aha/Documentation/perf_counter/design.txt

Performance Counters for Linux
------------------------------

Performance counters are special hardware registers available on most modern
CPUs. These registers count the number of certain types of hw events: such
as instructions executed, cachemisses suffered, or branches mis-predicted -
without slowing down the kernel or applications. These registers can also
trigger interrupts when a threshold number of events have passed - and can
thus be used to profile the code that runs on that CPU.

The Linux Performance Counter subsystem provides an abstraction of these
hardware capabilities. It provides per task and per CPU counters, counter
groups, and it provides event capabilities on top of those.  It
provides "virtual" 64-bit counters, regardless of the width of the
underlying hardware counters.

Performance counters are accessed via special file descriptors.
There's one file descriptor per virtual counter used.

The special file descriptor is opened via the perf_counter_open()
system call:

   int sys_perf_counter_open(struct perf_counter_hw_event *hw_event_uptr,
			     pid_t pid, int cpu, int group_fd,
			     unsigned long flags);

The syscall returns the new fd. The fd can be used via the normal
VFS system calls: read() can be used to read the counter, fcntl()
can be used to set the blocking mode, etc.

Multiple counters can be kept open at a time, and the counters
can be poll()ed.

When creating a new counter fd, 'perf_counter_hw_event' is:

/*
 * Event to monitor via a performance monitoring counter:
 */
struct perf_counter_hw_event {
	__u64			event_config;

	__u64			irq_period;
	__u64			record_type;
	__u64			read_format;

	__u64			disabled       :  1, /* off by default        */
				nmi	       :  1, /* NMI sampling          */
				inherit	       :  1, /* children inherit it   */
				pinned	       :  1, /* must always be on PMU */
				exclusive      :  1, /* only group on PMU     */
				exclude_user   :  1, /* don't count user      */
				exclude_kernel :  1, /* ditto kernel          */
				exclude_hv     :  1, /* ditto hypervisor      */
				exclude_idle   :  1, /* don't count when idle */

				__reserved_1   : 55;

	__u32			extra_config_len;

	__u32			__reserved_4;
	__u64			__reserved_2;
	__u64			__reserved_3;
};

The 'event_config' field specifies what the counter should count.  It
is divided into 3 bit-fields:

raw_type: 1 bit (most significant bit)		0x8000_0000_0000_0000
type:	  7 bits (next most significant)	0x7f00_0000_0000_0000
event_id: 56 bits (least significant)		0x00ff_0000_0000_0000

If 'raw_type' is 1, then the counter will count a hardware event
specified by the remaining 63 bits of event_config.  The encoding is
machine-specific.

If 'raw_type' is 0, then the 'type' field says what kind of counter
this is, with the following encoding:

enum perf_event_types {
	PERF_TYPE_HARDWARE		= 0,
	PERF_TYPE_SOFTWARE		= 1,
	PERF_TYPE_TRACEPOINT		= 2,
};

A counter of PERF_TYPE_HARDWARE will count the hardware event
specified by 'event_id':

/*
 * Generalized performance counter event types, used by the hw_event.event_id
 * parameter of the sys_perf_counter_open() syscall:
 */
enum hw_event_ids {
	/*
	 * Common hardware events, generalized by the kernel:
	 */
	PERF_COUNT_CPU_CYCLES		= 0,
	PERF_COUNT_INSTRUCTIONS		= 1,
	PERF_COUNT_CACHE_REFERENCES	= 2,
	PERF_COUNT_CACHE_MISSES		= 3,
	PERF_COUNT_BRANCH_INSTRUCTIONS	= 4,
	PERF_COUNT_BRANCH_MISSES	= 5,
	PERF_COUNT_BUS_CYCLES		= 6,
};

These are standardized types of events that work relatively uniformly
on all CPUs that implement Performance Counters support under Linux,
although there may be variations (e.g., different CPUs might count
cache references and misses at different levels of the cache hierarchy).
If a CPU is not able to count the selected event, then the system call
will return -EINVAL.

More hw_event_types are supported as well, but they are CPU-specific
and accessed as raw events.  For example, to count "External bus
cycles while bus lock signal asserted" events on Intel Core CPUs, pass
in a 0x4064 event_id value and set hw_event.raw_type to 1.

A counter of type PERF_TYPE_SOFTWARE will count one of the available
software events, selected by 'event_id':

/*
 * Special "software" counters provided by the kernel, even if the hardware
 * does not support performance counters. These counters measure various
 * physical and sw events of the kernel (and allow the profiling of them as
 * well):
 */
enum sw_event_ids {
	PERF_COUNT_CPU_CLOCK		= 0,
	PERF_COUNT_TASK_CLOCK		= 1,
	PERF_COUNT_PAGE_FAULTS		= 2,
	PERF_COUNT_CONTEXT_SWITCHES	= 3,
	PERF_COUNT_CPU_MIGRATIONS	= 4,
	PERF_COUNT_PAGE_FAULTS_MIN	= 5,
	PERF_COUNT_PAGE_FAULTS_MAJ	= 6,
};

Counters come in two flavours: counting counters and sampling
counters.  A "counting" counter is one that is used for counting the
number of events that occur, and is characterised by having
irq_period = 0 and record_type = PERF_RECORD_SIMPLE.  A read() on a
counting counter simply returns the current value of the counter as
an 8-byte number.

A "sampling" counter is one that is set up to generate an interrupt
every N events, where N is given by 'irq_period'.  A sampling counter
has irq_period > 0 and record_type != PERF_RECORD_SIMPLE.  The
record_type controls what data is recorded on each interrupt, and the
available values are currently:

/*
 * IRQ-notification data record type:
 */
enum perf_counter_record_type {
	PERF_RECORD_SIMPLE		= 0,
	PERF_RECORD_IRQ			= 1,
	PERF_RECORD_GROUP		= 2,
};

A record_type value of PERF_RECORD_IRQ will record the instruction
pointer (IP) at which the interrupt occurred.  A record_type value of
PERF_RECORD_GROUP will record the event_config and counter value of
all of the other counters in the group, and should only be used on a
group leader (see below).  Currently these two values are mutually
exclusive, but record_type will become a bit-mask in future and
support other values.

A sampling counter has an event queue, into which an event is placed
on each interrupt.  A read() on a sampling counter will read the next
event from the event queue.  If the queue is empty, the read() will
either block or return an EAGAIN error, depending on whether the fd
has been set to non-blocking mode or not.

The 'disabled' bit specifies whether the counter starts out disabled
or enabled.  If it is initially disabled, it can be enabled by ioctl
or prctl (see below).

The 'nmi' bit specifies, for hardware events, whether the counter
should be set up to request non-maskable interrupts (NMIs) or normal
interrupts.  This bit is ignored if the user doesn't have
CAP_SYS_ADMIN privilege (i.e. is not root) or if the CPU doesn't
generate NMIs from hardware counters.

The 'inherit' bit, if set, specifies that this counter should count
events on descendant tasks as well as the task specified.  This only
applies to new descendents, not to any existing descendents at the
time the counter is created (nor to any new descendents of existing
descendents).

The 'pinned' bit, if set, specifies that the counter should always be
on the CPU if at all possible.  It only applies to hardware counters
and only to group leaders.  If a pinned counter cannot be put onto the
CPU (e.g. because there are not enough hardware counters or because of
a conflict with some other event), then the counter goes into an
'error' state, where reads return end-of-file (i.e. read() returns 0)
until the counter is subsequently enabled or disabled.

The 'exclusive' bit, if set, specifies that when this counter's group
is on the CPU, it should be the only group using the CPU's counters.
In future, this will allow sophisticated monitoring programs to supply
extra configuration information via 'extra_config_len' to exploit
advanced features of the CPU's Performance Monitor Unit (PMU) that are
not otherwise accessible and that might disrupt other hardware
counters.

The 'exclude_user', 'exclude_kernel' and 'exclude_hv' bits provide a
way to request that counting of events be restricted to times when the
CPU is in user, kernel and/or hypervisor mode.


The 'pid' parameter to the perf_counter_open() system call allows the
counter to be specific to a task:

 pid == 0: if the pid parameter is zero, the counter is attached to the
 current task.

 pid > 0: the counter is attached to a specific task (if the current task
 has sufficient privilege to do so)

 pid < 0: all tasks are counted (per cpu counters)

The 'cpu' parameter allows a counter to be made specific to a CPU:

 cpu >= 0: the counter is restricted to a specific CPU
 cpu == -1: the counter counts on all CPUs

(Note: the combination of 'pid == -1' and 'cpu == -1' is not valid.)

A 'pid > 0' and 'cpu == -1' counter is a per task counter that counts
events of that task and 'follows' that task to whatever CPU the task
gets schedule to. Per task counters can be created by any user, for
their own tasks.

A 'pid == -1' and 'cpu == x' counter is a per CPU counter that counts
all events on CPU-x. Per CPU counters need CAP_SYS_ADMIN privilege.

The 'flags' parameter is currently unused and must be zero.

The 'group_fd' parameter allows counter "groups" to be set up.  A
counter group has one counter which is the group "leader".  The leader
is created first, with group_fd = -1 in the perf_counter_open call
that creates it.  The rest of the group members are created
subsequently, with group_fd giving the fd of the group leader.
(A single counter on its own is created with group_fd = -1 and is
considered to be a group with only 1 member.)

A counter group is scheduled onto the CPU as a unit, that is, it will
only be put onto the CPU if all of the counters in the group can be
put onto the CPU.  This means that the values of the member counters
can be meaningfully compared, added, divided (to get ratios), etc.,
with each other, since they have counted events for the same set of
executed instructions.

Counters can be enabled and disabled in two ways: via ioctl and via
prctl.  When a counter is disabled, it doesn't count or generate
events but does continue to exist and maintain its count value.

An individual counter or counter group can be enabled with

	ioctl(fd, PERF_COUNTER_IOC_ENABLE);

or disabled with

	ioctl(fd, PERF_COUNTER_IOC_DISABLE);

Enabling or disabling the leader of a group enables or disables the
whole group; that is, while the group leader is disabled, none of the
counters in the group will count.  Enabling or disabling a member of a
group other than the leader only affects that counter - disabling an
non-leader stops that counter from counting but doesn't affect any
other counter.

A process can enable or disable all the counter groups that are
attached to it, using prctl:

	prctl(PR_TASK_PERF_COUNTERS_ENABLE);

	prctl(PR_TASK_PERF_COUNTERS_DISABLE);

This applies to all counters on the current process, whether created
by this process or by another, and doesn't affect any counters that
this process has created on other processes.  It only enables or
disables the group leaders, not any other members in the groups.