summaryrefslogtreecommitdiffstats
path: root/security
Commit message (Collapse)AuthorAgeFilesLines
* Merge branch 'userns-linus' of ↵Linus Torvalds2018-04-241-0/+2
|\ | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace Pull userns bug fix from Eric Biederman: "Just a small fix to properly set the return code on error" * 'userns-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: commoncap: Handle memory allocation failure.
| * commoncap: Handle memory allocation failure.Tetsuo Handa2018-04-101-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | syzbot is reporting NULL pointer dereference at xattr_getsecurity() [1], for cap_inode_getsecurity() is returning sizeof(struct vfs_cap_data) when memory allocation failed. Return -ENOMEM if memory allocation failed. [1] https://syzkaller.appspot.com/bug?id=a55ba438506fe68649a5f50d2d82d56b365e0107 Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Fixes: 8db6c34f1dbc8e06 ("Introduce v3 namespaced file capabilities") Reported-by: syzbot <syzbot+9369930ca44f29e60e2d@syzkaller.appspotmail.com> Cc: stable <stable@vger.kernel.org> # 4.14+ Acked-by: Serge E. Hallyn <serge@hallyn.com> Acked-by: James Morris <james.morris@microsoft.com> Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
* | Merge tag 'apparmor-pr-2018-04-10' of ↵Linus Torvalds2018-04-1332-523/+2117
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/jj/linux-apparmor Pull apparmor updates from John Johansen: "Features: - add base infrastructure for socket mediation. ABI bump and additional checks to ensure only v8 compliant policy uses socket af mediation. - improve and cleanup dfa verification - improve profile attachment logic - improve overlapping expression handling - add the xattr matching to the attachment logic - improve signal mediation handling with stacked labels - improve handling of no_new_privs in a label stack Cleanups and changes: - use dfa to parse string split - bounded version of label_parse - proper line wrap nulldfa.in - split context out into task and cred naming to better match usage - simplify code in aafs Bug fixes: - fix display of .ns_name for containers - fix resource audit messages when auditing peer - fix logging of the existence test for signals - fix resource audit messages when auditing peer - fix display of .ns_name for containers - fix an error code in verify_table_headers() - fix memory leak on buffer on error exit path - fix error returns checks by making size a ssize_t" * tag 'apparmor-pr-2018-04-10' of git://git.kernel.org/pub/scm/linux/kernel/git/jj/linux-apparmor: (36 commits) apparmor: fix memory leak on buffer on error exit path apparmor: fix dangling symlinks to policy rawdata after replacement apparmor: Fix an error code in verify_table_headers() apparmor: fix error returns checks by making size a ssize_t apparmor: update MAINTAINERS file git and wiki locations apparmor: remove POLICY_MEDIATES_SAFE apparmor: add base infastructure for socket mediation apparmor: improve overlapping domain attachment resolution apparmor: convert attaching profiles via xattrs to use dfa matching apparmor: Add support for attaching profiles via xattr, presence and value apparmor: cleanup: simplify code to get ns symlink name apparmor: cleanup create_aafs() error path apparmor: dfa split verification of table headers apparmor: dfa add support for state differential encoding apparmor: dfa move character match into a macro apparmor: update domain transitions that are subsets of confinement at nnp apparmor: move context.h to cred.h apparmor: move task related defines and fns to task.X files apparmor: cleanup, drop unused fn __aa_task_is_confined() apparmor: cleanup fixup description of aa_replace_profiles ...
| * | apparmor: fix memory leak on buffer on error exit pathColin Ian King2018-03-301-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently on the error exit path the allocated buffer is not free'd causing a memory leak. Fix this by kfree'ing it. Detected by CoverityScan, CID#1466876 ("Resource leaks") Fixes: 1180b4c757aa ("apparmor: fix dangling symlinks to policy rawdata after replacement") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: fix dangling symlinks to policy rawdata after replacementJohn Johansen2018-03-231-31/+95
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When policy replacement occurs the symlinks in the profile directory need to be updated to point to the new rawdata, otherwise once the old rawdata is removed the symlink becomes broken. Fix this by dynamically generating the symlink everytime it is read. These links are used enough that their value needs to be cached and this way we can avoid needing locking to read and update the link value. Fixes: a481f4d917835 ("apparmor: add custom apparmorfs that will be used by policy namespace files") BugLink: http://bugs.launchpad.net/bugs/1755563 Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: Fix an error code in verify_table_headers()Dan Carpenter2018-03-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We accidentally return a positive EPROTO instead of a negative -EPROTO. Since 71 is not an error pointer, that means it eventually results in an Oops in the caller. Fixes: d901d6a298dc ("apparmor: dfa split verification of table headers") Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: fix error returns checks by making size a ssize_tColin Ian King2018-03-231-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently variable size is a unsigned size_t, hence comparisons to see if it is less than zero (for error checking) will always be false. Fix this by making size a ssize_t Detected by CoverityScan, CID#1466080 ("Unsigned compared against 0") Fixes: 8e51f9087f40 ("apparmor: Add support for attaching profiles via xattr, presence and value") Signed-off-by: Colin Ian King <colin.king@canonical.com> Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: remove POLICY_MEDIATES_SAFEJohn Johansen2018-03-132-12/+2
| | | | | | | | | | | | | | | | | | | | | The unpack code now makes sure every profile has a dfa so the safe version of POLICY_MEDIATES is no longer needed. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: add base infastructure for socket mediationJohn Johansen2018-03-1313-8/+786
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | version 2 - Force an abi break. Network mediation will only be available in v8 abi complaint policy. Provide a basic mediation of sockets. This is not a full net mediation but just whether a spcific family of socket can be used by an application, along with setting up some basic infrastructure for network mediation to follow. the user space rule hav the basic form of NETWORK RULE = [ QUALIFIERS ] 'network' [ DOMAIN ] [ TYPE | PROTOCOL ] DOMAIN = ( 'inet' | 'ax25' | 'ipx' | 'appletalk' | 'netrom' | 'bridge' | 'atmpvc' | 'x25' | 'inet6' | 'rose' | 'netbeui' | 'security' | 'key' | 'packet' | 'ash' | 'econet' | 'atmsvc' | 'sna' | 'irda' | 'pppox' | 'wanpipe' | 'bluetooth' | 'netlink' | 'unix' | 'rds' | 'llc' | 'can' | 'tipc' | 'iucv' | 'rxrpc' | 'isdn' | 'phonet' | 'ieee802154' | 'caif' | 'alg' | 'nfc' | 'vsock' | 'mpls' | 'ib' | 'kcm' ) ',' TYPE = ( 'stream' | 'dgram' | 'seqpacket' | 'rdm' | 'raw' | 'packet' ) PROTOCOL = ( 'tcp' | 'udp' | 'icmp' ) eg. network, network inet, Signed-off-by: John Johansen <john.johansen@canonical.com> Acked-by: Seth Arnold <seth.arnold@canonical.com>
| * | apparmor: improve overlapping domain attachment resolutionJohn Johansen2018-02-094-14/+158
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Overlapping domain attachments using the current longest left exact match fail in some simple cases, and with the fix to ensure consistent behavior by failing unresolvable attachments it becomes important to do a better job. eg. under the current match the following are unresolvable where the alternation is clearly a better match under the most specific left match rule. /** /{bin/,}usr/ Use a counting match that detects when a loop in the state machine is enter, and return the match count to provide a better specific left match resolution. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: convert attaching profiles via xattrs to use dfa matchingJohn Johansen2018-02-095-57/+43
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This converts profile attachment based on xattrs to a fixed extended conditional using dfa matching. This has a couple of advantages - pattern matching can be used for the xattr match - xattrs can be optional for an attachment or marked as required - the xattr attachment conditional will be able to be combined with other extended conditionals when the flexible extended conditional work lands. The xattr fixed extended conditional is appended to the xmatch conditional. If an xattr attachment is specified the profile xmatch will be generated regardless of whether there is a pattern match on the executable name. Signed-off-by: John Johansen <john.johansen@canonical.com> Acked-by: Seth Arnold <seth.arnold@canonical.com>
| * | apparmor: Add support for attaching profiles via xattr, presence and valueMatthew Garrett2018-02-094-34/+217
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Make it possible to tie Apparmor profiles to the presence of one or more extended attributes, and optionally their values. An example usecase for this is to automatically transition to a more privileged Apparmor profile if an executable has a valid IMA signature, which can then be appraised by the IMA subsystem. Signed-off-by: Matthew Garrett <mjg59@google.com> Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: cleanup: simplify code to get ns symlink nameJohn Johansen2018-02-091-19/+6
| | | | | | | | | | | | | | | | | | ns_get_name() is called in only one place and can be folded in. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: cleanup create_aafs() error pathJohn Johansen2018-02-091-20/+12
| | | | | | | | | | | | Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: dfa split verification of table headersJohn Johansen2018-02-091-48/+68
| | | | | | | | | | | | | | | | | | | | | separate the different types of verification so they are logically separate and can be reused separate of each other. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: dfa add support for state differential encodingJohn Johansen2018-02-092-1/+29
| | | | | | | | | | | | | | | | | | | | | State differential encoding can provide better compression for apparmor policy, without having significant impact on match time. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: dfa move character match into a macroJohn Johansen2018-02-091-47/+27
| | | | | | | | | | | | Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: update domain transitions that are subsets of confinement at nnpJohn Johansen2018-02-094-65/+110
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Domain transition so far have been largely blocked by no new privs, unless the transition has been provably a subset of the previous confinement. There was a couple problems with the previous implementations, - transitions that weren't explicitly a stack but resulted in a subset of confinement were disallowed - confinement subsets were only calculated from the previous confinement instead of the confinement being enforced at the time of no new privs, so transitions would have to get progressively tighter. Fix this by detecting and storing a reference to the task's confinement at the "time" no new privs is set. This reference is then used to determine whether a transition is a subsystem of the confinement at the time no new privs was set. Unfortunately the implementation is less than ideal in that we have to detect no new privs after the fact when a task attempts a domain transition. This is adequate for the currently but will not work in a stacking situation where no new privs could be conceivably be set in both the "host" and in the container. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: move context.h to cred.hJohn Johansen2018-02-0915-14/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | Now that file contexts have been moved into file, and task context fns() and data have been split from the context, only the cred context remains in context.h so rename to cred.h to better reflect what it deals with. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: move task related defines and fns to task.X filesJohn Johansen2018-02-096-98/+105
| | | | | | | | | | | | Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: cleanup, drop unused fn __aa_task_is_confined()John Johansen2018-02-091-11/+0
| | | | | | | | | | | | Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: cleanup fixup description of aa_replace_profilesJohn Johansen2018-02-091-2/+3
| | | | | | | | | | | | Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: rename tctx to ctxJohn Johansen2018-02-093-30/+29
| | | | | | | | | | | | | | | | | | | | | now that cred_ctx has been removed we can rename task_ctxs from tctx without causing confusion. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: drop cred_ctx and reference the label directlyJohn Johansen2018-02-094-129/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | | With the task domain change information now stored in the task->security context, the cred->security context only stores the label. We can get rid of the cred_ctx and directly reference the label, removing a layer of indirection, and unneeded extra allocations. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: move task domain change info to task securityJohn Johansen2018-02-094-52/+132
| | | | | | | | | | | | | | | | | | | | | | | | The task domain change info is task specific and its and abuse of the cred to store the information in there. Now that a task->security field exists store it in the proper place. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: rename task_ctx to the more accurate cred_ctxJohn Johansen2018-02-095-46/+45
| | | | | | | | | | | | Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: audit unknown signal numbersJohn Johansen2018-02-093-4/+12
| | | | | | | | | | | | | | | | | | | | | | | | Allow apparmor to audit the number of a signal that it does not provide a mapping for and is currently being reported only as unknown. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: make signal label match work when matching stacked labelsJohn Johansen2018-02-091-28/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Given a label with a profile stack of A//&B or A//&C ... A ptrace rule should be able to specify a generic trace pattern with a rule like signal send A//&**, however this is failing because while the correct label match routine is called, it is being done post label decomposition so it is always being done against a profile instead of the stacked label. To fix this refactor the cross check to pass the full peer label in to the label_match. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | security: apparmor: remove duplicate includesPravin Shedge2018-02-091-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | These duplicate includes have been found with scripts/checkincludes.pl but they have been removed manually to avoid removing false positives. Signed-off-by: Pravin Shedge <pravin.shedge4linux@gmail.com> Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: root view labels should not be under user controlJohn Johansen2018-02-091-3/+2
| | | | | | | | | | | | | | | | | | | | | | | | The root view of the label parse should not be exposed to user control. Signed-off-by: John Johansen <john.johansen@canonical.com> Acked-by: Seth Arnold <seth.arnold@canonical.com>
| * | apparmor: cleanup add proper line wrapping to nulldfa.inJohn Johansen2018-02-091-1/+107
| | | | | | | | | | | | | | | | | | | | | nulldfa.in makes for a very long unwrapped line, which certain tools do not like. So add line breaks. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: provide a bounded version of label_parseJohn Johansen2018-02-092-11/+27
| | | | | | | | | | | | | | | | | | | | | | | | some label/context sources might not be guaranteed to be null terminiated provide a size bounded version of label parse to deal with these. Signed-off-by: John Johansen <john.johansen@canonical.com> Acked-by: Seth Arnold <seth.arnold@canonical.com>
| * | apparmor: use the dfa to do label parse string splittingJohn Johansen2018-02-095-11/+170
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current split scheme is actually wrong in that it splits ///& where that is invalid and should fail. Use the dfa to do a proper bounded split without having to worry about getting the string processing right in code. Signed-off-by: John Johansen <john.johansen@canonical.com> Acked-by: Seth Arnold <seth.arnold@canonical.com>
| * | apparmor: add first substr match to dfaJohn Johansen2018-02-092-0/+124
| | | | | | | | | | | | | | | Signed-off-by: John Johansen <john.johansen@canonical.com> Acked-by: Seth Arnold <seth.arnold@canonical.com>
| * | apparmor: split load data into management struct and data blobJohn Johansen2018-02-092-4/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Splitting the management struct from the actual data blob will allow us in the future to do some sharing and other data reduction techniques like replacing the the raw data with compressed data. Prepare for this by separating the management struct from the data blob. Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: fix logging of the existence test for signalsJohn Johansen2018-02-092-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The existence test is not being properly logged as the signal mapping maps it to the last entry in the named signal table. This is done to help catch bugs by making the 0 mapped signal value invalid so that we can catch the signal value not being filled in. When fixing the off-by-one comparision logic the reporting of the existence test was broken, because the logic behind the mapped named table was hidden. Fix this by adding a define for the name lookup and using it. Cc: Stable <stable@vger.kernel.org> Fixes: f7dc4c9a855a1 ("apparmor: fix off-by-one comparison on MAXMAPPED_SIG") Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: fix resource audit messages when auditing peerJohn Johansen2018-02-091-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Resource auditing is using the peer field which is not available when the rlim data struct is used, because it is a different element of the same union. Accessing peer during resource auditing could cause garbage log entries or even oops the kernel. Move the rlim data block into the same struct as the peer field so they can be used together. CC: <stable@vger.kernel.org> Fixes: 86b92cb782b3 ("apparmor: move resource checks to using labels") Signed-off-by: John Johansen <john.johansen@canonical.com>
| * | apparmor: fix display of .ns_name for containersJohn Johansen2018-02-091-3/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The .ns_name should not be virtualized by the current ns view. It needs to report the ns base name as that is being used during startup as part of determining apparmor policy namespace support. BugLink: http://bugs.launchpad.net/bugs/1746463 Fixes: d9f02d9c237aa ("apparmor: fix display of ns name") Cc: Stable <stable@vger.kernel.org> Reported-by: Serge Hallyn <serge@hallyn.com> Tested-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: John Johansen <john.johansen@canonical.com>
* | | ipc/msg: introduce msgctl(MSG_STAT_ANY)Davidlohr Bueso2018-04-112-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is a permission discrepancy when consulting msq ipc object metadata between /proc/sysvipc/msg (0444) and the MSG_STAT shmctl command. The later does permission checks for the object vs S_IRUGO. As such there can be cases where EACCESS is returned via syscall but the info is displayed anyways in the procfs files. While this might have security implications via info leaking (albeit no writing to the msq metadata), this behavior goes way back and showing all the objects regardless of the permissions was most likely an overlook - so we are stuck with it. Furthermore, modifying either the syscall or the procfs file can cause userspace programs to break (ie ipcs). Some applications require getting the procfs info (without root privileges) and can be rather slow in comparison with a syscall -- up to 500x in some reported cases for shm. This patch introduces a new MSG_STAT_ANY command such that the msq ipc object permissions are ignored, and only audited instead. In addition, I've left the lsm security hook checks in place, as if some policy can block the call, then the user has no other choice than just parsing the procfs file. Link: http://lkml.kernel.org/r/20180215162458.10059-4-dave@stgolabs.net Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Reported-by: Robert Kettler <robert.kettler@outlook.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Kees Cook <keescook@chromium.org> Cc: Manfred Spraul <manfred@colorfullife.com> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Michal Hocko <mhocko@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | | ipc/sem: introduce semctl(SEM_STAT_ANY)Davidlohr Bueso2018-04-112-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is a permission discrepancy when consulting shm ipc object metadata between /proc/sysvipc/sem (0444) and the SEM_STAT semctl command. The later does permission checks for the object vs S_IRUGO. As such there can be cases where EACCESS is returned via syscall but the info is displayed anyways in the procfs files. While this might have security implications via info leaking (albeit no writing to the sma metadata), this behavior goes way back and showing all the objects regardless of the permissions was most likely an overlook - so we are stuck with it. Furthermore, modifying either the syscall or the procfs file can cause userspace programs to break (ie ipcs). Some applications require getting the procfs info (without root privileges) and can be rather slow in comparison with a syscall -- up to 500x in some reported cases for shm. This patch introduces a new SEM_STAT_ANY command such that the sem ipc object permissions are ignored, and only audited instead. In addition, I've left the lsm security hook checks in place, as if some policy can block the call, then the user has no other choice than just parsing the procfs file. Link: http://lkml.kernel.org/r/20180215162458.10059-3-dave@stgolabs.net Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Reported-by: Robert Kettler <robert.kettler@outlook.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Kees Cook <keescook@chromium.org> Cc: Manfred Spraul <manfred@colorfullife.com> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Michal Hocko <mhocko@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | | ipc/shm: introduce shmctl(SHM_STAT_ANY)Davidlohr Bueso2018-04-112-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Patch series "sysvipc: introduce STAT_ANY commands", v2. The following patches adds the discussed (see [1]) new command for shm as well as for sems and msq as they are subject to the same discrepancies for ipc object permission checks between the syscall and via procfs. These new commands are justified in that (1) we are stuck with this semantics as changing syscall and procfs can break userland; and (2) some users can benefit from performance (for large amounts of shm segments, for example) from not having to parse the procfs interface. Once merged, I will submit the necesary manpage updates. But I'm thinking something like: : diff --git a/man2/shmctl.2 b/man2/shmctl.2 : index 7bb503999941..bb00bbe21a57 100644 : --- a/man2/shmctl.2 : +++ b/man2/shmctl.2 : @@ -41,6 +41,7 @@ : .\" 2005-04-25, mtk -- noted aberrant Linux behavior w.r.t. new : .\" attaches to a segment that has already been marked for deletion. : .\" 2005-08-02, mtk: Added IPC_INFO, SHM_INFO, SHM_STAT descriptions. : +.\" 2018-02-13, dbueso: Added SHM_STAT_ANY description. : .\" : .TH SHMCTL 2 2017-09-15 "Linux" "Linux Programmer's Manual" : .SH NAME : @@ -242,6 +243,18 @@ However, the : argument is not a segment identifier, but instead an index into : the kernel's internal array that maintains information about : all shared memory segments on the system. : +.TP : +.BR SHM_STAT_ANY " (Linux-specific)" : +Return a : +.I shmid_ds : +structure as for : +.BR SHM_STAT . : +However, the : +.I shm_perm.mode : +is not checked for read access for : +.IR shmid , : +resembing the behaviour of : +/proc/sysvipc/shm. : .PP : The caller can prevent or allow swapping of a shared : memory segment with the following \fIcmd\fP values: : @@ -287,7 +300,7 @@ operation returns the index of the highest used entry in the : kernel's internal array recording information about all : shared memory segments. : (This information can be used with repeated : -.B SHM_STAT : +.B SHM_STAT/SHM_STAT_ANY : operations to obtain information about all shared memory segments : on the system.) : A successful : @@ -328,7 +341,7 @@ isn't accessible. : \fIshmid\fP is not a valid identifier, or \fIcmd\fP : is not a valid command. : Or: for a : -.B SHM_STAT : +.B SHM_STAT/SHM_STAT_ANY : operation, the index value specified in : .I shmid : referred to an array slot that is currently unused. This patch (of 3): There is a permission discrepancy when consulting shm ipc object metadata between /proc/sysvipc/shm (0444) and the SHM_STAT shmctl command. The later does permission checks for the object vs S_IRUGO. As such there can be cases where EACCESS is returned via syscall but the info is displayed anyways in the procfs files. While this might have security implications via info leaking (albeit no writing to the shm metadata), this behavior goes way back and showing all the objects regardless of the permissions was most likely an overlook - so we are stuck with it. Furthermore, modifying either the syscall or the procfs file can cause userspace programs to break (ie ipcs). Some applications require getting the procfs info (without root privileges) and can be rather slow in comparison with a syscall -- up to 500x in some reported cases. This patch introduces a new SHM_STAT_ANY command such that the shm ipc object permissions are ignored, and only audited instead. In addition, I've left the lsm security hook checks in place, as if some policy can block the call, then the user has no other choice than just parsing the procfs file. [1] https://lkml.org/lkml/2017/12/19/220 Link: http://lkml.kernel.org/r/20180215162458.10059-2-dave@stgolabs.net Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Acked-by: Michal Hocko <mhocko@suse.com> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Cc: Manfred Spraul <manfred@colorfullife.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Kees Cook <keescook@chromium.org> Cc: Robert Kettler <robert.kettler@outlook.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | | Merge tag 'trace-v4.17' of ↵Linus Torvalds2018-04-101-1/+7
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace Pull tracing updates from Steven Rostedt: "New features: - Tom Zanussi's extended histogram work. This adds the synthetic events to have histograms from multiple event data Adds triggers "onmatch" and "onmax" to call the synthetic events Several updates to the histogram code from this - Allow way to nest ring buffer calls in the same context - Allow absolute time stamps in ring buffer - Rewrite of filter code parsing based on Al Viro's suggestions - Setting of trace_clock to global if TSC is unstable (on boot) - Better OOM handling when allocating large ring buffers - Added initcall tracepoints (consolidated initcall_debug code with them) And other various fixes and clean ups" * tag 'trace-v4.17' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace: (68 commits) init: Have initcall_debug still work without CONFIG_TRACEPOINTS init, tracing: Have printk come through the trace events for initcall_debug init, tracing: instrument security and console initcall trace events init, tracing: Add initcall trace events tracing: Add rcu dereference annotation for test func that touches filter->prog tracing: Add rcu dereference annotation for filter->prog tracing: Fixup logic inversion on setting trace_global_clock defaults tracing: Hide global trace clock from lockdep ring-buffer: Add set/clear_current_oom_origin() during allocations ring-buffer: Check if memory is available before allocation lockdep: Add print_irqtrace_events() to __warn vsprintf: Do not preprocess non-dereferenced pointers for bprintf (%px and %pK) tracing: Uninitialized variable in create_tracing_map_fields() tracing: Make sure variable string fields are NULL-terminated tracing: Add action comparisons when testing matching hist triggers tracing: Don't add flag strings when displaying variable references tracing: Fix display of hist trigger expressions containing timestamps ftrace: Drop a VLA in module_exists() tracing: Mention trace_clock=global when warning about unstable clocks tracing: Default to using trace_global_clock if sched_clock is unstable ...
| * | | init, tracing: instrument security and console initcall trace eventsAbderrahmane Benbachir2018-04-061-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Trace events have been added around the initcall functions defined in init/main.c. But console and security have their own initcalls. This adds the trace events associated for those initcall functions. Link: http://lkml.kernel.org/r/1521765208.19745.2.camel@polymtl.ca Cc: Ingo Molnar <mingo@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Abderrahmane Benbachir <abderrahmane.benbachir@polymtl.ca> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
* | | | selinux: fix missing dput() before selinuxfs unmountStephen Smalley2018-04-091-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Commit 0619f0f5e36f ("selinux: wrap selinuxfs state") triggers a BUG when SELinux is runtime-disabled (i.e. systemd or equivalent disables SELinux before initial policy load via /sys/fs/selinux/disable based on /etc/selinux/config SELINUX=disabled). This does not manifest if SELinux is disabled via kernel command line argument or if SELinux is enabled (permissive or enforcing). Before: SELinux: Disabled at runtime. BUG: Dentry 000000006d77e5c7{i=17,n=null} still in use (1) [unmount of selinuxfs selinuxfs] After: SELinux: Disabled at runtime. Fixes: 0619f0f5e36f ("selinux: wrap selinuxfs state") Reported-by: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | | | Merge branch 'next-integrity' of ↵Linus Torvalds2018-04-0716-68/+207
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security Pull integrity updates from James Morris: "A mixture of bug fixes, code cleanup, and continues to close IMA-measurement, IMA-appraisal, and IMA-audit gaps. Also note the addition of a new cred_getsecid LSM hook by Matthew Garrett: For IMA purposes, we want to be able to obtain the prepared secid in the bprm structure before the credentials are committed. Add a cred_getsecid hook that makes this possible. which is used by a new CREDS_CHECK target in IMA: In ima_bprm_check(), check with both the existing process credentials and the credentials that will be committed when the new process is started. This will not change behaviour unless the system policy is extended to include CREDS_CHECK targets - BPRM_CHECK will continue to check the same credentials that it did previously" * 'next-integrity' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: ima: Fallback to the builtin hash algorithm ima: Add smackfs to the default appraise/measure list evm: check for remount ro in progress before writing ima: Improvements in ima_appraise_measurement() ima: Simplify ima_eventsig_init() integrity: Remove unused macro IMA_ACTION_RULE_FLAGS ima: drop vla in ima_audit_measurement() ima: Fix Kconfig to select TPM 2.0 CRB interface evm: Constify *integrity_status_msg[] evm: Move evm_hmac and evm_hash from evm_main.c to evm_crypto.c fuse: define the filesystem as untrusted ima: fail signature verification based on policy ima: clear IMA_HASH ima: re-evaluate files on privileged mounted filesystems ima: fail file signature verification on non-init mounted filesystems IMA: Support using new creds in appraisal policy security: Add a cred_getsecid hook
| * | | | ima: Fallback to the builtin hash algorithmPetr Vorel2018-03-252-0/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | IMA requires having it's hash algorithm be compiled-in due to it's early use. The default IMA algorithm is protected by Kconfig to be compiled-in. The ima_hash kernel parameter allows to choose the hash algorithm. When the specified algorithm is not available or available as a module, IMA initialization fails, which leads to a kernel panic (mknodat syscall calls ima_post_path_mknod()). Therefore as fallback we force IMA to use the default builtin Kconfig hash algorithm. Fixed crash: $ grep CONFIG_CRYPTO_MD4 .config CONFIG_CRYPTO_MD4=m [ 0.000000] Command line: BOOT_IMAGE=/boot/vmlinuz-4.12.14-2.3-default root=UUID=74ae8202-9ca7-4e39-813b-22287ec52f7a video=1024x768-16 plymouth.ignore-serial-consoles console=ttyS0 console=tty resume=/dev/disk/by-path/pci-0000:00:07.0-part3 splash=silent showopts ima_hash=md4 ... [ 1.545190] ima: Can not allocate md4 (reason: -2) ... [ 2.610120] BUG: unable to handle kernel NULL pointer dereference at (null) [ 2.611903] IP: ima_match_policy+0x23/0x390 [ 2.612967] PGD 0 P4D 0 [ 2.613080] Oops: 0000 [#1] SMP [ 2.613080] Modules linked in: autofs4 [ 2.613080] Supported: Yes [ 2.613080] CPU: 0 PID: 1 Comm: systemd Not tainted 4.12.14-2.3-default #1 [ 2.613080] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.0.0-prebuilt.qemu-project.org 04/01/2014 [ 2.613080] task: ffff88003e2d0040 task.stack: ffffc90000190000 [ 2.613080] RIP: 0010:ima_match_policy+0x23/0x390 [ 2.613080] RSP: 0018:ffffc90000193e88 EFLAGS: 00010296 [ 2.613080] RAX: 0000000000000000 RBX: 000000000000000c RCX: 0000000000000004 [ 2.613080] RDX: 0000000000000010 RSI: 0000000000000001 RDI: ffff880037071728 [ 2.613080] RBP: 0000000000008000 R08: 0000000000000000 R09: 0000000000000000 [ 2.613080] R10: 0000000000000008 R11: 61c8864680b583eb R12: 00005580ff10086f [ 2.613080] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000008000 [ 2.613080] FS: 00007f5c1da08940(0000) GS:ffff88003fc00000(0000) knlGS:0000000000000000 [ 2.613080] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 2.613080] CR2: 0000000000000000 CR3: 0000000037002000 CR4: 00000000003406f0 [ 2.613080] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 2.613080] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 2.613080] Call Trace: [ 2.613080] ? shmem_mknod+0xbf/0xd0 [ 2.613080] ima_post_path_mknod+0x1c/0x40 [ 2.613080] SyS_mknod+0x210/0x220 [ 2.613080] entry_SYSCALL_64_fastpath+0x1a/0xa5 [ 2.613080] RIP: 0033:0x7f5c1bfde570 [ 2.613080] RSP: 002b:00007ffde1c90dc8 EFLAGS: 00000246 ORIG_RAX: 0000000000000085 [ 2.613080] RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f5c1bfde570 [ 2.613080] RDX: 0000000000000000 RSI: 0000000000008000 RDI: 00005580ff10086f [ 2.613080] RBP: 00007ffde1c91040 R08: 00005580ff10086f R09: 0000000000000000 [ 2.613080] R10: 0000000000104000 R11: 0000000000000246 R12: 00005580ffb99660 [ 2.613080] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000002 [ 2.613080] Code: 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 41 57 41 56 44 8d 14 09 41 55 41 54 55 53 44 89 d3 09 cb 48 83 ec 38 48 8b 05 c5 03 29 01 <4c> 8b 20 4c 39 e0 0f 84 d7 01 00 00 4c 89 44 24 08 89 54 24 20 [ 2.613080] RIP: ima_match_policy+0x23/0x390 RSP: ffffc90000193e88 [ 2.613080] CR2: 0000000000000000 [ 2.613080] ---[ end trace 9a9f0a8a73079f6a ]--- [ 2.673052] Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000009 [ 2.673052] [ 2.675337] Kernel Offset: disabled [ 2.676405] ---[ end Kernel panic - not syncing: Attempted to kill init! exitcode=0x00000009 Signed-off-by: Petr Vorel <pvorel@suse.cz> Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
| * | | | ima: Add smackfs to the default appraise/measure listMartin Townsend2018-03-251-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is required to use SMACK and IMA/EVM together. Add it to the default nomeasure/noappraise list like other pseudo filesystems. Signed-off-by: Martin Townsend <mtownsend1973@gmail.com> Acked-by: Casey Schaufler <casey@schaufler-ca.com> Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
| * | | | evm: check for remount ro in progress before writingSascha Hauer2018-03-251-2/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | EVM might update the evm xattr while the VFS performs a remount to readonly mode. This is not properly checked for, additionally check the s_readonly_remount superblock flag before writing. The bug can for example be observed with UBIFS. UBIFS checks the free space on the device before and after a remount. With EVM enabled the free space sometimes differs between both checks. Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de> Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
| * | | | ima: Improvements in ima_appraise_measurement()Thiago Jung Bauermann2018-03-251-13/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Replace nested ifs in the EVM xattr verification logic with a switch statement, making the code easier to understand. Also, add comments to the if statements in the out section and constify the cause variable. Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com> Signed-off-by: Thiago Jung Bauermann <bauerman@linux.vnet.ibm.com> Acked-by: Serge Hallyn <serge@hallyn.com>
| * | | | ima: Simplify ima_eventsig_init()Thiago Jung Bauermann2018-03-251-8/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The "goto out" statement doesn't have any purpose since there's no cleanup to be done when returning early, so remove it. This also makes the rc variable unnecessary so remove it as well. Also, the xattr_len and fmt variables are redundant so remove them as well. Signed-off-by: Thiago Jung Bauermann <bauerman@linux.vnet.ibm.com> Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
OpenPOWER on IntegriCloud