-
-
Notifications
You must be signed in to change notification settings - Fork 31.5k
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
bpo-45923: Add RESUME_QUICK
#31244
bpo-45923: Add RESUME_QUICK
#31244
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The oparg < 2
check can be done when quickening.
TARGET(RESUME_QUICK) { | ||
PREDICTED(RESUME_QUICK); | ||
assert(tstate->cframe == &cframe); | ||
assert(frame == cframe.current_frame); | ||
frame->f_state = FRAME_EXECUTING; | ||
if (_Py_atomic_load_relaxed(eval_breaker) && oparg < 2) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The specializer should handle the oparg < 2
.
When you're done making the requested changes, leave the comment: |
I note that #31240 also shows a ~1% speedup. |
I'm pretty sure they're independent, since even though the All four combinations of PR branches and
I know the UK benchmarking machine is a bit more stable than the US one, though, so if you want to verify those numbers I won't be offended. 🙃 |
I think your analysis is sound, and they look like good changes anyway, even if there were some correlation. |
OK, merging this. We can quicken |
Also, clean up
JUMP_ABSOLUTE
/JUMP_ABSOLUTE_QUICK
to reduce code duplication.1% faster:
https://bugs.python.org/issue45923