Re: [PATCHSET v4 0/4] perf stat: Enable BPF counters with --for-each-cgroup

[Namhyung Kim <namhyung@kernel.org>]

On Sun, Jun 27, 2021 at 8:29 AM Namhyung Kim <namhyung@kernel.org> wrote:
>
> On Fri, Jun 25, 2021 at 12:18 AM Namhyung Kim <namhyung@kernel.org> wrote:
> >
> > Hello,
> >
> > This is to add BPF support for --for-each-cgroup to handle many cgroup
> > events on big machines.  You can use the --bpf-counters to enable the
> > new behavior.
> >
> >  * changes in v4
> >   - convert cgrp_readings to a per-cpu array map
> >   - remove now-unused cpu_idx map
> >   - move common functions to a header file
> >   - reuse bpftool bootstrap binary
> >   - fix build error in the cgroup code
> >
> >  * changes in v3
> >   - support cgroup hierarchy with ancestor ids
> >   - add and trigger raw_tp BPF program
> >   - add a build rule for vmlinux.h
> >
> >  * changes in v2
> >   - remove incorrect use of BPF_F_PRESERVE_ELEMS
> >   - add missing map elements after lookup
> >   - handle cgroup v1
> >
> > Basic idea is to use a single set of per-cpu events to count
> > interested events and aggregate them to each cgroup.  I used bperf
> > mechanism to use a BPF program for cgroup-switches and save the
> > results in a matching map element for given cgroups.
> >
> > Without this, we need to have separate events for cgroups, and it
> > creates unnecessary multiplexing overhead (and PMU programming) when
> > tasks in different cgroups are switched.  I saw this makes a big
> > difference on 256 cpu machines with hundreds of cgroups.
> >
> > Actually this is what I wanted to do it in the kernel [1], but we can
> > do the job using BPF!
>
> Ugh, I found the current kernel bpf verifier doesn't accept the
> bpf_get_current_ancestor_cgroup_id() helper.  Will send the fix
> to BPF folks.

The fix landed on the bpf-next tree.

Thanks,
Namhyung