perf_event_open always returns -1

Question

I run the following program which call perf_event_open syscall in: Linux sama-desktop 3.18.0-20-rpi2 #21-Ubuntu SMP PREEMPT Sun Apr 5 01:56:02 UTC 2015 armv7l armv7l armv7l GNU/Linux

The program:

#define _GNU_SOURCE 1

#include <asm/unistd.h>
#include <fcntl.h>
#include <linux/perf_event.h>
#include <signal.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>

long perf_event_open(struct perf_event_attr* event_attr, pid_t pid, int cpu, int group_fd, unsigned long flags)
{
    return syscall(__NR_perf_event_open, event_attr, pid, cpu, group_fd, flags);
}

static void perf_event_handler(int signum, siginfo_t* info, void* ucontext) {
    if(info->si_code != POLL_HUP) {
        // Only POLL_HUP should happen.
        exit(EXIT_FAILURE);
    }

    ioctl(info->si_fd, PERF_EVENT_IOC_REFRESH, 1);
}

int main(int argc, char** argv)
{
    // Configure signal handler
    struct sigaction sa;
    memset(&sa, 0, sizeof(struct sigaction));
    sa.sa_sigaction = perf_event_handler;
    sa.sa_flags = SA_SIGINFO;

    // Setup signal handler
    if (sigaction(SIGIO, &sa, NULL) < 0) {
        fprintf(stderr,"Error setting up signal handler\n");
        perror("sigaction");
        exit(EXIT_FAILURE);
    }

    // Configure perf_event_attr struct
    struct perf_event_attr pe;
    memset(&pe, 0, sizeof(struct perf_event_attr));
    pe.type = PERF_TYPE_HARDWARE;
    pe.size = sizeof(struct perf_event_attr);
    pe.config = PERF_COUNT_HW_INSTRUCTIONS;     // Count retired hardware instructions
    pe.disabled = 1;        // Event is initially disabled
    pe.sample_type = PERF_SAMPLE_IP;
    pe.sample_period = 1000;
    pe.exclude_kernel = 1;  // excluding events that happen in the kernel-space
    pe.exclude_hv = 1;      // excluding events that happen in the hypervisor

    pid_t pid = 0;  // measure the current process/thread
    int cpu = -1;   // measure on any cpu
    int group_fd = -1;
    unsigned long flags = 0;

    int fd = perf_event_open(&pe, pid, cpu, group_fd, flags);
    if (fd == -1) {
        fprintf(stderr, "Error opening leader %llx\n", pe.config);
        perror("perf_event_open");
        exit(EXIT_FAILURE);
    }
    // Setup event handler for overflow signals
    fcntl(fd, F_SETFL, O_NONBLOCK|O_ASYNC);
    fcntl(fd, F_SETSIG, SIGIO);
    fcntl(fd, F_SETOWN, getpid());

    ioctl(fd, PERF_EVENT_IOC_RESET, 0);     // Reset event counter to 0
    ioctl(fd, PERF_EVENT_IOC_REFRESH, 1);   // 

// Start monitoring

    long loopCount = 1000000;
    long c = 0;
    long i = 0;

    // Some sample payload.
    for(i = 0; i < loopCount; i++) {
        c += 1;
    }

// End monitoring

    ioctl(fd, PERF_EVENT_IOC_DISABLE, 0);   // Disable event

    long long counter;
    read(fd, &counter, sizeof(long long));  // Read event counter value

    printf("Used %lld instructions\n", counter);

    close(fd);
}

which returns Error opening leader. I checked the fd and it looks like that it always returns -1.

I used the second example from the perf syscall manual which have the same problem(Error opening leader triggered by fd=-1). Here is the sample code for the perf in the manual:

  #include <stdlib.h>
   #include <stdio.h>
   #include <unistd.h>
   #include <string.h>
   #include <sys/ioctl.h>
   #include <linux/perf_event.h>
   #include <asm/unistd.h>

   static long
   perf_event_open(struct perf_event_attr *hw_event, pid_t pid,
                   int cpu, int group_fd, unsigned long flags)
   {
       int ret;

       ret = syscall(__NR_perf_event_open, hw_event, pid, cpu,
                      group_fd, flags);
       return ret;
   }

   int
   main(int argc, char **argv)
   {
       struct perf_event_attr pe;
       long long count;
       int fd;

       memset(&pe, 0, sizeof(struct perf_event_attr));
       pe.type = PERF_TYPE_HARDWARE;
       pe.size = sizeof(struct perf_event_attr);
       pe.config = PERF_COUNT_HW_INSTRUCTIONS;
       pe.disabled = 1;
       pe.exclude_kernel = 1;
       pe.exclude_hv = 1;

       fd = perf_event_open(&pe, 0, -1, -1, 0);
       if (fd == -1) {
          fprintf(stderr, "Error opening leader %llx\n", pe.config);
          exit(EXIT_FAILURE);
       }

       ioctl(fd, PERF_EVENT_IOC_RESET, 0);
       ioctl(fd, PERF_EVENT_IOC_ENABLE, 0);

       printf("Measuring instruction count for this printf\n");

       ioctl(fd, PERF_EVENT_IOC_DISABLE, 0);
       read(fd, &count, sizeof(long long));

       printf("Used %lld instructions\n", count);

       close(fd);
   }

I also made my own code to just check if the perf having problem accessing the PMU registers. Therefor, I made a kernel module to enable user mode access to PMU registers.

the kernel mode only execute the following:

    asm volatile("mrc p15, 0, %0, c9, c14, 0" :: "r"(1));
    asm volatile("mcr p15, 0, %0, c9, c14, 0" :: "r"(1));

and then I tried to run the perf_event_open

init(void)
{
        static struct perf_event_attr attr;
        attr.type = PERF_TYPE_HARDWARE;
//      attr.config = PERF_COUNT_HW_INSTRUCTIONS;
        attr.config = PERF_COUNT_HW_CPU_CYCLES;
        fddev = syscall(__NR_perf_event_open, &attr, 0, -1, -1, 0);
        printf("The fddev value is %d", fddev);
}

which returns -1. I also used this repo which again the fd returns -1.

I also looked at kallsyms to make sure the syscall for perf_event_open is there.

root@sama-desktop:/home/sama# cat /proc/kallsyms | grep "perf_event_open"
800f3178 T SyS_perf_event_open
800f3178 T sys_perf_event_open

and here is the output from /boot/config-3.18.0-20-rpi2:

# 
# Kernel Performance Events And Counters
#
CONFIG_PERF_EVENTS=y
# CONFIG_DEBUG_PERF_USE_VMALLOC is not set
CONFIG_VM_EVENT_COUNTERS=y
# CONFIG_COMPAT_BRK is not set
CONFIG_SLAB=y
# CONFIG_SLUB is not set
# CONFIG_SLOB is not set
# CONFIG_SYSTEM_TRUSTED_KEYRING is not set
CONFIG_PROFILING=y
CONFIG_TRACEPOINTS=y
CONFIG_OPROFILE=m
CONFIG_HAVE_OPROFILE=y
CONFIG_KPROBES=y
CONFIG_JUMP_LABEL=y
CONFIG_UPROBES=y
# CONFIG_HAVE_64BIT_ALIGNED_ACCESS is not set
CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS=y
CONFIG_ARCH_USE_BUILTIN_BSWAP=y
CONFIG_KRETPROBES=y
CONFIG_HAVE_KPROBES=y
CONFIG_HAVE_KRETPROBES=y
CONFIG_HAVE_ARCH_TRACEHOOK=y
CONFIG_HAVE_DMA_ATTRS=y
CONFIG_HAVE_DMA_CONTIGUOUS=y
CONFIG_GENERIC_SMP_IDLE_THREAD=y
CONFIG_GENERIC_IDLE_POLL_SETUP=y
CONFIG_HAVE_REGS_AND_STACK_ACCESS_API=y
CONFIG_HAVE_CLK=y
CONFIG_HAVE_DMA_API_DEBUG=y
CONFIG_HAVE_HW_BREAKPOINT=y
CONFIG_HAVE_PERF_REGS=y
CONFIG_HAVE_PERF_USER_STACK_DUMP=y
CONFIG_HAVE_ARCH_JUMP_LABEL=y
CONFIG_ARCH_WANT_IPC_PARSE_VERSION=y
CONFIG_HAVE_ARCH_SECCOMP_FILTER=y
CONFIG_SECCOMP_FILTER=y
CONFIG_HAVE_CC_STACKPROTECTOR=y
CONFIG_CC_STACKPROTECTOR=y
# CONFIG_CC_STACKPROTECTOR_NONE is not set
CONFIG_CC_STACKPROTECTOR_REGULAR=y
# CONFIG_CC_STACKPROTECTOR_STRONG is not set
CONFIG_HAVE_CONTEXT_TRACKING=y
CONFIG_HAVE_VIRT_CPU_ACCOUNTING_GEN=y
CONFIG_HAVE_IRQ_TIME_ACCOUNTING=y
CONFIG_HAVE_MOD_ARCH_SPECIFIC=y
CONFIG_MODULES_USE_ELF_REL=y
CONFIG_CLONE_BACKWARDS=y
CONFIG_OLD_SIGSUSPEND3=y
CONFIG_OLD_SIGACTION=y

and here is output from dmesg:

root@sama-desktop:/boot# dmesg | grep "perf"
[    0.003891] Initializing cgroup subsys perf_event

and here is the output for device tree:

root@sama-desktop:# ls -la /sys/bus/event_source/devices
total 0
drwxr-xr-x 2 root root 0 jul 18 20:15 .
drwxr-xr-x 4 root root 0 jan  1  1970 ..
lrwxrwxrwx 1 root root 0 jan  1  1970 breakpoint -> ../../../devices/breakpoint
lrwxrwxrwx 1 root root 0 jan  1  1970 software -> ../../../devices/software
lrwxrwxrwx 1 root root 0 jan  1  1970 tracepoint -> ../../../devices/tracepoint

I really do not know why perf_event_open returns -1.

Answer 1

From the lack of anything relevant in dmesg and sysfs, it should hopefully now be apparent that the PMU isn't being described to the kernel. Thus perf events doesn't know anything about the hardware event you're asking for, so it's little surprise that it fails to open it. What you need to do is make sure the kernel does know about the PMU, so that the driver picks it up - said driver should already be built-in via CONFIG_HW_PERF_EVENTS, which is on by default with CONFIG_PERF_EVENTS and doesn't look to be disabled in your config, but it might be worth double-checking.

It looks like the PMU is described in the devicetree in their 3.18 kernel, so my best guess is that your board might be booting using the legacy boardfile rather than FDT. I don't know much about Raspberry Pi specifics, but judging by this fairly exhaustive article (I'd say skip directly to section 3.1), it seems relatively straightforward to reconfigure the bootloader to use FDT.