Re: [PATCH 5/8] perf/hw_breakpoint: Remove useless code related to flexible breakpoints

[Marco Elver]

On Thu, 9 Jun 2022 at 15:41, Dmitry Vyukov <dvyukov@xxxxxxxxxx> wrote:
>
> On Thu, 9 Jun 2022 at 14:04, Dmitry Vyukov <dvyukov@xxxxxxxxxx> wrote:
> >
> > On Thu, 9 Jun 2022 at 13:31, Marco Elver <elver@xxxxxxxxxx> wrote:
> > >
> > > Flexible breakpoints have never been implemented, with
> > > bp_cpuinfo::flexible always being 0. Unfortunately, they still occupy 4
> > > bytes in each bp_cpuinfo and bp_busy_slots, as well as computing the max
> > > flexible count in fetch_bp_busy_slots().
> > >
> > > This again causes suboptimal code generation, when we always know that
> > > `!!slots.flexible` will be 0.
> > >
> > > Just get rid of the flexible "placeholder" and remove all real code
> > > related to it. Make a note in the comment related to the constraints
> > > algorithm but don't remove them from the algorithm, so that if in future
> > > flexible breakpoints need supporting, it should be trivial to revive
> > > them (along with reverting this change).
> > >
> > > Signed-off-by: Marco Elver <elver@xxxxxxxxxx>
> >
> > Was added in 2009.
> >
> > Acked-by: Dmitry Vyukov <dvyukov@xxxxxxxxxx>
> >
> > > ---
> > >  kernel/events/hw_breakpoint.c | 12 +++---------
> > >  1 file changed, 3 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/kernel/events/hw_breakpoint.c b/kernel/events/hw_breakpoint.c
> > > index 5f40c8dfa042..afe0a6007e96 100644
> > > --- a/kernel/events/hw_breakpoint.c
> > > +++ b/kernel/events/hw_breakpoint.c
> > > @@ -46,8 +46,6 @@ struct bp_cpuinfo {
> > >  #else
> > >         unsigned int    *tsk_pinned;
> > >  #endif
> > > -       /* Number of non-pinned cpu/task breakpoints in a cpu */
> > > -       unsigned int    flexible; /* XXX: placeholder, see fetch_this_slot() */
> > >  };
> > >
> > >  static DEFINE_PER_CPU(struct bp_cpuinfo, bp_cpuinfo[TYPE_MAX]);
> > > @@ -71,7 +69,6 @@ static bool constraints_initialized __ro_after_init;
> > >  /* Gather the number of total pinned and un-pinned bp in a cpuset */
> > >  struct bp_busy_slots {
>
> Do we also want to remove this struct altogether? Now it becomes just
> an int counter.

Yes, that actually can simplify a bunch of things, including
fetch_bp_busy_slots() just returning an int and fetch_this_slot() can
be removed (it'll be even cleaner if we remove the overridable
weight).

I'll simplify unless I hear objections.