flat assembler
Message board for the users of flat assembler.
![]() |
Author |
|
Tomasz Grysztar 06 Aug 2010, 15:34
Kicer wrote: small files are sucessfuly processed, big ones generate problem (listing is interrupted). Just a quick test that demonstrates that it's not the size that is the problem - get this piece of code: Code: db 'use32',10 rept 1000000 i { db 'label',`i,':',10 db 'mov eax,[ebx+label',`i,']',10 db 'jmp label',`i,10 } Quote:
|
|||
![]() |
|
Kicer 06 Aug 2010, 19:37
Code: [michal@Kicer tmp]$ cat breed.asm db 'use32',10 rept 1000000 i { db 'label',`i,':',10 db 'mov eax,[ebx+label',`i,']',10 db 'jmp label',`i,10 } [michal@Kicer tmp]$ fasm breed.asm test.asm -m 1000000 flat assembler version 1.69.14 (1000000 kilobytes memory) 1 passes, 11.0 seconds, 54666694 bytes. [michal@Kicer tmp]$ fasm test.asm -s test.fas -m 1000000 flat assembler version 1.69.14 (1000000 kilobytes memory) 1 passes, 9.0 seconds, 7999921 bytes. [michal@Kicer tmp]$ listing test.fas test.lst Naruszenie ochrony pamici [michal@Kicer tmp]$ for one who doesn't know: "Naruszenie ochrony pamięci" means "Segmentation Fault" ![]() i'm wondering if problem is somewhere outside fasm: I've not changed my asm file for a few weeks. It used to work, just today i've noticed it doesn't. maybe some changens in my system are the reason? I've updated fasm meanwhile, but as i said problem is also with version i've previously used. it may be also related to my another problem: http://board.flatassembler.net/topic.php?t=11756 I always recompile original fasm binaries, to be fully compatible with my libc. |
|||
![]() |
|
Tomasz Grysztar 06 Aug 2010, 19:41
It appears it is a solely problem of "listing" executable. How did you link it?
|
|||
![]() |
|
Kicer 06 Aug 2010, 20:31
just gcc -m32 -o listing listing.o
|
|||
![]() |
|
Tomasz Grysztar 06 Aug 2010, 20:49
Haven't you used some older sources for the listing by mistake? Please compare them with the source from the latest package.
|
|||
![]() |
|
Kicer 07 Aug 2010, 05:14
the problem was actually with fasm...
when i used my fasm (i used sources from "libc" directory), generated *.fsm was different than one generated by fasm executable from original package. now i use fasm from Linux directory and it's ok. edit: and also my problem with "8" exit code has disappeared |
|||
![]() |
|
Kicer 03 Sep 2010, 09:32
Problem returned in fasm .17 and .18
listing gives segmentation fault (with my code and "breed" code) |
|||
![]() |
|
Tomasz Grysztar 03 Sep 2010, 11:29
Oh, I see now - the new "ccall" macro (doing additional stack alignment to make the tools work on MacOS systems) was not preserving EBP, and I forgot that listing tool uses this register. I updated the packages with a corrected macro.
|
|||
![]() |
|
Kicer 04 Sep 2010, 20:06
seems to work now, thx
|
|||
![]() |
|
< Last Thread | Next Thread > |
Forum Rules:
|
Copyright © 1999-2023, Tomasz Grysztar. Also on GitHub, YouTube.
Website powered by rwasa.