Loading...
1#
2# Security configuration
3#
4
5menu "Security options"
6
7config KEYS
8 bool "Enable access key retention support"
9 help
10 This option provides support for retaining authentication tokens and
11 access keys in the kernel.
12
13 It also includes provision of methods by which such keys might be
14 associated with a process so that network filesystems, encryption
15 support and the like can find them.
16
17 Furthermore, a special type of key is available that acts as keyring:
18 a searchable sequence of keys. Each process is equipped with access
19 to five standard keyrings: UID-specific, GID-specific, session,
20 process and thread.
21
22 If you are unsure as to whether this is required, answer N.
23
24config TRUSTED_KEYS
25 tristate "TRUSTED KEYS"
26 depends on KEYS && TCG_TPM
27 select CRYPTO
28 select CRYPTO_HMAC
29 select CRYPTO_SHA1
30 help
31 This option provides support for creating, sealing, and unsealing
32 keys in the kernel. Trusted keys are random number symmetric keys,
33 generated and RSA-sealed by the TPM. The TPM only unseals the keys,
34 if the boot PCRs and other criteria match. Userspace will only ever
35 see encrypted blobs.
36
37 If you are unsure as to whether this is required, answer N.
38
39config ENCRYPTED_KEYS
40 tristate "ENCRYPTED KEYS"
41 depends on KEYS && TRUSTED_KEYS
42 select CRYPTO_AES
43 select CRYPTO_CBC
44 select CRYPTO_SHA256
45 select CRYPTO_RNG
46 help
47 This option provides support for create/encrypting/decrypting keys
48 in the kernel. Encrypted keys are kernel generated random numbers,
49 which are encrypted/decrypted with a 'master' symmetric key. The
50 'master' key can be either a trusted-key or user-key type.
51 Userspace only ever sees/stores encrypted blobs.
52
53 If you are unsure as to whether this is required, answer N.
54
55config KEYS_DEBUG_PROC_KEYS
56 bool "Enable the /proc/keys file by which keys may be viewed"
57 depends on KEYS
58 help
59 This option turns on support for the /proc/keys file - through which
60 can be listed all the keys on the system that are viewable by the
61 reading process.
62
63 The only keys included in the list are those that grant View
64 permission to the reading process whether or not it possesses them.
65 Note that LSM security checks are still performed, and may further
66 filter out keys that the current process is not authorised to view.
67
68 Only key attributes are listed here; key payloads are not included in
69 the resulting table.
70
71 If you are unsure as to whether this is required, answer N.
72
73config SECURITY_DMESG_RESTRICT
74 bool "Restrict unprivileged access to the kernel syslog"
75 default n
76 help
77 This enforces restrictions on unprivileged users reading the kernel
78 syslog via dmesg(8).
79
80 If this option is not selected, no restrictions will be enforced
81 unless the dmesg_restrict sysctl is explicitly set to (1).
82
83 If you are unsure how to answer this question, answer N.
84
85config SECURITY
86 bool "Enable different security models"
87 depends on SYSFS
88 help
89 This allows you to choose different security modules to be
90 configured into your kernel.
91
92 If this option is not selected, the default Linux security
93 model will be used.
94
95 If you are unsure how to answer this question, answer N.
96
97config SECURITYFS
98 bool "Enable the securityfs filesystem"
99 help
100 This will build the securityfs filesystem. It is currently used by
101 the TPM bios character driver and IMA, an integrity provider. It is
102 not used by SELinux or SMACK.
103
104 If you are unsure how to answer this question, answer N.
105
106config SECURITY_NETWORK
107 bool "Socket and Networking Security Hooks"
108 depends on SECURITY
109 help
110 This enables the socket and networking security hooks.
111 If enabled, a security module can use these hooks to
112 implement socket and networking access controls.
113 If you are unsure how to answer this question, answer N.
114
115config SECURITY_NETWORK_XFRM
116 bool "XFRM (IPSec) Networking Security Hooks"
117 depends on XFRM && SECURITY_NETWORK
118 help
119 This enables the XFRM (IPSec) networking security hooks.
120 If enabled, a security module can use these hooks to
121 implement per-packet access controls based on labels
122 derived from IPSec policy. Non-IPSec communications are
123 designated as unlabelled, and only sockets authorized
124 to communicate unlabelled data can send without using
125 IPSec.
126 If you are unsure how to answer this question, answer N.
127
128config SECURITY_PATH
129 bool "Security hooks for pathname based access control"
130 depends on SECURITY
131 help
132 This enables the security hooks for pathname based access control.
133 If enabled, a security module can use these hooks to
134 implement pathname based access controls.
135 If you are unsure how to answer this question, answer N.
136
137config INTEL_TXT
138 bool "Enable Intel(R) Trusted Execution Technology (Intel(R) TXT)"
139 depends on HAVE_INTEL_TXT
140 help
141 This option enables support for booting the kernel with the
142 Trusted Boot (tboot) module. This will utilize
143 Intel(R) Trusted Execution Technology to perform a measured launch
144 of the kernel. If the system does not support Intel(R) TXT, this
145 will have no effect.
146
147 Intel TXT will provide higher assurance of system configuration and
148 initial state as well as data reset protection. This is used to
149 create a robust initial kernel measurement and verification, which
150 helps to ensure that kernel security mechanisms are functioning
151 correctly. This level of protection requires a root of trust outside
152 of the kernel itself.
153
154 Intel TXT also helps solve real end user concerns about having
155 confidence that their hardware is running the VMM or kernel that
156 it was configured with, especially since they may be responsible for
157 providing such assurances to VMs and services running on it.
158
159 See <http://www.intel.com/technology/security/> for more information
160 about Intel(R) TXT.
161 See <http://tboot.sourceforge.net> for more information about tboot.
162 See Documentation/intel_txt.txt for a description of how to enable
163 Intel TXT support in a kernel boot.
164
165 If you are unsure as to whether this is required, answer N.
166
167config LSM_MMAP_MIN_ADDR
168 int "Low address space for LSM to protect from user allocation"
169 depends on SECURITY && SECURITY_SELINUX
170 default 32768 if ARM
171 default 65536
172 help
173 This is the portion of low virtual memory which should be protected
174 from userspace allocation. Keeping a user from writing to low pages
175 can help reduce the impact of kernel NULL pointer bugs.
176
177 For most ia64, ppc64 and x86 users with lots of address space
178 a value of 65536 is reasonable and should cause no problems.
179 On arm and other archs it should not be higher than 32768.
180 Programs which use vm86 functionality or have some need to map
181 this low address space will need the permission specific to the
182 systems running LSM.
183
184source security/selinux/Kconfig
185source security/smack/Kconfig
186source security/tomoyo/Kconfig
187source security/apparmor/Kconfig
188
189source security/integrity/ima/Kconfig
190
191choice
192 prompt "Default security module"
193 default DEFAULT_SECURITY_SELINUX if SECURITY_SELINUX
194 default DEFAULT_SECURITY_SMACK if SECURITY_SMACK
195 default DEFAULT_SECURITY_TOMOYO if SECURITY_TOMOYO
196 default DEFAULT_SECURITY_APPARMOR if SECURITY_APPARMOR
197 default DEFAULT_SECURITY_DAC
198
199 help
200 Select the security module that will be used by default if the
201 kernel parameter security= is not specified.
202
203 config DEFAULT_SECURITY_SELINUX
204 bool "SELinux" if SECURITY_SELINUX=y
205
206 config DEFAULT_SECURITY_SMACK
207 bool "Simplified Mandatory Access Control" if SECURITY_SMACK=y
208
209 config DEFAULT_SECURITY_TOMOYO
210 bool "TOMOYO" if SECURITY_TOMOYO=y
211
212 config DEFAULT_SECURITY_APPARMOR
213 bool "AppArmor" if SECURITY_APPARMOR=y
214
215 config DEFAULT_SECURITY_DAC
216 bool "Unix Discretionary Access Controls"
217
218endchoice
219
220config DEFAULT_SECURITY
221 string
222 default "selinux" if DEFAULT_SECURITY_SELINUX
223 default "smack" if DEFAULT_SECURITY_SMACK
224 default "tomoyo" if DEFAULT_SECURITY_TOMOYO
225 default "apparmor" if DEFAULT_SECURITY_APPARMOR
226 default "" if DEFAULT_SECURITY_DAC
227
228endmenu
229
1# SPDX-License-Identifier: GPL-2.0-only
2#
3# Security configuration
4#
5
6menu "Security options"
7
8source "security/keys/Kconfig"
9
10config SECURITY_DMESG_RESTRICT
11 bool "Restrict unprivileged access to the kernel syslog"
12 default n
13 help
14 This enforces restrictions on unprivileged users reading the kernel
15 syslog via dmesg(8).
16
17 If this option is not selected, no restrictions will be enforced
18 unless the dmesg_restrict sysctl is explicitly set to (1).
19
20 If you are unsure how to answer this question, answer N.
21
22choice
23 prompt "Allow /proc/pid/mem access override"
24 default PROC_MEM_ALWAYS_FORCE
25 help
26 Traditionally /proc/pid/mem allows users to override memory
27 permissions for users like ptrace, assuming they have ptrace
28 capability.
29
30 This allows people to limit that - either never override, or
31 require actual active ptrace attachment.
32
33 Defaults to the traditional behavior (for now)
34
35config PROC_MEM_ALWAYS_FORCE
36 bool "Traditional /proc/pid/mem behavior"
37 help
38 This allows /proc/pid/mem accesses to override memory mapping
39 permissions if you have ptrace access rights.
40
41config PROC_MEM_FORCE_PTRACE
42 bool "Require active ptrace() use for access override"
43 help
44 This allows /proc/pid/mem accesses to override memory mapping
45 permissions for active ptracers like gdb.
46
47config PROC_MEM_NO_FORCE
48 bool "Never"
49 help
50 Never override memory mapping permissions
51
52endchoice
53
54config SECURITY
55 bool "Enable different security models"
56 depends on SYSFS
57 depends on MULTIUSER
58 help
59 This allows you to choose different security modules to be
60 configured into your kernel.
61
62 If this option is not selected, the default Linux security
63 model will be used.
64
65 If you are unsure how to answer this question, answer N.
66
67config SECURITYFS
68 bool "Enable the securityfs filesystem"
69 help
70 This will build the securityfs filesystem. It is currently used by
71 various security modules (AppArmor, IMA, SafeSetID, TOMOYO, TPM).
72
73 If you are unsure how to answer this question, answer N.
74
75config SECURITY_NETWORK
76 bool "Socket and Networking Security Hooks"
77 depends on SECURITY
78 help
79 This enables the socket and networking security hooks.
80 If enabled, a security module can use these hooks to
81 implement socket and networking access controls.
82 If you are unsure how to answer this question, answer N.
83
84config SECURITY_INFINIBAND
85 bool "Infiniband Security Hooks"
86 depends on SECURITY && INFINIBAND
87 help
88 This enables the Infiniband security hooks.
89 If enabled, a security module can use these hooks to
90 implement Infiniband access controls.
91 If you are unsure how to answer this question, answer N.
92
93config SECURITY_NETWORK_XFRM
94 bool "XFRM (IPSec) Networking Security Hooks"
95 depends on XFRM && SECURITY_NETWORK
96 help
97 This enables the XFRM (IPSec) networking security hooks.
98 If enabled, a security module can use these hooks to
99 implement per-packet access controls based on labels
100 derived from IPSec policy. Non-IPSec communications are
101 designated as unlabelled, and only sockets authorized
102 to communicate unlabelled data can send without using
103 IPSec.
104 If you are unsure how to answer this question, answer N.
105
106config SECURITY_PATH
107 bool "Security hooks for pathname based access control"
108 depends on SECURITY
109 help
110 This enables the security hooks for pathname based access control.
111 If enabled, a security module can use these hooks to
112 implement pathname based access controls.
113 If you are unsure how to answer this question, answer N.
114
115config INTEL_TXT
116 bool "Enable Intel(R) Trusted Execution Technology (Intel(R) TXT)"
117 depends on HAVE_INTEL_TXT
118 help
119 This option enables support for booting the kernel with the
120 Trusted Boot (tboot) module. This will utilize
121 Intel(R) Trusted Execution Technology to perform a measured launch
122 of the kernel. If the system does not support Intel(R) TXT, this
123 will have no effect.
124
125 Intel TXT will provide higher assurance of system configuration and
126 initial state as well as data reset protection. This is used to
127 create a robust initial kernel measurement and verification, which
128 helps to ensure that kernel security mechanisms are functioning
129 correctly. This level of protection requires a root of trust outside
130 of the kernel itself.
131
132 Intel TXT also helps solve real end user concerns about having
133 confidence that their hardware is running the VMM or kernel that
134 it was configured with, especially since they may be responsible for
135 providing such assurances to VMs and services running on it.
136
137 See <https://www.intel.com/technology/security/> for more information
138 about Intel(R) TXT.
139 See <http://tboot.sourceforge.net> for more information about tboot.
140 See Documentation/arch/x86/intel_txt.rst for a description of how to enable
141 Intel TXT support in a kernel boot.
142
143 If you are unsure as to whether this is required, answer N.
144
145config LSM_MMAP_MIN_ADDR
146 int "Low address space for LSM to protect from user allocation"
147 depends on SECURITY && SECURITY_SELINUX
148 default 32768 if ARM || (ARM64 && COMPAT)
149 default 65536
150 help
151 This is the portion of low virtual memory which should be protected
152 from userspace allocation. Keeping a user from writing to low pages
153 can help reduce the impact of kernel NULL pointer bugs.
154
155 For most ia64, ppc64 and x86 users with lots of address space
156 a value of 65536 is reasonable and should cause no problems.
157 On arm and other archs it should not be higher than 32768.
158 Programs which use vm86 functionality or have some need to map
159 this low address space will need the permission specific to the
160 systems running LSM.
161
162config HARDENED_USERCOPY
163 bool "Harden memory copies between kernel and userspace"
164 imply STRICT_DEVMEM
165 help
166 This option checks for obviously wrong memory regions when
167 copying memory to/from the kernel (via copy_to_user() and
168 copy_from_user() functions) by rejecting memory ranges that
169 are larger than the specified heap object, span multiple
170 separately allocated pages, are not on the process stack,
171 or are part of the kernel text. This prevents entire classes
172 of heap overflow exploits and similar kernel memory exposures.
173
174config FORTIFY_SOURCE
175 bool "Harden common str/mem functions against buffer overflows"
176 depends on ARCH_HAS_FORTIFY_SOURCE
177 # https://github.com/llvm/llvm-project/issues/53645
178 depends on !CC_IS_CLANG || !X86_32
179 help
180 Detect overflows of buffers in common string and memory functions
181 where the compiler can determine and validate the buffer sizes.
182
183config STATIC_USERMODEHELPER
184 bool "Force all usermode helper calls through a single binary"
185 help
186 By default, the kernel can call many different userspace
187 binary programs through the "usermode helper" kernel
188 interface. Some of these binaries are statically defined
189 either in the kernel code itself, or as a kernel configuration
190 option. However, some of these are dynamically created at
191 runtime, or can be modified after the kernel has started up.
192 To provide an additional layer of security, route all of these
193 calls through a single executable that can not have its name
194 changed.
195
196 Note, it is up to this single binary to then call the relevant
197 "real" usermode helper binary, based on the first argument
198 passed to it. If desired, this program can filter and pick
199 and choose what real programs are called.
200
201 If you wish for all usermode helper programs are to be
202 disabled, choose this option and then set
203 STATIC_USERMODEHELPER_PATH to an empty string.
204
205config STATIC_USERMODEHELPER_PATH
206 string "Path to the static usermode helper binary"
207 depends on STATIC_USERMODEHELPER
208 default "/sbin/usermode-helper"
209 help
210 The binary called by the kernel when any usermode helper
211 program is wish to be run. The "real" application's name will
212 be in the first argument passed to this program on the command
213 line.
214
215 If you wish for all usermode helper programs to be disabled,
216 specify an empty string here (i.e. "").
217
218source "security/selinux/Kconfig"
219source "security/smack/Kconfig"
220source "security/tomoyo/Kconfig"
221source "security/apparmor/Kconfig"
222source "security/loadpin/Kconfig"
223source "security/yama/Kconfig"
224source "security/safesetid/Kconfig"
225source "security/lockdown/Kconfig"
226source "security/landlock/Kconfig"
227source "security/ipe/Kconfig"
228
229source "security/integrity/Kconfig"
230
231choice
232 prompt "First legacy 'major LSM' to be initialized"
233 default DEFAULT_SECURITY_SELINUX if SECURITY_SELINUX
234 default DEFAULT_SECURITY_SMACK if SECURITY_SMACK
235 default DEFAULT_SECURITY_TOMOYO if SECURITY_TOMOYO
236 default DEFAULT_SECURITY_APPARMOR if SECURITY_APPARMOR
237 default DEFAULT_SECURITY_DAC
238
239 help
240 This choice is there only for converting CONFIG_DEFAULT_SECURITY
241 in old kernel configs to CONFIG_LSM in new kernel configs. Don't
242 change this choice unless you are creating a fresh kernel config,
243 for this choice will be ignored after CONFIG_LSM has been set.
244
245 Selects the legacy "major security module" that will be
246 initialized first. Overridden by non-default CONFIG_LSM.
247
248 config DEFAULT_SECURITY_SELINUX
249 bool "SELinux" if SECURITY_SELINUX=y
250
251 config DEFAULT_SECURITY_SMACK
252 bool "Simplified Mandatory Access Control" if SECURITY_SMACK=y
253
254 config DEFAULT_SECURITY_TOMOYO
255 bool "TOMOYO" if SECURITY_TOMOYO=y
256
257 config DEFAULT_SECURITY_APPARMOR
258 bool "AppArmor" if SECURITY_APPARMOR=y
259
260 config DEFAULT_SECURITY_DAC
261 bool "Unix Discretionary Access Controls"
262
263endchoice
264
265config LSM
266 string "Ordered list of enabled LSMs"
267 default "landlock,lockdown,yama,loadpin,safesetid,smack,selinux,tomoyo,apparmor,ipe,bpf" if DEFAULT_SECURITY_SMACK
268 default "landlock,lockdown,yama,loadpin,safesetid,apparmor,selinux,smack,tomoyo,ipe,bpf" if DEFAULT_SECURITY_APPARMOR
269 default "landlock,lockdown,yama,loadpin,safesetid,tomoyo,ipe,bpf" if DEFAULT_SECURITY_TOMOYO
270 default "landlock,lockdown,yama,loadpin,safesetid,ipe,bpf" if DEFAULT_SECURITY_DAC
271 default "landlock,lockdown,yama,loadpin,safesetid,selinux,smack,tomoyo,apparmor,ipe,bpf"
272 help
273 A comma-separated list of LSMs, in initialization order.
274 Any LSMs left off this list, except for those with order
275 LSM_ORDER_FIRST and LSM_ORDER_LAST, which are always enabled
276 if selected in the kernel configuration, will be ignored.
277 This can be controlled at boot with the "lsm=" parameter.
278
279 If unsure, leave this as the default.
280
281source "security/Kconfig.hardening"
282
283endmenu
284