Re: [PATCH v3 0/4] Enabling patches of 2 early consoles for Intel MID platforms

[Ingo Molnar <mingo@elte.hu>]

* Feng Tang <feng.tang@intel.com> wrote:

> Hi all,
> 
> This serie mainly enables 2 early consoles named mrst and hsu for
> Intel MID platforms. please help to review, thanks,
> 
> - Feng
> 
> Changelog:
> 	v3:
> 	     * Fix a warning for mrst.h
> 	     * Add 2 helper funcs for fixed address mapping as
> 	       suggested by Ingo
> 	     * Rename the mrst_earlyprintk.c to early_printk_mrst.c
> 	     * Clean the code and fix typoes upon Ingo's comments
> 	v2:
> 	     * Code clean as suggested by Ingo 

Thanks, this series looks really good!

> Feng Tang (4):
>   x86, mrst: make mrst_identify_cpu() inlne inside mrst.h
>   x86: add two helper macros for fixed address mapping
>   x86, earlyprintk: add earlyprintk for Intel Moorestown platform
>   x86, earlyprintk: add hsu early console for Intel Medfield platform
> 
>  arch/x86/Kconfig.debug              |    4 +
>  arch/x86/include/asm/fixmap.h       |   15 ++
>  arch/x86/include/asm/mrst.h         |   10 +-
>  arch/x86/kernel/Makefile            |    1 +
>  arch/x86/kernel/early_printk.c      |   13 ++
>  arch/x86/kernel/early_printk_mrst.c |  319 +++++++++++++++++++++++++++++++++++
>  6 files changed, 361 insertions(+), 1 deletions(-)
>  mode change 100644 => 100755 arch/x86/include/asm/fixmap.h
>  create mode 100644 arch/x86/kernel/early_printk_mrst.c

Wrt. integration: i'd prefer to carry these four patches in the 
tip:x86/early-printk tree (to be merged into v2.6.37), because
there's some interaction with other x86 bits.

AFAICS the two serial patches:

 [PATCH 1/2] serial: mrst_max3110: some code cleanup
 [PATCH 2/2] mrst_max3110: Make the IRQ option runtime

Are structurally independent from the earlyprintk patches and could be 
carried separately in the serial tree.

Alan, Greg, what's your preference?

Thanks,

	Ingo