Scroll to navigation

perf_event_open(2) System Calls Manual perf_event_open(2)

NAME

perf_event_open - set up performance monitoring

LIBRARY

Standard C library (libc, -lc)

SYNOPSIS

#include <linux/perf_event.h>    /* Definition of PERF_* constants */
#include <linux/hw_breakpoint.h> /* Definition of HW_* constants */
#include <sys/syscall.h>         /* Definition of SYS_* constants */
#include <unistd.h>
int syscall(SYS_perf_event_open, struct perf_event_attr *attr,
            pid_t pid, int cpu, int group_fd, unsigned long flags);

Note: glibc provides no wrapper for perf_event_open(), necessitating the use of syscall(2).

DESCRIPTION

Given a list of parameters, perf_event_open() returns a file descriptor, for use in subsequent system calls (read(2), mmap(2), prctl(2), fcntl(2), etc.).

A call to perf_event_open() creates a file descriptor that allows measuring performance information. Each file descriptor corresponds to one event that is measured; these can be grouped together to measure multiple events simultaneously.

Events can be enabled and disabled in two ways: via ioctl(2) and via prctl(2). When an event is disabled it does not count or generate overflows but does continue to exist and maintain its count value.

Events come in two flavors: counting and sampled. A counting event is one that is used for counting the aggregate number of events that occur. In general, counting event results are gathered with a read(2) call. A sampling event periodically writes measurements to a buffer that can then be accessed via mmap(2).

Arguments

The pid and cpu arguments allow specifying which process and CPU to monitor:

pid == 0 and cpu == -1
This measures the calling process/thread on any CPU.
pid == 0 and cpu >= 0
This measures the calling process/thread only when running on the specified CPU.
pid > 0 and cpu == -1
This measures the specified process/thread on any CPU.
pid > 0 and cpu >= 0
This measures the specified process/thread only when running on the specified CPU.
pid == -1 and cpu >= 0
This measures all processes/threads on the specified CPU. This requires CAP_PERFMON (since Linux 5.8) or CAP_SYS_ADMIN capability or a /proc/sys/kernel/perf_event_paranoid value of less than 1.
pid == -1 and cpu == -1
This setting is invalid and will return an error.

When pid is greater than zero, permission to perform this system call is governed by CAP_PERFMON (since Linux 5.9) and a ptrace access mode PTRACE_MODE_READ_REALCREDS check on older Linux versions; see ptrace(2).

The group_fd argument allows event groups to be created. An event group has one event which is the group leader. The leader is created first, with group_fd = -1. The rest of the group members are created with subsequent perf_event_open() calls with group_fd being set to the file descriptor of the group leader. (A single event on its own is created with group_fd = -1 and is considered to be a group with only 1 member.) An event group is scheduled onto the CPU as a unit: it will be put onto the CPU only if all of the events in the group can be put onto the CPU. This means that the values of the member events can be meaningfully compared —added, divided (to get ratios), and so on— with each other, since they have counted events for the same set of executed instructions.

The flags argument is formed by ORing together zero or more of the following values:

This flag enables the close-on-exec flag for the created event file descriptor, so that the file descriptor is automatically closed on execve(2). Setting the close-on-exec flags at creation time, rather than later with fcntl(2), avoids potential race conditions where the calling thread invokes perf_event_open() and fcntl(2) at the same time as another thread calls fork(2) then execve(2).
This flag tells the event to ignore the group_fd parameter except for the purpose of setting up output redirection using the PERF_FLAG_FD_OUTPUT flag.
This flag re-routes the event's sampled output to instead be included in the mmap buffer of the event specified by group_fd.
This flag activates per-container system-wide monitoring. A container is an abstraction that isolates a set of resources for finer-grained control (CPUs, memory, etc.). In this mode, the event is measured only if the thread running on the monitored CPU belongs to the designated container (cgroup). The cgroup is identified by passing a file descriptor opened on its directory in the cgroupfs filesystem. For instance, if the cgroup to monitor is called test, then a file descriptor opened on /dev/cgroup/test (assuming cgroupfs is mounted on /dev/cgroup) must be passed as the pid parameter. cgroup monitoring is available only for system-wide events and may therefore require extra permissions.

The perf_event_attr structure provides detailed configuration information for the event being created.


struct perf_event_attr {

__u32 type; /* Type of event */
__u32 size; /* Size of attribute structure */
__u64 config; /* Type-specific configuration */
union {
__u64 sample_period; /* Period of sampling */
__u64 sample_freq; /* Frequency of sampling */
};
__u64 sample_type; /* Specifies values included in sample */
__u64 read_format; /* Specifies values returned in read */
__u64 disabled : 1, /* off by default */
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, /* don't count kernel */
exclude_hv : 1, /* don't count hypervisor */
exclude_idle : 1, /* don't count when idle */
mmap : 1, /* include mmap data */
comm : 1, /* include comm data */
freq : 1, /* use freq, not period */
inherit_stat : 1, /* per task counts */
enable_on_exec : 1, /* next exec enables */
task : 1, /* trace fork/exit */
watermark : 1, /* wakeup_watermark */
precise_ip : 2, /* skid constraint */
mmap_data : 1, /* non-exec mmap data */
sample_id_all : 1, /* sample_type all events */
exclude_host : 1, /* don't count in host */
exclude_guest : 1, /* don't count in guest */
exclude_callchain_kernel : 1,
/* exclude kernel callchains */
exclude_callchain_user : 1,
/* exclude user callchains */
mmap2 : 1, /* include mmap with inode data */
comm_exec : 1, /* flag comm events that are
due to exec */
use_clockid : 1, /* use clockid for time fields */
context_switch : 1, /* context switch data */
write_backward : 1, /* Write ring buffer from end
to beginning */
namespaces : 1, /* include namespaces data */
ksymbol : 1, /* include ksymbol events */
bpf_event : 1, /* include bpf events */
aux_output : 1, /* generate AUX records
instead of events */
cgroup : 1, /* include cgroup events */
text_poke : 1, /* include text poke events */
build_id : 1, /* use build id in mmap2 events */
inherit_thread : 1, /* children only inherit */
/* if cloned with CLONE_THREAD */
remove_on_exec : 1, /* event is removed from task
on exec */
sigtrap : 1, /* send synchronous SIGTRAP
on event */
__reserved_1 : 26;
union {
__u32 wakeup_events; /* wakeup every n events */
__u32 wakeup_watermark; /* bytes before wakeup */
};
__u32 bp_type; /* breakpoint type */
union {
__u64 bp_addr; /* breakpoint address */
__u64 kprobe_func; /* for perf_kprobe */
__u64 uprobe_path; /* for perf_uprobe */
__u64 config1; /* extension of config */
};
union {
__u64 bp_len; /* breakpoint length */
__u64 kprobe_addr; /* with kprobe_func == NULL */
__u64 probe_offset; /* for perf_[k,u]probe */
__u64 config2; /* extension of config1 */
};
__u64 branch_sample_type; /* enum perf_branch_sample_type */
__u64 sample_regs_user; /* user regs to dump on samples */
__u32 sample_stack_user; /* size of stack to dump on
samples */
__s32 clockid; /* clock to use for time fields */
__u64 sample_regs_intr; /* regs to dump on samples */
__u32 aux_watermark; /* aux bytes before wakeup */
__u16 sample_max_stack; /* max frames in callchain */
__u16 __reserved_2; /* align to u64 */
__u32 aux_sample_size; /* max aux sample size */
__u32 __reserved_3; /* align to u64 */
__u64 sig_data; /* user data for sigtrap */ };

The fields of the perf_event_attr structure are described in more detail below:

This field specifies the overall event type. It has one of the following values:
This indicates one of the "generalized" hardware events provided by the kernel. See the config field definition for more details.
This indicates one of the software-defined events provided by the kernel (even if no hardware support is available).
This indicates a tracepoint provided by the kernel tracepoint infrastructure.
This indicates a hardware cache event. This has a special encoding, described in the config field definition.
This indicates a "raw" implementation-specific event in the config field.
This indicates a hardware breakpoint as provided by the CPU. Breakpoints can be read/write accesses to an address as well as execution of an instruction address.
Since Linux 2.6.38, perf_event_open() can support multiple PMUs. To enable this, a value exported by the kernel can be used in the type field to indicate which PMU to use. The value to use can be found in the sysfs filesystem: there is a subdirectory per PMU instance under /sys/bus/event_source/devices. In each subdirectory there is a type file whose content is an integer that can be used in the type field. For instance, /sys/bus/event_source/devices/cpu/type contains the value for the core CPU PMU, which is usually 4.
These two dynamic PMUs create a kprobe/uprobe and attach it to the file descriptor generated by perf_event_open. The kprobe/uprobe will be destroyed on the destruction of the file descriptor. See fields kprobe_func, uprobe_path, kprobe_addr, and probe_offset for more details.
The size of the perf_event_attr structure for forward/backward compatibility. Set this using sizeof(struct perf_event_attr) to allow the kernel to see the struct size at the time of compilation.
The related define PERF_ATTR_SIZE_VER0 is set to 64; this was the size of the first published struct. PERF_ATTR_SIZE_VER1 is 72, corresponding to the addition of breakpoints in Linux 2.6.33. PERF_ATTR_SIZE_VER2 is 80 corresponding to the addition of branch sampling in Linux 3.4. PERF_ATTR_SIZE_VER3 is 96 corresponding to the addition of sample_regs_user and sample_stack_user in Linux 3.7. PERF_ATTR_SIZE_VER4 is 104 corresponding to the addition of sample_regs_intr in Linux 3.19. PERF_ATTR_SIZE_VER5 is 112 corresponding to the addition of aux_watermark in Linux 4.1.
This specifies which event you want, in conjunction with the type field. The config1 and config2 fields are also taken into account in cases where 64 bits is not enough to fully specify the event. The encoding of these fields are event dependent.
There are various ways to set the config field that are dependent on the value of the previously described type field. What follows are various possible settings for config separated out by type.
If type is PERF_TYPE_HARDWARE, we are measuring one of the generalized hardware CPU events. Not all of these are available on all platforms. Set config to one of the following:
Total cycles. Be wary of what happens during CPU frequency scaling.
Retired instructions. Be careful, these can be affected by various issues, most notably hardware interrupt counts.
Cache accesses. Usually this indicates Last Level Cache accesses but this may vary depending on your CPU. This may include prefetches and coherency messages; again this depends on the design of your CPU.
Cache misses. Usually this indicates Last Level Cache misses; this is intended to be used in conjunction with the PERF_COUNT_HW_CACHE_REFERENCES event to calculate cache miss rates.
Retired branch instructions. Prior to Linux 2.6.35, this used the wrong event on AMD processors.
Mispredicted branch instructions.
Bus cycles, which can be different from total cycles.
Stalled cycles during issue.
Stalled cycles during retirement.
Total cycles; not affected by CPU frequency scaling.
If type is PERF_TYPE_SOFTWARE, we are measuring software events provided by the kernel. Set config to one of the following:
This reports the CPU clock, a high-resolution per-CPU timer.
This reports a clock count specific to the task that is running.
This reports the number of page faults.
This counts context switches. Until Linux 2.6.34, these were all reported as user-space events, after that they are reported as happening in the kernel.
This reports the number of times the process has migrated to a new CPU.
This counts the number of minor page faults. These did not require disk I/O to handle.
This counts the number of major page faults. These required disk I/O to handle.
This counts the number of alignment faults. These happen when unaligned memory accesses happen; the kernel can handle these but it reduces performance. This happens only on some architectures (never on x86).
This counts the number of emulation faults. The kernel sometimes traps on unimplemented instructions and emulates them for user space. This can negatively impact performance.
This is a placeholder event that counts nothing. Informational sample record types such as mmap or comm must be associated with an active event. This dummy event allows gathering such records without requiring a counting event.
This is used to generate raw sample data from BPF. BPF programs can write to this event using bpf_perf_event_output helper.
This counts context switches to a task in a different cgroup. In other words, if the next task is in the same cgroup, it won't count the switch.

If type is PERF_TYPE_TRACEPOINT, then we are measuring kernel tracepoints. The value to use in config can be obtained from under debugfs tracing/events/*/*/id if ftrace is enabled in the kernel.

If type is PERF_TYPE_HW_CACHE, then we are measuring a hardware CPU cache event. To calculate the appropriate config value, use the following equation:


config = (perf_hw_cache_id) |

(perf_hw_cache_op_id << 8) |
(perf_hw_cache_op_result_id << 16);

where perf_hw_cache_id is one of:

for measuring Level 1 Data Cache
for measuring Level 1 Instruction Cache
for measuring Last-Level Cache
for measuring the Data TLB
for measuring the Instruction TLB
for measuring the branch prediction unit
for measuring local memory accesses

and perf_hw_cache_op_id is one of:

for read accesses
for write accesses
for prefetch accesses

and perf_hw_cache_op_result_id is one of:

If type is PERF_TYPE_RAW, then a custom "raw" config value is needed. Most CPUs support events that are not covered by the "generalized" events. These are implementation defined; see your CPU manual (for example the Intel Volume 3B documentation or the AMD BIOS and Kernel Developer Guide). The libpfm4 library can be used to translate from the name in the architectural manuals to the raw hex value perf_event_open() expects in this field.

If type is PERF_TYPE_BREAKPOINT, then leave config set to zero. Its parameters are set in other places.

If type is kprobe or uprobe, set retprobe (bit 0 of config, see /sys/bus/event_source/devices/[k,u]probe/format/retprobe) for kretprobe/uretprobe. See fields kprobe_func, uprobe_path, kprobe_addr, and probe_offset for more details.

These fields describe the kprobe/uprobe for dynamic PMUs kprobe and uprobe. For kprobe: use kprobe_func and probe_offset, or use kprobe_addr and leave kprobe_func as NULL. For uprobe: use uprobe_path and probe_offset.
A "sampling" event is one that generates an overflow notification every N events, where N is given by sample_period. A sampling event has sample_period > 0. When an overflow occurs, requested data is recorded in the mmap buffer. The sample_type field controls what data is recorded on each overflow.
sample_freq can be used if you wish to use frequency rather than period. In this case, you set the freq flag. The kernel will adjust the sampling period to try and achieve the desired rate. The rate of adjustment is a timer tick.
The various bits in this field specify which values to include in the sample. They will be recorded in a ring-buffer, which is available to user space using mmap(2). The order in which the values are saved in the sample are documented in the MMAP Layout subsection below; it is not the enum perf_event_sample_format order.
Records instruction pointer.
Records the process and thread IDs.
Records a timestamp.
Records an address, if applicable.
Record counter values for all events in a group, not just the group leader.
Records the callchain (stack backtrace).
Records a unique ID for the opened event's group leader.
Records CPU number.
Records the current sampling period.
Records a unique ID for the opened event. Unlike PERF_SAMPLE_ID the actual ID is returned, not the group leader. This ID is the same as the one returned by PERF_FORMAT_ID.
Records additional data, if applicable. Usually returned by tracepoint events.
This provides a record of recent branches, as provided by CPU branch sampling hardware (such as Intel Last Branch Record). Not all hardware supports this feature.
See the branch_sample_type field for how to filter which branches are reported.
Records the current user-level CPU register state (the values in the process before the kernel was called).
Records the user level stack, allowing stack unwinding.
Records a hardware provided weight value that expresses how costly the sampled event was. This allows the hardware to highlight expensive events in a profile.
Records the data source: where in the memory hierarchy the data associated with the sampled instruction came from. This is available only if the underlying hardware supports this feature.
Places the SAMPLE_ID value in a fixed position in the record, either at the beginning (for sample events) or at the end (if a non-sample event).
This was necessary because a sample stream may have records from various different event sources with different sample_type settings. Parsing the event stream properly was not possible because the format of the record was needed to find SAMPLE_ID, but the format could not be found without knowing what event the sample belonged to (causing a circular dependency).
The PERF_SAMPLE_IDENTIFIER setting makes the event stream always parsable by putting SAMPLE_ID in a fixed location, even though it means having duplicate SAMPLE_ID values in records.
Records reasons for transactional memory abort events (for example, from Intel TSX transactional memory support).
The precise_ip setting must be greater than 0 and a transactional memory abort event must be measured or no values will be recorded. Also note that some perf_event measurements, such as sampled cycle counting, may cause extraneous aborts (by causing an interrupt during a transaction).
Records a subset of the current CPU register state as specified by sample_regs_intr. Unlike PERF_SAMPLE_REGS_USER the register values will return kernel register state if the overflow happened while kernel code is running. If the CPU supports hardware sampling of register state (i.e., PEBS on Intel x86) and precise_ip is set higher than zero then the register values returned are those captured by hardware at the time of the sampled instruction's retirement.
Records physical address of data like in PERF_SAMPLE_ADDR.
Records (perf_event) cgroup ID of the process. This corresponds to the id field in the PERF_RECORD_CGROUP event.
Records page size of data like in PERF_SAMPLE_ADDR.
Records page size of ip like in PERF_SAMPLE_IP.
Records hardware provided weight values like in PERF_SAMPLE_WEIGHT, but it can represent multiple values in a struct. This shares the same space as PERF_SAMPLE_WEIGHT, so users can apply either of those, not both. It has the following format and the meaning of each field is dependent on the hardware implementation.


union perf_sample_weight {

u64 full; /* PERF_SAMPLE_WEIGHT */
struct { /* PERF_SAMPLE_WEIGHT_STRUCT */
u32 var1_dw;
u16 var2_w;
u16 var3_w;
}; };

This field specifies the format of the data returned by read(2) on a perf_event_open() file descriptor.
Adds the 64-bit time_enabled field. This can be used to calculate estimated totals if the PMU is overcommitted and multiplexing is happening.
Adds the 64-bit time_running field. This can be used to calculate estimated totals if the PMU is overcommitted and multiplexing is happening.
Adds a 64-bit unique value that corresponds to the event group.
Allows all counter values in an event group to be read with one read.
Adds a 64-bit value that is the number of lost samples for this event. This would be only meaningful when sample_period or sample_freq is set.
The disabled bit specifies whether the counter starts out disabled or enabled. If disabled, the event can later be enabled by ioctl(2), prctl(2), or enable_on_exec.
When creating an event group, typically the group leader is initialized with disabled set to 1 and any child events are initialized with disabled set to 0. Despite disabled being 0, the child events will not start until the group leader is enabled.
The inherit bit specifies that this counter should count events of child tasks as well as the task specified. This applies only to new children, not to any existing children at the time the counter is created (nor to any new children of existing children).
Inherit does not work for some combinations of read_format values, such as PERF_FORMAT_GROUP.
The pinned bit specifies that the counter should always be on the CPU if at all possible. It applies only 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(2) returns 0) until the counter is subsequently enabled or disabled.
The exclusive bit specifies that when this counter's group is on the CPU, it should be the only group using the CPU's counters. In the future this may allow monitoring programs to support PMU features that need to run alone so that they do not disrupt other hardware counters.
Note that many unexpected situations may prevent events with the exclusive bit set from ever running. This includes any users running a system-wide measurement as well as any kernel use of the performance counters (including the commonly enabled NMI Watchdog Timer interface).
If this bit is set, the count excludes events that happen in user space.
If this bit is set, the count excludes events that happen in kernel space.
If this bit is set, the count excludes events that happen in the hypervisor. This is mainly for PMUs that have built-in support for handling this (such as POWER). Extra support is needed for handling hypervisor measurements on most machines.
If set, don't count when the CPU is running the idle task. While you can currently enable this for any event type, it is ignored for all but software events.
The mmap bit enables generation of PERF_RECORD_MMAP samples for every mmap(2) call that has PROT_EXEC set. This allows tools to notice new executable code being mapped into a program (dynamic shared libraries for example) so that addresses can be mapped back to the original code.
The comm bit enables tracking of process command name as modified by the execve(2) and prctl(PR_SET_NAME) system calls as well as writing to /proc/self/comm. If the comm_exec flag is also successfully set (possible since Linux 3.16), then the misc flag PERF_RECORD_MISC_COMM_EXEC can be used to differentiate the execve(2) case from the others.
If this bit is set, then sample_frequency not sample_period is used when setting up the sampling interval.
This bit enables saving of event counts on context switch for inherited tasks. This is meaningful only if the inherit field is set.
If this bit is set, a counter is automatically enabled after a call to execve(2).
If this bit is set, then fork/exit notifications are included in the ring buffer.
If set, have an overflow notification happen when we cross the wakeup_watermark boundary. Otherwise, overflow notifications happen after wakeup_events samples.
This controls the amount of skid. Skid is how many instructions execute between an event of interest happening and the kernel being able to stop and record the event. Smaller skid is better and allows more accurate reporting of which events correspond to which instructions, but hardware is often limited with how small this can be.
The possible values of this field are the following:
0
SAMPLE_IP can have arbitrary skid.
1
SAMPLE_IP must have constant skid.
2
SAMPLE_IP requested to have 0 skid.
3
SAMPLE_IP must have 0 skid. See also the description of PERF_RECORD_MISC_EXACT_IP.
This is the counterpart of the mmap field. This enables generation of PERF_RECORD_MMAP samples for mmap(2) calls that do not have PROT_EXEC set (for example data and SysV shared memory).
If set, then TID, TIME, ID, STREAM_ID, and CPU can additionally be included in non-PERF_RECORD_SAMPLEs if the corresponding sample_type is selected.
If PERF_SAMPLE_IDENTIFIER is specified, then an additional ID value is included as the last value to ease parsing the record stream. This may lead to the id value appearing twice.
The layout is described by this pseudo-structure:

struct sample_id {

{ u32 pid, tid; } /* if PERF_SAMPLE_TID set */
{ u64 time; } /* if PERF_SAMPLE_TIME set */
{ u64 id; } /* if PERF_SAMPLE_ID set */
{ u64 stream_id;} /* if PERF_SAMPLE_STREAM_ID set */
{ u32 cpu, res; } /* if PERF_SAMPLE_CPU set */
{ u64 id; } /* if PERF_SAMPLE_IDENTIFIER set */ };

When conducting measurements that include processes running VM instances (i.e., have executed a KVM_RUN ioctl(2)), only measure events happening inside a guest instance. This is only meaningful outside the guests; this setting does not change counts gathered inside of a guest. Currently, this functionality is x86 only.
When conducting measurements that include processes running VM instances (i.e., have executed a KVM_RUN ioctl(2)), do not measure events happening inside guest instances. This is only meaningful outside the guests; this setting does not change counts gathered inside of a guest. Currently, this functionality is x86 only.
Do not include kernel callchains.
Do not include user callchains.
Generate an extended executable mmap record that contains enough additional information to uniquely identify shared mappings. The mmap flag must also be set for this to work.
This is purely a feature-detection flag, it does not change kernel behavior. If this flag can successfully be set, then, when comm is enabled, the PERF_RECORD_MISC_COMM_EXEC flag will be set in the misc field of a comm record header if the rename event being reported was caused by a call to execve(2). This allows tools to distinguish between the various types of process renaming.
This allows selecting which internal Linux clock to use when generating timestamps via the clockid field. This can make it easier to correlate perf sample times with timestamps generated by other tools.
This enables the generation of PERF_RECORD_SWITCH records when a context switch occurs. It also enables the generation of PERF_RECORD_SWITCH_CPU_WIDE records when sampling in CPU-wide mode. This functionality is in addition to existing tracepoint and software events for measuring context switches. The advantage of this method is that it will give full information even with strict perf_event_paranoid settings.
This causes the ring buffer to be written from the end to the beginning. This is to support reading from overwritable ring buffer.
This enables the generation of PERF_RECORD_NAMESPACES records when a task enters a new namespace. Each namespace has a combination of device and inode numbers.
This enables the generation of PERF_RECORD_KSYMBOL records when new kernel symbols are registered or unregistered. This is analyzing dynamic kernel functions like eBPF.
This enables the generation of PERF_RECORD_BPF_EVENT records when an eBPF program is loaded or unloaded.
This allows normal (non-AUX) events to generate data for AUX events if the hardware supports it.
This enables the generation of PERF_RECORD_CGROUP records when a new cgroup is created (and activated).
This enables the generation of PERF_RECORD_TEXT_POKE records when there's a change to the kernel text (i.e., self-modifying code).
This changes the contents in the PERF_RECORD_MMAP2 to have a build-id instead of device and inode numbers.
This disables the inheritance of the event to a child process. Only new threads in the same process (which is cloned with CLONE_THREAD) will inherit the event.
This closes the event when it starts a new process image by execve(2).
This enables synchronous signal delivery of SIGTRAP on event overflow.
This union sets how many samples (wakeup_events) or bytes (wakeup_watermark) happen before an overflow notification happens. Which one is used is selected by the watermark bit flag.
wakeup_events counts only PERF_RECORD_SAMPLE record types. To receive overflow notification for all PERF_RECORD types choose watermark and set wakeup_watermark to 1.
Prior to Linux 3.0, setting wakeup_events to 0 resulted in no overflow notifications; more recent kernels treat 0 the same as 1.
This chooses the breakpoint type. It is one of:
No breakpoint.
Count when we read the memory location.
Count when we write the memory location.
Count when we read or write the memory location.
Count when we execute code at the memory location.

The values can be combined via a bitwise or, but the combination of HW_BREAKPOINT_R or HW_BREAKPOINT_W with HW_BREAKPOINT_X is not allowed.

This is the address of the breakpoint. For execution breakpoints, this is the memory address of the instruction of interest; for read and write breakpoints, it is the memory address of the memory location of interest.
config1 is used for setting events that need an extra register or otherwise do not fit in the regular config field. Raw OFFCORE_EVENTS on Nehalem/Westmere/SandyBridge use this field on Linux 3.3 and later kernels.
bp_len is the length of the breakpoint being measured if type is PERF_TYPE_BREAKPOINT. Options are HW_BREAKPOINT_LEN_1, HW_BREAKPOINT_LEN_2, HW_BREAKPOINT_LEN_4, and HW_BREAKPOINT_LEN_8. For an execution breakpoint, set this to sizeof(long).
config2 is a further extension of the config1 field.
If PERF_SAMPLE_BRANCH_STACK is enabled, then this specifies what branches to include in the branch record.
The first part of the value is the privilege level, which is a combination of one of the values listed below. If the user does not set privilege level explicitly, the kernel will use the event's privilege level. Event and branch privilege levels do not have to match.
Branch target is in user space.
Branch target is in kernel space.
Branch target is in hypervisor.
A convenience value that is the three preceding values ORed together.

In addition to the privilege value, at least one or more of the following bits must be set.

Any branch type.
Any call branch (includes direct calls, indirect calls, and far jumps).
Indirect calls.
Direct calls.
Any return branch.
Indirect jumps.
Conditional branches.
Transactional memory aborts.
Branch in transactional memory transaction.
Branch not in transactional memory transaction. PERF_SAMPLE_BRANCH_CALL_STACK (since Linux 4.1) Branch is part of a hardware-generated call stack. This requires hardware support, currently only found on Intel x86 Haswell or newer.
This bit mask defines the set of user CPU registers to dump on samples. The layout of the register mask is architecture-specific and is described in the kernel header file arch/ARCH/include/uapi/asm/perf_regs.h.
This defines the size of the user stack to dump if PERF_SAMPLE_STACK_USER is specified.
If use_clockid is set, then this field selects which internal Linux timer to use for timestamps. The available timers are defined in linux/time.h, with CLOCK_MONOTONIC, CLOCK_MONOTONIC_RAW, CLOCK_REALTIME, CLOCK_BOOTTIME, and CLOCK_TAI currently supported.
This specifies how much data is required to trigger a PERF_RECORD_AUX sample.
When sample_type includes PERF_SAMPLE_CALLCHAIN, this field specifies how many stack frames to report when generating the callchain.
When PERF_SAMPLE_AUX flag is set, specify the desired size of AUX data. Note that it can get smaller data than the specified size.
This data will be copied to user's signal handler (through si_perf in the siginfo_t) to disambiguate which event triggered the signal.

Reading results

Once a perf_event_open() file descriptor has been opened, the values of the events can be read from the file descriptor. The values that are there are specified by the read_format field in the attr structure at open time.

If you attempt to read into a buffer that is not big enough to hold the data, the error ENOSPC results.

Here is the layout of the data returned by a read:

If PERF_FORMAT_GROUP was specified to allow reading all events in a group at once:

struct read_format {

u64 nr; /* The number of events */
u64 time_enabled; /* if PERF_FORMAT_TOTAL_TIME_ENABLED */
u64 time_running; /* if PERF_FORMAT_TOTAL_TIME_RUNNING */
struct {
u64 value; /* The value of the event */
u64 id; /* if PERF_FORMAT_ID */
u64 lost; /* if PERF_FORMAT_LOST */
} values[nr]; };

If PERF_FORMAT_GROUP was not specified:

struct read_format {

u64 value; /* The value of the event */
u64 time_enabled; /* if PERF_FORMAT_TOTAL_TIME_ENABLED */
u64 time_running; /* if PERF_FORMAT_TOTAL_TIME_RUNNING */
u64 id; /* if PERF_FORMAT_ID */
u64 lost; /* if PERF_FORMAT_LOST */ };

The values read are as follows:

The number of events in this file descriptor. Available only if PERF_FORMAT_GROUP was specified.
Total time the event was enabled and running. Normally these values are the same. Multiplexing happens if the number of events is more than the number of available PMU counter slots. In that case the events run only part of the time and the time_enabled and time running values can be used to scale an estimated value for the count.
An unsigned 64-bit value containing the counter result.
A globally unique value for this particular event; only present if PERF_FORMAT_ID was specified in read_format.
The number of lost samples of this event; only present if PERF_FORMAT_LOST was specified in read_format.

MMAP layout

When using perf_event_open() in sampled mode, asynchronous events (like counter overflow or PROT_EXEC mmap tracking) are logged into a ring-buffer. This ring-buffer is created and accessed through mmap(2).

The mmap size should be 1+2^n pages, where the first page is a metadata page (struct perf_event_mmap_page) that contains various bits of information such as where the ring-buffer head is.

Before Linux 2.6.39, there is a bug that means you must allocate an mmap ring buffer when sampling even if you do not plan to access it.

The structure of the first metadata mmap page is as follows:


struct perf_event_mmap_page {

__u32 version; /* version number of this structure */
__u32 compat_version; /* lowest version this is compat with */
__u32 lock; /* seqlock for synchronization */
__u32 index; /* hardware counter identifier */
__s64 offset; /* add to hardware counter value */
__u64 time_enabled; /* time event active */
__u64 time_running; /* time event on CPU */
union {
__u64 capabilities;
struct {
__u64 cap_usr_time / cap_usr_rdpmc / cap_bit0 : 1,
cap_bit0_is_deprecated : 1,
cap_user_rdpmc : 1,
cap_user_time : 1,
cap_user_time_zero : 1,
};
};
__u16 pmc_width;
__u16 time_shift;
__u32 time_mult;
__u64 time_offset;
__u64 __reserved[120]; /* Pad to 1 k */
__u64 data_head; /* head in the data section */
__u64 data_tail; /* user-space written tail */
__u64 data_offset; /* where the buffer starts */
__u64 data_size; /* data buffer size */
__u64 aux_head;
__u64 aux_tail;
__u64 aux_offset;
__u64 aux_size; }

The following list describes the fields in the perf_event_mmap_page structure in more detail:

Version number of this structure.
The lowest version this is compatible with.
A seqlock for synchronization.
A unique hardware counter identifier.
When using rdpmc for reads this offset value must be added to the one returned by rdpmc to get the current total event count.
Time the event was active.
Time the event was running.
There was a bug in the definition of cap_usr_time and cap_usr_rdpmc from Linux 3.4 until Linux 3.11. Both bits were defined to point to the same location, so it was impossible to know if cap_usr_time or cap_usr_rdpmc were actually set.
Starting with Linux 3.12, these are renamed to cap_bit0 and you should use the cap_user_time and cap_user_rdpmc fields instead.
If set, this bit indicates that the kernel supports the properly separated cap_user_time and cap_user_rdpmc bits.
If not-set, it indicates an older kernel where cap_usr_time and cap_usr_rdpmc map to the same bit and thus both features should be used with caution.
If the hardware supports user-space read of performance counters without syscall (this is the "rdpmc" instruction on x86), then the following code can be used to do a read:

u32 seq, time_mult, time_shift, idx, width;
u64 count, enabled, running;
u64 cyc, time_offset;
do {

seq = pc->lock;
barrier();
enabled = pc->time_enabled;
running = pc->time_running;
if (pc->cap_usr_time && enabled != running) {
cyc = rdtsc();
time_offset = pc->time_offset;
time_mult = pc->time_mult;
time_shift = pc->time_shift;
}
idx = pc->index;
count = pc->offset;
if (pc->cap_usr_rdpmc && idx) {
width = pc->pmc_width;
count += rdpmc(idx - 1);
}
barrier(); } while (pc->lock != seq);

This bit indicates the hardware has a constant, nonstop timestamp counter (TSC on x86).
Indicates the presence of time_zero which allows mapping timestamp values to the hardware clock.
If cap_usr_rdpmc, this field provides the bit-width of the value read using the rdpmc or equivalent instruction. This can be used to sign extend the result like:

pmc <<= 64 - pmc_width;
pmc >>= 64 - pmc_width; // signed shift right
count += pmc;
    

If cap_usr_time, these fields can be used to compute the time delta since time_enabled (in nanoseconds) using rdtsc or similar.

u64 quot, rem;
u64 delta;
quot  = cyc >> time_shift;
rem   = cyc & (((u64)1 << time_shift) - 1);
delta = time_offset + quot * time_mult +

((rem * time_mult) >> time_shift);

Where time_offset, time_mult, time_shift, and cyc are read in the seqcount loop described above. This delta can then be added to enabled and possible running (if idx), improving the scaling:

enabled += delta;
if (idx)

running += delta; quot = count / running; rem = count % running; count = quot * enabled + (rem * enabled) / running;

If cap_usr_time_zero is set, then the hardware clock (the TSC timestamp counter on x86) can be calculated from the time_zero, time_mult, and time_shift values:

time = timestamp - time_zero;
quot = time / time_mult;
rem  = time % time_mult;
cyc  = (quot << time_shift) + (rem << time_shift) / time_mult;
    

And vice versa:

quot = cyc >> time_shift;
rem  = cyc & (((u64)1 << time_shift) - 1);
timestamp = time_zero + quot * time_mult +

((rem * time_mult) >> time_shift);

This points to the head of the data section. The value continuously increases, it does not wrap. The value needs to be manually wrapped by the size of the mmap buffer before accessing the samples.
On SMP-capable platforms, after reading the data_head value, user space should issue an rmb().
When the mapping is PROT_WRITE, the data_tail value should be written by user space to reflect the last read data. In this case, the kernel will not overwrite unread data.
Contains the offset of the location in the mmap buffer where perf sample data begins.
Contains the size of the perf sample region within the mmap buffer.
The AUX region allows mmap(2)-ing a separate sample buffer for high-bandwidth data streams (separate from the main perf sample buffer). An example of a high-bandwidth stream is instruction tracing support, as is found in newer Intel processors.
To set up an AUX area, first aux_offset needs to be set with an offset greater than data_offset+data_size and aux_size needs to be set to the desired buffer size. The desired offset and size must be page aligned, and the size must be a power of two. These values are then passed to mmap in order to map the AUX buffer. Pages in the AUX buffer are included as part of the RLIMIT_MEMLOCK resource limit (see setrlimit(2)), and also as part of the perf_event_mlock_kb allowance.
By default, the AUX buffer will be truncated if it will not fit in the available space in the ring buffer. If the AUX buffer is mapped as a read only buffer, then it will operate in ring buffer mode where old data will be overwritten by new. In overwrite mode, it might not be possible to infer where the new data began, and it is the consumer's job to disable measurement while reading to avoid possible data races.
The aux_head and aux_tail ring buffer pointers have the same behavior and ordering rules as the previous described data_head and data_tail.

The following 2^n ring-buffer pages have the layout described below.

If perf_event_attr.sample_id_all is set, then all event types will have the sample_type selected fields related to where/when (identity) an event took place (TID, TIME, ID, CPU, STREAM_ID) described in PERF_RECORD_SAMPLE below, it will be stashed just after the perf_event_header and the fields already present for the existing fields, that is, at the end of the payload. This allows a newer perf.data file to be supported by older perf tools, with the new optional fields being ignored.

The mmap values start with a header:


struct perf_event_header {

__u32 type;
__u16 misc;
__u16 size; };

Below, we describe the perf_event_header fields in more detail. For ease of reading, the fields with shorter descriptions are presented first.

This indicates the size of the record.
The misc field contains additional information about the sample.
The CPU mode can be determined from this value by masking with PERF_RECORD_MISC_CPUMODE_MASK and looking for one of the following (note these are not bit masks, only one can be set at a time):
Unknown CPU mode.
Sample happened in the kernel.
Sample happened in user code.
Sample happened in the hypervisor.
Sample happened in the guest kernel.
Sample happened in guest user code.

Since the following three statuses are generated by different record types, they alias to the same bit:
This is set when the mapping is not executable; otherwise the mapping is executable.
This is set for a PERF_RECORD_COMM record on kernels more recent than Linux 3.16 if a process name change was caused by an execve(2) system call.
When a PERF_RECORD_SWITCH or PERF_RECORD_SWITCH_CPU_WIDE record is generated, this bit indicates that the context switch is away from the current process (instead of into the current process).

In addition, the following bits can be set:
This indicates that the content of PERF_SAMPLE_IP points to the actual instruction that triggered the event. See also perf_event_attr.precise_ip.
When a PERF_RECORD_SWITCH or PERF_RECORD_SWITCH_CPU_WIDE record is generated, this indicates the context switch was a preemption.
This indicates that the content of PERF_SAMPLE_MMAP2 contains build-ID data instead of device major and minor numbers as well as the inode number.
This indicates there is extended data available (currently not used).
This bit is not set by the kernel. It is reserved for the user-space perf utility to indicate that /proc/pid/maps parsing was taking too long and was stopped, and thus the mmap records may be truncated.
The type value is one of the below. The values in the corresponding record (that follows the header) depend on the type selected as shown.
The MMAP events record the PROT_EXEC mappings so that we can correlate user-space IPs to code. They have the following structure:

struct {

struct perf_event_header header;
u32 pid, tid;
u64 addr;
u64 len;
u64 pgoff;
char filename[]; };

is the process ID.
is the thread ID.
is the address of the allocated memory. len is the length of the allocated memory. pgoff is the page offset of the allocated memory. filename is a string describing the backing of the allocated memory.
This record indicates when events are lost.

struct {

struct perf_event_header header;
u64 id;
u64 lost;
struct sample_id sample_id; };

is the unique event ID for the samples that were lost.
is the number of events that were lost.
This record indicates a change in the process name.

struct {

struct perf_event_header header;
u32 pid;
u32 tid;
char comm[];
struct sample_id sample_id; };

is the process ID.
is the thread ID.
is a string containing the new name of the process.
This record indicates a process exit event.

struct {

struct perf_event_header header;
u32 pid, ppid;
u32 tid, ptid;
u64 time;
struct sample_id sample_id; };

This record indicates a throttle/unthrottle event.

struct {

struct perf_event_header header;
u64 time;
u64 id;
u64 stream_id;
struct sample_id sample_id; };

This record indicates a fork event.

struct {

struct perf_event_header header;
u32 pid, ppid;
u32 tid, ptid;
u64 time;
struct sample_id sample_id; };

This record indicates a read event.

struct {

struct perf_event_header header;
u32 pid, tid;
struct read_format values;
struct sample_id sample_id; };

This record indicates a sample.

struct {

struct perf_event_header header;
u64 sample_id; /* if PERF_SAMPLE_IDENTIFIER */
u64 ip; /* if PERF_SAMPLE_IP */
u32 pid, tid; /* if PERF_SAMPLE_TID */
u64 time; /* if PERF_SAMPLE_TIME */
u64 addr; /* if PERF_SAMPLE_ADDR */
u64 id; /* if PERF_SAMPLE_ID */
u64 stream_id; /* if PERF_SAMPLE_STREAM_ID */
u32 cpu, res; /* if PERF_SAMPLE_CPU */
u64 period; /* if PERF_SAMPLE_PERIOD */
struct read_format v;
/* if PERF_SAMPLE_READ */
u64 nr; /* if PERF_SAMPLE_CALLCHAIN */
u64 ips[nr]; /* if PERF_SAMPLE_CALLCHAIN */
u32 size; /* if PERF_SAMPLE_RAW */
char data[size]; /* if PERF_SAMPLE_RAW */
u64 bnr; /* if PERF_SAMPLE_BRANCH_STACK */
struct perf_branch_entry lbr[bnr];
/* if PERF_SAMPLE_BRANCH_STACK */
u64 abi; /* if PERF_SAMPLE_REGS_USER */
u64 regs[weight(mask)];
/* if PERF_SAMPLE_REGS_USER */
u64 size; /* if PERF_SAMPLE_STACK_USER */
char data[size]; /* if PERF_SAMPLE_STACK_USER */
u64 dyn_size; /* if PERF_SAMPLE_STACK_USER &&
size != 0 */
union perf_sample_weight weight;
/* if PERF_SAMPLE_WEIGHT */
/* || PERF_SAMPLE_WEIGHT_STRUCT */
u64 data_src; /* if PERF_SAMPLE_DATA_SRC */
u64 transaction; /* if PERF_SAMPLE_TRANSACTION */
u64 abi; /* if PERF_SAMPLE_REGS_INTR */
u64 regs[weight(mask)];
/* if PERF_SAMPLE_REGS_INTR */
u64 phys_addr; /* if PERF_SAMPLE_PHYS_ADDR */
u64 cgroup; /* if PERF_SAMPLE_CGROUP */
u64 data_page_size;
/* if PERF_SAMPLE_DATA_PAGE_SIZE */
u64 code_page_size;
/* if PERF_SAMPLE_CODE_PAGE_SIZE */
u64 size; /* if PERF_SAMPLE_AUX */
char data[size]; /* if PERF_SAMPLE_AUX */ };

If PERF_SAMPLE_IDENTIFIER is enabled, a 64-bit unique ID is included. This is a duplication of the PERF_SAMPLE_ID id value, but included at the beginning of the sample so parsers can easily obtain the value.
If PERF_SAMPLE_IP is enabled, then a 64-bit instruction pointer value is included.
If PERF_SAMPLE_TID is enabled, then a 32-bit process ID and 32-bit thread ID are included.
If PERF_SAMPLE_TIME is enabled, then a 64-bit timestamp is included. This is obtained via local_clock() which is a hardware timestamp if available and the jiffies value if not.
If PERF_SAMPLE_ADDR is enabled, then a 64-bit address is included. This is usually the address of a tracepoint, breakpoint, or software event; otherwise the value is 0.
If PERF_SAMPLE_ID is enabled, a 64-bit unique ID is included. If the event is a member of an event group, the group leader ID is returned. This ID is the same as the one returned by PERF_FORMAT_ID.
If PERF_SAMPLE_STREAM_ID is enabled, a 64-bit unique ID is included. Unlike PERF_SAMPLE_ID the actual ID is returned, not the group leader. This ID is the same as the one returned by PERF_FORMAT_ID.
If PERF_SAMPLE_CPU is enabled, this is a 32-bit value indicating which CPU was being used, in addition to a reserved (unused) 32-bit value.
If PERF_SAMPLE_PERIOD is enabled, a 64-bit value indicating the current sampling period is written.
If PERF_SAMPLE_READ is enabled, a structure of type read_format is included which has values for all events in the event group. The values included depend on the read_format value used at perf_event_open() time.
If PERF_SAMPLE_CALLCHAIN is enabled, then a 64-bit number is included which indicates how many following 64-bit instruction pointers will follow. This is the current callchain.
If PERF_SAMPLE_RAW is enabled, then a 32-bit value indicating size is included followed by an array of 8-bit values of length size. The values are padded with 0 to have 64-bit alignment.
This RAW record data is opaque with respect to the ABI. The ABI doesn't make any promises with respect to the stability of its content, it may vary depending on event, hardware, and kernel version.
If PERF_SAMPLE_BRANCH_STACK is enabled, then a 64-bit value indicating the number of records is included, followed by bnr perf_branch_entry structures which each include the fields:
This indicates the source instruction (may not be a branch).
The branch target.
The branch target was mispredicted.
The branch target was predicted.
The branch was in a transactional memory transaction.
The branch was in an aborted transactional memory transaction.
This reports the number of cycles elapsed since the previous branch stack update.

The entries are from most to least recent, so the first entry has the most recent branch.

Support for mispred, predicted, and cycles is optional; if not supported, those values will be 0.

The type of branches recorded is specified by the branch_sample_type field.

If PERF_SAMPLE_REGS_USER is enabled, then the user CPU registers are recorded.
The abi field is one of PERF_SAMPLE_REGS_ABI_NONE, PERF_SAMPLE_REGS_ABI_32, or PERF_SAMPLE_REGS_ABI_64.
The regs field is an array of the CPU registers that were specified by the sample_regs_user attr field. The number of values is the number of bits set in the sample_regs_user bit mask.
If PERF_SAMPLE_STACK_USER is enabled, then the user stack is recorded. This can be used to generate stack backtraces. size is the size requested by the user in sample_stack_user or else the maximum record size. data is the stack data (a raw dump of the memory pointed to by the stack pointer at the time of sampling). dyn_size is the amount of data actually dumped (can be less than size). Note that dyn_size is omitted if size is 0.
If PERF_SAMPLE_WEIGHT or PERF_SAMPLE_WEIGHT_STRUCT is enabled, then a 64-bit value provided by the hardware is recorded that indicates how costly the event was. This allows expensive events to stand out more clearly in profiles.
If PERF_SAMPLE_DATA_SRC is enabled, then a 64-bit value is recorded that is made up of the following fields:
Type of opcode, a bitwise combination of:
Not available
Load instruction
Store instruction
Prefetch
Executable code
Memory hierarchy level hit or miss, a bitwise combination of the following, shifted left by PERF_MEM_LVL_SHIFT:
Not available
Hit
Miss
Level 1 cache
Line fill buffer
Level 2 cache
Level 3 cache
Local DRAM
Remote DRAM 1 hop
Remote DRAM 2 hops
Remote cache 1 hop
Remote cache 2 hops
I/O memory
Uncached memory
Snoop mode, a bitwise combination of the following, shifted left by PERF_MEM_SNOOP_SHIFT:
Not available
No snoop
Snoop hit
Snoop miss
Snoop hit modified
Lock instruction, a bitwise combination of the following, shifted left by PERF_MEM_LOCK_SHIFT:
Not available
Locked transaction
TLB access hit or miss, a bitwise combination of the following, shifted left by PERF_MEM_TLB_SHIFT:
Not available
Hit
Miss
Level 1 TLB
Level 2 TLB
Hardware walker
OS fault handler
If the PERF_SAMPLE_TRANSACTION flag is set, then a 64-bit field is recorded describing the sources of any transactional memory aborts.
The field is a bitwise combination of the following values:
Abort from an elision type transaction (Intel-CPU-specific).
Abort from a generic transaction.
Synchronous abort (related to the reported instruction).
Asynchronous abort (not related to the reported instruction).
Retryable abort (retrying the transaction may have succeeded).
Abort due to memory conflicts with other threads.
Abort due to write capacity overflow.
Abort due to read capacity overflow.
In addition, a user-specified abort code can be obtained from the high 32 bits of the field by shifting right by PERF_TXN_ABORT_SHIFT and masking with the value PERF_TXN_ABORT_MASK.
If PERF_SAMPLE_REGS_INTR is enabled, then the user CPU registers are recorded.
The abi field is one of PERF_SAMPLE_REGS_ABI_NONE, PERF_SAMPLE_REGS_ABI_32, or PERF_SAMPLE_REGS_ABI_64.
The regs field is an array of the CPU registers that were specified by the sample_regs_intr attr field. The number of values is the number of bits set in the sample_regs_intr bit mask.
If the PERF_SAMPLE_PHYS_ADDR flag is set, then the 64-bit physical address is recorded.
If the PERF_SAMPLE_CGROUP flag is set, then the 64-bit cgroup ID (for the perf_event subsystem) is recorded. To get the pathname of the cgroup, the ID should match to one in a PERF_RECORD_CGROUP.
If the PERF_SAMPLE_DATA_PAGE_SIZE flag is set, then the 64-bit page size value of the data address is recorded.
If the PERF_SAMPLE_CODE_PAGE_SIZE flag is set, then the 64-bit page size value of the ip address is recorded.
If PERF_SAMPLE_AUX is enabled, a snapshot of the aux buffer is recorded.
This record includes extended information on mmap(2) calls returning executable mappings. The format is similar to that of the PERF_RECORD_MMAP record, but includes extra values that allow uniquely identifying shared mappings. Depending on the PERF_RECORD_MISC_MMAP_BUILD_ID bit in the header, the extra values have different layout and meanings.

struct {

struct perf_event_header header;
u32 pid;
u32 tid;
u64 addr;
u64 len;
u64 pgoff;
union {
struct {
u32 maj;
u32 min;
u64 ino;
u64 ino_generation;
};
struct { /* if PERF_RECORD_MISC_MMAP_BUILD_ID */
u8 build_id_size;
u8 __reserved_1;
u16 __reserved_2;
u8 build_id[20];
};
};
u32 prot;
u32 flags;
char filename[];
struct sample_id sample_id; };

is the process ID.
is the thread ID.
is the address of the allocated memory.
is the length of the allocated memory.
is the page offset of the allocated memory.
is the major ID of the underlying device.
is the minor ID of the underlying device.
is the inode number.
is the inode generation.
is the actual size of build_id field (up to 20).
is a raw data to identify a binary.
is the protection information.
is the flags information.
is a string describing the backing of the allocated memory.
This record reports that new data is available in the separate AUX buffer region.

struct {

struct perf_event_header header;
u64 aux_offset;
u64 aux_size;
u64 flags;
struct sample_id sample_id; };

offset in the AUX mmap region where the new data begins.
size of the data made available.
describes the AUX update.
if set, then the data returned was truncated to fit the available buffer size.
if set, then the data returned has overwritten previous data.
This record indicates which process has initiated an instruction trace event, allowing tools to properly correlate the instruction addresses in the AUX buffer with the proper executable.

struct {

struct perf_event_header header;
u32 pid;
u32 tid; };

process ID of the thread starting an instruction trace.
thread ID of the thread starting an instruction trace.
When using hardware sampling (such as Intel PEBS) this record indicates some number of samples that may have been lost.

struct {

struct perf_event_header header;
u64 lost;
struct sample_id sample_id; };

the number of potentially lost samples.
This record indicates a context switch has happened. The PERF_RECORD_MISC_SWITCH_OUT bit in the misc field indicates whether it was a context switch into or away from the current process.

struct {

struct perf_event_header header;
struct sample_id sample_id; };

As with PERF_RECORD_SWITCH this record indicates a context switch has happened, but it only occurs when sampling in CPU-wide mode and provides additional information on the process being switched to/from. The PERF_RECORD_MISC_SWITCH_OUT bit in the misc field indicates whether it was a context switch into or away from the current process.

struct {

struct perf_event_header header;
u32 next_prev_pid;
u32 next_prev_tid;
struct sample_id sample_id; };

The process ID of the previous (if switching in) or next (if switching out) process on the CPU.
The thread ID of the previous (if switching in) or next (if switching out) thread on the CPU.
This record includes various namespace information of a process.

struct {

struct perf_event_header header;
u32 pid;
u32 tid;
u64 nr_namespaces;
struct { u64 dev, inode } [nr_namespaces];
struct sample_id sample_id; };

is the process ID
is the thread ID
is the number of namespaces in this record
Each namespace has dev and inode fields and is recorded in the fixed position like below:
Network namespace
UTS namespace
IPC namespace
PID namespace
User namespace
Mount namespace
Cgroup namespace
This record indicates kernel symbol register/unregister events.

struct {

struct perf_event_header header;
u64 addr;
u32 len;
u16 ksym_type;
u16 flags;
char name[];
struct sample_id sample_id; };

is the address of the kernel symbol.
is the length of the kernel symbol.
is the type of the kernel symbol. Currently the following types are available:
The kernel symbol is a BPF function.
If the PERF_RECORD_KSYMBOL_FLAGS_UNREGISTER is set, then this event is for unregistering the kernel symbol.
This record indicates BPF program is loaded or unloaded.

struct {

struct perf_event_header header;
u16 type;
u16 flags;
u32 id;
u8 tag[BPF_TAG_SIZE];
struct sample_id sample_id; };

is one of the following values:
A BPF program is loaded
A BPF program is unloaded
is the ID of the BPF program.
is the tag of the BPF program. Currently, BPF_TAG_SIZE is defined as 8.
This record indicates a new cgroup is created and activated.

struct {

struct perf_event_header header;
u64 id;
char path[];
struct sample_id sample_id; };

is the cgroup identifier. This can be also retrieved by name_to_handle_at(2) on the cgroup path (as a file handle).
is the path of the cgroup from the root.
This record indicates a change in the kernel text. This includes addition and removal of the text and the corresponding length is zero in this case.

struct {

struct perf_event_header header;
u64 addr;
u16 old_len;
u16 new_len;
u8 bytes[];
struct sample_id sample_id; };

is the address of the change
is the old length
is the new length
contains old bytes immediately followed by new bytes.

Overflow handling

Events can be set to notify when a threshold is crossed, indicating an overflow. Overflow conditions can be captured by monitoring the event file descriptor with poll(2), select(2), or epoll(7). Alternatively, the overflow events can be captured via sa signal handler, by enabling I/O signaling on the file descriptor; see the discussion of the F_SETOWN and F_SETSIG operations in fcntl(2).

Overflows are generated only by sampling events (sample_period must have a nonzero value).

There are two ways to generate overflow notifications.

The first is to set a wakeup_events or wakeup_watermark value that will trigger if a certain number of samples or bytes have been written to the mmap ring buffer. In this case, POLL_IN is indicated.

The other way is by use of the PERF_EVENT_IOC_REFRESH ioctl. This ioctl adds to a counter that decrements each time the event overflows. When nonzero, POLL_IN is indicated, but once the counter reaches 0 POLL_HUP is indicated and the underlying event is disabled.

Refreshing an event group leader refreshes all siblings and refreshing with a parameter of 0 currently enables infinite refreshes; these behaviors are unsupported and should not be relied on.

Starting with Linux 3.18, POLL_HUP is indicated if the event being monitored is attached to a different process and that process exits.

rdpmc instruction

Starting with Linux 3.4 on x86, you can use the rdpmc instruction to get low-latency reads without having to enter the kernel. Note that using rdpmc is not necessarily faster than other methods for reading event values.

Support for this can be detected with the cap_usr_rdpmc field in the mmap page; documentation on how to calculate event values can be found in that section.

Originally, when rdpmc support was enabled, any process (not just ones with an active perf event) could use the rdpmc instruction to access the counters. Starting with Linux 4.0, rdpmc support is only allowed if an event is currently enabled in a process's context. To restore the old behavior, write the value 2 to /sys/devices/cpu/rdpmc.

perf_event ioctl calls

Various ioctls act on perf_event_open() file descriptors:

This enables the individual event or event group specified by the file descriptor argument.
If the PERF_IOC_FLAG_GROUP bit is set in the ioctl argument, then all events in a group are enabled, even if the event specified is not the group leader (but see BUGS).
This disables the individual counter or event group specified by the file descriptor argument.
Enabling or disabling the leader of a group enables or disables the entire 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 affects only that counter; disabling a non-leader stops that counter from counting but doesn't affect any other counter.
If the PERF_IOC_FLAG_GROUP bit is set in the ioctl argument, then all events in a group are disabled, even if the event specified is not the group leader (but see BUGS).
Non-inherited overflow counters can use this to enable a counter for a number of overflows specified by the argument, after which it is disabled. Subsequent calls of this ioctl add the argument value to the current count. An overflow notification with POLL_IN set will happen on each overflow until the count reaches 0; when that happens a notification with POLL_HUP set is sent and the event is disabled. Using an argument of 0 is considered undefined behavior.
Reset the event count specified by the file descriptor argument to zero. This resets only the counts; there is no way to reset the multiplexing time_enabled or time_running values.
If the PERF_IOC_FLAG_GROUP bit is set in the ioctl argument, then all events in a group are reset, even if the event specified is not the group leader (but see BUGS).
This updates the overflow period for the event.
Since Linux 3.7 (on ARM) and Linux 3.14 (all other architectures), the new period takes effect immediately. On older kernels, the new period did not take effect until after the next overflow.
The argument is a pointer to a 64-bit value containing the desired new period.
Prior to Linux 2.6.36, this ioctl always failed due to a bug in the kernel.
This tells the kernel to report event notifications to the specified file descriptor rather than the default one. The file descriptors must all be on the same CPU.
The argument specifies the desired file descriptor, or -1 if output should be ignored.
This adds an ftrace filter to this event.
The argument is a pointer to the desired ftrace filter.
This returns the event ID value for the given event file descriptor.
The argument is a pointer to a 64-bit unsigned integer to hold the result.
This allows attaching a Berkeley Packet Filter (BPF) program to an existing kprobe tracepoint event. You need CAP_PERFMON (since Linux 5.8) or CAP_SYS_ADMIN privileges to use this ioctl.
The argument is a BPF program file descriptor that was created by a previous bpf(2) system call.
This allows pausing and resuming the event's ring-buffer. A paused ring-buffer does not prevent generation of samples, but simply discards them. The discarded samples are considered lost, and cause a PERF_RECORD_LOST sample to be generated when possible. An overflow signal may still be triggered by the discarded sample even though the ring-buffer remains empty.
The argument is an unsigned 32-bit integer. A nonzero value pauses the ring-buffer, while a zero value resumes the ring-buffer.
This allows modifying an existing event without the overhead of closing and reopening a new event. Currently this is supported only for breakpoint events.
The argument is a pointer to a perf_event_attr structure containing the updated event settings.
This allows querying which Berkeley Packet Filter (BPF) programs are attached to an existing kprobe tracepoint. You can only attach one BPF program per event, but you can have multiple events attached to a tracepoint. Querying this value on one tracepoint event returns the ID of all BPF programs in all events attached to the tracepoint. You need CAP_PERFMON (since Linux 5.8) or CAP_SYS_ADMIN privileges to use this ioctl.
The argument is a pointer to a structure
struct perf_event_query_bpf {

__u32 ids_len;
__u32 prog_cnt;
__u32 ids[0]; };

The ids_len field indicates the number of ids that can fit in the provided ids array. The prog_cnt value is filled in by the kernel with the number of attached BPF programs. The ids array is filled with the ID of each attached BPF program. If there are more programs than will fit in the array, then the kernel will return ENOSPC and ids_len will indicate the number of program IDs that were successfully copied.

Using prctl(2)

A process can enable or disable all currently open event groups using the prctl(2) PR_TASK_PERF_EVENTS_ENABLE and PR_TASK_PERF_EVENTS_DISABLE operations. This applies only to events created locally by the calling process. This does not apply to events created by other processes attached to the calling process or inherited events from a parent process. Only group leaders are enabled and disabled, not any other members of the groups.

Files in /proc/sys/kernel/

/proc/sys/kernel/perf_event_paranoid
The perf_event_paranoid file can be set to restrict access to the performance counters.
2
allow only user-space measurements (default since Linux 4.6).
1
allow both kernel and user measurements (default before Linux 4.6).
0
allow access to CPU-specific data but not raw tracepoint samples.
-1
no restrictions.
The existence of the perf_event_paranoid file is the official method for determining if a kernel supports perf_event_open().
/proc/sys/kernel/perf_event_max_sample_rate
This sets the maximum sample rate. Setting this too high can allow users to sample at a rate that impacts overall machine performance and potentially lock up the machine. The default value is 100000 (samples per second).
/proc/sys/kernel/perf_event_max_stack
This file sets the maximum depth of stack frame entries reported when generating a call trace.
/proc/sys/kernel/perf_event_mlock_kb
Maximum number of pages an unprivileged user can mlock(2). The default is 516 (kB).

Files in /sys/bus/event_source/devices/

Since Linux 2.6.34, the kernel supports having multiple PMUs available for monitoring. Information on how to program these PMUs can be found under /sys/bus/event_source/devices/. Each subdirectory corresponds to a different PMU.
/sys/bus/event_source/devices/*/type (since Linux 2.6.38)
This contains an integer that can be used in the type field of perf_event_attr to indicate that you wish to use this PMU.
/sys/bus/event_source/devices/cpu/rdpmc (since Linux 3.4)
If this file is 1, then direct user-space access to the performance counter registers is allowed via the rdpmc instruction. This can be disabled by echoing 0 to the file.
As of Linux 4.0 the behavior has changed, so that 1 now means only allow access to processes with active perf events, with 2 indicating the old allow-anyone-access behavior.
/sys/bus/event_source/devices/*/format/ (since Linux 3.4)
This subdirectory contains information on the architecture-specific subfields available for programming the various config fields in the perf_event_attr struct.
The content of each file is the name of the config field, followed by a colon, followed by a series of integer bit ranges separated by commas. For example, the file event may contain the value config1:1,6-10,44 which indicates that event is an attribute that occupies bits 1,6–10, and 44 of perf_event_attr::config1.
/sys/bus/event_source/devices/*/events/ (since Linux 3.4)
This subdirectory contains files with predefined events. The contents are strings describing the event settings expressed in terms of the fields found in the previously mentioned ./format/ directory. These are not necessarily complete lists of all events supported by a PMU, but usually a subset of events deemed useful or interesting.
The content of each file is a list of attribute names separated by commas. Each entry has an optional value (either hex or decimal). If no value is specified, then it is assumed to be a single-bit field with a value of 1. An example entry may look like this: event=0x2,inv,ldlat=3.
/sys/bus/event_source/devices/*/uevent
This file is the standard kernel device interface for injecting hotplug events.
/sys/bus/event_source/devices/*/cpumask (since Linux 3.7)
The cpumask file contains a comma-separated list of integers that indicate a representative CPU number for each socket (package) on the motherboard. This is needed when setting up uncore or northbridge events, as those PMUs present socket-wide events.

RETURN VALUE

On success, perf_event_open() returns the new file descriptor. On error, -1 is returned and errno is set to indicate the error.

ERRORS

The errors returned by perf_event_open() can be inconsistent, and may vary across processor architectures and performance monitoring units.

Returned if the perf_event_attr size value is too small (smaller than PERF_ATTR_SIZE_VER0), too big (larger than the page size), or larger than the kernel supports and the extra bytes are not zero. When E2BIG is returned, the perf_event_attr size field is overwritten by the kernel to be the size of the structure it was expecting.
Returned when the requested event requires CAP_PERFMON (since Linux 5.8) or CAP_SYS_ADMIN permissions (or a more permissive perf_event paranoid setting). Some common cases where an unprivileged process may encounter this error: attaching to a process owned by a different user; monitoring all processes on a given CPU (i.e., specifying the pid argument as -1); and not setting exclude_kernel when the paranoid setting requires it.
Returned if the group_fd file descriptor is not valid, or, if PERF_FLAG_PID_CGROUP is set, the cgroup file descriptor in pid is not valid.
Returned if another event already has exclusive access to the PMU.
Returned if the attr pointer points at an invalid memory address.
Returned when trying to mix perf and ftrace handling for a uprobe.
Returned if the specified event is invalid. There are many possible reasons for this. A not-exhaustive list: sample_freq is higher than the maximum setting; the cpu to monitor does not exist; read_format is out of range; sample_type is out of range; the flags value is out of range; exclusive or pinned set and the event is not a group leader; the event config values are out of range or set reserved bits; the generic event selected is not supported; or there is not enough room to add the selected event.
Each opened event uses one file descriptor. If a large number of events are opened, the per-process limit on the number of open file descriptors will be reached, and no more events can be created.
Returned when the event involves a feature not supported by the current CPU.
Returned if the type setting is not valid. This error is also returned for some unsupported generic events.
Prior to Linux 3.3, if there was not enough room for the event, ENOSPC was returned. In Linux 3.3, this was changed to EINVAL. ENOSPC is still returned if you try to add more breakpoint events than supported by the hardware.
Returned if PERF_SAMPLE_STACK_USER is set in sample_type and it is not supported by hardware.
Returned if an event requiring a specific hardware feature is requested but there is no hardware support. This includes requesting low-skid events if not supported, branch tracing if it is not available, sampling if no PMU interrupt is available, and branch stacks for software events.
Returned if PERF_SAMPLE_CALLCHAIN is requested and sample_max_stack is larger than the maximum specified in /proc/sys/kernel/perf_event_max_stack.
Returned on many (but not all) architectures when an unsupported exclude_hv, exclude_idle, exclude_user, or exclude_kernel setting is specified.
It can also happen, as with EACCES, when the requested event requires CAP_PERFMON (since Linux 5.8) or CAP_SYS_ADMIN permissions (or a more permissive perf_event paranoid setting). This includes setting a breakpoint on a kernel address, and (since Linux 3.13) setting a kernel function-trace tracepoint.
Returned if attempting to attach to a process that does not exist.

STANDARDS

Linux.

HISTORY

perf_event_open() was introduced in Linux 2.6.31 but was called perf_counter_open(). It was renamed in Linux 2.6.32.

NOTES

The official way of knowing if perf_event_open() support is enabled is checking for the existence of the file /proc/sys/kernel/perf_event_paranoid.

CAP_PERFMON capability (since Linux 5.8) provides secure approach to performance monitoring and observability operations in a system according to the principal of least privilege (POSIX IEEE 1003.1e). Accessing system performance monitoring and observability operations using CAP_PERFMON rather than the much more powerful CAP_SYS_ADMIN excludes chances to misuse credentials and makes operations more secure. CAP_SYS_ADMIN usage for secure system performance monitoring and observability is discouraged in favor of the CAP_PERFMON capability.

BUGS

The F_SETOWN_EX option to fcntl(2) is needed to properly get overflow signals in threads. This was introduced in Linux 2.6.32.

Prior to Linux 2.6.33 (at least for x86), the kernel did not check if events could be scheduled together until read time. The same happens on all known kernels if the NMI watchdog is enabled. This means to see if a given set of events works you have to perf_event_open(), start, then read before you know for sure you can get valid measurements.

Prior to Linux 2.6.34, event constraints were not enforced by the kernel. In that case, some events would silently return "0" if the kernel scheduled them in an improper counter slot.

Prior to Linux 2.6.34, there was a bug when multiplexing where the wrong results could be returned.

Kernels from Linux 2.6.35 to Linux 2.6.39 can quickly crash the kernel if "inherit" is enabled and many threads are started.

Prior to Linux 2.6.35, PERF_FORMAT_GROUP did not work with attached processes.

There is a bug in the kernel code between Linux 2.6.36 and Linux 3.0 that ignores the "watermark" field and acts as if a wakeup_event was chosen if the union has a nonzero value in it.

From Linux 2.6.31 to Linux 3.4, the PERF_IOC_FLAG_GROUP ioctl argument was broken and would repeatedly operate on the event specified rather than iterating across all sibling events in a group.

From Linux 3.4 to Linux 3.11, the mmap cap_usr_rdpmc and cap_usr_time bits mapped to the same location. Code should migrate to the new cap_user_rdpmc and cap_user_time fields instead.

Always double-check your results! Various generalized events have had wrong values. For example, retired branches measured the wrong thing on AMD machines until Linux 2.6.35.

EXAMPLES

The following is a short example that measures the total instruction count of a call to printf(3).

#include <linux/perf_event.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/ioctl.h>
#include <sys/syscall.h>
#include <unistd.h>
static long
perf_event_open(struct perf_event_attr *hw_event, pid_t pid,

int cpu, int group_fd, unsigned long flags) {
int ret;
ret = syscall(SYS_perf_event_open, hw_event, pid, cpu,
group_fd, flags);
return ret; } int main(void) {
int fd;
long long count;
struct perf_event_attr pe;
memset(&pe, 0, sizeof(pe));
pe.type = PERF_TYPE_HARDWARE;
pe.size = sizeof(pe);
pe.config = PERF_COUNT_HW_INSTRUCTIONS;
pe.disabled = 1;
pe.exclude_kernel = 1;
pe.exclude_hv = 1;
fd = perf_event_open(&pe, 0, -1, -1, 0);
if (fd == -1) {
fprintf(stderr, "Error opening leader %llx\n", pe.config);
exit(EXIT_FAILURE);
}
ioctl(fd, PERF_EVENT_IOC_RESET, 0);
ioctl(fd, PERF_EVENT_IOC_ENABLE, 0);
printf("Measuring instruction count for this printf\n");
ioctl(fd, PERF_EVENT_IOC_DISABLE, 0);
read(fd, &count, sizeof(count));
printf("Used %lld instructions\n", count);
close(fd); }

SEE ALSO

perf(1), fcntl(2), mmap(2), open(2), prctl(2), read(2)

Documentation/admin-guide/perf-security.rst in the kernel source tree

2023-04-03 Linux man-pages 6.04