commit 1ea0ce40 ("selinux: allow
changing labels for cgroupfs") broke the Android init program,
which looks up security contexts whenever creating directories
and attempts to assign them via setfscreatecon().
When creating subdirectories in cgroup mounts, this would previously
be ignored since cgroup did not support userspace setting of security
contexts.  However, after the commit, SELinux would attempt to honor
the requested context on cgroup directories and fail due to permission
denial.  Avoid breaking existing userspace/policy by wrapping this change
with a conditional on a new cgroup_seclabel policy capability.  This
preserves existing behavior until/unless a new policy explicitly enables
this capability.
Reported-by: NJohn Stultz <john.stultz@linaro.org>
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <paul@paul-moore.com>
Signed-off-by: NJames Morris <james.l.morris@oracle.com>

Now that %z is standartised in C99 there is no reason to support %Z.
Unlike %L it doesn't even make format strings smaller.

Use BUILD_BUG_ON in a couple ATM drivers.

In case anyone didn't notice lib/vsprintf.o is about half of SLUB which
is in my opinion is quite an achievement.  Hopefully this patch inspires
someone else to trim vsprintf.c more.

Link: http://lkml.kernel.org/r/20170103230126.GA30170@avx2Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Cc: Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Extend SELinux to support distinctions among all network address families
implemented by the kernel by defining new socket security classes
and mapping to them. Otherwise, many sockets are mapped to the generic
socket class and are indistinguishable in policy.  This has come up
previously with regard to selectively allowing access to bluetooth sockets,
and more recently with regard to selectively allowing access to AF_ALG
sockets.  Guido Trentalancia submitted a patch that took a similar approach
to add only support for distinguishing AF_ALG sockets, but this generalizes
his approach to handle all address families implemented by the kernel.
Socket security classes are also added for ICMP and SCTP sockets.
Socket security classes were not defined for AF_* values that are reserved
but unimplemented in the kernel, e.g. AF_NETBEUI, AF_SECURITY, AF_ASH,
AF_ECONET, AF_SNA, AF_WANPIPE.

Backward compatibility is provided by only enabling the finer-grained
socket classes if a new policy capability is set in the policy; older
policies will behave as before.  The legacy redhat1 policy capability
that was only ever used in testing within Fedora for ptrace_child
is reclaimed for this purpose; as far as I can tell, this policy
capability is not enabled in any supported distro policy.

Add a pair of conditional compilation guards to detect when new AF_* values
are added so that we can update SELinux accordingly rather than having to
belatedly update it long after new address families are introduced.
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <paul@paul-moore.com>

Long long ago the kernel log buffer was a buffered stream of bytes, very
much like stdio in user space.  It supported log levels by scanning the
stream and noticing the log level markers at the beginning of each line,
but if you wanted to print a partial line in multiple chunks, you just
did multiple printk() calls, and it just automatically worked.

Except when it didn't, and you had very confusing output when different
lines got all mixed up with each other.  Then you got fragment lines
mixing with each other, or with non-fragment lines, because it was
traditionally impossible to tell whether a printk() call was a
continuation or not.

To at least help clarify the issue of continuation lines, we added a
KERN_CONT marker back in 2007 to mark continuation lines:

  47492527 ("printk: add KERN_CONT annotation").

That continuation marker was initially an empty string, and didn't
actuall make any semantic difference.  But it at least made it possible
to annotate the source code, and have check-patch notice that a printk()
didn't need or want a log level marker, because it was a continuation of
a previous line.

To avoid the ambiguity between a continuation line that had that
KERN_CONT marker, and a printk with no level information at all, we then
in 2009 made KERN_CONT be a real log level marker which meant that we
could now reliably tell the difference between the two cases.

  5fd29d6c ("printk: clean up handling of log-levels and newlines")

and we could take advantage of that to make sure we didn't mix up
continuation lines with lines that just didn't have any loglevel at all.

Then, in 2012, the kernel log buffer was changed to be a "record" based
log, where each line was a record that has a loglevel and a timestamp.

You can see the beginning of that conversion in commits

  e11fea92 ("kmsg: export printk records to the /dev/kmsg interface")
  7ff9554b ("printk: convert byte-buffer to variable-length record buffer")

with a number of follow-up commits to fix some painful fallout from that
conversion.  Over all, it took a couple of months to sort out most of
it.  But the upside was that you could have concurrent readers (and
writers) of the kernel log and not have lines with mixed output in them.

And one particular pain-point for the record-based kernel logging was
exactly the fragmentary lines that are generated in smaller chunks.  In
order to still log them as one recrod, the continuation lines need to be
attached to the previous record properly.

However the explicit continuation record marker that is actually useful
for this exact case was actually removed in aroundm the same time by commit

  61e99ab8 ("printk: remove the now unnecessary "C" annotation for KERN_CONT")

due to the incorrect belief that KERN_CONT wasn't meaningful.  The
ambiguity between "is this a continuation line" or "is this a plain
printk with no log level information" was reintroduced, and in fact
became an even bigger pain point because there was now the whole
record-level merging of kernel messages going on.

This patch reinstates the KERN_CONT as a real non-empty string marker,
so that the ambiguity is fixed once again.

But it's not a plain revert of that original removal: in the four years
since we made KERN_CONT an empty string again, not only has the format
of the log level markers changed, we've also had some usage changes in
this area.

For example, some ACPI code seems to use KERN_CONT _together_ with a log
level, and now uses both the KERN_CONT marker and (for example) a
KERN_INFO marker to show that it's an informational continuation of a
line.

Which is actually not a bad idea - if the continuation line cannot be
attached to its predecessor, without the log level information we don't
know what log level to assign to it (and we traditionally just assigned
it the default loglevel).  So having both a log level and the KERN_CONT
marker is not necessarily a bad idea, but it does mean that we need to
actually iterate over potentially multiple markers, rather than just a
single one.

Also, since KERN_CONT was still conceptually needed, and encouraged, but
didn't actually _do_ anything, we've also had the reverse problem:
rather than having too many annotations it has too few, and there is bit
rot with code that no longer marks the continuation lines with the
KERN_CONT marker.

So this patch not only re-instates the non-empty KERN_CONT marker, it
also fixes up the cases of bit-rot I noticed in my own logs.

There are probably other cases where KERN_CONT will be needed to be
added, either because it is new code that never dealt with the need for
KERN_CONT, or old code that has bitrotted without anybody noticing.

That said, we should strive to avoid the need for KERN_CONT.  It does
result in real problems for logging, and should generally not be seen as
a good feature.  If we some day can get rid of the feature entirely,
because nobody does any fragmented printk calls, that would be lovely.

But until that point, let's at mark the code that relies on the hacky
multi-fragment kernel printk's.  Not only does it avoid the ambiguity,
it also annotates code as "maybe this would be good to fix some day".

(That said, particularly during single-threaded bootup, the downsides of
KERN_CONT are very limited.  Things get much hairier when you have
multiple threads going on and user level reading and writing logs too).
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix to return error code -EINVAL from the error handling case instead
of 0 (rc is overwrite to 0 when policyvers >=
POLICYDB_VERSION_ROLETRANS), as done elsewhere in this function.
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
[PM: normalize "selinux" in patch subject, description line wrap]
Signed-off-by: NPaul Moore <paul@paul-moore.com>

Throughout the SELinux LSM, values taken from sepolicy are
used in places where length == 0 or length == <saturated>
matter, find and fix these.
Signed-off-by: NWilliam Roberts <william.c.roberts@intel.com>
Signed-off-by: NPaul Moore <paul@paul-moore.com>

libsepol pointed out an issue where its possible to have
an unitialized jmp and invalid dereference, fix this.
While we're here, zero allocate all the *_val_to_struct
structures.
Signed-off-by: NWilliam Roberts <william.c.roberts@intel.com>
Signed-off-by: NPaul Moore <paul@paul-moore.com>

When count is 0 and the highbit is not zero, the ebitmap is not
valid and the internal node is not allocated. This causes issues
when routines, like mls_context_isvalid() attempt to use the
ebitmap_for_each_bit() and ebitmap_node_get_bit() as they assume
a highbit > 0 will have a node allocated.
Signed-off-by: NWilliam Roberts <william.c.roberts@intel.com>
Signed-off-by: NPaul Moore <paul@paul-moore.com>

The existing ebitmap_netlbl_import() code didn't correctly handle the
case where the ebitmap_node was not aligned/sized to a power of two,
this patch fixes this (on x86_64 ebitmap_node contains six bitmaps
making a range of 0..383).
Signed-off-by: NPaul Moore <paul@paul-moore.com>

The current bounds checking of both source and target types
requires allowing any domain that has access to the child
domain to also have the same permissions to the parent, which
is undesirable.  Drop the target bounds checking.

KaiGai Kohei originally removed all use of target bounds in
commit 7d52a155 ("selinux: remove dead code in
type_attribute_bounds_av()") but this was reverted in
commit 2ae3ba39 ("selinux: libsepol: remove dead code in
check_avtab_hierarchy_callback()") because it would have
required explicitly allowing the parent any permissions
to the child that the child is allowed to itself.

This change in contrast retains the logic for the case where both
source and target types are bounded, thereby allowing access
if the parent of the source is allowed the corresponding
permissions to the parent of the target.  Further, this change
reworks the logic such that we only perform a single computation
for each case and there is no ambiguity as to how to resolve
a bounds violation.

Under the new logic, if the source type and target types are both
bounded, then the parent of the source type must be allowed the same
permissions to the parent of the target type.  If only the source
type is bounded, then the parent of the source type must be allowed
the same permissions to the target type.

Examples of the new logic and comparisons with the old logic:
1. If we have:
	typebounds A B;
then:
	allow B self:process <permissions>;
will satisfy the bounds constraint iff:
	allow A self:process <permissions>;
is also allowed in policy.

Under the old logic, the allow rule on B satisfies the
bounds constraint if any of the following three are allowed:
	allow A B:process <permissions>; or
	allow B A:process <permissions>; or
	allow A self:process <permissions>;
However, either of the first two ultimately require the third to
satisfy the bounds constraint under the old logic, and therefore
this degenerates to the same result (but is more efficient - we only
need to perform one compute_av call).

2. If we have:
	typebounds A B;
	typebounds A_exec B_exec;
then:
	allow B B_exec:file <permissions>;
will satisfy the bounds constraint iff:
	allow A A_exec:file <permissions>;
is also allowed in policy.

This is essentially the same as #1; it is merely included as
an example of dealing with object types related to a bounded domain
in a manner that satisfies the bounds relationship.  Note that
this approach is preferable to leaving B_exec unbounded and having:
	allow A B_exec:file <permissions>;
in policy because that would allow B's entrypoints to be used to
enter A.  Similarly for _tmp or other related types.

3. If we have:
	typebounds A B;
and an unbounded type T, then:
	allow B T:file <permissions>;
will satisfy the bounds constraint iff:
	allow A T:file <permissions>;
is allowed in policy.

The old logic would have been identical for this example.

4. If we have:
	typebounds A B;
and an unbounded domain D, then:
	allow D B:unix_stream_socket <permissions>;
is not subject to any bounds constraints under the new logic
because D is not bounded.  This is desirable so that we can
allow a domain to e.g. connectto a child domain without having
to allow it to do the same to its parent.

The old logic would have required:
	allow D A:unix_stream_socket <permissions>;
to also be allowed in policy.
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
[PM: re-wrapped description to appease checkpatch.pl]
Signed-off-by: NPaul Moore <paul@paul-moore.com>

security_get_bool_value(int bool) argument "bool" conflicts with
in-kernel macros such as BUILD_BUG().  This patch changes this to
index which isn't a type.

Cc: Paul Moore <paul@paul-moore.com>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Cc: Eric Paris <eparis@parisplace.org>
Cc: James Morris <james.l.morris@oracle.com>
Cc: "Serge E. Hallyn" <serge@hallyn.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Andrew Perepechko <anserper@ya.ru>
Cc: Jeff Vander Stoep <jeffv@google.com>
Cc: selinux@tycho.nsa.gov
Cc: Eric Paris <eparis@redhat.com>
Cc: Paul Moore <pmoore@redhat.com>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: NPrarit Bhargava <prarit@redhat.com>
Acked-by: NDavid Howells <dhowells@redhat.com>
[PM: wrapped description for checkpatch.pl, use "selinux:..." as subj]
Signed-off-by: NPaul Moore <paul@paul-moore.com>

Make validatetrans decisions available through selinuxfs.
"/validatetrans" is added to selinuxfs for this purpose.
This functionality is needed by file system servers
implemented in userspace or kernelspace without the VFS
layer.

Writing "$oldcontext $newcontext $tclass $taskcontext"
to /validatetrans is expected to return 0 if the transition
is allowed and -EPERM otherwise.
Signed-off-by: NAndrew Perepechko <anserper@ya.ru>
CC: andrew.perepechko@seagate.com
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

commit fa1aa143 ("selinux: extended permissions for ioctls")
introduced a bug into the handling of conditional rules, skipping the
processing entirely when the caller does not provide an extended
permissions (xperms) structure.  Access checks from userspace using
/sys/fs/selinux/access do not include such a structure since that
interface does not presently expose extended permission information.
As a result, conditional rules were being ignored entirely on userspace
access requests, producing denials when access was allowed by
conditional rules in the policy.  Fix the bug by only skipping
computation of extended permissions in this situation, not the entire
conditional rules processing.
Reported-by: NLaurent Bigonville <bigon@debian.org>
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
[PM: fixed long lines in patch description]
Cc: stable@vger.kernel.org # 4.3
Signed-off-by: NPaul 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: NRasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

This is much simpler.
Signed-off-by: NRasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Signed-off-by: NRasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul 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: NRasmus Villemoes <linux@rasmusvillemoes.dk>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Add extended permissions logic to selinux. Extended permissions
provides additional permissions in 256 bit increments. Extend the
generic ioctl permission check to use the extended permissions for
per-command filtering. Source/target/class sets including the ioctl
permission may additionally include a set of commands. Example:

allowxperm <source> <target>:<class> ioctl unpriv_app_socket_cmds
auditallowxperm <source> <target>:<class> ioctl priv_gpu_cmds

Where unpriv_app_socket_cmds and priv_gpu_cmds are macros
representing commonly granted sets of ioctl commands.

When ioctl commands are omitted only the permissions are checked.
This feature is intended to provide finer granularity for the ioctl
permission that may be too imprecise. For example, the same driver
may use ioctls to provide important and benign functionality such as
driver version or socket type as well as dangerous capabilities such
as debugging features, read/write/execute to physical memory or
access to sensitive data. Per-command filtering provides a mechanism
to reduce the attack surface of the kernel, and limit applications
to the subset of commands required.

The format of the policy binary has been modified to include ioctl
commands, and the policy version number has been incremented to
POLICYDB_VERSION_XPERMS_IOCTL=30 to account for the format
change.

The extended permissions logic is deliberately generic to allow
components to be reused e.g. netlink filters
Signed-off-by: NJeff Vander Stoep <jeffv@google.com>
Acked-by: NNick Kralevich <nnk@google.com>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

At present we don't create efficient ebitmaps when importing NetLabel
category bitmaps.  This can present a problem when comparing ebitmaps
since ebitmap_cmp() is very strict about these things and considers
these wasteful ebitmaps not equal when compared to their more
efficient counterparts, even if their values are the same.  This isn't
likely to cause problems on 64-bit systems due to a bit of luck on
how NetLabel/CIPSO works and the default ebitmap size, but it can be
a problem on 32-bit systems.

This patch fixes this problem by being a bit more intelligent when
importing NetLabel category bitmaps by skipping over empty sections
which should result in a nice, efficient ebitmap.

Cc: stable@vger.kernel.org # 3.17
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Now that we can safely increase the avtab max buckets without
triggering high order allocations and have a hash function that
will make better use of the larger number of buckets, increase
the max buckets to 2^16.

Original:
101421 entries and 2048/2048 buckets used, longest chain length 374

With new hash function:
101421 entries and 2048/2048 buckets used, longest chain length 81

With increased max buckets:
101421 entries and 31078/32768 buckets used, longest chain length 12
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

This function, based on murmurhash3, has much better distribution than
the original. Using the current default of 2048 buckets, there are many
fewer collisions:

Before:
101421 entries and 2048/2048 buckets used, longest chain length 374
After:
101421 entries and 2048/2048 buckets used, longest chain length 81

The difference becomes much more significant when buckets are increased.
A naive attempt to expand the current function to larger outputs doesn't
yield any significant improvement; so this function is a prerequisite
for increasing the bucket size.

sds:  Adapted from the original patches for libsepol to the kernel.
Signed-off-by: NJohn Brooks <john.brooks@jolla.com>
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Previously we shrank the avtab max hash buckets to avoid
high order memory allocations, but this causes avtab lookups to
degenerate to very long linear searches for the Fedora policy. Convert to
using a flex_array instead so that we can increase the buckets
without such limitations.

This change does not alter the max hash buckets; that is left to a
separate follow-on change.
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Move the NetLabel secattr MLS category import logic into
mls_import_netlbl_cat() where it belongs, and use the
mls_import_netlbl_cat() function in security_netlbl_secattr_to_sid().
Reported-by: NRickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

If hashtab_create() returns a NULL pointer then we should return -ENOMEM
but instead the current code returns success.
Signed-off-by: NDan Carpenter <dan.carpenter@oracle.com>
Acked-by: NSerge Hallyn <serge.hallyn@canonical.com>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Restructure to keyword=value pairs without spaces.  Drop superfluous words in
text.  Make invalid_context a keyword.  Change result= keyword to seresult=.
Signed-off-by: NRichard Guy Briggs <rgb@redhat.com>
[Minor rewrite to the patch subject line]
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Historically the NetLabel LSM secattr catmap functions and data
structures have had very long names which makes a mess of the NetLabel
code and anyone who uses NetLabel.  This patch renames the catmap
functions and structures from "*_secattr_catmap_*" to just "*_catmap_*"
which improves things greatly.

There are no substantial code or logic changes in this patch.
Signed-off-by: NPaul Moore <pmoore@redhat.com>
Tested-by: NCasey Schaufler <casey@schaufler-ca.com>

The NetLabel secattr catmap functions, and the SELinux import/export
glue routines, were broken in many horrible ways and the SELinux glue
code fiddled with the NetLabel catmap structures in ways that we
probably shouldn't allow.  At some point this "worked", but that was
likely due to a bit of dumb luck and sub-par testing (both inflicted
by yours truly).  This patch corrects these problems by basically
gutting the code in favor of something less obtuse and restoring the
NetLabel abstractions in the SELinux catmap glue code.

Everything is working now, and if it decides to break itself in the
future this code will be much easier to debug than the code it
replaces.

One noteworthy side effect of the changes is that it is no longer
necessary to allocate a NetLabel catmap before calling one of the
NetLabel APIs to set a bit in the catmap.  NetLabel will automatically
allocate the catmap nodes when needed, resulting in less allocations
when the lowest bit is greater than 255 and less code in the LSMs.

Cc: stable@vger.kernel.org
Reported-by: NChristian Evans <frodox@zoho.com>
Signed-off-by: NPaul Moore <pmoore@redhat.com>
Tested-by: NCasey Schaufler <casey@schaufler-ca.com>

With the introduction of fair queued rwlock, recursive read_lock()
may hang the offending process if there is a write_lock() somewhere
in between.

With recursive read_lock checking enabled, the following error was
reported:

=============================================
[ INFO: possible recursive locking detected ]
3.16.0-rc1 #2 Tainted: G            E
---------------------------------------------
load_policy/708 is trying to acquire lock:
 (policy_rwlock){.+.+..}, at: [<ffffffff8125b32a>]
security_genfs_sid+0x3a/0x170

but task is already holding lock:
 (policy_rwlock){.+.+..}, at: [<ffffffff8125b48c>]
security_fs_use+0x2c/0x110

other info that might help us debug this:
 Possible unsafe locking scenario:

       CPU0
       ----
  lock(policy_rwlock);
  lock(policy_rwlock);

This patch fixes the occurrence of recursive read_lock() of
policy_rwlock by adding a helper function __security_genfs_sid()
which requires caller to take the lock before calling it. The
security_fs_use() was then modified to call the new helper function.
Signed-off-by: NWaiman Long <Waiman.Long@hp.com>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

The cond_read_node() should free the given node on error path as it's
not linked to p->cond_list yet.  This is done via cond_node_destroy()
but it's not called when next_entry() fails before the expr loop.
Signed-off-by: NNamhyung Kim <namhyung@kernel.org>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

The node->cur_state and len can be read in a single call of next_entry().
And setting len before reading is a dead write so can be eliminated.
Signed-off-by: NNamhyung Kim <namhyung@kernel.org>
(Minor tweak to the length parameter in the call to next_entry())
Signed-off-by: NPaul Moore <pmoore@redhat.com>

There're some code duplication for reading a string value during
policydb_read().  Add str_read() helper to fix it.
Signed-off-by: NNamhyung Kim <namhyung@kernel.org>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

ARRAY_SIZE is more concise to use when the size of an array is divided
by the size of its type or the size of its first element.

The Coccinelle semantic patch that makes this change is as follows:

// <smpl>
@@
type T;
T[] E;
@@

- (sizeof(E)/sizeof(E[...]))
+ ARRAY_SIZE(E)
// </smpl>
Signed-off-by: NHimangi Saraogi <himangi774@gmail.com>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

After silencing the sleeping warning in mls_convert_context() I started
seeing similar traces from hashtab_insert. Do a cond_resched there too.
Signed-off-by: NDave Jones <davej@redhat.com>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

On a slow machine (with debugging enabled), upgrading selinux policy may take
a considerable amount of time. Long enough that the softlockup detector
gets triggered.

The backtrace looks like this..

 > BUG: soft lockup - CPU#2 stuck for 23s! [load_policy:19045]
 > Call Trace:
 >  [<ffffffff81221ddf>] symcmp+0xf/0x20
 >  [<ffffffff81221c27>] hashtab_search+0x47/0x80
 >  [<ffffffff8122e96c>] mls_convert_context+0xdc/0x1c0
 >  [<ffffffff812294e8>] convert_context+0x378/0x460
 >  [<ffffffff81229170>] ? security_context_to_sid_core+0x240/0x240
 >  [<ffffffff812221b5>] sidtab_map+0x45/0x80
 >  [<ffffffff8122bb9f>] security_load_policy+0x3ff/0x580
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff810786dd>] ? sched_clock_local+0x1d/0x80
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff8103096a>] ? __change_page_attr_set_clr+0x82a/0xa50
 >  [<ffffffff810786dd>] ? sched_clock_local+0x1d/0x80
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff8103096a>] ? __change_page_attr_set_clr+0x82a/0xa50
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff81534ddc>] ? retint_restore_args+0xe/0xe
 >  [<ffffffff8109c82d>] ? trace_hardirqs_on_caller+0xfd/0x1c0
 >  [<ffffffff81279a2e>] ? trace_hardirqs_on_thunk+0x3a/0x3f
 >  [<ffffffff810d28a8>] ? rcu_irq_exit+0x68/0xb0
 >  [<ffffffff81534ddc>] ? retint_restore_args+0xe/0xe
 >  [<ffffffff8121e947>] sel_write_load+0xa7/0x770
 >  [<ffffffff81139633>] ? vfs_write+0x1c3/0x200
 >  [<ffffffff81210e8e>] ? security_file_permission+0x1e/0xa0
 >  [<ffffffff8113952b>] vfs_write+0xbb/0x200
 >  [<ffffffff811581c7>] ? fget_light+0x397/0x4b0
 >  [<ffffffff81139c27>] SyS_write+0x47/0xa0
 >  [<ffffffff8153bde4>] tracesys+0xdd/0xe2

Stephen Smalley suggested:

 > Maybe put a cond_resched() within the ebitmap_for_each_positive_bit()
 > loop in mls_convert_context()?

That seems to do the trick. Tested by downgrading and re-upgrading selinux-policy-targeted.
Signed-off-by: NDave Jones <davej@redhat.com>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

After silencing the sleeping warning in mls_convert_context() I started
seeing similar traces from hashtab_insert. Do a cond_resched there too.
Signed-off-by: NDave Jones <davej@redhat.com>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

On a slow machine (with debugging enabled), upgrading selinux policy may take
a considerable amount of time. Long enough that the softlockup detector
gets triggered.

The backtrace looks like this..

 > BUG: soft lockup - CPU#2 stuck for 23s! [load_policy:19045]
 > Call Trace:
 >  [<ffffffff81221ddf>] symcmp+0xf/0x20
 >  [<ffffffff81221c27>] hashtab_search+0x47/0x80
 >  [<ffffffff8122e96c>] mls_convert_context+0xdc/0x1c0
 >  [<ffffffff812294e8>] convert_context+0x378/0x460
 >  [<ffffffff81229170>] ? security_context_to_sid_core+0x240/0x240
 >  [<ffffffff812221b5>] sidtab_map+0x45/0x80
 >  [<ffffffff8122bb9f>] security_load_policy+0x3ff/0x580
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff810786dd>] ? sched_clock_local+0x1d/0x80
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff8103096a>] ? __change_page_attr_set_clr+0x82a/0xa50
 >  [<ffffffff810786dd>] ? sched_clock_local+0x1d/0x80
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff8103096a>] ? __change_page_attr_set_clr+0x82a/0xa50
 >  [<ffffffff810788a8>] ? sched_clock_cpu+0xa8/0x100
 >  [<ffffffff81534ddc>] ? retint_restore_args+0xe/0xe
 >  [<ffffffff8109c82d>] ? trace_hardirqs_on_caller+0xfd/0x1c0
 >  [<ffffffff81279a2e>] ? trace_hardirqs_on_thunk+0x3a/0x3f
 >  [<ffffffff810d28a8>] ? rcu_irq_exit+0x68/0xb0
 >  [<ffffffff81534ddc>] ? retint_restore_args+0xe/0xe
 >  [<ffffffff8121e947>] sel_write_load+0xa7/0x770
 >  [<ffffffff81139633>] ? vfs_write+0x1c3/0x200
 >  [<ffffffff81210e8e>] ? security_file_permission+0x1e/0xa0
 >  [<ffffffff8113952b>] vfs_write+0xbb/0x200
 >  [<ffffffff811581c7>] ? fget_light+0x397/0x4b0
 >  [<ffffffff81139c27>] SyS_write+0x47/0xa0
 >  [<ffffffff8153bde4>] tracesys+0xdd/0xe2

Stephen Smalley suggested:

 > Maybe put a cond_resched() within the ebitmap_for_each_positive_bit()
 > loop in mls_convert_context()?

That seems to do the trick. Tested by downgrading and re-upgrading selinux-policy-targeted.
Signed-off-by: NDave Jones <davej@redhat.com>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Signed-off-by: NPaul Moore <pmoore@redhat.com>

security_xfrm_policy_alloc can be called in atomic context so the
allocation should be done with GFP_ATOMIC. Add an argument to let the
callers choose the appropriate way. In order to do so a gfp argument
needs to be added to the method xfrm_policy_alloc_security in struct
security_operations and to the internal function
selinux_xfrm_alloc_user. After that switch to GFP_ATOMIC in the atomic
callers and leave GFP_KERNEL as before for the rest.
The path that needed the gfp argument addition is:
security_xfrm_policy_alloc -> security_ops.xfrm_policy_alloc_security ->
all users of xfrm_policy_alloc_security (e.g. selinux_xfrm_policy_alloc) ->
selinux_xfrm_alloc_user (here the allocation used to be GFP_KERNEL only)

Now adding a gfp argument to selinux_xfrm_alloc_user requires us to also
add it to security_context_to_sid which is used inside and prior to this
patch did only GFP_KERNEL allocation. So add gfp argument to
security_context_to_sid and adjust all of its callers as well.

CC: Paul Moore <paul@paul-moore.com>
CC: Dave Jones <davej@redhat.com>
CC: Steffen Klassert <steffen.klassert@secunet.com>
CC: Fan Du <fan.du@windriver.com>
CC: David S. Miller <davem@davemloft.net>
CC: LSM list <linux-security-module@vger.kernel.org>
CC: SELinux list <selinux@tycho.nsa.gov>
Signed-off-by: NNikolay Aleksandrov <nikolay@redhat.com>
Acked-by: NPaul Moore <paul@paul-moore.com>
Signed-off-by: NSteffen Klassert <steffen.klassert@secunet.com>

When writing policy via /sys/fs/selinux/policy I wrote the type and class
of filename trans rules in CPU endian instead of little endian.  On
x86_64 this works just fine, but it means that on big endian arch's like
ppc64 and s390 userspace reads the policy and converts it from
le32_to_cpu.  So the values are all screwed up.  Write the values in le
format like it should have been to start.
Signed-off-by: NEric Paris <eparis@redhat.com>
Acked-by: NStephen Smalley <sds@tycho.nsa.gov>
Cc: stable@vger.kernel.org
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Setting an empty security context (length=0) on a file will
lead to incorrectly dereferencing the type and other fields
of the security context structure, yielding a kernel BUG.
As a zero-length security context is never valid, just reject
all such security contexts whether coming from userspace
via setxattr or coming from the filesystem upon a getxattr
request by SELinux.

Setting a security context value (empty or otherwise) unknown to
SELinux in the first place is only possible for a root process
(CAP_MAC_ADMIN), and, if running SELinux in enforcing mode, only
if the corresponding SELinux mac_admin permission is also granted
to the domain by policy.  In Fedora policies, this is only allowed for
specific domains such as livecd for setting down security contexts
that are not defined in the build host policy.

Reproducer:
su
setenforce 0
touch foo
setfattr -n security.selinux foo

Caveat:
Relabeling or removing foo after doing the above may not be possible
without booting with SELinux disabled.  Any subsequent access to foo
after doing the above will also trigger the BUG.

BUG output from Matthew Thode:
[  473.893141] ------------[ cut here ]------------
[  473.962110] kernel BUG at security/selinux/ss/services.c:654!
[  473.995314] invalid opcode: 0000 [#6] SMP
[  474.027196] Modules linked in:
[  474.058118] CPU: 0 PID: 8138 Comm: ls Tainted: G      D   I
3.13.0-grsec #1
[  474.116637] Hardware name: Supermicro X8ST3/X8ST3, BIOS 2.0
07/29/10
[  474.149768] task: ffff8805f50cd010 ti: ffff8805f50cd488 task.ti:
ffff8805f50cd488
[  474.183707] RIP: 0010:[<ffffffff814681c7>]  [<ffffffff814681c7>]
context_struct_compute_av+0xce/0x308
[  474.219954] RSP: 0018:ffff8805c0ac3c38  EFLAGS: 00010246
[  474.252253] RAX: 0000000000000000 RBX: ffff8805c0ac3d94 RCX:
0000000000000100
[  474.287018] RDX: ffff8805e8aac000 RSI: 00000000ffffffff RDI:
ffff8805e8aaa000
[  474.321199] RBP: ffff8805c0ac3cb8 R08: 0000000000000010 R09:
0000000000000006
[  474.357446] R10: 0000000000000000 R11: ffff8805c567a000 R12:
0000000000000006
[  474.419191] R13: ffff8805c2b74e88 R14: 00000000000001da R15:
0000000000000000
[  474.453816] FS:  00007f2e75220800(0000) GS:ffff88061fc00000(0000)
knlGS:0000000000000000
[  474.489254] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  474.522215] CR2: 00007f2e74716090 CR3: 00000005c085e000 CR4:
00000000000207f0
[  474.556058] Stack:
[  474.584325]  ffff8805c0ac3c98 ffffffff811b549b ffff8805c0ac3c98
ffff8805f1190a40
[  474.618913]  ffff8805a6202f08 ffff8805c2b74e88 00068800d0464990
ffff8805e8aac860
[  474.653955]  ffff8805c0ac3cb8 000700068113833a ffff880606c75060
ffff8805c0ac3d94
[  474.690461] Call Trace:
[  474.723779]  [<ffffffff811b549b>] ? lookup_fast+0x1cd/0x22a
[  474.778049]  [<ffffffff81468824>] security_compute_av+0xf4/0x20b
[  474.811398]  [<ffffffff8196f419>] avc_compute_av+0x2a/0x179
[  474.843813]  [<ffffffff8145727b>] avc_has_perm+0x45/0xf4
[  474.875694]  [<ffffffff81457d0e>] inode_has_perm+0x2a/0x31
[  474.907370]  [<ffffffff81457e76>] selinux_inode_getattr+0x3c/0x3e
[  474.938726]  [<ffffffff81455cf6>] security_inode_getattr+0x1b/0x22
[  474.970036]  [<ffffffff811b057d>] vfs_getattr+0x19/0x2d
[  475.000618]  [<ffffffff811b05e5>] vfs_fstatat+0x54/0x91
[  475.030402]  [<ffffffff811b063b>] vfs_lstat+0x19/0x1b
[  475.061097]  [<ffffffff811b077e>] SyS_newlstat+0x15/0x30
[  475.094595]  [<ffffffff8113c5c1>] ? __audit_syscall_entry+0xa1/0xc3
[  475.148405]  [<ffffffff8197791e>] system_call_fastpath+0x16/0x1b
[  475.179201] Code: 00 48 85 c0 48 89 45 b8 75 02 0f 0b 48 8b 45 a0 48
8b 3d 45 d0 b6 00 8b 40 08 89 c6 ff ce e8 d1 b0 06 00 48 85 c0 49 89 c7
75 02 <0f> 0b 48 8b 45 b8 4c 8b 28 eb 1e 49 8d 7d 08 be 80 01 00 00 e8
[  475.255884] RIP  [<ffffffff814681c7>]
context_struct_compute_av+0xce/0x308
[  475.296120]  RSP <ffff8805c0ac3c38>
[  475.328734] ---[ end trace f076482e9d754adc ]---
Reported-by: NMatthew Thode <mthode@mthode.org>
Signed-off-by: NStephen Smalley <sds@tycho.nsa.gov>
Cc: stable@vger.kernel.org
Signed-off-by: NPaul Moore <pmoore@redhat.com>

Two of the conditions in selinux_audit_rule_match() should never happen and
the third indicates a race that should be retried.  Remove the calls to
audit_log() (which call audit_log_start()) and deal with the errors in the
caller, logging only once if the condition is met.  Calling audit_log_start()
in this location makes buffer allocation and locking more complicated in the
calling tree (audit_filter_user()).
Signed-off-by: NRichard Guy Briggs <rgb@redhat.com>
Signed-off-by: NEric Paris <eparis@redhat.com>