seccomp_unotify allows more privileged processes do actions on behalf
of less privileged processes.

In many cases, the workflow is fully synchronous. It means a target
process triggers a system call and passes controls to a supervisor
process that handles the system call and returns controls back to the
target process. In this context, "synchronous" means that only one
process is running and another one is waiting.

The new WF_CURRENT_CPU flag advises the scheduler to move the wakee to
the current CPU. For such synchronous workflows, it makes context
switches a few times faster.

Right now, each interaction takes 12µs. With this patch, it takes about
3µs.

v2: clean up the first patch and add the test.
v3: update commit messages and a few fixes suggested by Kees Cook.
v4: update the third patch to avoid code duplications (suggested by
    Peter Zijlstra)
    Add the benchmark to the perf bench set.
v5: Update the author email. No code changes.

Kees is ready to take this patch set, but wants to get Acks from the
sched folks.

Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Christian Brauner <brauner@kernel.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Peter Oskolkov <posk@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tycho Andersen <tycho@tycho.pizza>
Cc: Will Drewry <wad@chromium.org>
Cc: Vincent Guittot <vincent.guittot@linaro.org>

Andrei Vagin (4):
  seccomp: don't use semaphore and wait_queue together
  sched: add a few helpers to wake up tasks on the current cpu
  seccomp: add the synchronous mode for seccomp_unotify
  selftest/seccomp: add a new test for the sync mode of
    seccomp_user_notify

Peter Oskolkov (1):
  sched: add WF_CURRENT_CPU and externise ttwu

 include/linux/completion.h                    |   1 +
 include/linux/swait.h                         |   2 +-
 include/linux/wait.h                          |   3 +
 include/uapi/linux/seccomp.h                  |   4 +
 kernel/sched/completion.c                     |  26 ++-
 kernel/sched/core.c                           |   5 +-
 kernel/sched/fair.c                           |   4 +
 kernel/sched/sched.h                          |  13 +-
 kernel/sched/swait.c                          |   8 +-
 kernel/sched/wait.c                           |   5 +
 kernel/seccomp.c                              |  72 +++++++-
 tools/arch/x86/include/uapi/asm/unistd_32.h   |   3 +
 tools/arch/x86/include/uapi/asm/unistd_64.h   |   3 +
 tools/perf/bench/Build                        |   1 +
 tools/perf/bench/bench.h                      |   1 +
 tools/perf/bench/sched-seccomp-notify.c       | 167 ++++++++++++++++++
 tools/perf/builtin-bench.c                    |   1 +
 tools/testing/selftests/seccomp/seccomp_bpf.c |  55 ++++++
 18 files changed, 346 insertions(+), 28 deletions(-)
 create mode 100644 tools/perf/bench/sched-seccomp-notify.c

-- 
2.37.2

On Tue, Mar 7, 2023 at 11:32 PM Andrei Vagin <avagin@google.com> wrote:
>
> seccomp_unotify allows more privileged processes do actions on behalf
> of less privileged processes.
>
> In many cases, the workflow is fully synchronous. It means a target
> process triggers a system call and passes controls to a supervisor
> process that handles the system call and returns controls back to the
> target process. In this context, "synchronous" means that only one
> process is running and another one is waiting.
>
> The new WF_CURRENT_CPU flag advises the scheduler to move the wakee to
> the current CPU. For such synchronous workflows, it makes context
> switches a few times faster.
>
> Right now, each interaction takes 12µs. With this patch, it takes about
> 3µs.
>
> v2: clean up the first patch and add the test.
> v3: update commit messages and a few fixes suggested by Kees Cook.
> v4: update the third patch to avoid code duplications (suggested by
>     Peter Zijlstra)
>     Add the benchmark to the perf bench set.
> v5: Update the author email. No code changes.
>
> Kees is ready to take this patch set, but wants to get Acks from the
> sched folks.

Peter, could you review the second and third patches of this series?

Thanks,
Andrei

On Tue, Mar 07, 2023 at 11:31:55PM -0800, Andrei Vagin wrote:

> Kees is ready to take this patch set, but wants to get Acks from the
> sched folks.
> 

> Andrei Vagin (4):
>   seccomp: don't use semaphore and wait_queue together
>   sched: add a few helpers to wake up tasks on the current cpu
>   seccomp: add the synchronous mode for seccomp_unotify
>   selftest/seccomp: add a new test for the sync mode of
>     seccomp_user_notify
> 
> Peter Oskolkov (1):
>   sched: add WF_CURRENT_CPU and externise ttwu

Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

On Mon, Mar 27, 2023 at 3:27 AM Peter Zijlstra <peterz@infradead.org> wrote:
>
> On Tue, Mar 07, 2023 at 11:31:55PM -0800, Andrei Vagin wrote:
>
> > Kees is ready to take this patch set, but wants to get Acks from the
> > sched folks.
> >
>
> > Andrei Vagin (4):
> >   seccomp: don't use semaphore and wait_queue together
> >   sched: add a few helpers to wake up tasks on the current cpu
> >   seccomp: add the synchronous mode for seccomp_unotify
> >   selftest/seccomp: add a new test for the sync mode of
> >     seccomp_user_notify
> >
> > Peter Oskolkov (1):
> >   sched: add WF_CURRENT_CPU and externise ttwu
>
> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>

Kees,

Could you look at this patch set? You wrote to one of the previous
versions that you are ready to take it if sched maintainers approve it.
Here is no major changes from that moment. The sched-related part has
been cleaned up according with Peter's comments, and I moved the perf
test to the perf tool.

Thanks,
Andrei

On April 3, 2023 11:32:00 AM PDT, Andrei Vagin <avagin@gmail.com> wrote:
>On Mon, Mar 27, 2023 at 3:27 AM Peter Zijlstra <peterz@infradead.org> wrote:
>>
>> On Tue, Mar 07, 2023 at 11:31:55PM -0800, Andrei Vagin wrote:
>>
>> > Kees is ready to take this patch set, but wants to get Acks from the
>> > sched folks.
>> >
>>
>> > Andrei Vagin (4):
>> >   seccomp: don't use semaphore and wait_queue together
>> >   sched: add a few helpers to wake up tasks on the current cpu
>> >   seccomp: add the synchronous mode for seccomp_unotify
>> >   selftest/seccomp: add a new test for the sync mode of
>> >     seccomp_user_notify
>> >
>> > Peter Oskolkov (1):
>> >   sched: add WF_CURRENT_CPU and externise ttwu
>>
>> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
>
>Kees,
>
>Could you look at this patch set? You wrote to one of the previous
>versions that you are ready to take it if sched maintainers approve it.
>Here is no major changes from that moment. The sched-related part has
>been cleaned up according with Peter's comments, and I moved the perf
>test to the perf tool.

Hi!

Yes, thanks for keeping this going! I'm catching up after some vacation, but this is on my TODO list. :)

-Kees


-- 
Kees Cook

On Wed, Apr 5, 2023 at 8:19 PM Kees Cook <kees@kernel.org> wrote:
>
> On April 3, 2023 11:32:00 AM PDT, Andrei Vagin <avagin@gmail.com> wrote:
> >On Mon, Mar 27, 2023 at 3:27 AM Peter Zijlstra <peterz@infradead.org> wrote:
> >>
> >> On Tue, Mar 07, 2023 at 11:31:55PM -0800, Andrei Vagin wrote:
> >>
> >> > Kees is ready to take this patch set, but wants to get Acks from the
> >> > sched folks.
> >> >
> >>
> >> > Andrei Vagin (4):
> >> >   seccomp: don't use semaphore and wait_queue together
> >> >   sched: add a few helpers to wake up tasks on the current cpu
> >> >   seccomp: add the synchronous mode for seccomp_unotify
> >> >   selftest/seccomp: add a new test for the sync mode of
> >> >     seccomp_user_notify
> >> >
> >> > Peter Oskolkov (1):
> >> >   sched: add WF_CURRENT_CPU and externise ttwu
> >>
> >> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> >
> >Kees,
> >
> >Could you look at this patch set? You wrote to one of the previous
> >versions that you are ready to take it if sched maintainers approve it.
> >Here is no major changes from that moment. The sched-related part has
> >been cleaned up according with Peter's comments, and I moved the perf
> >test to the perf tool.
>
> Hi!
>
> Yes, thanks for keeping this going! I'm catching up after some vacation, but this is on my TODO list. :)

Hi Kees. Do you have any updates on this series?

>
> -Kees
>
>
> --
> Kees Cook

On Wed, Jun 28, 2023 at 11:44:02AM -0700, Andrei Vagin wrote:
> On Wed, Apr 5, 2023 at 8:19 PM Kees Cook <kees@kernel.org> wrote:
> >
> > On April 3, 2023 11:32:00 AM PDT, Andrei Vagin <avagin@gmail.com> wrote:
> > >On Mon, Mar 27, 2023 at 3:27 AM Peter Zijlstra <peterz@infradead.org> wrote:
> > >>
> > >> On Tue, Mar 07, 2023 at 11:31:55PM -0800, Andrei Vagin wrote:
> > >>
> > >> > Kees is ready to take this patch set, but wants to get Acks from the
> > >> > sched folks.
> > >> >
> > >>
> > >> > Andrei Vagin (4):
> > >> >   seccomp: don't use semaphore and wait_queue together
> > >> >   sched: add a few helpers to wake up tasks on the current cpu
> > >> >   seccomp: add the synchronous mode for seccomp_unotify
> > >> >   selftest/seccomp: add a new test for the sync mode of
> > >> >     seccomp_user_notify
> > >> >
> > >> > Peter Oskolkov (1):
> > >> >   sched: add WF_CURRENT_CPU and externise ttwu
> > >>
> > >> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> > >
> > >Kees,
> > >
> > >Could you look at this patch set? You wrote to one of the previous
> > >versions that you are ready to take it if sched maintainers approve it.
> > >Here is no major changes from that moment. The sched-related part has
> > >been cleaned up according with Peter's comments, and I moved the perf
> > >test to the perf tool.
> >
> > Hi!
> >
> > Yes, thanks for keeping this going! I'm catching up after some vacation, but this is on my TODO list. :)
> 
> Hi Kees. Do you have any updates on this series?

Apologies for the delay!

I've added this to the seccomp tree -- it should show up in -next soon.

-Kees

-- 
Kees Cook