linux-toradex.git/kernel/hw_breakpoint.c, branch v2.6.34

Merge branch 'perf/core' into perf/urgent

2010-03-04T10:47:52+00:00

Merge reason: Switch from pre-merge topical split to the post-merge urgent track

Signed-off-by: Ingo Molnar

hw-breakpoints: Remove stub unthrottle callback

2010-02-28T19:51:15+00:00

We support event unthrottling in breakpoint events. It means
that if we have more than sysctl_perf_event_sample_rate/HZ,
perf will throttle, ignoring subsequent events until the next
tick.

So if ptrace exceeds this max rate, it will omit events, which
breaks the ptrace determinism that is supposed to report every
triggered breakpoints. This is likely to happen if we set
sysctl_perf_event_sample_rate to 1.

This patch removes support for unthrottling in breakpoint
events to break throttling and restore ptrace determinism.

Signed-off-by: Frederic Weisbecker 
Cc: 2.6.33.x 
Cc: Peter Zijlstra 
Cc: K.Prasad 
Cc: Paul Mackerras

percpu: Add __percpu sparse annotations to hw_breakpoint

2010-02-27T15:23:39+00:00

Add __percpu sparse annotations to hw_breakpoint.

These annotations are to make sparse consider percpu variables to be
in a different address space and warn if accessed without going
through percpu accessors.  This patch doesn't affect normal builds.

In kernel/hw_breakpoint.c, per_cpu(nr_task_bp_pinned, cpu)'s will
trigger spurious noderef related warnings from sparse.  Changing it to
&per_cpu(nr_task_bp_pinned[0], cpu) will work around the problem but
deemed to ugly by the maintainer.  Leave it alone until better
solution can be found.

Signed-off-by: Tejun Heo 
Cc: Stephen Rothwell 
Cc: K.Prasad 
LKML-Reference: <4B7B4B7A.9050902@kernel.org>
Signed-off-by: Frederic Weisbecker

perf: Make bp_len type to u64 generic across the arch

2010-02-04T00:07:12+00:00

Change 'bp_len' type to __u64 to make it work across archs as
the s390 architecture watch point length can be upto 2^64.

reference:
	http://lkml.org/lkml/2010/1/25/212

This is an ABI change that is not backward compatible with
the previous hardware breakpoint info layout integrated in this
development cycle, a rebuilt of perf tools is necessary for
versions based on 2.6.33-rc1 - 2.6.33-rc6 to work with a
kernel based on this patch.

Signed-off-by: Mahesh Salgaonkar 
Acked-by: Peter Zijlstra 
Cc: Ananth N Mavinakayanahalli 
Cc: "K. Prasad" 
Cc: Maneesh Soni 
Cc: Heiko Carstens 
Cc: Martin 
LKML-Reference: <20100130045518.GA20776@in.ibm.com>
Signed-off-by: Frederic Weisbecker

perf, hw_breakpoint, kgdb: Do not take mutex for kernel debugger

2010-01-30T07:42:21+00:00

This patch fixes the regression in functionality where the
kernel debugger and the perf API do not nicely share hw
breakpoint reservations.

The kernel debugger cannot use any mutex_lock() calls because it
can start the kernel running from an invalid context.

A mutex free version of the reservation API needed to get
created for the kernel debugger to safely update hw breakpoint
reservations.

The possibility for a breakpoint reservation to be concurrently
processed at the time that kgdb interrupts the system is
improbable. Should this corner case occur the end user is
warned, and the kernel debugger will prohibit updating the
hardware breakpoint reservations.

Any time the kernel debugger reserves a hardware breakpoint it
will be a system wide reservation.

Signed-off-by: Jason Wessel 
Acked-by: Frederic Weisbecker 
Cc: kgdb-bugreport@lists.sourceforge.net
Cc: K.Prasad 
Cc: Peter Zijlstra 
Cc: Alan Stern 
Cc: torvalds@linux-foundation.org
LKML-Reference: <1264719883-7285-3-git-send-email-jason.wessel@windriver.com>
Signed-off-by: Ingo Molnar

hw_breakpoints: Release the bp slot if arch_validate_hwbkpt_settings() fails.

2010-01-28T13:15:51+00:00

On a given architecture, when hardware breakpoint registration fails
due to un-supported access type (read/write/execute), we lose the bp
slot since register_perf_hw_breakpoint() does not release the bp slot
on failure.
Hence, any subsequent hardware breakpoint registration starts failing
with 'no space left on device' error.

This patch introduces error handling in register_perf_hw_breakpoint()
function and releases bp slot on error.

Signed-off-by: Mahesh Salgaonkar 
Cc: Ananth N Mavinakayanahalli 
Cc: K. Prasad 
Cc: Maneesh Soni 
LKML-Reference: <20100121125516.GA32521@in.ibm.com>
Signed-off-by: Frederic Weisbecker

Merge branch 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip

2009-12-31T19:52:01+00:00

* 'tracing-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  tracing: Fix sign fields in ftrace_define_fields_##call()
  tracing/syscalls: Fix typo in SYSCALL_DEFINE0
  tracing/kprobe: Show sign of fields in trace_kprobe format files
  ksym_tracer: Remove trace_stat
  ksym_tracer: Fix race when incrementing count
  ksym_tracer: Fix to allow writing newline to ksym_trace_filter
  ksym_tracer: Fix to make the tracer work
  tracing: Kconfig spelling fixes and cleanups
  tracing: Fix setting tracer specific options
  Documentation: Update ftrace-design.txt
  Documentation: Update tracepoint-analysis.txt
  Documentation: Update mmiotrace.txt

ksym_tracer: Fix to make the tracer work

2009-12-30T06:50:47+00:00

ksym tracer doesn't work:

 # echo tasklist_lock:rw- > ksym_trace_filter
 -bash: echo: write error: No such device

It's because we pass to perf_event_create_kernel_counter()
a cpu number which is not present.

Signed-off-by: Li Zefan 
Cc: Steven Rostedt 
Cc: K.Prasad 
Cc: Frederic Weisbecker 
LKML-Reference: <4B3AF19E.1010201@cn.fujitsu.com>
Signed-off-by: Ingo Molnar

perf_event: Convert to raw_spinlock

2009-12-14T22:55:34+00:00

Convert locks which cannot be sleeping locks in preempt-rt to
raw_spinlocks.

Signed-off-by: Thomas Gleixner 
Acked-by: Peter Zijlstra 
Acked-by: Ingo Molnar

hw-breakpoints: Modify breakpoints without unregistering them

2009-12-09T08:48:20+00:00

Currently, when ptrace needs to modify a breakpoint, like disabling
it, changing its address, type or len, it calls
modify_user_hw_breakpoint(). This latter will perform the heavy and
racy task of unregistering the old breakpoint and registering a new
one.

This is racy as someone else might steal the reserved breakpoint
slot under us, which is undesired as the breakpoint is only
supposed to be modified, sometimes in the middle of a debugging
workflow. We don't want our slot to be stolen in the middle.

So instead of unregistering/registering the breakpoint, just
disable it while we modify its breakpoint fields and re-enable it
after if necessary.

Signed-off-by: Frederic Weisbecker 
Cc: Peter Zijlstra 
Cc: Prasad 
LKML-Reference: <1260347148-5519-1-git-send-regression-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar