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: <fdd9ce1d-146a-5fbf-75c5-3a9384603312@gmx.de> <5478a950-4355-8084-ea7d-fe8b270bf2e3@infradead.org>
 <5392275.BHAU0OPJTB@kreacher> <4b21c095-fbe5-1138-b977-a505baa41a2b@gmx.de>
In-Reply-To: <4b21c095-fbe5-1138-b977-a505baa41a2b@gmx.de>
From:   "Rafael J. Wysocki" <rafael@kernel.org>
Date:   Fri, 17 Apr 2020 20:52:10 +0200
Message-ID: <CAJZ5v0icdVL6_yGpfsorqszdi9GcLxzYdvDqTJyG4ENzkOG2pQ@mail.gmail.com>
Subject: Re: regression 5.6.4->5.6.5 at drivers/acpi/ec.c
To:     =?UTF-8?Q?Toralf_F=C3=B6rster?= <toralf.foerster@gmx.de>
Cc:     "Rafael J. Wysocki" <rjw@rjwysocki.net>,
        Randy Dunlap <rdunlap@infradead.org>,
        Linux Kernel <linux-kernel@vger.kernel.org>,
        ACPI Devel Mailing List <linux-acpi@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 8BIT
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Fri, Apr 17, 2020 at 6:36 PM Toralf Förster <toralf.foerster@gmx.de> wrote:
>
> On 4/17/20 5:53 PM, Rafael J. Wysocki wrote:
> > Does the patch below (untested) make any difference?
> >
> > ---
> >  drivers/acpi/ec.c |    5 ++++-
> >  1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > Index: linux-pm/drivers/acpi/ec.c
> > ===================================================================
> > --- linux-pm.orig/drivers/acpi/ec.c
> > +++ linux-pm/drivers/acpi/ec.c
> > @@ -2067,7 +2067,10 @@ static struct acpi_driver acpi_ec_driver
> >               .add = acpi_ec_add,
> >               .remove = acpi_ec_remove,
> >               },
> > -     .drv.pm = &acpi_ec_pm,
> > +     .drv = {
> > +             .probe_type = PROBE_FORCE_SYNCHRONOUS,
> > +             .pm = &acpi_ec_pm,
> > +     },
> >  };
> >
> >  static void acpi_ec_destroy_workqueues(void)
> I'd say no, but for completeness:

OK, it looks like mainline commit

65a691f5f8f0 ("ACPI: EC: Do not clear boot_ec_is_ecdt in acpi_ec_add()")

was backported into 5.6.5 by mistake.

Can you please revert that patch and retest?