__GFP_WAIT was used to signal that the caller was in atomic context and
could not sleep. Now it is possible to distinguish between true atomic
context and callers that are not willing to sleep. The latter should
clear __GFP_DIRECT_RECLAIM so kswapd will still wake. As clearing
__GFP_WAIT behaves differently, there is a risk that people will clear the
wrong flags. This patch renames __GFP_WAIT to __GFP_RECLAIM to clearly
indicate what it does -- setting it allows all reclaim activity, clearing
them prevents it.
[akpm@linux-foundation.org: fix build]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Christoph Lameter <cl@linux.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Vitaly Wool <vitalywool@gmail.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull security subsystem update from James Morris:
"This is mostly maintenance updates across the subsystem, with a
notable update for TPM 2.0, and addition of Jarkko Sakkinen as a
maintainer of that"
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (40 commits)
apparmor: clarify CRYPTO dependency
selinux: Use a kmem_cache for allocation struct file_security_struct
selinux: ioctl_has_perm should be static
selinux: use sprintf return value
selinux: use kstrdup() in security_get_bools()
selinux: use kmemdup in security_sid_to_context_core()
selinux: remove pointless cast in selinux_inode_setsecurity()
selinux: introduce security_context_str_to_sid
selinux: do not check open perm on ftruncate call
selinux: change CONFIG_SECURITY_SELINUX_CHECKREQPROT_VALUE default
KEYS: Merge the type-specific data with the payload data
KEYS: Provide a script to extract a module signature
KEYS: Provide a script to extract the sys cert list from a vmlinux file
keys: Be more consistent in selection of union members used
certs: add .gitignore to stop git nagging about x509_certificate_list
KEYS: use kvfree() in add_key
Smack: limited capability for changing process label
TPM: remove unnecessary little endian conversion
vTPM: support little endian guests
char: Drop owner assignment from i2c_driver
...
The crypto framework can be built as a loadable module, but the
apparmor hash code can only be built-in, which then causes a
link error:
security/built-in.o: In function `aa_calc_profile_hash':
integrity_audit.c:(.text+0x21610): undefined reference to `crypto_shash_update'
security/built-in.o: In function `init_profile_hash':
integrity_audit.c:(.init.text+0xb4c): undefined reference to `crypto_alloc_shash'
This changes Apparmor to use 'select CRYPTO' like a lot of other
subsystems do.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: John Johansen <john.johansen@canonical.com>
Signed-off-by: James Morris <james.l.morris@oracle.com>
The size of struct file_security_struct is 16byte at my setup.
But, the real allocation size for per each file_security_struct
is 64bytes in my setup that kmalloc min size is 64bytes
because ARCH_DMA_MINALIGN is 64.
This allocation is called every times at file allocation(alloc_file()).
So, the total slack memory size(allocated size - request size)
is increased exponentially.
E.g) Min Kmalloc Size : 64bytes, Unit : bytes
Allocated Size | Request Size | Slack Size | Allocation Count
---------------------------------------------------------------
770048 | 192512 | 577536 | 12032
At the result, this change reduce memory usage 42bytes per each
file_security_struct
Signed-off-by: Sangwoo <sangwoo2.park@lge.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
[PM: removed extra subject prefix]
Signed-off-by: Paul Moore <pmoore@redhat.com>
Fixes the following sparse warning:
security/selinux/hooks.c:3242:5: warning: symbol 'ioctl_has_perm' was
not declared. Should it be static?
Signed-off-by: Geliang Tang <geliangtang@163.com>
Acked-by: Jeff Vander Stoep <jeffv@google.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Paul Moore <pmoore@redhat.com>
sprintf returns the number of characters printed (excluding '\0'), so
we can use that and avoid duplicating the length computation.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Paul Moore <pmoore@redhat.com>
This is much simpler.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Paul Moore <pmoore@redhat.com>
security_context_to_sid() expects a const char* argument, so there's
no point in casting away the const qualifier of value.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Paul Moore <pmoore@redhat.com>
There seems to be a little confusion as to whether the scontext_len
parameter of security_context_to_sid() includes the nul-byte or
not. Reading security_context_to_sid_core(), it seems that the
expectation is that it does not (both the string copying and the test
for scontext_len being zero hint at that).
Introduce the helper security_context_str_to_sid() to do the strlen()
call and fix all callers.
Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Paul Moore <pmoore@redhat.com>
Use the ATTR_FILE attribute to distinguish between truncate()
and ftruncate() system calls. The two other cases where
do_truncate is called with a filp (and therefore ATTR_FILE is set)
are for coredump files and for open(O_TRUNC). In both of those cases
the open permission has already been checked during file open and
therefore does not need to be repeated.
Commit 95dbf73931 ("SELinux: check OPEN on truncate calls")
fixed a major issue where domains were allowed to truncate files
without the open permission. However, it introduced a new bug where
a domain with the write permission can no longer ftruncate files
without the open permission, even when they receive an already open
file.
Signed-off-by: Jeff Vander Stoep <jeffv@google.com>
Acked-by: Stephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: Paul Moore <pmoore@redhat.com>
Change the SELinux checkreqprot default value to 0 so that SELinux
performs access control checking on the actual memory protections
used by the kernel and not those requested by the application.
Signed-off-by: Paul Moore <pmoore@redhat.com>
key->description and key->index_key.description are same because
they are unioned. But, for readability, using same name for
duplication and validation seems better.
Signed-off-by: Insu Yun <wuninsu@gmail.com>
Signed-off-by: David Howells <dhowells@redhat.com>
There is no need to make a flag to tell that this memory is allocated by
kmalloc or vmalloc. Just use kvfree to free the memory.
Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: David Howells <dhowells@redhat.com>
This feature introduces new kernel interface:
- <smack_fs>/relabel-self - for setting transition labels list
This list is used to control smack label transition mechanism.
List is set by, and per process. Process can transit to new label only if
label is on the list. Only process with CAP_MAC_ADMIN capability can add
labels to this list. With this list, process can change it's label without
CAP_MAC_ADMIN but only once. After label changing, list is unset.
Changes in v2:
* use list_for_each_entry instead of _rcu during label write
* added missing description in security/Smack.txt
Changes in v3:
* squashed into one commit
Changes in v4:
* switch from global list to per-task list
* since the per-task list is accessed only by the task itself
there is no need to use synchronization mechanisms on it
Changes in v5:
* change smackfs interface of relabel-self to the one used for onlycap
multiple labels are accepted, separated by space, which
replace the previous list upon write
Signed-off-by: Zbigniew Jasinski <z.jasinski@samsung.com>
Signed-off-by: Rafal Krypa <r.krypa@samsung.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
If request_key() is used to find a keyring, only do the search part - don't
do the construction part if the keyring was not found by the search. We
don't really want keyrings in the negative instantiated state since the
rejected/negative instantiation error value in the payload is unioned with
keyring metadata.
Now the kernel gives an error:
request_key("keyring", "#selinux,bdekeyring", "keyring", KEY_SPEC_USER_SESSION_KEYRING) = -1 EPERM (Operation not permitted)
Signed-off-by: David Howells <dhowells@redhat.com>
Call tpm_seal_trusted() and tpm_unseal_trusted() for TPM 2.0 chips.
We require explicit 'keyhandle=' option because there's no a fixed
storage root key inside TPM2 chips.
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Reviewed-by: Andreas Fuchs <andreas.fuchs@sit.fraunhofer.de>
Tested-by: Mimi Zohar <zohar@linux.vnet.ibm.com> (on TPM 1.2)
Tested-by: Chris J Arges <chris.j.arges@canonical.com>
Tested-by: Colin Ian King <colin.king@canonical.com>
Tested-by: Kevin Strasser <kevin.strasser@intel.com>
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
Moved struct trusted_key_options to trustes-type.h so that the fields
can be accessed from drivers/char/tpm.
Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Signed-off-by: Peter Huewe <peterhuewe@gmx.de>
This merge resolves conflicts with 75aec9df3a ("bridge: Remove
br_nf_push_frag_xmit_sk") as part of Eric Biederman's effort to improve
netns support in the network stack that reached upstream via David's
net-next tree.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Conflicts:
net/bridge/br_netfilter_hooks.c
since commit 8405a8fff3 ("netfilter: nf_qeueue: Drop queue entries on
nf_unregister_hook") all pending queued entries are discarded.
So we can simply remove all of the owner handling -- when module is
removed it also needs to unregister all its hooks.
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
The following sequence of commands:
i=`keyctl add user a a @s`
keyctl request2 keyring foo bar @t
keyctl unlink $i @s
tries to invoke an upcall to instantiate a keyring if one doesn't already
exist by that name within the user's keyring set. However, if the upcall
fails, the code sets keyring->type_data.reject_error to -ENOKEY or some
other error code. When the key is garbage collected, the key destroy
function is called unconditionally and keyring_destroy() uses list_empty()
on keyring->type_data.link - which is in a union with reject_error.
Subsequently, the kernel tries to unlink the keyring from the keyring names
list - which oopses like this:
BUG: unable to handle kernel paging request at 00000000ffffff8a
IP: [<ffffffff8126e051>] keyring_destroy+0x3d/0x88
...
Workqueue: events key_garbage_collector
...
RIP: 0010:[<ffffffff8126e051>] keyring_destroy+0x3d/0x88
RSP: 0018:ffff88003e2f3d30 EFLAGS: 00010203
RAX: 00000000ffffff82 RBX: ffff88003bf1a900 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 000000003bfc6901 RDI: ffffffff81a73a40
RBP: ffff88003e2f3d38 R08: 0000000000000152 R09: 0000000000000000
R10: ffff88003e2f3c18 R11: 000000000000865b R12: ffff88003bf1a900
R13: 0000000000000000 R14: ffff88003bf1a908 R15: ffff88003e2f4000
...
CR2: 00000000ffffff8a CR3: 000000003e3ec000 CR4: 00000000000006f0
...
Call Trace:
[<ffffffff8126c756>] key_gc_unused_keys.constprop.1+0x5d/0x10f
[<ffffffff8126ca71>] key_garbage_collector+0x1fa/0x351
[<ffffffff8105ec9b>] process_one_work+0x28e/0x547
[<ffffffff8105fd17>] worker_thread+0x26e/0x361
[<ffffffff8105faa9>] ? rescuer_thread+0x2a8/0x2a8
[<ffffffff810648ad>] kthread+0xf3/0xfb
[<ffffffff810647ba>] ? kthread_create_on_node+0x1c2/0x1c2
[<ffffffff815f2ccf>] ret_from_fork+0x3f/0x70
[<ffffffff810647ba>] ? kthread_create_on_node+0x1c2/0x1c2
Note the value in RAX. This is a 32-bit representation of -ENOKEY.
The solution is to only call ->destroy() if the key was successfully
instantiated.
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Tested-by: Dmitry Vyukov <dvyukov@google.com>
selinux needs few changes to accommodate fact that SYNACK messages
can be attached to a request socket, lacking sk_security pointer
(Only syncookies are still attached to a TCP_LISTEN socket)
Adds a new sk_listener() helper, and use it in selinux and sch_fq
Fixes: ca6fb06518 ("tcp: attach SYNACK messages to request sockets instead of listener")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported by: kernel test robot <ying.huang@linux.intel.com>
Cc: Paul Moore <paul@paul-moore.com>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Cc: Eric Paris <eparis@parisplace.org>
Acked-by: Paul Moore <paul@paul-moore.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This fix writes the task label when
smack_d_instantiate is called, before the
label of the superblock was written on the
pipe's inode.
Signed-off-by: Roman Kubiak <r.kubiak@samsung.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
This change has two goals:
- delay the setting of 'smack_enabled' until
it will be really effective
- ensure that smackfs is valid only if 'smack_enabled'
is set (it is already the case in smack_netfilter.c)
Signed-off-by: José Bollo <jose.bollo@iot.bzh>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
Fixes the following sparse warning:
security/smack/smack_lsm.c:55:1: warning: symbol 'smk_ipv6_port_list'
was not declared. Should it be static?
Signed-off-by: Geliang Tang <geliangtang@163.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
'commit e774ad683f ("smack: pass error code through pointers")'
made this function return proper error codes instead of NULL. Reflect that.
This is a fix for a NULL dereference introduced in
'commit 21abb1ec41 ("Smack: IPv6 host labeling")'
echo "$SOME_IPV6_ADDR \"test" > /smack/ipv6host
(this should return EINVAL, it doesn't)
cat /smack/ipv6host
(derefences 0x000a)
Signed-off-by: Lukasz Pawelczyk <l.pawelczyk@samsung.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
If IMA_LOAD_X509 is enabled, either directly or indirectly via
IMA_APPRAISE_SIGNED_INIT, certificates are loaded onto the IMA
trusted keyring by the kernel via key_create_or_update(). When
the KEY_ALLOC_TRUSTED flag is provided, certificates are loaded
without first verifying the certificate is properly signed by a
trusted key on the system keyring. This patch removes the
KEY_ALLOC_TRUSTED flag.
Signed-off-by: Dmitry Kasatkin <dmitry.kasatkin@huawei.com>
Cc: <stable@vger.kernel.org> # 3.19+
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Conflicts:
net/ipv4/arp.c
The net/ipv4/arp.c conflict was one commit adding a new
local variable while another commit was deleting one.
Signed-off-by: David S. Miller <davem@davemloft.net>
There appears to be a race between:
(1) key_gc_unused_keys() which frees key->security and then calls
keyring_destroy() to unlink the name from the name list
(2) find_keyring_by_name() which calls key_permission(), thus accessing
key->security, on a key before checking to see whether the key usage is 0
(ie. the key is dead and might be cleaned up).
Fix this by calling ->destroy() before cleaning up the core key data -
including key->security.
Reported-by: Petr Matousek <pmatouse@redhat.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Only pass the void *priv parameter out of the nf_hook_ops. That is
all any of the functions are interested now, and by limiting what is
passed it becomes simpler to change implementation details.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
With two exceptions (drm/qxl and drm/radeon) all vm_operations_struct
structs should be constant.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull security subsystem updates from James Morris:
"Highlights:
- PKCS#7 support added to support signed kexec, also utilized for
module signing. See comments in 3f1e1bea.
** NOTE: this requires linking against the OpenSSL library, which
must be installed, e.g. the openssl-devel on Fedora **
- Smack
- add IPv6 host labeling; ignore labels on kernel threads
- support smack labeling mounts which use binary mount data
- SELinux:
- add ioctl whitelisting (see
http://kernsec.org/files/lss2015/vanderstoep.pdf)
- fix mprotect PROT_EXEC regression caused by mm change
- Seccomp:
- add ptrace options for suspend/resume"
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (57 commits)
PKCS#7: Add OIDs for sha224, sha284 and sha512 hash algos and use them
Documentation/Changes: Now need OpenSSL devel packages for module signing
scripts: add extract-cert and sign-file to .gitignore
modsign: Handle signing key in source tree
modsign: Use if_changed rule for extracting cert from module signing key
Move certificate handling to its own directory
sign-file: Fix warning about BIO_reset() return value
PKCS#7: Add MODULE_LICENSE() to test module
Smack - Fix build error with bringup unconfigured
sign-file: Document dependency on OpenSSL devel libraries
PKCS#7: Appropriately restrict authenticated attributes and content type
KEYS: Add a name for PKEY_ID_PKCS7
PKCS#7: Improve and export the X.509 ASN.1 time object decoder
modsign: Use extract-cert to process CONFIG_SYSTEM_TRUSTED_KEYS
extract-cert: Cope with multiple X.509 certificates in a single file
sign-file: Generate CMS message as signature instead of PKCS#7
PKCS#7: Support CMS messages also [RFC5652]
X.509: Change recorded SKID & AKID to not include Subject or Issuer
PKCS#7: Check content type and versions
MAINTAINERS: The keyrings mailing list has moved
...
Many file systems that implement the show_options hook fail to correctly
escape their output which could lead to unescaped characters (e.g. new
lines) leaking into /proc/mounts and /proc/[pid]/mountinfo files. This
could lead to confusion, spoofed entries (resulting in things like
systemd issuing false d-bus "mount" notifications), and who knows what
else. This looks like it would only be the root user stepping on
themselves, but it's possible weird things could happen in containers or
in other situations with delegated mount privileges.
Here's an example using overlay with setuid fusermount trusting the
contents of /proc/mounts (via the /etc/mtab symlink). Imagine the use
of "sudo" is something more sneaky:
$ BASE="ovl"
$ MNT="$BASE/mnt"
$ LOW="$BASE/lower"
$ UP="$BASE/upper"
$ WORK="$BASE/work/ 0 0
none /proc fuse.pwn user_id=1000"
$ mkdir -p "$LOW" "$UP" "$WORK"
$ sudo mount -t overlay -o "lowerdir=$LOW,upperdir=$UP,workdir=$WORK" none /mnt
$ cat /proc/mounts
none /root/ovl/mnt overlay rw,relatime,lowerdir=ovl/lower,upperdir=ovl/upper,workdir=ovl/work/ 0 0
none /proc fuse.pwn user_id=1000 0 0
$ fusermount -u /proc
$ cat /proc/mounts
cat: /proc/mounts: No such file or directory
This fixes the problem by adding new seq_show_option and
seq_show_option_n helpers, and updating the vulnerable show_option
handlers to use them as needed. Some, like SELinux, need to be open
coded due to unusual existing escape mechanisms.
[akpm@linux-foundation.org: add lost chunk, per Kees]
[keescook@chromium.org: seq_show_option should be using const parameters]
Signed-off-by: Kees Cook <keescook@chromium.org>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Acked-by: Jan Kara <jack@suse.com>
Acked-by: Paul Moore <paul@paul-moore.com>
Cc: J. R. Okajima <hooanon05g@gmail.com>
Signed-off-by: Kees Cook <keescook@chromium.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Per Andrew Morgan's request, add a securebit to allow admins to disable
PR_CAP_AMBIENT_RAISE. This securebit will prevent processes from adding
capabilities to their ambient set.
For simplicity, this disables PR_CAP_AMBIENT_RAISE entirely rather than
just disabling setting previously cleared bits.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Acked-by: Andrew G. Morgan <morgan@kernel.org>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Aaron Jones <aaronmdjones@gmail.com>
Cc: Ted Ts'o <tytso@mit.edu>
Cc: Andrew G. Morgan <morgan@kernel.org>
Cc: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Austin S Hemmelgarn <ahferroin7@gmail.com>
Cc: Markku Savela <msa@moth.iki.fi>
Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: James Morris <james.l.morris@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Credit where credit is due: this idea comes from Christoph Lameter with
a lot of valuable input from Serge Hallyn. This patch is heavily based
on Christoph's patch.
===== The status quo =====
On Linux, there are a number of capabilities defined by the kernel. To
perform various privileged tasks, processes can wield capabilities that
they hold.
Each task has four capability masks: effective (pE), permitted (pP),
inheritable (pI), and a bounding set (X). When the kernel checks for a
capability, it checks pE. The other capability masks serve to modify
what capabilities can be in pE.
Any task can remove capabilities from pE, pP, or pI at any time. If a
task has a capability in pP, it can add that capability to pE and/or pI.
If a task has CAP_SETPCAP, then it can add any capability to pI, and it
can remove capabilities from X.
Tasks are not the only things that can have capabilities; files can also
have capabilities. A file can have no capabilty information at all [1].
If a file has capability information, then it has a permitted mask (fP)
and an inheritable mask (fI) as well as a single effective bit (fE) [2].
File capabilities modify the capabilities of tasks that execve(2) them.
A task that successfully calls execve has its capabilities modified for
the file ultimately being excecuted (i.e. the binary itself if that
binary is ELF or for the interpreter if the binary is a script.) [3] In
the capability evolution rules, for each mask Z, pZ represents the old
value and pZ' represents the new value. The rules are:
pP' = (X & fP) | (pI & fI)
pI' = pI
pE' = (fE ? pP' : 0)
X is unchanged
For setuid binaries, fP, fI, and fE are modified by a moderately
complicated set of rules that emulate POSIX behavior. Similarly, if
euid == 0 or ruid == 0, then fP, fI, and fE are modified differently
(primary, fP and fI usually end up being the full set). For nonroot
users executing binaries with neither setuid nor file caps, fI and fP
are empty and fE is false.
As an extra complication, if you execute a process as nonroot and fE is
set, then the "secure exec" rules are in effect: AT_SECURE gets set,
LD_PRELOAD doesn't work, etc.
This is rather messy. We've learned that making any changes is
dangerous, though: if a new kernel version allows an unprivileged
program to change its security state in a way that persists cross
execution of a setuid program or a program with file caps, this
persistent state is surprisingly likely to allow setuid or file-capped
programs to be exploited for privilege escalation.
===== The problem =====
Capability inheritance is basically useless.
If you aren't root and you execute an ordinary binary, fI is zero, so
your capabilities have no effect whatsoever on pP'. This means that you
can't usefully execute a helper process or a shell command with elevated
capabilities if you aren't root.
On current kernels, you can sort of work around this by setting fI to
the full set for most or all non-setuid executable files. This causes
pP' = pI for nonroot, and inheritance works. No one does this because
it's a PITA and it isn't even supported on most filesystems.
If you try this, you'll discover that every nonroot program ends up with
secure exec rules, breaking many things.
This is a problem that has bitten many people who have tried to use
capabilities for anything useful.
===== The proposed change =====
This patch adds a fifth capability mask called the ambient mask (pA).
pA does what most people expect pI to do.
pA obeys the invariant that no bit can ever be set in pA if it is not
set in both pP and pI. Dropping a bit from pP or pI drops that bit from
pA. This ensures that existing programs that try to drop capabilities
still do so, with a complication. Because capability inheritance is so
broken, setting KEEPCAPS, using setresuid to switch to nonroot uids, and
then calling execve effectively drops capabilities. Therefore,
setresuid from root to nonroot conditionally clears pA unless
SECBIT_NO_SETUID_FIXUP is set. Processes that don't like this can
re-add bits to pA afterwards.
The capability evolution rules are changed:
pA' = (file caps or setuid or setgid ? 0 : pA)
pP' = (X & fP) | (pI & fI) | pA'
pI' = pI
pE' = (fE ? pP' : pA')
X is unchanged
If you are nonroot but you have a capability, you can add it to pA. If
you do so, your children get that capability in pA, pP, and pE. For
example, you can set pA = CAP_NET_BIND_SERVICE, and your children can
automatically bind low-numbered ports. Hallelujah!
Unprivileged users can create user namespaces, map themselves to a
nonzero uid, and create both privileged (relative to their namespace)
and unprivileged process trees. This is currently more or less
impossible. Hallelujah!
You cannot use pA to try to subvert a setuid, setgid, or file-capped
program: if you execute any such program, pA gets cleared and the
resulting evolution rules are unchanged by this patch.
Users with nonzero pA are unlikely to unintentionally leak that
capability. If they run programs that try to drop privileges, dropping
privileges will still work.
It's worth noting that the degree of paranoia in this patch could
possibly be reduced without causing serious problems. Specifically, if
we allowed pA to persist across executing non-pA-aware setuid binaries
and across setresuid, then, naively, the only capabilities that could
leak as a result would be the capabilities in pA, and any attacker
*already* has those capabilities. This would make me nervous, though --
setuid binaries that tried to privilege-separate might fail to do so,
and putting CAP_DAC_READ_SEARCH or CAP_DAC_OVERRIDE into pA could have
unexpected side effects. (Whether these unexpected side effects would
be exploitable is an open question.) I've therefore taken the more
paranoid route. We can revisit this later.
An alternative would be to require PR_SET_NO_NEW_PRIVS before setting
ambient capabilities. I think that this would be annoying and would
make granting otherwise unprivileged users minor ambient capabilities
(CAP_NET_BIND_SERVICE or CAP_NET_RAW for example) much less useful than
it is with this patch.
===== Footnotes =====
[1] Files that are missing the "security.capability" xattr or that have
unrecognized values for that xattr end up with has_cap set to false.
The code that does that appears to be complicated for no good reason.
[2] The libcap capability mask parsers and formatters are dangerously
misleading and the documentation is flat-out wrong. fE is *not* a mask;
it's a single bit. This has probably confused every single person who
has tried to use file capabilities.
[3] Linux very confusingly processes both the script and the interpreter
if applicable, for reasons that elude me. The results from thinking
about a script's file capabilities and/or setuid bits are mostly
discarded.
Preliminary userspace code is here, but it needs updating:
https://git.kernel.org/cgit/linux/kernel/git/luto/util-linux-playground.git/commit/?h=cap_ambient&id=7f5afbd175d2
Here is a test program that can be used to verify the functionality
(from Christoph):
/*
* Test program for the ambient capabilities. This program spawns a shell
* that allows running processes with a defined set of capabilities.
*
* (C) 2015 Christoph Lameter <cl@linux.com>
* Released under: GPL v3 or later.
*
*
* Compile using:
*
* gcc -o ambient_test ambient_test.o -lcap-ng
*
* This program must have the following capabilities to run properly:
* Permissions for CAP_NET_RAW, CAP_NET_ADMIN, CAP_SYS_NICE
*
* A command to equip the binary with the right caps is:
*
* setcap cap_net_raw,cap_net_admin,cap_sys_nice+p ambient_test
*
*
* To get a shell with additional caps that can be inherited by other processes:
*
* ./ambient_test /bin/bash
*
*
* Verifying that it works:
*
* From the bash spawed by ambient_test run
*
* cat /proc/$$/status
*
* and have a look at the capabilities.
*/
#include <stdlib.h>
#include <stdio.h>
#include <errno.h>
#include <cap-ng.h>
#include <sys/prctl.h>
#include <linux/capability.h>
/*
* Definitions from the kernel header files. These are going to be removed
* when the /usr/include files have these defined.
*/
#define PR_CAP_AMBIENT 47
#define PR_CAP_AMBIENT_IS_SET 1
#define PR_CAP_AMBIENT_RAISE 2
#define PR_CAP_AMBIENT_LOWER 3
#define PR_CAP_AMBIENT_CLEAR_ALL 4
static void set_ambient_cap(int cap)
{
int rc;
capng_get_caps_process();
rc = capng_update(CAPNG_ADD, CAPNG_INHERITABLE, cap);
if (rc) {
printf("Cannot add inheritable cap\n");
exit(2);
}
capng_apply(CAPNG_SELECT_CAPS);
/* Note the two 0s at the end. Kernel checks for these */
if (prctl(PR_CAP_AMBIENT, PR_CAP_AMBIENT_RAISE, cap, 0, 0)) {
perror("Cannot set cap");
exit(1);
}
}
int main(int argc, char **argv)
{
int rc;
set_ambient_cap(CAP_NET_RAW);
set_ambient_cap(CAP_NET_ADMIN);
set_ambient_cap(CAP_SYS_NICE);
printf("Ambient_test forking shell\n");
if (execv(argv[1], argv + 1))
perror("Cannot exec");
return 0;
}
Signed-off-by: Christoph Lameter <cl@linux.com> # Original author
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Acked-by: Serge E. Hallyn <serge.hallyn@ubuntu.com>
Acked-by: Kees Cook <keescook@chromium.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Aaron Jones <aaronmdjones@gmail.com>
Cc: Ted Ts'o <tytso@mit.edu>
Cc: Andrew G. Morgan <morgan@kernel.org>
Cc: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: Austin S Hemmelgarn <ahferroin7@gmail.com>
Cc: Markku Savela <msa@moth.iki.fi>
Cc: Jarkko Sakkinen <jarkko.sakkinen@linux.intel.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: James Morris <james.l.morris@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
f78f5b90c4 ("rcu: Rename rcu_lockdep_assert() to RCU_LOCKDEP_WARN()")
introduced a bug by incorrectly inverting the condition when moving from
rcu_lockdep_assert() to RCU_LOCKDEP_WARN(). This commit therefore fixes
the inversion.
Reported-by: Felipe Balbi <balbi@ti.com>
Reported-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Tested-by: Josh Boyer <jwboyer@fedoraproject.org>
Pull user namespace updates from Eric Biederman:
"This finishes up the changes to ensure proc and sysfs do not start
implementing executable files, as the there are application today that
are only secure because such files do not exist.
It akso fixes a long standing misfeature of /proc/<pid>/mountinfo that
did not show the proper source for files bind mounted from
/proc/<pid>/ns/*.
It also straightens out the handling of clone flags related to user
namespaces, fixing an unnecessary failure of unshare(CLONE_NEWUSER)
when files such as /proc/<pid>/environ are read while <pid> is calling
unshare. This winds up fixing a minor bug in unshare flag handling
that dates back to the first version of unshare in the kernel.
Finally, this fixes a minor regression caused by the introduction of
sysfs_create_mount_point, which broke someone's in house application,
by restoring the size of /sys/fs/cgroup to 0 bytes. Apparently that
application uses the directory size to determine if a tmpfs is mounted
on /sys/fs/cgroup.
The bind mount escape fixes are present in Al Viros for-next branch.
and I expect them to come from there. The bind mount escape is the
last of the user namespace related security bugs that I am aware of"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
fs: Set the size of empty dirs to 0.
userns,pidns: Force thread group sharing, not signal handler sharing.
unshare: Unsharing a thread does not require unsharing a vm
nsfs: Add a show_path method to fix mountinfo
mnt: fs_fully_visible enforce noexec and nosuid if !SB_I_NOEXEC
vfs: Commit to never having exectuables on proc and sysfs.
Pull RCU updates from Ingo Molnar:
"The main RCU changes in this cycle are:
- the combination of tree geometry-initialization simplifications and
OS-jitter-reduction changes to expedited grace periods. These two
are stacked due to the large number of conflicts that would
otherwise result.
- privatize smp_mb__after_unlock_lock().
This commit moves the definition of smp_mb__after_unlock_lock() to
kernel/rcu/tree.h, in recognition of the fact that RCU is the only
thing using this, that nothing else is likely to use it, and that
it is likely to go away completely.
- documentation updates.
- torture-test updates.
- misc fixes"
* 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (60 commits)
rcu,locking: Privatize smp_mb__after_unlock_lock()
rcu: Silence lockdep false positive for expedited grace periods
rcu: Don't disable CPU hotplug during OOM notifiers
scripts: Make checkpatch.pl warn on expedited RCU grace periods
rcu: Update MAINTAINERS entry
rcu: Clarify CONFIG_RCU_EQS_DEBUG help text
rcu: Fix backwards RCU_LOCKDEP_WARN() in synchronize_rcu_tasks()
rcu: Rename rcu_lockdep_assert() to RCU_LOCKDEP_WARN()
rcu: Make rcu_is_watching() really notrace
cpu: Wait for RCU grace periods concurrently
rcu: Create a synchronize_rcu_mult()
rcu: Fix obsolete priority-boosting comment
rcu: Use WRITE_ONCE in RCU_INIT_POINTER
rcu: Hide RCU_NOCB_CPU behind RCU_EXPERT
rcu: Add RCU-sched flavors of get-state and cond-sync
rcu: Add fastpath bypassing funnel locking
rcu: Rename RCU_GP_DONE_FQS to RCU_GP_DOING_FQS
rcu: Pull out wait_event*() condition into helper function
documentation: Describe new expedited stall warnings
rcu: Add stall warnings to synchronize_sched_expedited()
...
While in most cases commit b1d9e6b064 ("LSM: Switch to lists of hooks")
retained previous error returns, in three cases it altered them without
any explanation in the commit message. Restore all of them - in the
security_old_inode_init_security() case this led to reiserfs using
uninitialized data, sooner or later crashing the system (the only other
user of this function - ocfs2 - was unaffected afaict, since it passes
pre-initialized structures).
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: James Morris <james.l.morris@oracle.com>
Mel Gorman