_domain();
>>>>> +
>>>>> +	if (!dom)
>>>>> +		return 0;
>>>>> +
>>>>> +	/* Check if it's a TCP socket. */
>>>>> +	if (sock->type != SOCK_STREAM)
>>>>> +		return 0;
>>>>> +
>>>>> +	/* Check if the hook is AF_INET* socket's action. */
>>>>> +	switch (address->sa_family) {
>>>>> +	case AF_INET:
>>>>> +#if IS_ENABLED(CONFIG_IPV6)
>>>>> +	case AF_INET6:
>>>>> +#endif
>>>>> +		return check_socket_access(dom, get_port(address),
>>>>> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
>>>>> +	case AF_UNSPEC: {
>>>>> +		u16 i;
>>>>
>>>> You can move "i" after the "dom" declaration to remove the extra braces.
>>>>
>>>      Ok. Thanks.
>>>>
>>>>> +
>>>>> +		/*
>>>>> +		 * If just in a layer a mask supports connect access,
>>>>> +		 * the socket_connect() hook with AF_UNSPEC family flag
>>>>> +		 * must be banned. This prevents from disconnecting already
>>>>> +		 * connected sockets.
>>>>> +		 */
>>>>> +		for (i = 0; i < dom->num_layers; i++) {
>>>>> +			if (landlock_get_net_access_mask(dom, i) &
>>>>> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
>>>>> +				return -EACCES;
>>>>
>>>> I'm wondering if this is the right error code for this case. EPERM may
>>>> be more appropriate.
>>>
>>>      Ok. Will be refactored.
>>>>
>>>> Thinking more about this case, I don't understand what is the rationale
>>>> to deny such action. What would be the consequence to always allow
>>>> connection with AF_UNSPEC (i.e. to disconnect a socket)?
>>>>
>>>      I thought we have come to a conclusion about connect(...AF_UNSPEC..)
>>>     behaviour in the patchset V3:
>>> https://lore.kernel.org/linux-security-module/19ad3a01-d76e-0e73-7833-99acd4afd97e@huawei.com/
>>
>> The conclusion was that AF_UNSPEC disconnects a socket, but I'm asking
>> if this is a security issue. I don't think it is more dangerous than a
>> new (unconnected) socket. Am I missing something? Which kind of rule
>> could be bypassed? What are we protecting against by restricting AF_UNSPEC?
> 
> I just follow Willem de Bruijn concerns about this issue:
> 
> quote: "It is valid to pass an address with AF_UNSPEC to a PF_INET(6)
> socket. And there are legitimate reasons to want to deny this. Such as
> passing a connection to a unprivileged process and disallow it from
> disconnect and opening a different new connection."
> 
> https://lore.kernel.org/linux-security-module/CA+FuTSf4EjgjBCCOiu-PHJcTMia41UkTh8QJ0+qdxL_J8445EA@mail.gmail.com/

I agree with the fact that we want to deny this, but in this example the 
new connection should still be restricted by the Landlock domain. Using 
AF_UNSPEC on a connected socket should not make this socket allowed to 
create any connection if the process is restricted with TCP_CONNECT. 
Being allowed to close a connection should not be an issue, and any new 
connection must be vetted by Landlock.

> 
> 
> quote: "The intended use-case is for a privileged process to open a
> connection (i.e., bound and connected socket) and pass that to a
> restricted process. The intent is for that process to only be allowed to
> communicate over this pre-established channel.
> 
> In practice, it is able to disconnect (while staying bound) and
> elevate its privileges to that of a listening server: ..."
> 
> https://lore.kernel.org/linux-security-module/CA+FuTScaoby-=xRKf_Dz3koSYHqrMN0cauCg4jMmy_nDxwPADA@mail.gmail.com/
> 
> Looks like it's a security issue here.

It the provided example, if child_process() is restricted with 
TCP_CONNECT and TCP_BIND, any call to connect() or bind() will return an 
access error. listen() and accept() would work if the socket is bound, 
which is the case here, and then implicitly allowed by the parent 
process. I don' see any security issue. Am I missing something?

In fact, connect with AF_UNSPEC should always be allowed to be 
consistent with close(2), which is a way to drop privileges.


What Willem said:
> It would be good to also
> ensure that a now-bound socket cannot call listen.

This is not relevant for Landlock because the security model is to check 
process's requests to get new accesses (e.g. create a new file 
descriptor), but not to check passed accesses (e.g. inherited from a 
parent process, or pass through a unix socket) which are delegated to 
the sender/parent. The goal of a sandbox is to limit the set of new 
access requested (to the kernel) from within this sandbox. All already 
opened file descriptors were previously vetted by Landlock (and other 
access control systems).

> 
>>
>> We could then reduce the hook codes to just:
>> return current_check_access_socket(sock, address, LANDLOCK_ACCESS_NET_*);
>> .

As for SELinux, the connect hook should first do this check (with an 
appropriate comment):
if (address->sa_family == AF_UNSPEC)
	return 0;

Konstantin Meskhidze (A) Dec. 5, 2022, 2:55 a.m. UTC | #6

12/2/2022 4:01 PM, Mickaël Salaün пишет:
> 
> On 02/12/2022 04:13, Konstantin Meskhidze (A) wrote:
>> 
>> 
>> 11/29/2022 12:00 AM, Mickaël Salaün пишет:
>>> The previous commit provides an interface to theoretically restrict
>>> network access (i.e. ruleset handled network accesses), but in fact this
>>> is not enforced until this commit. I like this split but to avoid any
>>> inconsistency, please squash this commit into the previous one: "7/12
>>> landlock: Add network rules support"
>>> You should keep all the commit messages but maybe tweak them a bit.
>>>
>>     Ok. Will be squashed.
>>>
>>> On 28/11/2022 09:21, Konstantin Meskhidze (A) wrote:
>>>>
>>>>
>>>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>>>
>>>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>>>> particular ports.
>>>>>
>>>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>>>> restrict TCP socket binding and connection to specific ports.
>>>>>
>>>>      Ok. Thanks.
>>>>>
>>>>>>
>>>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>>>> ---
>>>
>>> [...]
>>>
>>>>>> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
>>>>>> +			       int addrlen)
>>>>>> +{
>>>>>> +	const struct landlock_ruleset *const dom =
>>>>>> +		landlock_get_current_domain();
>>>>>> +
>>>>>> +	if (!dom)
>>>>>> +		return 0;
>>>>>> +
>>>>>> +	/* Check if it's a TCP socket. */
>>>>>> +	if (sock->type != SOCK_STREAM)
>>>>>> +		return 0;
>>>>>> +
>>>>>> +	/* Check if the hook is AF_INET* socket's action. */
>>>>>> +	switch (address->sa_family) {
>>>>>> +	case AF_INET:
>>>>>> +#if IS_ENABLED(CONFIG_IPV6)
>>>>>> +	case AF_INET6:
>>>>>> +#endif
>>>>>> +		return check_socket_access(dom, get_port(address),
>>>>>> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
>>>>>> +	case AF_UNSPEC: {
>>>>>> +		u16 i;
>>>>>
>>>>> You can move "i" after the "dom" declaration to remove the extra braces.
>>>>>
>>>>      Ok. Thanks.
>>>>>
>>>>>> +
>>>>>> +		/*
>>>>>> +		 * If just in a layer a mask supports connect access,
>>>>>> +		 * the socket_connect() hook with AF_UNSPEC family flag
>>>>>> +		 * must be banned. This prevents from disconnecting already
>>>>>> +		 * connected sockets.
>>>>>> +		 */
>>>>>> +		for (i = 0; i < dom->num_layers; i++) {
>>>>>> +			if (landlock_get_net_access_mask(dom, i) &
>>>>>> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
>>>>>> +				return -EACCES;
>>>>>
>>>>> I'm wondering if this is the right error code for this case. EPERM may
>>>>> be more appropriate.
>>>>
>>>>      Ok. Will be refactored.
>>>>>
>>>>> Thinking more about this case, I don't understand what is the rationale
>>>>> to deny such action. What would be the consequence to always allow
>>>>> connection with AF_UNSPEC (i.e. to disconnect a socket)?
>>>>>
>>>>      I thought we have come to a conclusion about connect(...AF_UNSPEC..)
>>>>     behaviour in the patchset V3:
>>>> https://lore.kernel.org/linux-security-module/19ad3a01-d76e-0e73-7833-99acd4afd97e@huawei.com/
>>>
>>> The conclusion was that AF_UNSPEC disconnects a socket, but I'm asking
>>> if this is a security issue. I don't think it is more dangerous than a
>>> new (unconnected) socket. Am I missing something? Which kind of rule
>>> could be bypassed? What are we protecting against by restricting AF_UNSPEC?
>> 
>> I just follow Willem de Bruijn concerns about this issue:
>> 
>> quote: "It is valid to pass an address with AF_UNSPEC to a PF_INET(6)
>> socket. And there are legitimate reasons to want to deny this. Such as
>> passing a connection to a unprivileged process and disallow it from
>> disconnect and opening a different new connection."
>> 
>> https://lore.kernel.org/linux-security-module/CA+FuTSf4EjgjBCCOiu-PHJcTMia41UkTh8QJ0+qdxL_J8445EA@mail.gmail.com/
> 
> I agree with the fact that we want to deny this, but in this example the
> new connection should still be restricted by the Landlock domain. Using
> AF_UNSPEC on a connected socket should not make this socket allowed to
> create any connection if the process is restricted with TCP_CONNECT.
> Being allowed to close a connection should not be an issue, and any new
> connection must be vetted by Landlock.
> 

   You are right. This makes sense. Thanks for the comment.
>> 
>> 
>> quote: "The intended use-case is for a privileged process to open a
>> connection (i.e., bound and connected socket) and pass that to a
>> restricted process. The intent is for that process to only be allowed to
>> communicate over this pre-established channel.
>> 
>> In practice, it is able to disconnect (while staying bound) and
>> elevate its privileges to that of a listening server: ..."
>> 
>> https://lore.kernel.org/linux-security-module/CA+FuTScaoby-=xRKf_Dz3koSYHqrMN0cauCg4jMmy_nDxwPADA@mail.gmail.com/
>> 
>> Looks like it's a security issue here.
> 
> It the provided example, if child_process() is restricted with
> TCP_CONNECT and TCP_BIND, any call to connect() or bind() will return an
> access error. listen() and accept() would work if the socket is bound,
> which is the case here, and then implicitly allowed by the parent
> process. I don' see any security issue. Am I missing something?
> 
> In fact, connect with AF_UNSPEC should always be allowed to be
> consistent with close(2), which is a way to drop privileges.
> 

  It should be allowed with checking:
"return check_socket_access(dom, get_port(address),
                                  LANDLOCK_ACCESS_NET_CONNECT_TCP);
> 
> What Willem said:
>> It would be good to also
>> ensure that a now-bound socket cannot call listen.
> 
> This is not relevant for Landlock because the security model is to check
> process's requests to get new accesses (e.g. create a new file
> descriptor), but not to check passed accesses (e.g. inherited from a
> parent process, or pass through a unix socket) which are delegated to
> the sender/parent. The goal of a sandbox is to limit the set of new
> access requested (to the kernel) from within this sandbox. All already
> opened file descriptors were previously vetted by Landlock (and other
> access control systems).

    I got your point. Thanks.
> 
>> 
>>>
>>> We could then reduce the hook codes to just:
>>> return current_check_access_socket(sock, address, LANDLOCK_ACCESS_NET_*);
>>> .
> 
> As for SELinux, the connect hook should first do this check (with an
> appropriate comment):
> if (address->sa_family == AF_UNSPEC)
> 	return 0;

   In case of Landlock it looks like a landlocked process could connnect 
to the ports it's not allowed to connect to.
So we need just to return check_socket_access(dom, get_port(address),
				   LANDLOCK_ACCESS_NET_CONNECT_TCP);
I'm I correct? Did I miss something?
> .

Mickaël Salaün Dec. 5, 2022, 1:18 p.m. UTC | #7

On 05/12/2022 03:55, Konstantin Meskhidze (A) wrote:
> 
> 
> 12/2/2022 4:01 PM, Mickaël Salaün пишет:
>>
>> On 02/12/2022 04:13, Konstantin Meskhidze (A) wrote:
>>>
>>>
>>> 11/29/2022 12:00 AM, Mickaël Salaün пишет:
>>>> The previous commit provides an interface to theoretically restrict
>>>> network access (i.e. ruleset handled network accesses), but in fact this
>>>> is not enforced until this commit. I like this split but to avoid any
>>>> inconsistency, please squash this commit into the previous one: "7/12
>>>> landlock: Add network rules support"
>>>> You should keep all the commit messages but maybe tweak them a bit.
>>>>
>>>      Ok. Will be squashed.
>>>>
>>>> On 28/11/2022 09:21, Konstantin Meskhidze (A) wrote:
>>>>>
>>>>>
>>>>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>>>>
>>>>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>>>>> particular ports.
>>>>>>
>>>>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>>>>> restrict TCP socket binding and connection to specific ports.
>>>>>>
>>>>>       Ok. Thanks.
>>>>>>
>>>>>>>
>>>>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>>>>> ---
>>>>
>>>> [...]
>>>>
>>>>>>> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
>>>>>>> +			       int addrlen)
>>>>>>> +{
>>>>>>> +	const struct landlock_ruleset *const dom =
>>>>>>> +		landlock_get_current_domain();
>>>>>>> +
>>>>>>> +	if (!dom)
>>>>>>> +		return 0;
>>>>>>> +
>>>>>>> +	/* Check if it's a TCP socket. */
>>>>>>> +	if (sock->type != SOCK_STREAM)
>>>>>>> +		return 0;
>>>>>>> +
>>>>>>> +	/* Check if the hook is AF_INET* socket's action. */
>>>>>>> +	switch (address->sa_family) {
>>>>>>> +	case AF_INET:
>>>>>>> +#if IS_ENABLED(CONFIG_IPV6)
>>>>>>> +	case AF_INET6:
>>>>>>> +#endif
>>>>>>> +		return check_socket_access(dom, get_port(address),
>>>>>>> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
>>>>>>> +	case AF_UNSPEC: {
>>>>>>> +		u16 i;
>>>>>>
>>>>>> You can move "i" after the "dom" declaration to remove the extra braces.
>>>>>>
>>>>>       Ok. Thanks.
>>>>>>
>>>>>>> +
>>>>>>> +		/*
>>>>>>> +		 * If just in a layer a mask supports connect access,
>>>>>>> +		 * the socket_connect() hook with AF_UNSPEC family flag
>>>>>>> +		 * must be banned. This prevents from disconnecting already
>>>>>>> +		 * connected sockets.
>>>>>>> +		 */
>>>>>>> +		for (i = 0; i < dom->num_layers; i++) {
>>>>>>> +			if (landlock_get_net_access_mask(dom, i) &
>>>>>>> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
>>>>>>> +				return -EACCES;
>>>>>>
>>>>>> I'm wondering if this is the right error code for this case. EPERM may
>>>>>> be more appropriate.
>>>>>
>>>>>       Ok. Will be refactored.
>>>>>>
>>>>>> Thinking more about this case, I don't understand what is the rationale
>>>>>> to deny such action. What would be the consequence to always allow
>>>>>> connection with AF_UNSPEC (i.e. to disconnect a socket)?
>>>>>>
>>>>>       I thought we have come to a conclusion about connect(...AF_UNSPEC..)
>>>>>      behaviour in the patchset V3:
>>>>> https://lore.kernel.org/linux-security-module/19ad3a01-d76e-0e73-7833-99acd4afd97e@huawei.com/
>>>>
>>>> The conclusion was that AF_UNSPEC disconnects a socket, but I'm asking
>>>> if this is a security issue. I don't think it is more dangerous than a
>>>> new (unconnected) socket. Am I missing something? Which kind of rule
>>>> could be bypassed? What are we protecting against by restricting AF_UNSPEC?
>>>
>>> I just follow Willem de Bruijn concerns about this issue:
>>>
>>> quote: "It is valid to pass an address with AF_UNSPEC to a PF_INET(6)
>>> socket. And there are legitimate reasons to want to deny this. Such as
>>> passing a connection to a unprivileged process and disallow it from
>>> disconnect and opening a different new connection."
>>>
>>> https://lore.kernel.org/linux-security-module/CA+FuTSf4EjgjBCCOiu-PHJcTMia41UkTh8QJ0+qdxL_J8445EA@mail.gmail.com/
>>
>> I agree with the fact that we want to deny this, but in this example the
>> new connection should still be restricted by the Landlock domain. Using
>> AF_UNSPEC on a connected socket should not make this socket allowed to
>> create any connection if the process is restricted with TCP_CONNECT.
>> Being allowed to close a connection should not be an issue, and any new
>> connection must be vetted by Landlock.
>>
> 
>     You are right. This makes sense. Thanks for the comment.
>>>
>>>
>>> quote: "The intended use-case is for a privileged process to open a
>>> connection (i.e., bound and connected socket) and pass that to a
>>> restricted process. The intent is for that process to only be allowed to
>>> communicate over this pre-established channel.
>>>
>>> In practice, it is able to disconnect (while staying bound) and
>>> elevate its privileges to that of a listening server: ..."
>>>
>>> https://lore.kernel.org/linux-security-module/CA+FuTScaoby-=xRKf_Dz3koSYHqrMN0cauCg4jMmy_nDxwPADA@mail.gmail.com/
>>>
>>> Looks like it's a security issue here.
>>
>> It the provided example, if child_process() is restricted with
>> TCP_CONNECT and TCP_BIND, any call to connect() or bind() will return an
>> access error. listen() and accept() would work if the socket is bound,
>> which is the case here, and then implicitly allowed by the parent
>> process. I don' see any security issue. Am I missing something?
>>
>> In fact, connect with AF_UNSPEC should always be allowed to be
>> consistent with close(2), which is a way to drop privileges.
>>
> 
>    It should be allowed with checking:
> "return check_socket_access(dom, get_port(address),
>                                    LANDLOCK_ACCESS_NET_CONNECT_TCP);
>>
>> What Willem said:
>>> It would be good to also
>>> ensure that a now-bound socket cannot call listen.
>>
>> This is not relevant for Landlock because the security model is to check
>> process's requests to get new accesses (e.g. create a new file
>> descriptor), but not to check passed accesses (e.g. inherited from a
>> parent process, or pass through a unix socket) which are delegated to
>> the sender/parent. The goal of a sandbox is to limit the set of new
>> access requested (to the kernel) from within this sandbox. All already
>> opened file descriptors were previously vetted by Landlock (and other
>> access control systems).
> 
>      I got your point. Thanks.
>>
>>>
>>>>
>>>> We could then reduce the hook codes to just:
>>>> return current_check_access_socket(sock, address, LANDLOCK_ACCESS_NET_*);

This current_check_access_socket() helper should contain all the access 
control code.

>>>> .
>>
>> As for SELinux, the connect hook should first do this check (with an
>> appropriate comment):
>> if (address->sa_family == AF_UNSPEC)
>> 	return 0;
> 
>     In case of Landlock it looks like a landlocked process could connnect
> to the ports it's not allowed to connect to.
> So we need just to return check_socket_access(dom, get_port(address),
> 				   LANDLOCK_ACCESS_NET_CONNECT_TCP);
> I'm I correct? Did I miss something?

Using AF_UNSPEC with connect(2) doesn't connect the socket to a port, 
and in fact completely ignore the port. We can move the AF_UNSPEC check 
to the current_check_access_socket() helper:

  	switch (address->sa_family) {
  	case AF_UNSPEC:
+		/*
+		 * Connecting to an address with AF_UNSPEC dissolves the TCP
+		 * association, which have the same effect as closing the
+		 * connection while retaining the socket object (i.e., the file
+		 * descriptor).  As for dropping privileges, closing
+		 * connections is always allowed.
+		 */
+		if (access_request == LANDLOCK_ACCESS_NET_CONNECT_TCP)
+			return 0;
+
+		/*
+		 * For compatibility reason, accept AF_UNSPEC for bind
+		 * accesses (mapped to AF_INET) only if the address is
+		 * INADDR_ANY (cf. __inet_bind).  Checking the address is
+		 * required to not wrongfully return -EACCES instead of
+		 * -EAFNOSUPPORT.
+		 */
+		if (access_request == LANDLOCK_ACCESS_NET_BIND_TCP) {
+			const struct sockaddr_in *const sockaddr =
+				(struct sockaddr_in *)address;
+
+			if (sockaddr->sin_addr.s_addr != htonl(INADDR_ANY))
+				return -EAFNOSUPPORT;
+		}
+
+		fallthrough;
  	case AF_INET:
  #if IS_ENABLED(CONFIG_IPV6)
  	case AF_INET6:


I also added another check (copied from SELinux) with the appropriate 
explanation. All this needs dedicated tests to make sure everything is 
covered.

We also need to add extra checks (and related tests) for addrlen as do 
other LSMs.

Konstantin Meskhidze (A) Jan. 5, 2023, 8:57 a.m. UTC | #8

11/17/2022 9:43 PM, Mickaël Salaün пишет:
> 
> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>> This patch adds support of socket_bind() and socket_connect() hooks.
>> It's possible to restrict binding and connecting of TCP sockets to
>> particular ports.
> 
> Implement socket_bind() and socket_connect LSM hooks, which enable to
> restrict TCP socket binding and connection to specific ports.
> 
> 
>> 
>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>> ---
>> 
>> Changes since v7:
>> * Minor fixes.
>> * Refactors commit message.
>> 
>> Changes since v6:
>> * Updates copyright.
>> * Refactors landlock_append_net_rule() and check_socket_access()
>>    functions with landlock_id type.
>> 
>> Changes since v5:
>> * Fixes some logic errors.
>> * Formats code with clang-format-14.
>> 
>> Changes since v4:
>> * Factors out CONFIG_INET into make file.
>> * Refactors check_socket_access().
>> * Adds helper get_port().
>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>> functions to support AF_INET6 family.
>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>> functions.
>> * Refactors add_rule_net_service() and landlock_add_rule
>> syscall to support network rule inserting.
>> * Refactors init_layer_masks() to support network rules.
>> 
>> Changes since v3:
>> * Splits commit.
>> * Adds SECURITY_NETWORK in config.
>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>> * Adds hook_socket_bind and hook_socket_connect hooks.
>> 
>> ---
>>   security/landlock/Kconfig    |   1 +
>>   security/landlock/Makefile   |   2 +
>>   security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>   security/landlock/net.h      |  26 ++++++
>>   security/landlock/setup.c    |   2 +
>>   security/landlock/syscalls.c |  59 ++++++++++++-
>>   6 files changed, 251 insertions(+), 3 deletions(-)
>>   create mode 100644 security/landlock/net.c
>>   create mode 100644 security/landlock/net.h
>> 
>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>> index 8e33c4e8ffb8..10c099097533 100644
>> --- a/security/landlock/Kconfig
>> +++ b/security/landlock/Kconfig
>> @@ -3,6 +3,7 @@
>>   config SECURITY_LANDLOCK
>>   	bool "Landlock support"
>>   	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>> +	select SECURITY_NETWORK
>>   	select SECURITY_PATH
>>   	help
>>   	  Landlock is a sandboxing mechanism that enables processes to restrict
>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>> index 7bbd2f413b3e..53d3c92ae22e 100644
>> --- a/security/landlock/Makefile
>> +++ b/security/landlock/Makefile
>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>> 
>>   landlock-y := setup.o syscalls.o object.o ruleset.o \
>>   	cred.o ptrace.o fs.o
>> +
>> +landlock-$(CONFIG_INET) += net.o
>> \ No newline at end of file
>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>> new file mode 100644
>> index 000000000000..39e8a156a1f4
>> --- /dev/null
>> +++ b/security/landlock/net.c
>> @@ -0,0 +1,164 @@
>> +// SPDX-License-Identifier: GPL-2.0-only
>> +/*
>> + * Landlock LSM - Network management and hooks
>> + *
>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>> + * Copyright © 2022 Microsoft Corporation
>> + */
>> +
>> +#include <linux/in.h>
>> +#include <linux/net.h>
>> +#include <linux/socket.h>
>> +#include <net/ipv6.h>
>> +
>> +#include "common.h"
>> +#include "cred.h"
>> +#include "limits.h"
>> +#include "net.h"
>> +#include "ruleset.h"
>> +
>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>> +			     const u16 port, access_mask_t access_rights)
>> +{
>> +	int err;
>> +	const struct landlock_id id = {
>> +		.key.data = port,
>> +		.type = LANDLOCK_KEY_NET_PORT,
>> +	};
>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>> +
>> +	/* Transforms relative access rights to absolute ones. */
>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>> +
>> +	mutex_lock(&ruleset->lock);
>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>> +	mutex_unlock(&ruleset->lock);
>> +
>> +	return err;
>> +}
>> +
>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>> +			       u16 port, access_mask_t access_request)
>> +{
>> +	bool allowed = false;
>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>> +	const struct landlock_rule *rule;
>> +	access_mask_t handled_access;
>> +	const struct landlock_id id = {
>> +		.key.data = port,
>> +		.type = LANDLOCK_KEY_NET_PORT,
>> +	};
>> +
>> +	if (WARN_ON_ONCE(!domain))
>> +		return 0;
>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>> +		return -EACCES;
>> +
>> +	rule = landlock_find_rule(domain, id);
>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>> +					  LANDLOCK_KEY_NET_PORT);
>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>> +				ARRAY_SIZE(layer_masks));
>> +
>> +	return allowed ? 0 : -EACCES;
>> +}
>> +
>> +static u16 get_port(const struct sockaddr *const address)
> 
> get_port() should return a __be16 type. This enables to avoid converting
> port when checking a rule.

   In this case a user must do a coverting port into __be16:

   struct landlock_net_service_attr net_service = {
                 .allowed_access = LANDLOCK_ACCESS_NET_BIND_TCP,

                 .port = htons(sock_port),
         };
  I think that a user should not think about this conversion cause it 
makes UAPI more complex to use. Lets do this under kernel's hood and let 
it as it is now -> u16 port.

What do you think?


> 
> make C=2 security/landlock/ must not print any warning.
> 
> 
>> +{
>> +	/* Gets port value in host byte order. */
>> +	switch (address->sa_family) {
>> +	case AF_UNSPEC:
>> +	case AF_INET: {
>> +		const struct sockaddr_in *const sockaddr =
>> +			(struct sockaddr_in *)address;
>> +		return ntohs(sockaddr->sin_port);
>> +	}
>> +#if IS_ENABLED(CONFIG_IPV6)
>> +	case AF_INET6: {
>> +		const struct sockaddr_in6 *const sockaddr_ip6 =
>> +			(struct sockaddr_in6 *)address;
>> +		return ntohs(sockaddr_ip6->sin6_port);
>> +	}
>> +#endif
>> +	}
>> +	WARN_ON_ONCE(1);
>> +	return 0;
>> +}
>> +
>> +static int hook_socket_bind(struct socket *sock, struct sockaddr *address,
>> +			    int addrlen)
>> +{
>> +	const struct landlock_ruleset *const dom =
>> +		landlock_get_current_domain();
>> +
>> +	if (!dom)
>> +		return 0;
>> +
>> +	/* Check if it's a TCP socket. */
>> +	if (sock->type != SOCK_STREAM)
>> +		return 0;
>> +
>> +	switch (address->sa_family) {
>> +	case AF_UNSPEC:
>> +	case AF_INET:
>> +#if IS_ENABLED(CONFIG_IPV6)
>> +	case AF_INET6:
>> +#endif
>> +		return check_socket_access(dom, get_port(address),
>> +					   LANDLOCK_ACCESS_NET_BIND_TCP);
>> +	default:
>> +		return 0;
> 
> You can remove this default case and move the return 0 at the end of the
> function.
> 
> 
>> +	}
>> +}
>> +
>> +static int hook_socket_connect(struct socket *sock, struct sockaddr *address,
>> +			       int addrlen)
>> +{
>> +	const struct landlock_ruleset *const dom =
>> +		landlock_get_current_domain();
>> +
>> +	if (!dom)
>> +		return 0;
>> +
>> +	/* Check if it's a TCP socket. */
>> +	if (sock->type != SOCK_STREAM)
>> +		return 0;
>> +
>> +	/* Check if the hook is AF_INET* socket's action. */
>> +	switch (address->sa_family) {
>> +	case AF_INET:
>> +#if IS_ENABLED(CONFIG_IPV6)
>> +	case AF_INET6:
>> +#endif
>> +		return check_socket_access(dom, get_port(address),
>> +					   LANDLOCK_ACCESS_NET_CONNECT_TCP);
>> +	case AF_UNSPEC: {
>> +		u16 i;
> 
> You can move "i" after the "dom" declaration to remove the extra braces.
> 
> 
>> +
>> +		/*
>> +		 * If just in a layer a mask supports connect access,
>> +		 * the socket_connect() hook with AF_UNSPEC family flag
>> +		 * must be banned. This prevents from disconnecting already
>> +		 * connected sockets.
>> +		 */
>> +		for (i = 0; i < dom->num_layers; i++) {
>> +			if (landlock_get_net_access_mask(dom, i) &
>> +			    LANDLOCK_ACCESS_NET_CONNECT_TCP)
>> +				return -EACCES;
> 
> I'm wondering if this is the right error code for this case. EPERM may
> be more appropriate.
> 
> Thinking more about this case, I don't understand what is the rationale
> to deny such action. What would be the consequence to always allow
> connection with AF_UNSPEC (i.e. to disconnect a socket)?
> 
> 
>> +		}
>> +	}
>> +	}
>> +	return 0;
>> +}
>> +
>> +static struct security_hook_list landlock_hooks[] __lsm_ro_after_init = {
>> +	LSM_HOOK_INIT(socket_bind, hook_socket_bind),
>> +	LSM_HOOK_INIT(socket_connect, hook_socket_connect),
>> +};
>> +
>> +__init void landlock_add_net_hooks(void)
>> +{
>> +	security_add_hooks(landlock_hooks, ARRAY_SIZE(landlock_hooks),
>> +			   LANDLOCK_NAME);
>> +}
>> diff --git a/security/landlock/net.h b/security/landlock/net.h
>> new file mode 100644
>> index 000000000000..0da1d9dff5ab
>> --- /dev/null
>> +++ b/security/landlock/net.h
>> @@ -0,0 +1,26 @@
>> +/* SPDX-License-Identifier: GPL-2.0-only */
>> +/*
>> + * Landlock LSM - Network management and hooks
>> + *
>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>> + */
>> +
>> +#ifndef _SECURITY_LANDLOCK_NET_H
>> +#define _SECURITY_LANDLOCK_NET_H
>> +
>> +#include "common.h"
>> +#include "ruleset.h"
>> +#include "setup.h"
>> +
>> +#if IS_ENABLED(CONFIG_INET)
>> +__init void landlock_add_net_hooks(void);
>> +
>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>> +			     const u16 port, access_mask_t access_rights);
>> +#else /* IS_ENABLED(CONFIG_INET) */
>> +static inline void landlock_add_net_hooks(void)
>> +{
>> +}
>> +#endif /* IS_ENABLED(CONFIG_INET) */
>> +
>> +#endif /* _SECURITY_LANDLOCK_NET_H */
>> diff --git a/security/landlock/setup.c b/security/landlock/setup.c
>> index 3f196d2ce4f9..7e4a598177b8 100644
>> --- a/security/landlock/setup.c
>> +++ b/security/landlock/setup.c
>> @@ -14,6 +14,7 @@
>>   #include "fs.h"
>>   #include "ptrace.h"
>>   #include "setup.h"
>> +#include "net.h"
>> 
>>   bool landlock_initialized __lsm_ro_after_init = false;
>> 
>> @@ -29,6 +30,7 @@ static int __init landlock_init(void)
>>   	landlock_add_cred_hooks();
>>   	landlock_add_ptrace_hooks();
>>   	landlock_add_fs_hooks();
>> +	landlock_add_net_hooks();
>>   	landlock_initialized = true;
>>   	pr_info("Up and running.\n");
>>   	return 0;
>> diff --git a/security/landlock/syscalls.c b/security/landlock/syscalls.c
>> index c5a6ad4e2fca..7853f32e8325 100644
>> --- a/security/landlock/syscalls.c
>> +++ b/security/landlock/syscalls.c
>> @@ -29,6 +29,7 @@
>>   #include "cred.h"
>>   #include "fs.h"
>>   #include "limits.h"
>> +#include "net.h"
>>   #include "ruleset.h"
>>   #include "setup.h"
>> 
>> @@ -74,7 +75,8 @@ static void build_check_abi(void)
>>   {
>>   	struct landlock_ruleset_attr ruleset_attr;
>>   	struct landlock_path_beneath_attr path_beneath_attr;
>> -	size_t ruleset_size, path_beneath_size;
>> +	struct landlock_net_service_attr net_service_attr;
>> +	size_t ruleset_size, path_beneath_size, net_service_size;
>> 
>>   	/*
>>   	 * For each user space ABI structures, first checks that there is no
>> @@ -90,6 +92,11 @@ static void build_check_abi(void)
>>   	path_beneath_size += sizeof(path_beneath_attr.parent_fd);
>>   	BUILD_BUG_ON(sizeof(path_beneath_attr) != path_beneath_size);
>>   	BUILD_BUG_ON(sizeof(path_beneath_attr) != 12);
>> +
>> +	net_service_size = sizeof(net_service_attr.allowed_access);
>> +	net_service_size += sizeof(net_service_attr.port);
>> +	BUILD_BUG_ON(sizeof(net_service_attr) != net_service_size);
>> +	BUILD_BUG_ON(sizeof(net_service_attr) != 10);
>>   }
>> 
>>   /* Ruleset handling */
>> @@ -322,13 +329,54 @@ static int add_rule_path_beneath(struct landlock_ruleset *const ruleset,
>>   	return err;
>>   }
>> 
>> +static int add_rule_net_service(struct landlock_ruleset *ruleset,
>> +				const void __user *const rule_attr)
>> +{
>> +#if IS_ENABLED(CONFIG_INET)
>> +	struct landlock_net_service_attr net_service_attr;
>> +	int res;
>> +	u32 mask;
> 
> access_mask_t mask;
> 
> 
>> +
>> +	/* Copies raw user space buffer, only one type for now. */
>> +	res = copy_from_user(&net_service_attr, rule_attr,
>> +			     sizeof(net_service_attr));
>> +	if (res)
>> +		return -EFAULT;
>> +
>> +	/*
>> +	 * Informs about useless rule: empty allowed_access (i.e. deny rules)
>> +	 * are ignored by network actions.
>> +	 */
>> +	if (!net_service_attr.allowed_access)
>> +		return -ENOMSG;
>> +
>> +	/*
>> +	 * Checks that allowed_access matches the @ruleset constraints
>> +	 * (ruleset->access_masks[0] is automatically upgraded to 64-bits).
>> +	 */
>> +	mask = landlock_get_net_access_mask(ruleset, 0);
>> +	if ((net_service_attr.allowed_access | mask) != mask)
>> +		return -EINVAL;
>> +
>> +	/* Denies inserting a rule with port 0. */
>> +	if (net_service_attr.port == 0)
>> +		return -EINVAL;
>> +
>> +	/* Imports the new rule. */
>> +	return landlock_append_net_rule(ruleset, net_service_attr.port,
>> +					net_service_attr.allowed_access);
>> +#else /* IS_ENABLED(CONFIG_INET) */
>> +	return -EAFNOSUPPORT;
>> +#endif /* IS_ENABLED(CONFIG_INET) */
>> +}
>> +
>>   /**
>>    * sys_landlock_add_rule - Add a new rule to a ruleset
>>    *
>>    * @ruleset_fd: File descriptor tied to the ruleset that should be extended
>>    *		with the new rule.
>> - * @rule_type: Identify the structure type pointed to by @rule_attr (only
>> - *             %LANDLOCK_RULE_PATH_BENEATH for now).
>> + * @rule_type: Identify the structure type pointed to by @rule_attr:
>> + *             %LANDLOCK_RULE_PATH_BENEATH or %LANDLOCK_RULE_NET_SERVICE.
>>    * @rule_attr: Pointer to a rule (only of type &struct
>>    *             landlock_path_beneath_attr for now).
>>    * @flags: Must be 0.
>> @@ -339,6 +387,8 @@ static int add_rule_path_beneath(struct landlock_ruleset *const ruleset,
>>    * Possible returned errors are:
>>    *
>>    * - %EOPNOTSUPP: Landlock is supported by the kernel but disabled at boot time;
>> + * - %EAFNOSUPPORT: @rule_type is LANDLOCK_RULE_NET_SERVICE but TCP/IP is not
> 
> %LANDLOCK_RULE_NET_SERVICE
> 
> 
>> + *   supported by the running kernel;
>>    * - %EINVAL: @flags is not 0, or inconsistent access in the rule (i.e.
>>    *   &landlock_path_beneath_attr.allowed_access is not a subset of the
>>    *   ruleset handled accesses);
>> @@ -373,6 +423,9 @@ SYSCALL_DEFINE4(landlock_add_rule, const int, ruleset_fd,
>>   	case LANDLOCK_RULE_PATH_BENEATH:
>>   		err = add_rule_path_beneath(ruleset, rule_attr);
>>   		break;
>> +	case LANDLOCK_RULE_NET_SERVICE:
>> +		err = add_rule_net_service(ruleset, rule_attr);
>> +		break;
>>   	default:
>>   		err = -EINVAL;
>>   		break;
>> --
>> 2.25.1
>> 
> .

Mickaël Salaün Jan. 6, 2023, 7:30 p.m. UTC | #9

On 05/01/2023 09:57, Konstantin Meskhidze (A) wrote:
> 
> 
> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>
>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>> It's possible to restrict binding and connecting of TCP sockets to
>>> particular ports.
>>
>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>> restrict TCP socket binding and connection to specific ports.
>>
>>
>>>
>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>> ---
>>>
>>> Changes since v7:
>>> * Minor fixes.
>>> * Refactors commit message.
>>>
>>> Changes since v6:
>>> * Updates copyright.
>>> * Refactors landlock_append_net_rule() and check_socket_access()
>>>     functions with landlock_id type.
>>>
>>> Changes since v5:
>>> * Fixes some logic errors.
>>> * Formats code with clang-format-14.
>>>
>>> Changes since v4:
>>> * Factors out CONFIG_INET into make file.
>>> * Refactors check_socket_access().
>>> * Adds helper get_port().
>>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>>> functions to support AF_INET6 family.
>>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>>> functions.
>>> * Refactors add_rule_net_service() and landlock_add_rule
>>> syscall to support network rule inserting.
>>> * Refactors init_layer_masks() to support network rules.
>>>
>>> Changes since v3:
>>> * Splits commit.
>>> * Adds SECURITY_NETWORK in config.
>>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>>> * Adds hook_socket_bind and hook_socket_connect hooks.
>>>
>>> ---
>>>    security/landlock/Kconfig    |   1 +
>>>    security/landlock/Makefile   |   2 +
>>>    security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>>    security/landlock/net.h      |  26 ++++++
>>>    security/landlock/setup.c    |   2 +
>>>    security/landlock/syscalls.c |  59 ++++++++++++-
>>>    6 files changed, 251 insertions(+), 3 deletions(-)
>>>    create mode 100644 security/landlock/net.c
>>>    create mode 100644 security/landlock/net.h
>>>
>>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>>> index 8e33c4e8ffb8..10c099097533 100644
>>> --- a/security/landlock/Kconfig
>>> +++ b/security/landlock/Kconfig
>>> @@ -3,6 +3,7 @@
>>>    config SECURITY_LANDLOCK
>>>    	bool "Landlock support"
>>>    	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>>> +	select SECURITY_NETWORK
>>>    	select SECURITY_PATH
>>>    	help
>>>    	  Landlock is a sandboxing mechanism that enables processes to restrict
>>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>>> index 7bbd2f413b3e..53d3c92ae22e 100644
>>> --- a/security/landlock/Makefile
>>> +++ b/security/landlock/Makefile
>>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>>>
>>>    landlock-y := setup.o syscalls.o object.o ruleset.o \
>>>    	cred.o ptrace.o fs.o
>>> +
>>> +landlock-$(CONFIG_INET) += net.o
>>> \ No newline at end of file
>>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>>> new file mode 100644
>>> index 000000000000..39e8a156a1f4
>>> --- /dev/null
>>> +++ b/security/landlock/net.c
>>> @@ -0,0 +1,164 @@
>>> +// SPDX-License-Identifier: GPL-2.0-only
>>> +/*
>>> + * Landlock LSM - Network management and hooks
>>> + *
>>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>>> + * Copyright © 2022 Microsoft Corporation
>>> + */
>>> +
>>> +#include <linux/in.h>
>>> +#include <linux/net.h>
>>> +#include <linux/socket.h>
>>> +#include <net/ipv6.h>
>>> +
>>> +#include "common.h"
>>> +#include "cred.h"
>>> +#include "limits.h"
>>> +#include "net.h"
>>> +#include "ruleset.h"
>>> +
>>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>>> +			     const u16 port, access_mask_t access_rights)
>>> +{
>>> +	int err;
>>> +	const struct landlock_id id = {
>>> +		.key.data = port,
>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>> +	};
>>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>>> +
>>> +	/* Transforms relative access rights to absolute ones. */
>>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>>> +
>>> +	mutex_lock(&ruleset->lock);
>>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>>> +	mutex_unlock(&ruleset->lock);
>>> +
>>> +	return err;
>>> +}
>>> +
>>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>>> +			       u16 port, access_mask_t access_request)
>>> +{
>>> +	bool allowed = false;
>>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>>> +	const struct landlock_rule *rule;
>>> +	access_mask_t handled_access;
>>> +	const struct landlock_id id = {
>>> +		.key.data = port,
>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>> +	};
>>> +
>>> +	if (WARN_ON_ONCE(!domain))
>>> +		return 0;
>>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>>> +		return -EACCES;
>>> +
>>> +	rule = landlock_find_rule(domain, id);
>>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>>> +					  LANDLOCK_KEY_NET_PORT);
>>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>>> +				ARRAY_SIZE(layer_masks));
>>> +
>>> +	return allowed ? 0 : -EACCES;
>>> +}
>>> +
>>> +static u16 get_port(const struct sockaddr *const address)
>>
>> get_port() should return a __be16 type. This enables to avoid converting
>> port when checking a rule.
> 
>     In this case a user must do a coverting port into __be16:
> 
>     struct landlock_net_service_attr net_service = {
>                   .allowed_access = LANDLOCK_ACCESS_NET_BIND_TCP,
> 
>                   .port = htons(sock_port),
>           };
>    I think that a user should not think about this conversion cause it
> makes UAPI more complex to use. Lets do this under kernel's hood and let
> it as it is now -> u16 port.
> 
> What do you think?

BE and LE conversions may be error prone without strong typing, but the 
current Linux network UAPI uses this convention (see related syscalls), 
so developers already use htons() in their applications. I think it is 
less hazardous to use the same convention. It would be nice to have the 
point of view of network and API folks though.

Konstantin Meskhidze (A) Jan. 9, 2023, 8:07 a.m. UTC | #10

1/6/2023 10:30 PM, Mickaël Salaün пишет:
> 
> On 05/01/2023 09:57, Konstantin Meskhidze (A) wrote:
>> 
>> 
>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>
>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>> particular ports.
>>>
>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>> restrict TCP socket binding and connection to specific ports.
>>>
>>>
>>>>
>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>> ---
>>>>
>>>> Changes since v7:
>>>> * Minor fixes.
>>>> * Refactors commit message.
>>>>
>>>> Changes since v6:
>>>> * Updates copyright.
>>>> * Refactors landlock_append_net_rule() and check_socket_access()
>>>>     functions with landlock_id type.
>>>>
>>>> Changes since v5:
>>>> * Fixes some logic errors.
>>>> * Formats code with clang-format-14.
>>>>
>>>> Changes since v4:
>>>> * Factors out CONFIG_INET into make file.
>>>> * Refactors check_socket_access().
>>>> * Adds helper get_port().
>>>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>>>> functions to support AF_INET6 family.
>>>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>>>> functions.
>>>> * Refactors add_rule_net_service() and landlock_add_rule
>>>> syscall to support network rule inserting.
>>>> * Refactors init_layer_masks() to support network rules.
>>>>
>>>> Changes since v3:
>>>> * Splits commit.
>>>> * Adds SECURITY_NETWORK in config.
>>>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>>>> * Adds hook_socket_bind and hook_socket_connect hooks.
>>>>
>>>> ---
>>>>    security/landlock/Kconfig    |   1 +
>>>>    security/landlock/Makefile   |   2 +
>>>>    security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>>>    security/landlock/net.h      |  26 ++++++
>>>>    security/landlock/setup.c    |   2 +
>>>>    security/landlock/syscalls.c |  59 ++++++++++++-
>>>>    6 files changed, 251 insertions(+), 3 deletions(-)
>>>>    create mode 100644 security/landlock/net.c
>>>>    create mode 100644 security/landlock/net.h
>>>>
>>>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>>>> index 8e33c4e8ffb8..10c099097533 100644
>>>> --- a/security/landlock/Kconfig
>>>> +++ b/security/landlock/Kconfig
>>>> @@ -3,6 +3,7 @@
>>>>    config SECURITY_LANDLOCK
>>>>    	bool "Landlock support"
>>>>    	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>>>> +	select SECURITY_NETWORK
>>>>    	select SECURITY_PATH
>>>>    	help
>>>>    	  Landlock is a sandboxing mechanism that enables processes to restrict
>>>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>>>> index 7bbd2f413b3e..53d3c92ae22e 100644
>>>> --- a/security/landlock/Makefile
>>>> +++ b/security/landlock/Makefile
>>>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>>>>
>>>>    landlock-y := setup.o syscalls.o object.o ruleset.o \
>>>>    	cred.o ptrace.o fs.o
>>>> +
>>>> +landlock-$(CONFIG_INET) += net.o
>>>> \ No newline at end of file
>>>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>>>> new file mode 100644
>>>> index 000000000000..39e8a156a1f4
>>>> --- /dev/null
>>>> +++ b/security/landlock/net.c
>>>> @@ -0,0 +1,164 @@
>>>> +// SPDX-License-Identifier: GPL-2.0-only
>>>> +/*
>>>> + * Landlock LSM - Network management and hooks
>>>> + *
>>>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>>>> + * Copyright © 2022 Microsoft Corporation
>>>> + */
>>>> +
>>>> +#include <linux/in.h>
>>>> +#include <linux/net.h>
>>>> +#include <linux/socket.h>
>>>> +#include <net/ipv6.h>
>>>> +
>>>> +#include "common.h"
>>>> +#include "cred.h"
>>>> +#include "limits.h"
>>>> +#include "net.h"
>>>> +#include "ruleset.h"
>>>> +
>>>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>>>> +			     const u16 port, access_mask_t access_rights)
>>>> +{
>>>> +	int err;
>>>> +	const struct landlock_id id = {
>>>> +		.key.data = port,
>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>> +	};
>>>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>>>> +
>>>> +	/* Transforms relative access rights to absolute ones. */
>>>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>>>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>>>> +
>>>> +	mutex_lock(&ruleset->lock);
>>>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>>>> +	mutex_unlock(&ruleset->lock);
>>>> +
>>>> +	return err;
>>>> +}
>>>> +
>>>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>>>> +			       u16 port, access_mask_t access_request)
>>>> +{
>>>> +	bool allowed = false;
>>>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>>>> +	const struct landlock_rule *rule;
>>>> +	access_mask_t handled_access;
>>>> +	const struct landlock_id id = {
>>>> +		.key.data = port,
>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>> +	};
>>>> +
>>>> +	if (WARN_ON_ONCE(!domain))
>>>> +		return 0;
>>>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>>>> +		return -EACCES;
>>>> +
>>>> +	rule = landlock_find_rule(domain, id);
>>>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>>>> +					  LANDLOCK_KEY_NET_PORT);
>>>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>>>> +				ARRAY_SIZE(layer_masks));
>>>> +
>>>> +	return allowed ? 0 : -EACCES;
>>>> +}
>>>> +
>>>> +static u16 get_port(const struct sockaddr *const address)
>>>
>>> get_port() should return a __be16 type. This enables to avoid converting
>>> port when checking a rule.
>> 
>>     In this case a user must do a coverting port into __be16:
>> 
>>     struct landlock_net_service_attr net_service = {
>>                   .allowed_access = LANDLOCK_ACCESS_NET_BIND_TCP,
>> 
>>                   .port = htons(sock_port),
>>           };
>>    I think that a user should not think about this conversion cause it
>> makes UAPI more complex to use. Lets do this under kernel's hood and let
>> it as it is now -> u16 port.
>> 
>> What do you think?
> 
> BE and LE conversions may be error prone without strong typing, but the
> current Linux network UAPI uses this convention (see related syscalls),
> so developers already use htons() in their applications. I think it is
> less hazardous to use the same convention. It would be nice to have the
> point of view of network and API folks though.

   Ok. Thanks. Let ports be in BE format like in network packets.

   What should a selftest with port conversion be like?

   1. Set a port with a Landlock rule with no conversion. get an error 
wit bind/connect actions.
   2. Convert a port with htons(sock_port). get no error.

   What do you think?
> .

Mickaël Salaün Jan. 9, 2023, 12:38 p.m. UTC | #11

On 09/01/2023 09:07, Konstantin Meskhidze (A) wrote:
> 
> 
> 1/6/2023 10:30 PM, Mickaël Salaün пишет:
>>
>> On 05/01/2023 09:57, Konstantin Meskhidze (A) wrote:
>>>
>>>
>>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>>
>>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>>> particular ports.
>>>>
>>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>>> restrict TCP socket binding and connection to specific ports.
>>>>
>>>>
>>>>>
>>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>>> ---
>>>>>
>>>>> Changes since v7:
>>>>> * Minor fixes.
>>>>> * Refactors commit message.
>>>>>
>>>>> Changes since v6:
>>>>> * Updates copyright.
>>>>> * Refactors landlock_append_net_rule() and check_socket_access()
>>>>>      functions with landlock_id type.
>>>>>
>>>>> Changes since v5:
>>>>> * Fixes some logic errors.
>>>>> * Formats code with clang-format-14.
>>>>>
>>>>> Changes since v4:
>>>>> * Factors out CONFIG_INET into make file.
>>>>> * Refactors check_socket_access().
>>>>> * Adds helper get_port().
>>>>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>>>>> functions to support AF_INET6 family.
>>>>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>>>>> functions.
>>>>> * Refactors add_rule_net_service() and landlock_add_rule
>>>>> syscall to support network rule inserting.
>>>>> * Refactors init_layer_masks() to support network rules.
>>>>>
>>>>> Changes since v3:
>>>>> * Splits commit.
>>>>> * Adds SECURITY_NETWORK in config.
>>>>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>>>>> * Adds hook_socket_bind and hook_socket_connect hooks.
>>>>>
>>>>> ---
>>>>>     security/landlock/Kconfig    |   1 +
>>>>>     security/landlock/Makefile   |   2 +
>>>>>     security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>>>>     security/landlock/net.h      |  26 ++++++
>>>>>     security/landlock/setup.c    |   2 +
>>>>>     security/landlock/syscalls.c |  59 ++++++++++++-
>>>>>     6 files changed, 251 insertions(+), 3 deletions(-)
>>>>>     create mode 100644 security/landlock/net.c
>>>>>     create mode 100644 security/landlock/net.h
>>>>>
>>>>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>>>>> index 8e33c4e8ffb8..10c099097533 100644
>>>>> --- a/security/landlock/Kconfig
>>>>> +++ b/security/landlock/Kconfig
>>>>> @@ -3,6 +3,7 @@
>>>>>     config SECURITY_LANDLOCK
>>>>>     	bool "Landlock support"
>>>>>     	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>>>>> +	select SECURITY_NETWORK
>>>>>     	select SECURITY_PATH
>>>>>     	help
>>>>>     	  Landlock is a sandboxing mechanism that enables processes to restrict
>>>>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>>>>> index 7bbd2f413b3e..53d3c92ae22e 100644
>>>>> --- a/security/landlock/Makefile
>>>>> +++ b/security/landlock/Makefile
>>>>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>>>>>
>>>>>     landlock-y := setup.o syscalls.o object.o ruleset.o \
>>>>>     	cred.o ptrace.o fs.o
>>>>> +
>>>>> +landlock-$(CONFIG_INET) += net.o
>>>>> \ No newline at end of file
>>>>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>>>>> new file mode 100644
>>>>> index 000000000000..39e8a156a1f4
>>>>> --- /dev/null
>>>>> +++ b/security/landlock/net.c
>>>>> @@ -0,0 +1,164 @@
>>>>> +// SPDX-License-Identifier: GPL-2.0-only
>>>>> +/*
>>>>> + * Landlock LSM - Network management and hooks
>>>>> + *
>>>>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>>>>> + * Copyright © 2022 Microsoft Corporation
>>>>> + */
>>>>> +
>>>>> +#include <linux/in.h>
>>>>> +#include <linux/net.h>
>>>>> +#include <linux/socket.h>
>>>>> +#include <net/ipv6.h>
>>>>> +
>>>>> +#include "common.h"
>>>>> +#include "cred.h"
>>>>> +#include "limits.h"
>>>>> +#include "net.h"
>>>>> +#include "ruleset.h"
>>>>> +
>>>>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>>>>> +			     const u16 port, access_mask_t access_rights)
>>>>> +{
>>>>> +	int err;
>>>>> +	const struct landlock_id id = {
>>>>> +		.key.data = port,
>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>> +	};
>>>>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>>>>> +
>>>>> +	/* Transforms relative access rights to absolute ones. */
>>>>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>>>>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>>>>> +
>>>>> +	mutex_lock(&ruleset->lock);
>>>>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>>>>> +	mutex_unlock(&ruleset->lock);
>>>>> +
>>>>> +	return err;
>>>>> +}
>>>>> +
>>>>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>>>>> +			       u16 port, access_mask_t access_request)
>>>>> +{
>>>>> +	bool allowed = false;
>>>>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>>>>> +	const struct landlock_rule *rule;
>>>>> +	access_mask_t handled_access;
>>>>> +	const struct landlock_id id = {
>>>>> +		.key.data = port,
>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>> +	};
>>>>> +
>>>>> +	if (WARN_ON_ONCE(!domain))
>>>>> +		return 0;
>>>>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>>>>> +		return -EACCES;
>>>>> +
>>>>> +	rule = landlock_find_rule(domain, id);
>>>>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>>>>> +					  LANDLOCK_KEY_NET_PORT);
>>>>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>>>>> +				ARRAY_SIZE(layer_masks));
>>>>> +
>>>>> +	return allowed ? 0 : -EACCES;
>>>>> +}
>>>>> +
>>>>> +static u16 get_port(const struct sockaddr *const address)
>>>>
>>>> get_port() should return a __be16 type. This enables to avoid converting
>>>> port when checking a rule.
>>>
>>>      In this case a user must do a coverting port into __be16:
>>>
>>>      struct landlock_net_service_attr net_service = {
>>>                    .allowed_access = LANDLOCK_ACCESS_NET_BIND_TCP,
>>>
>>>                    .port = htons(sock_port),
>>>            };
>>>     I think that a user should not think about this conversion cause it
>>> makes UAPI more complex to use. Lets do this under kernel's hood and let
>>> it as it is now -> u16 port.
>>>
>>> What do you think?
>>
>> BE and LE conversions may be error prone without strong typing, but the
>> current Linux network UAPI uses this convention (see related syscalls),
>> so developers already use htons() in their applications. I think it is
>> less hazardous to use the same convention. It would be nice to have the
>> point of view of network and API folks though.
> 
>     Ok. Thanks. Let ports be in BE format like in network packets.
> 
>     What should a selftest with port conversion be like?
> 
>     1. Set a port with a Landlock rule with no conversion. get an error
> wit bind/connect actions.
>     2. Convert a port with htons(sock_port). get no error.
> 
>     What do you think?

Right, you can do both on a LE architecture (that must be checked in the 
test or it should be skipped), test with a port value that has different 
representation in LE and BE.

Konstantin Meskhidze (A) Jan. 10, 2023, 4:45 a.m. UTC | #12

1/9/2023 3:38 PM, Mickaël Salaün пишет:
> 
> On 09/01/2023 09:07, Konstantin Meskhidze (A) wrote:
>> 
>> 
>> 1/6/2023 10:30 PM, Mickaël Salaün пишет:
>>>
>>> On 05/01/2023 09:57, Konstantin Meskhidze (A) wrote:
>>>>
>>>>
>>>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>>>
>>>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>>>> particular ports.
>>>>>
>>>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>>>> restrict TCP socket binding and connection to specific ports.
>>>>>
>>>>>
>>>>>>
>>>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>>>> ---
>>>>>>
>>>>>> Changes since v7:
>>>>>> * Minor fixes.
>>>>>> * Refactors commit message.
>>>>>>
>>>>>> Changes since v6:
>>>>>> * Updates copyright.
>>>>>> * Refactors landlock_append_net_rule() and check_socket_access()
>>>>>>      functions with landlock_id type.
>>>>>>
>>>>>> Changes since v5:
>>>>>> * Fixes some logic errors.
>>>>>> * Formats code with clang-format-14.
>>>>>>
>>>>>> Changes since v4:
>>>>>> * Factors out CONFIG_INET into make file.
>>>>>> * Refactors check_socket_access().
>>>>>> * Adds helper get_port().
>>>>>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>>>>>> functions to support AF_INET6 family.
>>>>>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>>>>>> functions.
>>>>>> * Refactors add_rule_net_service() and landlock_add_rule
>>>>>> syscall to support network rule inserting.
>>>>>> * Refactors init_layer_masks() to support network rules.
>>>>>>
>>>>>> Changes since v3:
>>>>>> * Splits commit.
>>>>>> * Adds SECURITY_NETWORK in config.
>>>>>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>>>>>> * Adds hook_socket_bind and hook_socket_connect hooks.
>>>>>>
>>>>>> ---
>>>>>>     security/landlock/Kconfig    |   1 +
>>>>>>     security/landlock/Makefile   |   2 +
>>>>>>     security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>>>>>     security/landlock/net.h      |  26 ++++++
>>>>>>     security/landlock/setup.c    |   2 +
>>>>>>     security/landlock/syscalls.c |  59 ++++++++++++-
>>>>>>     6 files changed, 251 insertions(+), 3 deletions(-)
>>>>>>     create mode 100644 security/landlock/net.c
>>>>>>     create mode 100644 security/landlock/net.h
>>>>>>
>>>>>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>>>>>> index 8e33c4e8ffb8..10c099097533 100644
>>>>>> --- a/security/landlock/Kconfig
>>>>>> +++ b/security/landlock/Kconfig
>>>>>> @@ -3,6 +3,7 @@
>>>>>>     config SECURITY_LANDLOCK
>>>>>>     	bool "Landlock support"
>>>>>>     	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>>>>>> +	select SECURITY_NETWORK
>>>>>>     	select SECURITY_PATH
>>>>>>     	help
>>>>>>     	  Landlock is a sandboxing mechanism that enables processes to restrict
>>>>>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>>>>>> index 7bbd2f413b3e..53d3c92ae22e 100644
>>>>>> --- a/security/landlock/Makefile
>>>>>> +++ b/security/landlock/Makefile
>>>>>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>>>>>>
>>>>>>     landlock-y := setup.o syscalls.o object.o ruleset.o \
>>>>>>     	cred.o ptrace.o fs.o
>>>>>> +
>>>>>> +landlock-$(CONFIG_INET) += net.o
>>>>>> \ No newline at end of file
>>>>>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>>>>>> new file mode 100644
>>>>>> index 000000000000..39e8a156a1f4
>>>>>> --- /dev/null
>>>>>> +++ b/security/landlock/net.c
>>>>>> @@ -0,0 +1,164 @@
>>>>>> +// SPDX-License-Identifier: GPL-2.0-only
>>>>>> +/*
>>>>>> + * Landlock LSM - Network management and hooks
>>>>>> + *
>>>>>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>>>>>> + * Copyright © 2022 Microsoft Corporation
>>>>>> + */
>>>>>> +
>>>>>> +#include <linux/in.h>
>>>>>> +#include <linux/net.h>
>>>>>> +#include <linux/socket.h>
>>>>>> +#include <net/ipv6.h>
>>>>>> +
>>>>>> +#include "common.h"
>>>>>> +#include "cred.h"
>>>>>> +#include "limits.h"
>>>>>> +#include "net.h"
>>>>>> +#include "ruleset.h"
>>>>>> +
>>>>>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>>>>>> +			     const u16 port, access_mask_t access_rights)
>>>>>> +{
>>>>>> +	int err;
>>>>>> +	const struct landlock_id id = {
>>>>>> +		.key.data = port,
>>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>>> +	};
>>>>>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>>>>>> +
>>>>>> +	/* Transforms relative access rights to absolute ones. */
>>>>>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>>>>>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>>>>>> +
>>>>>> +	mutex_lock(&ruleset->lock);
>>>>>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>>>>>> +	mutex_unlock(&ruleset->lock);
>>>>>> +
>>>>>> +	return err;
>>>>>> +}
>>>>>> +
>>>>>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>>>>>> +			       u16 port, access_mask_t access_request)
>>>>>> +{
>>>>>> +	bool allowed = false;
>>>>>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>>>>>> +	const struct landlock_rule *rule;
>>>>>> +	access_mask_t handled_access;
>>>>>> +	const struct landlock_id id = {
>>>>>> +		.key.data = port,
>>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>>> +	};
>>>>>> +
>>>>>> +	if (WARN_ON_ONCE(!domain))
>>>>>> +		return 0;
>>>>>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>>>>>> +		return -EACCES;
>>>>>> +
>>>>>> +	rule = landlock_find_rule(domain, id);
>>>>>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>>>>>> +					  LANDLOCK_KEY_NET_PORT);
>>>>>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>>>>>> +				ARRAY_SIZE(layer_masks));
>>>>>> +
>>>>>> +	return allowed ? 0 : -EACCES;
>>>>>> +}
>>>>>> +
>>>>>> +static u16 get_port(const struct sockaddr *const address)
>>>>>
>>>>> get_port() should return a __be16 type. This enables to avoid converting
>>>>> port when checking a rule.
>>>>
>>>>      In this case a user must do a coverting port into __be16:
>>>>
>>>>      struct landlock_net_service_attr net_service = {
>>>>                    .allowed_access = LANDLOCK_ACCESS_NET_BIND_TCP,
>>>>
>>>>                    .port = htons(sock_port),
>>>>            };
>>>>     I think that a user should not think about this conversion cause it
>>>> makes UAPI more complex to use. Lets do this under kernel's hood and let
>>>> it as it is now -> u16 port.
>>>>
>>>> What do you think?
>>>
>>> BE and LE conversions may be error prone without strong typing, but the
>>> current Linux network UAPI uses this convention (see related syscalls),
>>> so developers already use htons() in their applications. I think it is
>>> less hazardous to use the same convention. It would be nice to have the
>>> point of view of network and API folks though.
>> 
>>     Ok. Thanks. Let ports be in BE format like in network packets.
>> 
>>     What should a selftest with port conversion be like?
>> 
>>     1. Set a port with a Landlock rule with no conversion. get an error
>> wit bind/connect actions.
>>     2. Convert a port with htons(sock_port). get no error.
>> 
>>     What do you think?
> 
> Right, you can do both on a LE architecture (that must be checked in the
> test or it should be skipped), test with a port value that has different
> representation in LE and BE.

   Do you mean to check architecture in a test first and then port 
representaton? What about BE architectures? My current VM is X86-64 
architecture a LE one. I can test just it now.
> .

Mickaël Salaün Jan. 10, 2023, 5:24 p.m. UTC | #13

On 10/01/2023 05:45, Konstantin Meskhidze (A) wrote:
> 
> 
> 1/9/2023 3:38 PM, Mickaël Salaün пишет:
>>
>> On 09/01/2023 09:07, Konstantin Meskhidze (A) wrote:
>>>
>>>
>>> 1/6/2023 10:30 PM, Mickaël Salaün пишет:
>>>>
>>>> On 05/01/2023 09:57, Konstantin Meskhidze (A) wrote:
>>>>>
>>>>>
>>>>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>>>>
>>>>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>>>>> particular ports.
>>>>>>
>>>>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>>>>> restrict TCP socket binding and connection to specific ports.
>>>>>>
>>>>>>
>>>>>>>
>>>>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>>>>> ---
>>>>>>>
>>>>>>> Changes since v7:
>>>>>>> * Minor fixes.
>>>>>>> * Refactors commit message.
>>>>>>>
>>>>>>> Changes since v6:
>>>>>>> * Updates copyright.
>>>>>>> * Refactors landlock_append_net_rule() and check_socket_access()
>>>>>>>       functions with landlock_id type.
>>>>>>>
>>>>>>> Changes since v5:
>>>>>>> * Fixes some logic errors.
>>>>>>> * Formats code with clang-format-14.
>>>>>>>
>>>>>>> Changes since v4:
>>>>>>> * Factors out CONFIG_INET into make file.
>>>>>>> * Refactors check_socket_access().
>>>>>>> * Adds helper get_port().
>>>>>>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>>>>>>> functions to support AF_INET6 family.
>>>>>>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>>>>>>> functions.
>>>>>>> * Refactors add_rule_net_service() and landlock_add_rule
>>>>>>> syscall to support network rule inserting.
>>>>>>> * Refactors init_layer_masks() to support network rules.
>>>>>>>
>>>>>>> Changes since v3:
>>>>>>> * Splits commit.
>>>>>>> * Adds SECURITY_NETWORK in config.
>>>>>>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>>>>>>> * Adds hook_socket_bind and hook_socket_connect hooks.
>>>>>>>
>>>>>>> ---
>>>>>>>      security/landlock/Kconfig    |   1 +
>>>>>>>      security/landlock/Makefile   |   2 +
>>>>>>>      security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>>>>>>      security/landlock/net.h      |  26 ++++++
>>>>>>>      security/landlock/setup.c    |   2 +
>>>>>>>      security/landlock/syscalls.c |  59 ++++++++++++-
>>>>>>>      6 files changed, 251 insertions(+), 3 deletions(-)
>>>>>>>      create mode 100644 security/landlock/net.c
>>>>>>>      create mode 100644 security/landlock/net.h
>>>>>>>
>>>>>>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>>>>>>> index 8e33c4e8ffb8..10c099097533 100644
>>>>>>> --- a/security/landlock/Kconfig
>>>>>>> +++ b/security/landlock/Kconfig
>>>>>>> @@ -3,6 +3,7 @@
>>>>>>>      config SECURITY_LANDLOCK
>>>>>>>      	bool "Landlock support"
>>>>>>>      	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>>>>>>> +	select SECURITY_NETWORK
>>>>>>>      	select SECURITY_PATH
>>>>>>>      	help
>>>>>>>      	  Landlock is a sandboxing mechanism that enables processes to restrict
>>>>>>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>>>>>>> index 7bbd2f413b3e..53d3c92ae22e 100644
>>>>>>> --- a/security/landlock/Makefile
>>>>>>> +++ b/security/landlock/Makefile
>>>>>>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>>>>>>>
>>>>>>>      landlock-y := setup.o syscalls.o object.o ruleset.o \
>>>>>>>      	cred.o ptrace.o fs.o
>>>>>>> +
>>>>>>> +landlock-$(CONFIG_INET) += net.o
>>>>>>> \ No newline at end of file
>>>>>>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>>>>>>> new file mode 100644
>>>>>>> index 000000000000..39e8a156a1f4
>>>>>>> --- /dev/null
>>>>>>> +++ b/security/landlock/net.c
>>>>>>> @@ -0,0 +1,164 @@
>>>>>>> +// SPDX-License-Identifier: GPL-2.0-only
>>>>>>> +/*
>>>>>>> + * Landlock LSM - Network management and hooks
>>>>>>> + *
>>>>>>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>>>>>>> + * Copyright © 2022 Microsoft Corporation
>>>>>>> + */
>>>>>>> +
>>>>>>> +#include <linux/in.h>
>>>>>>> +#include <linux/net.h>
>>>>>>> +#include <linux/socket.h>
>>>>>>> +#include <net/ipv6.h>
>>>>>>> +
>>>>>>> +#include "common.h"
>>>>>>> +#include "cred.h"
>>>>>>> +#include "limits.h"
>>>>>>> +#include "net.h"
>>>>>>> +#include "ruleset.h"
>>>>>>> +
>>>>>>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>>>>>>> +			     const u16 port, access_mask_t access_rights)
>>>>>>> +{
>>>>>>> +	int err;
>>>>>>> +	const struct landlock_id id = {
>>>>>>> +		.key.data = port,
>>>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>>>> +	};
>>>>>>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>>>>>>> +
>>>>>>> +	/* Transforms relative access rights to absolute ones. */
>>>>>>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>>>>>>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>>>>>>> +
>>>>>>> +	mutex_lock(&ruleset->lock);
>>>>>>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>>>>>>> +	mutex_unlock(&ruleset->lock);
>>>>>>> +
>>>>>>> +	return err;
>>>>>>> +}
>>>>>>> +
>>>>>>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>>>>>>> +			       u16 port, access_mask_t access_request)
>>>>>>> +{
>>>>>>> +	bool allowed = false;
>>>>>>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>>>>>>> +	const struct landlock_rule *rule;
>>>>>>> +	access_mask_t handled_access;
>>>>>>> +	const struct landlock_id id = {
>>>>>>> +		.key.data = port,
>>>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>>>> +	};
>>>>>>> +
>>>>>>> +	if (WARN_ON_ONCE(!domain))
>>>>>>> +		return 0;
>>>>>>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>>>>>>> +		return -EACCES;
>>>>>>> +
>>>>>>> +	rule = landlock_find_rule(domain, id);
>>>>>>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>>>>>>> +					  LANDLOCK_KEY_NET_PORT);
>>>>>>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>>>>>>> +				ARRAY_SIZE(layer_masks));
>>>>>>> +
>>>>>>> +	return allowed ? 0 : -EACCES;
>>>>>>> +}
>>>>>>> +
>>>>>>> +static u16 get_port(const struct sockaddr *const address)
>>>>>>
>>>>>> get_port() should return a __be16 type. This enables to avoid converting
>>>>>> port when checking a rule.
>>>>>
>>>>>       In this case a user must do a coverting port into __be16:
>>>>>
>>>>>       struct landlock_net_service_attr net_service = {
>>>>>                     .allowed_access = LANDLOCK_ACCESS_NET_BIND_TCP,
>>>>>
>>>>>                     .port = htons(sock_port),
>>>>>             };
>>>>>      I think that a user should not think about this conversion cause it
>>>>> makes UAPI more complex to use. Lets do this under kernel's hood and let
>>>>> it as it is now -> u16 port.
>>>>>
>>>>> What do you think?
>>>>
>>>> BE and LE conversions may be error prone without strong typing, but the
>>>> current Linux network UAPI uses this convention (see related syscalls),
>>>> so developers already use htons() in their applications. I think it is
>>>> less hazardous to use the same convention. It would be nice to have the
>>>> point of view of network and API folks though.
>>>
>>>      Ok. Thanks. Let ports be in BE format like in network packets.
>>>
>>>      What should a selftest with port conversion be like?
>>>
>>>      1. Set a port with a Landlock rule with no conversion. get an error
>>> wit bind/connect actions.
>>>      2. Convert a port with htons(sock_port). get no error.
>>>
>>>      What do you think?
>>
>> Right, you can do both on a LE architecture (that must be checked in the
>> test or it should be skipped), test with a port value that has different
>> representation in LE and BE.
> 
>     Do you mean to check architecture in a test first and then port
> representaton? What about BE architectures? My current VM is X86-64
> architecture a LE one. I can test just it now.

It's just that tests should pass whatever architecture they are run on. 
So we need to check that the current architecture is LE to check against 
an LE result but not against a BE one, and vice versa. In fact no test 
should be skipped, just the result to compare with adjusted (i.e. either 
it pass or it failed).

Konstantin Meskhidze (A) Jan. 11, 2023, 1:54 a.m. UTC | #14

1/10/2023 8:24 PM, Mickaël Salaün пишет:
> 
> On 10/01/2023 05:45, Konstantin Meskhidze (A) wrote:
>> 
>> 
>> 1/9/2023 3:38 PM, Mickaël Salaün пишет:
>>>
>>> On 09/01/2023 09:07, Konstantin Meskhidze (A) wrote:
>>>>
>>>>
>>>> 1/6/2023 10:30 PM, Mickaël Salaün пишет:
>>>>>
>>>>> On 05/01/2023 09:57, Konstantin Meskhidze (A) wrote:
>>>>>>
>>>>>>
>>>>>> 11/17/2022 9:43 PM, Mickaël Salaün пишет:
>>>>>>>
>>>>>>> On 21/10/2022 17:26, Konstantin Meskhidze wrote:
>>>>>>>> This patch adds support of socket_bind() and socket_connect() hooks.
>>>>>>>> It's possible to restrict binding and connecting of TCP sockets to
>>>>>>>> particular ports.
>>>>>>>
>>>>>>> Implement socket_bind() and socket_connect LSM hooks, which enable to
>>>>>>> restrict TCP socket binding and connection to specific ports.
>>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>> Signed-off-by: Konstantin Meskhidze <konstantin.meskhidze@huawei.com>
>>>>>>>> ---
>>>>>>>>
>>>>>>>> Changes since v7:
>>>>>>>> * Minor fixes.
>>>>>>>> * Refactors commit message.
>>>>>>>>
>>>>>>>> Changes since v6:
>>>>>>>> * Updates copyright.
>>>>>>>> * Refactors landlock_append_net_rule() and check_socket_access()
>>>>>>>>       functions with landlock_id type.
>>>>>>>>
>>>>>>>> Changes since v5:
>>>>>>>> * Fixes some logic errors.
>>>>>>>> * Formats code with clang-format-14.
>>>>>>>>
>>>>>>>> Changes since v4:
>>>>>>>> * Factors out CONFIG_INET into make file.
>>>>>>>> * Refactors check_socket_access().
>>>>>>>> * Adds helper get_port().
>>>>>>>> * Adds CONFIG_IPV6 in get_port(), hook_socket_bind/connect
>>>>>>>> functions to support AF_INET6 family.
>>>>>>>> * Adds AF_UNSPEC family support in hook_socket_bind/connect
>>>>>>>> functions.
>>>>>>>> * Refactors add_rule_net_service() and landlock_add_rule
>>>>>>>> syscall to support network rule inserting.
>>>>>>>> * Refactors init_layer_masks() to support network rules.
>>>>>>>>
>>>>>>>> Changes since v3:
>>>>>>>> * Splits commit.
>>>>>>>> * Adds SECURITY_NETWORK in config.
>>>>>>>> * Adds IS_ENABLED(CONFIG_INET) if a kernel has no INET configuration.
>>>>>>>> * Adds hook_socket_bind and hook_socket_connect hooks.
>>>>>>>>
>>>>>>>> ---
>>>>>>>>      security/landlock/Kconfig    |   1 +
>>>>>>>>      security/landlock/Makefile   |   2 +
>>>>>>>>      security/landlock/net.c      | 164 +++++++++++++++++++++++++++++++++++
>>>>>>>>      security/landlock/net.h      |  26 ++++++
>>>>>>>>      security/landlock/setup.c    |   2 +
>>>>>>>>      security/landlock/syscalls.c |  59 ++++++++++++-
>>>>>>>>      6 files changed, 251 insertions(+), 3 deletions(-)
>>>>>>>>      create mode 100644 security/landlock/net.c
>>>>>>>>      create mode 100644 security/landlock/net.h
>>>>>>>>
>>>>>>>> diff --git a/security/landlock/Kconfig b/security/landlock/Kconfig
>>>>>>>> index 8e33c4e8ffb8..10c099097533 100644
>>>>>>>> --- a/security/landlock/Kconfig
>>>>>>>> +++ b/security/landlock/Kconfig
>>>>>>>> @@ -3,6 +3,7 @@
>>>>>>>>      config SECURITY_LANDLOCK
>>>>>>>>      	bool "Landlock support"
>>>>>>>>      	depends on SECURITY && !ARCH_EPHEMERAL_INODES
>>>>>>>> +	select SECURITY_NETWORK
>>>>>>>>      	select SECURITY_PATH
>>>>>>>>      	help
>>>>>>>>      	  Landlock is a sandboxing mechanism that enables processes to restrict
>>>>>>>> diff --git a/security/landlock/Makefile b/security/landlock/Makefile
>>>>>>>> index 7bbd2f413b3e..53d3c92ae22e 100644
>>>>>>>> --- a/security/landlock/Makefile
>>>>>>>> +++ b/security/landlock/Makefile
>>>>>>>> @@ -2,3 +2,5 @@ obj-$(CONFIG_SECURITY_LANDLOCK) := landlock.o
>>>>>>>>
>>>>>>>>      landlock-y := setup.o syscalls.o object.o ruleset.o \
>>>>>>>>      	cred.o ptrace.o fs.o
>>>>>>>> +
>>>>>>>> +landlock-$(CONFIG_INET) += net.o
>>>>>>>> \ No newline at end of file
>>>>>>>> diff --git a/security/landlock/net.c b/security/landlock/net.c
>>>>>>>> new file mode 100644
>>>>>>>> index 000000000000..39e8a156a1f4
>>>>>>>> --- /dev/null
>>>>>>>> +++ b/security/landlock/net.c
>>>>>>>> @@ -0,0 +1,164 @@
>>>>>>>> +// SPDX-License-Identifier: GPL-2.0-only
>>>>>>>> +/*
>>>>>>>> + * Landlock LSM - Network management and hooks
>>>>>>>> + *
>>>>>>>> + * Copyright © 2022 Huawei Tech. Co., Ltd.
>>>>>>>> + * Copyright © 2022 Microsoft Corporation
>>>>>>>> + */
>>>>>>>> +
>>>>>>>> +#include <linux/in.h>
>>>>>>>> +#include <linux/net.h>
>>>>>>>> +#include <linux/socket.h>
>>>>>>>> +#include <net/ipv6.h>
>>>>>>>> +
>>>>>>>> +#include "common.h"
>>>>>>>> +#include "cred.h"
>>>>>>>> +#include "limits.h"
>>>>>>>> +#include "net.h"
>>>>>>>> +#include "ruleset.h"
>>>>>>>> +
>>>>>>>> +int landlock_append_net_rule(struct landlock_ruleset *const ruleset,
>>>>>>>> +			     const u16 port, access_mask_t access_rights)
>>>>>>>> +{
>>>>>>>> +	int err;
>>>>>>>> +	const struct landlock_id id = {
>>>>>>>> +		.key.data = port,
>>>>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>>>>> +	};
>>>>>>>> +	BUILD_BUG_ON(sizeof(port) > sizeof(id.key.data));
>>>>>>>> +
>>>>>>>> +	/* Transforms relative access rights to absolute ones. */
>>>>>>>> +	access_rights |= LANDLOCK_MASK_ACCESS_NET &
>>>>>>>> +			 ~landlock_get_net_access_mask(ruleset, 0);
>>>>>>>> +
>>>>>>>> +	mutex_lock(&ruleset->lock);
>>>>>>>> +	err = landlock_insert_rule(ruleset, id, access_rights);
>>>>>>>> +	mutex_unlock(&ruleset->lock);
>>>>>>>> +
>>>>>>>> +	return err;
>>>>>>>> +}
>>>>>>>> +
>>>>>>>> +static int check_socket_access(const struct landlock_ruleset *const domain,
>>>>>>>> +			       u16 port, access_mask_t access_request)
>>>>>>>> +{
>>>>>>>> +	bool allowed = false;
>>>>>>>> +	layer_mask_t layer_masks[LANDLOCK_NUM_ACCESS_NET] = {};
>>>>>>>> +	const struct landlock_rule *rule;
>>>>>>>> +	access_mask_t handled_access;
>>>>>>>> +	const struct landlock_id id = {
>>>>>>>> +		.key.data = port,
>>>>>>>> +		.type = LANDLOCK_KEY_NET_PORT,
>>>>>>>> +	};
>>>>>>>> +
>>>>>>>> +	if (WARN_ON_ONCE(!domain))
>>>>>>>> +		return 0;
>>>>>>>> +	if (WARN_ON_ONCE(domain->num_layers < 1))
>>>>>>>> +		return -EACCES;
>>>>>>>> +
>>>>>>>> +	rule = landlock_find_rule(domain, id);
>>>>>>>> +	handled_access = init_layer_masks(domain, access_request, &layer_masks,
>>>>>>>> +					  LANDLOCK_KEY_NET_PORT);
>>>>>>>> +	allowed = unmask_layers(rule, handled_access, &layer_masks,
>>>>>>>> +				ARRAY_SIZE(layer_masks));
>>>>>>>> +
>>>>>>>> +	return allowed ? 0 : -EACCES;
>>>>>>>> +}
>>>>>>>> +
>>>>>>>> +static u16 get_port(const struct sockaddr *const address)
>>>>>>>
>>>>>>> get_port() should return a __be16 type. This enables to avoid converting
>>>>>>> port when checking a rule.
>>>>>>
>>>>>>       In this case a user must do a coverting port into __be16:
>>>>>>
>>>>>>       struct landlock_net_service_attr net_service = {
>>>>>>                     .allowed_access = LANDLOCK_ACCESS_NET_BIND_TCP,
>>>>>>
>>>>>>                     .port = htons(sock_port),
>>>>>>             };
>>>>>>      I think that a user should not think about this conversion cause it
>>>>>> makes UAPI more complex to use. Lets do this under kernel's hood and let
>>>>>> it as it is now -> u16 port.
>>>>>>
>>>>>> What do you think?
>>>>>
>>>>> BE and LE conversions may be error prone without strong typing, but the
>>>>> current Linux network UAPI uses this convention (see related syscalls),
>>>>> so developers already use htons() in their applications. I think it is
>>>>> less hazardous to use the same convention. It would be nice to have the
>>>>> point of view of network and API folks though.
>>>>
>>>>      Ok. Thanks. Let ports be in BE format like in network packets.
>>>>
>>>>      What should a selftest with port conversion be like?
>>>>
>>>>      1. Set a port with a Landlock rule with no conversion. get an error
>>>> wit bind/connect actions.
>>>>      2. Convert a port with htons(sock_port). get no error.
>>>>
>>>>      What do you think?
>>>
>>> Right, you can do both on a LE architecture (that must be checked in the
>>> test or it should be skipped), test with a port value that has different
>>> representation in LE and BE.
>> 
>>     Do you mean to check architecture in a test first and then port
>> representaton? What about BE architectures? My current VM is X86-64
>> architecture a LE one. I can test just it now.
> 
> It's just that tests should pass whatever architecture they are run on.
> So we need to check that the current architecture is LE to check against
> an LE result but not against a BE one, and vice versa. In fact no test
> should be skipped, just the result to compare with adjusted (i.e. either
> it pass or it failed).

   Ok. Thanks.
> .

[v8,08/12] landlock: Implement TCP network hooks

Checks

Commit Message

Comments

Patch