.\" Copyright (c) 2012, Vincent Weaver .\" .\" SPDX-License-Identifier: GPL-2.0-or-later .\" .\" This document is based on the perf_event.h header file, the .\" tools/perf/design.txt file, and a lot of bitter experience. .\" .TH perf_event_open 2 2024-05-02 "Linux man-pages 6.8" .SH NAME perf_event_open \- set up performance monitoring .SH LIBRARY Standard C library .RI ( libc ", " \-lc ) .SH SYNOPSIS .nf .BR "#include " " /* Definition of " PERF_* " constants */" .BR "#include " " /* Definition of " HW_* " constants */" .BR "#include " " /* Definition of " SYS_* " constants */" .B #include .P .BI "int syscall(SYS_perf_event_open, struct perf_event_attr *" attr , .BI " pid_t " pid ", int " cpu ", int " group_fd \ ", unsigned long " flags ); .fi .P .IR Note : glibc provides no wrapper for .BR perf_event_open (), necessitating the use of .BR syscall (2). .SH DESCRIPTION Given a list of parameters, .BR perf_event_open () returns a file descriptor, for use in subsequent system calls (\c .BR read (2), .BR mmap (2), .BR prctl (2), .BR fcntl (2), etc.). .PP A call to .BR 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. .P Events can be enabled and disabled in two ways: via .BR ioctl (2) and via .BR prctl (2). When an event is disabled it does not count or generate overflows but does continue to exist and maintain its count value. .P Events come in two flavors: counting and sampled. A .I 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 .BR read (2) call. A .I sampling event periodically writes measurements to a buffer that can then be accessed via .BR mmap (2). .SS Arguments The .I pid and .I cpu arguments allow specifying which process and CPU to monitor: .TP .BR "pid == 0" " and " "cpu == \-1" This measures the calling process/thread on any CPU. .TP .BR "pid == 0" " and " "cpu >= 0" This measures the calling process/thread only when running on the specified CPU. .TP .BR "pid > 0" " and " "cpu == \-1" This measures the specified process/thread on any CPU. .TP .BR "pid > 0" " and " "cpu >= 0" This measures the specified process/thread only when running on the specified CPU. .TP .BR "pid == \-1" " and " "cpu >= 0" This measures all processes/threads on the specified CPU. This requires .B CAP_PERFMON (since Linux 5.8) or .B CAP_SYS_ADMIN capability or a .I /proc/sys/kernel/perf_event_paranoid value of less than 1. .TP .BR "pid == \-1" " and " "cpu == \-1" This setting is invalid and will return an error. .P When .I pid is greater than zero, permission to perform this system call is governed by .B CAP_PERFMON (since Linux 5.9) and a ptrace access mode .B PTRACE_MODE_READ_REALCREDS check on older Linux versions; see .BR ptrace (2). .P The .I 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 .IR group_fd " = \-1." The rest of the group members are created with subsequent .BR perf_event_open () calls with .I group_fd being set to the file descriptor of the group leader. (A single event on its own is created with .IR 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 \[em]added, divided (to get ratios), and so on\[em] with each other, since they have counted events for the same set of executed instructions. .P The .I flags argument is formed by ORing together zero or more of the following values: .TP .BR PERF_FLAG_FD_CLOEXEC " (since Linux 3.14)" .\" commit a21b0b354d4ac39be691f51c53562e2c24443d9e This flag enables the close-on-exec flag for the created event file descriptor, so that the file descriptor is automatically closed on .BR execve (2). Setting the close-on-exec flags at creation time, rather than later with .BR fcntl (2), avoids potential race conditions where the calling thread invokes .BR perf_event_open () and .BR fcntl (2) at the same time as another thread calls .BR fork (2) then .BR execve (2). .TP .B PERF_FLAG_FD_NO_GROUP This flag tells the event to ignore the .I group_fd parameter except for the purpose of setting up output redirection using the .B PERF_FLAG_FD_OUTPUT flag. .TP .BR PERF_FLAG_FD_OUTPUT " (broken since Linux 2.6.35)" .\" commit ac9721f3f54b27a16c7e1afb2481e7ee95a70318 This flag re-routes the event's sampled output to instead be included in the mmap buffer of the event specified by .IR group_fd . .TP .BR PERF_FLAG_PID_CGROUP " (since Linux 2.6.39)" .\" commit e5d1367f17ba6a6fed5fd8b74e4d5720923e0c25 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 .IR test , then a file descriptor opened on .I /dev/cgroup/test (assuming cgroupfs is mounted on .IR /dev/cgroup ) must be passed as the .I pid parameter. cgroup monitoring is available only for system-wide events and may therefore require extra permissions. .P The .I perf_event_attr structure provides detailed configuration information for the event being created. .P .in +4n .EX 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\[aq]t count user */ exclude_kernel : 1, /* don\[aq]t count kernel */ exclude_hv : 1, /* don\[aq]t count hypervisor */ exclude_idle : 1, /* don\[aq]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\[aq]t count in host */ exclude_guest : 1, /* don\[aq]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 */ \& }; .EE .in .P The fields of the .I perf_event_attr structure are described in more detail below: .TP .I type This field specifies the overall event type. It has one of the following values: .RS .TP .B PERF_TYPE_HARDWARE This indicates one of the "generalized" hardware events provided by the kernel. See the .I config field definition for more details. .TP .B PERF_TYPE_SOFTWARE This indicates one of the software-defined events provided by the kernel (even if no hardware support is available). .TP .B PERF_TYPE_TRACEPOINT This indicates a tracepoint provided by the kernel tracepoint infrastructure. .TP .B PERF_TYPE_HW_CACHE This indicates a hardware cache event. This has a special encoding, described in the .I config field definition. .TP .B PERF_TYPE_RAW This indicates a "raw" implementation-specific event in the .IR config " field." .TP .BR PERF_TYPE_BREAKPOINT " (since Linux 2.6.33)" .\" commit 24f1e32c60c45c89a997c73395b69c8af6f0a84e 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. .TP dynamic PMU Since Linux 2.6.38, .\" commit 2e80a82a49c4c7eca4e35734380f28298ba5db19 .BR perf_event_open () can support multiple PMUs. To enable this, a value exported by the kernel can be used in the .I 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 .IR /sys/bus/event_source/devices . In each subdirectory there is a .I type file whose content is an integer that can be used in the .I type field. For instance, .I /sys/bus/event_source/devices/cpu/type contains the value for the core CPU PMU, which is usually 4. .TP .BR kprobe " and " uprobe " (since Linux 4.17)" .\" commit 65074d43fc77bcae32776724b7fa2696923c78e4 .\" commit e12f03d7031a977356e3d7b75a68c2185ff8d155 .\" commit 33ea4b24277b06dbc55d7f5772a46f029600255e 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 .IR kprobe_func , .IR uprobe_path , .IR kprobe_addr , and .I probe_offset for more details. .RE .TP .I "size" The size of the .I perf_event_attr structure for forward/backward compatibility. Set this using .I sizeof(struct perf_event_attr) to allow the kernel to see the struct size at the time of compilation. .IP The related define .B PERF_ATTR_SIZE_VER0 is set to 64; this was the size of the first published struct. .B PERF_ATTR_SIZE_VER1 is 72, corresponding to the addition of breakpoints in Linux 2.6.33. .\" commit cb5d76999029ae7a517cb07dfa732c1b5a934fc2 .\" this was added much later when PERF_ATTR_SIZE_VER2 happened .\" but the actual attr_size had increased in Linux 2.6.33 .B PERF_ATTR_SIZE_VER2 is 80 corresponding to the addition of branch sampling in Linux 3.4. .\" commit cb5d76999029ae7a517cb07dfa732c1b5a934fc2 .B PERF_ATTR_SIZE_VER3 is 96 corresponding to the addition of .I sample_regs_user and .I sample_stack_user in Linux 3.7. .\" commit 1659d129ed014b715b0b2120e6fd929bdd33ed03 .B PERF_ATTR_SIZE_VER4 is 104 corresponding to the addition of .I sample_regs_intr in Linux 3.19. .\" commit 60e2364e60e86e81bc6377f49779779e6120977f .B PERF_ATTR_SIZE_VER5 is 112 corresponding to the addition of .I aux_watermark in Linux 4.1. .\" commit 1a5941312414c71dece6717da9a0fa1303127afa .TP .I "config" This specifies which event you want, in conjunction with the .I type field. The .I config1 and .I 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. .IP There are various ways to set the .I config field that are dependent on the value of the previously described .I type field. What follows are various possible settings for .I config separated out by .IR type . .IP If .I type is .BR PERF_TYPE_HARDWARE , we are measuring one of the generalized hardware CPU events. Not all of these are available on all platforms. Set .I config to one of the following: .RS 12 .TP .B PERF_COUNT_HW_CPU_CYCLES Total cycles. Be wary of what happens during CPU frequency scaling. .TP .B PERF_COUNT_HW_INSTRUCTIONS Retired instructions. Be careful, these can be affected by various issues, most notably hardware interrupt counts. .TP .B PERF_COUNT_HW_CACHE_REFERENCES 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. .TP .B PERF_COUNT_HW_CACHE_MISSES Cache misses. Usually this indicates Last Level Cache misses; this is intended to be used in conjunction with the .B PERF_COUNT_HW_CACHE_REFERENCES event to calculate cache miss rates. .TP .B PERF_COUNT_HW_BRANCH_INSTRUCTIONS Retired branch instructions. Prior to Linux 2.6.35, this used the wrong event on AMD processors. .\" commit f287d332ce835f77a4f5077d2c0ef1e3f9ea42d2 .TP .B PERF_COUNT_HW_BRANCH_MISSES Mispredicted branch instructions. .TP .B PERF_COUNT_HW_BUS_CYCLES Bus cycles, which can be different from total cycles. .TP .BR PERF_COUNT_HW_STALLED_CYCLES_FRONTEND " (since Linux 3.0)" .\" commit 8f62242246351b5a4bc0c1f00c0c7003edea128a Stalled cycles during issue. .TP .BR PERF_COUNT_HW_STALLED_CYCLES_BACKEND " (since Linux 3.0)" .\" commit 8f62242246351b5a4bc0c1f00c0c7003edea128a Stalled cycles during retirement. .TP .BR PERF_COUNT_HW_REF_CPU_CYCLES " (since Linux 3.3)" .\" commit c37e17497e01fc0f5d2d6feb5723b210b3ab8890 Total cycles; not affected by CPU frequency scaling. .RE .IP If .I type is .BR PERF_TYPE_SOFTWARE , we are measuring software events provided by the kernel. Set .I config to one of the following: .RS 12 .TP .B PERF_COUNT_SW_CPU_CLOCK This reports the CPU clock, a high-resolution per-CPU timer. .TP .B PERF_COUNT_SW_TASK_CLOCK This reports a clock count specific to the task that is running. .TP .B PERF_COUNT_SW_PAGE_FAULTS This reports the number of page faults. .TP .B PERF_COUNT_SW_CONTEXT_SWITCHES 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. .\" commit e49a5bd38159dfb1928fd25b173bc9de4bbadb21 .TP .B PERF_COUNT_SW_CPU_MIGRATIONS This reports the number of times the process has migrated to a new CPU. .TP .B PERF_COUNT_SW_PAGE_FAULTS_MIN This counts the number of minor page faults. These did not require disk I/O to handle. .TP .B PERF_COUNT_SW_PAGE_FAULTS_MAJ This counts the number of major page faults. These required disk I/O to handle. .TP .BR PERF_COUNT_SW_ALIGNMENT_FAULTS " (since Linux 2.6.33)" .\" commit f7d7986060b2890fc26db6ab5203efbd33aa2497 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). .TP .BR PERF_COUNT_SW_EMULATION_FAULTS " (since Linux 2.6.33)" .\" commit f7d7986060b2890fc26db6ab5203efbd33aa2497 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. .TP .BR PERF_COUNT_SW_DUMMY " (since Linux 3.12)" .\" commit fa0097ee690693006ab1aea6c01ad3c851b65c77 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. .TP .BR PERF_COUNT_SW_BPF_OUTPUT " (since Linux 4.4)" .\" commit a43eec304259a6c637f4014a6d4767159b6a3aa3 This is used to generate raw sample data from BPF. BPF programs can write to this event using .B bpf_perf_event_output helper. .TP .BR PERF_COUNT_SW_CGROUP_SWITCHES " (since Linux 5.13)" .\" commit d0d1dd628527c77db2391ce0293c1ed344b2365f 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. .RE .P .RS If .I type is .BR PERF_TYPE_TRACEPOINT , then we are measuring kernel tracepoints. The value to use in .I config can be obtained from under debugfs .I tracing/events/*/*/id if ftrace is enabled in the kernel. .RE .P .RS If .I type is .BR PERF_TYPE_HW_CACHE , then we are measuring a hardware CPU cache event. To calculate the appropriate .I config value, use the following equation: .RS 4 .P .in +4n .EX config = (perf_hw_cache_id) | (perf_hw_cache_op_id << 8) | (perf_hw_cache_op_result_id << 16); .EE .in .P where .I perf_hw_cache_id is one of: .RS 4 .TP .B PERF_COUNT_HW_CACHE_L1D for measuring Level 1 Data Cache .TP .B PERF_COUNT_HW_CACHE_L1I for measuring Level 1 Instruction Cache .TP .B PERF_COUNT_HW_CACHE_LL for measuring Last-Level Cache .TP .B PERF_COUNT_HW_CACHE_DTLB for measuring the Data TLB .TP .B PERF_COUNT_HW_CACHE_ITLB for measuring the Instruction TLB .TP .B PERF_COUNT_HW_CACHE_BPU for measuring the branch prediction unit .TP .BR PERF_COUNT_HW_CACHE_NODE " (since Linux 3.1)" .\" commit 89d6c0b5bdbb1927775584dcf532d98b3efe1477 for measuring local memory accesses .RE .P and .I perf_hw_cache_op_id is one of: .RS 4 .TP .B PERF_COUNT_HW_CACHE_OP_READ for read accesses .TP .B PERF_COUNT_HW_CACHE_OP_WRITE for write accesses .TP .B PERF_COUNT_HW_CACHE_OP_PREFETCH for prefetch accesses .RE .P and .I perf_hw_cache_op_result_id is one of: .RS 4 .TP .B PERF_COUNT_HW_CACHE_RESULT_ACCESS to measure accesses .TP .B PERF_COUNT_HW_CACHE_RESULT_MISS to measure misses .RE .RE .P If .I type is .BR PERF_TYPE_RAW , then a custom "raw" .I 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 .BR perf_event_open () expects in this field. .P If .I type is .BR PERF_TYPE_BREAKPOINT , then leave .I config set to zero. Its parameters are set in other places. .P If .I type is .B kprobe or .BR uprobe , set .I retprobe (bit 0 of .IR config , see .IR /sys/bus/event_source/devices/[k,u]probe/format/retprobe ) for kretprobe/uretprobe. See fields .IR kprobe_func , .IR uprobe_path , .IR kprobe_addr , and .I probe_offset for more details. .RE .TP .I kprobe_func .TQ .I uprobe_path .TQ .I kprobe_addr .TQ .I probe_offset These fields describe the kprobe/uprobe for dynamic PMUs .B kprobe and .BR uprobe . For .BR kprobe : use .I kprobe_func and .IR probe_offset , or use .I kprobe_addr and leave .I kprobe_func as NULL. For .BR uprobe : use .I uprobe_path and .IR probe_offset . .TP .I sample_period .TQ .I sample_freq A "sampling" event is one that generates an overflow notification every N events, where N is given by .IR sample_period . A sampling event has .IR sample_period " > 0." When an overflow occurs, requested data is recorded in the mmap buffer. The .I sample_type field controls what data is recorded on each overflow. .IP .I sample_freq can be used if you wish to use frequency rather than period. In this case, you set the .I freq flag. The kernel will adjust the sampling period to try and achieve the desired rate. The rate of adjustment is a timer tick. .TP .I sample_type 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 .BR 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 .I "enum perf_event_sample_format" order. .RS .TP .B PERF_SAMPLE_IP Records instruction pointer. .TP .B PERF_SAMPLE_TID Records the process and thread IDs. .TP .B PERF_SAMPLE_TIME Records a timestamp. .TP .B PERF_SAMPLE_ADDR Records an address, if applicable. .TP .B PERF_SAMPLE_READ Record counter values for all events in a group, not just the group leader. .TP .B PERF_SAMPLE_CALLCHAIN Records the callchain (stack backtrace). .TP .B PERF_SAMPLE_ID Records a unique ID for the opened event's group leader. .TP .B PERF_SAMPLE_CPU Records CPU number. .TP .B PERF_SAMPLE_PERIOD Records the current sampling period. .TP .B PERF_SAMPLE_STREAM_ID Records a unique ID for the opened event. Unlike .B PERF_SAMPLE_ID the actual ID is returned, not the group leader. This ID is the same as the one returned by .BR PERF_FORMAT_ID . .TP .B PERF_SAMPLE_RAW Records additional data, if applicable. Usually returned by tracepoint events. .TP .BR PERF_SAMPLE_BRANCH_STACK " (since Linux 3.4)" .\" commit bce38cd53e5ddba9cb6d708c4ef3d04a4016ec7e 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. .IP See the .I branch_sample_type field for how to filter which branches are reported. .TP .BR PERF_SAMPLE_REGS_USER " (since Linux 3.7)" .\" commit 4018994f3d8785275ef0e7391b75c3462c029e56 Records the current user-level CPU register state (the values in the process before the kernel was called). .TP .BR PERF_SAMPLE_STACK_USER " (since Linux 3.7)" .\" commit c5ebcedb566ef17bda7b02686e0d658a7bb42ee7 Records the user level stack, allowing stack unwinding. .TP .BR PERF_SAMPLE_WEIGHT " (since Linux 3.10)" .\" commit c3feedf2aaf9ac8bad6f19f5d21e4ee0b4b87e9c 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. .TP .BR PERF_SAMPLE_DATA_SRC " (since Linux 3.10)" .\" commit d6be9ad6c960f43800a6f118932bc8a5a4eadcd1 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. .TP .BR PERF_SAMPLE_IDENTIFIER " (since Linux 3.12)" .\" commit ff3d527cebc1fa3707c617bfe9e74f53fcfb0955 Places the .B 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). .IP This was necessary because a sample stream may have records from various different event sources with different .I sample_type settings. Parsing the event stream properly was not possible because the format of the record was needed to find .BR SAMPLE_ID , but the format could not be found without knowing what event the sample belonged to (causing a circular dependency). .IP The .B PERF_SAMPLE_IDENTIFIER setting makes the event stream always parsable by putting .B SAMPLE_ID in a fixed location, even though it means having duplicate .B SAMPLE_ID values in records. .TP .BR PERF_SAMPLE_TRANSACTION " (since Linux 3.13)" .\" commit fdfbbd07e91f8fe387140776f3fd94605f0c89e5 Records reasons for transactional memory abort events (for example, from Intel TSX transactional memory support). .IP The .I 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). .TP .BR PERF_SAMPLE_REGS_INTR " (since Linux 3.19)" .\" commit 60e2364e60e86e81bc6377f49779779e6120977f Records a subset of the current CPU register state as specified by .IR sample_regs_intr . Unlike .B 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 .I 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. .TP .BR PERF_SAMPLE_PHYS_ADDR " (since Linux 4.13)" .\" commit fc7ce9c74c3ad232b084d80148654f926d01ece7 Records physical address of data like in .BR PERF_SAMPLE_ADDR . .TP .BR PERF_SAMPLE_CGROUP " (since Linux 5.7)" .\" commit 96aaab686505c449e24d76e76507290dcc30e008 Records (perf_event) cgroup ID of the process. This corresponds to the .I id field in the .B PERF_RECORD_CGROUP event. .TP .BR PERF_SAMPLE_DATA_PAGE_SIZE " (since Linux 5.11)" .\" commit 8d97e71811aaafe4abf611dc24822fd6e73df1a1 Records page size of data like in .BR PERF_SAMPLE_ADDR . .TP .BR PERF_SAMPLE_CODE_PAGE_SIZE " (since Linux 5.11)" .\" commit 995f088efebe1eba0282a6ffa12411b37f8990c2 Records page size of ip like in .BR PERF_SAMPLE_IP . .TP .BR PERF_SAMPLE_WEIGHT_STRUCT " (since Linux 5.12)" .\" commit 2a6c6b7d7ad346f0679d0963cb19b3f0ea7ef32c Records hardware provided weight values like in .BR PERF_SAMPLE_WEIGHT , but it can represent multiple values in a struct. This shares the same space as .BR 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. .P .in +4n .EX union perf_sample_weight { u64 full; /* PERF_SAMPLE_WEIGHT */ struct { /* PERF_SAMPLE_WEIGHT_STRUCT */ u32 var1_dw; u16 var2_w; u16 var3_w; }; }; .EE .in .RE .TP .I read_format This field specifies the format of the data returned by .BR read (2) on a .BR perf_event_open () file descriptor. .RS .TP .B PERF_FORMAT_TOTAL_TIME_ENABLED Adds the 64-bit .I time_enabled field. This can be used to calculate estimated totals if the PMU is overcommitted and multiplexing is happening. .TP .B PERF_FORMAT_TOTAL_TIME_RUNNING Adds the 64-bit .I time_running field. This can be used to calculate estimated totals if the PMU is overcommitted and multiplexing is happening. .TP .B PERF_FORMAT_ID Adds a 64-bit unique value that corresponds to the event group. .TP .B PERF_FORMAT_GROUP Allows all counter values in an event group to be read with one read. .TP .B PERF_FORMAT_LOST " (since Linux 6.0)" .\" commit 119a784c81270eb88e573174ed2209225d646656 Adds a 64-bit value that is the number of lost samples for this event. This would be only meaningful when .I sample_period or .I sample_freq is set. .RE .TP .I disabled The .I disabled bit specifies whether the counter starts out disabled or enabled. If disabled, the event can later be enabled by .BR ioctl (2), .BR prctl (2), or .IR enable_on_exec . .IP When creating an event group, typically the group leader is initialized with .I disabled set to 1 and any child events are initialized with .I disabled set to 0. Despite .I disabled being 0, the child events will not start until the group leader is enabled. .TP .I inherit The .I 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). .IP Inherit does not work for some combinations of .I read_format values, such as .BR PERF_FORMAT_GROUP . .TP .I pinned The .I 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., .BR read (2) returns 0) until the counter is subsequently enabled or disabled. .TP .I exclusive The .I 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. .IP Note that many unexpected situations may prevent events with the .I 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). .TP .I exclude_user If this bit is set, the count excludes events that happen in user space. .TP .I exclude_kernel If this bit is set, the count excludes events that happen in kernel space. .TP .I exclude_hv 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. .TP .I exclude_idle 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. .TP .I mmap The .I mmap bit enables generation of .B PERF_RECORD_MMAP samples for every .BR mmap (2) call that has .B 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. .TP .I comm The .I comm bit enables tracking of process command name as modified by the .BR execve (2) and .BR prctl (PR_SET_NAME) system calls as well as writing to .IR /proc/self/comm . If the .I comm_exec flag is also successfully set (possible since Linux 3.16), .\" commit 82b897782d10fcc4930c9d4a15b175348fdd2871 then the misc flag .B PERF_RECORD_MISC_COMM_EXEC can be used to differentiate the .BR execve (2) case from the others. .TP .I freq If this bit is set, then .I sample_frequency not .I sample_period is used when setting up the sampling interval. .TP .I inherit_stat This bit enables saving of event counts on context switch for inherited tasks. This is meaningful only if the .I inherit field is set. .TP .I enable_on_exec If this bit is set, a counter is automatically enabled after a call to .BR execve (2). .TP .I task If this bit is set, then fork/exit notifications are included in the ring buffer. .TP .I watermark If set, have an overflow notification happen when we cross the .I wakeup_watermark boundary. Otherwise, overflow notifications happen after .I wakeup_events samples. .TP .IR precise_ip " (since Linux 2.6.35)" .\" commit ab608344bcbde4f55ec4cd911b686b0ce3eae076 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. .IP The possible values of this field are the following: .RS .TP .B 0 .B SAMPLE_IP can have arbitrary skid. .TP .B 1 .B SAMPLE_IP must have constant skid. .TP .B 2 .B SAMPLE_IP requested to have 0 skid. .TP .B 3 .B SAMPLE_IP must have 0 skid. See also the description of .BR PERF_RECORD_MISC_EXACT_IP . .RE .TP .IR mmap_data " (since Linux 2.6.36)" .\" commit 3af9e859281bda7eb7c20b51879cf43aa788ac2e This is the counterpart of the .I mmap field. This enables generation of .B PERF_RECORD_MMAP samples for .BR mmap (2) calls that do not have .B PROT_EXEC set (for example data and SysV shared memory). .TP .IR sample_id_all " (since Linux 2.6.38)" .\" commit c980d1091810df13f21aabbce545fd98f545bbf7 If set, then TID, TIME, ID, STREAM_ID, and CPU can additionally be included in .RB non- PERF_RECORD_SAMPLE s if the corresponding .I sample_type is selected. .IP If .B 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 .I id value appearing twice. .IP The layout is described by this pseudo-structure: .IP .in +4n .EX 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 */ }; .EE .in .TP .IR exclude_host " (since Linux 3.2)" .\" commit a240f76165e6255384d4bdb8139895fac7988799 When conducting measurements that include processes running VM instances (i.e., have executed a .B KVM_RUN .BR 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. .TP .IR exclude_guest " (since Linux 3.2)" .\" commit a240f76165e6255384d4bdb8139895fac7988799 When conducting measurements that include processes running VM instances (i.e., have executed a .B KVM_RUN .BR 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. .TP .IR exclude_callchain_kernel " (since Linux 3.7)" .\" commit d077526485d5c9b12fe85d0b2b3b7041e6bc5f91 Do not include kernel callchains. .TP .IR exclude_callchain_user " (since Linux 3.7)" .\" commit d077526485d5c9b12fe85d0b2b3b7041e6bc5f91 Do not include user callchains. .TP .IR mmap2 " (since Linux 3.16)" .\" commit 13d7a2410fa637f450a29ecb515ac318ee40c741 .\" This is tricky; was committed during 3.12 development .\" but right before release was disabled. .\" So while you could select mmap2 starting with Linux 3.12 .\" it did not work until Linux 3.16 .\" commit a5a5ba72843dd05f991184d6cb9a4471acce1005 Generate an extended executable mmap record that contains enough additional information to uniquely identify shared mappings. The .I mmap flag must also be set for this to work. .TP .IR comm_exec " (since Linux 3.16)" .\" commit 82b897782d10fcc4930c9d4a15b175348fdd2871 This is purely a feature-detection flag, it does not change kernel behavior. If this flag can successfully be set, then, when .I comm is enabled, the .B PERF_RECORD_MISC_COMM_EXEC flag will be set in the .I misc field of a comm record header if the rename event being reported was caused by a call to .BR execve (2). This allows tools to distinguish between the various types of process renaming. .TP .IR use_clockid " (since Linux 4.1)" .\" commit 34f439278cef7b1177f8ce24f9fc81dfc6221d3b This allows selecting which internal Linux clock to use when generating timestamps via the .I clockid field. This can make it easier to correlate perf sample times with timestamps generated by other tools. .TP .IR context_switch " (since Linux 4.3)" .\" commit 45ac1403f564f411c6a383a2448688ba8dd705a4 This enables the generation of .B PERF_RECORD_SWITCH records when a context switch occurs. It also enables the generation of .B 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 .I perf_event_paranoid settings. .TP .IR write_backward " (since Linux 4.6)" .\" commit 9ecda41acb971ebd07c8fb35faf24005c0baea12 This causes the ring buffer to be written from the end to the beginning. This is to support reading from overwritable ring buffer. .TP .IR namespaces " (since Linux 4.11)" .\" commit e422267322cd319e2695a535e47c5b1feeac45eb This enables the generation of .B PERF_RECORD_NAMESPACES records when a task enters a new namespace. Each namespace has a combination of device and inode numbers. .TP .IR ksymbol " (since Linux 5.0)" .\" commit 76193a94522f1d4edf2447a536f3f796ce56343b This enables the generation of .B PERF_RECORD_KSYMBOL records when new kernel symbols are registered or unregistered. This is analyzing dynamic kernel functions like eBPF. .TP .IR bpf_event " (since Linux 5.0)" .\" commit 6ee52e2a3fe4ea35520720736e6791df1fb67106 This enables the generation of .B PERF_RECORD_BPF_EVENT records when an eBPF program is loaded or unloaded. .TP .IR aux_output " (since Linux 5.4)" .\" commit ab43762ef010967e4ccd53627f70a2eecbeafefb This allows normal (non-AUX) events to generate data for AUX events if the hardware supports it. .TP .IR cgroup " (since Linux 5.7)" .\" commit 96aaab686505c449e24d76e76507290dcc30e008 This enables the generation of .B PERF_RECORD_CGROUP records when a new cgroup is created (and activated). .TP .IR text_poke " (since Linux 5.8)" .\" commit e17d43b93e544f5016c0251d2074c15568d5d963 This enables the generation of .B PERF_RECORD_TEXT_POKE records when there's a change to the kernel text (i.e., self-modifying code). .TP .IR build_id " (since Linux 5.12)" .\" commit 88a16a1309333e43d328621ece3e9fa37027e8eb This changes the contents in the .B PERF_RECORD_MMAP2 to have a build-id instead of device and inode numbers. .TP .IR inherit_thread " (since Linux 5.13)" .\" commit 2b26f0aa004995f49f7b6f4100dd0e4c39a9ed5f This disables the inheritance of the event to a child process. Only new threads in the same process (which is cloned with .BR CLONE_THREAD ) will inherit the event. .TP .IR remove_on_exec " (since Linux 5.13)" .\" commit 2e498d0a74e5b88a6689ae1b811f247f91ff188e This closes the event when it starts a new process image by .BR execve (2). .TP .IR sigtrap " (since Linux 5.13)" .\" commit 97ba62b278674293762c3d91f724f1bb922f04e0 This enables synchronous signal delivery of .B SIGTRAP on event overflow. .TP .I wakeup_events .TQ .I wakeup_watermark This union sets how many samples .RI ( wakeup_events ) or bytes .RI ( wakeup_watermark ) happen before an overflow notification happens. Which one is used is selected by the .I watermark bit flag. .IP .I wakeup_events counts only .B PERF_RECORD_SAMPLE record types. To receive overflow notification for all .B PERF_RECORD types choose watermark and set .I wakeup_watermark to 1. .IP Prior to Linux 3.0, setting .\" commit f506b3dc0ec454a16d40cab9ee5d75435b39dc50 .I wakeup_events to 0 resulted in no overflow notifications; more recent kernels treat 0 the same as 1. .TP .IR bp_type " (since Linux 2.6.33)" .\" commit 24f1e32c60c45c89a997c73395b69c8af6f0a84e This chooses the breakpoint type. It is one of: .RS .TP .B HW_BREAKPOINT_EMPTY No breakpoint. .TP .B HW_BREAKPOINT_R Count when we read the memory location. .TP .B HW_BREAKPOINT_W Count when we write the memory location. .TP .B HW_BREAKPOINT_RW Count when we read or write the memory location. .TP .B HW_BREAKPOINT_X Count when we execute code at the memory location. .P The values can be combined via a bitwise or, but the combination of .B HW_BREAKPOINT_R or .B HW_BREAKPOINT_W with .B HW_BREAKPOINT_X is not allowed. .RE .TP .IR bp_addr " (since Linux 2.6.33)" .\" commit 24f1e32c60c45c89a997c73395b69c8af6f0a84e 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. .TP .IR config1 " (since Linux 2.6.39)" .\" commit a7e3ed1e470116c9d12c2f778431a481a6be8ab6 .I 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. .TP .IR bp_len " (since Linux 2.6.33)" .\" commit 24f1e32c60c45c89a997c73395b69c8af6f0a84e .I bp_len is the length of the breakpoint being measured if .I type is .BR PERF_TYPE_BREAKPOINT . Options are .BR HW_BREAKPOINT_LEN_1 , .BR HW_BREAKPOINT_LEN_2 , .BR HW_BREAKPOINT_LEN_4 , and .BR HW_BREAKPOINT_LEN_8 . For an execution breakpoint, set this to .IR sizeof(long) . .TP .IR config2 " (since Linux 2.6.39)" .\" commit a7e3ed1e470116c9d12c2f778431a481a6be8ab6 .I config2 is a further extension of the .I config1 field. .TP .IR branch_sample_type " (since Linux 3.4)" .\" commit bce38cd53e5ddba9cb6d708c4ef3d04a4016ec7e If .B PERF_SAMPLE_BRANCH_STACK is enabled, then this specifies what branches to include in the branch record. .IP 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. .RS .TP .B PERF_SAMPLE_BRANCH_USER Branch target is in user space. .TP .B PERF_SAMPLE_BRANCH_KERNEL Branch target is in kernel space. .TP .B PERF_SAMPLE_BRANCH_HV Branch target is in hypervisor. .TP .B PERF_SAMPLE_BRANCH_PLM_ALL A convenience value that is the three preceding values ORed together. .P In addition to the privilege value, at least one or more of the following bits must be set. .TP .B PERF_SAMPLE_BRANCH_ANY Any branch type. .TP .B PERF_SAMPLE_BRANCH_ANY_CALL Any call branch (includes direct calls, indirect calls, and far jumps). .TP .B PERF_SAMPLE_BRANCH_IND_CALL Indirect calls. .TP .BR PERF_SAMPLE_BRANCH_CALL " (since Linux 4.4)" .\" commit c229bf9dc179d2023e185c0f705bdf68484c1e73 Direct calls. .TP .B PERF_SAMPLE_BRANCH_ANY_RETURN Any return branch. .TP .BR PERF_SAMPLE_BRANCH_IND_JUMP " (since Linux 4.2)" .\" commit c9fdfa14c3792c0160849c484e83aa57afd80ccc Indirect jumps. .TP .BR PERF_SAMPLE_BRANCH_COND " (since Linux 3.16)" .\" commit bac52139f0b7ab31330e98fd87fc5a2664951050 Conditional branches. .TP .BR PERF_SAMPLE_BRANCH_ABORT_TX " (since Linux 3.11)" .\" commit 135c5612c460f89657c4698fe2ea753f6f667963 Transactional memory aborts. .TP .BR PERF_SAMPLE_BRANCH_IN_TX " (since Linux 3.11)" .\" commit 135c5612c460f89657c4698fe2ea753f6f667963 Branch in transactional memory transaction. .TP .BR PERF_SAMPLE_BRANCH_NO_TX " (since Linux 3.11)" .\" commit 135c5612c460f89657c4698fe2ea753f6f667963 Branch not in transactional memory transaction. .BR PERF_SAMPLE_BRANCH_CALL_STACK " (since Linux 4.1)" .\" commit 2c44b1936bb3b135a3fac8b3493394d42e51cf70 Branch is part of a hardware-generated call stack. This requires hardware support, currently only found on Intel x86 Haswell or newer. .RE .TP .IR sample_regs_user " (since Linux 3.7)" .\" commit 4018994f3d8785275ef0e7391b75c3462c029e56 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 .IR arch/ARCH/include/uapi/asm/perf_regs.h . .TP .IR sample_stack_user " (since Linux 3.7)" .\" commit c5ebcedb566ef17bda7b02686e0d658a7bb42ee7 This defines the size of the user stack to dump if .B PERF_SAMPLE_STACK_USER is specified. .TP .IR clockid " (since Linux 4.1)" .\" commit 34f439278cef7b1177f8ce24f9fc81dfc6221d3b If .I use_clockid is set, then this field selects which internal Linux timer to use for timestamps. The available timers are defined in .IR linux/time.h , with .BR CLOCK_MONOTONIC , .BR CLOCK_MONOTONIC_RAW , .BR CLOCK_REALTIME , .BR CLOCK_BOOTTIME , and .B CLOCK_TAI currently supported. .TP .IR aux_watermark " (since Linux 4.1)" .\" commit 1a5941312414c71dece6717da9a0fa1303127afa This specifies how much data is required to trigger a .B PERF_RECORD_AUX sample. .TP .IR sample_max_stack " (since Linux 4.8)" .\" commit 97c79a38cd454602645f0470ffb444b3b75ce574 When .I sample_type includes .BR PERF_SAMPLE_CALLCHAIN , this field specifies how many stack frames to report when generating the callchain. .TP .IR aux_sample_size " (since Linux 5.5)" .\" commit a4faf00d994c40e64f656805ac375c65e324eefb When .B PERF_SAMPLE_AUX flag is set, specify the desired size of AUX data. Note that it can get smaller data than the specified size. .TP .IR sig_data " (since Linux 5.13)" .\" commit 97ba62b278674293762c3d91f724f1bb922f04e0 This data will be copied to user's signal handler (through .I si_perf in the .IR siginfo_t ) to disambiguate which event triggered the signal. .SS Reading results Once a .BR 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 .I read_format field in the .I attr structure at open time. .P If you attempt to read into a buffer that is not big enough to hold the data, the error .B ENOSPC results. .P Here is the layout of the data returned by a read: .IP \[bu] 3 If .B PERF_FORMAT_GROUP was specified to allow reading all events in a group at once: .IP .in +4n .EX 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]; }; .EE .in .IP \[bu] If .B PERF_FORMAT_GROUP was .I not specified: .IP .in +4n .EX 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 */ }; .EE .in .P The values read are as follows: .TP .I nr The number of events in this file descriptor. Available only if .B PERF_FORMAT_GROUP was specified. .TP .I time_enabled .TQ .I time_running 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 .I time_enabled and .I time running values can be used to scale an estimated value for the count. .TP .I value An unsigned 64-bit value containing the counter result. .TP .I id A globally unique value for this particular event; only present if .B PERF_FORMAT_ID was specified in .IR read_format . .TP .I lost The number of lost samples of this event; only present if .B PERF_FORMAT_LOST was specified in .IR read_format . .SS MMAP layout When using .BR perf_event_open () in sampled mode, asynchronous events (like counter overflow or .B PROT_EXEC mmap tracking) are logged into a ring-buffer. This ring-buffer is created and accessed through .BR mmap (2). .P The mmap size should be 1+2\[ha]n pages, where the first page is a metadata page .RI ( "struct perf_event_mmap_page" ) that contains various bits of information such as where the ring-buffer head is. .P 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. .P The structure of the first metadata mmap page is as follows: .P .in +4n .EX 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; \& } .EE .in .P The following list describes the fields in the .I perf_event_mmap_page structure in more detail: .TP .I version Version number of this structure. .TP .I compat_version The lowest version this is compatible with. .TP .I lock A seqlock for synchronization. .TP .I index A unique hardware counter identifier. .TP .I offset When using rdpmc for reads this offset value must be added to the one returned by rdpmc to get the current total event count. .TP .I time_enabled Time the event was active. .TP .I time_running Time the event was running. .TP .IR cap_usr_time " / " cap_usr_rdpmc " / " cap_bit0 " (since Linux 3.4)" .\" commit c7206205d00ab375839bd6c7ddb247d600693c09 There was a bug in the definition of .I cap_usr_time and .I 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 .I cap_usr_time or .I cap_usr_rdpmc were actually set. .IP Starting with Linux 3.12, these are renamed to .\" commit fa7315871046b9a4c48627905691dbde57e51033 .I cap_bit0 and you should use the .I cap_user_time and .I cap_user_rdpmc fields instead. .TP .IR cap_bit0_is_deprecated " (since Linux 3.12)" .\" commit fa7315871046b9a4c48627905691dbde57e51033 If set, this bit indicates that the kernel supports the properly separated .I cap_user_time and .I cap_user_rdpmc bits. .IP If not-set, it indicates an older kernel where .I cap_usr_time and .I cap_usr_rdpmc map to the same bit and thus both features should be used with caution. .TP .IR cap_user_rdpmc " (since Linux 3.12)" .\" commit fa7315871046b9a4c48627905691dbde57e51033 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: .IP .in +4n .EX 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); .EE .in .TP .IR cap_user_time " (since Linux 3.12)" .\" commit fa7315871046b9a4c48627905691dbde57e51033 This bit indicates the hardware has a constant, nonstop timestamp counter (TSC on x86). .TP .IR cap_user_time_zero " (since Linux 3.12)" .\" commit fa7315871046b9a4c48627905691dbde57e51033 Indicates the presence of .I time_zero which allows mapping timestamp values to the hardware clock. .TP .I pmc_width If .IR 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: .IP .in +4n .EX pmc <<= 64 \- pmc_width; pmc >>= 64 \- pmc_width; // signed shift right count += pmc; .EE .in .TP .I time_shift .TQ .I time_mult .TQ .I time_offset .IP If .IR cap_usr_time , these fields can be used to compute the time delta since .I time_enabled (in nanoseconds) using rdtsc or similar. .IP .in +4n .EX 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); .EE .in .IP Where .IR time_offset , .IR time_mult , .IR time_shift , and .I 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: .IP .in +4n .EX enabled += delta; if (idx) running += delta; quot = count / running; rem = count % running; count = quot * enabled + (rem * enabled) / running; .EE .in .TP .IR time_zero " (since Linux 3.12)" .\" commit fa7315871046b9a4c48627905691dbde57e51033 .IP If .I cap_usr_time_zero is set, then the hardware clock (the TSC timestamp counter on x86) can be calculated from the .IR time_zero , .IR time_mult , and .I time_shift values: .IP .in +4n .EX time = timestamp \- time_zero; quot = time / time_mult; rem = time % time_mult; cyc = (quot << time_shift) + (rem << time_shift) / time_mult; .EE .in .IP And vice versa: .IP .in +4n .EX quot = cyc >> time_shift; rem = cyc & (((u64)1 << time_shift) \- 1); timestamp = time_zero + quot * time_mult + ((rem * time_mult) >> time_shift); .EE .in .TP .I data_head 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. .IP On SMP-capable platforms, after reading the .I data_head value, user space should issue an rmb(). .TP .I data_tail When the mapping is .BR PROT_WRITE , the .I 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. .TP .IR data_offset " (since Linux 4.1)" .\" commit e8c6deac69629c0cb97c3d3272f8631ef17f8f0f Contains the offset of the location in the mmap buffer where perf sample data begins. .TP .IR data_size " (since Linux 4.1)" .\" commit e8c6deac69629c0cb97c3d3272f8631ef17f8f0f Contains the size of the perf sample region within the mmap buffer. .TP .I aux_head .TQ .I aux_tail .TQ .I aux_offset .TQ .I aux_size " (since Linux 4.1)" .\" commit 45bfb2e50471abbbfd83d40d28c986078b0d24ff The AUX region allows .BR 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. .IP To set up an AUX area, first .I aux_offset needs to be set with an offset greater than .IR data_offset + data_size and .I 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 .B RLIMIT_MEMLOCK resource limit (see .BR setrlimit (2)), and also as part of the .I perf_event_mlock_kb allowance. .IP 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. .IP The .I aux_head and .I aux_tail ring buffer pointers have the same behavior and ordering rules as the previous described .I data_head and .IR data_tail . .P The following 2^n ring-buffer pages have the layout described below. .P If .I 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 .B PERF_RECORD_SAMPLE below, it will be stashed just after the .I 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. .P The mmap values start with a header: .P .in +4n .EX struct perf_event_header { __u32 type; __u16 misc; __u16 size; }; .EE .in .P Below, we describe the .I perf_event_header fields in more detail. For ease of reading, the fields with shorter descriptions are presented first. .TP .I size This indicates the size of the record. .TP .I misc The .I misc field contains additional information about the sample. .IP The CPU mode can be determined from this value by masking with .B 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): .RS .TP .B PERF_RECORD_MISC_CPUMODE_UNKNOWN Unknown CPU mode. .TP .B PERF_RECORD_MISC_KERNEL Sample happened in the kernel. .TP .B PERF_RECORD_MISC_USER Sample happened in user code. .TP .B PERF_RECORD_MISC_HYPERVISOR Sample happened in the hypervisor. .TP .BR PERF_RECORD_MISC_GUEST_KERNEL " (since Linux 2.6.35)" .\" commit 39447b386c846bbf1c56f6403c5282837486200f Sample happened in the guest kernel. .TP .B PERF_RECORD_MISC_GUEST_USER " (since Linux 2.6.35)" .\" commit 39447b386c846bbf1c56f6403c5282837486200f Sample happened in guest user code. .RE .P .RS Since the following three statuses are generated by different record types, they alias to the same bit: .TP .BR PERF_RECORD_MISC_MMAP_DATA " (since Linux 3.10)" .\" commit 2fe85427e3bf65d791700d065132772fc26e4d75 This is set when the mapping is not executable; otherwise the mapping is executable. .TP .BR PERF_RECORD_MISC_COMM_EXEC " (since Linux 3.16)" .\" commit 82b897782d10fcc4930c9d4a15b175348fdd2871 This is set for a .B PERF_RECORD_COMM record on kernels more recent than Linux 3.16 if a process name change was caused by an .BR execve (2) system call. .TP .BR PERF_RECORD_MISC_SWITCH_OUT " (since Linux 4.3)" .\" commit 45ac1403f564f411c6a383a2448688ba8dd705a4 When a .B PERF_RECORD_SWITCH or .B 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). .RE .P .RS In addition, the following bits can be set: .TP .B PERF_RECORD_MISC_EXACT_IP This indicates that the content of .B PERF_SAMPLE_IP points to the actual instruction that triggered the event. See also .IR perf_event_attr.precise_ip . .TP .BR PERF_RECORD_MISC_SWITCH_OUT_PREEMPT " (since Linux 4.17)" .\" commit 101592b4904ecf6b8ed2a4784d41d180319d95a1 When a .B PERF_RECORD_SWITCH or .B PERF_RECORD_SWITCH_CPU_WIDE record is generated, this indicates the context switch was a preemption. .TP .BR PERF_RECORD_MISC_MMAP_BUILD_ID " (since Linux 5.12)" .\" commit 88a16a1309333e43d328621ece3e9fa37027e8eb This indicates that the content of .B PERF_SAMPLE_MMAP2 contains build-ID data instead of device major and minor numbers as well as the inode number. .TP .BR PERF_RECORD_MISC_EXT_RESERVED " (since Linux 2.6.35)" .\" commit 1676b8a077c352085d52578fb4f29350b58b6e74 This indicates there is extended data available (currently not used). .TP .B PERF_RECORD_MISC_PROC_MAP_PARSE_TIMEOUT .\" commit 930e6fcd2bcce9bcd9d4aa7e755678d33f3fe6f4 This bit is not set by the kernel. It is reserved for the user-space perf utility to indicate that .IR /proc/ pid /maps parsing was taking too long and was stopped, and thus the mmap records may be truncated. .RE .TP .I type The .I type value is one of the below. The values in the corresponding record (that follows the header) depend on the .I type selected as shown. .RS .TP 4 .B PERF_RECORD_MMAP The MMAP events record the .B PROT_EXEC mappings so that we can correlate user-space IPs to code. They have the following structure: .IP .in +4n .EX struct { struct perf_event_header header; u32 pid, tid; u64 addr; u64 len; u64 pgoff; char filename[]; }; .EE .in .RS .TP .I pid is the process ID. .TP .I tid is the thread ID. .TP .I addr is the address of the allocated memory. .I len is the length of the allocated memory. .I pgoff is the page offset of the allocated memory. .I filename is a string describing the backing of the allocated memory. .RE .TP .B PERF_RECORD_LOST This record indicates when events are lost. .IP .in +4n .EX struct { struct perf_event_header header; u64 id; u64 lost; struct sample_id sample_id; }; .EE .in .RS .TP .I id is the unique event ID for the samples that were lost. .TP .I lost is the number of events that were lost. .RE .TP .B PERF_RECORD_COMM This record indicates a change in the process name. .IP .in +4n .EX struct { struct perf_event_header header; u32 pid; u32 tid; char comm[]; struct sample_id sample_id; }; .EE .in .RS .TP .I pid is the process ID. .TP .I tid is the thread ID. .TP .I comm is a string containing the new name of the process. .RE .TP .B PERF_RECORD_EXIT This record indicates a process exit event. .IP .in +4n .EX struct { struct perf_event_header header; u32 pid, ppid; u32 tid, ptid; u64 time; struct sample_id sample_id; }; .EE .in .TP .B PERF_RECORD_THROTTLE .TQ .B PERF_RECORD_UNTHROTTLE This record indicates a throttle/unthrottle event. .IP .in +4n .EX struct { struct perf_event_header header; u64 time; u64 id; u64 stream_id; struct sample_id sample_id; }; .EE .in .TP .B PERF_RECORD_FORK This record indicates a fork event. .IP .in +4n .EX struct { struct perf_event_header header; u32 pid, ppid; u32 tid, ptid; u64 time; struct sample_id sample_id; }; .EE .in .TP .B PERF_RECORD_READ This record indicates a read event. .IP .in +4n .EX struct { struct perf_event_header header; u32 pid, tid; struct read_format values; struct sample_id sample_id; }; .EE .in .TP .B PERF_RECORD_SAMPLE This record indicates a sample. .IP .in +4n .EX 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 */ }; .EE .in .RS 4 .TP 4 .I sample_id If .B PERF_SAMPLE_IDENTIFIER is enabled, a 64-bit unique ID is included. This is a duplication of the .B PERF_SAMPLE_ID .I id value, but included at the beginning of the sample so parsers can easily obtain the value. .TP .I ip If .B PERF_SAMPLE_IP is enabled, then a 64-bit instruction pointer value is included. .TP .I pid .TQ .I tid If .B PERF_SAMPLE_TID is enabled, then a 32-bit process ID and 32-bit thread ID are included. .TP .I time If .B 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. .TP .I addr If .B 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. .TP .I id If .B 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 .BR PERF_FORMAT_ID . .TP .I stream_id If .B PERF_SAMPLE_STREAM_ID is enabled, a 64-bit unique ID is included. Unlike .B PERF_SAMPLE_ID the actual ID is returned, not the group leader. This ID is the same as the one returned by .BR PERF_FORMAT_ID . .TP .I cpu .TQ .I res If .B 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. .TP .I period If .B PERF_SAMPLE_PERIOD is enabled, a 64-bit value indicating the current sampling period is written. .TP .I v If .B 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 .I read_format value used at .BR perf_event_open () time. .TP .I nr .TQ .I ips[nr] If .B 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. .TP .I size .TQ .I data[size] If .B 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. .IP 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. .TP .I bnr .TQ .I lbr[bnr] If .B PERF_SAMPLE_BRANCH_STACK is enabled, then a 64-bit value indicating the number of records is included, followed by .I bnr .I perf_branch_entry structures which each include the fields: .RS .TP .I from This indicates the source instruction (may not be a branch). .TP .I to The branch target. .TP .I mispred The branch target was mispredicted. .TP .I predicted The branch target was predicted. .TP .IR in_tx " (since Linux 3.11)" .\" commit 135c5612c460f89657c4698fe2ea753f6f667963 The branch was in a transactional memory transaction. .TP .IR abort " (since Linux 3.11)" .\" commit 135c5612c460f89657c4698fe2ea753f6f667963 The branch was in an aborted transactional memory transaction. .TP .IR cycles " (since Linux 4.3)" .\" commit 71ef3c6b9d4665ee7afbbe4c208a98917dcfc32f This reports the number of cycles elapsed since the previous branch stack update. .P The entries are from most to least recent, so the first entry has the most recent branch. .P Support for .IR mispred , .IR predicted , and .I cycles is optional; if not supported, those values will be 0. .P The type of branches recorded is specified by the .I branch_sample_type field. .RE .TP .I abi .TQ .I regs[weight(mask)] If .B PERF_SAMPLE_REGS_USER is enabled, then the user CPU registers are recorded. .IP The .I abi field is one of .BR PERF_SAMPLE_REGS_ABI_NONE , .BR PERF_SAMPLE_REGS_ABI_32 , or .BR PERF_SAMPLE_REGS_ABI_64 . .IP The .I regs field is an array of the CPU registers that were specified by the .I sample_regs_user attr field. The number of values is the number of bits set in the .I sample_regs_user bit mask. .TP .I size .TQ .I data[size] .TQ .I dyn_size If .B PERF_SAMPLE_STACK_USER is enabled, then the user stack is recorded. This can be used to generate stack backtraces. .I size is the size requested by the user in .I sample_stack_user or else the maximum record size. .I data is the stack data (a raw dump of the memory pointed to by the stack pointer at the time of sampling). .I dyn_size is the amount of data actually dumped (can be less than .IR size ). Note that .I dyn_size is omitted if .I size is 0. .TP .I weight If .B PERF_SAMPLE_WEIGHT or .B 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. .TP .I data_src If .B PERF_SAMPLE_DATA_SRC is enabled, then a 64-bit value is recorded that is made up of the following fields: .RS .TP 4 .I mem_op Type of opcode, a bitwise combination of: .IP .PD 0 .RS .TP 24 .B PERF_MEM_OP_NA Not available .TP .B PERF_MEM_OP_LOAD Load instruction .TP .B PERF_MEM_OP_STORE Store instruction .TP .B PERF_MEM_OP_PFETCH Prefetch .TP .B PERF_MEM_OP_EXEC Executable code .RE .PD .TP .I mem_lvl Memory hierarchy level hit or miss, a bitwise combination of the following, shifted left by .BR PERF_MEM_LVL_SHIFT : .IP .PD 0 .RS .TP 24 .B PERF_MEM_LVL_NA Not available .TP .B PERF_MEM_LVL_HIT Hit .TP .B PERF_MEM_LVL_MISS Miss .TP .B PERF_MEM_LVL_L1 Level 1 cache .TP .B PERF_MEM_LVL_LFB Line fill buffer .TP .B PERF_MEM_LVL_L2 Level 2 cache .TP .B PERF_MEM_LVL_L3 Level 3 cache .TP .B PERF_MEM_LVL_LOC_RAM Local DRAM .TP .B PERF_MEM_LVL_REM_RAM1 Remote DRAM 1 hop .TP .B PERF_MEM_LVL_REM_RAM2 Remote DRAM 2 hops .TP .B PERF_MEM_LVL_REM_CCE1 Remote cache 1 hop .TP .B PERF_MEM_LVL_REM_CCE2 Remote cache 2 hops .TP .B PERF_MEM_LVL_IO I/O memory .TP .B PERF_MEM_LVL_UNC Uncached memory .RE .PD .TP .I mem_snoop Snoop mode, a bitwise combination of the following, shifted left by .BR PERF_MEM_SNOOP_SHIFT : .IP .PD 0 .RS .TP 24 .B PERF_MEM_SNOOP_NA Not available .TP .B PERF_MEM_SNOOP_NONE No snoop .TP .B PERF_MEM_SNOOP_HIT Snoop hit .TP .B PERF_MEM_SNOOP_MISS Snoop miss .TP .B PERF_MEM_SNOOP_HITM Snoop hit modified .RE .PD .TP .I mem_lock Lock instruction, a bitwise combination of the following, shifted left by .BR PERF_MEM_LOCK_SHIFT : .IP .PD 0 .RS .TP 24 .B PERF_MEM_LOCK_NA Not available .TP .B PERF_MEM_LOCK_LOCKED Locked transaction .RE .PD .TP .I mem_dtlb TLB access hit or miss, a bitwise combination of the following, shifted left by .BR PERF_MEM_TLB_SHIFT : .IP .PD 0 .RS .TP 24 .B PERF_MEM_TLB_NA Not available .TP .B PERF_MEM_TLB_HIT Hit .TP .B PERF_MEM_TLB_MISS Miss .TP .B PERF_MEM_TLB_L1 Level 1 TLB .TP .B PERF_MEM_TLB_L2 Level 2 TLB .TP .B PERF_MEM_TLB_WK Hardware walker .TP .B PERF_MEM_TLB_OS OS fault handler .RE .PD .RE .TP .I transaction If the .B PERF_SAMPLE_TRANSACTION flag is set, then a 64-bit field is recorded describing the sources of any transactional memory aborts. .IP The field is a bitwise combination of the following values: .RS .TP .B PERF_TXN_ELISION Abort from an elision type transaction (Intel-CPU-specific). .TP .B PERF_TXN_TRANSACTION Abort from a generic transaction. .TP .B PERF_TXN_SYNC Synchronous abort (related to the reported instruction). .TP .B PERF_TXN_ASYNC Asynchronous abort (not related to the reported instruction). .TP .B PERF_TXN_RETRY Retryable abort (retrying the transaction may have succeeded). .TP .B PERF_TXN_CONFLICT Abort due to memory conflicts with other threads. .TP .B PERF_TXN_CAPACITY_WRITE Abort due to write capacity overflow. .TP .B PERF_TXN_CAPACITY_READ Abort due to read capacity overflow. .RE .IP In addition, a user-specified abort code can be obtained from the high 32 bits of the field by shifting right by .B PERF_TXN_ABORT_SHIFT and masking with the value .BR PERF_TXN_ABORT_MASK . .TP .I abi .TQ .I regs[weight(mask)] If .B PERF_SAMPLE_REGS_INTR is enabled, then the user CPU registers are recorded. .IP The .I abi field is one of .BR PERF_SAMPLE_REGS_ABI_NONE , .BR PERF_SAMPLE_REGS_ABI_32 , or .BR PERF_SAMPLE_REGS_ABI_64 . .IP The .I regs field is an array of the CPU registers that were specified by the .I sample_regs_intr attr field. The number of values is the number of bits set in the .I sample_regs_intr bit mask. .TP .I phys_addr If the .B PERF_SAMPLE_PHYS_ADDR flag is set, then the 64-bit physical address is recorded. .TP .I cgroup If the .B 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 .BR PERF_RECORD_CGROUP . .TP .I data_page_size If the .B PERF_SAMPLE_DATA_PAGE_SIZE flag is set, then the 64-bit page size value of the .B data address is recorded. .TP .I code_page_size If the .B PERF_SAMPLE_CODE_PAGE_SIZE flag is set, then the 64-bit page size value of the .B ip address is recorded. .TP .I size .TQ .IR data [ size ] If .B PERF_SAMPLE_AUX is enabled, a snapshot of the aux buffer is recorded. .RE .TP .B PERF_RECORD_MMAP2 This record includes extended information on .BR mmap (2) calls returning executable mappings. The format is similar to that of the .B PERF_RECORD_MMAP record, but includes extra values that allow uniquely identifying shared mappings. Depending on the .B PERF_RECORD_MISC_MMAP_BUILD_ID bit in the header, the extra values have different layout and meanings. .IP .in +4n .EX 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; }; .EE .in .RS .TP .I pid is the process ID. .TP .I tid is the thread ID. .TP .I addr is the address of the allocated memory. .TP .I len is the length of the allocated memory. .TP .I pgoff is the page offset of the allocated memory. .TP .I maj is the major ID of the underlying device. .TP .I min is the minor ID of the underlying device. .TP .I ino is the inode number. .TP .I ino_generation is the inode generation. .TP .I build_id_size is the actual size of .I build_id field (up to 20). .TP .I build_id is a raw data to identify a binary. .TP .I prot is the protection information. .TP .I flags is the flags information. .TP .I filename is a string describing the backing of the allocated memory. .RE .TP .BR PERF_RECORD_AUX " (since Linux 4.1)" .\" commit 68db7e98c3a6ebe7284b6cf14906ed7c55f3f7f0 This record reports that new data is available in the separate AUX buffer region. .IP .in +4n .EX struct { struct perf_event_header header; u64 aux_offset; u64 aux_size; u64 flags; struct sample_id sample_id; }; .EE .in .RS .TP .I aux_offset offset in the AUX mmap region where the new data begins. .TP .I aux_size size of the data made available. .TP .I flags describes the AUX update. .RS .TP .B PERF_AUX_FLAG_TRUNCATED if set, then the data returned was truncated to fit the available buffer size. .TP .B PERF_AUX_FLAG_OVERWRITE .\" commit 2023a0d2829e521fe6ad6b9907f3f90bfbf57142 if set, then the data returned has overwritten previous data. .RE .RE .TP .BR PERF_RECORD_ITRACE_START " (since Linux 4.1)" .\" ec0d7729bbaed4b9d2d3fada693278e13a3d1368 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. .IP .in +4n .EX struct { struct perf_event_header header; u32 pid; u32 tid; }; .EE .in .RS .TP .I pid process ID of the thread starting an instruction trace. .TP .I tid thread ID of the thread starting an instruction trace. .RE .TP .BR PERF_RECORD_LOST_SAMPLES " (since Linux 4.2)" .\" f38b0dbb491a6987e198aa6b428db8692a6480f8 When using hardware sampling (such as Intel PEBS) this record indicates some number of samples that may have been lost. .IP .in +4n .EX struct { struct perf_event_header header; u64 lost; struct sample_id sample_id; }; .EE .in .RS .TP .I lost the number of potentially lost samples. .RE .TP .BR PERF_RECORD_SWITCH " (since Linux 4.3)" .\" commit 45ac1403f564f411c6a383a2448688ba8dd705a4 This record indicates a context switch has happened. The .B PERF_RECORD_MISC_SWITCH_OUT bit in the .I misc field indicates whether it was a context switch into or away from the current process. .IP .in +4n .EX struct { struct perf_event_header header; struct sample_id sample_id; }; .EE .in .TP .BR PERF_RECORD_SWITCH_CPU_WIDE " (since Linux 4.3)" .\" commit 45ac1403f564f411c6a383a2448688ba8dd705a4 As with .B 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 .B PERF_RECORD_MISC_SWITCH_OUT bit in the .I misc field indicates whether it was a context switch into or away from the current process. .IP .in +4n .EX struct { struct perf_event_header header; u32 next_prev_pid; u32 next_prev_tid; struct sample_id sample_id; }; .EE .in .RS .TP .I next_prev_pid The process ID of the previous (if switching in) or next (if switching out) process on the CPU. .TP .I next_prev_tid The thread ID of the previous (if switching in) or next (if switching out) thread on the CPU. .RE .TP .BR PERF_RECORD_NAMESPACES " (since Linux 4.11)" .\" commit e422267322cd319e2695a535e47c5b1feeac45eb This record includes various namespace information of a process. .IP .in +4n .EX struct { struct perf_event_header header; u32 pid; u32 tid; u64 nr_namespaces; struct { u64 dev, inode } [nr_namespaces]; struct sample_id sample_id; }; .EE .in .RS .TP .I pid is the process ID .TP .I tid is the thread ID .TP .I nr_namespace is the number of namespaces in this record .RE .IP Each namespace has .I dev and .I inode fields and is recorded in the fixed position like below: .RS .TP .BR NET_NS_INDEX = 0 Network namespace .TP .BR UTS_NS_INDEX = 1 UTS namespace .TP .BR IPC_NS_INDEX = 2 IPC namespace .TP .BR PID_NS_INDEX = 3 PID namespace .TP .BR USER_NS_INDEX = 4 User namespace .TP .BR MNT_NS_INDEX = 5 Mount namespace .TP .BR CGROUP_NS_INDEX = 6 Cgroup namespace .RE .TP .BR PERF_RECORD_KSYMBOL " (since Linux 5.0)" .\" commit 76193a94522f1d4edf2447a536f3f796ce56343b This record indicates kernel symbol register/unregister events. .IP .in +4n .EX struct { struct perf_event_header header; u64 addr; u32 len; u16 ksym_type; u16 flags; char name[]; struct sample_id sample_id; }; .EE .in .RS .TP .I addr is the address of the kernel symbol. .TP .I len is the length of the kernel symbol. .TP .I ksym_type is the type of the kernel symbol. Currently the following types are available: .RS .TP .B PERF_RECORD_KSYMBOL_TYPE_BPF The kernel symbol is a BPF function. .RE .TP .I flags If the .B PERF_RECORD_KSYMBOL_FLAGS_UNREGISTER is set, then this event is for unregistering the kernel symbol. .RE .TP .BR PERF_RECORD_BPF_EVENT " (since Linux 5.0)" .\" commit 6ee52e2a3fe4ea35520720736e6791df1fb67106 This record indicates BPF program is loaded or unloaded. .IP .in +4n .EX struct { struct perf_event_header header; u16 type; u16 flags; u32 id; u8 tag[BPF_TAG_SIZE]; struct sample_id sample_id; }; .EE .in .RS .TP .I type is one of the following values: .RS .TP .B PERF_BPF_EVENT_PROG_LOAD A BPF program is loaded .TP .B PERF_BPF_EVENT_PROG_UNLOAD A BPF program is unloaded .RE .TP .I id is the ID of the BPF program. .TP .I tag is the tag of the BPF program. Currently, .B BPF_TAG_SIZE is defined as 8. .RE .TP .BR PERF_RECORD_CGROUP " (since Linux 5.7)" .\" commit 96aaab686505c449e24d76e76507290dcc30e008 This record indicates a new cgroup is created and activated. .IP .in +4n .EX struct { struct perf_event_header header; u64 id; char path[]; struct sample_id sample_id; }; .EE .in .RS .TP .I id is the cgroup identifier. This can be also retrieved by .BR name_to_handle_at (2) on the cgroup path (as a file handle). .TP .I path is the path of the cgroup from the root. .RE .TP .BR PERF_RECORD_TEXT_POKE " (since Linux 5.8)" .\" commit e17d43b93e544f5016c0251d2074c15568d5d963 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. .IP .in +4n .EX struct { struct perf_event_header header; u64 addr; u16 old_len; u16 new_len; u8 bytes[]; struct sample_id sample_id; }; .EE .in .RS .TP .I addr is the address of the change .TP .I old_len is the old length .TP .I new_len is the new length .TP .I bytes contains old bytes immediately followed by new bytes. .RE .RE .SS 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 .BR poll (2), .BR select (2), or .BR 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 .B F_SETOWN and .B F_SETSIG operations in .BR fcntl (2). .P Overflows are generated only by sampling events .RI ( sample_period must have a nonzero value). .P There are two ways to generate overflow notifications. .P The first is to set a .I wakeup_events or .I 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, .B POLL_IN is indicated. .P The other way is by use of the .B PERF_EVENT_IOC_REFRESH ioctl. This ioctl adds to a counter that decrements each time the event overflows. When nonzero, .B POLL_IN is indicated, but once the counter reaches 0 .B POLL_HUP is indicated and the underlying event is disabled. .P 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. .\" See https://lkml.org/lkml/2011/5/24/337 .P Starting with Linux 3.18, .\" commit 179033b3e064d2cd3f5f9945e76b0a0f0fbf4883 .B POLL_HUP is indicated if the event being monitored is attached to a different process and that process exits. .SS rdpmc instruction Starting with Linux 3.4 on x86, you can use the .\" commit c7206205d00ab375839bd6c7ddb247d600693c09 .I rdpmc instruction to get low-latency reads without having to enter the kernel. Note that using .I rdpmc is not necessarily faster than other methods for reading event values. .P Support for this can be detected with the .I cap_usr_rdpmc field in the mmap page; documentation on how to calculate event values can be found in that section. .P 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, .\" 7911d3f7af14a614617e38245fedf98a724e46a9 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 .IR /sys/devices/cpu/rdpmc . .SS perf_event ioctl calls Various ioctls act on .BR perf_event_open () file descriptors: .TP .B PERF_EVENT_IOC_ENABLE This enables the individual event or event group specified by the file descriptor argument. .IP If the .B 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). .TP .B PERF_EVENT_IOC_DISABLE This disables the individual counter or event group specified by the file descriptor argument. .IP 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. .IP If the .B 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). .TP .B PERF_EVENT_IOC_REFRESH 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 .B POLL_IN set will happen on each overflow until the count reaches 0; when that happens a notification with .B POLL_HUP set is sent and the event is disabled. Using an argument of 0 is considered undefined behavior. .TP .B PERF_EVENT_IOC_RESET 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 .I time_enabled or .I time_running values. .IP If the .B 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). .TP .B PERF_EVENT_IOC_PERIOD This updates the overflow period for the event. .IP Since Linux 3.7 (on ARM) .\" commit 3581fe0ef37ce12ac7a4f74831168352ae848edc and Linux 3.14 (all other architectures), .\" commit bad7192b842c83e580747ca57104dd51fe08c223 the new period takes effect immediately. On older kernels, the new period did not take effect until after the next overflow. .IP The argument is a pointer to a 64-bit value containing the desired new period. .IP Prior to Linux 2.6.36, .\" commit ad0cf3478de8677f720ee06393b3147819568d6a this ioctl always failed due to a bug in the kernel. .TP .B PERF_EVENT_IOC_SET_OUTPUT 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. .IP The argument specifies the desired file descriptor, or \-1 if output should be ignored. .TP .BR PERF_EVENT_IOC_SET_FILTER " (since Linux 2.6.33)" .\" commit 6fb2915df7f0747d9044da9dbff5b46dc2e20830 This adds an ftrace filter to this event. .IP The argument is a pointer to the desired ftrace filter. .TP .BR PERF_EVENT_IOC_ID " (since Linux 3.12)" .\" commit cf4957f17f2a89984915ea808876d9c82225b862 This returns the event ID value for the given event file descriptor. .IP The argument is a pointer to a 64-bit unsigned integer to hold the result. .TP .BR PERF_EVENT_IOC_SET_BPF " (since Linux 4.1)" .\" commit 2541517c32be2531e0da59dfd7efc1ce844644f5 This allows attaching a Berkeley Packet Filter (BPF) program to an existing kprobe tracepoint event. You need .B CAP_PERFMON (since Linux 5.8) or .B CAP_SYS_ADMIN privileges to use this ioctl. .IP The argument is a BPF program file descriptor that was created by a previous .BR bpf (2) system call. .TP .BR PERF_EVENT_IOC_PAUSE_OUTPUT " (since Linux 4.7)" .\" commit 86e7972f690c1017fd086cdfe53d8524e68c661c 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 .B 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. .IP The argument is an unsigned 32-bit integer. A nonzero value pauses the ring-buffer, while a zero value resumes the ring-buffer. .TP .BR PERF_EVENT_MODIFY_ATTRIBUTES " (since Linux 4.17)" .\" commit 32ff77e8cc9e66cc4fb38098f64fd54cc8f54573 This allows modifying an existing event without the overhead of closing and reopening a new event. Currently this is supported only for breakpoint events. .IP The argument is a pointer to a .I perf_event_attr structure containing the updated event settings. .TP .BR PERF_EVENT_IOC_QUERY_BPF " (since Linux 4.16)" .\" commit f371b304f12e31fe30207c41ca7754564e0ea4dc 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 .B CAP_PERFMON (since Linux 5.8) or .B CAP_SYS_ADMIN privileges to use this ioctl. .IP The argument is a pointer to a structure .in +4n .EX struct perf_event_query_bpf { __u32 ids_len; __u32 prog_cnt; __u32 ids[0]; }; .EE .in .IP The .I ids_len field indicates the number of ids that can fit in the provided .I ids array. The .I prog_cnt value is filled in by the kernel with the number of attached BPF programs. The .I 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 .B ENOSPC and .I ids_len will indicate the number of program IDs that were successfully copied. .\" .SS Using prctl(2) A process can enable or disable all currently open event groups using the .BR prctl (2) .B PR_TASK_PERF_EVENTS_ENABLE and .B 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. .SS perf_event related configuration files Files in .I /proc/sys/kernel/ .RS 4 .TP .I /proc/sys/kernel/perf_event_paranoid The .I perf_event_paranoid file can be set to restrict access to the performance counters. .IP .PD 0 .RS .TP .B 2 allow only user-space measurements (default since Linux 4.6). .\" default changed in commit 0161028b7c8aebef64194d3d73e43bc3b53b5c66 .TP .B 1 allow both kernel and user measurements (default before Linux 4.6). .TP .B 0 allow access to CPU-specific data but not raw tracepoint samples. .TP .B \-1 no restrictions. .RE .PD .IP The existence of the .I perf_event_paranoid file is the official method for determining if a kernel supports .BR perf_event_open (). .TP .I /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). .TP .I /proc/sys/kernel/perf_event_max_stack .\" Introduced in c5dfd78eb79851e278b7973031b9ca363da87a7e This file sets the maximum depth of stack frame entries reported when generating a call trace. .TP .I /proc/sys/kernel/perf_event_mlock_kb Maximum number of pages an unprivileged user can .BR mlock (2). The default is 516 (kB). .RE .P Files in .I /sys/bus/event_source/devices/ .P .RS 4 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 .IR /sys/bus/event_source/devices/ . Each subdirectory corresponds to a different PMU. .TP .IR /sys/bus/event_source/devices/*/type " (since Linux 2.6.38)" .\" commit abe43400579d5de0078c2d3a760e6598e183f871 This contains an integer that can be used in the .I type field of .I perf_event_attr to indicate that you wish to use this PMU. .TP .IR /sys/bus/event_source/devices/cpu/rdpmc " (since Linux 3.4)" .\" commit 0c9d42ed4cee2aa1dfc3a260b741baae8615744f 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. .IP As of Linux 4.0 .\" a66734297f78707ce39d756b656bfae861d53f62 .\" 7911d3f7af14a614617e38245fedf98a724e46a9 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. .TP .IR /sys/bus/event_source/devices/*/format/ " (since Linux 3.4)" .\" commit 641cc938815dfd09f8fa1ec72deb814f0938ac33 This subdirectory contains information on the architecture-specific subfields available for programming the various .I config fields in the .I perf_event_attr struct. .IP 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 .I event may contain the value .I config1:1,6\-10,44 which indicates that event is an attribute that occupies bits 1,6\[en]10, and 44 of .IR perf_event_attr::config1 . .TP .IR /sys/bus/event_source/devices/*/events/ " (since Linux 3.4)" .\" commit 641cc938815dfd09f8fa1ec72deb814f0938ac33 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 .I ./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. .IP 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: .IR event=0x2,inv,ldlat=3 . .TP .I /sys/bus/event_source/devices/*/uevent This file is the standard kernel device interface for injecting hotplug events. .TP .IR /sys/bus/event_source/devices/*/cpumask " (since Linux 3.7)" .\" commit 314d9f63f385096580e9e2a06eaa0745d92fe4ac The .I 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. .RE .SH RETURN VALUE On success, .BR perf_event_open () returns the new file descriptor. On error, \-1 is returned and .I errno is set to indicate the error. .SH ERRORS The errors returned by .BR perf_event_open () can be inconsistent, and may vary across processor architectures and performance monitoring units. .TP .B E2BIG Returned if the .I perf_event_attr .I size value is too small (smaller than .BR 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 .B E2BIG is returned, the .I perf_event_attr .I size field is overwritten by the kernel to be the size of the structure it was expecting. .TP .B EACCES Returned when the requested event requires .B CAP_PERFMON (since Linux 5.8) or .B 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 .I pid argument as \-1); and not setting .I exclude_kernel when the paranoid setting requires it. .TP .B EBADF Returned if the .I group_fd file descriptor is not valid, or, if .B PERF_FLAG_PID_CGROUP is set, the cgroup file descriptor in .I pid is not valid. .TP .BR EBUSY " (since Linux 4.1)" .\" bed5b25ad9c8a2f5d735ef0bc746ec870c01c1b0 Returned if another event already has exclusive access to the PMU. .TP .B EFAULT Returned if the .I attr pointer points at an invalid memory address. .TP .B EINTR Returned when trying to mix perf and ftrace handling for a uprobe. .TP .B EINVAL Returned if the specified event is invalid. There are many possible reasons for this. A not-exhaustive list: .I sample_freq is higher than the maximum setting; the .I cpu to monitor does not exist; .I read_format is out of range; .I sample_type is out of range; the .I flags value is out of range; .I exclusive or .I pinned set and the event is not a group leader; the event .I 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. .TP .B EMFILE 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. .TP .B ENODEV Returned when the event involves a feature not supported by the current CPU. .TP .B ENOENT Returned if the .I type setting is not valid. This error is also returned for some unsupported generic events. .TP .B ENOSPC Prior to Linux 3.3, if there was not enough room for the event, .\" commit aa2bc1ade59003a379ffc485d6da2d92ea3370a6 .B ENOSPC was returned. In Linux 3.3, this was changed to .BR EINVAL . .B ENOSPC is still returned if you try to add more breakpoint events than supported by the hardware. .TP .B ENOSYS Returned if .B PERF_SAMPLE_STACK_USER is set in .I sample_type and it is not supported by hardware. .TP .B EOPNOTSUPP 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. .TP .BR EOVERFLOW " (since Linux 4.8)" .\" 97c79a38cd454602645f0470ffb444b3b75ce574 Returned if .B PERF_SAMPLE_CALLCHAIN is requested and .I sample_max_stack is larger than the maximum specified in .IR /proc/sys/kernel/perf_event_max_stack . .TP .B EPERM Returned on many (but not all) architectures when an unsupported .IR exclude_hv ", " exclude_idle ", " exclude_user ", or " exclude_kernel setting is specified. .IP It can also happen, as with .BR EACCES , when the requested event requires .B CAP_PERFMON (since Linux 5.8) or .B 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. .\" commit a4e95fc2cbb31d70a65beffeaf8773f881328c34 .TP .B ESRCH Returned if attempting to attach to a process that does not exist. .SH STANDARDS Linux. .SH HISTORY .BR perf_event_open () was introduced in Linux 2.6.31 but was called .\" commit 0793a61d4df8daeac6492dbf8d2f3e5713caae5e .BR perf_counter_open (). It was renamed in Linux 2.6.32. .\" commit cdd6c482c9ff9c55475ee7392ec8f672eddb7be6 .SH NOTES The official way of knowing if .BR perf_event_open () support is enabled is checking for the existence of the file .IR /proc/sys/kernel/perf_event_paranoid . .P .B 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 .B CAP_PERFMON rather than the much more powerful .B CAP_SYS_ADMIN excludes chances to misuse credentials and makes operations more secure. .B CAP_SYS_ADMIN usage for secure system performance monitoring and observability is discouraged in favor of the .B CAP_PERFMON capability. .SH BUGS The .B F_SETOWN_EX option to .BR fcntl (2) is needed to properly get overflow signals in threads. This was introduced in Linux 2.6.32. .\" commit ba0a6c9f6fceed11c6a99e8326f0477fe383e6b5 .P Prior to Linux 2.6.33 (at least for x86), .\" commit b690081d4d3f6a23541493f1682835c3cd5c54a1 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 .BR perf_event_open (), start, then read before you know for sure you can get valid measurements. .P Prior to Linux 2.6.34, .\" FIXME . cannot find a kernel commit for this one 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. .P Prior to Linux 2.6.34, there was a bug when multiplexing where the wrong results could be returned. .\" commit 45e16a6834b6af098702e5ea6c9a40de42ff77d8 .P 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. .\" commit 38b435b16c36b0d863efcf3f07b34a6fac9873fd .P Prior to Linux 2.6.35, .\" commit 050735b08ca8a016bbace4445fa025b88fee770b .B PERF_FORMAT_GROUP did not work with attached processes. .P 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. .\" commit 4ec8363dfc1451f8c8f86825731fe712798ada02 .P From Linux 2.6.31 to Linux 3.4, the .B 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. .\" commit 724b6daa13e100067c30cfc4d1ad06629609dc4e .P From Linux 3.4 to Linux 3.11, the mmap .\" commit fa7315871046b9a4c48627905691dbde57e51033 .I cap_usr_rdpmc and .I cap_usr_time bits mapped to the same location. Code should migrate to the new .I cap_user_rdpmc and .I cap_user_time fields instead. .P 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. .\" commit f287d332ce835f77a4f5077d2c0ef1e3f9ea42d2 .SH EXAMPLES The following is a short example that measures the total instruction count of a call to .BR printf (3). .P .\" SRC BEGIN (perf_event_open.c) .EX #include #include #include #include #include #include #include #include \& 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\en", 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\en"); \& ioctl(fd, PERF_EVENT_IOC_DISABLE, 0); read(fd, &count, sizeof(count)); \& printf("Used %lld instructions\en", count); \& close(fd); } .EE .\" SRC END .SH SEE ALSO .BR perf (1), .BR fcntl (2), .BR mmap (2), .BR open (2), .BR prctl (2), .BR read (2) .P .I Documentation/admin\-guide/perf\-security.rst in the kernel source tree