[U-Boot] [PATCH v2 5/5] x86: Remove unused real mode code

[Simon Glass]

Hi Graham,

On Thu, Feb 14, 2013 at 7:47 PM, Graeme Russ <graeme.russ at gmail.com> wrote:
> Hi Simon,
>
> On Fri, Feb 15, 2013 at 2:35 PM, Simon Glass <sjg at chromium.org> wrote:
>> Hi Graeme,
>>
>> On Thu, Feb 14, 2013 at 2:36 PM, Graeme Russ <graeme.russ at gmail.com> wrote:
>>> Hi Simon,
>>>
>>> On Fri, Feb 15, 2013 at 1:18 AM, Simon Glass <sjg at chromium.org> wrote:
>>>> This code is pretty old and we want to support only 32-bit systems now.
>>>> Keep the real mode-to-protected code in case we want to boot an x86 from
>>>> power-up, but invert the polarity of the option.
>>>>
>>>> Signed-off-by: Simon Glass <sjg at chromium.org>
>
> [snip]
>
>>>> diff --git a/README b/README
>>>> index 6106e0d..7b50720 100644
>>>> --- a/README
>>>> +++ b/README
>>>> @@ -3820,10 +3820,9 @@ Low Level (hardware related) configuration options:
>>>>                 If defined, the x86 reset vector code is included. This is not
>>>>                 needed when U-Boot is running from Coreboot.
>>>>
>>>> -- CONFIG_X86_NO_REAL_MODE
>>>> -               If defined, x86 real mode code is omitted. This assumes a
>>>> -               32-bit environment where such code is not needed. You will
>>>> -               need to do this when U-Boot is running from Coreboot.
>>>> +- CONFIG_X86_REAL_MODE
>>>> +               If defined, some 86 real mode code is included. This is not
>>>> +               needed when U-Boot is running from Coreboot.
>>>
>>> I think this config option can be punted completely. This define is
>>> only used to pull in the real-mode trampoline code which switches from
>>> protected-mode to real-mode before jumping into the Linux kernel's
>>> real-mode stub. The code for the initial real-mode to protected-mode
>>> switch will be pulled in by CONFIG_X86_RESET_VECTOR
>>
>> I suppose I am confused about what is needed here. I will drop the
>> flag, and arch/x86/lib/realmode_switch.S. Is that right?
>
> Correct. All you need to keep is:
>
> arch/x86/cpu/resetvec.S
> arch/x86/cpu/start16
>
> And these are conditional on CONFIG_X86_RESET_VECTOR
>
> Looking at the most recent patch, I think you have it nailed now

OK, thanks for your help with it

Regards,
Simon

>
> Regards,
>
> Graeme