Re: [RFC] store function address in BTF

[Jiri Olsa <jolsa@redhat.com>]

On Wed, Oct 06, 2021 at 07:53:31AM -0700, Alexei Starovoitov wrote:
> On Wed, Oct 6, 2021 at 1:44 AM Jiri Olsa <jolsa@redhat.com> wrote:
> >
> > On Wed, Oct 06, 2021 at 10:41:41AM +0200, Jiri Olsa wrote:
> > > hi,
> > > I'm hitting performance issue and soft lock ups with the new version
> > > of the patchset and the reason seems to be kallsyms lookup that we
> > > need to do for each btf id we want to attach
> >
> > ugh, I meant to sent this as reply to the patchset mentioned above,
> > nevermind, here's the patchset:
> >   https://lore.kernel.org/bpf/20210605111034.1810858-1-jolsa@kernel.org/
> >
> > jirka
> >
> > >
> > > I tried to change kallsyms_lookup_name linear search into rbtree search,
> > > but it has its own pitfalls like duplicate function names and it still
> > > seems not to be fast enough when you want to attach like 30k functions
> > >
> > > so I wonder we could 'fix this' by storing function address in BTF,
> > > which would cut kallsyms lookup completely, because it'd be done in
> > > compile time
> > >
> > > my first thought was to add extra BTF section for that, after discussion
> > > with Arnaldo perhaps we could be able to store extra 8 bytes after
> > > BTF_KIND_FUNC record, using one of the 'unused' bits in btf_type to
> > > indicate that? or new BTF_KIND_FUNC2 type?
> > >
> > > thoughts?
> 
> That would be on top of your next patch set?
> Please post it first.

ok, will do

jirka