On Mon, Aug 06, 2012 at 04:24:06PM +0200, Uwe Kleine-K?nig wrote:

On Tue, Jul 31, 2012 at 08:32:06PM +0200, Uwe Kleine-K?nig wrote:

quoted

Hello Sascha,

On Tue, Jul 31, 2012 at 07:43:58PM +0200, Sascha Hauer wrote:

quoted

On Tue, Jul 31, 2012 at 03:44:39PM +0200, Uwe Kleine-K?nig wrote:

quoted

Otherwise printch et al. and printk with earlyprintk and keep_bootcon
becomes unfunctional when the clk framework disables all unused clocks.

Why don't you use the regular console? Letting the clock framework
disable the clocks was pretty much intentional.

I don't question that disabling unused clocks is sensible. But usually
if you enable DEBUG_LL you want to use it even after the clock framework
disabled all (apparently) unused clocks. And in my case the regular
console didn't work yet and so I relied on earlyprintk.
Even if the problem proved to be the result of my stupidity this patch
asserts that earlyprintk just works as it should be for debug aids.

ping

Honestly, I don't think this is a good idea. You fixed one UART on one
SoC, so ARRAY_SIZE(i.MX1, i.MX21, i.MX25, i.MX27, i.MX31, i.MX35,
i.MX51, i.MX53, i.MX6) * NUM_UARTS remain broken for this usecase. IMO
DEBUG_LL is to get some stuff out of an otherwise silent board, nothing
more.

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 |