Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Sun, 8 Jul 2018 10:31:30 +0300
From:   Mika Westerberg <mika.westerberg@linux.intel.com>
To:     Lukas Wunner <lukas@wunner.de>
Cc:     linux-kernel@vger.kernel.org,
        Andreas Noever <andreas.noever@gmail.com>,
        Michael Jamet <michael.jamet@intel.com>,
        Yehezkel Bernat <YehezkelShB@gmail.com>,
        "Rafael J. Wysocki" <rjw@rjwysocki.net>,
        Christian Kellner <christian@kellner.me>,
        Mario Limonciello <mario.limonciello@dell.com>
Subject: Re: [PATCH 5/5] thunderbolt: Add support for runtime PM
Message-ID: <20180708073130.GR2534@lahna.fi.intel.com>
References: <20180618110731.57427-1-mika.westerberg@linux.intel.com>
 <20180618110731.57427-6-mika.westerberg@linux.intel.com>
 <20180707211401.GA13111@wunner.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20180707211401.GA13111@wunner.de>
Organization: Intel Finland Oy - BIC 0357606-4 - Westendinkatu 7, 02160 Espoo
User-Agent: Mutt/1.10.0 (2018-05-17)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Sat, Jul 07, 2018 at 11:14:01PM +0200, Lukas Wunner wrote:
> On Mon, Jun 18, 2018 at 02:07:31PM +0300, Mika Westerberg wrote:
> > Implement this using standard Linux runtime PM APIs so that when all the
> > children devices are runtime suspended, the Thunderbolt host controller
> > PCI device is runtime suspended as well. The ICM firmware then starts
> > powering down power domains towards RTD3 but it can prevent this if it
> > detects that there is an active Display Port stream (this is not visible
> > to the software, though).
> > 
> > The Thunderbolt host controller will be runtime resumed either when
> > there is a remote wake event (device is connected or disconnected), or
> > when there is access from userspace that requires hardware access.
> 
> IIUC, if there is no xdomain, after 15 s all switch devices as well as
> the domain device and the NHI will have runtime suspended.  The control
> channel is torn down as well, so you can no longer receive notifications
> over it.  Then how is wakeup of the NHI signalled on hotplug/unplug?
> Do you get a PME for the NHI device?

Yes, it sends PME.

> Because I'm fairly certain that
> I do not get a PME for the Light Ridge in my MacBook Pro, but I'll test
> this once more and modify negotiate_os_control() to grant PME control
> to the OS.

I think in case of Apple hardware, they handle the in some different
means than PME (possibly part of chipset driver or ACPI method/event).