Ensure that cpus specified with the isolcpus= boot commandline
option stay outside of the load balancing in the kernel scheduler.
Operations like load balancing can introduce unwanted latencies,
which is exactly what the isolcpus= commandline is there to prevent.
Previously, simply creating a new cpuset, without even touching the
cpuset.cpus field inside the new cpuset, would undo the effects of
isolcpus=, by creating a scheduler domain spanning the whole system,
and setting up load balancing inside that domain. The cpuset root
cpuset.cpus file is read-only, so there was not even a way to undo
that effect.
This does not impact the majority of cpusets users, since isolcpus=
is a fairly specialized feature used for realtime purposes.
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Clark Williams <williams@redhat.com>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: cgroups@vger.kernel.org
Signed-off-by: Rik van Riel <riel@redhat.com>
Tested-by: David Rientjes <rientjes@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Needed by the next patch. Also makes cpu_isolated_map present
when compiled without SMP and/or with CONFIG_NR_CPUS=1, like
the other cpu masks.
At some point we may want to clean things up so cpumasks do
not exist in UP kernels. Maybe something for the CONFIG_TINY
crowd.
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Clark Williams <williams@redhat.com>
Cc: Li Zefan <lizefan@huawei.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Luiz Capitulino <lcapitulino@redhat.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: cgroups@vger.kernel.org
Signed-off-by: Rik van Riel <riel@redhat.com>
Acked-by: Zefan Li <lizefan@huawei.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Pull livepatching fix from Jiri Kosina:
- fix for potential race with module loading, from Petr Mladek.
The race is very unlikely to be seen in real world and has been found
by code inspection, but should be fixed for 4.0 anyway.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jikos/livepatching:
livepatch: Fix subtle race with coming and going modules
Occasionally, the system can't come back up after suspend/resume
due to problems of device suspending phase. This patch make
PM_TRACE infrastructure cover device suspending phase of
suspend/resume process, and the information in RTC can tell
developers which device suspending function make system hang.
Signed-off-by: Zhonghui Fu <zhonghui.fu@linux.intel.com>
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Add a tuning knob so we can adjust the dirtytime expiration timeout,
which is very useful for testing lazytime.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Jan Kara <jack@suse.cz>
There is a notifier that handles live patches for coming and going modules.
It takes klp_mutex lock to avoid races with coming and going patches but
it does not keep the lock all the time. Therefore the following races are
possible:
1. The notifier is called sometime in STATE_MODULE_COMING. The module
is visible by find_module() in this state all the time. It means that
new patch can be registered and enabled even before the notifier is
called. It might create wrong order of stacked patches, see below
for an example.
2. New patch could still see the module in the GOING state even after
the notifier has been called. It will try to initialize the related
object structures but the module could disappear at any time. There
will stay mess in the structures. It might even cause an invalid
memory access.
This patch solves the problem by adding a boolean variable into struct module.
The value is true after the coming and before the going handler is called.
New patches need to be applied when the value is true and they need to ignore
the module when the value is false.
Note that we need to know state of all modules on the system. The races are
related to new patches. Therefore we do not know what modules will get
patched.
Also note that we could not simply ignore going modules. The code from the
module could be called even in the GOING state until mod->exit() finishes.
If we start supporting patches with semantic changes between function
calls, we need to apply new patches to any still usable code.
See below for an example.
Finally note that the patch solves only the situation when a new patch is
registered. There are no such problems when the patch is being removed.
It does not matter who disable the patch first, whether the normal
disable_patch() or the module notifier. There is nothing to do
once the patch is disabled.
Alternative solutions:
======================
+ reject new patches when a patched module is coming or going; this is ugly
+ wait with adding new patch until the module leaves the COMING and GOING
states; this might be dangerous and complicated; we would need to release
kgr_lock in the middle of the patch registration to avoid a deadlock
with the coming and going handlers; also we might need a waitqueue for
each module which seems to be even bigger overhead than the boolean
+ stop modules from entering COMING and GOING states; wait until modules
leave these states when they are already there; looks complicated; we would
need to ignore the module that asked to stop the others to avoid a deadlock;
also it is unclear what to do when two modules asked to stop others and
both are in COMING state (situation when two new patches are applied)
+ always register/enable new patches and fix up the potential mess (registered
patches order) in klp_module_init(); this is nasty and prone to regressions
in the future development
+ add another MODULE_STATE where the kallsyms are visible but the module is not
used yet; this looks too complex; the module states are checked on "many"
locations
Example of patch stacking breakage:
===================================
The notifier could _not_ _simply_ ignore already initialized module objects.
For example, let's have three patches (P1, P2, P3) for functions a() and b()
where a() is from vmcore and b() is from a module M. Something like:
a() b()
P1 a1() b1()
P2 a2() b2()
P3 a3() b3(3)
If you load the module M after all patches are registered and enabled.
The ftrace ops for function a() and b() has listed the functions in this
order:
ops_a->func_stack -> list(a3,a2,a1)
ops_b->func_stack -> list(b3,b2,b1)
, so the pointer to b3() is the first and will be used.
Then you might have the following scenario. Let's start with state when patches
P1 and P2 are registered and enabled but the module M is not loaded. Then ftrace
ops for b() does not exist. Then we get into the following race:
CPU0 CPU1
load_module(M)
complete_formation()
mod->state = MODULE_STATE_COMING;
mutex_unlock(&module_mutex);
klp_register_patch(P3);
klp_enable_patch(P3);
# STATE 1
klp_module_notify(M)
klp_module_notify_coming(P1);
klp_module_notify_coming(P2);
klp_module_notify_coming(P3);
# STATE 2
The ftrace ops for a() and b() then looks:
STATE1:
ops_a->func_stack -> list(a3,a2,a1);
ops_b->func_stack -> list(b3);
STATE2:
ops_a->func_stack -> list(a3,a2,a1);
ops_b->func_stack -> list(b2,b1,b3);
therefore, b2() is used for the module but a3() is used for vmcore
because they were the last added.
Example of the race with going modules:
=======================================
CPU0 CPU1
delete_module() #SYSCALL
try_stop_module()
mod->state = MODULE_STATE_GOING;
mutex_unlock(&module_mutex);
klp_register_patch()
klp_enable_patch()
#save place to switch universe
b() # from module that is going
a() # from core (patched)
mod->exit();
Note that the function b() can be called until we call mod->exit().
If we do not apply patch against b() because it is in MODULE_STATE_GOING,
it will call patched a() with modified semantic and things might get wrong.
[jpoimboe@redhat.com: use one boolean instead of two]
Signed-off-by: Petr Mladek <pmladek@suse.cz>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Pull full dynticks support for virt guests from Frederic Weisbecker:
"Some measurements showed that disabling the tick on the host while the
guest is running can be interesting on some workloads. Indeed the
host tick is irrelevant while a vcpu runs, it consumes CPU time and cache
footprint for no good reasons.
Full dynticks already works in every context, but RCU prevents it to
be effective outside userspace, because the CPU needs to take part of
RCU grace period completion as long as RCU may be used on it, which is
the case in kernel context.
However guest is similar to userspace and idle in that we know RCU is
unused on such context. Therefore a CPU in guest/userspace/idle context
can let other CPUs report its own RCU quiescent state on its behalf
and shut down the tick safely, provided it isn't needed for other
reasons than RCU. This is called RCU extended quiescent state.
This was already implemented for idle and userspace. This patchset now
brings it for guest contexts through the following steps:
- Generalize the context tracking APIs to also track guest state
- Rename/sanitize a few CPP symbols accordingly
- Report guest entry/exit to RCU and define this context area as an RCU
extended quiescent state."
Signed-off-by: Ingo Molnar <mingo@kernel.org>
introduce user accessible mirror of in-kernel 'struct sk_buff':
struct __sk_buff {
__u32 len;
__u32 pkt_type;
__u32 mark;
__u32 queue_mapping;
};
bpf programs can do:
int bpf_prog(struct __sk_buff *skb)
{
__u32 var = skb->pkt_type;
which will be compiled to bpf assembler as:
dst_reg = *(u32 *)(src_reg + 4) // 4 == offsetof(struct __sk_buff, pkt_type)
bpf verifier will check validity of access and will convert it to:
dst_reg = *(u8 *)(src_reg + offsetof(struct sk_buff, __pkt_type_offset))
dst_reg &= 7
since skb->pkt_type is a bitfield.
Signed-off-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the possibility to obtain raw_smp_processor_id() in
eBPF. Currently, this is only possible in classic BPF where commit
da2033c282 ("filter: add SKF_AD_RXHASH and SKF_AD_CPU") has added
facilities for this.
Perhaps most importantly, this would also allow us to track per CPU
statistics with eBPF maps, or to implement a poor-man's per CPU data
structure through eBPF maps.
Example function proto-type looks like:
u32 (*smp_processor_id)(void) = (void *)BPF_FUNC_get_smp_processor_id;
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This work is similar to commit 4cd3675ebf ("filter: added BPF
random opcode") and adds a possibility for packet sampling in eBPF.
Currently, this is only possible in classic BPF and useful to
combine sampling with f.e. packet sockets, possible also with tc.
Example function proto-type looks like:
u32 (*prandom_u32)(void) = (void *)BPF_FUNC_get_prandom_u32;
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
This proves to be useful with stacked domains, when the current
domain doesn't implement wake-up, but expect the parent to do so.
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Link: https://lkml.kernel.org/r/1426088629-15377-2-git-send-email-marc.zyngier@arm.com
Signed-off-by: Jason Cooper <jason@lakedaemon.net>
After commit 3e1d0bb622 ("audit: Convert int limit
uses to u32"), by converting an int to u32, few conditions will always evaluate
to false.
These warnings were emitted during compilation:
kernel/audit.c: In function ‘audit_set_enabled’:
kernel/audit.c:347:2: warning: comparison of unsigned expression < 0 is always
false [-Wtype-limits]
if (state < AUDIT_OFF || state > AUDIT_LOCKED)
^
kernel/audit.c: In function ‘audit_receive_msg’:
kernel/audit.c:880:9: warning: comparison of unsigned expression < 0 is
always false [-Wtype-limits]
if (s.backlog_wait_time < 0 ||
The following patch removes those unnecessary conditions.
Signed-off-by: Pranith Kumar <bobby.prani@gmail.com>
Signed-off-by: Paul Moore <pmoore@redhat.com>
Commit:
a83fe28e2e ("perf: Fix put_event() ctx lock")
changed the locking logic in put_event() by replacing mutex_lock_nested()
with perf_event_ctx_lock_nested(), but didn't fix the subsequent
mutex_unlock() with a correct counterpart, perf_event_ctx_unlock().
Contexts are thus leaked as a result of incremented refcount
in perf_event_ctx_lock_nested().
Signed-off-by: Leon Yu <chianglungyu@gmail.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Fixes: a83fe28e2e ("perf: Fix put_event() ctx lock")
Link: http://lkml.kernel.org/r/1424954613-5034-1-git-send-email-chianglungyu@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Ingo requested this function be renamed to improve readability,
so I've renamed __clocksource_updatefreq_scale() as well as the
__clocksource_updatefreq_hz/khz() functions to avoid
squishedtogethernames.
This touches some of the sh clocksources, which I've not tested.
The arch/arm/plat-omap change is just a comment change for
consistency.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Daniel Lezcano <daniel.lezcano@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-13-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Print the mask, max_cycles, and max_idle_ns values for
clocksources being registered.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-12-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A long running project has been to clean up remaining uses
of clocksource_register(), replacing it with the simpler
clocksource_register_khz/hz() functions.
However, there are a few cases where we need to self-define
our mult/shift values, so switch the function to a more
obviously internal __clocksource_register() name, and
consolidate much of the internal logic so we don't have
duplication.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: David S. Miller <davem@davemloft.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-10-git-send-email-john.stultz@linaro.org
[ Minor cleanups. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The clocksource watchdog reporting has been less helpful
then desired, as it just printed the delta between
the two clocksources. This prevents any useful analysis
of why the skew occurred.
Thus this patch tries to improve the output when we
mark a clocksource as unstable, printing out the cycle
last and now values for both the current clocksource
and the watchdog clocksource. This will allow us to see
if the result was due to a false positive caused by
a problematic watchdog.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-9-git-send-email-john.stultz@linaro.org
[ Minor cleanups of kernel messages. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
It was suggested that the underflow/overflow protection
should probably throw some sort of warning out, rather
than just silently fixing the issue.
So this patch adds some warnings here. The flag variables
used are not protected by locks, but since we can't print
from the reading functions, just being able to say we
saw an issue in the update interval is useful enough,
and can be slightly racy without real consequence.
The big complication is that we're only under a read
seqlock, so the data could shift under us during
our calculation to see if there was a problem. This
patch avoids this issue by nesting another seqlock
which allows us to snapshot the just required values
atomically. So we shouldn't see false positives.
I also added some basic rate-limiting here, since
on one build machine w/ skewed TSCs it was fairly
noisy at bootup.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-8-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In the case where there is a broken clocksource
where there are multiple actual clocks that
aren't perfectly aligned, we may see small "negative"
deltas when we subtract 'now' from 'cycle_last'.
The values are actually negative with respect to the
clocksource mask value, not necessarily negative
if cast to a s64, but we can check by checking the
delta to see if it is a small (relative to the mask)
negative value (again negative relative to the mask).
If so, we assume we jumped backwards somehow and
instead use zero for our delta.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-7-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When calculating the current delta since the last tick, we
currently have no hard protections to prevent a multiplication
overflow from occuring.
This patch introduces infrastructure to allow a cap that
limits the clocksource read delta value to the 'max_cycles' value,
which is where an overflow would occur.
Since this is in the hotpath, it adds the extra checking under
CONFIG_DEBUG_TIMEKEEPING=y.
There was some concern that capping time like this could cause
problems as we may stop expiring timers, which could go circular
if the timer that triggers time accumulation were mis-scheduled
too far in the future, which would cause time to stop.
However, since the mult overflow would result in a smaller time
value, we would effectively have the same problem there.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-6-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Recently there's been requests for better sanity
checking in the time code, so that it's more clear
when something is going wrong, since timekeeping issues
could manifest in a large number of strange ways in
various subsystems.
Thus, this patch adds some extra infrastructure to
add a check to update_wall_time() to print two new
warnings:
1) if we see the call delayed beyond the 'max_cycles'
overflow point,
2) or if we see the call delayed beyond the clocksource's
'max_idle_ns' value, which is currently 50% of the
overflow point.
This extra infrastructure is conditional on
a new CONFIG_DEBUG_TIMEKEEPING option, also
added in this patch - default off.
Tested this a bit by halting qemu for specified
lengths of time to trigger the warnings.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-5-git-send-email-john.stultz@linaro.org
[ Improved the changelog and the messages a bit. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
There is no need to pass the total request length in the kiocb, as
we already get passed in through the iov_iter argument.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Current approach in handling shadow memory for modules is broken.
Shadow memory could be freed only after memory shadow corresponds it is no
longer used. vfree() called from interrupt context could use memory its
freeing to store 'struct llist_node' in it:
void vfree(const void *addr)
{
...
if (unlikely(in_interrupt())) {
struct vfree_deferred *p = this_cpu_ptr(&vfree_deferred);
if (llist_add((struct llist_node *)addr, &p->list))
schedule_work(&p->wq);
Later this list node used in free_work() which actually frees memory.
Currently module_memfree() called in interrupt context will free shadow
before freeing module's memory which could provoke kernel crash.
So shadow memory should be freed after module's memory. However, such
deallocation order could race with kasan_module_alloc() in module_alloc().
Free shadow right before releasing vm area. At this point vfree()'d
memory is not used anymore and yet not available for other allocations.
New VM_KASAN flag used to indicate that vm area has dynamically allocated
shadow memory so kasan frees shadow only if it was previously allocated.
Signed-off-by: Andrey Ryabinin <a.ryabinin@samsung.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
At grace-period initialization time, RCU checks that all quiescent
states were really reported for the previous grace period. Now that
grace-period cleanup has been split out of grace-period initialization,
this commit also performs those checks at grace-period cleanup time.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
This commit informs RCU of an outgoing CPU just before that CPU invokes
arch_cpu_idle_dead() during its last pass through the idle loop (via a
new CPU_DYING_IDLE notifier value). This change means that RCU need not
deal with outgoing CPUs passing through the scheduler after informing
RCU that they are no longer online. Note that removing the CPU from
the rcu_node ->qsmaskinit bit masks is done at CPU_DYING_IDLE time,
and orphaning callbacks is still done at CPU_DEAD time, the reason being
that at CPU_DEAD time we have another CPU that can adopt them.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
This commit uses a per-CPU variable to make the CPU-offline code path
through the idle loop more precise, so that the outgoing CPU is
guaranteed to make it into the idle loop before it is powered off.
This commit is in preparation for putting the RCU offline-handling
code on this code path, which will eliminate the magic one-jiffy
wait that RCU uses as the maximum time for an outgoing CPU to get
all the way through the scheduler.
The magic one-jiffy wait for incoming CPUs remains a separate issue.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Because that RCU grace-period initialization need no longer exclude
CPU-hotplug operations, this commit eliminates the ->onoff_mutex and
its uses.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Races between CPU hotplug and grace periods can be difficult to resolve,
so the ->onoff_mutex is used to exclude the two events. Unfortunately,
this means that it is impossible for an outgoing CPU to perform the
last bits of its offlining from its last pass through the idle loop,
because sleeplocks cannot be acquired in that context.
This commit avoids these problems by buffering online and offline events
in a new ->qsmaskinitnext field in the leaf rcu_node structures. When a
grace period starts, the events accumulated in this mask are applied to
the ->qsmaskinit field, and, if needed, up the rcu_node tree. The special
case of all CPUs corresponding to a given leaf rcu_node structure being
offline while there are still elements in that structure's ->blkd_tasks
list is handled using a new ->wait_blkd_tasks field. In this case,
propagating the offline bits up the tree is deferred until the beginning
of the grace period after all of the tasks have exited their RCU read-side
critical sections and removed themselves from the list, at which point
the ->wait_blkd_tasks flag is cleared. If one of that leaf rcu_node
structure's CPUs comes back online before the list empties, then the
->wait_blkd_tasks flag is simply cleared.
This of course means that RCU's notion of which CPUs are offline can be
out of date. This is OK because RCU need only wait on CPUs that were
online at the time that the grace period started. In addition, RCU's
force-quiescent-state actions will handle the case where a CPU goes
offline after the grace period starts.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
The rcu_report_unblock_qs_rnp() function is invoked when the
last task blocking the current grace period exits its outermost
RCU read-side critical section. Previously, this was called only
from rcu_read_unlock_special(), and was therefore defined only when
CONFIG_RCU_PREEMPT=y. However, this function will be invoked even when
CONFIG_RCU_PREEMPT=n once CPU-hotplug operations are processed only at
the beginnings of RCU grace periods. The reason for this change is that
the last task on a given leaf rcu_node structure's ->blkd_tasks list
might well exit its RCU read-side critical section between the time that
recent CPU-hotplug operations were applied and when the new grace period
was initialized. This situation could result in RCU waiting forever on
that leaf rcu_node structure, because if all that structure's CPUs were
already offline, there would be no quiescent-state events to drive that
structure's part of the grace period.
This commit therefore moves rcu_report_unblock_qs_rnp() to common code
that is built unconditionally so that the quiescent-state-forcing code
can clean up after this situation, avoiding the grace-period stall.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Currently, the rcu_node tree ->expmask bitmasks are initially set to
reflect the online CPUs. This is pointless, because only the CPUs
preempted within RCU read-side critical sections by the preceding
synchronize_sched_expedited() need to be tracked. This commit therefore
instead sets up these bitmasks based on the state of the ->blkd_tasks
lists.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
I noticed that a helper function with argument type ARG_ANYTHING does
not need to have an initialized value (register).
This can worst case lead to unintented stack memory leakage in future
helper functions if they are not carefully designed, or unintended
application behaviour in case the application developer was not careful
enough to match a correct helper function signature in the API.
The underlying issue is that ARG_ANYTHING should actually be split
into two different semantics:
1) ARG_DONTCARE for function arguments that the helper function
does not care about (in other words: the default for unused
function arguments), and
2) ARG_ANYTHING that is an argument actually being used by a
helper function and *guaranteed* to be an initialized register.
The current risk is low: ARG_ANYTHING is only used for the 'flags'
argument (r4) in bpf_map_update_elem() that internally does strict
checking.
Fixes: 17a5267067 ("bpf: verifier (add verifier core)")
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@plumgrid.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to facilitate clocksource validation, add a
'max_cycles' field to the clocksource structure which
will hold the maximum cycle value that can safely be
multiplied without potentially causing an overflow.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-4-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The clocksource logic has a number of places where we try to
include a safety margin. Most of these are 12% safety margins,
but they are inconsistently applied and sometimes are applied
on top of each other.
Additionally, in the previous patch, we corrected an issue
where we unintentionally in effect created a 50% safety margin,
which these 12.5% margins where then added to.
So to simplify the logic here, this patch removes the various
12.5% margins, and consolidates adding the margin in one place:
clocks_calc_max_nsecs().
Additionally, Linus prefers a 50% safety margin, as it allows
bad clock values to be more easily caught. This should really
have no net effect, due to the corrected issue earlier which
caused greater then 50% margins to be used w/o issue.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Acked-by: Stephen Boyd <sboyd@codeaurora.org> (for the sched_clock.c bit)
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-3-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The previous clocks_calc_max_nsecs() code had some unecessarily
complex bit logic to find the max interval that could cause
multiplication overflows. Since this is not in the hot
path, just do the divide to make it easier to read.
The previous implementation also had a subtle issue
that it avoided overflows with signed 64-bit values, where
as the intervals are always unsigned. This resulted in
overly conservative intervals, which other safety margins
were then added to, reducing the intended interval length.
Signed-off-by: John Stultz <john.stultz@linaro.org>
Cc: Dave Jones <davej@codemonkey.org.uk>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Richard Cochran <richardcochran@gmail.com>
Cc: Stephen Boyd <sboyd@codeaurora.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1426133800-29329-2-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Offline CPUs cannot safely invoke trace events, but such CPUs do execute
within rcu_cpu_notify(). Therefore, this commit removes the trace events
from rcu_cpu_notify(). These trace events are for utilization, against
which rcu_cpu_notify() execution time should be negligible.
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Grace-period initialization normally proceeds quite quickly, so
that it is very difficult to reproduce races against grace-period
initialization. This commit therefore allows grace-period
initialization to be artificially slowed down, increasing
race-reproduction probability. A pair of new Kconfig parameters are
provided, CONFIG_RCU_TORTURE_TEST_SLOW_INIT to enable the slowdowns, and
CONFIG_RCU_TORTURE_TEST_SLOW_INIT_DELAY to specify the number of jiffies
of slowdown to apply. A boot-time parameter named rcutree.gp_init_delay
allows boot-time delay to be specified. By default, no delay will be
applied even if CONFIG_RCU_TORTURE_TEST_SLOW_INIT is set.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
If all CPUs have passed through quiescent states, then stalls might be
due to starvation of the grace-period kthread or to failure to propagate
the quiescent states up the rcu_node combining tree. The current stall
warning messages do not differentiate, so this commit adds a printout
of the root rcu_node structure's ->qsmask field.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Currently, both rcu_cleanup_dead_cpu() and rcu_send_cbs_to_orphanage()
initialize the outgoing CPU's callback list. However, only
rcu_cleanup_dead_cpu() invokes rcu_send_cbs_to_orphanage(), and
it does so unconditionally, which means that only one of these
initializations is required. This commit therefore consolidates the
callback-list initialization with the rest of the callback handling in
rcu_send_cbs_to_orphanage().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
RCU ignores offlined CPUs, so they cannot safely run RCU read-side code.
(They -can- use SRCU, but not RCU.) This means that any use of RCU
during or after the call to arch_cpu_idle_dead(). Unfortunately,
commit 2ed53c0d6c added a complete() call, which will contain RCU
read-side critical sections if there is a task waiting to be awakened.
Which, as it turns out, there almost never is. In my qemu/KVM testing,
the to-be-awakened task is not yet asleep more than 99.5% of the time.
In current mainline, failure is even harder to reproduce, requiring a
virtualized environment that delays the outgoing CPU by at least three
jiffies between the time it exits its stop_machine() task at CPU_DYING
time and the time it calls arch_cpu_idle_dead() from the idle loop.
However, this problem really can occur, especially in virtualized
environments, and therefore really does need to be fixed
This suggests moving back to the polling loop, but using a much shorter
wait, with gentle exponential backoff instead of the old 100-millisecond
wait. Most of the time, the loop will exit without waiting at all,
and almost all of the remaining uses will wait only five microseconds.
If the outgoing CPU is preempted, a loop will wait one jiffy, then
increase the wait by a factor of 11/10ths, rounding up. As before, there
is a five-second timeout.
This commit therefore provides common-code infrastructure to do the
dying-to-surviving CPU handoff in a safe manner. This code also
provides an indication at CPU-online of whether the CPU to be onlined
previously timed out on offline. The new cpu_check_up_prepare() function
returns -EBUSY if this CPU previously took more than five seconds to
go offline, or -EAGAIN if it has not yet managed to go offline. The
rationale for -EAGAIN is that it might still be preempted, so an additional
wait might well find it correctly offlined. Architecture-specific code
can decide how to handle these conditions. Systems in which CPUs take
themselves completely offline might respond to an -EBUSY return as if
it was a zero (success) return. Systems in which the surviving CPU must
take some action might take it at this time, or might simply mark the
other CPU as unusable.
Note that architectures that take the easy way out and simply pass the
-EBUSY and -EAGAIN upwards will change the sysfs API.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: <linux-api@vger.kernel.org>
Cc: <linux-arch@vger.kernel.org>
[ paulmck: Fixed state machine for architectures that don't check earlier
CPU-hotplug results as suggested by James Hogan. ]
This patch adds rq->clock update skip for SCHED_DEADLINE task yield,
to tell update_rq_clock() that we've just updated the clock, so that
we don't do a microscopic update in schedule() and double the
fastpath cost.
Signed-off-by: Wanpeng Li <wanpeng.li@linux.intel.com>
Cc: Juri Lelli <juri.lelli@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1425961200-3809-1-git-send-email-wanpeng.li@linux.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Conflicts:
drivers/net/ethernet/cadence/macb.c
Overlapping changes in macb driver, mostly fixes and cleanups
in 'net' overlapping with the integration of at91_ether into
macb in 'net-next'.
Signed-off-by: David S. Miller <davem@davemloft.net>
contains fixes to ftrace when /proc/sys/kernel/ftrace_enabled and
function tracing are started. Doing the following causes some issues:
# echo 0 > /proc/sys/kernel/ftrace_enabled
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
# echo 1 > /proc/sys/kernel/ftrace_enabled
# echo nop > /sys/kernel/debug/tracing/current_tracer
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
As well as with function tracing too. Pratyush Anand first reported
this issue to me and supplied a patch. When I tested this on my x86
test box, it caused thousands of backtraces and warnings to appear in
dmesg, which also caused a denial of service (a warning for every
function that was listed). I applied Pratyush's patch but it did not
fix the issue for me. I looked into it and found a slight problem
with trampoline accounting. I fixed it and sent Pratyush a patch, but
he said that it did not fix the issue for him.
I later learned tha Pratyush was using an ARM64 server, and when I tested
on my ARM board, I was able to reproduce the same issue as Pratyush.
After applying his patch, it fixed the problem. The above test uncovered
two different bugs, one in x86 and one in ARM and ARM64. As this looked
like it would affect PowerPC, I tested it on my PPC64 box. It too broke,
but neither the patch that fixed ARM or x86 fixed this box (the changes
were all in generic code!). The above test, uncovered two more bugs that
affected PowerPC. Again, the changes were only done to generic code.
It's the way the arch code expected things to be done that was different
between the archs. Some where more sensitive than others.
The rest of this series fixes the PPC bugs as well.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJU/cQSAAoJEEjnJuOKh9lde9sH/1MAPq+6jr7YaEFru0GKajE9
rVHjw8rde/I4tN2UxIVk+Qm6pXRZYpv3OKxHT48EHzkvgm++voioykpJP4IEVrP5
mEDuIcYe28csE2nV5u5Q9kwnZoC86TQW5nVV6zB1Gx/3IEzA8Z046jAov40Jya0y
zqHc/U43JeeVIDIOkwjzbH6OaFEDP13FkF3TO502WJhJLqMo+kPOalIgv0eauKzy
lVCQBSC4WS3rVsgW4W3dSrEBaUxbJxgunjxOuV2DwHj5eghHq0M2MKeIUxBz0PuN
wnhTrpf5cAfshTvYHxKlE0uItdyYfVb7UChAD5zTbBL4kMUFhpb183zVKH8K8kU=
=8R8y
-----END PGP SIGNATURE-----
Merge tag 'trace-fixes-v4.0-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace
Pull seq-buf/ftrace fixes from Steven Rostedt:
"This includes fixes for seq_buf_bprintf() truncation issue. It also
contains fixes to ftrace when /proc/sys/kernel/ftrace_enabled and
function tracing are started. Doing the following causes some issues:
# echo 0 > /proc/sys/kernel/ftrace_enabled
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
# echo 1 > /proc/sys/kernel/ftrace_enabled
# echo nop > /sys/kernel/debug/tracing/current_tracer
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
As well as with function tracing too. Pratyush Anand first reported
this issue to me and supplied a patch. When I tested this on my x86
test box, it caused thousands of backtraces and warnings to appear in
dmesg, which also caused a denial of service (a warning for every
function that was listed). I applied Pratyush's patch but it did not
fix the issue for me. I looked into it and found a slight problem
with trampoline accounting. I fixed it and sent Pratyush a patch, but
he said that it did not fix the issue for him.
I later learned tha Pratyush was using an ARM64 server, and when I
tested on my ARM board, I was able to reproduce the same issue as
Pratyush. After applying his patch, it fixed the problem. The above
test uncovered two different bugs, one in x86 and one in ARM and
ARM64. As this looked like it would affect PowerPC, I tested it on my
PPC64 box. It too broke, but neither the patch that fixed ARM or x86
fixed this box (the changes were all in generic code!). The above
test, uncovered two more bugs that affected PowerPC. Again, the
changes were only done to generic code. It's the way the arch code
expected things to be done that was different between the archs. Some
where more sensitive than others.
The rest of this series fixes the PPC bugs as well"
* tag 'trace-fixes-v4.0-rc2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rostedt/linux-trace:
ftrace: Fix ftrace enable ordering of sysctl ftrace_enabled
ftrace: Fix en(dis)able graph caller when en(dis)abling record via sysctl
ftrace: Clear REGS_EN and TRAMP_EN flags on disabling record via sysctl
seq_buf: Fix seq_buf_bprintf() truncation
seq_buf: Fix seq_buf_vprintf() truncation
Pull cgroup fixes from Tejun Heo:
"The cgroup iteration update two years ago and the recent cpuset
restructuring introduced regressions in subset of cpuset
configurations. Three patches to fix them.
All are marked for -stable"
* 'for-4.0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup:
cpuset: Fix cpuset sched_relax_domain_level
cpuset: fix a warning when clearing configured masks in old hierarchy
cpuset: initialize effective masks when clone_children is enabled
Pull workqueue fix from Tejun Heo:
"One fix patch for a subtle livelock condition which can happen on
PREEMPT_NONE kernels involving two racing cancel_work calls. Whoever
comes in the second has to wait for the previous one to finish. This
was implemented by making the later one block for the same condition
that the former would be (work item completion) and then loop and
retest; unfortunately, depending on the wake up order, the later one
could lock out the former one to finish by busy looping on the cpu.
This is fixed by implementing explicit wait mechanism. Work item
might not belong anywhere at this point and there's remote possibility
of thundering herd problem. I originally tried to use bit_waitqueue
but it didn't work for static work items on modules. It's currently
using single wait queue with filtering wake up function and exclusive
wakeup. If this ever becomes a problem, which is not very likely, we
can try to figure out a way to piggy back on bit_waitqueue"
* 'for-4.0-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/wq:
workqueue: fix hang involving racing cancel[_delayed]_work_sync()'s for PREEMPT_NONE
Some archs (specifically PowerPC), are sensitive with the ordering of
the enabling of the calls to function tracing and setting of the
function to use to be traced.
That is, update_ftrace_function() sets what function the ftrace_caller
trampoline should call. Some archs require this to be set before
calling ftrace_run_update_code().
Another bug was discovered, that ftrace_startup_sysctl() called
ftrace_run_update_code() directly. If the function the ftrace_caller
trampoline changes, then it will not be updated. Instead a call
to ftrace_startup_enable() should be called because it tests to see
if the callback changed since the code was disabled, and will
tell the arch to update appropriately. Most archs do not need this
notification, but PowerPC does.
The problem could be seen by the following commands:
# echo 0 > /proc/sys/kernel/ftrace_enabled
# echo function > /sys/kernel/debug/tracing/current_tracer
# echo 1 > /proc/sys/kernel/ftrace_enabled
# cat /sys/kernel/debug/tracing/trace
The trace will show that function tracing was not active.
Cc: stable@vger.kernel.org # 2.6.27+
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When ftrace is enabled globally through the proc interface, we must check if
ftrace_graph_active is set. If it is set, then we should also pass the
FTRACE_START_FUNC_RET command to ftrace_run_update_code(). Similarly, when
ftrace is disabled globally through the proc interface, we must check if
ftrace_graph_active is set. If it is set, then we should also pass the
FTRACE_STOP_FUNC_RET command to ftrace_run_update_code().
Consider the following situation.
# echo 0 > /proc/sys/kernel/ftrace_enabled
After this ftrace_enabled = 0.
# echo function_graph > /sys/kernel/debug/tracing/current_tracer
Since ftrace_enabled = 0, ftrace_enable_ftrace_graph_caller() is never
called.
# echo 1 > /proc/sys/kernel/ftrace_enabled
Now ftrace_enabled will be set to true, but still
ftrace_enable_ftrace_graph_caller() will not be called, which is not
desired.
Further if we execute the following after this:
# echo nop > /sys/kernel/debug/tracing/current_tracer
Now since ftrace_enabled is set it will call
ftrace_disable_ftrace_graph_caller(), which causes a kernel warning on
the ARM platform.
On the ARM platform, when ftrace_enable_ftrace_graph_caller() is called,
it checks whether the old instruction is a nop or not. If it's not a nop,
then it returns an error. If it is a nop then it replaces instruction at
that address with a branch to ftrace_graph_caller.
ftrace_disable_ftrace_graph_caller() behaves just the opposite. Therefore,
if generic ftrace code ever calls either ftrace_enable_ftrace_graph_caller()
or ftrace_disable_ftrace_graph_caller() consecutively two times in a row,
then it will return an error, which will cause the generic ftrace code to
raise a warning.
Note, x86 does not have an issue with this because the architecture
specific code for ftrace_enable_ftrace_graph_caller() and
ftrace_disable_ftrace_graph_caller() does not check the previous state,
and calling either of these functions twice in a row has no ill effect.
Link: http://lkml.kernel.org/r/e4fbe64cdac0dd0e86a3bf914b0f83c0b419f146.1425666454.git.panand@redhat.com
Cc: stable@vger.kernel.org # 2.6.31+
Signed-off-by: Pratyush Anand <panand@redhat.com>
[
removed extra if (ftrace_start_up) and defined ftrace_graph_active as 0
if CONFIG_FUNCTION_GRAPH_TRACER is not set.
]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
