reset16.inc and 32bit reset vector

[Jeremy Jackson]

. = 0x8;
         .code32
         jmp     protected_start
         .previous

What in the world would use this?  Is there am i386 CPU that starts in 
protected mode?

Jeremy Jackson  writes:
>      . = 0x8;
>      .code32
>      jmp     protected_start
>      .previous
>
> What in the world would use this?  Is there am i386 CPU that starts in protected
> mode?

The normal LinuxBIOS image.

The fallback LinuxBIOS image starts at the reset vector.
Then it checks the cmos options to see if everything is ok.
Decrements the counter in the cmos in case the count gets too high.
Then it jumps to the end of the normal image.

Essentially it is easier to fix the end address then to fix the
start address.

Eric

[Jeremy Jackson]

Eric W. Biederman wrote:
> The normal LinuxBIOS image.
>
> The fallback LinuxBIOS image starts at the reset vector.
> Then it checks the cmos options to see if everything is ok.
> Decrements the counter in the cmos in case the count gets too high.
> Then it jumps to the end of the normal image.

I could use some pointers, I find this a bit cryptic.

[ron minnich]

On Wed, 1 Sep 2004, Jeremy Jackson wrote:
>      . = 0x8;
>      .code32
>      jmp     protected_start
>      .previous

note the offset of 8. this is called by code once we're in protected mode. 
This is not the reset vector

ron

[Jeremy Jackson]

ron minnich wrote:
> On Wed, 1 Sep 2004, Jeremy Jackson wrote:
> >     . = 0x8;
> >     .code32
> >     jmp     protected_start
> >     .previous
>
> note the offset of 8. this is called by code once we're in protected mode. 
> This is not the reset vector

I gathered that, I was wondering what code might be calling it, and does 
it have to be there for that code to find it.