Skip to content

L
linux-davinci
Switch branch/tag
  1. 11 Dec, 2008 10 commits
    • Ingo Molnar's avatar
      perf counters: implement PERF_COUNT_TASK_CLOCK · bae43c99
      Ingo Molnar authored 
      Impact: add new perf-counter type

The 'task clock' counter counts the amount of time a task is executing,
in nanoseconds. It stops ticking when a task is scheduled out either due
to it blocking, sleeping or it being preempted.

This counter type is a Linux kernel based abstraction, it is available
even if the hardware does not support native hardware performance counters.
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      bae43c99
    • Ingo Molnar's avatar
      perf counters: consolidate hw_perf save/restore APIs · 01b2838c
      Ingo Molnar authored 
      Impact: cleanup

Rename them to better match up the usual IRQ disable/enable APIs:

 hw_perf_disable_all()  => hw_perf_save_disable()
 hw_perf_restore_ctrl() => hw_perf_restore()
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      01b2838c
    • Ingo Molnar's avatar
      perf counters: implement PERF_COUNT_CPU_CLOCK · 5c92d124
      Ingo Molnar authored 
      Impact: add new perf-counter type

The 'CPU clock' counter counts the amount of CPU clock time that is
elapsing, in nanoseconds. (regardless of how much of it the task is
spending on a CPU executing)

This counter type is a Linux kernel based abstraction, it is available
even if the hardware does not support native hardware performance counters.
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      5c92d124
    • Ingo Molnar's avatar
      perf counters: hw driver API · 621a01ea
      Ingo Molnar authored 
      Impact: restructure code, introduce hw_ops driver abstraction

Introduce this abstraction to handle counter details:

 struct hw_perf_counter_ops {
	void (*hw_perf_counter_enable)	(struct perf_counter *counter);
	void (*hw_perf_counter_disable)	(struct perf_counter *counter);
	void (*hw_perf_counter_read)	(struct perf_counter *counter);
 };

This will be useful to support assymetric hw details, and it will also
be useful to implement "software counters". (Counters that count kernel
managed sw events such as pagefaults, context-switches, wall-clock time
or task-local time.)
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      621a01ea
    • Ingo Molnar's avatar
      perf counters: group counter, fixes · ccff286d
      Ingo Molnar authored 
      Impact: bugfix

Check that a group does not span outside the context of a CPU or a task.

Also, do not allow deep recursive hierarchies.
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      ccff286d
    • Ingo Molnar's avatar
      perf counters: add support for group counters · 04289bb9
      Ingo Molnar authored 
      Impact: add group counters

This patch adds the "counter groups" abstraction.

Groups of counters behave much like normal 'single' counters, with a
few semantic and behavioral extensions on top of that.

A counter group is created by creating a new counter with the open()
syscall's group-leader group_fd file descriptor parameter pointing
to another, already existing counter.

Groups of counters are scheduled in and out in one atomic group, and
they are also roundrobin-scheduled atomically.

Counters that are member of a group can also record events with an
(atomic) extended timestamp that extends to all members of the group,
if the record type is set to PERF_RECORD_GROUP.
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      04289bb9
    • Ingo Molnar's avatar
      perf counters: restructure the API · 9f66a381
      Ingo Molnar authored 
      Impact: clean up new API

Thorough cleanup of the new perf counters API, we now get clean separation
of the various concepts:

 - introduce perf_counter_hw_event to separate out the event source details

 - move special type flags into separate attributes: PERF_COUNT_NMI,
   PERF_COUNT_RAW

 - extend the type to u64 and reserve it fully to the architecture in the
   raw type case.

And make use of all these changes in the core and x86 perfcounters code.

Also change the syscall signature to:

  asmlinkage int sys_perf_counter_open(

	struct perf_counter_hw_event	*hw_event_uptr		__user,
	pid_t				pid,
	int				cpu,
	int				group_fd);

( Note that group_fd is unused for now - it's reserved for the counter
  groups abstraction. )
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      9f66a381
    • Thomas Gleixner's avatar
      perf counters: expand use of counter->event · dfa7c899
      Thomas Gleixner authored 
      Impact: change syscall, cleanup

Make use of the new perf_counters event type.
Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      dfa7c899
    • Thomas Gleixner's avatar
      perf counters: clean up 'raw' type API · eab656ae
      Thomas Gleixner authored 
      Impact: cleanup

Introduce a separate hw_event type.
Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      eab656ae
    • Thomas Gleixner's avatar
      perf counters: protect them against CSTATE transitions · 4ac13294
      Thomas Gleixner authored 
      Impact: fix rare lost events problem

There are CPUs whose performance counters misbehave on CSTATE transitions,
so provide a way to just disable/enable them around deep idle methods.

(hw_perf_enable_all() is cheap on x86.)
Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
Signed-off-by: default avatarIngo Molnar <mingo@elte.hu>
      4ac13294
  3. 09 Dec, 2008 3 commits
  5. 08 Dec, 2008 7 commits
  7. 05 Dec, 2008 20 commits