.TH LIBPFM 3  "September, 2009" "" "Linux Programmer's Manual"

.SH NAME

pfm_get_perf_event_encoding \- encode event for perf_event API

.SH SYNOPSIS

.nf

.B #include <perfmon/pfmlib_perf_event.h>

.sp

.BI "int pfm_get_perf_event_encoding(const char *" str ", int " dfl_plm ", struct perf_event_attr *" attr ", char **" fstr ", int *" idx ");"

.sp

.SH DESCRIPTION

This function can be used in conjunction with the perf_events Linux kernel API which

provides access to hardware performance counters, kernel software counters and tracepoints.

The function takes an event string in \fBstr\fR and a default privilege level mask in \fBdfl_plm\fR

and fills out the relevant parts of the perf_events specific data structure in \fBattr\fR.

This function is \fBdeprecated\fR. It is superseded by \fBpfm_get_os_event_encoding()\fR

with the OS argument set to either \fBPFM_OS_PERF_EVENT\fR or \fBPFM_OS_PERF_EVENT_EXT\fR.

Using this function provides extended support for perf_events. Certain perf_event configuration

option are only available through this new interface.

The following examples illustrates the transition:

.nf

   struct perf_event_attr attr;

   int i, count = 0;

   uint64_t *codes;

   memset(&attr, 0, sizeof(attr));

   ret = pfm_get_perf_event_encoding("RETIRED_INSTRUCTIONS", PFM_PLM3, &attrs, NULL, NULL);

   if (ret != PFM_SUCCESS)

      err(1", cannot get encoding %s", pfm_strerror(ret));

.fi

is equivalent to:

.nf

   #include <perfmon/pfmlib_perf_event.h>

   struct perf_event_attr attr;

   pfm_perf_encode_arg_t arg;

   memset(&arg, 0, sizeof(arg));

   arg.size = sizeof(arg);

   arg.attr = &att;;

   ret = pfm_get_os_event_encoding("RETIRED_INSTRUCTIONS", PFM_PLM3, PFM_OS_PERF, &arg;;

   if (ret != PFM_SUCCESS)

      err(1", cannot get encoding %s", pfm_strerror(ret));

.nf

The \fBdfl_plm\fR cannot be zero, though it may not necessarily be used by the event.

Depending on the event, combination of the following privilege levels may be used:

.TP

.B PFM_PLM3

Measure at privilege level 3. This usually corresponds to user level. On X86, it corresponds

to privilege levels 3, 2, 1. Check the PMU specific man page to verify if this level

is supported by your PMU model.

.TP

.B PFM_PLM2

Measure at privilege level 2. Check the PMU specific man page to verify if this level

is supported by your PMU model.

.TP

.B PFM_PLM1

Measure at privilege level 1. Check the PMU specific man page to verify if this level

is supported by your PMU model.

.TP

.B PFM_PLM0

Measure at privilege level 0. This usually corresponds to kernel level. Check the PMU

specific man page to verify if this level is supported by your PMU model.

.TP

.B PFM_PLMH

Measure at hypervisor privilege level. This is used in conjunction with hardware virtualization.

Check the PMU specific man page to verify if this level is supported by your PMU model.

.PP

If \fBfstr\fR is not NULL, the function will make it point to the fully qualified event string,

i.e., a string with the event name, all unit masks set, and the value of all modifiers.

The library will allocate memory to store the event string but it is the responsibility of the

caller to eventually free that string using free().

If \fBidx\fR is not NULL, it returns the corresponding unique event identifier.

Only select fields are modified by the function, the others are untouched.

The following fields in \fBattr\fR are modified:

.TP

.B type

The type of the event

.TP

.B config

The encoding of the event

.TP

.B exclude_user

Whether or not user level execution should be excluded from monitoring. The definition

of user is PMU model specific.

.TP

.B exclude_kernel

Whether or not kernel level execution should be excluded from monitoring. The definition

of kernel is PMU model specific.

.TP

.B exclude_hv

Whether or not hypervisor level execution should be excluded from monitoring. The definition

of hypervisor is PMU model specific.

.PP

By default, if no privilege level modifier is specified in the event string, the library clears

\fBexclude_user\fR, \fBexclude_kernel\fR and \fBexclude_hv\fR, resulting in the event being

measured at all levels subject to hardware support.

The function is able to work on only one event at a time. For convenience, it accepts

event strings with commas. In that case, it will translate the first event up to the

first comma. This is handy in case tools gets passed events as a comma-separated list.

.SH RETURN

The function returns in \fBattr\fR the perf_event encoding which corresponds to the event

string. If \fBidx\fR is not NULL, then it will contain the unique event identifier upon

successful return. The value \fBPFM_SUCCESS\fR is returned if successful, otherwise a negative

error code is returned.

.SH ERRORS

.TP

.B PFM_ERR_TOOSMALL

The \fBcode\fR argument is too small for the encoding.

.TP

.B PFM_ERR_INVAL

The \fBattr\fR argument is \fBNULL\fR.

.TP

.B PFM_ERR_NOMEM

Not enough memory.

.TP

.B PFM_ERR_NOTFOUND

Event not found.

.TP

.B PFM_ERR_ATTR

Invalid event attribute (unit mask or modifier)

.TP

.B PFM_ERR_ATTR_VAL

Invalid modifier value.

.TP

.B PFM_ERR_ATTR_SET

attribute already set, cannot be changed.

.TP

.B PFM_ERR_ATTR_UMASK

Missing unit mask.

.TP

.B PFM_ERR_ATTR_FEATCOMB

Unit masks or features cannot be combined into a single event.

.SH AUTHOR

Stephane Eranian <eranian@gmail.com>

.SH SEE ALSO

pfm_get_os_event_encoding(3)