Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
References: <20200713204300.345975-1-badhri@google.com> <20200714060520.GB657428@kroah.com>
 <040b4cde-91d4-7776-60b5-8f55b6e75831@roeck-us.net>
In-Reply-To: <040b4cde-91d4-7776-60b5-8f55b6e75831@roeck-us.net>
From:   Badhri Jagan Sridharan <badhri@google.com>
Date:   Tue, 14 Jul 2020 10:16:25 -0700
Message-ID: <CAPTae5JHSrp5H47Dr9mrkS_PF_W8MH1TFMzNCc4Geva0-b-B0A@mail.gmail.com>
Subject: Re: [PATCH] usb: typec: tcpm: Move to high priority workqueue for
 processing events
To:     Guenter Roeck <linux@roeck-us.net>
Cc:     reg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Heikki Krogerus <heikki.krogerus@linux.intel.com>,
        USB <linux-usb@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Jul 13, 2020 at 11:58 PM Guenter Roeck <linux@roeck-us.net> wrote:
>
> On 7/13/20 11:05 PM, reg Kroah-Hartman wrote:
> > On Mon, Jul 13, 2020 at 01:43:00PM -0700, Badhri Jagan Sridharan wrote:
> >> "tReceiverResponse 15 ms Section 6.6.2
> >> The receiver of a Message requiring a response Shall respond
> >> within tReceiverResponse in order to ensure that the
> >> sender=E2=80=99s SenderResponseTimer does not expire."
> >>
> >> When the cpu complex is busy running other lower priority
> >> work items, TCPM's work queue sometimes does not get scheduled
> >> on time to meet the above requirement from the spec.
> >> Elevating the TCPM's work queue to higher priority allows
> >> TCPM to meet tReceiverResponse in a busy system.
> >>
> >> Signed-off-by: Badhri Jagan Sridharan <badhri@google.com>
> >> ---
> >>  drivers/usb/typec/tcpm/tcpm.c | 2 +-
> >>  1 file changed, 1 insertion(+), 1 deletion(-)
> >>
> >> diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tc=
pm.c
> >> index 82b19ebd7838e0..088b6f1fa1ff89 100644
> >> --- a/drivers/usb/typec/tcpm/tcpm.c
> >> +++ b/drivers/usb/typec/tcpm/tcpm.c
> >> @@ -4747,7 +4747,7 @@ struct tcpm_port *tcpm_register_port(struct devi=
ce *dev, struct tcpc_dev *tcpc)
> >>      mutex_init(&port->lock);
> >>      mutex_init(&port->swap_lock);
> >>
> >> -    port->wq =3D create_singlethread_workqueue(dev_name(dev));
> >> +    port->wq =3D alloc_ordered_workqueue("%s", WQ_HIGHPRI, dev_name(d=
ev));
> >
> > How are you "guaranteeing" that this is really going to change anything
> > on a highly loaded machine?
> >
> > Yes, it might make things better, but if you have a hard deadline like
> > this, you need to do things a bit differently to always ensure that you
> > meet it.  I do not think this change is that fix, do you?
> >
Yes Greg I agree with you, moving to HIGHPRI was making it better but
is not going to
solve the problem always. I was wondering whether are there better
ways of doing this.

>
> Good point. The worker in drivers/watchdog/ !watchdog_dev.c might be
> useful as a starting point. There may be better examples - this is
> just one I know of which had a similar problem. See commits
> 38a1222ae4f3 and 1ff688209e2e.
>
> Guenter

Thanks a lot Guenter !! Very useful pointers, will review the
approaches in both the
commits !