From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gregory CLEMENT Subject: Re: [PATCH v2 2/4] Documentation: bindings: Add the regulator property to the sub-nodes AHCI bindings Date: Fri, 09 Jan 2015 17:20:44 +0100 Message-ID: <54AFFFDC.7020307@free-electrons.com> References: <1420799989-10645-1-git-send-email-gregory.clement@free-electrons.com> <1420799989-10645-3-git-send-email-gregory.clement@free-electrons.com> <54AFF7E3.4070506@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 8bit Return-path: Received: from down.free-electrons.com ([37.187.137.238]:41974 "EHLO mail.free-electrons.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751161AbbAIQVG (ORCPT ); Fri, 9 Jan 2015 11:21:06 -0500 In-Reply-To: <54AFF7E3.4070506@redhat.com> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Hans de Goede , linux-kernel@vger.kernel.org Cc: Tejun Heo , linux-ide@vger.kernel.org, =?windows-1252?Q?Antoine_T=E9nart?= , Thomas Petazzoni , Ezequiel Garcia , Maxime Ripard , Boris BREZILLON , Jason Cooper , Andrew Lunn , Sebastian Hesselbarth , linux-arm-kernel@lists.infradead.org, Lior Amsalem , Tawfik Bayouk , Nadav Haklai , Mark Rutland , devicetree@vger.kernel.org, Mark Brown Hi Hans, On 09/01/2015 16:46, Hans de Goede wrote: > Hi, > > On 09-01-15 11:39, Gregory CLEMENT wrote: >> It is now possible to use a regulator property for each port of the >> AHCI controller. >> >> Signed-off-by: Gregory CLEMENT >> --- >> Documentation/devicetree/bindings/ata/ahci-platform.txt | 9 ++++++--- >> 1 file changed, 6 insertions(+), 3 deletions(-) >> >> diff --git a/Documentation/devicetree/bindings/ata/ahci-platform.txt b/Documentation/devicetree/bindings/ata/ahci-platform.txt >> index 4ab09f2202d4..87416d71c758 100644 >> --- a/Documentation/devicetree/bindings/ata/ahci-platform.txt >> +++ b/Documentation/devicetree/bindings/ata/ahci-platform.txt >> @@ -37,9 +37,10 @@ Required properties when using sub-nodes: >> >> >> Sub-nodes required properties: >> -- reg : the port number >> -- phys : reference to the SATA PHY node >> - >> +- reg : the port number >> +And at least one of the following properties: >> +- phys : reference to the SATA PHY node >> +- target-port-supply : regulator for SATA target power for port >> >> Examples: >> sata@ffe08000 { >> @@ -68,10 +69,12 @@ With sub-nodes: >> sata0: sata-port@0 { >> reg = <0>; >> phys = <&sata_phy 0>; >> + target-port0-supply = <®_sata0>; >> }; >> > > I'm sorry, but I've to NAK this, I did not follow the regulator discussion., > thinking that it was just about the use of some utility function vs diy or > some such, I had no idea this would impact the dt-binding. > > The -port%d- bit is completely unnecessary from a dt pov. Devicetree is > supposed to be OS agnostic, we cannot make the dt-binding ugly just because > the regulator subsys in Linux does not provide the necessary functionality. > > I'm afraid you will have to get back to the regulator subsys people and tell > them that we really need an of_regulator_get, so that we can tell the > regulator subsys we want a regulator from a specific of_node. I didn't cc you especially, but I cc the linux-ide mailing list. The email was "[PATCH 2/2] regulator: core: Add the device tree version to the regulator_get family" (see http://thread.gmane.org/gmane.linux.kernel/1856154). Mark Brown didn't want a firmware specific interface. The regulator framework allows associating only a regulator to a device and currently, at least from the point of view of the device tree, the port subnodes are not devices. He suggested making the child node device. However, when I had a look on it, it didn't seem an easy task. As you know better this code, what do you think of it? (I added Mark Bronw in CC of this email so he can add more explanation if needed) Thanks, Gregory > > Please CC me when you re-open the discussion with the regulator people. > > Regards, > > Hans > -- Gregory Clement, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com From mboxrd@z Thu Jan 1 00:00:00 1970 From: gregory.clement@free-electrons.com (Gregory CLEMENT) Date: Fri, 09 Jan 2015 17:20:44 +0100 Subject: [PATCH v2 2/4] Documentation: bindings: Add the regulator property to the sub-nodes AHCI bindings In-Reply-To: <54AFF7E3.4070506@redhat.com> References: <1420799989-10645-1-git-send-email-gregory.clement@free-electrons.com> <1420799989-10645-3-git-send-email-gregory.clement@free-electrons.com> <54AFF7E3.4070506@redhat.com> Message-ID: <54AFFFDC.7020307@free-electrons.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Hans, On 09/01/2015 16:46, Hans de Goede wrote: > Hi, > > On 09-01-15 11:39, Gregory CLEMENT wrote: >> It is now possible to use a regulator property for each port of the >> AHCI controller. >> >> Signed-off-by: Gregory CLEMENT >> --- >> Documentation/devicetree/bindings/ata/ahci-platform.txt | 9 ++++++--- >> 1 file changed, 6 insertions(+), 3 deletions(-) >> >> diff --git a/Documentation/devicetree/bindings/ata/ahci-platform.txt b/Documentation/devicetree/bindings/ata/ahci-platform.txt >> index 4ab09f2202d4..87416d71c758 100644 >> --- a/Documentation/devicetree/bindings/ata/ahci-platform.txt >> +++ b/Documentation/devicetree/bindings/ata/ahci-platform.txt >> @@ -37,9 +37,10 @@ Required properties when using sub-nodes: >> >> >> Sub-nodes required properties: >> -- reg : the port number >> -- phys : reference to the SATA PHY node >> - >> +- reg : the port number >> +And at least one of the following properties: >> +- phys : reference to the SATA PHY node >> +- target-port-supply : regulator for SATA target power for port >> >> Examples: >> sata at ffe08000 { >> @@ -68,10 +69,12 @@ With sub-nodes: >> sata0: sata-port at 0 { >> reg = <0>; >> phys = <&sata_phy 0>; >> + target-port0-supply = <®_sata0>; >> }; >> > > I'm sorry, but I've to NAK this, I did not follow the regulator discussion., > thinking that it was just about the use of some utility function vs diy or > some such, I had no idea this would impact the dt-binding. > > The -port%d- bit is completely unnecessary from a dt pov. Devicetree is > supposed to be OS agnostic, we cannot make the dt-binding ugly just because > the regulator subsys in Linux does not provide the necessary functionality. > > I'm afraid you will have to get back to the regulator subsys people and tell > them that we really need an of_regulator_get, so that we can tell the > regulator subsys we want a regulator from a specific of_node. I didn't cc you especially, but I cc the linux-ide mailing list. The email was "[PATCH 2/2] regulator: core: Add the device tree version to the regulator_get family" (see http://thread.gmane.org/gmane.linux.kernel/1856154). Mark Brown didn't want a firmware specific interface. The regulator framework allows associating only a regulator to a device and currently, at least from the point of view of the device tree, the port subnodes are not devices. He suggested making the child node device. However, when I had a look on it, it didn't seem an easy task. As you know better this code, what do you think of it? (I added Mark Bronw in CC of this email so he can add more explanation if needed) Thanks, Gregory > > Please CC me when you re-open the discussion with the regulator people. > > Regards, > > Hans > -- Gregory Clement, Free Electrons Kernel, drivers, real-time and embedded Linux development, consulting, training and support. http://free-electrons.com