Nios® V/II Embedded Design Suite (EDS)
Support for Embedded Development Tools, Processors (SoCs and Nios® V/II processor), Embedded Development Suites (EDSs), Boot and Configuration, Operating Systems, C and C++
12589 Discussions

ARM PMU cycle counter not always incrementing at CPU frequency

Altera_Forum
Honored Contributor II
1,489 Views

Hello, 

 

I want to benchmark my program running on Linux on an Altera Cyclone V SoC board, but it turns out that the values returned from the ARM PMU cycle counter suggest that some sort of CPU frequency scaling is active, which is confusing me. 

 

So at first I tried to find out at which frequency the CPU was running and if any frequency scaling was happening, which turned out to be difficult. 

U-Boot tells me that the CPU is running at 925 MHz during boot, which is fine. But I could not find any way to read the current CPU speed from within Linux, and there seems to be no frequency scaling governor. I compiled Linux following the steps from the RocketBoards website [1], but modified .config to use only one CPU. The output of some commands that I tried is at the end of this post. 

 

The second thing I did was to benchmark some simple code using a cycle counter. I followed a blog post [2] which suggest to use perf [3]. I modified the perf call to measure all threads running on CPU 0. When I benchmarked doing calculations in a loop and measured time and cycles, the result was precisely matching 925 MHz. But when I measured sleep() it was counting much less cycles. 

So I tried reading the ARM PMU cycle counter basically following another code example ([2] and also [4]). The result was the same, during 1 second sleep() only roughly ~10000000 cycles are counted (but 1s * 925 MHz = 925000000 cycles). So I ran a second program to keep CPU usage at 100% while still benchmarking sleep(). The result was then again matching 925 MHz. 

Now I wonder if this implies some sort of frequency scaling? 

 

Even more confusing: when I configured perf to measure only cycles used by the calling (sleeping) thread and read the PMU cycle counter register at the same time, the result of the cycle counter unexpectedly matched the perf result. Even when a second program is utilizing 100% CPU at the same time. Configuring the ARM PMU event count registers [5] to measure cycles did not give different results.  

 

---------- 

Command outputs: 

 

> cpufreq-info 

cpufrequtils 008: cpufreq-info (C) Dominik Brodowski 2004-2009 

Report errors and bugs to cpufreq@vger.kernel.org, please. 

analyzing CPU 0: 

no or unknown cpufreq driver is active on this CPU 

----- 

> cat /proc/cpuinfo 

processor : 0 

model name : ARMv7 Processor rev 0 (v7l) 

BogoMIPS : 921.60 

Features : swp half thumb fastmult vfp edsp thumbee neon vfpv3 tls lpae  

CPU implementer : 0x41 

CPU architecture: 7 

CPU variant : 0x3 

CPU part : 0xc09 

CPU revision : 0 

Hardware : Altera SOCFPGA 

Revision : 0000 

Serial : 0000000000000000 

----- 

> ls /sys/devices/system/cpu/  

cpu0 kernel_max offline online possible present uevent 

----- 

> ls /sys/devices/system/cpu/cpu0/ 

subsystem topology uevent 

 

---------- 

 

[1] http://rocketboards.org/foswiki/view/documentation/gsrd141compilinglinux 

[2] http://neocontra.blogspot.de/2013/05/user-mode-performance-counters-for.html 

[3] http://web.eece.maine.edu/~vweaver/projects/perf_events/perf_event_open.html 

[4] http://stackoverflow.com/questions/3247373/how-to-measure-program-execution-time-in-arm-cortex-a8-processor/3250835#3250835 

[5] http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0464f/biibdhaf.html
0 Kudos
1 Reply
Altera_Forum
Honored Contributor II
642 Views

I already got the answer: Linux uses the WFI instruction to put the core into standby, during which the cycle count will not increase. The kernel parameter nohlt will disable WFI. 

http://community.arm.com/message/29414#29414
0 Kudos
Reply