.TH "PAPI_ipc" 3 "Thu Dec 14 2023" "Version 7.1.0.0" "PAPI" \" -*- nroff -*-
.ad l
.nh
.SH NAME
PAPI_ipc \- Simplified call to get instructions per cycle, real and processor time\&.  

.SH SYNOPSIS
.br
.PP
.SH "Detailed Description"
.PP 

.PP
\fBC Interface: \fP
.RS 4
#include <\fBpapi\&.h\fP> 
.br
int \fBPAPI_ipc( float *rtime, float *ptime, long long *ins, float *ipc )\fP;
.RE
.PP
\fBParameters\fP
.RS 4
\fI*rtime\fP realtime since the latest call 
.br
\fI*ptime\fP process time since the latest call 
.br
\fI*ins\fP instructions since the latest call 
.br
\fI*ipc\fP incremental instructions per cycle since the latest call
.RE
.PP
\fBReturn values\fP
.RS 4
\fIPAPI_EINVAL\fP The counters were already started by something other than \fBPAPI_ipc()\fP\&. 
.br
\fIPAPI_ENOEVNT\fP The events PAPI_TOT_INS and PAPI_TOT_CYC are not supported\&. 
.br
\fIPAPI_ENOMEM\fP Insufficient memory to complete the operation\&.
.RE
.PP
The first call to \fBPAPI_ipc()\fP will initialize the PAPI interface, set up the counters to monitor PAPI_TOT_INS and PAPI_TOT_CYC events and start the counters\&.
.PP
Subsequent calls will read the counters and return real time, process time, instructions and the IPC rate since the latest call to \fBPAPI_ipc()\fP\&.
.PP
\fBPAPI_ipc()\fP should return a ratio greater than 1\&.0, indicating instruction level parallelism within the chip\&. The larger this ratio the more effeciently the program is running\&. Note that \fBPAPI_ipc()\fP is thread-safe and can therefore be called by multiple threads\&.
.PP
\fBSee also\fP
.RS 4
\fBPAPI_flips_rate()\fP 
.PP
\fBPAPI_flops_rate()\fP 
.PP
\fBPAPI_epc()\fP 
.PP
\fBPAPI_rate_stop()\fP 
.RE
.PP


.SH "Author"
.PP 
Generated automatically by Doxygen for PAPI from the source code\&.