Re: [PATCH v3 2/3] perf: Reveal PMU type in fdinfo

[Chun-Tse Shao]

Hi Peter, just want to bring your attention to this patch since you
left comments on the previous version of patch.

Thanks,
CT

On Thu, Nov 21, 2024 at 1:18 PM Chun-Tse Shao <ctshao@xxxxxxxxxx> wrote:
>
> Hi, just a friendly ping for review, in case anyone missed this.
>
> Thanks,
> CT
>
> On Thu, Nov 14, 2024 at 10:30 AM Chun-Tse Shao <ctshao@xxxxxxxxxx> wrote:
> >
> > Hi Ian,
> >
> > On Thu, Nov 14, 2024 at 7:49 AM Ian Rogers <irogers@xxxxxxxxxx> wrote:
> > >
> > > On Tue, Nov 5, 2024 at 4:30 PM Chun-Tse Shao <ctshao@xxxxxxxxxx> wrote:
> > > >
> > > > It gives useful info on knowing which PMUs are reserved by this proceszs.
> > > > Also add config which would be useful.
> > > > Testing cycles:
> > > >
> > > >   $ ./perf stat -e cycles &
> > > >   $ cat /proc/`pidof perf`/fdinfo/3
> > > >   pos:    0
> > > >   flags:  02000002
> > > >   mnt_id: 16
> > > >   ino:    3081
> > > >   perf_event_attr.type:   0
> > > >   perf_event_attr.config: 0
> > > >
> > > > Testing L1-dcache-load-misses:
> > > >
> > > >   $ ./perf stat -e L1-dcache-load-misses &
> > > >   $ cat /proc/`pidof perf`/fdinfo/3
> > > >   pos:    0
> > > >   flags:  02000002
> > > >   mnt_id: 16
> > > >   ino:    1072
> > > >   perf_event_attr.type:   3
> > > >   perf_event_attr.config: 65536
> > > >
> > > > Signed-off-by: Chun-Tse Shao <ctshao@xxxxxxxxxx>
> > >
> > > Reviewed-by: Ian Rogers <irogers@xxxxxxxxxx>
> > >
> > > > Change-Id: Ibea5618aaf00bae6f48a9b2a6e7798ab2b7f23ce
> > > > ---
> > > >  kernel/events/core.c | 10 ++++++++++
> > > >  1 file changed, 10 insertions(+)
> > > >
> > > > diff --git a/kernel/events/core.c b/kernel/events/core.c
> > > > index cdd09769e6c56..398cac8b208b9 100644
> > > > --- a/kernel/events/core.c
> > > > +++ b/kernel/events/core.c
> > > > @@ -55,6 +55,7 @@
> > > >  #include <linux/pgtable.h>
> > > >  #include <linux/buildid.h>
> > > >  #include <linux/task_work.h>
> > > > +#include <linux/seq_file.h>
> > > >
> > > >  #include "internal.h"
> > > >
> > > > @@ -6820,6 +6821,14 @@ static int perf_fasync(int fd, struct file *filp, int on)
> > > >         return 0;
> > > >  }
> > > >
> > > > +static void perf_show_fdinfo(struct seq_file *m, struct file *f)
> > > > +{
> > > > +       struct perf_event *event = f->private_data;
> > > > +
> > > > +       seq_printf(m, "perf_event_attr.type:\t%u\n", event->orig_type);
> > > > +       seq_printf(m, "perf_event_attr.config:\t%llu\n", (unsigned long long)event->attr.config);
> > >
> > > nit: is the cast necessary? I don't see __u64 listed on:
> > > https://www.kernel.org/doc/Documentation/printk-formats.txt
> > > so I'm unsure.
> > >
> >
> > In this case I think it is safer to cast it to `unsigned long long`,
> > since I don't know if any architecture would have an exception on
> > __u64.
> >
> > Thanks,
> > CT
> >
> > > Thanks,
> > > Ian
> > >
> > > > +}
> > > > +
> > > >  static const struct file_operations perf_fops = {
> > > >         .release                = perf_release,
> > > >         .read                   = perf_read,
> > > > @@ -6828,6 +6837,7 @@ static const struct file_operations perf_fops = {
> > > >         .compat_ioctl           = perf_compat_ioctl,
> > > >         .mmap                   = perf_mmap,
> > > >         .fasync                 = perf_fasync,
> > > > +       .show_fdinfo            = perf_show_fdinfo,
> > > >  };
> > > >
> > > >  /*
> > > > --
> > > > 2.47.0.199.ga7371fff76-goog
> > > >