[RESEND PATCH v2] ARM: debug: add support for Palmchip BK-310x UART

[Marc Gonzalez]

On 29/01/2016 17:50, Måns Rullgård wrote:

> Marc Gonzalez <marc_gonzalez at sigmadesigns.com> writes:
> 
>> On 27/01/2016 01:44, Måns Rullgård wrote:
>>
>>> Kevin Hilman <khilman at baylibre.com> writes:
>>>
>>>> Marc Gonzalez <marc_gonzalez at sigmadesigns.com> writes:
>>>>
>>>>> From: Mans Rullgard <mans at mansr.com>
>>>>>
>>>>> Some SoCs use a Palmchip BK-310x UART which is mostly 16550 compatible
>>>>> but with a different register layout.
>>>>
>>>> ... because UART register layout is where we really need innovation. ;)
>>>>
>>>>> While this UART has previously
>>>>> only been supported in MIPS based chips (Alchemy, Ralink), the ARM based
>>>>> SMP87xx series from Sigma Designs also uses it.
>>>>>
>>>>> This patch allows the debug console to work with this type of UART.
>>>>>
>>>>> Signed-off-by: Mans Rullgard <mans at mansr.com>
>>>>> Signed-off-by: Marc Gonzalez <marc_gonzalez at sigmadesigns.com>
>>>>
>>>> It appears this was previously acked by Arnd also?
>>>> https://lkml.org/lkml/2015/11/9/116
>>>>
>>>>> ---
>>>>> Peter, this is the patch I've been using on Tango for earlyprintk support.
>>>>> arm-soc, this is just the Nov 8 patch with my Signed-off-by tag.
>>>>> ---
>>>>>  arch/arm/Kconfig.debug            |  9 +++++++++
>>>>>  arch/arm/include/debug/palmchip.S | 11 +++++++++++
>>>>>  2 files changed, 20 insertions(+)
>>>>>  create mode 100644 arch/arm/include/debug/palmchip.S
>>>>>
>>>>> diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug
>>>>> index 259c0ca9c99a..708aa9150c00 100644
>>>>> --- a/arch/arm/Kconfig.debug
>>>>> +++ b/arch/arm/Kconfig.debug
>>>>> @@ -1304,6 +1304,7 @@ config DEBUG_SIRFSOC_UART
>>>>>  config DEBUG_LL_INCLUDE
>>>>>  	string
>>>>>  	default "debug/sa1100.S" if DEBUG_SA1100
>>>>> +	default "debug/palmchip.S" if DEBUG_UART_8250_PALMCHIP
>>>>
>>>> minor nit: We're trying (not always with great success) to keep various
>>>> parts of this file sorted alphanumerically.  Can you please insert the
>>>> new entries accordingly?
>>>
>>> It has to precede the regular 8250 entry so as to take precedence over
>>> that.
>>
>> <confused> Are you all talking about these 2 lines:
>>
>> 	default "debug/palmchip.S" if DEBUG_UART_8250_PALMCHIP
>> 	default "debug/8250.S" if DEBUG_LL_UART_8250 || DEBUG_UART_8250
>>
>> default "debug/palmchip.S" is supposed to come before default "debug/8250.S" ?
>>
>> Because, in my config, there is no DEBUG_LL_UART_8250 to be seen, and
>> # CONFIG_DEBUG_UART_8250 is not set
> 
> CONFIG_DEBUG_LL_UART_8250=y here.  In fact DEBUG_UART_8250_PALMCHIP
> depends on "DEBUG_LL_UART_8250 || DEBUG_UART_8250" so one of those
> *must* be set for any of this to matter.

Doh! I was looking at a config where earlyprintk isn't enabled... o_O

config DEBUG_UART_8250_PALMCHIP
	bool "8250 UART is Palmchip BK-310x"
	depends on DEBUG_LL_UART_8250 || DEBUG_UART_8250

OK, I understand now. Thanks.

Regards.