This patchset continues the work that began in the sysv ipc semaphore
scaling series, see
https://lkml.org/lkml/2013/3/20/546
Just like semaphores used to be, sysv shared memory and msg queues also
abuse the ipc lock, unnecessarily holding it for operations such as
permission and security checks.
This patchset mostly deals with mqueues, and while shared mem can be
done in a very similar way, I want to get these patches out in the open
first. It also does some pending cleanups, mostly focused on the two
level locking we have in ipc code, taking care of ipc_addid() and
ipcctl_pre_down_nolock() - yes there are still functions that need to be
updated as well.
This patch:
Make all callers explicitly take and release the RCU read lock.
This addresses the two level locking seen in newary(), newseg() and
newqueue(). For the last two, explicitly unlock the ipc object and the
rcu lock, instead of calling the custom shm_unlock and msg_unlock
functions. The next patch will deal with the open coded locking for
->perm.lock
Signed-off-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The old audit PATH records for mq_open looked like this:
type=PATH msg=audit(1366282323.982:869): item=1 name=(null) inode=6777
dev=00:0c mode=041777 ouid=0 ogid=0 rdev=00:00
obj=system_u:object_r:tmpfs_t:s15:c0.c1023
type=PATH msg=audit(1366282323.982:869): item=0 name="test_mq" inode=26732
dev=00:0c mode=0100700 ouid=0 ogid=0 rdev=00:00
obj=staff_u:object_r:user_tmpfs_t:s15:c0.c1023
...with the audit related changes that went into 3.7, they now look like this:
type=PATH msg=audit(1366282236.776:3606): item=2 name=(null) inode=66655
dev=00:0c mode=0100700 ouid=0 ogid=0 rdev=00:00
obj=staff_u:object_r:user_tmpfs_t:s15:c0.c1023
type=PATH msg=audit(1366282236.776:3606): item=1 name=(null) inode=6926
dev=00:0c mode=041777 ouid=0 ogid=0 rdev=00:00
obj=system_u:object_r:tmpfs_t:s15:c0.c1023
type=PATH msg=audit(1366282236.776:3606): item=0 name="test_mq"
Both of these look wrong to me. As Steve Grubb pointed out:
"What we need is 1 PATH record that identifies the MQ. The other PATH
records probably should not be there."
Fix it to record the mq root as a parent, and flag it such that it
should be hidden from view when the names are logged, since the root of
the mq filesystem isn't terribly interesting. With this change, we get
a single PATH record that looks more like this:
type=PATH msg=audit(1368021604.836:484): item=0 name="test_mq" inode=16914
dev=00:0c mode=0100644 ouid=0 ogid=0 rdev=00:00
obj=unconfined_u:object_r:user_tmpfs_t:s0
In order to do this, a new audit_inode_parent_hidden() function is
added. If we do it this way, then we avoid having the existing callers
of audit_inode needing to do any sort of flag conversion if auditing is
inactive.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Reported-by: Jiri Jaburek <jjaburek@redhat.com>
Cc: Steve Grubb <sgrubb@redhat.com>
Cc: Eric Paris <eparis@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
do_smart_update_queue() is called when an operation (semop,
semctl(SETVAL), semctl(SETALL), ...) modified the array. It must check
which of the sleeping tasks can proceed.
do_smart_update_queue() missed a few wakeups:
- if a sleeping complex op was completed, then all per-semaphore queues
must be scanned - not only those that were modified by *sops
- if a sleeping simple op proceeded, then the global queue must be
scanned again
And:
- the test for "|sops == NULL) before scanning the global queue is not
required: If the global queue is empty, then it doesn't need to be
scanned - regardless of the reason for calling do_smart_update_queue()
The patch is not optimized, i.e. even completing a wait-for-zero
operation causes a rescan. This is done to keep the patch as simple as
possible.
Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The semctl GETNCNT returns the number of semops waiting for the
specified semaphore to become nonzero. After commit 9f1bc2c902
("ipc,sem: have only one list in struct sem_queue"), the semops waiting
on just one semaphore are waiting on that semaphore's list.
In order to return the correct count, we have to walk that list too, in
addition to the sem_array's list for complex operations.
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The semctl GETZCNT returns the number of semops waiting for the
specified semaphore to become zero. After commit 9f1bc2c902
("ipc,sem: have only one list in struct sem_queue"), the semops waiting
on just one semaphore are waiting on that semaphore's list.
In order to return the correct count, we have to walk that list too, in
addition to the sem_array's list for complex operations.
This bug broke dbench; it works again with this patch applied.
Signed-off-by: Rik van Riel <riel@redhat.com>
Reported-by: Kent Overstreet <koverstreet@google.com>
Tested-by: Kent Overstreet <koverstreet@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The current kernel returns -EINVAL unless a given mmap length is
"almost" hugepage aligned. This is because in sys_mmap_pgoff() the
given length is passed to vm_mmap_pgoff() as it is without being aligned
with hugepage boundary.
This is a regression introduced in commit 40716e2924 ("hugetlbfs: fix
alignment of huge page requests"), where alignment code is pushed into
hugetlb_file_setup() and the variable len in caller side is not changed.
To fix this, this patch partially reverts that commit, and adds
alignment code in caller side. And it also introduces hstate_sizelog()
in order to get proper hstate to specified hugepage size.
Addresses https://bugzilla.kernel.org/show_bug.cgi?id=56881
[akpm@linux-foundation.org: fix warning when CONFIG_HUGETLB_PAGE=n]
Signed-off-by: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: <iceman_dvd@yahoo.com>
Cc: Steven Truelove <steven.truelove@utoronto.ca>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This trivially combines two rcu_read_lock() calls in both sides of a
if-statement into one single one in front of the if-statement.
Split out as an independent cleanup from the previous commit.
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With various straight RCU lock/unlock movements, one common exit path
pattern had become
rcu_read_unlock();
goto out_wakeup;
and in fact there were no cases where we wanted to exit to out_wakeup
_without_ releasing the RCU read lock.
So replace that pattern with "goto out_rcu_wakeup", and remove the old
out_wakeup.
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
sem_obtain_lock() was another of those functions that returned with the
RCU lock held for reading in the success case. Move the RCU locking to
the caller (semtimedop()), making it more obvious. We already did RCU
locking elsewhere in that function.
Side note: why does semtimedop() re-do the semphore lookup after the
sleep, rather than just getting a reference to the semaphore it already
looked up originally?
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix another ipc locking buglet introduced by the scalability patches:
when semctl_down() was changed to delay the semaphore locking, one error
path for security_sem_semctl() went through the semaphore unlock logic
even though the semaphore had never been locked.
Introduced by commit 16df3674ef ("ipc,sem: do not hold ipc lock more
than necessary")
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is another ipc semaphore locking cleanup, trying to make the
locking more straightforward. We move the rcu read locking into the
callers of sem_lock_and_putref(), which in general means that we now
mostly do the rcu_read_lock() and rcu_read_unlock() in the same
function.
Mostly. We still have the ipc_addid/newary/freeary mess, and things
like ipcctl_pre_down_nolock().
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
ipc_rcu_putref() uses atomics for the refcount, and the games to lock
and unlock the semaphore just to try to keep the reference counting
working are no longer useful.
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The IPC locking is a mess, and sem_unlock() unlocks not only the
semaphore spinlock, it also drops the rcu read lock. Unlike sem_lock(),
which just gets the spin-lock, and expects the caller to get the rcu
read lock.
This all makes things very hard to follow, and it's very confusing when
you take the rcu read lock in one function, and then release it in
another. And it has caused actual bugs: the sem_obtain_lock() function
ended up dropping the RCU read lock twice in one error path, because it
first did the sem_unlock(), and then did a rcu_read_unlock() to match
the rcu_read_lock() it had done.
This is just a totally mindless "remove rcu_read_unlock() from
sem_unlock() and add it immediately after each caller" (except for the
aforementioned bug where we did too many rcu_read_unlock(), and in
find_alloc_undo() where we just got the rcu_read_lock() to correct for
the fact that sem_unlock would immediately drop it again).
We can (and should) clean things up further, but this fixes the bug with
the minimal amount of subtlety.
Reviewed-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We can step on WARN_ON_ONCE() in sem_getref() if a semaphore is removed
just as we are about to call sem_getref() from semctl_main(); results
are not pretty.
We should fail with -EIDRM, same as if IPC_RM happened while we'd been
doing allocation there. This also expands sem_getref() at its only
callsite (and fixed there), while sem_getref_and_unlock() is simply
killed off - it has no callers at all.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 32fcfd4071 ("make vfree() safe to call from interrupt
contexts") made it safe to do vfree directly from the RCU callback,
which allows us to simplify ipc/util.c a lot by getting rid of the
differences between vmalloc/kmalloc memory.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull VFS updates from Al Viro,
Misc cleanups all over the place, mainly wrt /proc interfaces (switch
create_proc_entry to proc_create(), get rid of the deprecated
create_proc_read_entry() in favor of using proc_create_data() and
seq_file etc).
7kloc removed.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (204 commits)
don't bother with deferred freeing of fdtables
proc: Move non-public stuff from linux/proc_fs.h to fs/proc/internal.h
proc: Make the PROC_I() and PDE() macros internal to procfs
proc: Supply a function to remove a proc entry by PDE
take cgroup_open() and cpuset_open() to fs/proc/base.c
ppc: Clean up scanlog
ppc: Clean up rtas_flash driver somewhat
hostap: proc: Use remove_proc_subtree()
drm: proc: Use remove_proc_subtree()
drm: proc: Use minor->index to label things, not PDE->name
drm: Constify drm_proc_list[]
zoran: Don't print proc_dir_entry data in debug
reiserfs: Don't access the proc_dir_entry in r_open(), r_start() r_show()
proc: Supply an accessor for getting the data from a PDE's parent
airo: Use remove_proc_subtree()
rtl8192u: Don't need to save device proc dir PDE
rtl8187se: Use a dir under /proc/net/r8180/
proc: Add proc_mkdir_data()
proc: Move some bits from linux/proc_fs.h to linux/{of.h,signal.h,tty.h}
proc: Move PDE_NET() to fs/proc/proc_net.c
...
Split the proc namespace stuff out into linux/proc_ns.h.
Signed-off-by: David Howells <dhowells@redhat.com>
cc: netdev@vger.kernel.org
cc: Serge E. Hallyn <serge.hallyn@ubuntu.com>
cc: Eric W. Biederman <ebiederm@xmission.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Trying to run an application which was trying to put data into half of
memory using shmget(), we found that having a shmall value below 8EiB-8TiB
would prevent us from using anything more than 8TiB. By setting
kernel.shmall greater than 8EiB-8TiB would make the job work.
In the newseg() function, ns->shm_tot which, at 8TiB is INT_MAX.
ipc/shm.c:
458 static int newseg(struct ipc_namespace *ns, struct ipc_params *params)
459 {
...
465 int numpages = (size + PAGE_SIZE -1) >> PAGE_SHIFT;
...
474 if (ns->shm_tot + numpages > ns->shm_ctlall)
475 return -ENOSPC;
[akpm@linux-foundation.org: make ipc/shm.c:newseg()'s numpages size_t, not int]
Signed-off-by: Robin Holt <holt@sgi.com>
Reported-by: Alex Thorlton <athorlton@sgi.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The ipc/msg.c code does its list operations by hand and it open-codes the
accesses, instead of using for_each_entry_[safe].
Signed-off-by: Nikola Pajkovsky <npajkovs@redhat.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Introduce finer grained locking for semtimedop, to handle the common case
of a program wanting to manipulate one semaphore from an array with
multiple semaphores.
If the call is a semop manipulating just one semaphore in an array with
multiple semaphores, only take the lock for that semaphore itself.
If the call needs to manipulate multiple semaphores, or another caller is
in a transaction that manipulates multiple semaphores, the sem_array lock
is taken, as well as all the locks for the individual semaphores.
On a 24 CPU system, performance numbers with the semop-multi
test with N threads and N semaphores, look like this:
vanilla Davidlohr's Davidlohr's + Davidlohr's +
threads patches rwlock patches v3 patches
10 610652 726325 1783589 2142206
20 341570 365699 1520453 1977878
30 288102 307037 1498167 2037995
40 290714 305955 1612665 2256484
50 288620 312890 1733453 2650292
60 289987 306043 1649360 2388008
70 291298 306347 1723167 2717486
80 290948 305662 1729545 2763582
90 290996 306680 1736021 2757524
100 292243 306700 1773700 3059159
[davidlohr.bueso@hp.com: do not call sem_lock when bogus sma]
[davidlohr.bueso@hp.com: make refcounter atomic]
Signed-off-by: Rik van Riel <riel@redhat.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Chegu Vinod <chegu_vinod@hp.com>
Cc: Jason Low <jason.low2@hp.com>
Reviewed-by: Michel Lespinasse <walken@google.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Tested-by: Emmanuel Benisty <benisty.e@gmail.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Having only one list in struct sem_queue, and only queueing simple
semaphore operations on the list for the semaphore involved, allows us to
introduce finer grained locking for semtimedop.
Signed-off-by: Rik van Riel <riel@redhat.com>
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Chegu Vinod <chegu_vinod@hp.com>
Cc: Emmanuel Benisty <benisty.e@gmail.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Rename sem_lock() to sem_obtain_lock(), so we can introduce a sem_lock()
later that only locks the sem_array and does nothing else.
Open code the locking from ipc_lock() in sem_obtain_lock() so we can
introduce finer grained locking for the sem_array in the next patch.
[akpm@linux-foundation.org: propagate the ipc_obtain_object() errno out of sem_obtain_lock()]
Signed-off-by: Rik van Riel <riel@redhat.com>
Acked-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Cc: Chegu Vinod <chegu_vinod@hp.com>
Cc: Emmanuel Benisty <benisty.e@gmail.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Various forms of ipc use ipcctl_pre_down() to retrieve an ipc object and
check permissions, mostly for IPC_RMID and IPC_SET commands.
Introduce ipcctl_pre_down_nolock(), a lockless version of this function.
The locking version is retained, yet modified to call the nolock version
without affecting its semantics, thus transparent to all ipc callers.
Signed-off-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Chegu Vinod <chegu_vinod@hp.com>
Cc: Emmanuel Benisty <benisty.e@gmail.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Through ipc_lock() and therefore ipc_lock_check() we currently return the
locked ipc object. This is not necessary for all situations and can,
therefore, cause unnecessary ipc lock contention.
Introduce analogous ipc_obtain_object() and ipc_obtain_object_check()
functions that only lookup and return the ipc object.
Both these functions must be called within the RCU read critical section.
[akpm@linux-foundation.org: propagate the ipc_obtain_object() errno from ipc_lock()]
Signed-off-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Reviewed-by: Chegu Vinod <chegu_vinod@hp.com>
Acked-by: Michel Lespinasse <walken@google.com>
Cc: Emmanuel Benisty <benisty.e@gmail.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This series makes the sysv semaphore code more scalable, by reducing the
time the semaphore lock is held, and making the locking more scalable for
semaphore arrays with multiple semaphores.
The first four patches were written by Davidlohr Buesso, and reduce the
hold time of the semaphore lock.
The last three patches change the sysv semaphore code locking to be more
fine grained, providing a performance boost when multiple semaphores in a
semaphore array are being manipulated simultaneously.
On a 24 CPU system, performance numbers with the semop-multi
test with N threads and N semaphores, look like this:
vanilla Davidlohr's Davidlohr's + Davidlohr's +
threads patches rwlock patches v3 patches
10 610652 726325 1783589 2142206
20 341570 365699 1520453 1977878
30 288102 307037 1498167 2037995
40 290714 305955 1612665 2256484
50 288620 312890 1733453 2650292
60 289987 306043 1649360 2388008
70 291298 306347 1723167 2717486
80 290948 305662 1729545 2763582
90 290996 306680 1736021 2757524
100 292243 306700 1773700 3059159
This patch:
There is no reason to be holding the ipc lock while reading ipcp->seq,
hence remove misleading comment.
Also simplify the return value for the function.
Signed-off-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: Chegu Vinod <chegu_vinod@hp.com>
Cc: Emmanuel Benisty <benisty.e@gmail.com>
Cc: Jason Low <jason.low2@hp.com>
Cc: Michel Lespinasse <walken@google.com>
Cc: Peter Hurley <peter@hurleysoftware.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Teach the helper routines about MSG_COPY so that msgtyp is preserved as
the message number to copy.
The security functions affected by this change were audited and no
additional changes are necessary.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Acked-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In preparation for refactoring the queue scan into a separate
function, relocate msg copying.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Acked-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull compat cleanup from Al Viro:
"Mostly about syscall wrappers this time; there will be another pile
with patches in the same general area from various people, but I'd
rather push those after both that and vfs.git pile are in."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/signal:
syscalls.h: slightly reduce the jungles of macros
get rid of union semop in sys_semctl(2) arguments
make do_mremap() static
sparc: no need to sign-extend in sync_file_range() wrapper
ppc compat wrappers for add_key(2) and request_key(2) are pointless
x86: trim sys_ia32.h
x86: sys32_kill and sys32_mprotect are pointless
get rid of compat_sys_semctl() and friends in case of ARCH_WANT_OLD_COMPAT_IPC
merge compat sys_ipc instances
consolidate compat lookup_dcookie()
convert vmsplice to COMPAT_SYSCALL_DEFINE
switch getrusage() to COMPAT_SYSCALL_DEFINE
switch epoll_pwait to COMPAT_SYSCALL_DEFINE
convert sendfile{,64} to COMPAT_SYSCALL_DEFINE
switch signalfd{,4}() to COMPAT_SYSCALL_DEFINE
make SYSCALL_DEFINE<n>-generated wrappers do asmlinkage_protect
make HAVE_SYSCALL_WRAPPERS unconditional
consolidate cond_syscall and SYSCALL_ALIAS declarations
teach SYSCALL_DEFINE<n> how to deal with long long/unsigned long long
get rid of duplicate logics in __SC_....[1-6] definitions
Squishes a statement-with-no-effect warning, removes some ifdefs and
shrinks .text by one byte!
Note that this code fails to check for blocking_notifier_chain_register()
failures.
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The only part of proc_dir_entry the code outside of fs/proc
really cares about is PDE(inode)->data. Provide a helper
for that; static inline for now, eventually will be moved
to fs/proc, along with the knowledge of struct proc_dir_entry
layout.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Make sure that msg pointer is set back to error value in case of
MSG_COPY flag is set and desired message to copy wasn't found. This
garantees that msg is either a error pointer or a copy address.
Otherwise the last message in queue will be freed without unlinking from
the queue (which leads to memory corruption) and the dummy allocated
copy won't be released.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull userns fixes from Eric W Biederman:
"The bulk of the changes are fixing the worst consequences of the user
namespace design oversight in not considering what happens when one
namespace starts off as a clone of another namespace, as happens with
the mount namespace.
The rest of the changes are just plain bug fixes.
Many thanks to Andy Lutomirski for pointing out many of these issues."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
userns: Restrict when proc and sysfs can be mounted
ipc: Restrict mounting the mqueue filesystem
vfs: Carefully propogate mounts across user namespaces
vfs: Add a mount flag to lock read only bind mounts
userns: Don't allow creation if the user is chrooted
yama: Better permission check for ptraceme
pid: Handle the exit of a multi-threaded init.
scm: Require CAP_SYS_ADMIN over the current pidns to spoof pids.
Only allow mounting the mqueue filesystem if the caller has CAP_SYS_ADMIN
rights over the ipc namespace. The principle here is if you create
or have capabilities over it you can mount it, otherwise you get to live
with what other people have mounted.
This information is not particularly sensitive and mqueue essentially
only reports which posix messages queues exist. Still when creating a
restricted environment for an application to live any extra
information may be of use to someone with sufficient creativity. The
historical if imperfect way this information has been restricted has
been not to allow mounts and restricting this to ipc namespace
creators maintains the spirit of the historical restriction.
Cc: stable@vger.kernel.org
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
mnt_drop_write() must be called only if mnt_want_write() succeeded,
otherwise the mnt_writers counter will diverge.
mnt_writers counters are used to check if remounting FS as read-only is
OK, so after an extra mnt_drop_write() call, it would be impossible to
remount mqueue FS as read-only. Besides, on umount a warning would be
printed like this one:
=====================================
[ BUG: bad unlock balance detected! ]
3.9.0-rc3 #5 Not tainted
-------------------------------------
a.out/12486 is trying to release lock (sb_writers) at:
mnt_drop_write+0x1f/0x30
but there are no more locks to release!
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Doug Ledford <dledford@redhat.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When MSG_COPY is set, a duplicate message must be allocated for the copy
before locking the queue. However, the copy could not be larger than was
sent which is limited to msg_ctlmax.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Acked-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If the src msg is > 4k, then dest->next points to the
next allocated segment; resetting it just prior to dereferencing
is bad.
Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
Acked-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
just have the bugger take unsigned long and deal with SETVAL
case (when we use an int member in the union) explicitly.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Convert to the much saner new idr interface.
The new interface doesn't directly translate to the way idr_pre_get()
was used around ipc_addid() as preloading disables preemption. From
my cursory reading, it seems like we should be able to do all
allocation from ipc_addid(), so I moved it there. Can you please
check whether this would be okay? If this is wrong and ipc_addid()
should be allowed to be called from non-sleepable context, I'd suggest
allocating id itself in the outer functions and later install the
pointer using idr_replace().
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Sedat Dilek <sedat.dilek@gmail.com>
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Cc: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs pile (part one) from Al Viro:
"Assorted stuff - cleaning namei.c up a bit, fixing ->d_name/->d_parent
locking violations, etc.
The most visible changes here are death of FS_REVAL_DOT (replaced with
"has ->d_weak_revalidate()") and a new helper getting from struct file
to inode. Some bits of preparation to xattr method interface changes.
Misc patches by various people sent this cycle *and* ocfs2 fixes from
several cycles ago that should've been upstream right then.
PS: the next vfs pile will be xattr stuff."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (46 commits)
saner proc_get_inode() calling conventions
proc: avoid extra pde_put() in proc_fill_super()
fs: change return values from -EACCES to -EPERM
fs/exec.c: make bprm_mm_init() static
ocfs2/dlm: use GFP_ATOMIC inside a spin_lock
ocfs2: fix possible use-after-free with AIO
ocfs2: Fix oops in ocfs2_fast_symlink_readpage() code path
get_empty_filp()/alloc_file() leave both ->f_pos and ->f_version zero
target: writev() on single-element vector is pointless
export kernel_write(), convert open-coded instances
fs: encode_fh: return FILEID_INVALID if invalid fid_type
kill f_vfsmnt
vfs: kill FS_REVAL_DOT by adding a d_weak_revalidate dentry op
nfsd: handle vfs_getattr errors in acl protocol
switch vfs_getattr() to struct path
default SET_PERSONALITY() in linux/elf.h
ceph: prepopulate inodes only when request is aborted
d_hash_and_lookup(): export, switch open-coded instances
9p: switch v9fs_set_create_acl() to inode+fid, do it before d_instantiate()
9p: split dropping the acls from v9fs_set_create_acl()
...
Pull user namespace and namespace infrastructure changes from Eric W Biederman:
"This set of changes starts with a few small enhnacements to the user
namespace. reboot support, allowing more arbitrary mappings, and
support for mounting devpts, ramfs, tmpfs, and mqueuefs as just the
user namespace root.
I do my best to document that if you care about limiting your
unprivileged users that when you have the user namespace support
enabled you will need to enable memory control groups.
There is a minor bug fix to prevent overflowing the stack if someone
creates way too many user namespaces.
The bulk of the changes are a continuation of the kuid/kgid push down
work through the filesystems. These changes make using uids and gids
typesafe which ensures that these filesystems are safe to use when
multiple user namespaces are in use. The filesystems converted for
3.9 are ceph, 9p, afs, ocfs2, gfs2, ncpfs, nfs, nfsd, and cifs. The
changes for these filesystems were a little more involved so I split
the changes into smaller hopefully obviously correct changes.
XFS is the only filesystem that remains. I was hoping I could get
that in this release so that user namespace support would be enabled
with an allyesconfig or an allmodconfig but it looks like the xfs
changes need another couple of days before it they are ready."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (93 commits)
cifs: Enable building with user namespaces enabled.
cifs: Convert struct cifs_ses to use a kuid_t and a kgid_t
cifs: Convert struct cifs_sb_info to use kuids and kgids
cifs: Modify struct smb_vol to use kuids and kgids
cifs: Convert struct cifsFileInfo to use a kuid
cifs: Convert struct cifs_fattr to use kuid and kgids
cifs: Convert struct tcon_link to use a kuid.
cifs: Modify struct cifs_unix_set_info_args to hold a kuid_t and a kgid_t
cifs: Convert from a kuid before printing current_fsuid
cifs: Use kuids and kgids SID to uid/gid mapping
cifs: Pass GLOBAL_ROOT_UID and GLOBAL_ROOT_GID to keyring_alloc
cifs: Use BUILD_BUG_ON to validate uids and gids are the same size
cifs: Override unmappable incoming uids and gids
nfsd: Enable building with user namespaces enabled.
nfsd: Properly compare and initialize kuids and kgids
nfsd: Store ex_anon_uid and ex_anon_gid as kuids and kgids
nfsd: Modify nfsd4_cb_sec to use kuids and kgids
nfsd: Handle kuids and kgids in the nfs4acl to posix_acl conversion
nfsd: Convert nfsxdr to use kuids and kgids
nfsd: Convert nfs3xdr to use kuids and kgids
...
do_mmap_pgoff() rounds up the desired size to the next PAGE_SIZE
multiple, however there was no equivalent code in mm_populate(), which
caused issues.
This could be fixed by introduced the same rounding in mm_populate(),
however I think it's preferable to make do_mmap_pgoff() return populate
as a size rather than as a boolean, so we don't have to duplicate the
size rounding logic in mm_populate().
Signed-off-by: Michel Lespinasse <walken@google.com>
Acked-by: Rik van Riel <riel@redhat.com>
Tested-by: Andy Lutomirski <luto@amacapital.net>
Cc: Greg Ungerer <gregungerer@westnet.com.au>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When creating new mappings using the MAP_POPULATE / MAP_LOCKED flags (or
with MCL_FUTURE in effect), we want to populate the pages within the
newly created vmas. This may take a while as we may have to read pages
from disk, so ideally we want to do this outside of the write-locked
mmap_sem region.
This change introduces mm_populate(), which is used to defer populating
such mappings until after the mmap_sem write lock has been released.
This is implemented as a generalization of the former do_mlock_pages(),
which accomplished the same task but was using during mlock() /
mlockall().
Signed-off-by: Michel Lespinasse <walken@google.com>
Reported-by: Andy Lutomirski <luto@amacapital.net>
Acked-by: Rik van Riel <riel@redhat.com>
Tested-by: Andy Lutomirski <luto@amacapital.net>
Cc: Greg Ungerer <gregungerer@westnet.com.au>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Allocating a file structure in function get_empty_filp() might fail because
of several reasons:
- not enough memory for file structures
- operation is not allowed
- user is over its limit
Currently the function returns NULL in all cases and we loose the exact
reason of the error. All callers of get_empty_filp() assume that the function
can fail with ENFILE only.
Return error through pointer. Change all callers to preserve this error code.
[AV: cleaned up a bit, carved the get_empty_filp() part out into a separate commit
(things remaining here deal with alloc_file()), removed pipe(2) behaviour change]
Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
This patch allow the unprivileged user to mount mqueuefs in
user ns.
If two userns share the same ipcns,the files in mqueue fs
should be seen in both these two userns.
If the userns has its own ipcns,it has its own mqueue fs too.
ipcns has already done this job well.
Signed-off-by: Gao feng <gaofeng@cn.fujitsu.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Remove the redundant and confusing fill_copy(). Also add copy_msg()
check for error. In this case exit from the function have to be done
instead of break, because further code interprets any error as EAGAIN.
Also define copy_msg() for the case when CONFIG_CHECKPOINT_RESTORE is
disabled.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This code works if CONFIG_CHECKPOINT_RESTORE is disabled.
[akpm@linux-foundation.org: remove __maybe_unused]
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Passing and checking of msgflg to free_copy() is redundant. This patch
sets copy to NULL on declaration instead and checks for non-NULL in
free_copy().
Note: in case of copy allocation failure, error is returned immediately.
So no need to check for IS_ERR() in free_copy().
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This test can be used to check wheither kernel supports IPC message queue
copy and restore features (required by CRIU project).
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch is required for checkpoint/restore in userspace.
c/r requires some way to get all pending IPC messages without deleting
them from the queue (checkpoint can fail and in this case tasks will be
resumed, so queue have to be valid).
To achive this, new operation flag MSG_COPY for sys_msgrcv() system call
was introduced. If this flag was specified, then mtype is interpreted as
number of the message to copy.
If MSG_COPY is set, then kernel will allocate dummy message with passed
size, and then use new copy_msg() helper function to copy desired message
(instead of unlinking it from the queue).
Notes:
1) Return -ENOSYS if MSG_COPY is specified, but
CONFIG_CHECKPOINT_RESTORE is not set.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move all message related manipulation into one function msg_fill().
Actually, two functions because of the compat one.
[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add 3 new variables and sysctls to tune them (by one "next_id" variable
for messages, semaphores and shared memory respectively). This variable
can be used to set desired id for next allocated IPC object. By default
it's equal to -1 and old behaviour is preserved. If this variable is
non-negative, then desired idr will be extracted from it and used as a
start value to search for free IDR slot.
Notes:
1) this patch doesn't guarantee that the new object will have desired
id. So it's up to user space how to handle new object with wrong id.
2) After a sucessful id allocation attempt, "next_id" will be set back
to -1 (if it was non-negative).
[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is a cleanup patch. The assignment is redundant.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Cc: Serge Hallyn <serge.hallyn@canonical.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull (again) user namespace infrastructure changes from Eric Biederman:
"Those bugs, those darn embarrasing bugs just want don't want to get
fixed.
Linus I just updated my mirror of your kernel.org tree and it appears
you successfully pulled everything except the last 4 commits that fix
those embarrasing bugs.
When you get a chance can you please repull my branch"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
userns: Fix typo in description of the limitation of userns_install
userns: Add a more complete capability subset test to commit_creds
userns: Require CAP_SYS_ADMIN for most uses of setns.
Fix cap_capable to only allow owners in the parent user namespace to have caps.
Pull user namespace changes from Eric Biederman:
"While small this set of changes is very significant with respect to
containers in general and user namespaces in particular. The user
space interface is now complete.
This set of changes adds support for unprivileged users to create user
namespaces and as a user namespace root to create other namespaces.
The tyranny of supporting suid root preventing unprivileged users from
using cool new kernel features is broken.
This set of changes completes the work on setns, adding support for
the pid, user, mount namespaces.
This set of changes includes a bunch of basic pid namespace
cleanups/simplifications. Of particular significance is the rework of
the pid namespace cleanup so it no longer requires sending out
tendrils into all kinds of unexpected cleanup paths for operation. At
least one case of broken error handling is fixed by this cleanup.
The files under /proc/<pid>/ns/ have been converted from regular files
to magic symlinks which prevents incorrect caching by the VFS,
ensuring the files always refer to the namespace the process is
currently using and ensuring that the ptrace_mayaccess permission
checks are always applied.
The files under /proc/<pid>/ns/ have been given stable inode numbers
so it is now possible to see if different processes share the same
namespaces.
Through the David Miller's net tree are changes to relax many of the
permission checks in the networking stack to allowing the user
namespace root to usefully use the networking stack. Similar changes
for the mount namespace and the pid namespace are coming through my
tree.
Two small changes to add user namespace support were commited here adn
in David Miller's -net tree so that I could complete the work on the
/proc/<pid>/ns/ files in this tree.
Work remains to make it safe to build user namespaces and 9p, afs,
ceph, cifs, coda, gfs2, ncpfs, nfs, nfsd, ocfs2, and xfs so the
Kconfig guard remains in place preventing that user namespaces from
being built when any of those filesystems are enabled.
Future design work remains to allow root users outside of the initial
user namespace to mount more than just /proc and /sys."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (38 commits)
proc: Usable inode numbers for the namespace file descriptors.
proc: Fix the namespace inode permission checks.
proc: Generalize proc inode allocation
userns: Allow unprivilged mounts of proc and sysfs
userns: For /proc/self/{uid,gid}_map derive the lower userns from the struct file
procfs: Print task uids and gids in the userns that opened the proc file
userns: Implement unshare of the user namespace
userns: Implent proc namespace operations
userns: Kill task_user_ns
userns: Make create_new_namespaces take a user_ns parameter
userns: Allow unprivileged use of setns.
userns: Allow unprivileged users to create new namespaces
userns: Allow setting a userns mapping to your current uid.
userns: Allow chown and setgid preservation
userns: Allow unprivileged users to create user namespaces.
userns: Ignore suid and sgid on binaries if the uid or gid can not be mapped
userns: fix return value on mntns_install() failure
vfs: Allow unprivileged manipulation of the mount namespace.
vfs: Only support slave subtrees across different user namespaces
vfs: Add a user namespace reference from struct mnt_namespace
...
Andy Lutomirski <luto@amacapital.net> found a nasty little bug in
the permissions of setns. With unprivileged user namespaces it
became possible to create new namespaces without privilege.
However the setns calls were relaxed to only require CAP_SYS_ADMIN in
the user nameapce of the targed namespace.
Which made the following nasty sequence possible.
pid = clone(CLONE_NEWUSER | CLONE_NEWNS);
if (pid == 0) { /* child */
system("mount --bind /home/me/passwd /etc/passwd");
}
else if (pid != 0) { /* parent */
char path[PATH_MAX];
snprintf(path, sizeof(path), "/proc/%u/ns/mnt");
fd = open(path, O_RDONLY);
setns(fd, 0);
system("su -");
}
Prevent this possibility by requiring CAP_SYS_ADMIN
in the current user namespace when joing all but the user namespace.
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
There was some desire in large applications using MAP_HUGETLB or
SHM_HUGETLB to use 1GB huge pages on some mappings, and stay with 2MB on
others. This is useful together with NUMA policy: use 2MB interleaving
on some mappings, but 1GB on local mappings.
This patch extends the IPC/SHM syscall interfaces slightly to allow
specifying the page size.
It borrows some upper bits in the existing flag arguments and allows
encoding the log of the desired page size in addition to the *_HUGETLB
flag. When 0 is specified the default size is used, this makes the
change fully compatible.
Extending the internal hugetlb code to handle this is straight forward.
Instead of a single mount it just keeps an array of them and selects the
right mount based on the specified page size. When no page size is
specified it uses the mount of the default page size.
The change is not visible in /proc/mounts because internal mounts don't
appear there. It also has very little overhead: the additional mounts
just consume a super block, but not more memory when not used.
I also exported the new flags to the user headers (they were previously
under __KERNEL__). Right now only symbols for x86 and some other
architecture for 1GB and 2MB are defined. The interface should already
work for all other architectures though. Only architectures that define
multiple hugetlb sizes actually need it (that is currently x86, tile,
powerpc). However tile and powerpc have user configurable hugetlb
sizes, so it's not easy to add defines. A program on those
architectures would need to query sysfs and use the appropiate log2.
[akpm@linux-foundation.org: cleanups]
[rientjes@google.com: fix build]
[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Hillf Danton <dhillf@gmail.com>
Signed-off-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Assign a unique proc inode to each namespace, and use that
inode number to ensure we only allocate at most one proc
inode for every namespace in proc.
A single proc inode per namespace allows userspace to test
to see if two processes are in the same namespace.
This has been a long requested feature and only blocked because
a naive implementation would put the id in a global space and
would ultimately require having a namespace for the names of
namespaces, making migration and certain virtualization tricks
impossible.
We still don't have per superblock inode numbers for proc, which
appears necessary for application unaware checkpoint/restart and
migrations (if the application is using namespace file descriptors)
but that is now allowd by the design if it becomes important.
I have preallocated the ipc and uts initial proc inode numbers so
their structures can be statically initialized.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Modify create_new_namespaces to explicitly take a user namespace
parameter, instead of implicitly through the task_struct.
This allows an implementation of unshare(CLONE_NEWUSER) where
the new user namespace is not stored onto the current task_struct
until after all of the namespaces are created.
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
- Push the permission check from the core setns syscall into
the setns install methods where the user namespace of the
target namespace can be determined, and used in a ns_capable
call.
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Keep a pointer to the audit_names "slot" in struct filename.
Have all of the audit_inode callers pass a struct filename ponter to
audit_inode instead of a string pointer. If the aname field is already
populated, then we can skip walking the list altogether and just use it
directly.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
getname() is intended to copy pathname strings from userspace into a
kernel buffer. The result is just a string in kernel space. It would
however be quite helpful to be able to attach some ancillary info to
the string.
For instance, we could attach some audit-related info to reduce the
amount of audit-related processing needed. When auditing is enabled,
we could also call getname() on the string more than once and not
need to recopy it from userspace.
This patchset converts the getname()/putname() interfaces to return
a struct instead of a string. For now, the struct just tracks the
string in kernel space and the original userland pointer for it.
Later, we'll add other information to the struct as it becomes
convenient.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Currently, this gets set mostly by happenstance when we call into
audit_inode_child. While that might be a little more efficient, it seems
wrong. If the syscall ends up failing before audit_inode_child ever gets
called, then you'll have an audit_names record that shows the full path
but has the parent inode info attached.
Fix this by passing in a parent flag when we call audit_inode that gets
set to the value of LOOKUP_PARENT. We can then fix up the pathname for
the audit entry correctly from the get-go.
While we're at it, clean up the no-op macro for audit_inode in the
!CONFIG_AUDITSYSCALL case.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Commit d6629859b3 ("ipc/mqueue: improve performance of send/recv") and
ce2d52cc ("ipc/mqueue: add rbtree node caching support") introduced an
rbtree of message priorities, and usage of rb_init_node() to initialize
the corresponding nodes. As it turns out, rb_init_node() is unnecessary
here, as the nodes are fully initialized on insertion by rb_link_node()
and the code doesn't access nodes that aren't inserted on the rbtree.
Removing the rb_init_node() calls as I removed that function during
rbtree API cleanups (the only other use of it was in a place that
similarly didn't require it).
Signed-off-by: Michel Lespinasse <walken@google.com>
Acked-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs update from Al Viro:
- big one - consolidation of descriptor-related logics; almost all of
that is moved to fs/file.c
(BTW, I'm seriously tempted to rename the result to fd.c. As it is,
we have a situation when file_table.c is about handling of struct
file and file.c is about handling of descriptor tables; the reasons
are historical - file_table.c used to be about a static array of
struct file we used to have way back).
A lot of stray ends got cleaned up and converted to saner primitives,
disgusting mess in android/binder.c is still disgusting, but at least
doesn't poke so much in descriptor table guts anymore. A bunch of
relatively minor races got fixed in process, plus an ext4 struct file
leak.
- related thing - fget_light() partially unuglified; see fdget() in
there (and yes, it generates the code as good as we used to have).
- also related - bits of Cyrill's procfs stuff that got entangled into
that work; _not_ all of it, just the initial move to fs/proc/fd.c and
switch of fdinfo to seq_file.
- Alex's fs/coredump.c spiltoff - the same story, had been easier to
take that commit than mess with conflicts. The rest is a separate
pile, this was just a mechanical code movement.
- a few misc patches all over the place. Not all for this cycle,
there'll be more (and quite a few currently sit in akpm's tree)."
Fix up trivial conflicts in the android binder driver, and some fairly
simple conflicts due to two different changes to the sock_alloc_file()
interface ("take descriptor handling from sock_alloc_file() to callers"
vs "net: Providing protocol type via system.sockprotoname xattr of
/proc/PID/fd entries" adding a dentry name to the socket)
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (72 commits)
MAX_LFS_FILESIZE should be a loff_t
compat: fs: Generic compat_sys_sendfile implementation
fs: push rcu_barrier() from deactivate_locked_super() to filesystems
btrfs: reada_extent doesn't need kref for refcount
coredump: move core dump functionality into its own file
coredump: prevent double-free on an error path in core dumper
usb/gadget: fix misannotations
fcntl: fix misannotations
ceph: don't abuse d_delete() on failure exits
hypfs: ->d_parent is never NULL or negative
vfs: delete surplus inode NULL check
switch simple cases of fget_light to fdget
new helpers: fdget()/fdput()
switch o2hb_region_dev_write() to fget_light()
proc_map_files_readdir(): don't bother with grabbing files
make get_file() return its argument
vhost_set_vring(): turn pollstart/pollstop into bool
switch prctl_set_mm_exe_file() to fget_light()
switch xfs_find_handle() to fget_light()
switch xfs_swapext() to fget_light()
...
Pull user namespace changes from Eric Biederman:
"This is a mostly modest set of changes to enable basic user namespace
support. This allows the code to code to compile with user namespaces
enabled and removes the assumption there is only the initial user
namespace. Everything is converted except for the most complex of the
filesystems: autofs4, 9p, afs, ceph, cifs, coda, fuse, gfs2, ncpfs,
nfs, ocfs2 and xfs as those patches need a bit more review.
The strategy is to push kuid_t and kgid_t values are far down into
subsystems and filesystems as reasonable. Leaving the make_kuid and
from_kuid operations to happen at the edge of userspace, as the values
come off the disk, and as the values come in from the network.
Letting compile type incompatible compile errors (present when user
namespaces are enabled) guide me to find the issues.
The most tricky areas have been the places where we had an implicit
union of uid and gid values and were storing them in an unsigned int.
Those places were converted into explicit unions. I made certain to
handle those places with simple trivial patches.
Out of that work I discovered we have generic interfaces for storing
quota by projid. I had never heard of the project identifiers before.
Adding full user namespace support for project identifiers accounts
for most of the code size growth in my git tree.
Ultimately there will be work to relax privlige checks from
"capable(FOO)" to "ns_capable(user_ns, FOO)" where it is safe allowing
root in a user names to do those things that today we only forbid to
non-root users because it will confuse suid root applications.
While I was pushing kuid_t and kgid_t changes deep into the audit code
I made a few other cleanups. I capitalized on the fact we process
netlink messages in the context of the message sender. I removed
usage of NETLINK_CRED, and started directly using current->tty.
Some of these patches have also made it into maintainer trees, with no
problems from identical code from different trees showing up in
linux-next.
After reading through all of this code I feel like I might be able to
win a game of kernel trivial pursuit."
Fix up some fairly trivial conflicts in netfilter uid/git logging code.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace: (107 commits)
userns: Convert the ufs filesystem to use kuid/kgid where appropriate
userns: Convert the udf filesystem to use kuid/kgid where appropriate
userns: Convert ubifs to use kuid/kgid
userns: Convert squashfs to use kuid/kgid where appropriate
userns: Convert reiserfs to use kuid and kgid where appropriate
userns: Convert jfs to use kuid/kgid where appropriate
userns: Convert jffs2 to use kuid and kgid where appropriate
userns: Convert hpfs to use kuid and kgid where appropriate
userns: Convert btrfs to use kuid/kgid where appropriate
userns: Convert bfs to use kuid/kgid where appropriate
userns: Convert affs to use kuid/kgid wherwe appropriate
userns: On alpha modify linux_to_osf_stat to use convert from kuids and kgids
userns: On ia64 deal with current_uid and current_gid being kuid and kgid
userns: On ppc convert current_uid from a kuid before printing.
userns: Convert s390 getting uid and gid system calls to use kuid and kgid
userns: Convert s390 hypfs to use kuid and kgid where appropriate
userns: Convert binder ipc to use kuids
userns: Teach security_path_chown to take kuids and kgids
userns: Add user namespace support to IMA
userns: Convert EVM to deal with kuids and kgids in it's hmac computation
...
- Store the ipc owner and creator with a kuid
- Store the ipc group and the crators group with a kgid.
- Add error handling to ipc_update_perms, allowing it to
fail if the uids and gids can not be converted to kuids
or kgids.
- Modify the proc files to display the ipc creator and
owner in the user namespace of the opener of the proc file.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
Rather than #define the options manually in the architecture code, add
Kconfig options for them and select them there instead. This also allows
us to select the compat IPC version parsing automatically for platforms
using the old compat IPC interface.
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The msgsnd and msgrcv system calls use size_t to represent the size of the
message being transferred. POSIX states that values of msgsz greater than
SSIZE_MAX cause the result to be implementation-defined. On Linux, this
equates to returning -EINVAL if (long) msgsz < 0.
For compat tasks where !CONFIG_ARCH_WANT_OLD_COMPAT_IPC and compat_size_t
is smaller than size_t, negative size values passed from userspace will be
interpreted as positive values by do_msg{rcv,snd} and will fail to exit
early with -EINVAL.
This patch changes the compat prototypes for msg{rcv,snd} so that the
message size is represented as a compat_ssize_t, which we cast to the
native ssize_t type for the core IPC code.
Cc: Arnd Bergmann <arnd@arndb.de>
Acked-by: Chris Metcalf <cmetcalf@tilera.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 48b25c43e6 ("ipc: provide generic compat versions of IPC
syscalls") added a new ARCH_WANT_OLD_COMPAT_IPC config option for
architectures to select if their compat target requires the old IPC
syscall interface.
For architectures (such as AArch64) that do not require the internal
calling conventions provided by this option, but have a compat target
where the C library passes the IPC_64 flag explicitly,
compat_ipc_parse_version no longer strips out the flag before calling
the native system call implementation, resulting in unknown SHM/IPC
commands and -EINVAL being returned to userspace.
This patch separates the selection of the internal calling conventions
for the IPC syscalls from the version parsing, allowing architectures to
select __ARCH_WANT_COMPAT_IPC_PARSE_VERSION if they want to use version
parsing whilst retaining the newer syscall calling conventions.
Acked-by: Chris Metcalf <cmetcalf@tilera.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If the SHMLBA definition for a native task differs from the definition for
a compat task, the do_shmat() function would need to handle both.
This patch introduces COMPAT_SHMLBA, which is used by the compat shmat
syscall when calling the ipc code and allows architectures such as AArch64
(where the native SHMLBA is 64k but the compat (AArch32) definition is
16k) to provide the correct semantics for compat IPC system calls.
Cc: David S. Miller <davem@davemloft.net>
Cc: Chris Zankel <chris@zankel.net>
Cc: Arnd Bergmann <arnd@arndb.de>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
boolean "does it have to be exclusive?" flag is passed instead;
Local filesystem should just ignore it - the object is guaranteed
not to be there yet.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Commit 17cf28afea ("mm/fs: remove truncate_range") removed the
truncate_range inode operation in favour of the fallocate file
operation.
When using SYSV IPC shared memory segments, calling madvise with the
MADV_REMOVE advice on an area of shared memory will attempt to invoke
the .fallocate function for the shm_file_operations, which is NULL and
therefore returns -EOPNOTSUPP to userspace. The previous behaviour
would inherit the inode_operations from the underlying tmpfs file and
invoke truncate_range there.
This patch restores the previous behaviour by wrapping the underlying
fallocate function in shm_fallocate, as we do for fsync.
[hughd@google.com: use -ENOTSUPP in shm_fallocate()]
Signed-off-by: Will Deacon <will.deacon@arm.com>
Acked-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs changes from Al Viro.
"A lot of misc stuff. The obvious groups:
* Miklos' atomic_open series; kills the damn abuse of
->d_revalidate() by NFS, which was the major stumbling block for
all work in that area.
* ripping security_file_mmap() and dealing with deadlocks in the
area; sanitizing the neighborhood of vm_mmap()/vm_munmap() in
general.
* ->encode_fh() switched to saner API; insane fake dentry in
mm/cleancache.c gone.
* assorted annotations in fs (endianness, __user)
* parts of Artem's ->s_dirty work (jff2 and reiserfs parts)
* ->update_time() work from Josef.
* other bits and pieces all over the place.
Normally it would've been in two or three pull requests, but
signal.git stuff had eaten a lot of time during this cycle ;-/"
Fix up trivial conflicts in Documentation/filesystems/vfs.txt (the
'truncate_range' inode method was removed by the VM changes, the VFS
update adds an 'update_time()' method), and in fs/btrfs/ulist.[ch] (due
to sparse fix added twice, with other changes nearby).
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (95 commits)
nfs: don't open in ->d_revalidate
vfs: retry last component if opening stale dentry
vfs: nameidata_to_filp(): don't throw away file on error
vfs: nameidata_to_filp(): inline __dentry_open()
vfs: do_dentry_open(): don't put filp
vfs: split __dentry_open()
vfs: do_last() common post lookup
vfs: do_last(): add audit_inode before open
vfs: do_last(): only return EISDIR for O_CREAT
vfs: do_last(): check LOOKUP_DIRECTORY
vfs: do_last(): make ENOENT exit RCU safe
vfs: make follow_link check RCU safe
vfs: do_last(): use inode variable
vfs: do_last(): inline walk_component()
vfs: do_last(): make exit RCU safe
vfs: split do_lookup()
Btrfs: move over to use ->update_time
fs: introduce inode operation ->update_time
reiserfs: get rid of resierfs_sync_super
reiserfs: mark the superblock as dirty a bit later
...
When I wrote the first patch that added the rbtree support for message
queue insertion, it sped up the case where the queue was very full
drastically from the original code. It, however, slowed down the case
where the queue was empty (not drastically though).
This patch caches the last freed rbtree node struct so we can quickly
reuse it when we get a new message. This is the common path for any queue
that very frequently goes from 0 to 1 then back to 0 messages in queue.
Andrew Morton didn't like that we were doing a GFP_ATOMIC allocation in
msg_insert, so this patch attempts to speculatively allocate a new node
struct outside of the spin lock when we know we need it, but will still
fall back to a GFP_ATOMIC allocation if it has to.
Once I added the caching, the necessary various ret = ; spin_unlock
gyrations in mq_timedsend were getting pretty ugly, so this also slightly
refactors that function to streamline the flow of the code and the
function exit.
Finally, while working on getting performance back I made sure that all of
the node structs were always fully initialized when they were first used,
rendering the use of kzalloc unnecessary and a waste of CPU cycles.
The net result of all of this is:
1) We will avoid a GFP_ATOMIC allocation when possible, but fall back
on it when necessary.
2) We will speculatively allocate a node struct using GFP_KERNEL if our
cache is empty (and save the struct to our cache if it's still empty
after we have obtained the spin lock).
3) The performance of the common queue empty case has significantly
improved and is now much more in line with the older performance for
this case.
The performance changes are:
Old mqueue new mqueue new mqueue + caching
queue empty
send/recv 305/288ns 349/318ns 310/322ns
I don't think we'll ever be able to get the recv performance back, but
that's because the old recv performance was a direct result and
consequence of the old methods abysmal send performance. The recv path
simply must do more so that the send path does not incur such a penalty
under higher queue depths.
As it turns out, the new caching code also sped up the various queue full
cases relative to my last patch. That could be because of the difference
between the syscall path in 3.3.4-rc5 and 3.3.4-rc6, or because of the
change in code flow in the mq_timedsend routine. Regardless, I'll take
it. It wasn't huge, and I *would* say it was within the margin for error,
but after many repeated runs what I'm seeing is that the old numbers trend
slightly higher (about 10 to 20ns depending on which test is the one
running).
[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Doug Ledford <dledford@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We already check the mq attr struct if it's passed in, but now that the
admin can set system wide defaults separate from maximums, it's actually
possible to set the defaults to something that would overflow. So, if
there is no attr struct passed in to the open call, check the default
values.
While we are at it, simplify mq_attr_ok() by making it return 0 or an
error condition, so that way if we add more tests to it later, we have the
option of what error should be returned instead of the calling location
having to pick a possibly inaccurate error code.
[akpm@linux-foundation.org: s/ENOMEM/EOVERFLOW/]
Signed-off-by: Doug Ledford <dledford@redhat.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Manfred Spraul <manfred@colorfullife.com>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
While working on the other parts of the mqueue stuff, I noticed that the
calculation for overflow in mq_attr_ok didn't actually match reality (this
is especially true since my last patch which changed how we account memory
slightly).
In particular, we used to test for overflow using:
msgs * msgsize + msgs * sizeof(struct msg_msg *)
That was never really correct because each message we allocate via
load_msg() is actually a struct msg_msg followed by the data for the
message (and if struct msg_msg + data exceeds PAGE_SIZE we end up
allocating struct msg_msgseg structs too, but accounting for them would
get really tedious, so let's ignore those...they're only a pointer in size
anyway). This patch updates the calculation to be more accurate in
regards to maximum possible memory consumption by the mqueue.
[akpm@linux-foundation.org: add a local to simplify overflow-checking expression]
Signed-off-by: Doug Ledford <dledford@redhat.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Manfred Spraul <manfred@colorfullife.com>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The existing implementation of the POSIX message queue send and recv
functions is, well, abysmal. Even worse than abysmal. I submitted a
patch to increase the maximum POSIX message queue limit to 65536 due to
customer needs, however, upon looking over the send/recv implementation, I
realized that my customer needs help with that too even if they don't know
it. The basic problem is that, given the fairly typical use case scenario
for a large queue of queueing lots of messages all at the same priority (I
verified with my customer that this is indeed what their app does), the
msg_insert routine is basically a frikkin' bubble sort. I mean, whoa,
that's *so* middle school.
OK, OK, to not slam the original author too much, I'm sure they didn't
envision a queue depth of 50,000+ messages. No one would think that
moving elements in an array, one at a time, and dereferencing each pointer
in that array to check priority of the message being pointed too, again
one at a time, for 50,000+ times would be good. So let's assume that, as
is typical, the users have found a way to break our code simply by using
it in a way we didn't envision. Fair enough.
"So, just how broken is it?", you ask. I wondered the same thing, so I
wrote an app to let me know. It's my next patch. It gave me some
interesting results. Here's what it tested:
Interference with other apps - In continuous mode, the app just sits there
and hits a message queue forever, while you go do something productive on
another terminal using other CPUs. You then measure how long it takes you
to do that something productive. Then you restart the app in fake
continuous mode, and it sits in a tight loop on a CPU while you repeat
your tests. The whole point of this is to keep one CPU tied up (so it
can't be used in your other work) but in one case tied up hitting the
mqueue code so we can see the effect of walking that 65,528 element array
one pointer at a time on the global CPU cache. If it's bad, then it will
slow down your app on the other CPUs just by polluting cache mercilessly.
In the fake case, it will be in a tight loop, but not polluting cache.
Testing the mqueue subsystem directly - Here we just run a number of tests
to see how the mqueue subsystem performs under different conditions. A
couple conditions are known to be worst case for the old system, and some
routines, so this tests all of them.
So, on to the results already:
Subsystem/Test Old New
Time to compile linux
kernel (make -j12 on a
6 core CPU)
Running mqueue test user 49m10.744s user 45m26.294s
sys 5m51.924s sys 4m59.894s
total 55m02.668s total 50m26.188s
Running fake test user 45m32.686s user 45m18.552s
sys 5m12.465s sys 4m56.468s
total 50m45.151s total 50m15.020s
% slowdown from mqueue
cache thrashing ~8% ~.5%
Avg time to send/recv (in nanoseconds per message)
when queue empty 305/288 349/318
when queue full (65528 messages)
constant priority 526589/823 362/314
increasing priority 403105/916 495/445
decreasing priority 73420/594 482/409
random priority 280147/920 546/436
Time to fill/drain queue (65528 messages, in seconds)
constant priority 17.37/.12 .13/.12
increasing priority 4.14/.14 .21/.18
decreasing priority 12.93/.13 .21/.18
random priority 8.88/.16 .22/.17
So, I think the results speak for themselves. It's possible this
implementation could be improved by cacheing at least one priority level
in the node tree (that would bring the queue empty performance more in
line with the old implementation), but this works and is *so* much better
than what we had, especially for the common case of a single priority in
use, that further refinements can be in follow on patches.
[akpm@linux-foundation.org: fix typo in comment, remove stray semicolon]
[levinsasha928@gmail.com: use correct gfp flags in msg_insert]
Signed-off-by: Doug Ledford <dledford@redhat.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Manfred Spraul <manfred@colorfullife.com>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
