While using AppArmor, SYS_CAP_RESOURCE is insufficient to call prlimit
on another task. The only other example of a AppArmor mediating access to
another, already running, task (ignoring fork+exec) is ptrace.
The AppArmor model for ptrace is that one of the following must be true:
1) The tracer is unconfined
2) The tracer is in complain mode
3) The tracer and tracee are confined by the same profile
4) The tracer is confined but has SYS_CAP_PTRACE
1), 2, and 3) are already true for setrlimit.
We can match the ptrace model just by allowing CAP_SYS_RESOURCE.
We still test the values of the rlimit since it can always be overridden
using a value that means unlimited for a particular resource.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
list_next_entry has been defined in list.h, so I replace list_entry_next
with it.
Signed-off-by: Geliang Tang <geliangtang@163.com>
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: John Johansen <john.johansen@canonical.com>
When finding a child profile via an rcu critical section, the profile
may be put and scheduled for deletion after the child is found but
before its refcount is incremented.
Protect against this by repeating the lookup if the profiles refcount
is 0 and is one its way to deletion.
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
The target profile name was not being correctly audited in a few
cases because the target variable was not being set and gotos
passed the code to set it at apply:
Since it is always based on new_profile just drop the target var
and conditionally report based on new_profile.
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
Currently logging of a successful profile load only logs the basename
of the profile. This can result in confusion when a child profile has
the same name as the another profile in the set. Logging the hname
will ensure there is no confusion.
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
currently only the profile that is causing the failure is logged. This
makes it more confusing than necessary about which profiles loaded
and which didn't. So make sure to log success and failure messages for
all profiles in the set being loaded.
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
Internal mounts are not mounted anywhere and as such should be treated
as disconnected paths.
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
Bind mounts can fail to be properly reconnected when PATH_CONNECT is
specified. Ensure that when PATH_CONNECT is specified the path has
a root.
BugLink: http://bugs.launchpad.net/bugs/1319984
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
The current behavior is confusing as it causes exec failures to report
the executable is missing instead of identifying that apparmor
caused the failure.
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
When set atomic replacement is used and the parent is updated before the
child, and the child did not exist in the old parent so there is no
direct replacement then the new child is incorrectly added to the old
parent. This results in the new parent not having the child(ren) that
it should and the old parent when being destroyed asserting the
following error.
AppArmor: policy_destroy: internal error, policy '<profile/name>' still
contains profiles
Signed-off-by: John Johansen <john.johansen@canonical.com>
Acked-by: Seth Arnold <seth.arnold@canonical.com>
Extend the PCR supplied as a parameter, instead of assuming that the
measurement entry uses the default configured PCR.
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
IMA avoids re-measuring files by storing the current state as a flag in
the integrity cache. It will then skip adding a new measurement log entry
if the cache reports the file as already measured.
If a policy measures an already measured file to a new PCR, the measurement
will not be added to the list. This patch implements a new bitfield for
specifying which PCR the file was measured into, rather than if it was
measured.
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Template entry duplicates are prevented from being added to the
measurement list by checking a hash table that contains the template
entry digests. However, the PCR value is not included in this comparison,
so duplicate template entry digests with differing PCRs may be dropped.
This patch redefines duplicate template entries as template entries with
the same digest and same PCR values.
Reported-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
IMA assumes that the same default Kconfig PCR is extended for each
entry. This patch replaces the default configured PCR with the policy
defined PCR.
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
The IMA measurement list entries include the Kconfig defined PCR value.
This patch defines a new ima_template_entry field for including the PCR
as specified in the policy rule.
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Different policy rules may extend different PCRs. This patch retrieves
the specific PCR for the matched rule. Subsequent patches will include
the rule specific PCR in the measurement list and extend the appropriate
PCR.
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
This patch defines a new IMA measurement policy rule option "pcr=",
which allows extending different PCRs on a per rule basis. For example,
the system independent files could extend the default IMA Kconfig
specified PCR, while the system dependent files could extend a different
PCR.
The following is an example of this usage with an SELinux policy; the
rule would extend PCR 11 with system configuration files:
measure func=FILE_CHECK mask=MAY_READ obj_type=system_conf_t pcr=11
Changelog v3:
- FIELD_SIZEOF returns bytes, not bits. Fixed INVALID_PCR
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
To keep track of which measurements have been extended to which PCRs, this
patch defines a new integrity_iint_cache field named measured_pcrs. This
field is a bitmask of the PCRs measured. Each bit corresponds to a PCR
index. For example, bit 10 corresponds to PCR 10.
Signed-off-by: Eric Richter <erichte@linux.vnet.ibm.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
This works in exactly the same way as the CIPSO label cache.
The idea is to allow the lsm to cache the result of a secattr
lookup so that it doesn't need to perform the lookup for
every skbuff.
It introduces two sysctl controls:
calipso_cache_enable - enables/disables the cache.
calipso_cache_bucket_size - sets the size of a cache bucket.
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
This makes it possible to route the error to the appropriate
labelling engine. CALIPSO is far less verbose than CIPSO
when encountering a bogus packet, so there is no need for a
CALIPSO error handler.
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
In some cases, the lsm needs to add the label to the skbuff directly.
A NF_INET_LOCAL_OUT IPv6 hook is added to selinux to match the IPv4
behaviour. This allows selinux to label the skbuffs that it requires.
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
Request sockets need to have a label that takes into account the
incoming connection as well as their parent's label. This is used
for the outgoing SYN-ACK and for their child full-socket.
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
If a socket has a netlabel in place then don't let setsockopt() alter
the socket's IPv6 hop-by-hop option. This is in the same spirit as
the existing check for IPv4.
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
CALIPSO is a hop-by-hop IPv6 option. A lot of this patch is based on
the equivalent CISPO code. The main difference is due to manipulating
the options in the hop-by-hop header.
Signed-off-by: Huw Davies <huw@codeweavers.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
avc_cache_threshold is of type unsigned int. Do not use a signed
new_value in sscanf(page, "%u", &new_value).
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
[PM: subject prefix fix, description cleanup]
Signed-off-by: Paul 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: Paul Moore <paul@paul-moore.com>
Kill with signal number 0 is commonly used for checking PID existence.
Smack treated such cases like any other kills, although no signal is
actually delivered when sig == 0.
Checking permissions when sig == 0 didn't prevent an unprivileged caller
from learning whether PID exists or not. When it existed, kernel returned
EPERM, when it didn't - ESRCH. The only effect of policy check in such
case is noise in audit logs.
This change lets Smack silently ignore kill() invocations with sig == 0.
Signed-off-by: Rafal Krypa <r.krypa@samsung.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
The code is doing the equivalent of the kthread_run macro.
Signed-off-by: Mike Danese <mikedanese@google.com>
Acked-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: James Morris <james.l.morris@oracle.com>
Serge Hallyn pointed out that the current implementation of
security_inode_getsecurity() works if there is only one hook
provided for it, but will fail if there is more than one and
the attribute requested isn't supplied by the first module.
This isn't a problem today, since only SELinux and Smack
provide this hook and there is (currently) no way to enable
both of those modules at the same time. Serge, however, wants
to introduce a capability attribute and an inode_getsecurity
hook in the capability security module to handle it. This
addresses that upcoming problem, will be required for "extreme
stacking" and is just a better implementation.
Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
Acked-by: Serge Hallyn <serge@hallyn.com>
Signed-off-by: James Morris <james.l.morris@oracle.com>
The values computed during Diffie-Hellman key exchange are often used
in combination with key derivation functions to create cryptographic
keys. Add a placeholder for a later implementation to configure a
key derivation function that will transform the Diffie-Hellman
result returned by the KEYCTL_DH_COMPUTE command.
[This patch was stripped down from a patch produced by Mat Martineau that
had a bug in the compat code - so for the moment Stephan's patch simply
requires that the placeholder argument must be NULL]
Original-signed-off-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: Stephan Mueller <smueller@chronox.de>
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: James Morris <james.l.morris@oracle.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 7d52a155e3 ("selinux: remove dead code in
type_attribute_bounds_av()") but this was reverted in
commit 2ae3ba3938 ("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: Stephen Smalley <sds@tycho.nsa.gov>
[PM: re-wrapped description to appease checkpatch.pl]
Signed-off-by: Paul Moore <paul@paul-moore.com>
Pull vfs fixes from Al Viro:
"Followups to the parallel lookup work:
- update docs
- restore killability of the places that used to take ->i_mutex
killably now that we have down_write_killable() merged
- Additionally, it turns out that I missed a prerequisite for
security_d_instantiate() stuff - ->getxattr() wasn't the only thing
that could be called before dentry is attached to inode; with smack
we needed the same treatment applied to ->setxattr() as well"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
switch ->setxattr() to passing dentry and inode separately
switch xattr_handler->set() to passing dentry and inode separately
restore killability of old mutex_lock_killable(&inode->i_mutex) users
add down_write_killable_nested()
update D/f/directory-locking
smack ->d_instantiate() uses ->setxattr(), so to be able to call it before
we'd hashed the new dentry and attached it to inode, we need ->setxattr()
instances getting the inode as an explicit argument rather than obtaining
it from dentry.
Similar change for ->getxattr() had been done in commit ce23e64. Unlike
->getxattr() (which is used by both selinux and smack instances of
->d_instantiate()) ->setxattr() is used only by smack one and unfortunately
it got missed back then.
Reported-by: Seung-Woo Kim <sw0312.kim@samsung.com>
Tested-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Commit 8a56038c2a ("Yama: consolidate error reporting") causes lockups
when someone hits a Yama denial. Call chain:
process_vm_readv -> process_vm_rw -> process_vm_rw_core -> mm_access
-> ptrace_may_access
task_lock(...) is taken
__ptrace_may_access -> security_ptrace_access_check
-> yama_ptrace_access_check -> report_access -> kstrdup_quotable_cmdline
-> get_cmdline -> access_process_vm -> get_task_mm
task_lock(...) is taken again
task_lock(p) just calls spin_lock(&p->alloc_lock), so at this point,
spin_lock() is called on a lock that is already held by the current
process.
Also: Since the alloc_lock is a spinlock, sleeping inside
security_ptrace_access_check hooks is probably not allowed at all? So it's
not even possible to print the cmdline from in there because that might
involve paging in userspace memory.
It would be tempting to rewrite ptrace_may_access() to drop the alloc_lock
before calling the LSM, but even then, ptrace_may_access() itself might be
called from various contexts in which you're not allowed to sleep; for
example, as far as I understand, to be able to hold a reference to another
task, usually an RCU read lock will be taken (see e.g. kcmp() and
get_robust_list()), so that also prohibits sleeping. (And using e.g. FUSE,
a user can cause pagefault handling to take arbitrary amounts of time -
see https://bugs.chromium.org/p/project-zero/issues/detail?id=808.)
Therefore, AFAIK, in order to print the name of a process below
security_ptrace_access_check(), you'd have to either grab a reference to
the mm_struct and defer the access violation reporting or just use the
"comm" value that's stored in kernelspace and accessible without big
complications. (Or you could try to use some kind of atomic remote VM
access that fails if the memory isn't paged in, similar to
copy_from_user_inatomic(), and if necessary fall back to comm, but
that'd be kind of ugly because the comm/cmdline choice would look
pretty random to the user.)
Fix it by deferring reporting of the access violation until current
exits kernelspace the next time.
v2: Don't oops on PTRACE_TRACEME, call report_access under
task_lock(current). Also fix nonsensical comment. And don't use
GPF_ATOMIC for memory allocation with no locks held.
This patch is tested both for ptrace attach and ptrace traceme.
Fixes: 8a56038c2a ("Yama: consolidate error reporting")
Signed-off-by: Jann Horn <jann@thejh.net>
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: James Morris <james.l.morris@oracle.com>
Pull security subsystem updates from James Morris:
"Highlights:
- A new LSM, "LoadPin", from Kees Cook is added, which allows forcing
of modules and firmware to be loaded from a specific device (this
is from ChromeOS, where the device as a whole is verified
cryptographically via dm-verity).
This is disabled by default but can be configured to be enabled by
default (don't do this if you don't know what you're doing).
- Keys: allow authentication data to be stored in an asymmetric key.
Lots of general fixes and updates.
- SELinux: add restrictions for loading of kernel modules via
finit_module(). Distinguish non-init user namespace capability
checks. Apply execstack check on thread stacks"
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (48 commits)
LSM: LoadPin: provide enablement CONFIG
Yama: use atomic allocations when reporting
seccomp: Fix comment typo
ima: add support for creating files using the mknodat syscall
ima: fix ima_inode_post_setattr
vfs: forbid write access when reading a file into memory
fs: fix over-zealous use of "const"
selinux: apply execstack check on thread stacks
selinux: distinguish non-init user namespace capability checks
LSM: LoadPin for kernel file loading restrictions
fs: define a string representation of the kernel_read_file_id enumeration
Yama: consolidate error reporting
string_helpers: add kstrdup_quotable_file
string_helpers: add kstrdup_quotable_cmdline
string_helpers: add kstrdup_quotable
selinux: check ss_initialized before revalidating an inode label
selinux: delay inode label lookup as long as possible
selinux: don't revalidate an inode's label when explicitly setting it
selinux: Change bool variable name to index.
KEYS: Add KEYCTL_DH_COMPUTE command
...
Pull networking updates from David Miller:
"Highlights:
1) Support SPI based w5100 devices, from Akinobu Mita.
2) Partial Segmentation Offload, from Alexander Duyck.
3) Add GMAC4 support to stmmac driver, from Alexandre TORGUE.
4) Allow cls_flower stats offload, from Amir Vadai.
5) Implement bpf blinding, from Daniel Borkmann.
6) Optimize _ASYNC_ bit twiddling on sockets, unless the socket is
actually using FASYNC these atomics are superfluous. From Eric
Dumazet.
7) Run TCP more preemptibly, also from Eric Dumazet.
8) Support LED blinking, EEPROM dumps, and rxvlan offloading in mlx5e
driver, from Gal Pressman.
9) Allow creating ppp devices via rtnetlink, from Guillaume Nault.
10) Improve BPF usage documentation, from Jesper Dangaard Brouer.
11) Support tunneling offloads in qed, from Manish Chopra.
12) aRFS offloading in mlx5e, from Maor Gottlieb.
13) Add RFS and RPS support to SCTP protocol, from Marcelo Ricardo
Leitner.
14) Add MSG_EOR support to TCP, this allows controlling packet
coalescing on application record boundaries for more accurate
socket timestamp sampling. From Martin KaFai Lau.
15) Fix alignment of 64-bit netlink attributes across the board, from
Nicolas Dichtel.
16) Per-vlan stats in bridging, from Nikolay Aleksandrov.
17) Several conversions of drivers to ethtool ksettings, from Philippe
Reynes.
18) Checksum neutral ILA in ipv6, from Tom Herbert.
19) Factorize all of the various marvell dsa drivers into one, from
Vivien Didelot
20) Add VF support to qed driver, from Yuval Mintz"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1649 commits)
Revert "phy dp83867: Fix compilation with CONFIG_OF_MDIO=m"
Revert "phy dp83867: Make rgmii parameters optional"
r8169: default to 64-bit DMA on recent PCIe chips
phy dp83867: Make rgmii parameters optional
phy dp83867: Fix compilation with CONFIG_OF_MDIO=m
bpf: arm64: remove callee-save registers use for tmp registers
asix: Fix offset calculation in asix_rx_fixup() causing slow transmissions
switchdev: pass pointer to fib_info instead of copy
net_sched: close another race condition in tcf_mirred_release()
tipc: fix nametable publication field in nl compat
drivers: net: Don't print unpopulated net_device name
qed: add support for dcbx.
ravb: Add missing free_irq() calls to ravb_close()
qed: Remove a stray tab
net: ethernet: fec-mpc52xx: use phy_ethtool_{get|set}_link_ksettings
net: ethernet: fec-mpc52xx: use phydev from struct net_device
bpf, doc: fix typo on bpf_asm descriptions
stmmac: hardware TX COE doesn't work when force_thresh_dma_mode is set
net: ethernet: fs-enet: use phy_ethtool_{get|set}_link_ksettings
net: ethernet: fs-enet: use phydev from struct net_device
...
Pull 'struct path' constification update from Al Viro:
"'struct path' is passed by reference to a bunch of Linux security
methods; in theory, there's nothing to stop them from modifying the
damn thing and LSM community being what it is, sooner or later some
enterprising soul is going to decide that it's a good idea.
Let's remove the temptation and constify all of those..."
* 'work.const-path' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
constify ima_d_path()
constify security_sb_pivotroot()
constify security_path_chroot()
constify security_path_{link,rename}
apparmor: remove useless checks for NULL ->mnt
constify security_path_{mkdir,mknod,symlink}
constify security_path_{unlink,rmdir}
apparmor: constify common_perm_...()
apparmor: constify aa_path_link()
apparmor: new helper - common_path_perm()
constify chmod_common/security_path_chmod
constify security_sb_mount()
constify chown_common/security_path_chown
tomoyo: constify assorted struct path *
apparmor_path_truncate(): path->mnt is never NULL
constify vfs_truncate()
constify security_path_truncate()
[apparmor] constify struct path * in a bunch of helpers
Pull parallel filesystem directory handling update from Al Viro.
This is the main parallel directory work by Al that makes the vfs layer
able to do lookup and readdir in parallel within a single directory.
That's a big change, since this used to be all protected by the
directory inode mutex.
The inode mutex is replaced by an rwsem, and serialization of lookups of
a single name is done by a "in-progress" dentry marker.
The series begins with xattr cleanups, and then ends with switching
filesystems over to actually doing the readdir in parallel (switching to
the "iterate_shared()" that only takes the read lock).
A more detailed explanation of the process from Al Viro:
"The xattr work starts with some acl fixes, then switches ->getxattr to
passing inode and dentry separately. This is the point where the
things start to get tricky - that got merged into the very beginning
of the -rc3-based #work.lookups, to allow untangling the
security_d_instantiate() mess. The xattr work itself proceeds to
switch a lot of filesystems to generic_...xattr(); no complications
there.
After that initial xattr work, the series then does the following:
- untangle security_d_instantiate()
- convert a bunch of open-coded lookup_one_len_unlocked() to calls of
that thing; one such place (in overlayfs) actually yields a trivial
conflict with overlayfs fixes later in the cycle - overlayfs ended
up switching to a variant of lookup_one_len_unlocked() sans the
permission checks. I would've dropped that commit (it gets
overridden on merge from #ovl-fixes in #for-next; proper resolution
is to use the variant in mainline fs/overlayfs/super.c), but I
didn't want to rebase the damn thing - it was fairly late in the
cycle...
- some filesystems had managed to depend on lookup/lookup exclusion
for *fs-internal* data structures in a way that would break if we
relaxed the VFS exclusion. Fixing hadn't been hard, fortunately.
- core of that series - parallel lookup machinery, replacing
->i_mutex with rwsem, making lookup_slow() take it only shared. At
that point lookups happen in parallel; lookups on the same name
wait for the in-progress one to be done with that dentry.
Surprisingly little code, at that - almost all of it is in
fs/dcache.c, with fs/namei.c changes limited to lookup_slow() -
making it use the new primitive and actually switching to locking
shared.
- parallel readdir stuff - first of all, we provide the exclusion on
per-struct file basis, same as we do for read() vs lseek() for
regular files. That takes care of most of the needed exclusion in
readdir/readdir; however, these guys are trickier than lookups, so
I went for switching them one-by-one. To do that, a new method
'->iterate_shared()' is added and filesystems are switched to it
as they are either confirmed to be OK with shared lock on directory
or fixed to be OK with that. I hope to kill the original method
come next cycle (almost all in-tree filesystems are switched
already), but it's still not quite finished.
- several filesystems get switched to parallel readdir. The
interesting part here is dealing with dcache preseeding by readdir;
that needs minor adjustment to be safe with directory locked only
shared.
Most of the filesystems doing that got switched to in those
commits. Important exception: NFS. Turns out that NFS folks, with
their, er, insistence on VFS getting the fuck out of the way of the
Smart Filesystem Code That Knows How And What To Lock(tm) have
grown the locking of their own. They had their own homegrown
rwsem, with lookup/readdir/atomic_open being *writers* (sillyunlink
is the reader there). Of course, with VFS getting the fuck out of
the way, as requested, the actual smarts of the smart filesystem
code etc. had become exposed...
- do_last/lookup_open/atomic_open cleanups. As the result, open()
without O_CREAT locks the directory only shared. Including the
->atomic_open() case. Backmerge from #for-linus in the middle of
that - atomic_open() fix got brought in.
- then comes NFS switch to saner (VFS-based ;-) locking, killing the
homegrown "lookup and readdir are writers" kinda-sorta rwsem. All
exclusion for sillyunlink/lookup is done by the parallel lookups
mechanism. Exclusion between sillyunlink and rmdir is a real rwsem
now - rmdir being the writer.
Result: NFS lookups/readdirs/O_CREAT-less opens happen in parallel
now.
- the rest of the series consists of switching a lot of filesystems
to parallel readdir; in a lot of cases ->llseek() gets simplified
as well. One backmerge in there (again, #for-linus - rockridge
fix)"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (74 commits)
ext4: switch to ->iterate_shared()
hfs: switch to ->iterate_shared()
hfsplus: switch to ->iterate_shared()
hostfs: switch to ->iterate_shared()
hpfs: switch to ->iterate_shared()
hpfs: handle allocation failures in hpfs_add_pos()
gfs2: switch to ->iterate_shared()
f2fs: switch to ->iterate_shared()
afs: switch to ->iterate_shared()
befs: switch to ->iterate_shared()
befs: constify stuff a bit
isofs: switch to ->iterate_shared()
get_acorn_filename(): deobfuscate a bit
btrfs: switch to ->iterate_shared()
logfs: no need to lock directory in lseek
switch ecryptfs to ->iterate_shared
9p: switch to ->iterate_shared()
fat: switch to ->iterate_shared()
romfs, squashfs: switch to ->iterate_shared()
more trivial ->iterate_shared conversions
...
