One additional 'rule' for using update_cfs_rq_load_avg() is that one
should call update_tg_load_avg() if it returns true.
Add a bunch of comments to hopefully clarify some of the rules:
o You need to update cfs_rq _before_ any entity attach/detach,
this is important, because while for mathmatical consisency this
isn't strictly needed, it is required for the physical
interpretation of the model, you attach/detach _now_.
o When you modify the cfs_rq avg, you have to then call
update_tg_load_avg() in order to propagate changes upwards.
o (Fair) entities are always attached, switched_{to,from}_fair()
deal with !fair. This directly follows from the definition of the
cfs_rq averages, namely that they are a direct sum of all
(runnable or blocked) entities on that rq.
It is the second rule that this patch enforces, but it adds comments
pertaining to all of them.
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>
Vincent and Yuyang found another few scenarios in which entity
tracking goes wobbly.
The scenarios are basically due to the fact that new tasks are not
immediately attached and thereby differ from the normal situation -- a
task is always attached to a cfs_rq load average (such that it
includes its blocked contribution) and are explicitly
detached/attached on migration to another cfs_rq.
Scenario 1: switch to fair class
p->sched_class = fair_class;
if (queued)
enqueue_task(p);
...
enqueue_entity()
enqueue_entity_load_avg()
migrated = !sa->last_update_time (true)
if (migrated)
attach_entity_load_avg()
check_class_changed()
switched_from() (!fair)
switched_to() (fair)
switched_to_fair()
attach_entity_load_avg()
If @p is a new task that hasn't been fair before, it will have
!last_update_time and, per the above, end up in
attach_entity_load_avg() _twice_.
Scenario 2: change between cgroups
sched_move_group(p)
if (queued)
dequeue_task()
task_move_group_fair()
detach_task_cfs_rq()
detach_entity_load_avg()
set_task_rq()
attach_task_cfs_rq()
attach_entity_load_avg()
if (queued)
enqueue_task();
...
enqueue_entity()
enqueue_entity_load_avg()
migrated = !sa->last_update_time (true)
if (migrated)
attach_entity_load_avg()
Similar as with scenario 1, if @p is a new task, it will have
!load_update_time and we'll end up in attach_entity_load_avg()
_twice_.
Furthermore, notice how we do a detach_entity_load_avg() on something
that wasn't attached to begin with.
As stated above; the problem is that the new task isn't yet attached
to the load tracking and thereby violates the invariant assumption.
This patch remedies this by ensuring a new task is indeed properly
attached to the load tracking on creation, through
post_init_entity_util_avg().
Of course, this isn't entirely as straightforward as one might think,
since the task is hashed before we call wake_up_new_task() and thus
can be poked at. We avoid this by adding TASK_NEW and teaching
cpu_cgroup_can_attach() to refuse such tasks.
Reported-by: Yuyang Du <yuyang.du@intel.com>
Reported-by: Vincent Guittot <vincent.guittot@linaro.org>
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>
A new fair task is detached and attached from/to task_group with:
cgroup_post_fork()
ss->fork(child) := cpu_cgroup_fork()
sched_move_task()
task_move_group_fair()
Which is wrong, because at this point in fork() the task isn't fully
initialized and it cannot 'move' to another group, because its not
attached to any group as yet.
In fact, cpu_cgroup_fork() needs a small part of sched_move_task() so we
can just call this small part directly instead sched_move_task(). And
the task doesn't really migrate because it is not yet attached so we
need the following sequence:
do_fork()
sched_fork()
__set_task_cpu()
cgroup_post_fork()
set_task_rq() # set task group and runqueue
wake_up_new_task()
select_task_rq() can select a new cpu
__set_task_cpu
post_init_entity_util_avg
attach_task_cfs_rq()
activate_task
enqueue_task
This patch makes that happen.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
[ Added TASK_SET_GROUP to set depth properly. ]
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>
Vincent reported that when a new task is moved into a new cgroup it
gets attached twice to the load tracking:
sched_move_task()
task_move_group_fair()
detach_task_cfs_rq()
set_task_rq()
attach_task_cfs_rq()
attach_entity_load_avg()
se->avg.last_load_update = cfs_rq->avg.last_load_update // == 0
enqueue_entity()
enqueue_entity_load_avg()
update_cfs_rq_load_avg()
now = clock()
__update_load_avg(&cfs_rq->avg)
cfs_rq->avg.last_load_update = now
// ages load/util for: now - 0, load/util -> 0
if (migrated)
attach_entity_load_avg()
se->avg.last_load_update = cfs_rq->avg.last_load_update; // now != 0
The problem is that we don't update cfs_rq load_avg before all
entity attach/detach operations. Only enqueue_task() and migrate_task()
do this.
By fixing this, the above will not happen, because the
sched_move_task() attach will have updated cfs_rq's last_load_update
time before attach, and in turn the attach will have set the entity's
last_load_update stamp.
Note that there is a further problem with sched_move_task() calling
detach on a task that hasn't yet been attached; this will be taken
care of in a subsequent patch.
Reported-by: Vincent Guittot <vincent.guittot@linaro.org>
Tested-by: Vincent Guittot <vincent.guittot@linaro.org>
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: Yuyang Du <yuyang.du@intel.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The task_fork_fair() callback already calls __set_task_cpu() and takes
rq->lock.
If we move the sched_class::task_fork callback in sched_fork() under
the existing p->pi_lock, right after its set_task_cpu() call, we can
avoid doing two such calls and omit the IRQ disabling on the rq->lock.
Change to __set_task_cpu() to skip the migration bits, this is a new
task, not a migration. Similarly, make wake_up_new_task() use
__set_task_cpu() for the same reason, the task hasn't actually
migrated as it hasn't ever ran.
This cures the problem of calling migrate_task_rq_fair(), which does
remove_entity_from_load_avg() on tasks that have never been added to
the load avg to begin with.
This bug would result in transiently messed up load_avg values, averaged
out after a few dozen milliseconds. This is probably the reason why
this bug was not found for such a long time.
Reported-by: Vincent Guittot <vincent.guittot@linaro.org>
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>
queued_spin_lock_slowpath() should not worry about another
queued_spin_lock_slowpath() running in interrupt context and
changing node->count by accident, because node->count keeps
the same value every time we enter/leave queued_spin_lock_slowpath().
On some architectures this_cpu_dec() will save/restore irq flags,
which has high overhead. Use the much cheaper __this_cpu_dec() instead.
Signed-off-by: Pan Xinhui <xinhui.pan@linux.vnet.ibm.com>
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: Waiman.Long@hpe.com
Link: http://lkml.kernel.org/r/1465886247-3773-1-git-send-email-xinhui.pan@linux.vnet.ibm.com
[ Rewrote changelog. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Commit:
fde7d22e01aa ("sched/fair: Fix overly small weight for interactive group entities")
did something non-obvious but also did it buggy yet latent.
The problem was exposed for real by a later commit in the v4.7 merge window:
2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
... after which tg->load_avg and cfs_rq->load.weight had different
units (10 bit fixed point and 20 bit fixed point resp.).
Add a comment to explain the use of cfs_rq->load.weight over the
'natural' cfs_rq->avg.load_avg and add scale_load_down() to correct
for the difference in unit.
Since this is (now, as per a previous commit) the only user of
calc_tg_weight(), collapse it.
The effects of this bug should be randomly inconsistent SMP-balancing
of cgroups workloads.
Reported-by: Jirka Hladky <jhladky@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
Fixes: fde7d22e01aa ("sched/fair: Fix overly small weight for interactive group entities")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Starting with the following commit:
fde7d22e01aa ("sched/fair: Fix overly small weight for interactive group entities")
calc_tg_weight() doesn't compute the right value as expected by effective_load().
The difference is in the 'correction' term. In order to ensure \Sum
rw_j >= rw_i we cannot use tg->load_avg directly, since that might be
lagging a correction on the current cfs_rq->avg.load_avg value.
Therefore we use tg->load_avg - cfs_rq->tg_load_avg_contrib +
cfs_rq->avg.load_avg.
Now, per the referenced commit, calc_tg_weight() doesn't use
cfs_rq->avg.load_avg, as is later used in @w, but uses
cfs_rq->load.weight instead.
So stop using calc_tg_weight() and do it explicitly.
The effects of this bug are wake_affine() making randomly
poor choices in cgroup-intense workloads.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org> # v4.3+
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: fde7d22e01aa ("sched/fair: Fix overly small weight for interactive group entities")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
With the following fix:
70595b479ce1 ("x86/power/64: Fix crash whan the hibernation code passes control to the image kernel")
... there is no longer a problem with hibernation resuming a
KASLR-booted kernel image, so remove the restriction.
Signed-off-by: Kees Cook <keescook@chromium.org>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Baoquan He <bhe@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Len Brown <len.brown@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Linux PM list <linux-pm@vger.kernel.org>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephen Smalley <sds@tycho.nsa.gov>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: linux-doc@vger.kernel.org
Link: http://lkml.kernel.org/r/20160613221002.GA29719@www.outflux.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Thomas Gleixner:
"A couple of scheduler fixes:
- force watchdog reset while processing sysrq-w
- fix a deadlock when enabling trace events in the scheduler
- fixes to the throttled next buddy logic
- fixes for the average accounting (missing serialization and
underflow handling)
- allow kernel threads for fallback to online but not active cpus"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/core: Allow kthreads to fall back to online && !active cpus
sched/fair: Do not announce throttled next buddy in dequeue_task_fair()
sched/fair: Initialize throttle_count for new task-groups lazily
sched/fair: Fix cfs_rq avg tracking underflow
kernel/sysrq, watchdog, sched/core: Reset watchdog on all CPUs while processing sysrq-w
sched/debug: Fix deadlock when enabling sched events
sched/fair: Fix post_init_entity_util_avg() serialization
Pull locking fix from Thomas Gleixner:
"A single fix to address a race in the static key logic"
* 'locking-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
locking/static_key: Fix concurrent static_key_slow_inc()
Commit b235beea9e99 ("Clarify naming of thread info/stack allocators")
breaks the build on some powerpc configs, where THREAD_SIZE < PAGE_SIZE:
kernel/fork.c:235:2: error: implicit declaration of function 'free_thread_stack'
kernel/fork.c:355:8: error: assignment from incompatible pointer type
stack = alloc_thread_stack_node(tsk, node);
^
Fix it by renaming free_stack() to free_thread_stack(), and updating the
return type of alloc_thread_stack_node().
Fixes: b235beea9e99 ("Clarify naming of thread info/stack allocators")
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge misc fixes from Andrew Morton:
"Two weeks worth of fixes here"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (41 commits)
init/main.c: fix initcall_blacklisted on ia64, ppc64 and parisc64
autofs: don't get stuck in a loop if vfs_write() returns an error
mm/page_owner: avoid null pointer dereference
tools/vm/slabinfo: fix spelling mistake: "Ocurrences" -> "Occurrences"
fs/nilfs2: fix potential underflow in call to crc32_le
oom, suspend: fix oom_reaper vs. oom_killer_disable race
ocfs2: disable BUG assertions in reading blocks
mm, compaction: abort free scanner if split fails
mm: prevent KASAN false positives in kmemleak
mm/hugetlb: clear compound_mapcount when freeing gigantic pages
mm/swap.c: flush lru pvecs on compound page arrival
memcg: css_alloc should return an ERR_PTR value on error
memcg: mem_cgroup_migrate() may be called with irq disabled
hugetlb: fix nr_pmds accounting with shared page tables
Revert "mm: disable fault around on emulated access bit architecture"
Revert "mm: make faultaround produce old ptes"
mailmap: add Boris Brezillon's email
mailmap: add Antoine Tenart's email
mm, sl[au]b: add __GFP_ATOMIC to the GFP reclaim mask
mm: mempool: kasan: don't poot mempool objects in quarantine
...
Tetsuo has reported the following potential oom_killer_disable vs.
oom_reaper race:
(1) freeze_processes() starts freezing user space threads.
(2) Somebody (maybe a kenrel thread) calls out_of_memory().
(3) The OOM killer calls mark_oom_victim() on a user space thread
P1 which is already in __refrigerator().
(4) oom_killer_disable() sets oom_killer_disabled = true.
(5) P1 leaves __refrigerator() and enters do_exit().
(6) The OOM reaper calls exit_oom_victim(P1) before P1 can call
exit_oom_victim(P1).
(7) oom_killer_disable() returns while P1 not yet finished
(8) P1 perform IO/interfere with the freezer.
This situation is unfortunate. We cannot move oom_killer_disable after
all the freezable kernel threads are frozen because the oom victim might
depend on some of those kthreads to make a forward progress to exit so
we could deadlock. It is also far from trivial to teach the oom_reaper
to not call exit_oom_victim() because then we would lose a guarantee of
the OOM killer and oom_killer_disable forward progress because
exit_mm->mmput might block and never call exit_oom_victim.
It seems the easiest way forward is to workaround this race by calling
try_to_freeze_tasks again after oom_killer_disable. This will make sure
that all the tasks are frozen or it bails out.
Fixes: 449d777d7ad6 ("mm, oom_reaper: clear TIF_MEMDIE for all tasks queued for oom_reaper")
Link: http://lkml.kernel.org/r/1466597634-16199-1-git-send-email-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We've had the thread info allocated together with the thread stack for
most architectures for a long time (since the thread_info was split off
from the task struct), but that is about to change.
But the patches that move the thread info to be off-stack (and a part of
the task struct instead) made it clear how confused the allocator and
freeing functions are.
Because the common case was that we share an allocation with the thread
stack and the thread_info, the two pointers were identical. That
identity then meant that we would have things like
ti = alloc_thread_info_node(tsk, node);
...
tsk->stack = ti;
which certainly _worked_ (since stack and thread_info have the same
value), but is rather confusing: why are we assigning a thread_info to
the stack? And if we move the thread_info away, the "confusing" code
just gets to be entirely bogus.
So remove all this confusion, and make it clear that we are doing the
stack allocation by renaming and clarifying the function names to be
about the stack. The fact that the thread_info then shares the
allocation is an implementation detail, and not really about the
allocation itself.
This is a pure renaming and type fix: we pass in the same pointer, it's
just that we clarify what the pointer means.
The ia64 code that actually only has one single allocation (for all of
task_struct, thread_info and kernel thread stack) now looks a bit odd,
but since "tsk->stack" is actually not even used there, that oddity
doesn't matter. It would be a separate thing to clean that up, I
intentionally left the ia64 changes as a pure brute-force renaming and
type change.
Acked-by: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently phys_to_pfn_t() is an exported symbol to allow nfit_test to
override it and indicate that nfit_test-pmem is not device-mapped. Now,
we want to enable nfit_test to operate without DMA_CMA and the pmem it
provides will no longer be physically contiguous, i.e. won't be capable
of supporting direct_access requests larger than a page. Make
pmem_direct_access() a weak symbol so that it can be replaced by the
tools/testing/nvdimm/ version, and move phys_to_pfn_t() to a static
inline now that it no longer needs to be overridden.
Acked-by: Johannes Thumshirn <jthumshirn@suse.de>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Capability sets attached to files must be ignored except in the
user namespaces where the mounter is privileged, i.e. s_user_ns
and its descendants. Otherwise a vector exists for gaining
privileges in namespaces where a user is not already privileged.
Add a new helper function, current_in_user_ns(), to test whether a user
namespace is the same as or a descendant of another namespace.
Use this helper to determine whether a file's capability set
should be applied to the caps constructed during exec.
--EWB Replaced in_userns with the simpler current_in_userns.
Acked-by: Serge Hallyn <serge.hallyn@canonical.com>
Signed-off-by: Seth Forshee <seth.forshee@canonical.com>
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com>
During CPU hotplug, CPU_ONLINE callbacks are run while the CPU is
online but not active. A CPU_ONLINE callback may create or bind a
kthread so that its cpus_allowed mask only allows the CPU which is
being brought online. The kthread may start executing before the CPU
is made active and can end up in select_fallback_rq().
In such cases, the expected behavior is selecting the CPU which is
coming online; however, because select_fallback_rq() only chooses from
active CPUs, it determines that the task doesn't have any viable CPU
in its allowed mask and ends up overriding it to cpu_possible_mask.
CPU_ONLINE callbacks should be able to put kthreads on the CPU which
is coming online. Update select_fallback_rq() so that it follows
cpu_online() rather than cpu_active() for kthreads.
Reported-by: Gautham R Shenoy <ego@linux.vnet.ibm.com>
Tested-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Abdul Haleem <abdhalee@linux.vnet.ibm.com>
Cc: Aneesh Kumar <aneesh.kumar@linux.vnet.ibm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: kernel-team@fb.com
Cc: linuxppc-dev@lists.ozlabs.org
Link: http://lkml.kernel.org/r/20160616193504.GB3262@mtj.duckdns.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Hierarchy could be already throttled at this point. Throttled next
buddy could trigger a NULL pointer dereference in pick_next_task_fair().
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ben Segall <bsegall@google.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/146608183552.21905.15924473394414832071.stgit@buzz
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cgroup created inside throttled group must inherit current throttle_count.
Broken throttle_count allows to nominate throttled entries as a next buddy,
later this leads to null pointer dereference in pick_next_task_fair().
This patch initialize cfs_rq->throttle_count at first enqueue: laziness
allows to skip locking all rq at group creation. Lazy approach also allows
to skip full sub-tree scan at throttling hierarchy (not in this patch).
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: bsegall@google.com
Link: http://lkml.kernel.org/r/146608182119.21870.8439834428248129633.stgit@buzz
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The following scenario is possible:
CPU 1 CPU 2
static_key_slow_inc()
atomic_inc_not_zero()
-> key.enabled == 0, no increment
jump_label_lock()
atomic_inc_return()
-> key.enabled == 1 now
static_key_slow_inc()
atomic_inc_not_zero()
-> key.enabled == 1, inc to 2
return
** static key is wrong!
jump_label_update()
jump_label_unlock()
Testing the static key at the point marked by (**) will follow the
wrong path for jumps that have not been patched yet. This can
actually happen when creating many KVM virtual machines with userspace
LAPIC emulation; just run several copies of the following program:
#include <fcntl.h>
#include <unistd.h>
#include <sys/ioctl.h>
#include <linux/kvm.h>
int main(void)
{
for (;;) {
int kvmfd = open("/dev/kvm", O_RDONLY);
int vmfd = ioctl(kvmfd, KVM_CREATE_VM, 0);
close(ioctl(vmfd, KVM_CREATE_VCPU, 1));
close(vmfd);
close(kvmfd);
}
return 0;
}
Every KVM_CREATE_VCPU ioctl will attempt a static_key_slow_inc() call.
The static key's purpose is to skip NULL pointer checks and indeed one
of the processes eventually dereferences NULL.
As explained in the commit that introduced the bug:
706249c222f6 ("locking/static_keys: Rework update logic")
jump_label_update() needs key.enabled to be true. The solution adopted
here is to temporarily make key.enabled == -1, and use go down the
slow path when key.enabled <= 0.
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: <stable@vger.kernel.org> # v4.3+
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 706249c222f6 ("locking/static_keys: Rework update logic")
Link: http://lkml.kernel.org/r/1466527937-69798-1-git-send-email-pbonzini@redhat.com
[ Small stylistic edits to the changelog and the code. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
# echo 1 > options/stacktrace
# echo 1 > events/sched/sched_switch/enable
# cat trace
<idle>-0 [002] d..2 1982.525169: <stack trace>
=> save_stack_trace
=> __ftrace_trace_stack
=> trace_buffer_unlock_commit_regs
=> event_trigger_unlock_commit
=> trace_event_buffer_commit
=> trace_event_raw_event_sched_switch
=> __schedule
=> schedule
=> schedule_preempt_disabled
=> cpu_startup_entry
=> start_secondary
The above shows that we are seeing 6 functions before ever making it to the
caller of the sched_switch event.
# echo stacktrace > events/sched/sched_switch/trigger
# cat trace
<idle>-0 [002] d..3 2146.335208: <stack trace>
=> trace_event_buffer_commit
=> trace_event_raw_event_sched_switch
=> __schedule
=> schedule
=> schedule_preempt_disabled
=> cpu_startup_entry
=> start_secondary
The stacktrace trigger isn't as bad, because it adds its own skip to the
stacktracing, but still has two events extra.
One issue is that if the stacktrace passes its own "regs" then there should
be no addition to the skip, as the regs will not include the functions being
called. This was an issue that was fixed by commit 7717c6be6999 ("tracing:
Fix stacktrace skip depth in trace_buffer_unlock_commit_regs()" as adding
the skip number for kprobes made the probes not have any stack at all.
But since this is only an issue when regs is being used, a skip should be
added if regs is NULL. Now we have:
# echo 1 > options/stacktrace
# echo 1 > events/sched/sched_switch/enable
# cat trace
<idle>-0 [000] d..2 1297.676333: <stack trace>
=> __schedule
=> schedule
=> schedule_preempt_disabled
=> cpu_startup_entry
=> rest_init
=> start_kernel
=> x86_64_start_reservations
=> x86_64_start_kernel
# echo stacktrace > events/sched/sched_switch/trigger
# cat trace
<idle>-0 [002] d..3 1370.759745: <stack trace>
=> __schedule
=> schedule
=> schedule_preempt_disabled
=> cpu_startup_entry
=> start_secondary
And kprobes are not touched.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
None of the code actually wants a thread_info, it all wants a
task_struct, and it's just converting back and forth between the two
("ti->task" to get the task_struct from the thread_info, and
"task_thread_info(task)" to go the other way).
No semantic change.
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The function irq_create_of_mapping() is used to create an interrupt
mapping. However, depending on whether the irqdomain, to which the
interrupt belongs, is part of a hierarchy, determines whether the
mapping is created via calling irq_domain_alloc_irqs() or
irq_create_mapping().
To dispose of the interrupt mapping, drivers call irq_dispose_mapping().
However, this function does not check to see if the irqdomain is part
of a hierarchy or not and simply assumes that it was mapped via calling
irq_create_mapping() so calls irq_domain_disassociate() to unmap the
interrupt.
Fix this by checking to see if the irqdomain is part of a hierarchy and
if so call irq_domain_free_irqs() to free/unmap the interrupt.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Cc: Marc Zyngier <marc.zyngier@arm.com>
Cc: Jiang Liu <jiang.liu@linux.intel.com>
Link: http://lkml.kernel.org/r/1466501002-16368-1-git-send-email-jonathanh@nvidia.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The `events_limit` variable needs to be formatted with %lld and not %ld.
This fixes the following warning discovered by kbuild test robot:
kernel/cgroup_pids.c: In function 'pids_events_show':
kernel/cgroup_pids.c:313:24: warning: format '%ld' expects argument of type
'long int', but argument 3 has type 'long long int' [-Wformat=]
seq_printf(sf, "max %ld\n", atomic64_read(&pids->events_limit));
^
tj: Added explicit (s64) cast as atomic64 switches between long long
and long depending on 32 or 64.
Signed-off-by: Kenny Yu <kennyyu@fb.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
This patch adds more visibility into the pids controller when the controller
rejects a fork request. Whenever fork fails because the limit on the number of
pids in the cgroup is reached, the controller will log this and also notify the
newly added cgroups events file. The `max` key in the events file represents
the number of times fork failed because of the pids controller.
This change also logs only the first time the `max` event counter is
incremented. This is to provide a hint to the user to understand why fork
failed, as users are not yet used to seeing fork failures because of the
pids controller.
Signed-off-by: Kenny Yu <kennyyu@fb.com>
Acked-by: Johannes Weiner <hannes <at> cmpxchg.org>
Signed-off-by: Tejun Heo <tj@kernel.org>
cgroup core expected css_alloc to return an ERR_PTR value on failure
and caused NULL deref if it returned NULL. It's an easy mistake to
make from an alloc function and there's no ambiguity in what's being
indicated. Update css_create() so that it interprets NULL return from
css_alloc as -ENOMEM.
Signed-off-by: Tejun Heo <tj@kernel.org>
Pull time(keeping) updates from John Stultz:
- Handle the 1ns issue with the old refusing to die vsyscall machinery
- More y2038 updates
- Documentation fixes
- Simplify clocksource handling
The tstats_show() function prints a ktime_t variable by converting
it to struct timespec first. The algorithm is ok, but we want to
stop using timespec in general because of the 32-bit time_t
overflow problem.
This changes the code to use struct timespec64, without any
functional change.
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
udelay_test_single() uses ktime_get_ts() to get two timespec values
and calculate the difference between them, while udelay_test_show()
uses the same to printk() the current monotonic time.
Both of these are y2038 safe on all machines, but we want to
get rid of struct timespec anyway, so this converts the code to
use ktime_get_ns() and ktime_get_ts64() respectively.
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: John Stultz <john.stultz@linaro.org>
time_to_tm() takes time_t as an argument.
time_t is not y2038 safe.
Add time64_to_tm() that takes time64_t as an argument
which is y2038 safe.
The plan is to eventually replace all calls to time_to_tm()
by time64_to_tm().
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
The user notices the problem in a raw and real time drift, calling
clock_gettime with CLOCK_REALTIME / CLOCK_MONOTONIC_RAW on a system
with no ntp correction taking place (no ntpd or ptp stuff running).
The problem is, that old_vsyscall_fixup adds an extra 1ns even though
xtime_nsec is already held in full nsecs and the remainder in this
case is 0. Do the rounding up buisness only if needed.
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Thomas Graziadei <thomas.graziadei@omicronenergy.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
In clocksource_enqueue(), it is unnecessary to continue looping
the list, if we find there is an entry that the value of rating
is smaller than the new one. It is safe to be out the loop,
because all of entry are inserted in descending order.
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Minfei Huang <mnghuan@gmail.com>
Signed-off-by: John Stultz <john.stultz@linaro.org>
o When trace_printk() is used with a non constant format descriptor,
it adds a NULL pointer into the trace format section, and the code
isn't prepared to deal with it. This bug appeared by a change that
was added in v3.5.
o The ftracetest (selftests section) can't handle testing histograms
when histograms are not configured. Currently it shows that they
fail the test, when they should state that they are unsupported.
This bug was added in the 4.7 merge window with the addition of
the historgram code.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJXZ/TGAAoJEKKk/i67LK/8OKEH/2pRnbWFh2EHScrhUgpnzzsn
9BSrxsON82KZtaRIl7QOCBXO8XGKdEfRzf6nz85q6I7GghURuzoGPpJ65YTx+InL
Ksg3TVIVPzMTRMoHQRMO6LNaX2Ks76KeyWxA8T7ib6qJFLJjHkgTYg/NE5A7cnNl
dNdz09GpHAgRt5zfETe/oS7pfc76w6x5wWSlt7oaeBXhY3goeq7WnX+/hqX3+/QI
fLlvJi8A8K0PY+xx9Tn/k1mn74cRuT475grHyyypWeEgo7HNU4CjLXKlZclT2hvX
qzv7tmJwccMmEVF0mdEpCttHmnnniJd8oZg4OQK/V3DX1RLKLMEz6OFUwBmXcbw=
=z/B5
-----END PGP SIGNATURE-----
Merge tag 'trace-v4.7-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull tracing fixes from Steven Rostedt:
"Two fixes for the tracing system:
- When trace_printk() is used with a non constant format descriptor,
it adds a NULL pointer into the trace format section, and the code
isn't prepared to deal with it. This bug appeared by a change that
was added in v3.5.
- The ftracetest (selftests section) can't handle testing histograms
when histograms are not configured. Currently it shows that they
fail the test, when they should state that they are unsupported.
This bug was added in the 4.7 merge window with the addition of the
historgram code"
* tag 'trace-v4.7-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftracetest: Fix hist unsupported result in hist selftests
tracing: Handle NULL formats in hold_module_trace_bprintk_format()
Previously, mmio_print_pcidev() put "user" addresses in the trace buffer.
On most architectures, these are the same as CPU physical addresses, but on
microblaze, mips, powerpc, and sparc, they may be something else, typically
a raw BAR value (a bus address as opposed to a CPU address).
Always expose the CPU physical address to avoid this arch-dependent
behavior.
This change should have no user-visible effect because this file currently
depends on CONFIG_HAVE_MMIOTRACE_SUPPORT, which is only defined for x86,
and pci_resource_to_user() is a no-op on x86.
Link: http://lkml.kernel.org/r/20160511190657.5898.4248.stgit@bhelgaas-glaptop2.roam.corp.google.com
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Because tracepoint callbacks are done with preemption enabled, the trace
events are always called with preempt disable due to the
rcu_read_lock_sched_notrace() in __DO_TRACE(). This causes the preempt count
shown in the recorded trace event to be inaccurate. It is always one more
that what the preempt_count was when the tracepoint was called.
If CONFIG_PREEMPT is enabled, subtract 1 from the preempt_count before
recording it in the trace buffer.
Link: http://lkml.kernel.org/r/20160525132537.GA10808@linutronix.de
Reported-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Currently, the trace_printk code chooses which static buffer to use based
on what type of atomic context (NMI, IRQ, etc) it's in. Simplify the
code and make it more robust: simply count the nesting depth and choose
a buffer based on the current nesting depth.
The new code will only drop an event if we nest more than 4 deep,
and the old code was guaranteed to malfunction if that happened.
Link: http://lkml.kernel.org/r/07ab03aecfba25fcce8f9a211b14c9c5e2865c58.1464289095.git.luto@kernel.org
Acked-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Andy Lutomirski <luto@kernel.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
ftrace is very quick to give up on saving the task command line (see
`trace_save_cmdline()`). The workaround for events which really care
about the command line is to explicitly assign it as part of the entry.
However, this doesn't work for kprobe events, as there's no
straightforward way to get access to current->comm. Add a kprobe/uprobe
event variable $comm which provides exactly that.
Link: http://lkml.kernel.org/r/f59b472033b943a370f5f48d0af37698f409108f.1465435894.git.osandov@fb.com
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Convert set_ftrace_pid to use the bitmap like set_event_pid does. This
allows for instances to use the pid filtering as well, and will allow for
function-fork option to set if the children of a traced function should be
traced or not.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The addition of PIDs into a pid_list via the write operation of
set_event_pid is a bit complex. The same operation will be needed for
function tracing pids. Move the code into its own generic function in
trace.c, so that we can avoid duplication of this code.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
To allow other aspects of ftrace to use the pid_list logic, we need to reuse
the seq_file functions. Making the generic part into functions that can be
called by other files will help in this regard.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
As the filtered_pid functions are going to be used by function tracer as
well as trace_events, move the code into the generic trace.c file.
The functions moved are:
trace_find_filtered_pid()
trace_ignore_this_task()
trace_filter_add_remove_task()
Kernel Doc text was also added.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Make the functions used for pid filtering global for tracing, such that the
function tracer can use the pid code as well.
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
If a task uses a non constant string for the format parameter in
trace_printk(), then the trace_printk_fmt variable is set to NULL. This
variable is then saved in the __trace_printk_fmt section.
The function hold_module_trace_bprintk_format() checks to see if duplicate
formats are used by modules, and reuses them if so (saves them to the list
if it is new). But this function calls lookup_format() that does a strcmp()
to the value (which is now NULL) and can cause a kernel oops.
This wasn't an issue till 3debb0a9ddb ("tracing: Fix trace_printk() to print
when not using bprintk()") which added "__used" to the trace_printk_fmt
variable, and before that, the kernel simply optimized it out (no NULL value
was saved).
The fix is simply to handle the NULL pointer in lookup_format() and have the
caller ignore the value if it was NULL.
Link: http://lkml.kernel.org/r/1464769870-18344-1-git-send-email-zhengjun.xing@intel.com
Reported-by: xingzhen <zhengjun.xing@intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Fixes: 3debb0a9ddb ("tracing: Fix trace_printk() to print when not using bprintk()")
Cc: stable@vger.kernel.org # v3.5+
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
As per commit:
b7fa30c9cc48 ("sched/fair: Fix post_init_entity_util_avg() serialization")
> the code generated from update_cfs_rq_load_avg():
>
> if (atomic_long_read(&cfs_rq->removed_load_avg)) {
> s64 r = atomic_long_xchg(&cfs_rq->removed_load_avg, 0);
> sa->load_avg = max_t(long, sa->load_avg - r, 0);
> sa->load_sum = max_t(s64, sa->load_sum - r * LOAD_AVG_MAX, 0);
> removed_load = 1;
> }
>
> turns into:
>
> ffffffff81087064: 49 8b 85 98 00 00 00 mov 0x98(%r13),%rax
> ffffffff8108706b: 48 85 c0 test %rax,%rax
> ffffffff8108706e: 74 40 je ffffffff810870b0 <update_blocked_averages+0xc0>
> ffffffff81087070: 4c 89 f8 mov %r15,%rax
> ffffffff81087073: 49 87 85 98 00 00 00 xchg %rax,0x98(%r13)
> ffffffff8108707a: 49 29 45 70 sub %rax,0x70(%r13)
> ffffffff8108707e: 4c 89 f9 mov %r15,%rcx
> ffffffff81087081: bb 01 00 00 00 mov $0x1,%ebx
> ffffffff81087086: 49 83 7d 70 00 cmpq $0x0,0x70(%r13)
> ffffffff8108708b: 49 0f 49 4d 70 cmovns 0x70(%r13),%rcx
>
> Which you'll note ends up with sa->load_avg -= r in memory at
> ffffffff8108707a.
So I _should_ have looked at other unserialized users of ->load_avg,
but alas. Luckily nikbor reported a similar /0 from task_h_load() which
instantly triggered recollection of this here problem.
Aside from the intermediate value hitting memory and causing problems,
there's another problem: the underflow detection relies on the signed
bit. This reduces the effective width of the variables, IOW its
effectively the same as having these variables be of signed type.
This patch changes to a different means of unsigned underflow
detection to not rely on the signed bit. This allows the variables to
use the 'full' unsigned range. And it does so with explicit LOAD -
STORE to ensure any intermediate value will never be visible in
memory, allowing these unserialized loads.
Note: GCC generates crap code for this, might warrant a look later.
Note2: I say 'full' above, if we end up at U*_MAX we'll still explode;
maybe we should do clamping on add too.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrey Ryabinin <aryabinin@virtuozzo.com>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
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: Yuyang Du <yuyang.du@intel.com>
Cc: bsegall@google.com
Cc: kernel@kyup.com
Cc: morten.rasmussen@arm.com
Cc: pjt@google.com
Cc: steve.muckle@linaro.org
Fixes: 9d89c257dfb9 ("sched/fair: Rewrite runnable load and utilization average tracking")
Link: http://lkml.kernel.org/r/20160617091948.GJ30927@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
This adds a software irq handler for controllers that multiplex
interrupts from multiple devices, but don't know which device generated
the interrupt. For these devices, the irq handler that demuxes must
check every action for every software irq using the same h/w irq in order
to find out which device generated the interrupt. This will inevitably
trigger spurious interrupt detection if we are noting the irq.
The new irq handler does not track the handling for spurious interrupt
detection. An irq that uses this also won't get stats tracked since it
didn't generate the interrupt, nor added to randomness since they are
not random.
Signed-off-by: Keith Busch <keith.busch@intel.com>
Cc: Bjorn Helgaas <bhelgaas@google.com>
Cc: linux-pci@vger.kernel.org
Cc: Jon Derrick <jonathan.derrick@intel.com>
Link: http://lkml.kernel.org/r/1466200821-29159-1-git-send-email-keith.busch@intel.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
