Re: [PATCH RFC bpf-next v6 3/6] bpf/helpers: introduce bpf_timer_set_sleepable_cb() kfunc

[Benjamin Tissoires]

On Mon, Apr 8, 2024 at 4:31 PM Eduard Zingerman <eddyz87@xxxxxxxxx> wrote:
>
> On Mon, 2024-04-08 at 10:09 +0200, Benjamin Tissoires wrote:
> [...]
>
> > diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
> > index fd05d4358b31..d6528359b3f4 100644
> > --- a/kernel/bpf/helpers.c
> > +++ b/kernel/bpf/helpers.c
>
> [...]
>
> > @@ -2726,6 +2764,7 @@ BTF_ID_FLAGS(func, bpf_dynptr_is_null)
> >  BTF_ID_FLAGS(func, bpf_dynptr_is_rdonly)
> >  BTF_ID_FLAGS(func, bpf_dynptr_size)
> >  BTF_ID_FLAGS(func, bpf_dynptr_clone)
> > +BTF_ID_FLAGS(func, bpf_timer_set_sleepable_cb_impl)
>
> Note:
> this hunk does not apply cleanly on top of current master.
> The line 'BTF_ID_FLAGS(func, bpf_modify_return_test_tp)'
> was added to the list since last time current patch-set was merged.


Oops, thanks for the update.

Just to be clear, I already mentioned it in the cover letter, but this
series is not intended to be merged just now (thus RFC again). The
plan is to add a new bpf_wq API on the side, and compare it with this
v6 to see which one is best, because I am trying to force the
workqueue API into a timer, when it's getting further and further away
from each other.

Cheers,
Benjamin


>
>
> >  BTF_KFUNCS_END(common_btf_ids)
> >
> >  static const struct btf_kfunc_id_set common_kfunc_set = {
>