flat assembler - FASMARM v1.44 - Cross assembler for ARM CPUs

Index > Non-x86 architectures > FASMARM v1.44 - Cross assembler for ARM CPUs

Goto page Previous 1, 2, 3 ... 11, 12, 13 ... 31, 32, 33 Next

Author

Thread

Madis731

Joined: 25 Sep 2003
Posts: 2139
Location: Estonia

Madis731 25 Jul 2007, 13:26

erm, QEMU ^o) ARM, MIPS, PPC, Sparc and... x86

25 Jul 2007, 13:26

0.1

Joined: 24 Jul 2007
Posts: 474
Location: India

0.1 25 Jul 2007, 13:29

^o) sounds interesting! I'll try it.
Could not find anything interesting in Google search please help!

25 Jul 2007, 13:29

0.1

Joined: 24 Jul 2007
Posts: 474
Location: India

0.1 25 Jul 2007, 13:32

erm search also gave no good results!

PS: Did you joke with me Shocked

25 Jul 2007, 13:32

0.1

Joined: 24 Jul 2007
Posts: 474
Location: India

0.1 25 Jul 2007, 13:36

When I searched for ARM tutorial, Google gave me lots of links.
But they were about arm twisting tricks, and a few about ARM were just not
good enough for me.

25 Jul 2007, 13:36

Dex4u

Joined: 08 Feb 2005
Posts: 1601
Location: web

Dex4u 25 Jul 2007, 15:02

Get a GBA emulator and start learning to code for the GBA and latter move on to the DS.
Heres a tut: http://www.patatersoft.info/gbaguy/gbaasm.htm

25 Jul 2007, 15:02

den_po

Joined: 17 Jul 2006
Posts: 23
Location: Russia, Cheboksary

den_po 15 Aug 2007, 12:58

is this a bug?

source:

Code:

code32
ldr r0,[ x ]

code16
ldr r0,[ x ]

align 4
x: DW 0x12345678

result:

Code:

                            CODE32
0C 00 90 E5                 LDR     R0, [R0,#0xC]
                            CODE16
C0 68                       LDR     R0, [R0,#0xC]

how else can i load data from a specified address? i don't like to use macros.
when i'm trying to use this command without brackets, fasmarm says:
error: Expecting square bracket "[".

_________________
JUST_DAn_PO

15 Aug 2007, 12:58

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20300
Location: In your JS exploiting you and your system

revolution 30 Aug 2007, 11:53

Hi den_po,

Sorry for the late reply, I have been away from the PC for a while.

The above code looks like a bug. If anything, the label x should be PC based and not R0 based. I don't know why you get the result above, I also get the same result. I will look into it shortly when I have sorted out my desk and the backlog of things to do. Once I have sorted out the problem I will post back here.

30 Aug 2007, 11:53

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20300
Location: In your JS exploiting you and your system

revolution 31 Aug 2007, 11:13

I have uploaded a new version v1.11 http://arm.flatassembler.net

Just a bug fix release.

FYI: The way to do what den_po mentions above is like this

Code:

code32
ldr r0,[r15,x-($+8)]
ldr r0,[y]

code16
ldr r0,[pc,x-($+4)]

virtual at r2+20
y dw 987654321
end virtual

align 4
x: DW 0x12345678

31 Aug 2007, 11:13

oyzzo

Joined: 25 Sep 2007
Posts: 3

oyzzo 25 Sep 2007, 21:38

how can i debug the ELF binaries created with fasmarm? i'm coding for the gp2x and gdb seems to not work because of the symbol table...

25 Sep 2007, 21:38

Dex4u

Joined: 08 Feb 2005
Posts: 1601
Location: web

Dex4u 26 Sep 2007, 02:52

I have written some program for the Gp2x with fasmarm. i may be able to help.

Last edited by Dex4u on 06 Oct 2007, 20:10; edited 1 time in total

26 Sep 2007, 02:52

oyzzo

Joined: 25 Sep 2007
Posts: 3

oyzzo 26 Sep 2007, 13:36

nope, it doesn't work. gdb can show me the bytecodes of the program but can't disassm them nor trace the program, so i can't debug it :S

26 Sep 2007, 13:36

oyzzo

Joined: 25 Sep 2007
Posts: 3

oyzzo 26 Sep 2007, 16:58

uhm, i can now debug with gdb. first i have to put a breakpoint in a place near to the entrypoint and then with display/i $pc i can do nexti. disas doesn't works but i think that's enough.

about the OS, i've read the forums about the arm port of Dex4u but there's not a lot of information. i don't have experience in OS develop but maybe i can help in something.

26 Sep 2007, 16:58

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20300
Location: In your JS exploiting you and your system

revolution 16 Oct 2007, 16:45

oyzzo, Can you please elaborate on the problem you have with the symbols.

I have been successfully using symbolic debugging using the tools from ARM without any issue, but those tools don't include gdb so I haven't used it.

Is it the symbol table format that is causing a problem? Or some other reason?

Feel free to post some example code that illustrates the problem Wink

16 Oct 2007, 16:45

IronM

Joined: 01 Nov 2007
Posts: 3
Location: Russia

IronM 01 Nov 2007, 05:34

revolution,
Hi!
can you include some instructions in your prog:
1. adr rx, label1

2. Ability to use label in ldr instructions, e.g.

Code:

ldr r4, my_data
......
align 4
my_data: dw 0x12345678

Now you need to use macros.....

3. Unicode support (it's necessary for correct representation russian symbols), e.g.

Code:

ldr r2, my_string
.....
align 4
my_string: db *'Hello, World!',0

instead of

Code:

ldr r2, my_string
........
align 4
my_string: db 'H',0,'e',0,'l',0,'l',0,'o',0,',',0,' ',0,'W',0,'o',0,'r',0,'l',0,'d',0,'!',0,0,0

01 Nov 2007, 05:34

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20300
Location: In your JS exploiting you and your system

revolution 01 Nov 2007, 17:43

Hi IronM,

IronM wrote:

1. adr rx, label1

You can use the macros I posted earlier in this thread.

Code:

macro adr reg,location {
        add reg,pc,location-$-8
}

macro adrl reg,location {
        sub reg,pc,(-location+$+8) and (0ffh shl 2)
        sub reg,reg,(-location+$+4) and (0ffffff00h shl 2)
}

IronM wrote:

2. Ability to use label in ldr instructions ... Now you need to use macros.....

You don't need macros, but you do need some PC relative tricks to do it.

Code:

code32
ldr r0,[r15,x-($+8)]
ldr r0,[y]

code16
ldr r0,[pc,x-($+4)]

virtual at r2+20
y dw 987654321
end virtual

align 4
x: DW 0x12345678

However I may look into some sort of automatic PC relative addressing for a later version.

IronM wrote:

3. Unicode support

Already there. Use du.

Code:

my_string: du 'Hello, World!',0

Last edited by revolution on 22 Feb 2010, 12:59; edited 1 time in total

01 Nov 2007, 17:43

vid
Verbosity in development

Joined: 05 Sep 2003
Posts: 7105
Location: Slovakia

vid 01 Nov 2007, 17:45

Quote:

However I may look into some sort of automatic PC relative addressing for a later version.

I think that would be great addition, similar to automatic RIP-relative addressing in 64bit FASM

01 Nov 2007, 17:45

IronM

Joined: 01 Nov 2007
Posts: 3
Location: Russia

IronM 01 Nov 2007, 18:31

1. I know about this macros, but i didn't like to use such simple instruction as macros.... Any disassembler shows it as adr, not like add pc, pc,0x??

2. It will be very handy

3.It doesn't support russian symbols

And what about case instruction? e.g.

Code:

cmp r0, 0x4
bcs exit
adr r1, case
ldrb r1, [r1,r0]
add pc, r1

case:
........

01 Nov 2007, 18:31

revolution
When all else fails, read the source

Joined: 24 Aug 2004
Posts: 20300
Location: In your JS exploiting you and your system

revolution 01 Nov 2007, 19:04

vid wrote:

I think that would be great addition, similar to automatic RIP-relative addressing in 64bit FASM

Okay, that makes two votes so far, I may put it down for inclusion in v1.12

IronM wrote:

1. I know about this macros, but i didn't like to use such simple instruction as macros.... Any disassembler shows it as adr, not like add pc, pc,0x??

Fasmarm currently supports just the instructions that the CPU's execute. This is why adr is a macro. This follows the original tradition of fasm (the x86 version). But, if you can make a strong case to include it as a pseudo instruction then I am willing to listen.

IronM wrote:

3.It doesn't support russian symbols

This is a consequence of fasm only supporting ascii encoded source files. There are also some macros from the current fasm package that does UTF encding (UTF8.INC), perhaps that is what you are looking for?

IronM wrote:

And what about case instruction? e.g.

Code:

cmp r0, 0x4
bcs exit
adr r1, case
ldrb r1, [r1,r0]
add pc, r1

case:
........

I'm not sure what you are asking here, "case" is not an instruction. Maybe you want a "case" macro? There are macros in the fasm package that do things like case. But you probably need some minor tweaks to convert them to fasmarm syntax compatibility.

01 Nov 2007, 19:04

IronM

Joined: 01 Nov 2007
Posts: 3
Location: Russia

IronM 02 Nov 2007, 04:26

OK. About russian symbols.... it is bearable:) Anyway i can use hex-values of russian symbols.
About case.... of course macro, but i think it'll difficult because of different variants of cases....

02 Nov 2007, 04:26

vid
Verbosity in development

Joined: 05 Sep 2003
Posts: 7105
Location: Slovakia

vid 02 Nov 2007, 07:00

IronM: russian symbols ARE supported, you just have to understand what you are doing. First, you must know how your editor encodes files. In most cases (FASMW for example) it is 8bit per character, but some editors allow UTF8.

To find difference, just look at saved file: if special characters have 2 bytes, then it is UTF8.

Next thing is to learn what character set are you using. Good chance is that it's CP1251. Include appropriate file for your encoding from INCLUDE\ENCODING\. Then, your "du" strings will support special characters.

02 Nov 2007, 07:00

Goto page Previous 1, 2, 3 ... 11, 12, 13 ... 31, 32, 33 Next

< 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