From mboxrd@z Thu Jan 1 00:00:00 1970 From: rmallon@gmail.com (Ryan Mallon) Date: Mon, 02 Jul 2012 08:57:11 +1000 Subject: [PATCH] arm: ep93xx: Enable i2c support for ep9302 In-Reply-To: References: <4FD534A5.1030301@gmail.com> <4FD66843.6000504@gmail.com> Message-ID: <4FF0D5C7.9090904@gmail.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 22/06/12 05:33, Martin Guy wrote: > On 11 June 2012 23:50, Ryan Mallon wrote: >> On 12/06/12 01:45, H Hartley Sweeten wrote: >> >>> On Sunday, June 10, 2012 4:58 PM, Ryan Mallon wrote: >>>> On 10/06/12 22:44, Todor Colov wrote: >>>>> >>>>> Enable i2c support for ep9302. i2c support is needed when external RTC is >>>>> used. >>>>> >>>>> Signed-off-by: Todor Colov >>>> >>>> Thanks. Applied to ep93xx-fixes. >>> >>> Ryan, >>> >>> I'm not sure about this patch... >>> >>> The EDB9302 does not have a built-on external RTC connected to the I2C >>> bus. The EECLK and EEDAT pins are simply routed to two headers. They are >>> not connected to any I2C devices built onto the board. >>> >>> This patch will cause _all_ edb9302 boards to bind the ds1337 driver to the >>> I2C bus but that device will only exist if a user has connected the part >>> externally to the board. >>> >>> Just my 2 cents... >> >> >> Hmm, the binding is not really a problem, since it will just fail if the >> ds1337 device is not present at that i2c address. I guess the main problem >> is if there are users of the edb9302 who want i2c on the headers and others >> who want to use the pins as gpio. Without being able to auto-detect such >> boards at runtime (which seems very unlikely) I can't think of a good >> solution to this, other than it being time for me to do more DT reading :-). >> >> I'm inclined to leave the patch in. The ep93xx doesn't have a lot of users, >> so I think it is okay to tailor it a bit to the users we do have. I'm happy >> to revert it, and find a better solution if this breaks someone else's >> setup. > > Hi folks > Can I add that the SimpleMachines.it Sim.One board, based on the > EP9307, has a DS1337 on the I2C bus, for which the define is > machine_is_sim_one() However, the existing Linux kernel already > read/writes this chip successfully, so I'm not sure what advantage > this change would bring for that board. Hi Martin, Sorry for the late reply, I've been on holiday. The change only affects the Cirrus EDB-9302 board. The Sim.one is managed by a separate board file. > If you would like to test > your patch on that board too, we may be able to ship you an enginering > sample; if that would be useful please reply by private email. I already have one of the Sim.one boards. Thanks. ~Ryan