Report function alignment accuracy #59
Conversation
reccmp/tools/asmcmp.py
Outdated
| @@ -322,15 +332,20 @@ def main(): | |||
|
|
|||
| implemented_funcs = function_count | |||
|
|
|||
| # Substitute an alternate value for the total number of functions in the file | |||
There was a problem hiding this comment.
I like that you added multiple explanations! How about Substitute an alternate value for the total number of functions in the target (e.g. to account for functions that have not been stubbed yet but have been counted by some analysis tool like Ghidra)? Or have I misunderstood the purpose of this argument?
There was a problem hiding this comment.
No, that's correct. I'll expand the comment.
There was a problem hiding this comment.
While we're here, should this really be function_count = max(function_count, int(args.total)) to prevent exceeding 100% accuracy?
reccmp/tools/asmcmp.py
Outdated
| if args.total: | ||
| function_count = int(args.total) | ||
|
|
||
| if function_count > 0: | ||
| effective_accuracy = total_effective_accuracy / function_count * 100 | ||
| actual_accuracy = total_accuracy / function_count * 100 | ||
| alignment_accuracy = functions_aligned_count / function_count * 100 |
There was a problem hiding this comment.
Maybe alignment_percentage? I think accuracy is fine for the other variables since the code can be matched to a fractional degree, but alignment is binary
reccmp-reccmpnow reports how many functions have the same virtual address in both the original and recompiled binary.For example:
Should we add this to SVG output also?