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()
+The special file descriptor is opened via the perf_event_open()
system call:
- int sys_perf_counter_open(struct perf_counter_hw_event *hw_event_uptr,
+ int sys_perf_event_open(struct perf_event_hw_event *hw_event_uptr,
pid_t pid, int cpu, int group_fd,
unsigned long flags);
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:
+When creating a new counter fd, 'perf_event_hw_event' is:
-struct perf_counter_hw_event {
+struct perf_event_hw_event {
/*
* The MSB of the config word signifies if the rest contains cpu
* specific (raw) counter configuration data, if unset, the next
/*
* Generalized performance counter event types, used by the hw_event.event_id
- * parameter of the sys_perf_counter_open() syscall:
+ * parameter of the sys_perf_event_open() syscall:
*/
enum hw_event_ids {
/*
* reads on the counter should return the indicated quantities,
* in increasing order of bit value, after the counter value.
*/
-enum perf_counter_read_format {
+enum perf_event_read_format {
PERF_FORMAT_TOTAL_TIME_ENABLED = 1,
PERF_FORMAT_TOTAL_TIME_RUNNING = 2,
};
* Bits that can be set in hw_event.record_type to request information
* in the overflow packets.
*/
-enum perf_counter_record_format {
+enum perf_event_record_format {
PERF_RECORD_IP = 1U << 0,
PERF_RECORD_TID = 1U << 1,
PERF_RECORD_TIME = 1U << 2,
The 'comm' bit allows tracking of process comm data on process creation.
This too is recorded in the ring-buffer (see below).
-The 'pid' parameter to the perf_counter_open() system call allows the
+The 'pid' parameter to the perf_event_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
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
+is created first, with group_fd = -1 in the perf_event_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
accessed through mmap().
The mmap size should be 1+2^n pages, where the first page is a meta-data page
-(struct perf_counter_mmap_page) that contains various bits of information such
+(struct perf_event_mmap_page) that contains various bits of information such
as where the ring-buffer head is.
/*
* Structure of the page that can be mapped via mmap
*/
-struct perf_counter_mmap_page {
+struct perf_event_mmap_page {
__u32 version; /* version number of this structure */
__u32 compat_version; /* lowest version this is compat with */
* Control data for the mmap() data buffer.
*
* User-space reading this value should issue an rmb(), on SMP capable
- * platforms, after reading this value -- see perf_counter_wakeup().
+ * platforms, after reading this value -- see perf_event_wakeup().
*/
__u32 data_head; /* head in the data section */
};
The following 2^n pages are the ring-buffer which contains events of the form:
-#define PERF_EVENT_MISC_KERNEL (1 << 0)
-#define PERF_EVENT_MISC_USER (1 << 1)
-#define PERF_EVENT_MISC_OVERFLOW (1 << 2)
+#define PERF_RECORD_MISC_KERNEL (1 << 0)
+#define PERF_RECORD_MISC_USER (1 << 1)
+#define PERF_RECORD_MISC_OVERFLOW (1 << 2)
struct perf_event_header {
__u32 type;
* char filename[];
* };
*/
- PERF_EVENT_MMAP = 1,
- PERF_EVENT_MUNMAP = 2,
+ PERF_RECORD_MMAP = 1,
+ PERF_RECORD_MUNMAP = 2,
/*
* struct {
* char comm[];
* };
*/
- PERF_EVENT_COMM = 3,
+ PERF_RECORD_COMM = 3,
/*
- * When header.misc & PERF_EVENT_MISC_OVERFLOW the event_type field
+ * When header.misc & PERF_RECORD_MISC_OVERFLOW the event_type field
* will be PERF_RECORD_*
*
* struct {
fcntl() managing signals.
Normally a notification is generated for every page filled, however one can
-additionally set perf_counter_hw_event.wakeup_events to generate one every
+additionally set perf_event_hw_event.wakeup_events to generate one every
so many counter overflow events.
Future work will include a splice() interface to the ring-buffer.
An individual counter or counter group can be enabled with
- ioctl(fd, PERF_COUNTER_IOC_ENABLE);
+ ioctl(fd, PERF_EVENT_IOC_ENABLE);
or disabled with
- ioctl(fd, PERF_COUNTER_IOC_DISABLE);
+ ioctl(fd, PERF_EVENT_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
Additionally, non-inherited overflow counters can use
- ioctl(fd, PERF_COUNTER_IOC_REFRESH, nr);
+ ioctl(fd, PERF_EVENT_IOC_REFRESH, nr);
to enable a counter for 'nr' events, after which it gets disabled again.
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_EVENTS_ENABLE);
- prctl(PR_TASK_PERF_COUNTERS_DISABLE);
+ prctl(PR_TASK_PERF_EVENTS_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
If your architecture does not have hardware performance metrics, you can
still use the generic software counters based on hrtimers for sampling.
-So to start with, in order to add HAVE_PERF_COUNTERS to your Kconfig, you
+So to start with, in order to add HAVE_PERF_EVENTS to your Kconfig, you
will need at least this:
- - asm/perf_counter.h - a basic stub will suffice at first
+ - asm/perf_event.h - a basic stub will suffice at first
- support for atomic64 types (and associated helper functions)
- - set_perf_counter_pending() implemented
+ - set_perf_event_pending() implemented
If your architecture does have hardware capabilities, you can override the
-weak stub hw_perf_counter_init() to register hardware counters.
+weak stub hw_perf_event_init() to register hardware counters.
+
+Architectures that have d-cache aliassing issues, such as Sparc and ARM,
+should select PERF_USE_VMALLOC in order to avoid these for perf mmap().