Re: [PATCH] Add OMAP4 Panda support

[Kevin Hilman <khilman@deeprootsystems.com>]

"Anders, David" <x0132446@ti.com> writes:

>> 
>> > The panda isn't 100% compatible with the blaze/sdp. There are/will
>> > be some significant differences. I understand the goal to make the
>> > "multi-boot" images and to reduce the amount of duplicated code in
>> > the kernel, however we can learn a little about this situation from
>> > the BeagleBoard. Due to it being a development board, many people
>> > who are new to embedded development will be purchasing the Panda and
>> > they generally need a clearly defined place to start. Blaze and SDP
>> > are officially support TI products and their machine files can be
>> > easily groomed and maintained. Having a clearly defined individual
>> > machine file and defconfig will make things a lot easier for
>> > starting development on the Panda.
>> >
>> > IMHO, I think it is a giant mistake not to have individual machine
>> > files and defconfigs for the panda. This doesn't preclude making
>> > sure that they files work properly with multi-boot images.
>> 
>> In my view, all this helps argue the case for having shared code
>> instead of the copy-paste duplicating.
>> 
>> I am not opposed to having a separate board file for Panda.  What I am
>> opposed to is having a separate board file if it's mostly identical to
>> another board.
>> 
>> If Panda and Blaze are mostly the same, they should share the same
>> board code and use their separate board files to express their
>> differences.  See how this was done for Zoom2 and Zoom3 which have a
>> shared set of peripherals and also a shared debug board.
>> 
>> When customers go to build their designs based on Panda, they can
>> still start from the Panda board file and start hacking away.  These
>> custom boards will then also use the shared common code instead of
>> duplicating.
>> 
>> Kevin
>
> Panda will have different Ethernet, different video outputs,
> different usb host controller configurations, and a number of
> different pin muxes. Is there a quantitative way of deciding if
> there are enough similarities/differences between boards for having
> combined/individual machine files?

Not really more of a feeling.

But these feel like important enough differences to merit a separate
board file.  I was under the impression that there was a lot more in
common.

So, I'm OK with the separate board file, please fix up the minor comments below...

> I agree that Blaze and SDP are similar enough not to have separate files. 
>
>> 
>> 
>> > -----Original Message-----
>> > From: Kevin Hilman [mailto:khilman@deeprootsystems.com]
>> > Sent: Friday, June 18, 2010 9:32 AM
>> > To: Anders, David
>> > Cc: linux-omap@vger.kernel.org; tony@atomide.com
>> > Subject: Re: [PATCH] Add OMAP4 Panda support
>> >
>> > "Anders, David" <x0132446@ti.com> writes:
>> >
>> >> Add initial support for the OMAP4430 based Panda board.
>> >>
>> >> Signed-off-by: David Anders <x0132446@ti.com>
>> >> ---
>> >>  arch/arm/configs/omap4_panda_defconfig | 1094
>> ++++++++++++++++++++++++++++++++
>> >>  arch/arm/mach-omap2/Kconfig            |    4 +
>> >>  arch/arm/mach-omap2/Makefile           |    1 +
>> >>  arch/arm/mach-omap2/board-omap4panda.c |   80 +++
>> >>  4 files changed, 1179 insertions(+), 0 deletions(-)
>> >>  create mode 100644 arch/arm/configs/omap4_panda_defconfig
>> >>  create mode 100644 arch/arm/mach-omap2/board-omap4panda.c
>> >
>> > [...]
>> >
>> >> diff --git a/arch/arm/mach-omap2/board-omap4panda.c b/arch/arm/mach-
>> omap2/board-omap4panda.c
>> >> new file mode 100644
>> >> index 0000000..95faec0
>> >> --- /dev/null
>> >> +++ b/arch/arm/mach-omap2/board-omap4panda.c
>> >> @@ -0,0 +1,80 @@
>> >> +/*
>> >> + * Board support file for OMAP4430 based Panda.
>> >> + *
>> >> + * Copyright (C) 2010 Texas Instruments
>> >> + *
>> >> + * Author: David Anders <x0132446@ti.com>
>> >> + *
>> >> + * Based on mach-omap2/board-4430sdp.c
>> >> + *
>> >> + * Author: Santosh Shilimkar <santosh.shilimkar@ti.com>
>> >> + *
>> >> + * Based on mach-omap2/board-3430sdp.c
>> >> + *
>> >> + * This program is free software; you can redistribute it and/or
>> modify
>> >> + * it under the terms of the GNU General Public License version 2 as
>> >> + * published by the Free Software Foundation.
>> >> + */
>> >> +
>> >> +#include <linux/kernel.h>
>> >> +#include <linux/init.h>
>> >> +#include <linux/platform_device.h>
>> >> +#include <linux/io.h>
>> >> +#include <linux/gpio.h>
>> >> +#include <linux/usb/otg.h>
>> >> +
>> >> +#include <mach/hardware.h>
>> >> +#include <mach/omap4-common.h>
>> >> +#include <asm/mach-types.h>
>> >> +#include <asm/mach/arch.h>
>> >> +#include <asm/mach/map.h>
>> >> +
>> >> +#include <plat/board.h>
>> >> +#include <plat/common.h>
>> >> +#include <plat/control.h>
>> >> +#include <plat/timer-gp.h>
>> >> +#include <plat/usb.h>
>> >> +
>> >> +static void __init omap4_panda_init_irq(void)
>> >> +{
>> >> +       omap2_init_common_hw(NULL, NULL);
>> >> +#ifdef CONFIG_OMAP_32K_TIMER
>> >> +       omap2_gp_clockevent_set_gptimer(1);
>> >> +#endif

This part is not needed (here or in board-4430sdp.c) as the default
timer is already 1.

>> >> +       gic_init_irq();
>> >> +       omap_gpio_init();
>> >> +}
>> >> +
>> >> +static struct omap_musb_board_data musb_board_data = {
>> >> +       .interface_type         = MUSB_INTERFACE_UTMI,
>> >> +       .mode                   = MUSB_PERIPHERAL,
>> >> +       .power                  = 100,
>> >> +};
>> >> +
>> >> +static void __init omap4_panda_init(void)
>> >> +{
>> >> +       omap_serial_init();
>> >> +       /* OMAP4 SDP uses internal transceiver so register nop
>> transceiver */
>> >

s/SDP/Panda/

Please change (or remove) this comment which was just copied from the
SDP code.

Kevin