X Tutup
Skip to content

GH-102300: Reuse objects with refcount == 1 in float specialized binary ops.#102301

Merged
markshannon merged 5 commits intopython:mainfrom
faster-cpython:check-refcnt-in-binary-op-experimental
Mar 13, 2023
Merged

GH-102300: Reuse objects with refcount == 1 in float specialized binary ops.#102301
markshannon merged 5 commits intopython:mainfrom
faster-cpython:check-refcnt-in-binary-op-experimental

Conversation

@markshannon
Copy link
Member

@markshannon markshannon commented Feb 27, 2023 •
edited
Loading

About 0.5% faster:

https://github.com/faster-cpython/benchmarking/tree/main/results/bm-20230227-3.12.0a5%2B-c29d369

The results are quite noisy.
But the results seem good, as the most float heavy benchmark, nbody, does show a 7% speedup.

@bedevere-bot bedevere-bot mentioned this pull request Feb 27, 2023
Closed
@Fidget-Spinner
Copy link
Member

Fidget-Spinner commented Feb 28, 2023

I can't see the results. I think external parties can't view that repo.

@brandtbucher
Copy link
Member

brandtbucher commented Feb 28, 2023

I can't see the results. I think external parties can't view that repo.

Sorry, you can change the repo name from benchmarking to benchmarking-public in the URL:

https://github.com/faster-cpython/benchmarking-public/tree/main/results/bm-20230227-3.12.0a5%2B-c29d369

brandtbucher
brandtbucher approved these changes Mar 2, 2023
View reviewed changes
Copy link
Member

@brandtbucher brandtbucher left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good (like the clever use of _Py_DECREF_NO_DEALLOC). Do we want a helper function or macro to reduce duplication?

In the original approach (#30594), I only considered the LHS, and added a special case for in-place ops followed by some variant of LOAD_FAST (where the refcount is 2). Not sure if we want to consider adding something similar later.

@markshannon
Copy link
Member Author

markshannon commented Mar 6, 2023

An inline function is tricky without introducing extra branches, so I've added a macro.

The refcount == 2 variant probably wants support from the bytecode compiler to mark candidates operations in x = x + (complicated expression).

@markshannon markshannon merged commit 233e32f into python:main Mar 13, 2023
warsaw pushed a commit to warsaw/cpython that referenced this pull request Apr 11, 2023
@markshannon @warsaw
pythonGH-102300: Reuse objects with refcount == 1 in float specialize…
a6955eb 
…d binary ops. (pythonGH-102301)
@markshannon markshannon deleted the check-refcnt-in-binary-op-experimental branch September 26, 2023 12:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@brandtbucher brandtbucher brandtbucher approved these changes

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@markshannon @Fidget-Spinner @brandtbucher @bedevere-bot
X Tutup