[英]Incorrect Bytes Being Read from disk using a Simple Boot sector, in QEmu
So I am trying to learn how to make a little OS, primarily using this tutorial thing. 所以我正在尝试学习如何制作一个小操作系统,主要是使用本教程的东西。
I am at page 28 of that tutorial, trying to do the little disk-loader. 我在该教程的第28页,试图做一些小磁盘加载器。
However, when I run qemu-system-i386 boot_sector.bin
, which worked for the previous boot sector examples, the hex output I get is 0xDADA 0xZRWV
, the first of which is correct, while the second is not. 但是,当我运行
qemu-system-i386 boot_sector.bin
,它对前面的引导扇区示例0xDADA 0xZRWV
,我得到的十六进制输出是0xDADA 0xZRWV
,第一个是正确的,而第二个不是。
My code is as follows. 我的代码如下。
Boot_sector.asm Boot_sector.asm
; Very simple boot sector to read sectors from disk
jmp main ; ensures that we start where we want to
%include "disk_load.asm"
BOOT_DRIVE:
db 0
[org 0x7c00] ; tell the assembler where this code will be loaded
main:
mov bp, 0x8000 ; Make sure the stack is out of our way
mov sp, bp
mov bx, 0x9000
mov dh, 2 ; Want to read 2 sectors
call disk_load
mov dx, [0x9000]
call hex_print ; Should output 0xDADA
mov dx, [0x9000 + 512]; Should output 0xFACE
call hex_print
hang:
jmp hang ; Continually jump to current location
; Add padding and magic BIOS number (to let BIOS know this is a boot sector)
times 510-($-$$) db 0 ; pad with zeroes
dw 0xaa55 ; magic number
times 256 dw 0xdada ; second sector
times 256 dw 0xface ; third sector
The disk_load.asm: disk_load.asm:
%include "bios_print.asm"
ERROR_MSG:
db "Disk read error!", 0
SECTOR_ERROR_MSG:
db "Wrong number of sectors read!", 0
disk_load:
push dx ; Store the number of sectors we wanted to read
mov ah, 0x02 ; BIOS read sector function
mov al, dh ; Number of sectors to be read
mov ch, 0x00 ; Cylinder 0
mov dh, 0x00 ; Head 0
mov cl, 0x02 ; Sector 2 (just after the boot sector)
int 0x13
jc disk_error ; if carry flag is set, report an error
pop dx
cmp dh, al
jne sector_error ; if we didn't read as many sectors as we wanted.
ret
disk_error:
mov bx, ERROR_MSG
call bios_print
jmp $
sector_error:
mov bx, SECTOR_ERROR_MSG
call bios_print
jmp $
And then bios_print.asm just has the functions to print the hex and so on: 然后bios_print.asm只具有打印十六进制的功能,依此类推:
bios_print:
pusha
mov si, bx
loop:
lodsb
or al, al
jz done
mov ah, 0x0e
int 0x10
jmp loop
done:
popa
ret
HEX_OUT:
db '0x0000', 0
hex_print:
pusha
mov cx, 4 ; Counter
char_loop:
dec cx
mov ax, dx
shr dx, 4
and ax, 0xf ; Mask to get the last part of ax
mov bx, HEX_OUT
add bx, 2 ; Skip the '0x' in HEX_OUT
add bx, cx
; If the character is a number, we just plop that out, but if it
; is a letter, we have to convert it to ASCII by adding 7 (because
; ASCII starts at 17, and the numbers end at 10)
cmp ax, 0xa
jl set_char
add byte [bx], 7
jl set_char
set_char:
add byte [bx], al ; Add the value of the byte to the char stored at bx
cmp cx, 0
je finished
jmp char_loop
finished:
mov bx, HEX_OUT
call bios_print
popa
ret
I have also tried running QEmu using the -fda flag, which was suggested in a few places, but no luck. 我也尝试使用-fda标志运行QEmu,这是在一些地方建议的,但没有运气。 Also, if I change the number of sectors to read to, say, 5 (as in the tutorial) I get a Drive read error (raised from the disk_load function).
另外,如果我将要读取的扇区数更改为5(如教程中所示),则会出现Drive read错误(从disk_load函数引发)。
This has got me very annoyed. 这让我非常生气。 Thanks in advance for any help.
在此先感谢您的帮助。
-_- I just worked out what the issue was. -_-我刚刚解决了问题所在。
After calling hex_print
, I was not resetting HEX_STRING
. 在调用
hex_print
,我没有重置HEX_STRING
。
I fixed this by simply copying the location of HEX_STRING
back out to BX, and then iterating through BX, indirectly setting each byte of HEX_STRING
to ASCII 0. 我通过简单地将
HEX_STRING
的位置复制回BX,然后通过BX迭代,间接地将HEX_STRING
每个字节设置为ASCII 0来解决这个问题。
This has fixed the problem. 这解决了这个问题。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.