Merge reason: merge almost-rc8 into perfcounters/core, which was -rc6
based - to pick up the latest upstream fixes.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When fork() fails we cannot use perf_counter_exit_task() since that
assumes to operate on current. Write a new helper that cleans up
unused/clean contexts.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Remove the local_irq_save() etc.. in routines that are smp function
calls, or have IRQs disabled by other means.
Then change the COMM, MMAP, and swcounter context iteration to
current->perf_counter_ctxp and RCU, since it really doesn't matter
which context they iterate, they're all folded.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Commit a63eaf34ae ("perf_counter: Dynamically allocate tasks'
perf_counter_context struct") broke COMM and MMAP notification for
cpu wide counters by dropping out early if there was no task context,
thereby also not iterating the cpu context.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This fixes a nasty crash and highlights a bug that we were
freeing failed-fork() counters incorrectly.
(the fix for that will come separately)
[ Impact: fix crashes/lockups with inherited counters ]
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Peter noticed that we are sometimes reading cpuctx->task_ctx with
interrupts enabled.
Noticed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Peter Zijlstra pointed out that under some circumstances, we can take
the mutex in a context or a counter and then swap that context or
counter to another task, potentially leading to lock order inversions
or the mutexes not protecting what they are supposed to protect.
This fixes the problem by making sure that we never take a mutex in a
context or counter which could get swapped to another task. Most of
the cases where we take a mutex is on a top-level counter or context,
i.e. a counter which has an fd associated with it or a context that
contains such a counter. This adds WARN_ON_ONCE statements to verify
that.
The two cases where we need to take the mutex on a context that is a
clone of another are in perf_counter_exit_task and
perf_counter_init_task. The perf_counter_exit_task case is solved by
uncloning the context before starting to remove the counters from it.
The perf_counter_init_task is a little trickier; we temporarily
disable context swapping for the parent (forking) task by setting its
ctx->parent_gen to the all-1s value after locking the context, if it
is a cloned context, and restore the ctx->parent_gen value at the end
if the context didn't get uncloned in the meantime.
This also moves the increment of the context generation count to be
within the same critical section, protected by the context mutex, that
adds the new counter to the context. That way, taking the mutex is
sufficient to ensure that both the counter list and the generation
count are stable.
[ Impact: fix hangs, races with inherited and PID counters ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <18975.31580.520676.619896@drongo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Needed in followon patch.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Commit 564c2b21 ("perf_counter: Optimize context switch between
identical inherited contexts") introduced a race where it is possible
that a counter being attached to a task could get attached to the
wrong task, if the task is one that has inherited its context from
another task via fork. This happens because the optimized context
switch could switch the context to another task after find_get_context
has read task->perf_counter_ctxp. In fact, it's possible that the
context could then get freed, if the other task then exits.
This fixes the problem by protecting both the context switch and the
critical code in find_get_context with spinlocks. The context switch
locks the cxt->lock of both the outgoing and incoming contexts before
swapping them. That means that once code such as find_get_context
has obtained the spinlock for the context associated with a task,
the context can't get swapped to another task. However, the context
may have been swapped in the interval between reading
task->perf_counter_ctxp and getting the lock, so it is necessary to
check and retry.
To make sure that none of the contexts being looked at in
find_get_context can get freed, this changes the context freeing code
to use RCU. Thus an rcu_read_lock() is sufficient to ensure that no
contexts can get freed. This part of the patch is lifted from a patch
posted by Peter Zijlstra.
This also adds a check to make sure that we can't add a counter to a
task that is exiting.
There is also a race between perf_counter_exit_task and
find_get_context; this solves the race by moving the get_ctx that
was in perf_counter_alloc into the locked region in find_get_context,
so that once find_get_context has got the context for a task, it
won't get freed even if the task calls perf_counter_exit_task. It
doesn't matter if new top-level (non-inherited) counters get attached
to the context after perf_counter_exit_task has detached the context
from the task. They will just stay there and never get scheduled in
until the counters' fds get closed, and then perf_release will remove
them from the context and eventually free the context.
With this, we are now doing the unclone in find_get_context rather
than when a counter was added to or removed from a context (actually,
we were missing the unclone_ctx() call when adding a counter to a
context). We don't need to unclone when removing a counter from a
context because we have no way to remove a counter from a cloned
context.
This also takes out the smp_wmb() in find_get_context, which Peter
Zijlstra pointed out was unnecessary because the cmpxchg implies a
full barrier anyway.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <18974.33033.667187.273886@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch adds __print_symbolic which is similar to __print_flags but
works for an enumeration type instead. That is, there is only a one to one
mapping between the values and the symbols. When a match is made, then
it is printed, otherwise the hex value is outputed.
[ Impact: add interface for showing symbol names in events ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Developers have been asking for the ability in the ftrace event tracer
to display names of bits in a flags variable.
Instead of printing out c2, it would be easier to read FOO|BAR|GOO,
assuming that FOO is bit 1, BAR is bit 6 and GOO is bit 7.
Some examples where this would be useful are the state flags in a context
switch, kmalloc flags, and even permision flags in accessing files.
[
v2 changes include:
Frederic Weisbecker's idea of using a mask instead of bits,
thus we can output GFP_KERNEL instead of GPF_WAIT|GFP_IO|GFP_FS.
Li Zefan's idea of allowing the caller of __print_flags to add their
own delimiter (or no delimiter) where we can get for file permissions
rwx instead of r|w|x.
]
[
v3 changes:
Christoph Hellwig's idea of using an array instead of va_args.
]
[ Impact: better displaying of flags in trace output ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
This shouldnt matter normally (and i have not seen any
misbehavior), because active counters always have a
proper ->oncpu value - but nevertheless initialize the
field properly to -1.
[ Impact: cleanup ]
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Always use ftrace_event_enable_disable() to enable/disable an event
so that we can factorize out the event toggling code.
[ Impact: factorize and cleanup event tracing code ]
Signed-off-by: Zhao Lei <zhaolei@cn.fujitsu.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <4A14FDFE.2080402@cn.fujitsu.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
I found that there is nothing to protect event_hash in
ftrace_find_event(). Rcu protects the event hashlist
but not the event itself while we use it after its extraction
through ftrace_find_event().
This lack of a proper locking in this spot opens a race
window between any event dereferencing and module removal.
Eg:
--Task A--
print_trace_line(trace) {
event = find_ftrace_event(trace)
--Task B--
trace_module_remove_events(mod) {
list_trace_events_module(ev, mod) {
unregister_ftrace_event(ev->event) {
hlist_del(ev->event->node)
list_del(....)
}
}
}
|--> module removed, the event has been dropped
--Task A--
event->print(trace); // Dereferencing freed memory
If the event retrieved belongs to a module and this module
is concurrently removed, we may end up dereferencing a data
from a freed module.
RCU could solve this, but it would add latency to the kernel and
forbid tracers output callbacks to call any sleepable code.
So this fix converts 'trace_event_mutex' to a read/write semaphore,
and adds trace_event_read_lock() to protect ftrace_find_event().
[ Impact: fix possible freed memory dereference in ftrace ]
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <4A114806.7090302@cn.fujitsu.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Introduce a generic per counter interrupt throttle.
This uses the perf_counter_overflow() quick disable to throttle a specific
counter when its going too fast when a pmu->unthrottle() method is provided
which can undo the quick disable.
Power needs to implement both the quick disable and the unthrottle method.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090525153931.703093461@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Ingo noticed that cpu counters had 0 context switches, even though
there was plenty scheduling on the cpu.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090525124600.419025548@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fail fork() when we fail inheritance for some reason (-ENOMEM most likely).
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090525124600.324656474@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Paul noted that the new ptcrl() didn't work on child counters.
Reported-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090525124600.203151469@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Initialize a task's perfcounters (inherit from parent, etc.) after
the child task's scheduler fields have been initialized already.
[ Impact: cleanup ]
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The problem occurs when async_synchronize_full_domain() is called when
the async_pending list is not empty. This will cause lowest_running()
to return the cookie of the first entry on the async_pending list, which
might be nothing at all to do with the domain being asked for and thus
cause the domain synchronization to wait for an unrelated domain. This
can cause a deadlock if domain synchronization is used from one domain
to wait for another.
Fix by running over the async_pending list to see if any pending items
actually belong to our domain (and return their cookies if they do).
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We shouldn't hold dpm_list_mtx while executing
[disable|enable]_nonboot_cpus(), because theoretically this may lead
to a deadlock as shown by the following example (provided by Johannes
Berg):
CPU 3 CPU 2 CPU 1
suspend/hibernate
something:
rtnl_lock() device_pm_lock()
-> mutex_lock(&dpm_list_mtx)
mutex_lock(&dpm_list_mtx)
linkwatch_work
-> rtnl_lock()
disable_nonboot_cpus()
-> flush CPU 3 workqueue
Fortunately, device drivers are supposed to stop any activities that
might lead to the registration of new device objects way before
disable_nonboot_cpus() is called, so it shouldn't be necessary to
hold dpm_list_mtx over the entire late part of device suspend and
early part of device resume.
Thus, during the late suspend and the early resume of devices acquire
dpm_list_mtx only when dpm_list is going to be traversed and release
it right after that.
This patch is reported to fix the regressions tracked as
http://bugzilla.kernel.org/show_bug.cgi?id=13245.
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Alan Stern <stern@rowland.harvard.edu>
Reported-by: Miles Lane <miles.lane@gmail.com>
Tested-by: Ming Lei <tom.leiming@gmail.com>
In a default 'perf top' run the tool will create a counter for
each online CPU. With enough CPUs this will eventually exhaust
the default limit.
So scale it up with the number of online CPUs.
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
now that pctrl() no longer disables other people's counters,
remove the PMU cache code that deals with that.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090523163013.032998331@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Instead of en/dis-abling all counters acting on a particular
task, en/dis- able all counters we created.
[ v2: fix crash on first counter enable ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090523163012.916937244@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use perf_counter_remove_from_context() to remove counters from
the context.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090523163012.796275849@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Ensure we're consistent with the context locks.
context->mutex
context->lock
list_{add,del}_counter();
so that either lock is sufficient to stabilize the context.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090523163012.618790733@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
s/counter->mutex/counter->child_mutex/ and make sure its only
used to protect child_list.
The usage in __perf_counter_exit_task() doesn't appear to be
problematic since ctx->mutex also covers anything related to fd
tear-down.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090523163012.533186528@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We call perf_adjust_freq() from perf_counter_task_tick() which
is is called under the rq->lock causing lock recursion.
However, it's no longer required to be called under the
rq->lock, so remove it from under it.
Also, fix up some related comments.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090523163012.476197912@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Presently non-legacy IRQs have their irq_desc allocated with
kzalloc_node(). This assumes that all callers of irq_to_desc_node_alloc()
will be sufficiently late in the boot process that kmalloc is available.
While porting sparseirq support to sh this blew up immediately, as at the
time that we register the CPU's interrupt vector map only bootmem is
available. Check slab_is_available() to work out which path to use.
[ Impact: fix SH early boot crash with sparseirq enabled ]
Signed-off-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
LKML-Reference: <20090522014008.GA2806@linux-sh.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When monitoring a process and its descendants with a set of inherited
counters, we can often get the situation in a context switch where
both the old (outgoing) and new (incoming) process have the same set
of counters, and their values are ultimately going to be added together.
In that situation it doesn't matter which set of counters are used to
count the activity for the new process, so there is really no need to
go through the process of reading the hardware counters and updating
the old task's counters and then setting up the PMU for the new task.
This optimizes the context switch in this situation. Instead of
scheduling out the perf_counter_context for the old task and
scheduling in the new context, we simply transfer the old context
to the new task and keep using it without interruption. The new
context gets transferred to the old task. This means that both
tasks still have a valid perf_counter_context, so no special case
is introduced when the old task gets scheduled in again, either on
this CPU or another CPU.
The equivalence of contexts is detected by keeping a pointer in
each cloned context pointing to the context it was cloned from.
To cope with the situation where a context is changed by adding
or removing counters after it has been cloned, we also keep a
generation number on each context which is incremented every time
a context is changed. When a context is cloned we take a copy
of the parent's generation number, and two cloned contexts are
equivalent only if they have the same parent and the same
generation number. In order that the parent context pointer
remains valid (and is not reused), we increment the parent
context's reference count for each context cloned from it.
Since we don't have individual fds for the counters in a cloned
context, the only thing that can make two clones of a given parent
different after they have been cloned is enabling or disabling all
counters with prctl. To account for this, we keep a count of the
number of enabled counters in each context. Two contexts must have
the same number of enabled counters to be considered equivalent.
Here are some measurements of the context switch time as measured with
the lat_ctx benchmark from lmbench, comparing the times obtained with
and without this patch series:
-----Unmodified----- With this patch series
Counters: none 2 HW 4H+4S none 2 HW 4H+4S
2 processes:
Average 3.44 6.45 11.24 3.12 3.39 3.60
St dev 0.04 0.04 0.13 0.05 0.17 0.19
8 processes:
Average 6.45 8.79 14.00 5.57 6.23 7.57
St dev 1.27 1.04 0.88 1.42 1.46 1.42
32 processes:
Average 5.56 8.43 13.78 5.28 5.55 7.15
St dev 0.41 0.47 0.53 0.54 0.57 0.81
The numbers are the mean and standard deviation of 20 runs of
lat_ctx. The "none" columns are lat_ctx run directly without any
counters. The "2 HW" columns are with lat_ctx run under perfstat,
counting cycles and instructions. The "4H+4S" columns are lat_ctx run
under perfstat with 4 hardware counters and 4 software counters
(cycles, instructions, cache references, cache misses, task
clock, context switch, cpu migrations, and page faults).
[ Impact: performance optimization of counter context-switches ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <18966.10666.517218.332164@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This replaces the struct perf_counter_context in the task_struct with
a pointer to a dynamically allocated perf_counter_context struct. The
main reason for doing is this is to allow us to transfer a
perf_counter_context from one task to another when we do lazy PMU
switching in a later patch.
This has a few side-benefits: the task_struct becomes a little smaller,
we save some memory because only tasks that have perf_counters attached
get a perf_counter_context allocated for them, and we can remove the
inclusion of <linux/perf_counter.h> in sched.h, meaning that we don't
end up recompiling nearly everything whenever perf_counter.h changes.
The perf_counter_context structures are reference-counted and freed
when the last reference is dropped. A context can have references
from its task and the counters on its task. Counters can outlive the
task so it is possible that a context will be freed well after its
task has exited.
Contexts are allocated on fork if the parent had a context, or
otherwise the first time that a per-task counter is created on a task.
In the latter case, we set the context pointer in the task struct
locklessly using an atomic compare-and-exchange operation in case we
raced with some other task in creating a context for the subject task.
This also removes the task pointer from the perf_counter struct. The
task pointer was not used anywhere and would make it harder to move a
context from one task to another. Anything that needed to know which
task a counter was attached to was already using counter->ctx->task.
The __perf_counter_init_context function moves up in perf_counter.c
so that it can be called from find_get_context, and now initializes
the refcount, but is otherwise unchanged.
We were potentially calling list_del_counter twice: once from
__perf_counter_exit_task when the task exits and once from
__perf_counter_remove_from_context when the counter's fd gets closed.
This adds a check in list_del_counter so it doesn't do anything if
the counter has already been removed from the lists.
Since perf_counter_task_sched_in doesn't do anything if the task doesn't
have a context, and leaves cpuctx->task_ctx = NULL, this adds code to
__perf_install_in_context to set cpuctx->task_ctx if necessary, i.e. in
the case where the current task adds the first counter to itself and
thus creates a context for itself.
This also adds similar code to __perf_counter_enable to handle a
similar situation which can arise when the counters have been disabled
using prctl; that also leaves cpuctx->task_ctx = NULL.
[ Impact: refactor counter context management to prepare for new feature ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <18966.10075.781053.231153@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
register_module_notifier() returns zero in the success case.
So fix the inverted fail case check in trace events modules
handler.
[ Impact: fix spurious warning on ftrace initialization]
Reported-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Avoid a function call for !group counters by directly calling the counter
function.
[ Impact: micro-optimize the code ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090520102553.511933670@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently we call hrtimer_cancel() unconditionally on disable of time based
software counters. Avoid when possible.
[ Impact: micro-optimize the code ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090520102553.388185031@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
For the dynamic irq_period code, log whenever we change the period so that
analyzing code can normalize the event flow.
[ Impact: add new feature to allow more precise profiling ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090520102553.298769743@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Instead of disabling RR scheduling of the counters, use a different list
that does not get rotated to iterate the counters on inheritance.
[ Impact: cleanup, optimization ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
LKML-Reference: <20090520102553.237504544@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
If the waiter has been requeued to the outer PI futex and is
interrupted by a signal and the thread handles the signal then
ERESTART_RESTARTBLOCK is changed to EINTR and the restart block is
discarded. That way we return an unexcpected EINTR to user space
instead of ending up in futex_lock_pi_restart.
But we do not need to restart the syscall because we know that the
condition has changed since we have been requeued. If we would simply
restart the syscall then we would drop out via the comparison of the
user space value with EWOULDBLOCK.
The user space side needs to handle EWOULDBLOCK anyway as the
enqueueing on the inner futex can race with a requeue/wake. So we can
simply return EWOULDBLOCK to user space which also signals that we did
not take the outer futex and let user space handle it in the same way
it has to handle the requeue/wake race.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The futex_wait_requeue_pi op should restart unconditionally like
futex_lock_pi. The user of that function e.g. pthread_cond_wait can
not be interrupted so we do not care about the SA_RESTART flag of the
signal. Clean up the FIXMEs.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Merge reason: this branch was on an pre -rc1 base, merge it up to -rc6+
to get the latest upstream fixes.
Conflicts:
kernel/futex.c
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Context rotation should not occur when we are in the middle of
walking the counter list when inheriting counters ...
[ Impact: fix occasionally incorrect perf stat results ]
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix counter lifetime bugs which explain the crashes reported by
Marcelo Tosatti and Arnaldo Carvalho de Melo.
The new rule is: flushing + freeing is only done for a task's
own counters, never for other tasks.
[ Impact: fix crashes/lockups with inherited counters ]
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Reported-by: Marcelo Tosatti <mtosatti@redhat.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The futex code installs a read only mapping via get_user_pages_fast()
even if the futex op function has to modify user space data. The
eventual fault was fixed up by futex_handle_fault() which walked the
VMA with mmap_sem held.
After the cleanup patches which removed the mmap_sem dependency of the
futex code commit 4dc5b7a36a49eff97050894cf1b3a9a02523717 (futex:
clean up fault logic) removed the private VMA walk logic from the
futex code. This change results in a stale RO mapping which is not
fixed up.
Instead of reintroducing the previous fault logic we set up the
mapping in get_user_pages_fast() read/write for all operations which
modify user space data. Also handle private futexes in the same way
and make the current unconditional access_ok(VERIFY_WRITE) depend on
the futex op.
Reported-by: Andreas Schwab <schwab@linux-m68k.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
CC: stable@kernel.org
debugfs directory entries for devices are not removed on some
of the failure pathes in do_blk_trace_setup().
One way to reproduce is to start blktrace on multiple devices
with insufficient Vmalloc space: Devices will fail with
a message like this:
BLKTRACESETUP(2) /dev/sdu failed: 5/Input/output error
If so, the respective entries in debugfs
(e.g. /sys/kernel/debug/block/sdu) will remain and subsequent
attempts to start blktrace on the respective devices will not
succeed due to existing directories.
[ Impact: fix /debug/tracing file cleanup corner case ]
Signed-off-by: Stefan Raspl <stefan.raspl@linux.vnet.ibm.com>
Acked-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: schwidefsky@de.ibm.com
Cc: heiko.carstens@de.ibm.com
LKML-Reference: <4A1266CC.5040801@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Properly document the variable-size structure tricks we are doing
wrt. struct sched_group and sched_domain, and use the field[0] GCC
extension instead of defining a vla array.
Dont use unions for this, as pointed out by Linus.
[ Impact: cleanup, un-confuse Sparse and LLVM ]
Reported-by: Jeff Garzik <jeff@garzik.org>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
LKML-Reference: <alpine.LFD.2.01.0905180850110.3301@localhost.localdomain>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'sched-fixes-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
sched: Fix fallback sched_clock()'s offset when using jiffies
* 'core-fixes-for-linus-2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
lockdep: increase MAX_LOCKDEP_ENTRIES and MAX_LOCKDEP_CHAINS
* 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
tracing: Append prompt in /debug/tracing/README file
x86/function-graph: fix constraint for recording old return value
return zero should be correct, so fix it.
[ Impact: eliminate incorrect syslog message ]
Signed-off-by: Ming Lei <tom.leiming@gmail.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: rostedt@goodmis.org
LKML-Reference: <1242545498-7285-1-git-send-email-tom.leiming@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Flushing counters in __exit_signal() with irqs disabled is not
a good idea as perf_counter_exit_task() acquires mutexes. So
flush it before acquiring the tasklist lock.
(Note, we still need a fix for when the PID has been unhashed.)
[ Impact: fix crash with inherited counters ]
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Clean up code that open-coded the list_{add,del}_counter() code in
__perf_counter_exit_task() which consequently diverged. This could
lead to software counter crashes.
Also, fold the ctx->nr_counter inc/dec into those functions and clean
up some of the related code.
[ Impact: fix potential sw counter crash, cleanup ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srivatsa Vaddagiri <vatsa@in.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Ian Campbell noticed that since "Eliminate thousands of warnings with
gcc 3.2 build" (commit 57adc4d2db) all
WARN_ON()'s currently appear to come from warn_slowpath_null(), eg:
WARNING: at kernel/softirq.c:143 warn_slowpath_null+0x1c/0x20()
because now that warn_slowpath_null() is in the call path, the
__builtin_return_address(0) returns that, rather than the place that
caused the warning.
Fix this by splitting up the warn_slowpath_null/fmt cases differently,
using a common helper function, and getting the return address in the
right place. This also happens to avoid the unnecessary stack usage for
the non-stdargs case, and just generally cleans things up.
Make the function name printout use %pS while at it.
Cc: Ian Campbell <ian.campbell@citrix.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Check the return value of sysdev_suspend(). I think this was a typo.
Without this change, the following "if" check is always false.
I also changed the error message so it's distinguishable from the
similar message a few lines above.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jwessel/linux-2.6-kgdb:
kgdb: gdb documentation fix
kgdb,i386: use address that SP register points to in the exception frame
sysrq, intel_fb: fix sysrq g collision
At present the values we put in overflow events for the misc
flags indicating processor mode and the instruction pointer are
obtained using the standard user_mode() and
instruction_pointer() functions. Those functions tell you where
the performance monitor interrupt was taken, which might not be
exactly where the counter overflow occurred, for example
because interrupts were disabled at the point where the
overflow occurred, or because the processor had many
instructions in flight and chose to complete some more
instructions beyond the one that caused the counter overflow.
Some architectures (e.g. powerpc) can supply more precise
information about where the counter overflow occurred and the
processor mode at that point. This introduces new functions,
perf_misc_flags() and perf_instruction_pointer(), which arch
code can override to provide more precise information if
available. They have default implementations which are
identical to the existing code.
This also adds a new misc flag value,
PERF_EVENT_MISC_HYPERVISOR, for the case where a counter
overflow occurred in the hypervisor. We encode the processor
mode in the 2 bits previously used to indicate user or kernel
mode; the values for user and kernel mode are unchanged and
hypervisor mode is indicated by both bits being set.
[ Impact: generalize perfcounter core facilities ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <18956.1272.818511.561835@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
avenrun is an rough estimate so we don't have to worry about
consistency of the three avenrun values. Remove the xtime lock
dependency and provide a function to scale the values. Cleanup the
users.
[ Impact: cleanup ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Dimitri Sivanich noticed that xtime_lock is held write locked across
calc_load() which iterates over all online CPUs. That can cause long
latencies for xtime_lock readers on large SMP systems.
The load average calculation is an rough estimate anyway so there is
no real need to protect the readers vs. the update. It's not a problem
when the avenrun array is updated while a reader copies the values.
Instead of iterating over all online CPUs let the scheduler_tick code
update the number of active tasks shortly before the avenrun update
happens. The avenrun update itself is handled by the CPU which calls
do_timer().
[ Impact: reduce xtime_lock write locked section ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Instead of specifying the irq_period for a counter, provide a target interrupt
frequency and dynamically adapt the irq_period to match this frequency.
[ Impact: new perf-counter attribute/feature ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <20090515132018.646195868@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Instead of a per-process mlock gift for perf-counters, use a
per-user gift so that there is less of a DoS potential.
[ Impact: allow less worst-case unprivileged memory consumption ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <20090515132018.496182835@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Now that ACPI idle doesn't use it anymore, remove the exports.
[ Impact: remove dead code/data ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <20090515132018.429826617@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The waitqueue which is used in struct futex_q is a leftover from the
futexfd implementation. There is no need to use a waitqueue at all, as
the waiting task is the only user of it. The waitqueue just adds
additional locking and a loop in the wake up path which both can be
avoided.
We have already a task reference in struct futex_q which is used for
PI futexes. Use it for normal futexes as well and just wake up the
task directly.
The logic of signalling the futex wakeup via setting q->lock_ptr to
NULL is kept with the difference that we set it NULL before doing the
wakeup. This opens an exit race window vs. a non futex wake up of the
to be woken up task, which we prevent with get_task_struct /
put_task_struct on the waiter.
[ Impact: simplification ]
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Commit 79e539453b introduced a
regression where you cannot use sysrq 'g' to enter kgdb. The solution
is to move the intel fb sysrq over to V for video instead of G for
graphics. The SMP VOYAGER code to register for the sysrq-v is not
anywhere to be found in the mainline kernel, so the comments in the
code were cleaned up as well.
This patch also cleans up the sysrq definitions for kgdb to make it
generic for the kernel debugger, such that the sysrq 'g' can be used
in the future to enter a gdbstub or another kernel debugger.
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This reverts commit fafd688e4c.
Work is progressing to switch away from pdflush as the process backing
for flushing out dirty data. So it seems pointless to add more knobs
to control pdflush threads. The original author of the patch did not
have any specific use cases for adding the knobs, so we can easily
revert this before 2.6.30 to avoid having to maintain this API
forever.
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
The current disable/enable mechanism is:
token = hw_perf_save_disable();
...
/* do bits */
...
hw_perf_restore(token);
This works well, provided that the use nests properly. Except we don't.
x86 NMI/INT throttling has non-nested use of this, breaking things. Therefore
provide a reference counter disable/enable interface, where the first disable
disables the hardware, and the last enable enables the hardware again.
[ Impact: refactor, simplify the PMU disable/enable logic ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The simple reservation test in perf_output_copy() failed to take
unsigned int overflow into account, fix this.
[ Impact: fix false positive warning with more than 4GB of profiling data ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
We should leave the last slot for the ending '\0'.
[ Impact: fix possible crash when the length of an operand is 128 ]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <4A0CDC8C.30602@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
[ Impact: fix deadlock in a rare case we fail to allocate memory ]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <4A0CDC6F.7070200@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The stack tracer stores eight entries in the ring buffer when an event
traces the stack. The output outputs all eight entries regardless of
how many entries were recorded.
This patch breaks out of the loop when a null entry is discovered.
[ Impact: only print the stack that is recorded ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
I noticed that when enabling a group via the PERF_COUNTER_IOC_ENABLE
ioctl on the group leader, the counters weren't enabled and counting
immediately on return from the ioctl, but did start counting a little
while later (presumably after a context switch).
The reason was that __perf_counter_enable calls group_sched_in which
calls hw_perf_group_sched_in, which on powerpc assumes that the caller
has called hw_perf_save_disable already. Until commit 46d686c6
("perf_counter: put whole group on when enabling group leader") it was
true that all callers of group_sched_in had called
hw_perf_save_disable first, and the powerpc hw_perf_group_sched_in
relies on that (there isn't an x86 version).
This fixes the problem by putting calls to hw_perf_save_disable /
hw_perf_restore around the calls to group_sched_in and
counter_sched_in in __perf_counter_enable. Having the calls to
hw_perf_save_disable/restore around the counter_sched_in call is
harmless and makes this call consistent with the other call sites
of counter_sched_in, which have all called hw_perf_save_disable first.
[ Impact: more precise counter group disable/enable functionality ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <18953.25733.53359.147452@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Merge reason: both topics modify the APIC code but were able to do it in
parallel so far. An upcoming patch generates a conflict so
merge them to avoid the conflict.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This is a bit of micro-optimizations. But since the ring buffer is used
in tracing every function call, it is an extreme hot path. Every nanosecond
counts.
This change shows over 5% improvement in the ring-buffer-benchmark.
[ Impact: more efficient code ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The ring_buffer_time_stamp that is exported adds a little more overhead
than is needed for using it internally. This patch adds an internal
timestamp function that can be inlined (a single line function)
and used internally for the ring buffer.
[ Impact: a little less overhead to the ring buffer ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Doing some small changes in the fast path of the ring buffer recording
saves over 3% in the ring-buffer-benchmark test.
[ Impact: a little faster ring buffer recording ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
A long ago, in days of yore, it all began with a god named Thor.
There were vikings and boats and some plans for a Linux kernel
header. Unfortunately, a single 8-bit field was used for bootloader
type and version. This has generally worked without *too* much pain,
but we're getting close to flat running out of ID fields.
Add extension fields for both type and version. The type will be
extended if it the old field is 0xE; the version is a simple MSB
extension.
Keep /proc/sys/kernel/bootloader_type containing
(type << 4) + (ver & 0xf) for backwards compatiblity, but also add
/proc/sys/kernel/bootloader_version which contains the full version
number.
[ Impact: new feature to support more bootloaders ]
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
The event length is calculated and passed in to rb_reserve_next_event
in two different locations. Having rb_reserve_next_event do the
calculations directly makes only one location to do the change and
causes the calculation to be inlined by gcc.
Before:
text data bss dec hex filename
16538 24 12 16574 40be kernel/trace/ring_buffer.o
After:
text data bss dec hex filename
16490 24 12 16526 408e kernel/trace/ring_buffer.o
[ Impact: smaller more efficient code ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The rb_reserve_next_event is only called for the data type (type = 0).
There is no reason to pass in the type to the function.
Before:
text data bss dec hex filename
16554 24 12 16590 40ce kernel/trace/ring_buffer.o
After:
text data bss dec hex filename
16538 24 12 16574 40be kernel/trace/ring_buffer.o
[ Impact: cleaner, smaller and slightly more efficient code ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Although we check if "missed" is not zero, we divide by hit + missed,
and the addition can possible overflow and become a divide by zero.
This patch checks for this case, and will report it when it happens
then modify "hit" to make the calculation be non zero.
[ Impact: prevent possible divide by zero in ring-buffer-benchmark ]
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The use of numeric constants is discouraged. It is cleaner and more
descriptive to use macros for constant time conversions.
This patch also removes an extra new line.
[ Impact: more descriptive time conversions ]
Reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
A compile warning triggered because we are calling
atomic_set(&counter->count). But since counter->count
is an atomic64_t, we have to use atomic64_set.
So the count can be set short, resulting in the reset ioctl
only resetting the low word.
[ Impact: clear counter properly during the reset ioctl ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <18951.48285.270311.981806@drongo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The context-switch software counter gives inflated values at present
because each scheduler tick and each process-wide counter
enable/disable prctl gets counted as a context switch.
This happens because perf_counter_task_tick, perf_counter_task_disable
and perf_counter_task_enable all call perf_counter_task_sched_out,
which calls perf_swcounter_event to record a context switch event.
This fixes it by introducing a variant of perf_counter_task_sched_out
with two underscores in front for internal use within the perf_counter
code, and makes perf_counter_task_{tick,disable,enable} call it. This
variant doesn't record a context switch event, and takes a struct
perf_counter_context *. This adds the new variant rather than
changing the behaviour or interface of perf_counter_task_sched_out
because that is called from other code.
[ Impact: fix inflated context-switch event counts ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <18951.48034.485580.498953@drongo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently, if you have a group where the leader is disabled and there
are siblings that are enabled, and then you enable the leader, we only
put the leader on the PMU, and not its enabled siblings. This is
incorrect, since the enabled group members should be all on or all off
at any given point.
This fixes it by adding a call to group_sched_in in
__perf_counter_enable in the case where we're enabling a group leader.
To avoid the need for a forward declaration this also moves
group_sched_in up before __perf_counter_enable. The actual content of
group_sched_in is unchanged by this patch.
[ Impact: fix bug in counter enable code ]
Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <18951.34946.451546.691693@drongo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
struct request has had a few different ways to represent some
properties of a request. ->hard_* represent block layer's view of the
request progress (completion cursor) and the ones without the prefix
are supposed to represent the issue cursor and allowed to be updated
as necessary by the low level drivers. The thing is that as block
layer supports partial completion, the two cursors really aren't
necessary and only cause confusion. In addition, manual management of
request detail from low level drivers is cumbersome and error-prone at
the very least.
Another interesting duplicate fields are rq->[hard_]nr_sectors and
rq->{hard_cur|current}_nr_sectors against rq->data_len and
rq->bio->bi_size. This is more convoluted than the hard_ case.
rq->[hard_]nr_sectors are initialized for requests with bio but
blk_rq_bytes() uses it only for !pc requests. rq->data_len is
initialized for all request but blk_rq_bytes() uses it only for pc
requests. This causes good amount of confusion throughout block layer
and its drivers and determining the request length has been a bit of
black magic which may or may not work depending on circumstances and
what the specific LLD is actually doing.
rq->{hard_cur|current}_nr_sectors represent the number of sectors in
the contiguous data area at the front. This is mainly used by drivers
which transfers data by walking request segment-by-segment. This
value always equals rq->bio->bi_size >> 9. However, data length for
pc requests may not be multiple of 512 bytes and using this field
becomes a bit confusing.
In general, having multiple fields to represent the same property
leads only to confusion and subtle bugs. With recent block low level
driver cleanups, no driver is accessing or manipulating these
duplicate fields directly. Drop all the duplicates. Now rq->sector
means the current sector, rq->data_len the current total length and
rq->bio->bi_size the current segment length. Everything else is
defined in terms of these three and available only through accessors.
* blk_recalc_rq_sectors() is collapsed into blk_update_request() and
now handles pc and fs requests equally other than rq->sector update.
This means that now pc requests can use partial completion too (no
in-kernel user yet tho).
* bio_cur_sectors() is replaced with bio_cur_bytes() as block layer
now uses byte count as the primary data length.
* blk_rq_pos() is now guranteed to be always correct. In-block users
converted.
* blk_rq_bytes() is now guaranteed to be always valid as is
blk_rq_sectors(). In-block users converted.
* blk_rq_sectors() is now guaranteed to equal blk_rq_bytes() >> 9.
More convenient one is used.
* blk_rq_bytes() and blk_rq_cur_bytes() are now inlined and take const
pointer to request.
[ Impact: API cleanup, single way to represent one property of a request ]
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Boaz Harrosh <bharrosh@panasas.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Implement accessors - blk_rq_pos(), blk_rq_sectors() and
blk_rq_cur_sectors() which return rq->hard_sector, rq->hard_nr_sectors
and rq->hard_cur_sectors respectively and convert direct references of
the said fields to the accessors.
This is in preparation of request data length handling cleanup.
Geert : suggested adding const to struct request * parameter to accessors
Sergei : spotted error in patch description
[ Impact: cleanup ]
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Geert Uytterhoeven <Geert.Uytterhoeven@sonycom.com>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Tested-by: Grant Likely <grant.likely@secretlab.ca>
Acked-by: Grant Likely <grant.likely@secretlab.ca>
Ackec-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Cc: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Borislav Petkov <petkovbb@googlemail.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
Rename cred_exec_mutex to reflect that it's a guard against foreign
intervention on a process's credential state, such as is made by ptrace(). The
attachment of a debugger to a process affects execve()'s calculation of the new
credential state - _and_ also setprocattr()'s calculation of that state.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
Account for the initial offset to the jiffy count.
[ Impact: fix printk timestamps on architectures using fallback sched_clock() ]
Signed-off-by: Ron Lee <ron@debian.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix kprobes to lock text_mutex around some arch_arm/disarm_kprobe() which
are newly added by commit de5bd88d5a.
Signed-off-by: Masami Hiramatsu <mhiramat@redhat.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Cc: Jim Keniston <jkenisto@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Other parts of the kernel may need to be able to enable or disable
specific events. Especially parts that create trace events.
[ Impact: allow enabling of trace events by those that create the event ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Commit 8f31bfe538
tracing/events: clean up for ftrace_set_clr_event()
Moved out the code for ftrace_set_clr_event into a helper funciton but
did not initialize the return value. As a result, we do not warn about
a typo in the echoing of events in set_event.
This patch restores the old warning:
# echo foobar > set_event
-bash: echo: write error: Invalid argument
[ Impact: restore warning of invalid entries to set_event ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Allow recording the CPU number the event was generated on.
RFC: this leaves a u32 as reserved, should we fill in the
node_id() there, or leave this open for future extention,
as userspace can already easily do the cpu->node mapping
if needed.
[ Impact: extend perfcounter output record format ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090508170029.008627711@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Much like CONFIG_RECORD_GROUP records the hw_event.config to
identify the values, allow to record this for all counters.
[ Impact: extend perfcounter output record format ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090508170028.923228280@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Corey noticed that ioctl()s on grouped counters didn't work on
the whole group. This extends the ioctl() interface to take a
second argument that is interpreted as a flags field. We then
provide PERF_IOC_FLAG_GROUP to toggle the behaviour.
Having this flag gives the greatest flexibility, allowing you
to individually enable/disable/reset counters in a group, or
all together.
[ Impact: fix group counter enable/disable semantics ]
Reported-by: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <20090508170028.837558214@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
perf_counter_task_tick() does way too much work to find out
there's nothing to do. Provide an easy short-circuit for the
normal case where there are no counters on the system.
[ Impact: micro-optimization ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090508170028.750619201@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
A smarter way to figure out the output of an enable file.
[ Impact: clean up ]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A0399A5.2080603@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add a helper function __ftrace_set_clr_event(), and replace some
ftrace_set_clr_event() calls with this helper, thus we don't need any
kstrdup() or kmalloc().
As a side effect, this patch fixes an issue in self tests code, which is
similar to the one fixed in commit d6bf81ef0f
("tracing: append ":*" to internal setting of system events")
It's a small issue and won't cause any bug in fact, but we should do things
right anyway.
[ Impact: prevent spurious event-enabling in tracing self-tests ]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <4A03998E.3020503@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Conflicts:
arch/frv/include/asm/pgtable.h
arch/x86/include/asm/required-features.h
arch/x86/xen/mmu.c
Merge reason: x86/xen was on a .29 base still, move it to a fresher
branch and pick up Xen fixes as well, plus resolve
conflicts
Signed-off-by: Ingo Molnar <mingo@elte.hu>
There's a WARN_ON in the ring buffer code that makes sure preemption
is disabled. It checks "!preempt_count()". But when CONFIG_PREEMPT is not
enabled, preempt_count() is always zero, and this will trigger the warning.
[ Impact: prevent false warning on non preemptible kernels ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
It is nice to see the overhead of the benchmark test when tracing is
disabled. That is, we turn off the ring buffer just to see what the
cost of running the loop that calls into the ring buffer is.
Currently, if no entries wer made, we get 0. This is not informative.
This patch changes it to check if we had any "missed" (non recorded)
events. If so, a total count is also reported.
[ Impact: evaluate the over head of the ring buffer benchmark test ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Calling cond_resched at every iteration of the loop adds a bit of
overhead to the benchmark.
This patch does two things.
1) only calls cond-resched when CONFIG_PREEMPT is not enabled
2) only calls cond-resched after so many traces has been performed.
[ Impact: less overhead to the ring-buffer-benchmark ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Tracing can be very helpful to debug the kernel. When DEBUG_KERNEL is
enabled it is nice to enable the trace menu as well.
This patch only make the tracing menu enabled by default, it does not
make any of the tracers enabled. And the menu is only enabled by
default if DEBUG_KERNEL is enabled.
[ Impact: show tracing options to those debugging the kernel ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The system enabling of events uses the same code as the set_event file.
It passes in the name of the system to the parser and that will enable
all the events that has that system as a name.
The problem is that it will also enable events with the same name as the
system.
If you have system name foo, and system name bar, but within the system
bar, there exists an event called foo. By setting the system name foo,
you will also be enabling the event foo in the system bar. This is not
an expected result.
The solution is to pass in "foo:*", which will only enable the system
foo and not events called foo.
[ Impact: prevent accidental enabling of events with same name as a system ]
Reported-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Ingo Molnar thought that the code to calculate the time in cond_resched
is a bit too ugly and is not needed. This patch removes it and replaces
it with a simple call to cond_resched. I kept the comment that explains
the reason for the cond_resched.
[ Impact: remove ugly code ]
Reported-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Merge reason: this topic is ready for upstream now. It passed
Oleg's review and Andrew had no further mm/*
objections/observations either.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Merge reason: tracing/core was on a .30-rc1 base and was missing out on
on a handful of tracing fixes present in .30-rc5-almost.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
In filter_add_subsystem_pred() we should release event_mutex before
calling filter_free_subsystem_preds(), since both functions hold
event_mutex.
[ Impact: fix deadlock when writing invalid pred into subsystem filter ]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: tzanussi@gmail.com
Cc: a.p.zijlstra@chello.nl
Cc: fweisbec@gmail.com
Cc: rostedt@goodmis.org
LKML-Reference: <4A028993.7020509@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When we set a filter for an event, such as:
echo "name == my_lock_name" > \
/debug/tracing/events/lockdep/lock_acquired/filter
then the following order of token type is parsed:
- space
- operator
- parentheses
- operand
Because the operators and parentheses have a higher precedence
than the operand characters, which is normal, then we can't
use any string containing such special characters:
()=<>!&|
To get this support and also avoid ambiguous intepretation from
the parser or the human, we can do it using double quotes so that
we keep the usual languages habits.
Then after this patch you can still declare string condition like
before:
echo name == myname
But if you want to compare against a string containing an operator
character, you can use double quotes:
echo 'name == "&myname"'
Don't forget to include the whole expression into single quotes or
the double ones will be eaten by echo.
[ Impact: support strings with special characters for tracing filters ]
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Zhaolei <zhaolei@cn.fujitsu.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Currently the filtering infrastructure supports well the
numeric types and fixed sized array types.
But the recently added __string() field uses a specific
indirect offset mechanism which requires a specific
predicate. Until now it wasn't supported.
This patch adds this support and implies very few changes,
only a new predicate is needed, the management of this specific
field can be done through the usual string helpers in the
filtering infrastructure.
[ Impact: support all kinds of strings in the tracing filters ]
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Zhaolei <zhaolei@cn.fujitsu.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
When a thread is oom killed and fails to exit, it's helpful to know which
threads have access to memory reserves if the machine livelocks. This is
done by testing for the TIF_MEMDIE thread info flag and should be
displayed alongside stack traces to identify tasks that have access to
such reserves but are still stuck allocating pages, for instance.
It would probably be helpful in other cases as well, so all thread info
flags are emitted when showing a task.
( v2: fix warning reported by Stephen Rothwell )
[ Impact: extend debug printout info ]
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
LKML-Reference: <alpine.DEB.2.00.0905040136390.15831@chino.kir.corp.google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
With the current event directory, you can only enable individual events.
The file debugfs/tracing/set_event is used to be able to enable or
disable several events at once. But that can still be awkward.
This patch adds hierarchical enabling of events. That is, each directory
in debugfs/tracing/events has an "enable" file. This file can enable
or disable all events within the directory and below.
# echo 1 > /debugfs/tracing/events/enable
will enable all events.
# echo 1 > /debugfs/tracing/events/sched/enable
will enable all events in the sched subsystem.
# echo 1 > /debugfs/tracing/events/enable
# echo 0 > /debugfs/tracing/events/irq/enable
will enable all events, but then disable just the irq subsystem events.
When reading one of these enable files, there are four results:
0 - all events this file affects are disabled
1 - all events this file affects are enabled
X - there is a mixture of events enabled and disabled
? - this file does not affect any event
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Li Zefan found that there's a race using the event ids of events and
modules. When a module is loaded, an event id is incremented. We only
have 16 bits for event ids (65536) and there is a possible (but highly
unlikely) race that we could load and unload a module that registers
events so many times that the event id counter overflows.
When it overflows, it then restarts and goes looking for available
ids. An id is available if it was added by a module and released.
The race is if you have one module add an id, and then is removed.
Another module loaded can use that same event id. But if the old module
still had events in the ring buffer, the new module's call back would
get bogus data. At best (and most likely) the output would just be
garbage. But if the module for some reason used pointers (not recommended)
then this could potentially crash.
The safest thing to do is just reset the ring buffer if a module that
registered events is removed.
[ Impact: prevent unpredictable results of event id overflows ]
Reported-by: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <49FEAFD0.30106@cn.fujitsu.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When building with gcc 3.2 I get thousands of warnings such as
include/linux/gfp.h: In function `allocflags_to_migratetype':
include/linux/gfp.h:105: warning: null format string
due to passing a NULL format string to warn_slowpath() in
#define __WARN() warn_slowpath(__FILE__, __LINE__, NULL)
Split this case out into a separate call. This also shrinks the kernel
slightly:
text data bss dec hex filename
4802274 707668 712704 6222646 5ef336 vmlinux
text data bss dec hex filename
4799027 703572 712704 6215303 5ed687 vmlinux
due to removeing one argument from the commonly-called __WARN().
[akpm@linux-foundation.org: reduce scope of `empty']
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There is what we believe to be a false positive reported by lockdep.
inotify_inode_queue_event() => take inotify_mutex => kernel_event() =>
kmalloc() => SLOB => alloc_pages_node() => page reclaim => slab reclaim =>
dcache reclaim => inotify_inode_is_dead => take inotify_mutex => deadlock
The plan is to fix this via lockdep annotation, but that is proving to be
quite involved.
The patch flips the allocation over to GFP_NFS to shut the warning up, for
the 2.6.30 release.
Hopefully we will fix this for real in 2.6.31. I'll queue a patch in -mm
to switch it back to GFP_KERNEL so we don't forget.
=================================
[ INFO: inconsistent lock state ]
2.6.30-rc2-next-20090417 #203
---------------------------------
inconsistent {RECLAIM_FS-ON-W} -> {IN-RECLAIM_FS-W} usage.
kswapd0/380 [HC0[0]:SC0[0]:HE1:SE1] takes:
(&inode->inotify_mutex){+.+.?.}, at: [<ffffffff8112f1b5>] inotify_inode_is_dead+0x35/0xb0
{RECLAIM_FS-ON-W} state was registered at:
[<ffffffff81079188>] mark_held_locks+0x68/0x90
[<ffffffff810792a5>] lockdep_trace_alloc+0xf5/0x100
[<ffffffff810f5261>] __kmalloc_node+0x31/0x1e0
[<ffffffff81130652>] kernel_event+0xe2/0x190
[<ffffffff81130826>] inotify_dev_queue_event+0x126/0x230
[<ffffffff8112f096>] inotify_inode_queue_event+0xc6/0x110
[<ffffffff8110444d>] vfs_create+0xcd/0x140
[<ffffffff8110825d>] do_filp_open+0x88d/0xa20
[<ffffffff810f6b68>] do_sys_open+0x98/0x140
[<ffffffff810f6c50>] sys_open+0x20/0x30
[<ffffffff8100c272>] system_call_fastpath+0x16/0x1b
[<ffffffffffffffff>] 0xffffffffffffffff
irq event stamp: 690455
hardirqs last enabled at (690455): [<ffffffff81564fe4>] _spin_unlock_irqrestore+0x44/0x80
hardirqs last disabled at (690454): [<ffffffff81565372>] _spin_lock_irqsave+0x32/0xa0
softirqs last enabled at (690178): [<ffffffff81052282>] __do_softirq+0x202/0x220
softirqs last disabled at (690157): [<ffffffff8100d50c>] call_softirq+0x1c/0x50
other info that might help us debug this:
2 locks held by kswapd0/380:
#0: (shrinker_rwsem){++++..}, at: [<ffffffff810d0bd7>] shrink_slab+0x37/0x180
#1: (&type->s_umount_key#17){++++..}, at: [<ffffffff8110cfbf>] shrink_dcache_memory+0x11f/0x1e0
stack backtrace:
Pid: 380, comm: kswapd0 Not tainted 2.6.30-rc2-next-20090417 #203
Call Trace:
[<ffffffff810789ef>] print_usage_bug+0x19f/0x200
[<ffffffff81018bff>] ? save_stack_trace+0x2f/0x50
[<ffffffff81078f0b>] mark_lock+0x4bb/0x6d0
[<ffffffff810799e0>] ? check_usage_forwards+0x0/0xc0
[<ffffffff8107b142>] __lock_acquire+0xc62/0x1ae0
[<ffffffff810f478c>] ? slob_free+0x10c/0x370
[<ffffffff8107c0a1>] lock_acquire+0xe1/0x120
[<ffffffff8112f1b5>] ? inotify_inode_is_dead+0x35/0xb0
[<ffffffff81562d43>] mutex_lock_nested+0x63/0x420
[<ffffffff8112f1b5>] ? inotify_inode_is_dead+0x35/0xb0
[<ffffffff8112f1b5>] ? inotify_inode_is_dead+0x35/0xb0
[<ffffffff81012fe9>] ? sched_clock+0x9/0x10
[<ffffffff81077165>] ? lock_release_holdtime+0x35/0x1c0
[<ffffffff8112f1b5>] inotify_inode_is_dead+0x35/0xb0
[<ffffffff8110c9dc>] dentry_iput+0xbc/0xe0
[<ffffffff8110cb23>] d_kill+0x33/0x60
[<ffffffff8110ce23>] __shrink_dcache_sb+0x2d3/0x350
[<ffffffff8110cffa>] shrink_dcache_memory+0x15a/0x1e0
[<ffffffff810d0cc5>] shrink_slab+0x125/0x180
[<ffffffff810d1540>] kswapd+0x560/0x7a0
[<ffffffff810ce160>] ? isolate_pages_global+0x0/0x2c0
[<ffffffff81065a30>] ? autoremove_wake_function+0x0/0x40
[<ffffffff8107953d>] ? trace_hardirqs_on+0xd/0x10
[<ffffffff810d0fe0>] ? kswapd+0x0/0x7a0
[<ffffffff8106555b>] kthread+0x5b/0xa0
[<ffffffff8100d40a>] child_rip+0xa/0x20
[<ffffffff8100cdd0>] ? restore_args+0x0/0x30
[<ffffffff81065500>] ? kthread+0x0/0xa0
[<ffffffff8100d400>] ? child_rip+0x0/0x20
[eparis@redhat.com: fix audit too]
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Eric Paris <eparis@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The ring buffer benchmark/test runs a producer for 10 seconds.
This is done with preemption and interrupts enabled. But if the kernel
is not compiled with CONFIG_PREEMPT, it basically stops everything
but interrupts for 10 seconds.
Although this is just a test and is not for production, this attribute
can be quite annoying. It can also spawn badness elsewhere.
This patch solves the issues by calling "cond_resched" when the system
is not compiled with CONFIG_PREEMPT. It also keeps track of the time
spent to call cond_resched such that it does not go against the
time calculations. That is, if the task schedules away, the time scheduled
out is removed from the test data. Note, this only works for non PREEMPT
because we do not know when the task is scheduled out if we have PREEMPT
enabled.
[ Impact: prevent test from stopping the world for 10 seconds ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Ingo Molnar thought the code would be cleaner if we used a function call
instead of a goto for moving the tail page. After implementing this,
it seems that gcc still inlines the result and the output is pretty much
the same. Since this is considered a cleaner approach, might as well
implement it.
[ Impact: code clean up ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The result of the allocation of the ring buffer read page in the
ring buffer bench mark does not check the return to see if a page
was actually allocated. This patch fixes that.
[ Impact: avoid NULL dereference ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The code in __rb_reserve_next checks on page overflow if it is the
original commiter and then resets the page back to the original
setting. Although this is fine, and the code is correct, it is
a bit fragil. Some experimental work I did breaks it easily.
The better and more robust solution is to have all commiters that
overflow the page, simply subtract what they added.
[ Impact: more robust ring buffer account management ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This compiler warning:
CC kernel/trace/trace_output.o
kernel/trace/trace_output.c: In function ‘register_ftrace_event’:
kernel/trace/trace_output.c:544: warning: ‘list’ may be used uninitialized in this function
Is wrong as 'list' is always initialized - but GCC (4.3.2) does not
recognize this relationship properly.
Work around the warning by initializing the variable to NULL.
[ Impact: fix false positive compiler warning ]
Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This attempts to clarify names utilized during block I/O remap
operations (partition, volume manager). It correctly matches up the
/from/ information for both device & sector. This takes in the concept
from Kosaki Motohiro and extends it to include better naming for the
"device_from" field.
[ Impact: cleanup ]
Signed-off-by: Alan D. Brunelle <alan.brunelle@hp.com>
Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
LKML-Reference: <49FF4FAE.3000301@hp.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The orig_cpu parameter in trace_sched_migrate_task() is not necessary,
it can be got by using task_cpu(p) in the probe.
[ Impact: micro-optimization ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
[ modified from Mathieu's patch. The original patch is at:
http://marc.info/?l=linux-kernel&m=123791201716239&w=2 ]
Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
Cc: fweisbec@gmail.com
Cc: rostedt@goodmis.org
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: zhaolei@cn.fujitsu.com
Cc: laijs@cn.fujitsu.com
LKML-Reference: <49FFFDB7.1050402@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
A module will add/remove its trace events when it gets loaded/unloaded, so
the ftrace_events list is not "const", and concurrent access needs to be
protected.
This patch thus fixes races between loading/unloding modules and read
'available_events' or read/write 'set_event', etc.
Below shows how to reproduce the race:
# for ((; ;)) { cat /mnt/tracing/available_events; } > /dev/null &
# for ((; ;)) { insmod trace-events-sample.ko; rmmod sample; } &
After a while:
BUG: unable to handle kernel paging request at 0010011c
IP: [<c1080f27>] t_next+0x1b/0x2d
...
Call Trace:
[<c10c90e6>] ? seq_read+0x217/0x30d
[<c10c8ecf>] ? seq_read+0x0/0x30d
[<c10b4c19>] ? vfs_read+0x8f/0x136
[<c10b4fc3>] ? sys_read+0x40/0x65
[<c1002a68>] ? sysenter_do_call+0x12/0x36
[ Impact: fix races when concurrent accessing ftrace_events list ]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Tom Zanussi <tzanussi@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <4A00F709.3080800@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When unloading a module, memory allocated by init_preds() and
trace_define_field() is not freed.
[ Impact: fix memory leak ]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Frederic Weisbecker <fweisbec@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: Tom Zanussi <tzanussi@gmail.com>
LKML-Reference: <4A00F6E0.3040503@cn.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Merge reason: we moved a mutex.h commit that originated from the
perfcounters tree into core/locking - but now merge
back that branch to solve a merge artifact and to
pick up cleanups of this commit that happened in
core/locking.
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch adds code that can benchmark the ring buffer as well as
test it. This code can be compiled into the kernel (not recommended)
or as a module.
A separate ring buffer is used to not interfer with other users, like
ftrace. It creates a producer and a consumer (option to disable creation
of the consumer) and will run for 10 seconds, then sleep for 10 seconds
and then repeat.
While running, the producer will write 10 byte loads into the ring
buffer with just putting in the current CPU number. The reader will
continually try to read the buffer. The reader will alternate from reading
the buffer via event by event, or by full pages.
The output is a pr_info, thus it will fill up the syslogs.
Starting ring buffer hammer
End ring buffer hammer
Time: 9000349 (usecs)
Overruns: 12578640
Read: 5358440 (by events)
Entries: 0
Total: 17937080
Missed: 0
Hit: 17937080
Entries per millisec: 1993
501 ns per entry
Sleeping for 10 secs
Starting ring buffer hammer
End ring buffer hammer
Time: 9936350 (usecs)
Overruns: 0
Read: 28146644 (by pages)
Entries: 74
Total: 28146718
Missed: 0
Hit: 28146718
Entries per millisec: 2832
353 ns per entry
Sleeping for 10 secs
Time: is the time the test ran
Overruns: the number of events that were overwritten and not read
Read: the number of events read (either by pages or events)
Entries: the number of entries left in the buffer
(the by pages will only read full pages)
Total: Entries + Read + Overruns
Missed: the number of entries that failed to write
Hit: the number of entries that were written
The above example shows that it takes ~353 nanosecs per entry when
there is a reader, reading by pages (and no overruns)
The event by event reader slowed the producer down to 501 nanosecs.
[ Impact: see how changes to the ring buffer affect stability and performance ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
In the hot path of the ring buffer "__rb_reserve_next" there's a big
if statement that does not even return back to the work flow.
code;
if (cross to next page) {
[ lots of code ]
return;
}
more code;
The condition is even the unlikely path, although we do not denote it
with an unlikely because gcc is fine with it. The condition is true when
the write crosses a page boundary, and we need to start at a new page.
Having this if statement makes it hard to read, but calling another
function to do the work is also not appropriate, because we are using a lot
of variables that were set before the if statement, and we do not want to
send them as parameters.
This patch changes it to a goto:
code;
if (cross to next page)
goto next_page;
more code;
return;
next_page:
[ lots of code]
This makes the code easier to understand, and a bit more obvious.
The output from gcc is practically identical. For some reason, gcc decided
to use different registers when I switched it to a goto. But other than that,
the logic is the same.
[ Impact: easier to read code ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When adding the EXPORT_SYMBOL to some of the tracing API, I accidently
used EXPORT_SYMBOL instead of EXPORT_SYMBOL_GPL. This patch fixes
that mistake.
[ Impact: export the tracing code only for GPL modules ]
Reported-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
As a precaution, it is best to disable writing to the ring buffers
when reseting them.
[ Impact: prevent weird things if write happens during reset ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
In the swap page ring buffer code that is used by the ftrace splice code,
we scan the page to increment the counter of entries read.
With the number of entries already in the page we simply need to add it.
[ Impact: speed up reading page from ring buffer ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
* 'irq/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
Revert "genirq: assert that irq handlers are indeed running in hardirq context"
* 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
tracing: x86, mmiotrace: fix range test
tracing: fix ref count in splice pages
Currently, when the ring buffer writer overflows the buffer and must
write over non consumed data, we increment the overrun counter by
reading the entries on the page we are about to overwrite. This reads
the entries one by one.
This is not very effecient. This patch adds another entry counter
into each buffer page descriptor that keeps track of the number of
entries on the page. Now on overwrite, the overrun counter simply
needs to add the number of entries that is on the page it is about
to overwrite.
[ Impact: speed up of ring buffer in overwrite mode ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
As a kernel thread, rcu_sched_grace_period() runs with all signals ignored.
It can never receive a signal even if it sleeps in TASK_INTERRUPTIBLE, it
needs the explicit allow_signal() to be visible for signals.
[ Impact: reduce kernel size, remove dead code ]
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <20090503211118.GA22973@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The entries counter in cpu buffer is not atomic. It can be updated by
other interrupts or from another CPU (readers).
But making entries into "atomic_t" causes an atomic operation that can
hurt performance. Instead we convert it to a local_t that will increment
a counter with a local CPU atomic operation (if the arch supports it).
Instead of fighting with readers and overwrites that decrement the counter,
I added a "read" counter. Every time a reader reads an entry it is
incremented.
We already have a overrun counter and with that, the entries counter and
the read counter, we can calculate the total number of entries in the
buffer with:
(entries - overrun) - read
As long as the total number of entries in the ring buffer is less than
the word size, this will work. But since the entries counter was previously
a long, this is no different than what we had before.
Thanks to Andrew Morton for pointing out in the first version that
atomic_t does not replace unsigned long. I switched to atomic_long_t
even though it is signed. A negative count is most likely a bug.
[ Impact: keep accurate count of cpu buffer entries ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Redirect the output to the parent counter and put in some sanity checks.
[ Impact: new perfcounter feature - inherited sampling counters ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090505155437.331556171@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use -1 instead of 0 as unlocked, since 0 is a valid cpu number.
( This is not an issue right now but will be once we allow multiple
counters to output to the same mmap area. )
[ Impact: prepare code for multi-counter profile output ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090505155437.232686598@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Provide a threshold to relax the mlock accounting, increasing usability.
Each counter gets perf_counter_mlock_kb for free.
[ Impact: allow more mmap buffering ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090505155437.112113632@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Provide a way to reset an existing counter - this eases PAPI
libraries around perfcounters.
Similar to read() it doesn't collapse pending child counters.
[ Impact: new perfcounter fd ioctl method to reset counters ]
Suggested-by: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <20090505155437.022272933@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Keep data_head up-to-date irrespective of notifications. This fixes
the case where you disable a counter and don't get a notification for
the last few pending events, and it also allows polling usage.
[ Impact: increase precision of perfcounter mmap-ed fields ]
Suggested-by: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
LKML-Reference: <20090505155436.925084300@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Thomas noted that we should disallow sysctl_sched_rt_runtime == 0 for
(!RT_GROUP) since the root group always has some RT tasks in it.
Further, update the documentation to inspire clue.
[ Impact: exclude corner-case sysctl_sched_rt_runtime value ]
Reported-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <20090505155436.863098054@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch adds stats to the ftrace ring buffers:
# cat /debugfs/tracing/per_cpu/cpu0/stats
entries: 42360
overrun: 30509326
commit overrun: 0
nmi dropped: 0
Where entries are the total number of data entries in the buffer.
overrun is the number of entries not consumed and were overwritten by
the writer.
commit overrun is the number of entries dropped due to nested writers
wrapping the buffer before the initial writer finished the commit.
nmi dropped is the number of entries dropped due to the ring buffer
lock being held when an nmi was going to write to the ring buffer.
Note, this field will be meaningless and will go away when the ring
buffer becomes lockless.
[ Impact: let userspace know what is happening in the ring buffers ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The WARN_ON in the ring buffer when a commit is preempted and the
buffer is filled by preceding writes can happen in normal operations.
The WARN_ON makes it look like a bug, not to mention, because
it does not stop tracing and calls printk which can also recurse, this
is prone to deadlock (the WARN_ON is not in a position to recurse).
This patch removes the WARN_ON and replaces it with a counter that
can be retrieved by a tracer. This counter is called commit_overrun.
While at it, I added a nmi_dropped counter to count any time an NMI entry
is dropped because the NMI could not take the spinlock.
[ Impact: prevent deadlock by printing normal case warning ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
I'm adding a module to do a series of tests on the ring buffer as well
as benchmarks. This module needs to have more of the ring buffer API
exported. There's nothing wrong with reading the ring buffer from a
module.
[ Impact: allow modules to read pages from the ring buffer ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Now percpu counters can be initialized very early. But the init
sequence uses mutex_lock(). Fortunately, perf_resource_mutex should
be a spinlock anyway, so convert it.
[ Impact: fix crash due to early init mutex use ]
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
percpu scheduling for perfcounters wants to take the context lock,
but that lock first needs to be initialized. Currently it is an
early_initcall() - but that is too late, the task tick runs much
sooner than that.
Call it explicitly from the scheduler init sequence instead.
[ Impact: fix access-before-init crash ]
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This used to be unstable when we had the rq->lock dependencies,
but now that they are that of the past we can turn on percpu
counter RR too.
[ Impact: handle counter over-commit for per-CPU counters too ]
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Some arches don't supply their own clocksource. This is mainly the
case in architectures that get their inter-tick times by reading the
counter on their interval timer. Since these timers wrap every tick,
they're not really useful as clocksources. Wrapping them to act like
one is possible but not very efficient. So we provide a callout these
arches can implement for use with the jiffies clocksource to provide
finer then tick granular time.
[ Impact: ease the migration to generic time keeping ]
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Setup clocksource mult_orig in clocksource_enable().
Clocksource drivers can save power by using keeping the
device clock disabled while the clocksource is unused.
In practice this means that the enable() and disable()
callbacks perform clk_enable() and clk_disable().
The enable() callback may also use clk_get_rate() to get
the clock rate from the clock framework. This information
can then be used to calculate the shift and mult variables.
Currently the mult_orig variable is setup from mult at
registration time only. This is conflicting with the above
case since the clock is disabled and the mult variable is
not yet calculated at the time of registration.
Moving the mult_orig setup code to clocksource_enable()
allows us to both handle the common case with no enable()
callback and the mult-changed-after-enable() case.
[ Impact: allow dynamic clock source usage ]
Signed-off-by: Magnus Damm <damm@igel.co.jp>
LKML-Reference: <20090501054546.8193.10688.sendpatchset@rx1.opensource.se>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
tick_handle_periodic() can lock up hard when a one shot clock event
device is used in combination with jiffies clocksource.
Avoid an endless loop issue by requiring that a highres valid
clocksource be installed before we call tick_periodic() in a loop when
using ONESHOT mode. The result is we will only increment jiffies once
per interrupt until a continuous hardware clocksource is available.
Without this, we can run into a endless loop, where each cycle through
the loop, jiffies is updated which increments time by tick_period or
more (due to clock steering), which can cause the event programming to
think the next event was before the newly incremented time and fail
causing tick_periodic() to be called again and the whole process loops
forever.
[ Impact: prevent hard lock up ]
Signed-off-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@kernel.org
move_irq_desc() will try to move irq_desc to the home node if
the allocated one is not correct, in create_irq_nr().
( This can happen on devices that are on different nodes that
are using MSI, when drivers are loaded and unloaded randomly. )
v2: fix non-smp build
v3: add NUMA_IRQ_DESC to eliminate #ifdefs
[ Impact: improve irq descriptor locality on NUMA systems ]
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
LKML-Reference: <49F95EAE.2050903@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This reverts commit 044d408409.
The commit added a warning when handle_IRQ_event() is called outside
of hard interrupt context. This breaks the generic tasklet based
interrupt resend mechanism which is used when the hardware has no way
to retrigger the interrupt. So we get a warning for a use case which
is correct and worked for years. Remove it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
When two (or more) contexts output to the same buffer, it is possible
to observe half written output.
Suppose we have CPU0 doing perf_counter_mmap(), CPU1 doing
perf_counter_overflow(). If CPU1 does a wakeup and exposes head to
user-space, then CPU2 can observe the data CPU0 is still writing.
[ Impact: fix occasionally corrupted profiling records ]
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
LKML-Reference: <20090501102533.007821627@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
I was never able to understand what should we actually do when
security_task_wait() fails, but the current code doesn't look right.
If ->task_wait() returns the error, we update *notask_error correctly.
But then we either reap the child (despite the fact this was forbidden)
or clear *notask_error (and hide the securiy policy problems).
This patch assumes that "stolen by ptrace" doesn't matter. If selinux
denies the child we should ignore it but make sure we report -EACCESS
instead of -ECHLD if there are no other eligible children.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
This is necessary to avoid the conflict of syscall numbers.
Conflicts:
arch/x86/ia32/ia32entry.S
arch/x86/include/asm/unistd_32.h
arch/x86/include/asm/unistd_64.h
Fixes up the borked syscall numbers of perfcounters versus
preadv/pwritev as well.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
sys_kill has the per thread counterpart sys_tgkill. sigqueueinfo is
missing a thread directed counterpart. Such an interface is important
for migrating applications from other OSes which have the per thread
delivery implemented.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: Roland McGrath <roland@redhat.com>
Acked-by: Ulrich Drepper <drepper@redhat.com>
Split out the code from do_tkill to make it reusable by the follow up
patch which implements sys_rt_tgsigqueueinfo
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
The new requeue PI futex op codes were modeled after the existing
FUTEX_REQUEUE and FUTEX_CMP_REQUEUE calls. I was unaware at the time
that FUTEX_REQUEUE was only around for compatibility reasons and
shouldn't be used in new code. Ulrich Drepper elaborates on this in his
Futexes are Tricky paper: http://people.redhat.com/drepper/futex.pdf.
The deprecated call doesn't catch changes to the futex corresponding to
the destination futex which can lead to deadlock.
Therefor, I feel it best to remove FUTEX_REQUEUE_PI and leave only
FUTEX_CMP_REQUEUE_PI as there are not yet any existing users of the API.
This patch does change the OP code value of FUTEX_CMP_REQUEUE_PI to 12
from 13. Since my test case is the only known user of this API, I felt
this was the right thing to do, rather than leave a hole in the
enumeration.
I chose to continue using the _CMP_ modifier in the OP code to make it
explicit to the user that the test is being done.
Builds, boots, and ran several hundred iterations requeue_pi.c.
Signed-off-by: Darren Hart <dvhltc@us.ibm.com>
LKML-Reference: <49ED580E.1050502@us.ibm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
include/linux/mutex.h:136: warning: 'mutex_lock' declared inline after being called
include/linux/mutex.h:136: warning: previous declaration of 'mutex_lock' was here
uninline it.
[ Impact: clean up and uninline, address compiler warning ]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Eric Paris <eparis@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <200904292318.n3TNIsi6028340@imap1.linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This adds my name to the list of copyright holders on the core
perf_counter.c, since I have contributed a significant amount of the
code in there.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Robert Richter <robert.richter@amd.com>
LKML-Reference: <18936.59200.888049.746658@cargo.ozlabs.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Sparse reports the following in kernel/posix-cpu-timers.c:
warning: symbol 'firing' shadows an earlier one
Signed-off-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Subrata Modak <subrata@linux.vnet.ibm.com>
LKML-Reference: <BD79186B4FD85F4B8E60E381CAEE1909016C1AFE@mi8nycmail19.Mi8.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Don't flush inherited SIGKILL during execve() in SELinux's post cred commit
hook. This isn't really a security problem: if the SIGKILL came before the
credentials were changed, then we were right to receive it at the time, and
should honour it; if it came after the creds were changed, then we definitely
should honour it; and in any case, all that will happen is that the process
will be scrapped before it ever returns to userspace.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: James Morris <jmorris@namei.org>
Two minor updates on functions documentation:
- Updated documentation for function rt_mutex_unlock(), which contained an
incorrect name
- Removed extra '*' from comment in function rt_mutex_destroy()
[ Impact: cleanup ]
Signed-off-by: Luis Henriques <henrix@sapo.pt>
Cc: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20090429205451.GA23154@hades.domain.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Andrew Gallatin reported that IRQ and SOFTIRQ times were
sometime not reported correctly on recent kernels, and even
bisected to commit 457533a7d3
([PATCH] fix scaled & unscaled cputime accounting) as the first
bad commit.
Further analysis pointed that commit
79741dd357 ([PATCH] idle cputime
accounting) was the real cause of the problem.
account_process_tick() was not taking into account timer IRQ
interrupting the idle task servicing a hard or soft irq.
On mostly idle cpu, irqs were thus not accounted and top or
mpstat could tell user/admin that cpu was 100 % idle, 0.00 %
irq, 0.00 % softirq, while it was not.
[ Impact: fix occasionally incorrect CPU statistics in top/mpstat ]
Reported-by: Andrew Gallatin <gallatin@myri.com>
Re-reported-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: rick.jones2@hp.com
Cc: brice@myri.com
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
LKML-Reference: <49F84BC1.7080602@cosmosbay.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch renames struct hw_perf_counter_ops into struct pmu. It
introduces a structure to describe a cpu specific pmu (performance
monitoring unit). It may contain ops and data. The new name of the
structure fits better, is shorter, and thus better to handle. Where it
was appropriate, names of function and variable have been changed too.
[ Impact: cleanup ]
Signed-off-by: Robert Richter <robert.richter@amd.com>
Cc: Paul Mackerras <paulus@samba.org>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1241002046-8832-7-git-send-email-robert.richter@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add a section to the memory barriers document to note the implied
memory barriers of sleep primitives (set_current_state() and wrappers)
and wake-up primitives (wake_up() and co.).
Also extend the in-code comments on the wake_up() functions to note
these implied barriers.
[ Impact: add documentation ]
Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
LKML-Reference: <20090428140138.1192.94723.stgit@warthog.procyon.org.uk>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
"tracing: create automated trace defines" causes this compile error on s390,
as reported by Sachin Sant against linux-next:
kernel/built-in.o: In function `__do_softirq':
(.text+0x1c680): undefined reference to `__tracepoint_softirq_entry'
This happens because the definitions of the softirq tracepoints were moved
from kernel/softirq.c to kernel/irq/handle.c. Since s390 doesn't support
generic hardirqs handle.c doesn't get compiled and the definitions are
missing.
So move the tracepoints to softirq.c again.
[ Impact: fix build failure on s390 ]
Reported-by: Sachin Sant <sachinp@in.ibm.com>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: fweisbec@gmail.com
LKML-Reference: <20090429135139.5fac79b8@osiris.boeblingen.de.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Replace the current event parser hack with a better one. Filters are
no longer specified predicate by predicate, but all at once and can
use parens and any of the following operators:
numeric fields:
==, !=, <, <=, >, >=
string fields:
==, !=
predicates can be combined with the logical operators:
&&, ||
examples:
"common_preempt_count > 4" > filter
"((sig >= 10 && sig < 15) || sig == 17) && comm != bash" > filter
If there was an error, the erroneous string along with an error
message can be seen by looking at the filter e.g.:
((sig >= 10 && sig < 15) || dsig == 17) && comm != bash
^
parse_error: Field not found
Currently the caret for an error always appears at the beginning of
the filter; a real position should be used, but the error message
should be useful even without it.
To clear a filter, '0' can be written to the filter file.
Filters can also be set or cleared for a complete subsystem by writing
the same filter as would be written to an individual event to the
filter file at the root of the subsytem. Note however, that if any
event in the subsystem lacks a field specified in the filter being
set, the set will fail and all filters in the subsytem are
automatically cleared. This change from the previous version was made
because using only the fields that happen to exist for a given event
would most likely result in a meaningless filter.
Because the logical operators are now implemented as predicates, the
maximum number of predicates in a filter was increased from 8 to 16.
[ Impact: add new, extended trace-filter implementation ]
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: fweisbec@gmail.com
Cc: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <1240905899.6416.121.camel@tropicana>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The new filter comparison ops need to be able to distinguish between
signed and unsigned field types, so add an is_signed flag/param to the
event field struct/trace_define_fields(). Also define a simple macro,
is_signed_type() to determine the signedness at compile time, used in the
trace macros. If the is_signed_type() macro won't work with a specific
type, a new slightly modified version of TRACE_FIELD() called
TRACE_FIELD_SIGN(), allows the signedness to be set explicitly.
[ Impact: extend trace-filter code for new feature ]
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: fweisbec@gmail.com
Cc: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <1240905893.6416.120.camel@tropicana>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Create a new event_filter object, and move the pred-related members
out of the call and subsystem objects and into the filter object - the
details of the filter implementation don't need to be exposed in the
call and subsystem in any case, and it will also help make the new
parser implementation a little cleaner.
[ Impact: refactor trace-filter code to prepare for new features ]
Signed-off-by: Tom Zanussi <tzanussi@gmail.com>
Acked-by: Steven Rostedt <rostedt@goodmis.org>
Cc: fweisbec@gmail.com
Cc: Li Zefan <lizf@cn.fujitsu.com>
LKML-Reference: <1240905887.6416.119.camel@tropicana>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The pages allocated for the splice binary buffer did not initialize
the ref count correctly. This caused pages not to be freed and causes
a drastic memory leak.
Thanks to logdev I was able to trace the tracer to find where the leak
was.
[ Impact: stop memory leak when using splice ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The warning output in trace_recursive_lock uses %d for a long when
it should be %ld.
[ Impact: fix compile warning ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Splice works with pages, it is much more effecient to use an entire
page than to copy bits over several pages.
Using logdev to trace the internals of the splice mechanism, I was
able to see that splice can be very aggressive. When tracing is
occurring, and the reader caught up to the writer, and the writer
is on the reader page, the reader will copy what is there into the
splice page. Splice may iterate over several pages and if the
writer is still writing to the page, the reader will keep copying
bits to new pages to pass to userspace.
This patch changes it to only pass data to userspace if the page
is full (the writer has left the page). This has a small side effect
that splice can not read a partial page, and must wait for the
page to fill. This should not be an issue. If tracing has stopped,
then a use of "read" will still read all of the page.
[ Impact: better performance for ring buffer splice code ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The splice code allocates a page even when the ring buffer is empty.
It detects the ring buffer being empty when it it fails to copy
anything from the ring buffer into the page.
This patch adds a check to see if there is anything in the ring buffer
before allocating a page.
Thanks to logdev for letting me trace the tracer to find this.
[ Impact: speed up due to removing unnecessary allocation ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
The pages allocated for the splice binary buffer did not initialize
the ref count correctly. This caused pages not to be freed and causes
a drastic memory leak.
Thanks to logdev I was able to trace the tracer to find where the leak
was.
[ Impact: stop memory leak when using splice ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
ftrace_dump is used for printing out the contents of the ftrace ring buffer
to the console on failure. Currently it uses a spinlock to synchronize
the output from multiple failures on different CPUs. This spin lock
currently is a normal spinlock and can cause issues with lockdep and
lock tracing.
This patch converts it to raw since it is for error handling only.
The lock is local to the ftrace_dump and is not used by any other
infrastructure.
[ Impact: prevent ftrace_dump from locking up by internal tracing ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
This simplifies the node awareness of the code. All our allocators
only deal with a NUMA node ID locality not with CPU ids anyway - so
there's no need to maintain (and transform) a CPU id all across the
IRq layer.
v2: keep move_irq_desc related
[ Impact: cleanup, prepare IRQ code to be NUMA-aware ]
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
LKML-Reference: <49F65536.2020300@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
