All of lore.kernel.org
 help / color / mirror / Atom feed
* Early prints on different UART's for same machine type
       [not found] <mailman.7147.1312868034.20023.linux-arm-kernel@lists.infradead.org>
@ 2011-08-09  5:56 ` Mohammed, Afzal
  2011-08-10  6:37   ` Paul Walmsley
  0 siblings, 1 reply; 2+ messages in thread
From: Mohammed, Afzal @ 2011-08-09  5:56 UTC (permalink / raw)
  To: linux-arm-kernel-bounces, linux-omap; +Cc: Nori, Sekhar

Hi,

This is regarding board with multiple variations using an upcoming SoC from TI.
Variation of the board is detected by reading EEPROM using I2C after
init_machine gets invoked. In one of the variation UART# used is different.
Because of this decompressor logs (and early prints) can't be captured for the
board variant using this different UART#. Our solution for this problem is to
use different mach-id for the board using different UART#. But this solution
seems to be overkill as only to capture decompressor logs, we have to use
a new machine id.

As I2C will not work at the decompressor stage, I2C can't be used to detect
the board type.

As far as I know, ATAG is not parsed in decompressor stage and so ATAG cannot
be used to determine board variant or UART#. Otherwise we will have to add
capability in decompressor to parse ATAG, detect console to be used, and
modify uncompress.h to select proper console for capturing decompressor logs.
This seems a bad solution.

Our bootloader has the capability to detect the board variant.

So the best solution for us so far seems to add a new machine id, if any
better solutions are possible, please let us know.

Regards
Afzal

^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: Early prints on different UART's for same machine type
  2011-08-09  5:56 ` Early prints on different UART's for same machine type Mohammed, Afzal
@ 2011-08-10  6:37   ` Paul Walmsley
  0 siblings, 0 replies; 2+ messages in thread
From: Paul Walmsley @ 2011-08-10  6:37 UTC (permalink / raw)
  To: Mohammed, Afzal; +Cc: linux-arm-kernel-bounces, linux-omap, Nori, Sekhar

Hi

On Tue, 9 Aug 2011, Mohammed, Afzal wrote:

> This is regarding board with multiple variations using an upcoming SoC from TI.
> Variation of the board is detected by reading EEPROM using I2C after
> init_machine gets invoked. In one of the variation UART# used is different.
> Because of this decompressor logs (and early prints) can't be captured for the
> board variant using this different UART#. Our solution for this problem is to
> use different mach-id for the board using different UART#. But this solution
> seems to be overkill as only to capture decompressor logs, we have to use
> a new machine id.
> 
> As I2C will not work at the decompressor stage, I2C can't be used to detect
> the board type.
> 
> As far as I know, ATAG is not parsed in decompressor stage and so ATAG cannot
> be used to determine board variant or UART#. Otherwise we will have to add
> capability in decompressor to parse ATAG, detect console to be used, and
> modify uncompress.h to select proper console for capturing decompressor logs.
> This seems a bad solution.
> 
> Our bootloader has the capability to detect the board variant.
> 
> So the best solution for us so far seems to add a new machine id, if any
> better solutions are possible, please let us know.

You might get more feedback if you E-mail 
<linux-arm-kernel@lists.infradead.org>, rather than 
<linux-arm-kernel-bounces@lists.infradead.org>.


- Paul

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2011-08-10  6:37 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <mailman.7147.1312868034.20023.linux-arm-kernel@lists.infradead.org>
2011-08-09  5:56 ` Early prints on different UART's for same machine type Mohammed, Afzal
2011-08-10  6:37   ` Paul Walmsley

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.