flat assembler
Message board for the users of flat assembler.

Index > OS Construction > E820 Map

Thread Post new topic Reply to topic

Joined: 14 Feb 2013
Posts: 58
Location: Alberta
Purpose is to build a map at the top of first 64K (1000H), decending order, growing downward and each entry 32 bytes. Padding is for easy visualization in Bochs and less complexity moving segment to next iteration, or back.
        mov     ax, es
        dec     ax
        dec     ax
        mov     es, ax                  ; Bump segment back 32 bytes    
        push     es
        add     word [esp], 2
        pop      es    

Due to lack of any comprehensive documention, I fashioned my algo somewhat after this wiki example and Pure64 code.

Out of curiosity sake, does anyone have the answers to the following;

1: Why does EDX need "SMAP", when it should be obvious what is required by calling routine with E820 in EAX.

2. You would think the callee be designed in such a way that if EAX <> EDX or EBX = 0 on first interation, CF would be set and if so EAX would have an error code.
        call    GetMapEntry
        jc      .Error
        cmp     eax, edx                ; Was magic returned in EAX?
        jnz     .Error
        test    ebx, ebx
        jnz     @F    

Similarly this. Of what value is an entry that doesn't have a size. You would think startup probing memory would eliminate this possibility.
   @@:  mov     eax, [es:8]
        or      eax, [es:12]            ; ZF=1 if entry has null length
        jz      .Skip    

Two more functions and then I can get into protected or long mode and put this BIOS crap behind me. Evil or Very Mad

Filename: Stage2.ASM
Filesize: 3.3 KB
Downloaded: 267 Time(s)

Post 20 Jul 2013, 16:57
View user's profile Send private message Visit poster's website Reply with quote

Joined: 25 Aug 2004
Posts: 618
It is likely that some BIOSs had a handler for ah=0xE8 int 0x15 but no check on al, so another discriminator is needed to prevent conflicts. For whatever reason "SMAP" was chosen (low chance of working anywhere else) so it just stayed that way.

I've never seen a zero length entry but it is possible that for the newer ACPI3 entries which return invalid/bad regions, it will zero the size for backward compatibility if using the old structure size. No doubt there is also another legacy reasons for it as well.

As you have no doubt learned the hard way, the BIOS is a nasty mess Laughing
Post 20 Jul 2013, 21:31
View user's profile Send private message Reply with quote
Display posts from previous:
Post new topic Reply to topic

Jump to:  

< Last Thread | Next Thread >
Forum Rules:
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum

Copyright © 1999-2020, Tomasz Grysztar. Also on GitHub, YouTube, Twitter.

Website powered by rwasa.