From mboxrd@z Thu Jan 1 00:00:00 1970 From: maxime.ripard@free-electrons.com (Maxime Ripard) Date: Tue, 26 Mar 2013 10:20:21 +0100 Subject: [PATCH v2 0/3] clock driver for sunxi In-Reply-To: <20130322164054.834.64867@quantum> References: <1358835176-7197-1-git-send-email-emilio@elopez.com.ar> <1361803468-9899-1-git-send-email-emilio@elopez.com.ar> <20130321215429.834.60042@quantum> <514C30E1.2020203@free-electrons.com> <20130322164054.834.64867@quantum> Message-ID: <51516855.3020609@free-electrons.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Mike, Le 22/03/2013 17:40, Mike Turquette a ?crit : > Quoting Maxime Ripard (2013-03-22 03:22:25) >> Hi Mike, >> >> Le 21/03/2013 22:54, Mike Turquette a ?crit : >>> Quoting Emilio L?pez (2013-02-25 06:44:25) >>>> Hello everyone, >>>> >>>> This patchset adds basic clock support for sunxi devices. Currently, it >>>> implements support for the two oscillators, the main PLL, the CPU mux, >>>> its three divisor clocks and APB1. With this in place, it is possible to >>>> write a cpufreq driver and have it work, and with a little more code, >>>> represent the UART clocks correctly (I will be sending a separate >>>> patchset for this). >>>> >>>> I have tested this driver successfully on a Cubieboard (A10, sun4i) >>>> using the cpufreq driver from the linux-sunxi tree after minor >>>> modifications (the clock names are not the same). >>>> >>>> Any feedback will be highly appreciated >>>> >>> >>> Emilio, >>> >>> This is a nice series. I appreciate the amount of comments in the code, >>> the kerneldoc and also the added sunxi clock documentation. >>> >>> Concerning the discussion from the v1 series, I prefer for the struct >>> clk instances in the tree to match 1-to-1 with hardware clocks. As the >>> number of clocks in a system scales into the hundreds there are >>> certainly advantages to having fewer clocks in the hierarchy, not the >>> least of which is more easily associating these objects with hardware >>> clocks during a deep debug session. >>> >>> That can be changed at a later time if you wish. Do you want me to take >>> the series into clk-next or do you just want my ack? >> >> Thanks for your feedback! >> >> Unless you have some objection, I was planning to merge it through my >> tree, with your Acked-by, of course. >> > > Acked-by: Mike Turquette Arnd would prefer to get this serie going through your tree, could you merge the patches 1/3 and 3/3 with my Acked-by? Thanks, Maxime -- Maxime Ripard, Free Electrons Embedded Linux, Kernel and Android engineering http://free-electrons.com