On Fri, Apr 5, 2024 at 8:46 AM Benjamin Tissoires
wrote:
>
> Alright, as of today (and I'm about to be AFK for the weekend), I got
> your changes in and working (I think). I'll review the series on
> Monday and send it back so we have a baseline to compare it to with
> bpf_wq.
Nice! Looking forwa
On Thu, Apr 4, 2024 at 8:29 PM Alexei Starovoitov
wrote:
>
> On Thu, Apr 4, 2024 at 10:56 AM Benjamin Tissoires
> wrote:
> >
> > On Thu, Apr 4, 2024 at 6:41 PM Alexei Starovoitov
> > wrote:
> > >
> > > On Thu, Apr 4, 2024 at 8:27 AM Benjamin Tissoires
> > > wrote:
> > > >
> > > >
> > > > >
> >
On Thu, Apr 4, 2024 at 10:56 AM Benjamin Tissoires
wrote:
>
> On Thu, Apr 4, 2024 at 6:41 PM Alexei Starovoitov
> wrote:
> >
> > On Thu, Apr 4, 2024 at 8:27 AM Benjamin Tissoires
> > wrote:
> > >
> > >
> > > >
> > > > So we need something like:
> > > >
> > > > struct bpf_hrtimer {
> > > > unio
On Thu, Apr 4, 2024 at 6:41 PM Alexei Starovoitov
wrote:
>
> On Thu, Apr 4, 2024 at 8:27 AM Benjamin Tissoires
> wrote:
> >
> >
> > >
> > > So we need something like:
> > >
> > > struct bpf_hrtimer {
> > > union {
> > > struct hrtimer timer;
> > > + struct work_struct work;
> > > };
> >
On Thu, Apr 4, 2024 at 8:27 AM Benjamin Tissoires
wrote:
>
>
> >
> > So we need something like:
> >
> > struct bpf_hrtimer {
> > union {
> > struct hrtimer timer;
> > + struct work_struct work;
> > };
> > struct bpf_map *map;
> > struct bpf_prog *prog;
> > void __rcu *callback_fn;
On Thu, Apr 4, 2024 at 4:44 AM Alexei Starovoitov
wrote:
>
> On Wed, Apr 3, 2024 at 6:01 PM Alexei Starovoitov
> wrote:
> >
> > On Wed, Apr 3, 2024 at 11:50 AM Alexei Starovoitov
> > wrote:
> > >
> > > On Wed, Mar 27, 2024 at 10:02 AM Benjamin Tissoires
> > > wrote:
> > > > > >
On Wed, Apr 3, 2024 at 6:01 PM Alexei Starovoitov
wrote:
>
> On Wed, Apr 3, 2024 at 11:50 AM Alexei Starovoitov
> wrote:
> >
> > On Wed, Mar 27, 2024 at 10:02 AM Benjamin Tissoires
> > wrote:
> > > > > goto out;
> > > > > }
> > > > > + spin_lock(&t->sleepable_lock);
On Wed, Apr 3, 2024 at 11:50 AM Alexei Starovoitov
wrote:
>
> On Wed, Mar 27, 2024 at 10:02 AM Benjamin Tissoires
> wrote:
> > > > goto out;
> > > > }
> > > > + spin_lock(&t->sleepable_lock);
> > > > drop_prog_refcnt(t);
> > > > + spin_unlock(&t->sleepa
On Wed, Mar 27, 2024 at 10:02 AM Benjamin Tissoires
wrote:
> > > goto out;
> > > }
> > > + spin_lock(&t->sleepable_lock);
> > > drop_prog_refcnt(t);
> > > + spin_unlock(&t->sleepable_lock);
> >
> > this also looks odd.
>
> I basically need to protect "t-
On Mon, Mar 25, 2024 at 1:50 AM Alexei Starovoitov
wrote:
>
> On Fri, Mar 22, 2024 at 7:56 AM Benjamin Tissoires wrote:
> >
> > They are implemented as a workqueue, which means that there are no
> > guarantees of timing nor ordering.
> >
> > Signed-off-by: Benjamin Tissoires
> >
> > ---
> >
> >
On Fri, Mar 22, 2024 at 7:56 AM Benjamin Tissoires wrote:
>
> They are implemented as a workqueue, which means that there are no
> guarantees of timing nor ordering.
>
> Signed-off-by: Benjamin Tissoires
>
> ---
>
> no changes in v5
>
> changes in v4:
> - dropped __bpf_timer_compute_key()
> - use
They are implemented as a workqueue, which means that there are no
guarantees of timing nor ordering.
Signed-off-by: Benjamin Tissoires
---
no changes in v5
changes in v4:
- dropped __bpf_timer_compute_key()
- use a spin_lock instead of a semaphore
- ensure bpf_timer_cancel_and_free is not com
12 matches
Mail list logo