From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Tue, 16 Feb 2016 10:33:45 +0100 From: Maxime Ripard To: Andre Przywara Cc: Hans de Goede , Emilio =?iso-8859-1?Q?L=F3pez?= , Michael Turquette , Chen-Yu Tsai , Stephen Boyd , linux-clk@vger.kernel.org, linux-sunxi@googlegroups.com, linux-arm-kernel@lists.infradead.org, Rob Herring , Mark Rutland , devicetree@vger.kernel.org, Grant Likely , Frank Rowand Subject: Re: Allwinner A64 MMC support Message-ID: <20160216093345.GJ4344@lukather> References: <1455300000-18723-1-git-send-email-andre.przywara@arm.com> <56BE29A1.3060107@redhat.com> <56C1A590.6030500@arm.com> <56C1C7BB.1040602@redhat.com> <56C1D9C0.6020601@arm.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="6J7GEvtanOfV9oXA" In-Reply-To: <56C1D9C0.6020601@arm.com> List-ID: --6J7GEvtanOfV9oXA Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Hi Andre, On Mon, Feb 15, 2016 at 01:59:28PM +0000, Andre Przywara wrote: > 2) The reset node is a allwinner,sun6i-a31-ahb1-reset, which does not > get registered automatically, but by an explicit call from mach-sunxi/. > I wonder why we do this (yes, I saw that comment, but still...) and if > it would work with a normal MODULE_DEVICE_TABLE() declaration. Because we need to access the reset controller at the time the timer is probed, which is way before the device model is initilized, and the "real" reset controller driver is probed. I'm not really fond of the solution either, but there's no other solution, if we set aside "rework the driver / device model so that it initializes properly". > There is and will be no equivalent to the mach- directory on arm64. Then you cannot use the HS timer. Maxime --=20 Maxime Ripard, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com --6J7GEvtanOfV9oXA Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJWwuz5AAoJEBx+YmzsjxAgrEsQALS4hnb2da81VuSuMKXJpjk9 yaKKiQ6kwqkJ955Sk8M6wnsUZCBG8cIi3yFaUILUzO+DSULU3qgp9M88gkM/lVuQ 5O+XrSm29KDlYcvaZHrk3mRiEUem6mOTKmUdI/8Q6Z/ay/4qOaDaFXHqw8ZIxLMn rtveRG1r2uuqF73Fi45S1+oWKR1SeR5CUL00/y9eK6vtiQWpE8QMCaAVHaEgRBUh zpwln0943VSIdYxstbdBPbZ+GKUB5MSBNISBdhkEvcnf/fW9rPWhEqBPfNDnPl/3 2VwgevoarWD4XkjwVDJEVeRSiR0zxHplKair3ehaIzRzipcpcf1KOl0+njhY5Hf+ /GTiikwRy2MHrCxI3HgPxCokCX5ZekogrQ5C9MwkKcukONLbzCB7acw9O3NIaw5L hRF3yLTwu0MkUuYIx8nbyvgbVLCidFlfTupcAn+V0GmGlEv8UWiyHfUFKPLgkKMX faUzjyPvNsKbptDKstSrZn0tPPPchHSgjKb18h2GZ8HtNiRQLbu7oPcmmONZxPzW DUkgNsr/wIqz/lFK2DfD1aVI1Q6trqcMVsujLP37jPrx0ANs5Zjhy+J20jQJDKvh UWOHk+lxY5ZyMZ/Ky1jAMyE9I69xSTeHgLUuZcZ2Pwt0AYVJiRpjTs75SkYxmXwT fPAwOyFoUGbqt7Zs98JO =8CZ+ -----END PGP SIGNATURE----- --6J7GEvtanOfV9oXA-- From mboxrd@z Thu Jan 1 00:00:00 1970 From: maxime.ripard@free-electrons.com (Maxime Ripard) Date: Tue, 16 Feb 2016 10:33:45 +0100 Subject: Allwinner A64 MMC support In-Reply-To: <56C1D9C0.6020601@arm.com> References: <1455300000-18723-1-git-send-email-andre.przywara@arm.com> <56BE29A1.3060107@redhat.com> <56C1A590.6030500@arm.com> <56C1C7BB.1040602@redhat.com> <56C1D9C0.6020601@arm.com> Message-ID: <20160216093345.GJ4344@lukather> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Andre, On Mon, Feb 15, 2016 at 01:59:28PM +0000, Andre Przywara wrote: > 2) The reset node is a allwinner,sun6i-a31-ahb1-reset, which does not > get registered automatically, but by an explicit call from mach-sunxi/. > I wonder why we do this (yes, I saw that comment, but still...) and if > it would work with a normal MODULE_DEVICE_TABLE() declaration. Because we need to access the reset controller at the time the timer is probed, which is way before the device model is initilized, and the "real" reset controller driver is probed. I'm not really fond of the solution either, but there's no other solution, if we set aside "rework the driver / device model so that it initializes properly". > There is and will be no equivalent to the mach- directory on arm64. Then you cannot use the HS timer. Maxime -- Maxime Ripard, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 819 bytes Desc: Digital signature URL: From mboxrd@z Thu Jan 1 00:00:00 1970 From: Maxime Ripard Subject: Re: Allwinner A64 MMC support Date: Tue, 16 Feb 2016 10:33:45 +0100 Message-ID: <20160216093345.GJ4344@lukather> References: <1455300000-18723-1-git-send-email-andre.przywara@arm.com> <56BE29A1.3060107@redhat.com> <56C1A590.6030500@arm.com> <56C1C7BB.1040602@redhat.com> <56C1D9C0.6020601@arm.com> Reply-To: maxime.ripard-wi1+55ScJUtKEb57/3fJTNBPR1lH4CV8@public.gmane.org Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="6J7GEvtanOfV9oXA" Return-path: Content-Disposition: inline In-Reply-To: <56C1D9C0.6020601-5wv7dgnIgG8@public.gmane.org> List-Post: , List-Help: , List-Archive: , List-Unsubscribe: , To: Andre Przywara Cc: Hans de Goede , Emilio =?iso-8859-1?Q?L=F3pez?= , Michael Turquette , Chen-Yu Tsai , Stephen Boyd , linux-clk-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-sunxi-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org, Rob Herring , Mark Rutland , devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Grant Likely , Frank Rowand List-Id: devicetree@vger.kernel.org --6J7GEvtanOfV9oXA Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline Hi Andre, On Mon, Feb 15, 2016 at 01:59:28PM +0000, Andre Przywara wrote: > 2) The reset node is a allwinner,sun6i-a31-ahb1-reset, which does not > get registered automatically, but by an explicit call from mach-sunxi/. > I wonder why we do this (yes, I saw that comment, but still...) and if > it would work with a normal MODULE_DEVICE_TABLE() declaration. Because we need to access the reset controller at the time the timer is probed, which is way before the device model is initilized, and the "real" reset controller driver is probed. I'm not really fond of the solution either, but there's no other solution, if we set aside "rework the driver / device model so that it initializes properly". > There is and will be no equivalent to the mach- directory on arm64. Then you cannot use the HS timer. Maxime -- Maxime Ripard, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com --6J7GEvtanOfV9oXA--