From mboxrd@z Thu Jan 1 00:00:00 1970 From: boris.brezillon@bootlin.com (Boris Brezillon) Date: Mon, 10 Sep 2018 11:49:27 +0200 Subject: [PATCH v2 13/16] nvmem: add support for cell lookups from machine code In-Reply-To: References: <20180907100750.14564-1-brgl@bgdev.pl> <20180907100750.14564-14-brgl@bgdev.pl> <484b6ec5-cd8e-e5c5-0c5c-2f11c504ea1c@linaro.org> <20180910102324.52ecd8f7@bbrezillon> Message-ID: <20180910114927.083e2de8@bbrezillon> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Mon, 10 Sep 2018 11:45:48 +0200 Bartosz Golaszewski wrote: > 2018-09-10 10:55 GMT+02:00 Srinivas Kandagatla : > > > > > > On 10/09/18 09:23, Boris Brezillon wrote: > >> > >> Well, if we get rid of nvmem-machine.h, the cell-lookup stuff > >> should go in nvmem-consumer.h not nvmem-provider.h. On the other hand, > >> everything that is related to cell creation should be placed in > >> nvmem-provider.h. > > > > Yes, this is how it should be! > > > > Any actual reason for not putting these definitions into a separate > 'machine' header? This approach is currently used by gpio, pinctrl, > iio and regulator framework because most systems use either DT or ACPI > and don't need to pull in any stuff aimed at board files. I'm perfectly fine with the separate header file, all I'm saying is, if Srinivas does not want nvmem-machine.h, definitions should be placed in nvmem-provider.h or nvmem-consumer.h depending on who they're meant to be used by (providers or consumers).