From mboxrd@z Thu Jan  1 00:00:00 1970
From: Arnd Bergmann <arnd@arndb.de>
Subject: Re: [PATCH v3 0/9] phy: use syscon framework APIs to set ctrl mod reg
Date: Tue, 15 Dec 2015 12:55:37 +0100
Message-ID: <11176831.mhgJlIuj0S@wuerfel>
References: <1450170968-11282-1-git-send-email-kishon@ti.com> <725110094.6UkMkcy8u7@wuerfel> <566FF621.2040307@ti.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7Bit
Return-path: <linux-pci-owner@vger.kernel.org>
In-Reply-To: <566FF621.2040307@ti.com>
Sender: linux-pci-owner@vger.kernel.org
To: Kishon Vijay Abraham I <kishon@ti.com>
Cc: devicetree@vger.kernel.org, linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, robh+dt@kernel.org, pawel.moll@arm.com, mark.rutland@arm.com, ijc+devicetree@hellion.org.uk, galak@codeaurora.org, rogerq@ti.com, nsekhar@ti.com, linux-pci@vger.kernel.org, linux-usb@vger.kernel.org, tony@atomide.com, t-kristo@ti.com
List-Id: devicetree@vger.kernel.org

On Tuesday 15 December 2015 16:44:41 Kishon Vijay Abraham I wrote:
> Hi Arnd,
> 
> On Tuesday 15 December 2015 04:26 PM, Arnd Bergmann wrote:
> > On Tuesday 15 December 2015 14:45:59 Kishon Vijay Abraham I wrote:
> >> This series is basically to deprecate using phy-omap-control and use
> >> syscon APIs to program the control module registers.
> >>
> >> Changes from v2:
> >> No changes.
> >>
> >> Changes from v1:
> >> *) cleanup ti_pipe3_probe in multiple steps
> >> *) other minor cleanups
> >>
> >> Changes from [1] in PHY patches include
> >> *) cleanup ti_pipe3_probe
> >> *) have mask, power_on and power_off values in usb_phy_data for
> >>    omap-usb2 phy
> >>
> >> The patches have been pushed to
> >>         git://git.ti.com/linux-phy/linux-phy.git syscon
> >>
> >> [1] -> https://lkml.org/lkml/2015/6/23/189
> >>
> >> All the testing was done both before applying the dt patches and after
> >> applying the dt patches (dt patches will be posted shortly).
> >>
> > 
> > Can you explain here what the conversion is good for? Why do you
> > prefer the syscon mapping over a high-level driver in this case?
> 
> phy-omap-control driver was added when there was no proper
> infrastructure for doing control module initializations. The
> phy-omap-control driver is not an 'actual' PHY driver and it
> was just a hack to do PHY related control module initializations.
> phy-omap-control is also getting unmanageable with the number of
> platforms each having number of modules (like USB, SATA, PCIe),
> using the same driver for control module initializations.
> 
> Now with SYSCON framework being added to the kernel, phy-omap-control
> shouldn't be needed and it also provides a uniform API across all the
> modules to program the control module.

Ok, so the "phy-control" devices were really just a few registers of
a system controller device that does a lot of other things as well, right?

Can you put your description above into the cover-letter for the series,
and the merge commit?

	Arnd