Re: syscalls performance

[Mauro Romano Trajber <trajber@xxxxxxxxx>]

Sure, the code is attached.

On Fri, Feb 25, 2011 at 5:15 PM, Daniel Baluta <daniel.baluta@xxxxxxxxx> wrote:

On Fri, Feb 25, 2011 at 8:22 PM, Mauro Romano Trajber <trajber@xxxxxxxxx> wrote:
> Thanks Enrico and Daniel, you're right. glibc was caching getpid(); but this
> is not the root cause of this behavior.
> Going further, I decide to use call getpid without glibc, using
>  syscall(SYS_getpid) to test this behavior and it happened again.
> Calling it once, the test consumes about 7k CPU cycles and 10 calls consumes
> about 10k CPU cycles.
> Any ideas ?

Can you post a pointer to your code and information about how you got
this numbers?

thanks,
Daniel.

#define _GNU_SOURCE

#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>
#include <unistd.h>
#include <sys/syscall.h>
#include <sched.h>

long mygetpid() {
	return syscall(SYS_getpid);
}

inline volatile long long RDTSC() {
	register long long TSC asm("eax");
	asm volatile (".byte 15, 49" : : : "eax", "edx");
	return TSC;
}

int main(int argc, char *argv[]) {
	cpu_set_t cpu;
	CPU_ZERO(&cpu);
	CPU_SET(0, &cpu);

	if (sched_setaffinity(0, sizeof(cpu), &cpu) < 0) {
		perror("sched_setaffinity");
		exit(1);
	}

	int i, qtd = 0;
	if (argc == 2) {
		qtd = atoi(argv[1]);
	}

	long long start = RDTSC();

	for (i = 0; i < qtd; i++) {
		mygetpid();
	}

	long long end = RDTSC();
	printf("%lld\n",  (end - start));

	return 0;
}

_______________________________________________
Kernelnewbies mailing list
Kernelnewbies@xxxxxxxxxxxxxxxxx
http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies