On Wed, Nov 07, 2018 at 11:12:35AM +0100, Ladislav Michl wrote: > On Mon, Nov 05, 2018 at 04:50:51PM +0100, Sascha Hauer wrote: > > The following series contains a bigger at91 header file cleanup. With > > this the SoC header files become completely SoC namespaced which means > > they can all be included as desired without conflicts. This makes the > > way free for more at91 multiboard support. > > > > I don't have that much AT91 hardware on my desk, so I am unable to test > > this properly. Testing feedback very much appreciated. Sam maybe? > > Tested on custom AT91SAM9G20 based board. However as patch > "ARM: at91: separate restart handler registration into SoC specific code" > does exactly what its commit message says it brings some issues with > devicetree based boot. Nearly everything works except NAND and that would > work too if one uses old binding. As that is not my case, I'm using pdata > to register NAND, so I need also: > > diff --git a/arch/arm/mach-at91/Makefile b/arch/arm/mach-at91/Makefile > index d81683ac1..75442bd72 100644 > --- a/arch/arm/mach-at91/Makefile > +++ b/arch/arm/mach-at91/Makefile > @@ -18,11 +18,12 @@ obj-$(CONFIG_ARCH_AT91RM9200) += at91rm9200.o at91rm9200_time.o at91rm9200_devic > obj-$(CONFIG_ARCH_AT91SAM9260) += at91sam9260.o at91sam9260_devices.o > obj-$(CONFIG_ARCH_AT91SAM9261) += at91sam9261.o at91sam9261_devices.o > obj-$(CONFIG_ARCH_AT91SAM9G10) += at91sam9261.o at91sam9261_devices.o > +obj-$(CONFIG_ARCH_AT91SAM9G20) += at91sam9260_devices.o > ifeq ($(CONFIG_OFDEVICE),) > obj-$(CONFIG_ARCH_AT91SAM9263) += at91sam9263.o at91sam9263_devices.o > obj-$(CONFIG_ARCH_SAMA5D3) += sama5d3.o sama5d3_devices.o > +obj-$(CONFIG_ARCH_AT91SAM9G20) += at91sam9260.o > endif > -obj-$(CONFIG_ARCH_AT91SAM9G20) += at91sam9260.o at91sam9260_devices.o > obj-$(CONFIG_ARCH_AT91SAM9G45) += at91sam9g45.o at91sam9g45_devices.o > obj-$(CONFIG_ARCH_AT91SAM9X5) += at91sam9x5.o at91sam9x5_devices.o > obj-$(CONFIG_ARCH_AT91SAM9N12) += at91sam9n12.o at91sam9n12_devices.o > > Note, that with above change, once we move to device tree, there's noone > to register restart handler. I guess, that will get eventually sorted > by writing proper reset driver, right? For now I put reset handler > registration into a board file. On i.MX I have split the SoC init code into code that has always to be executed and code that has only to be executed without devicetree support. We can do the same on AT91 as a next step. Writing a proper reset driver is probably a good step but not required. Sascha -- Pengutronix e.K. | | Industrial Linux Solutions | http://www.pengutronix.de/ | Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 | _______________________________________________ barebox mailing list barebox@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/barebox
