Currently we switch to the stable sched_clock if we guess the TSC is
usable, and then switch back to the unstable path if it turns out TSC
isn't stable during SMP bringup after all.
Delay switching to the stable path until after SMP bringup is
complete. This way we'll avoid switching during the time we detect the
worst of the TSC offences.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
There's no diagnostic checks for figuring out when we've accidentally
missed update_rq_clock() calls. Let's add some by piggybacking on the
rq_*pin_lock() wrappers.
The idea behind the diagnostic checks is that upon pining rq lock the
rq clock should be updated, via update_rq_clock(), before anybody
reads the clock with rq_clock() or rq_clock_task().
The exception to this rule is when updates have explicitly been
disabled with the rq_clock_skip_update() optimisation.
There are some functions that only unpin the rq lock in order to grab
some other lock and avoid deadlock. In that case we don't need to
update the clock again and the previous diagnostic state can be
carried over in rq_repin_lock() by saving the state in the rq_flags
context.
Since this patch adds a new clock update flag and some already exist
in rq::clock_skip_update, that field has now been renamed. An attempt
has been made to keep the flag manipulation code small and fast since
it's used in the heart of the __schedule() fast path.
For the !CONFIG_SCHED_DEBUG case the only object code change (other
than addresses) is the following change to reset RQCF_ACT_SKIP inside
of __schedule(),
- c7 83 38 09 00 00 00 movl $0x0,0x938(%rbx)
- 00 00 00
+ 83 a3 38 09 00 00 fc andl $0xfffffffc,0x938(%rbx)
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Matt Fleming <matt@codeblueprint.co.uk>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Byungchul Park <byungchul.park@lge.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Luca Abeni <luca.abeni@unitn.it>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Wanpeng Li <wanpeng.li@hotmail.com>
Cc: Yuyang Du <yuyang.du@intel.com>
Link: http://lkml.kernel.org/r/20160921133813.31976-8-matt@codeblueprint.co.uk
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add the update_rq_clock() call at the top of the callstack instead of
at the bottom where we find it missing, this to aid later effort to
minimize the number of update_rq_lock() calls.
WARNING: CPU: 30 PID: 194 at ../kernel/sched/sched.h:797 assert_clock_updated()
rq->clock_update_flags < RQCF_ACT_SKIP
Call Trace:
dump_stack()
__warn()
warn_slowpath_fmt()
assert_clock_updated.isra.63.part.64()
can_migrate_task()
load_balance()
pick_next_task_fair()
__schedule()
schedule()
worker_thread()
kthread()
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Instead of adding the update_rq_clock() all the way at the bottom of
the callstack, add one at the top, this to aid later effort to
minimize update_rq_lock() calls.
WARNING: CPU: 0 PID: 1 at ../kernel/sched/sched.h:797 detach_task_cfs_rq()
rq->clock_update_flags < RQCF_ACT_SKIP
Call Trace:
dump_stack()
__warn()
warn_slowpath_fmt()
detach_task_cfs_rq()
switched_from_fair()
__sched_setscheduler()
_sched_setscheduler()
sched_set_stop_task()
cpu_stop_create()
__smpboot_create_thread.part.2()
smpboot_register_percpu_thread_cpumask()
cpu_stop_init()
do_one_initcall()
? print_cpu_info()
kernel_init_freeable()
? rest_init()
kernel_init()
ret_from_fork()
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In the following scenario, thread #1 should back off its attempt to lock
ww1 and unlock ww2 (assuming the acquire context stamps are ordered
accordingly).
Thread #0 Thread #1
--------- ---------
successfully lock ww2
set ww1->base.owner
attempt to lock ww1
confirm ww1->ctx == NULL
enter mutex_spin_on_owner
set ww1->ctx
What was likely to happen previously is:
attempt to lock ww2
refuse to spin because
ww2->ctx != NULL
schedule()
detect thread #0 is off CPU
stop optimistic spin
return -EDEADLK
unlock ww2
wakeup thread #0
lock ww2
Now, we are more likely to see:
detect ww1->ctx != NULL
stop optimistic spin
return -EDEADLK
unlock ww2
successfully lock ww2
... because thread #1 will stop its optimistic spin as soon as possible.
The whole scenario is quite unlikely, since it requires thread #1 to get
between thread #0 setting the owner and setting the ctx. But since we're
idling here anyway, the additional check is basically free.
Found by inspection.
Signed-off-by: Nicolai Hähnle <Nicolai.Haehnle@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Maarten Lankhorst <dev@mblankhorst.nl>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dri-devel@lists.freedesktop.org
Link: http://lkml.kernel.org/r/1482346000-9927-10-git-send-email-nhaehnle@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Instead of inlining __mutex_lock_common() 5 times, once for each
{state,ww} variant. Reduce this to two, ww and !ww.
Then add __always_inline to mutex_optimistic_spin(), so that that will
get inlined all 4 remaining times, for all {waiter,ww} variants.
text data bss dec hex filename
6301 0 0 6301 189d defconfig-build/kernel/locking/mutex.o
4053 0 0 4053 fd5 defconfig-build/kernel/locking/mutex.o
4257 0 0 4257 10a1 defconfig-build/kernel/locking/mutex.o
This reduces total text size and better separates the ww and !ww mutex
code generation.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add regular waiters in stamp order. Keep adding waiters that have no
context in FIFO order and take care not to starve them.
While adding our task as a waiter, back off if we detect that there is
a waiter with a lower stamp in front of us.
Make sure to call lock_contended even when we back off early.
For w/w mutexes, being first in the wait list is only stable when
taking the lock without a context. Therefore, the purpose of the first
flag is split into two: 'first' remains to indicate whether we want to
spin optimistically, while 'handoff' indicates that we should be
prepared to accept a handoff.
For w/w locking with a context, we always accept handoffs after the
first schedule(), to handle the following sequence of events:
1. Task #0 unlocks and hands off to Task #2 which is first in line
2. Task #1 adds itself in front of Task #2
3. Task #2 wakes up and must accept the handoff even though it is no
longer first in line
Signed-off-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= <Nicolai.Haehnle@amd.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Maarten Lankhorst <dev@mblankhorst.nl>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dri-devel@lists.freedesktop.org
Link: http://lkml.kernel.org/r/1482346000-9927-7-git-send-email-nhaehnle@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While reviewing the ww_mutex patches, I noticed that it was still
possible to (incorrectly) succeed for (incorrect) code like:
mutex_lock(&a);
mutex_lock(&a);
This was possible if the second mutex_lock() would block (as expected)
but then receive a spurious wakeup. At that point it would find itself
at the front of the queue, request a handoff and instantly claim
ownership and continue, since owner would point to itself.
Avoid this scenario and simplify the code by introducing a third low
bit to signal handoff pickup. So once we request handoff, unlock
clears the handoff bit and sets the pickup bit along with the new
owner.
This also removes the need for the .handoff argument to
__mutex_trylock(), since that becomes superfluous with PICKUP.
In order to guarantee enough low bits, ensure task_struct alignment is
at least L1_CACHE_BYTES (which seems a good ideal regardless).
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 9d659ae14b ("locking/mutex: Add lock handoff to avoid starvation")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The use of any kind of wait queue is an overkill for pcpu-rwsems.
While one option would be to use the less heavy simple (swait)
flavor, this is still too much for what pcpu-rwsems needs. For one,
we do not care about any sort of queuing in that the only (rare) time
writers (and readers, for that matter) are queued is when trying to
acquire the regular contended rw_sem. There cannot be any further
queuing as writers are serialized by the rw_sem in the first place.
Given that percpu_down_write() must not be called after exit_notify(),
we can replace the bulky waitqueue with rcuwait such that a writer
can wait for its turn to take the lock. As such, we can avoid the
queue handling and locking overhead.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dave@stgolabs.net
Link: http://lkml.kernel.org/r/1484148146-14210-3-git-send-email-dave@stgolabs.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
rcuwait provides support for (single) RCU-safe task wait/wake functionality,
with the caveat that it must not be called after exit_notify(), such that
we avoid racing with rcu delayed_put_task_struct callbacks, task_struct
being rcu unaware in this context -- for which we similarly have
task_rcu_dereference() magic, but with different return semantics, which
can conflict with the wakeup side.
The interfaces are quite straightforward:
rcuwait_wait_event()
rcuwait_wake_up()
More details are in the comments, but it's perhaps worth mentioning at least,
that users must provide proper serialization when waiting on a condition, and
avoid corrupting a concurrent waiter. Also care must be taken between the task
and the condition for when calling the wakeup -- we cannot miss wakeups. When
porting users, this is for example, a given when using waitqueues in that
everything is done under the q->lock. As such, it can remove sources of non
preemptable unbounded work for realtime.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dave@stgolabs.net
Link: http://lkml.kernel.org/r/1484148146-14210-2-git-send-email-dave@stgolabs.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This is a nasty interface and setting the state of a foreign task must
not be done. As of the following commit:
be628be095 ("bcache: Make gc wakeup sane, remove set_task_state()")
... everyone in the kernel calls set_task_state() with current, allowing
the helper to be removed.
However, as the comment indicates, it is still around for those archs
where computing current is more expensive than using a pointer, at least
in theory. An important arch that is affected is arm64, however this has
been addressed now [1] and performance is up to par making no difference
with either calls.
Of all the callers, if any, it's the locking bits that would care most
about this -- ie: we end up passing a tsk pointer to a lot of the lock
slowpath, and setting ->state on that. The following numbers are based
on two tests: a custom ad-hoc microbenchmark that just measures
latencies (for ~65 million calls) between get_task_state() vs
get_current_state().
Secondly for a higher overview, an unlink microbenchmark was used,
which pounds on a single file with open, close,unlink combos with
increasing thread counts (up to 4x ncpus). While the workload is quite
unrealistic, it does contend a lot on the inode mutex or now rwsem.
[1] https://lkml.kernel.org/r/1483468021-8237-1-git-send-email-mark.rutland@arm.com
== 1. x86-64 ==
Avg runtime set_task_state(): 601 msecs
Avg runtime set_current_state(): 552 msecs
vanilla dirty
Hmean unlink1-processes-2 36089.26 ( 0.00%) 38977.33 ( 8.00%)
Hmean unlink1-processes-5 28555.01 ( 0.00%) 29832.55 ( 4.28%)
Hmean unlink1-processes-8 37323.75 ( 0.00%) 44974.57 ( 20.50%)
Hmean unlink1-processes-12 43571.88 ( 0.00%) 44283.01 ( 1.63%)
Hmean unlink1-processes-21 34431.52 ( 0.00%) 38284.45 ( 11.19%)
Hmean unlink1-processes-30 34813.26 ( 0.00%) 37975.17 ( 9.08%)
Hmean unlink1-processes-48 37048.90 ( 0.00%) 39862.78 ( 7.59%)
Hmean unlink1-processes-79 35630.01 ( 0.00%) 36855.30 ( 3.44%)
Hmean unlink1-processes-110 36115.85 ( 0.00%) 39843.91 ( 10.32%)
Hmean unlink1-processes-141 32546.96 ( 0.00%) 35418.52 ( 8.82%)
Hmean unlink1-processes-172 34674.79 ( 0.00%) 36899.21 ( 6.42%)
Hmean unlink1-processes-203 37303.11 ( 0.00%) 36393.04 ( -2.44%)
Hmean unlink1-processes-224 35712.13 ( 0.00%) 36685.96 ( 2.73%)
== 2. ppc64le ==
Avg runtime set_task_state(): 938 msecs
Avg runtime set_current_state: 940 msecs
vanilla dirty
Hmean unlink1-processes-2 19269.19 ( 0.00%) 30704.50 ( 59.35%)
Hmean unlink1-processes-5 20106.15 ( 0.00%) 21804.15 ( 8.45%)
Hmean unlink1-processes-8 17496.97 ( 0.00%) 17243.28 ( -1.45%)
Hmean unlink1-processes-12 14224.15 ( 0.00%) 17240.21 ( 21.20%)
Hmean unlink1-processes-21 14155.66 ( 0.00%) 15681.23 ( 10.78%)
Hmean unlink1-processes-30 14450.70 ( 0.00%) 15995.83 ( 10.69%)
Hmean unlink1-processes-48 16945.57 ( 0.00%) 16370.42 ( -3.39%)
Hmean unlink1-processes-79 15788.39 ( 0.00%) 14639.27 ( -7.28%)
Hmean unlink1-processes-110 14268.48 ( 0.00%) 14377.40 ( 0.76%)
Hmean unlink1-processes-141 14023.65 ( 0.00%) 16271.69 ( 16.03%)
Hmean unlink1-processes-172 13417.62 ( 0.00%) 16067.55 ( 19.75%)
Hmean unlink1-processes-203 15293.08 ( 0.00%) 15440.40 ( 0.96%)
Hmean unlink1-processes-234 13719.32 ( 0.00%) 16190.74 ( 18.01%)
Hmean unlink1-processes-265 16400.97 ( 0.00%) 16115.22 ( -1.74%)
Hmean unlink1-processes-296 14388.60 ( 0.00%) 16216.13 ( 12.70%)
Hmean unlink1-processes-320 15771.85 ( 0.00%) 15905.96 ( 0.85%)
x86-64 (known to be fast for get_current()/this_cpu_read_stable() caching)
and ppc64 (with paca) show similar improvements in the unlink microbenches.
The small delta for ppc64 (2ms), does not represent the gains on the unlink
runs. In the case of x86, there was a decent amount of variation in the
latency runs, but always within a 20 to 50ms increase), ppc was more constant.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dave@stgolabs.net
Cc: mark.rutland@arm.com
Link: http://lkml.kernel.org/r/1483479794-14013-5-git-send-email-dave@stgolabs.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
It's possible to set up PEBS events to get only errors and not
any data, like on SNB-X (model 45) and IVB-EP (model 62)
via 2 perf commands running simultaneously:
taskset -c 1 ./perf record -c 4 -e branches:pp -j any -C 10
This leads to a soft lock up, because the error path of the
intel_pmu_drain_pebs_nhm() does not account event->hw.interrupt
for error PEBS interrupts, so in case you're getting ONLY
errors you don't have a way to stop the event when it's over
the max_samples_per_tick limit:
NMI watchdog: BUG: soft lockup - CPU#22 stuck for 22s! [perf_fuzzer:5816]
...
RIP: 0010:[<ffffffff81159232>] [<ffffffff81159232>] smp_call_function_single+0xe2/0x140
...
Call Trace:
? trace_hardirqs_on_caller+0xf5/0x1b0
? perf_cgroup_attach+0x70/0x70
perf_install_in_context+0x199/0x1b0
? ctx_resched+0x90/0x90
SYSC_perf_event_open+0x641/0xf90
SyS_perf_event_open+0x9/0x10
do_syscall_64+0x6c/0x1f0
entry_SYSCALL64_slow_path+0x25/0x25
Add perf_event_account_interrupt() which does the interrupt
and frequency checks and call it from intel_pmu_drain_pebs_nhm()'s
error path.
We keep the pending_kill and pending_wakeup logic only in the
__perf_event_overflow() path, because they make sense only if
there's any data to deliver.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vince@deater.net>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Link: http://lkml.kernel.org/r/1482931866-6018-2-git-send-email-jolsa@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Di Shen reported a race between two concurrent sys_perf_event_open()
calls where both try and move the same pre-existing software group
into a hardware context.
The problem is exactly that described in commit:
f63a8daa58 ("perf: Fix event->ctx locking")
... where, while we wait for a ctx->mutex acquisition, the event->ctx
relation can have changed under us.
That very same commit failed to recognise sys_perf_event_context() as an
external access vector to the events and thereby didn't apply the
established locking rules correctly.
So while one sys_perf_event_open() call is stuck waiting on
mutex_lock_double(), the other (which owns said locks) moves the group
about. So by the time the former sys_perf_event_open() acquires the
locks, the context we've acquired is stale (and possibly dead).
Apply the established locking rules as per perf_event_ctx_lock_nested()
to the mutex_lock_double() for the 'move_group' case. This obviously means
we need to validate state after we acquire the locks.
Reported-by: Di Shen (Keen Lab)
Tested-by: John Dias <joaodias@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Min Chong <mchong@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Fixes: f63a8daa58 ("perf: Fix event->ctx locking")
Link: http://lkml.kernel.org/r/20170106131444.GZ3174@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
There is problem with installing an event in a task that is 'stuck' on
an offline CPU.
Blocked tasks are not dis-assosciated from offlined CPUs, after all, a
blocked task doesn't run and doesn't require a CPU etc.. Only on
wakeup do we ammend the situation and place the task on a available
CPU.
If we hit such a task with perf_install_in_context() we'll loop until
either that task wakes up or the CPU comes back online, if the task
waking depends on the event being installed, we're stuck.
While looking into this issue, I also spotted another problem, if we
hit a task with perf_install_in_context() that is in the middle of
being migrated, that is we observe the old CPU before sending the IPI,
but run the IPI (on the old CPU) while the task is already running on
the new CPU, things also go sideways.
Rework things to rely on task_curr() -- outside of rq->lock -- which
is rather tricky. Imagine the following scenario where we're trying to
install the first event into our task 't':
CPU0 CPU1 CPU2
(current == t)
t->perf_event_ctxp[] = ctx;
smp_mb();
cpu = task_cpu(t);
switch(t, n);
migrate(t, 2);
switch(p, t);
ctx = t->perf_event_ctxp[]; // must not be NULL
smp_function_call(cpu, ..);
generic_exec_single()
func();
spin_lock(ctx->lock);
if (task_curr(t)) // false
add_event_to_ctx();
spin_unlock(ctx->lock);
perf_event_context_sched_in();
spin_lock(ctx->lock);
// sees event
So its CPU0's store of t->perf_event_ctxp[] that must not go 'missing'.
Because if CPU2's load of that variable were to observe NULL, it would
not try to schedule the ctx and we'd have a task running without its
counter, which would be 'bad'.
As long as we observe !NULL, we'll acquire ctx->lock. If we acquire it
first and not see the event yet, then CPU0 must observe task_curr()
and retry. If the install happens first, then we must see the event on
sched-in and all is well.
I think we can translate the first part (until the 'must not be NULL')
of the scenario to a litmus test like:
C C-peterz
{
}
P0(int *x, int *y)
{
int r1;
WRITE_ONCE(*x, 1);
smp_mb();
r1 = READ_ONCE(*y);
}
P1(int *y, int *z)
{
WRITE_ONCE(*y, 1);
smp_store_release(z, 1);
}
P2(int *x, int *z)
{
int r1;
int r2;
r1 = smp_load_acquire(z);
smp_mb();
r2 = READ_ONCE(*x);
}
exists
(0:r1=0 /\ 2:r1=1 /\ 2:r2=0)
Where:
x is perf_event_ctxp[],
y is our tasks's CPU, and
z is our task being placed on the rq of CPU2.
The P0 smp_mb() is the one added by this patch, ordering the store to
perf_event_ctxp[] from find_get_context() and the load of task_cpu()
in task_function_call().
The smp_store_release/smp_load_acquire model the RCpc locking of the
rq->lock and the smp_mb() of P2 is the context switch switching from
whatever CPU2 was running to our task 't'.
This litmus test evaluates into:
Test C-peterz Allowed
States 7
0:r1=0; 2:r1=0; 2:r2=0;
0:r1=0; 2:r1=0; 2:r2=1;
0:r1=0; 2:r1=1; 2:r2=1;
0:r1=1; 2:r1=0; 2:r2=0;
0:r1=1; 2:r1=0; 2:r2=1;
0:r1=1; 2:r1=1; 2:r2=0;
0:r1=1; 2:r1=1; 2:r2=1;
No
Witnesses
Positive: 0 Negative: 7
Condition exists (0:r1=0 /\ 2:r1=1 /\ 2:r2=0)
Observation C-peterz Never 0 7
Hash=e427f41d9146b2a5445101d3e2fcaa34
And the strong and weak model agree.
Reported-by: Mark Rutland <mark.rutland@arm.com>
Tested-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Vince Weaver <vincent.weaver@maine.edu>
Cc: Will Deacon <will.deacon@arm.com>
Cc: jeremy.linton@arm.com
Link: http://lkml.kernel.org/r/20161209135900.GU3174@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Improve __kernel_text_address()/kernel_text_address() to return
true if the given address is on a kprobe's instruction slot
trampoline.
This can help stacktraces to determine the address is on a
text area or not.
To implement this atomically in is_kprobe_*_slot(), also change
the insn_cache page list to an RCU list.
This changes timings a bit (it delays page freeing to the RCU garbage
collection phase), but none of that is in the hot path.
Note: this change can add small overhead to stack unwinders because
it adds 2 additional checks to __kernel_text_address(). However, the
impact should be very small, because kprobe_insn_pages list has 1 entry
per 256 probes(on x86, on arm/arm64 it will be 1024 probes),
and kprobe_optinsn_pages has 1 entry per 32 probes(on x86).
In most use cases, the number of kprobe events may be less
than 20, which means that is_kprobe_*_slot() will check just one entry.
Tested-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ananth N Mavinakayanahalli <ananth@linux.vnet.ibm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andrey Konovalov <andreyknvl@google.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/148388747896.6869.6354262871751682264.stgit@devbox
[ Improved the changelog and coding style. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull VFIO fixes from Alex Williamson:
- Cleanups and bug fixes for the mtty sample driver (Dan Carpenter)
- Export and make use of has_capability() to fix incorrect use of
ns_capable() for testing task capabilities (Jike Song)
* tag 'vfio-v4.10-rc4' of git://github.com/awilliam/linux-vfio:
vfio/type1: Remove pid_namespace.h include
vfio iommu type1: fix the testing of capability for remote task
capability: export has_capability
vfio-mdev: remove some dead code
vfio-mdev: buffer overflow in ioctl()
vfio-mdev: return -EFAULT if copy_to_user() fails
Pull KVM fixes from Paolo Bonzini:
- fix for module unload vs deferred jump labels (note: there might be
other buggy modules!)
- two NULL pointer dereferences from syzkaller
- also syzkaller: fix emulation of fxsave/fxrstor/sgdt/sidt, problem
made worse during this merge window, "just" kernel memory leak on
releases
- fix emulation of "mov ss" - somewhat serious on AMD, less so on Intel
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
KVM: x86: fix emulation of "MOV SS, null selector"
KVM: x86: fix NULL deref in vcpu_scan_ioapic
KVM: eventfd: fix NULL deref irqbypass consumer
KVM: x86: Introduce segmented_write_std
KVM: x86: flush pending lapic jump label updates on module unload
jump_labels: API for flushing deferred jump label updates
As reported by yangshukui, a permission denial from security_task_wait()
can lead to a soft lockup in zap_pid_ns_processes() since it only expects
sys_wait4() to return 0 or -ECHILD. Further, security_task_wait() can
in general lead to zombies; in the absence of some way to automatically
reparent a child process upon a denial, the hook is not useful. Remove
the security hook and its implementations in SELinux and Smack. Smack
already removed its check from its hook.
Reported-by: yangshukui <yangshukui@huawei.com>
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Paul Moore <paul@paul-moore.com>
Merge core DEBUG_VIRTUAL changes from Laura Abbott. Later arm and arm64
support depends on these.
* aarch64/for-next/debug-virtual:
drivers: firmware: psci: Use __pa_symbol for kernel symbol
mm/usercopy: Switch to using lm_alias
mm/kasan: Switch to using __pa_symbol and lm_alias
kexec: Switch to __pa_symbol
mm: Introduce lm_alias
mm/cma: Cleanup highmem check
lib/Kconfig.debug: Add ARCH_HAS_DEBUG_VIRTUAL
When structs are used to store temporary state in cb[] buffer that is
used with programs and among tail calls, then the generated code will
not always access the buffer in bpf_w chunks. We can ease programming
of it and let this act more natural by allowing for aligned b/h/w/dw
sized access for cb[] ctx member. Various test cases are attached as
well for the selftest suite. Potentially, this can also be reused for
other program types to pass data around.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, when calling convert_ctx_access() callback for the various
program types, we pass in insn->dst_reg, insn->src_reg, insn->off from
the original instruction. This information is needed to rewrite the
instruction that is based on the user ctx structure into a kernel
representation for the ctx. As we'd like to allow access size beyond
just BPF_W, we'd need also insn->code for that in order to decode the
original access size. Given that, lets just pass insn directly to the
convert_ctx_access() callback and work on that to not clutter the
callback with even more arguments we need to pass when everything is
already contained in insn. So lets go through that once, no functional
change.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Modules that use static_key_deferred need a way to synchronize with
any delayed work that is still pending when the module is unloaded.
Introduce static_key_deferred_flush() which flushes any pending
jump label updates.
Signed-off-by: David Matlack <dmatlack@google.com>
Cc: stable@vger.kernel.org
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When the tick is stopped and an interrupt occurs afterward, we check on
that interrupt exit if the next tick needs to be rescheduled. If it
doesn't need any update, we don't want to do anything.
In order to check if the tick needs an update, we compare it against the
clockevent device deadline. Now that's a problem because the clockevent
device is at a lower level than the tick itself if it is implemented
on top of hrtimer.
Every hrtimer share this clockevent device. So comparing the next tick
deadline against the clockevent device deadline is wrong because the
device may be programmed for another hrtimer whose deadline collides
with the tick. As a result we may end up not reprogramming the tick
accidentally.
In a worst case scenario under full dynticks mode, the tick stops firing
as it is supposed to every 1hz, leaving /proc/stat stalled:
Task in a full dynticks CPU
----------------------------
* hrtimer A is queued 2 seconds ahead
* the tick is stopped, scheduled 1 second ahead
* tick fires 1 second later
* on tick exit, nohz schedules the tick 1 second ahead but sees
the clockevent device is already programmed to that deadline,
fooled by hrtimer A, the tick isn't rescheduled.
* hrtimer A is cancelled before its deadline
* tick never fires again until an interrupt happens...
In order to fix this, store the next tick deadline to the tick_sched
local structure and reuse that value later to check whether we need to
reprogram the clock after an interrupt.
On the other hand, ts->sleep_length still wants to know about the next
clock event and not just the tick, so we want to improve the related
comment to avoid confusion.
Reported-by: James Hartsock <hartsjc@redhat.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: Wanpeng Li <wanpeng.li@hotmail.com>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Acked-by: Rik van Riel <riel@redhat.com>
Link: http://lkml.kernel.org/r/1483539124-5693-1-git-send-email-fweisbec@gmail.com
Cc: stable@vger.kernel.org
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Since commit 00cd5c37af ("ptrace: permit ptracing of /sbin/init") we
can now trace init processes. init is initially protected with
SIGNAL_UNKILLABLE which will prevent fatal signals such as SIGSTOP, but
there are a number of paths during tracing where SIGNAL_UNKILLABLE can
be implicitly cleared.
This can result in init becoming stoppable/killable after tracing. For
example, running:
while true; do kill -STOP 1; done &
strace -p 1
and then stopping strace and the kill loop will result in init being
left in state TASK_STOPPED. Sending SIGCONT to init will resume it, but
init will now respond to future SIGSTOP signals rather than ignoring
them.
Make sure that when setting SIGNAL_STOP_CONTINUED/SIGNAL_STOP_STOPPED
that we don't clear SIGNAL_UNKILLABLE.
Link: http://lkml.kernel.org/r/20170104122017.25047-1-jamie.iles@oracle.com
Signed-off-by: Jamie Iles <jamie.iles@oracle.com>
Acked-by: Oleg Nesterov <oleg@redhat.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Both arch_add_memory() and arch_remove_memory() expect a single threaded
context.
For example, arch/x86/mm/init_64.c::kernel_physical_mapping_init() does
not hold any locks over this check and branch:
if (pgd_val(*pgd)) {
pud = (pud_t *)pgd_page_vaddr(*pgd);
paddr_last = phys_pud_init(pud, __pa(vaddr),
__pa(vaddr_end),
page_size_mask);
continue;
}
pud = alloc_low_page();
paddr_last = phys_pud_init(pud, __pa(vaddr), __pa(vaddr_end),
page_size_mask);
The result is that two threads calling devm_memremap_pages()
simultaneously can end up colliding on pgd initialization. This leads
to crash signatures like the following where the loser of the race
initializes the wrong pgd entry:
BUG: unable to handle kernel paging request at ffff888ebfff0000
IP: memcpy_erms+0x6/0x10
PGD 2f8e8fc067 PUD 0 /* <---- Invalid PUD */
Oops: 0000 [#1] SMP DEBUG_PAGEALLOC
CPU: 54 PID: 3818 Comm: systemd-udevd Not tainted 4.6.7+ #13
task: ffff882fac290040 ti: ffff882f887a4000 task.ti: ffff882f887a4000
RIP: memcpy_erms+0x6/0x10
[..]
Call Trace:
? pmem_do_bvec+0x205/0x370 [nd_pmem]
? blk_queue_enter+0x3a/0x280
pmem_rw_page+0x38/0x80 [nd_pmem]
bdev_read_page+0x84/0xb0
Hold the standard memory hotplug mutex over calls to
arch_{add,remove}_memory().
Fixes: 41e94a8513 ("add devm_memremap_pages")
Link: http://lkml.kernel.org/r/148357647831.9498.12606007370121652979.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 01b3f52157 ("bpf: fix allocation warnings in bpf maps and
integer overflow") has added checks for the maximum allocateable size.
It (ab)used KMALLOC_SHIFT_MAX for that purpose.
While this is not incorrect it is not very clean because we already have
KMALLOC_MAX_SIZE for this very reason so let's change both checks to use
KMALLOC_MAX_SIZE instead.
The original motivation for using KMALLOC_SHIFT_MAX was to work around
an incorrect KMALLOC_MAX_SIZE which could lead to allocation warnings
but it is no longer needed since "slab: make sure that KMALLOC_MAX_SIZE
will fit into MAX_ORDER".
Link: http://lkml.kernel.org/r/20161220130659.16461-3-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Andrey Konovalov <andreyknvl@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
