Stephen,
On Wed, Oct 3, 2012 at 1:38 PM, Stephen Warren swarren@wwwdotorg.org wrote:
On 10/02/2012 04:45 PM, Tom Warren wrote:
This patch adds basic Tegra30 (T30) build support - no specific board is targeted.
diff --git a/board/nvidia/common/board.c b/board/nvidia/common/board.c
+#ifdef CONFIG_USB_EHCI_TEGRA +#include <asm/arch/usb.h> +#endif
That seems unrelated.
It's related because I don't have a usb.h file for Tegra30 yet (not used/needed), and this is a common file, so I have to #ifdef the inclusion or Cardhu won't build. I could put an empty usb.h file in arch-tegra30, but I don't like that approach and there's a need for this if someone turns off USB support in a Tegra20 build. The code further down in the file uses the same #ifdef.
+#ifdef CONFIG_TEGRA30 +#include <asm/arch/pinmux-config-common.h> +#endif
...
/*
- Routine: pinmux_init
- Description: Do individual peripheral pinmux configs
- */
+static void pinmux_init(void) +{ +#if defined(CONFIG_TEGRA30)
pinmux_config_table(tegra3_pinmux_common,ARRAY_SIZE(tegra3_pinmux_common));pinmux_config_table(unused_pins_lowpower,ARRAY_SIZE(unused_pins_lowpower));+#endif +}
I think that should be done by the Cardhu board file, since the data is Cardhu-specific.
If I move pinmux_init to cardhu.c, I'll need matching (empty) funcs in all the other board files, since it's called in board_early_init_f later in board.c. Or it could be made a weak function that's only overridden in cardhu files. But all this is more that I wanted to change in this first pass of T30/Cardhu support - I didn't want to touch any of the Tegra20 builds that I didn't have to.
Tom