ret2jazzy/disasm.pro

A realtime assembler/disassembler (formerly known as disasm.ninja)

disasm.pro's Issues

Website down

https://disasm.pro/ is down. Would it be easy to fix?

BNE bug?

disasm.pro has bugs in MIPS dasm:

  BE: 14 40 00 0E
  LE: 0E 00 40 14

For some reason it uses 2-operand bne

  bnez $v0, 0x3C

while it should be 3-way bne instruction:

 bne $v0, $r0, 0x3C

While @loadzero/mipsdis is fine.

Assembler output no longer shows bytes with hexadecimal letters correctly

After the recent update to the website, the assembler has issues when the output bytes contain any A-F characters and outputs 0 for those bytes. Additionally, bytes where the most significant half is 0 will not show this leading 0 (e.g. 06 becomes just 6.) The raw string display option is fully broken as well and shows only zeroes. Disassembly still works as expected.

I think the issue is in using map over a Uint8Array in assemble.js. This returns a similarly typed array, so all byte strings are being parsed back to numbers, with non-number strings becoming 0.

disasm.pro down

Hey there - was using disasm.pro (my favorite online disas service btw) - and while i was working with aarch64 assembling the site went down.

It's been down for about 24 hours now so i wanted to ping you.

cheers!

Website setup

It would be really great to have instructions on how to set this up on a web server. I was able to get disasm installed and python3 ninja.py launches ok, but can't figure out how to get the web page part of it setup. The index.py tries to load a lib that doesn't exist. Not sure how to get Socket.io to communicate with the server. This was really a useful tool and sad that it's gone. Thanks

ARM64 incorrect operation

E8 0F 1F 32 is disassembled to mov w8, #0x1e when it should orr w8, wzr, #0x1e

https://shell-storm.org/online/Online-Assembler-and-Disassembler/ https://disasm.czbix.com/
does work correctly when it uses the same tools

PowerPC assembly seems to be broken

Take this with a grain of salt as I know very little about anything (I'm just poking at some Wii cheat codes), but it seems like PPC assembly is broken. Disassembly works fine, but all attempts to assemble any PPC code just result in "Error: Invalid operand".

Example, I'm trying to assemble some branches like this:

0x800015e0: 4BFFF3F1 bl 0x800009d0

I can paste that machine code 4BFFF3F1 into the disassembly field on the right and the left will autofill bl 0x800009d0. Success?

But I can paste the very same bl 0x800009d0 that disasm.pro fed to me back into the (left) assembly field and I get "Error: Invalid operand". Not success.

Help?

Missing a few common CTF hurdles

First of all, awesome project!

It could be a great addition to add some of the common hurdles seen in CTF. For example, the ability to blacklist bytes (ex: nullbytes), and have a maximum shellcode length.

EDIT : We should be able to see the current shellcode length too.