Bug in the code or bug in the fasm?

Index > Main > Bug in the code or bug in the fasm?

Author

Thread

Everhest

Joined: 26 Jun 2008
Posts: 83
Location: Russia

Everhest 24 Nov 2008, 15:44

Hello, explain please that causes the mistake? debugger i not was able to find the reasons...

Code:

        proc name#.append title, text                              ; Íàäî ðàñæåâàòü
                add     [name#.LenTable], sizeof.BOOK32_TABLE_ITEM
                stdcall MemoryReSize, name#.table, [name#.LenTable]
                mov     [name#.table+4], eax
                invoke  lstrcpy, name#.item.title, [title]
                cld
                lea     esi, [name#.item]
                lea     edi, [name#.table+4]
                add     edi, [name#.LenTable]
                sub     edi, sizeof.BOOK32_TABLE_ITEM
                mov     ecx, sizeof.BOOK32_TABLE_ITEM
                rep     movsb
                inc     [name#.head.count]
                add     [name#.LenSect], sizeof.BOOK32_SECTION
                stdcall MemoryReSize, name#.sections, [name#.LenSect]
                mov     [name#.sections+4], eax                       ; error
                invoke  lstrlen,[text]
                inc     eax
                mov     [name#.sect.lengthof], eax
                stdcall MemoryInit, name#.sect.handle, [name#.sect.lengthof]
                mov     [name#.sect.buffer], eax
                cld
                lea     esi, [name#.sect]
                lea     edi, [name#.sections+4]
                add     edi, [name#.LenSect]
                sub     edi, sizeof.BOOK32_SECTION
                mov     ecx, sizeof.BOOK32_SECTION
                rep     movsb
                invoke  lstrcpy, [name#.sect.buffer], [text]  ; error not copy text
                invoke  MessageBox,0,[name#.sect.buffer],0,0
                mov     eax, [name#.sect.lengthof]
                add     [name#.LenSect], eax
                stdcall MemoryReSize, name#.sections, [name#.LenSect]
                mov     [name#.sections+4], eax                ; error
                mov     esi, [text]
                mov     ecx, [name#.sect.lengthof]
                rep     movsb
                ret
        endp

Description:	All code project in the attachment...	Download
Filename:	TEST BK API.rar
Filesize:	9.14 KB
Downloaded:	411 Time(s)

_________________
Forgive for my bad english, I from russia...

24 Nov 2008, 15:44

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20451
Location: In your JS exploiting you and your system

revolution 24 Nov 2008, 15:46

But you didn't explain what your problem is!

24 Nov 2008, 15:46

Everhest

Joined: 26 Jun 2008
Posts: 83
Location: Russia

Everhest 24 Nov 2008, 16:13

Sory, lines noted as error, produce terminated application...I do not know as this it is correct to describe in english, if you will compile application you themselves all will understand... No viruses...

24 Nov 2008, 16:13

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20451
Location: In your JS exploiting you and your system

revolution 24 Nov 2008, 16:21

Is it an assembly time error? Or a runtime error? What did you expect to happen at the lines you marked, and what did happen? What is the application supposed to do anyway?

You need to provide more information.

24 Nov 2008, 16:21

LocoDelAssembly
Your code has a bug

Joined: 06 May 2005
Posts: 4624
Location: Argentina

LocoDelAssembly 24 Nov 2008, 16:43

I have compiled it and crash on runtime due to access to an invalid pointer. I haven't checked anymore than that.

24 Nov 2008, 16:43

vid
Verbosity in development

Joined: 05 Sep 2003
Posts: 7105
Location: Slovakia

vid 24 Nov 2008, 16:53

Everhest: just by a quick glance, you don't seem to be checking return value of functions "MemoryInit", "MemoryResize", etc. Do you check error cases inside them? If not, you are asking for this kind of problem.

24 Nov 2008, 16:53

Everhest

Joined: 26 Jun 2008
Posts: 83
Location: Russia

Everhest 24 Nov 2008, 16:54

compile with fasm - no error.
run program - no error.
step book.open - no error
enter book.append - no error

Code:

                stdcall MemoryReSize, name#.sections, [name#.LenSect] 
                mov     [name#.sections+4], eax ; this error a run drwatson. why?
                invoke  lstrlen,[text]

if set comment at < mov [name#.sections+4], eax>,then drwatson start in the line <invoke lstrcpy, [name#.sect.buffer], [text]> why? text not are copy... if set comment on this line, then

Code:

                stdcall MemoryReSize, name#.sections, [name#.LenSect] 
                mov     [name#.sections+4], eax  ; hi drwatson 
                mov     esi, [text]

main asm file:

Code:

format PE GUI 4.0
entry start

include '../../include/win32a.inc'
include 'book.imp'

ERR_AUTO        = 0
soFromBeginning = 0

section '.data' data readable writable

        name                    db 'file.bk',0
        error_format            db 'Íåâåðíûé ôîðìàò ôàéëà',0
        NubersOfBytesRead       dd ?

        buffer                  rb 260

        class <TBOOK book>

section '.code' code readable executable

  start:
        stdcall book.open, name

        stdcall book.append, name, error_format ; any text and any title

    ;    stdcall book.save, name
        stdcall book.close
  exit:
        invoke  ExitProcess,0

include 'winapi.inc'

24 Nov 2008, 16:54

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20451
Location: In your JS exploiting you and your system

revolution 24 Nov 2008, 17:39

Let me suggest a better debugger than DrWatson, Ollydbg.

24 Nov 2008, 17:39

Everhest

Joined: 26 Jun 2008
Posts: 83
Location: Russia

Everhest 24 Nov 2008, 17:41

im see code from ollydbg and im not find error. why exhibit works without mistake if im add seh construction?

Code:

        proc name#.append title, text    
                add     [name#.LenTable], sizeof.BOOK32_TABLE_ITEM
                stdcall MemoryReSize, name#.table, [name#.LenTable]
                mov     [name#.table+4], eax
                invoke  lstrcpy, name#.item.title, [title]
                cld
                lea     esi, [name#.item]
                lea     edi, [name#.table+4]
                add     edi, [name#.LenTable]
                sub     edi, sizeof.BOOK32_TABLE_ITEM
                mov     ecx, sizeof.BOOK32_TABLE_ITEM
                rep     movsb

             .seh_install:
                push    .error1
                pushd   dword[fs:0]
                mov     [fs:0], esp
                inc     [name#.head.count]
             .error1:

Can here errors OS, rather then in code of application.

24 Nov 2008, 17:41

Everhest

Joined: 26 Jun 2008
Posts: 83
Location: Russia

Everhest 24 Nov 2008, 17:48

vid

Code:

error_memory    db 'Íå âûïîëíåí çàïðîñ íà ïîëó÷åíèå\ðàñïðåäåëåíèå ãëîáàëüíîé ïàìÿòè äëÿ ïðèëîæåíèÿ.',0

proc MemoryInit HGLOBAL, size
        invoke  GlobalAlloc, GMEM_MOVEABLE + GMEM_ZEROINIT, [size]
        mov     ecx, [HGLOBAL]; retrieve pointer parameter (ecx==hglb1)
        mov     [ecx], eax; save heap pointer
        cmp     eax, 0
        je      mem_error
        invoke  GlobalLock, eax   ; Âîçðàùàåò àäðåñ ïàìÿòè
        cmp     eax, 0
        je      mem_error
        mov     [HGLOBAL+4], eax
        ret
endp

proc MemoryReSize HGLOBAL, size
        mov     ecx, [HGLOBAL]; retrieve pointer parameter (ecx==hglb1)er
        invoke  GlobalReAlloc, [ecx], [size], GMEM_MOVEABLE + GMEM_ZEROINIT
        xor     ecx, ecx
        cmp     eax, ecx
        je      mem_error
        mov     [HGLOBAL], eax
        ret
endp

proc MemoryFree HGLOBAL
        invoke  GlobalUnlock,[HGLOBAL]
        invoke  GlobalFree,[HGLOBAL]
        ret
endp

     mem_error:
        invoke  MessageBox,0,error_memory,0,MB_ICONERROR
        invoke  ExitProcess,0
        ret

24 Nov 2008, 17:48

baldr

Joined: 19 Mar 2008
Posts: 1651

baldr 24 Nov 2008, 17:52

Everhest,

The same error as before:

Code:

proc MemoryInit HGLOBAL, size
        invoke  GlobalAlloc, GMEM_MOVEABLE + GMEM_ZEROINIT, [size]
        mov     ecx, [HGLOBAL]; retrieve pointer parameter (ecx==hglb1)
        mov     [ecx], eax; save heap pointer
        cmp     eax, 0
        je      mem_error
        invoke  GlobalLock, eax   ; Âîçðàùàåò àäðåñ ïàìÿòè
        cmp     eax, 0
        je      mem_error
        mov     [HGLOBAL+4], eax; <- !!! LOOK HERE !!!
        ret
endp

proc MemoryReSize HGLOBAL, size
        mov     ecx, [HGLOBAL]; retrieve pointer parameter (ecx==hglb1)er
        invoke  GlobalReAlloc, [ecx], [size], GMEM_MOVEABLE + GMEM_ZEROINIT
        xor     ecx, ecx
        cmp     eax, ecx
        je      mem_error
        mov     [HGLOBAL], eax; <- !!! LOOK HERE !!!
        ret
endp

Remember, procedure parameters are just values on stack, you have to dereference them if they are addresses. My answer to your GlobalReAlloc question was not the exact solution, just a clue (copy/pasted even with my comments Wink

). mov [name#.table+4], eax after invoke is a workaround for erroneous code inside MemoryInit/MemoryReSize. Revise your code, mov [param], something inside proc is not a method to return something. Feel free to PM me, I speak Russian (as you may remember).

24 Nov 2008, 17:52

Everhest

Joined: 26 Jun 2008
Posts: 83
Location: Russia

Everhest 24 Nov 2008, 18:08

baldr Äëÿ ëó÷øåãî ïîíèìàíèÿ ïîæàëóéñòà íàïèøèòå ýòî íà ðóññêîì, ÿ ñëèøêîì ïëîõî çíàþ àíãëèéñêèé.

24 Nov 2008, 18:08

< 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