From mboxrd@z Thu Jan 1 00:00:00 1970 From: florian.vaussard@epfl.ch (Florian Vaussard) Date: Mon, 10 Sep 2012 10:30:14 +0200 Subject: [PATCH v8 0/3] GPMC driver conversion In-Reply-To: <20120907221034.GP1303@atomide.com> References: <20120907221034.GP1303@atomide.com> Message-ID: <504DA516.1020405@epfl.ch> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hello, Le 08/09/2012 00:10, Tony Lindgren a ?crit : > * Afzal Mohammed [120905 05:37]: >> Hi, >> >> Basic gpmc driver conversion series. Driver that is now created out of >> gpmc code is a simple one, it handles tasks that were earlier executed >> by gpmc_init. Now instead of relying on cpu_is_* checks, it obtains >> resources and clk handle in the standard Linux way. The existing gpmc >> interface works as was without this series. >> >> HWMOD patch also has been brought into this series back with v7 series >> >> As this creates only a basic driver, further gpmc driver work can be >> based over this, while having a driver first in place. >> >> This series is based on l-o/testing-cleanup as on 5-Sep-12, >> i.e. over commit, >> >> e3a5c14 ARM: OMAP1: Move SoC specific headers from plat to mach for omap1 >> >> per Tony's suggestion. >> >> It is available >> @git://gitorious.org/x0148406-public/linux-kernel.git gpmc-drv-v8 > Great, this all looks good to me. I suggest that on top of this > we add minimal devicetree binding that does not even attempt to > deal with the timings yet. > > Then once the minimal devicetree binding is in place, we can > call the current GPMC timing functions using the compatible flags > and auxdata in the gpmc.c driver. > > That way we get all the legacy boards booting with GPMC support > for smsc911x etc, and can even remove some less used board-*.c > files. This would be great. If you happen to have DT bindings for GPMC, I would be happy to test them with my Overo board, as well as with my CAN chips on a custom board. Regards, Florian