On Thu, Jul 04, 2019 at 03:05:09PM +0200, Peter Zijlstra wrote:
> On Wed, Jul 03, 2019 at 08:14:30PM -0700, Kris Van Hees wrote:
> > +int dt_bpf_attach(int event_id, int bpf_fd)
> > +{
> > + int event_fd;
> > + int rc;
> > + struct perf_event_attr attr = {};
> > +
> > + attr.type = PERF_TYPE_TRACEPOINT;
> > + attr.sample_type = PERF_SAMPLE_RAW;
> > + attr.sample_period = 1;
> > + attr.wakeup_events = 1;
> > + attr.config = event_id;
> > +
> > + /* Register the event (based on its id), and obtain a fd. */
> > + event_fd = perf_event_open(&attr, -1, 0, -1, 0);
> > + if (event_fd < 0) {
> > + perror("sys_perf_event_open");
> > + return -1;
> > + }
> > +
> > + /* Enable the probe. */
> > + rc = ioctl(event_fd, PERF_EVENT_IOC_ENABLE, 0);
>
> AFAICT you didn't use attr.disabled = 1, so this IOC_ENABLE is
> completely superfluous.
Oh yes, good point (and the same applies to the dt_buffer.c code where I set
up the events that own each buffer - no point in doing an explicit enable there
eiteher).
Thanks for catching this!
> > + if (rc < 0) {
> > + perror("PERF_EVENT_IOC_ENABLE");
> > + return -1;
> > + }
> > +
> > + /* Associate the BPF program with the event. */
> > + rc = ioctl(event_fd, PERF_EVENT_IOC_SET_BPF, bpf_fd);
> > + if (rc < 0) {
> > + perror("PERF_EVENT_IOC_SET_BPF");
> > + return -1;
> > + }
> > +
> > + return 0;
> > +}
