[U-Boot] [patch arm/next] dm355 evm support (v2)

David Brownell david-b at pacbell.net
Tue May 12 03:56:06 CEST 2009


On Monday 11 May 2009, Jean-Christophe PLAGNIOL-VILLARD wrote:

> > +#define BIT(x) (1 << (x))
>
> please remove
> or use set_bit

I think that should probably be added to all the bitops.h headers,
or someplace similar.  But, OK; __{set,clear}_bit() work here too.


> > +/*#define CONFIG_SYS_NAND_SMALLPAGE */
>
> please remove if no need

OK, I'll just make a later #ifdef have an #else for that,
kicking in when someone swaps out the largepage chip for
a smallpage one.


> > +#define CONFIG_SYS_NAND_LARGEPAGE
> > +#define CONFIG_SYS_NAND_BASE_LIST	{ 0x02000000, }
> > +/* socket has two chipselects, nCE0 gated by address BIT(14) */
> > +#define CONFIG_SYS_MAX_NAND_DEVICE	1
> > +#define CONFIG_SYS_NAND_MAX_CHIPS	2
> > +
> > +/* USB: OTG connector */
> > +/* #define CONFIG_USB_DAVINCI */
> please remove if no need

That's a "NYET".  It initializes a bit differently on this
board because the on-chip PHY is newer, and also interchanges
the D+/D- lines for better signal routing.  It's there as a
placeholder, and will be enabled later.


> > +
> > +/*#define CONFIG_BOOTDELAY	5 */
> please remove if no need

Again a NYET.  When the NAND support kicks in, there will
be an environment, and this can be enabled.  Until then,
it's a placeholder ... when booting from MMC, UART or JTAG,
with no environment, hands-off booting is ill-advised.

Updated patch coming soon.

- dave




More information about the U-Boot mailing list