-
Notifications
You must be signed in to change notification settings - Fork 446
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Line numbers wrong with Python 3.8 #190
Comments
Line number increments in the line number table can be negative occasionally, and since we were loading as a table of u8 this wasn't being handled appropiately. This might be causing issues like #190 Change to match the logic in https://github.com/python/cpython/blob/143a97f6/Objects/lnotab_notes.txt#L48-L49
I haven't managed to replicate this - but after looking into that code I realized there is a bug there that could be causing this issue =( I have a fix in #196 that I hope will fix this issue. |
I have the same issue. |
Yup, I can confirm that with the latest release v0.3.1 this issue still exists. @benfred Is there some information from my Python environment that I can give you that might help you in narrowing this down? |
#208 fixes this issue - will be the in the next release |
Fix is in v0.3.2 |
Yup, can confirm that line numbers work perfectly again. Thanks! |
After upgrading my local project to Python 3.8, the line numbers that py-spy reports are about 3-8 times too high (difference varies). This affects both the flamegraph and the speedscope output.
Environment:
Python 3.8.0
Windws 10 Pro, Version 10.0.18362 Build 18362
py-spy==0.3.0 (but I believe it affects earlier versions, too)
Note: This situation occurs when the program that is analyzed is running on Python 3.8 and it's independent of the Python environment that I start py-spy from. To clarify:
The text was updated successfully, but these errors were encountered: