when I jmp 0000:7C00h in pmode

when I disassemble it and it shows EA00:7C00
instead of 0000:7C00h ???

Which disassembler are you using? Check you set the use32/16 assembler mode correctly.

BTW: A jump to a zero segment in protected mode should cause a GP fault, since zero in the CS segment register is not permitted by design.

Are your descriptors zero based ?, if so you can just jmp 7C000h or you can try the address - the base, if they are not 0 based.
Anyway as your trying to jump to the boot code, you should be in realmode anyway

in pmode, you can jump to [base=7c0h,size=512]:0
in rel mode you can too.
then, what does the BIOS?
does it load at [0:7C00h]?
[7:c0]?, [7c:0]?,[7c0:0]
then, what is preferable?

it is also possible to make an org independant bootloader.
something that can be executed from anywhere in ram?

jmp 0:7C00h

or


push WORD [0x0000]
push WORD [0x7C00]
retf

or

org 0x7C00



real mode or PMode? or both.. which do you refer to..

So you are using real mode then. Those instructions above all would not work in pmode.

both, using only instructions common to both modes, or maybe a sort of LUT with a higher level instruction set.

if PM, then, translate with LUTPM
if RM, translate with LUTRM.

something like this.
this is possible to make a very simple object oriented design.

Then you WILL get a GPF because the selector value of ZERO is deliberately made invalid in PM. Also, you didn't reveal what disassembler you use nor whether you use 16-bit or 32-bit PM nor how you initialize it (GDT ???)

I used different disasmeblers.. but any way..
I think I have it now..

Thanks for the replies..