Date: Fri, 21 Feb 2014 18:42:23 -0700 From: Warner Losh <wlosh@bsdimp.com> To: Ian Lepore <ian@FreeBSD.org> Cc: arm@freebsd.org Subject: Re: svn commit: r262244 - head/sys/arm/freescale/imx Message-ID: <BE6D12EB-2B60-4694-97F4-133512B18411@bsdimp.com> In-Reply-To: <201402201429.s1KETxk3090247@svn.freebsd.org> References: <201402201429.s1KETxk3090247@svn.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Feb 20, 2014, at 7:29 AM, Ian Lepore wrote:
> + * Resist the temptation to change the #if 0 to #ifdef EARLY_PRINTF =
here. It
> + * makes sense now, but if multiple SOCs do that it will make =
early_putc another
> + * duplicate symbol to be eliminated on the path to a generic kernel.
> + */
> +#if 0=20
> +static void=20
> +imx6_early_putc(int c)
> +{
> + volatile uint32_t * UART_STAT_REG =3D (uint32_t *)0x02020098;
> + volatile uint32_t * UART_TX_REG =3D (uint32_t *)0x02020040;
> + const uint32_t UART_TXRDY =3D (1 << 3);
> +
> + while ((*UART_STAT_REG & UART_TXRDY) =3D=3D 0)
> + continue;
> + *UART_TX_REG =3D c;
> +}
> +early_putc_t *early_putc =3D imx6_early_putc;
> +#endif
I'm starting to think we should have code like this in a platform board =
specific file that's matched based on the top-level compatible =
strings...
Doing this is like 8th on my list right now...
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?BE6D12EB-2B60-4694-97F4-133512B18411>