zlacker

[parent] [thread] 7 comments
1. Xmd5a+(OP)[view] [source] 2025-12-06 15:26:24
This deserves a discussion
replies(2): >>ronsor+L1 >>stevem+z9
◧
2. ronsor+L1[view] [source] 2025-12-06 15:37:05
>>Xmd5a+(OP)
I've used LLMs to help with decompilation since the original release of GPT-4. They're excellent at recognizing the purpose of functions and refactoring IDA or Ghidra pseudo-C into readable code.
replies(2): >>galang+l2 >>eurode+J3
◧◩
3. galang+l2[view] [source] [discussion] 2025-12-06 15:42:19
>>ronsor+L1
How does it do on things that were originally written in assembly?
replies(1): >>saagar+i4
◧◩
4. eurode+J3[view] [source] [discussion] 2025-12-06 15:54:43
>>ronsor+L1
Someone please try this on an original (early 1980s) IBM-PC BIOS.
replies(2): >>mh-+lx >>tadfis+0M
◧◩◪
5. saagar+i4[view] [source] [discussion] 2025-12-06 15:58:31
>>galang+l2
This is typically easier because the code was written for humans already.
◧
6. stevem+z9[view] [source] 2025-12-06 16:39:31
>>Xmd5a+(OP)
We're very far away from this.
◧◩◪
7. mh-+lx[view] [source] [discussion] 2025-12-06 19:50:21
>>eurode+J3
Got a bin?
◧◩◪
8. tadfis+0M[view] [source] [discussion] 2025-12-06 22:08:46
>>eurode+J3
I don't believe that was written in a compiled language, so any old 8086 disassembler should suffice. I would love to see what comments an LLM adds to the assembly code, though.
[go to top]