cython/cython

Generalise parallel loops - piter()

Open

#2,628 opened on Sep 25, 2018

View on GitHub
 (0 comments) (0 reactions) (0 assignees)Python (1,517 forks)batch import
Cython Language Featurefeaturehelp wanted

Repository metrics

Stars
 (8,663 stars)
PR merge metrics
 (Avg merge 45d 12h) (52 merged PRs in 30d)

Description

Cython currently restricts OpenMP loops to integer iteration using cython.parallel.prange().

A more general cython.parallel.piter() to wrap iterables would allow extending this to C++ iteration, C arrays, bytes/unicode buffers and direct memory view iteration. Except for C++ iteration (which the C++ compiler can handle), all of these cases are mapped to integer iteration internally and would thus be converted to a normal prange() loop by Cython.

Contributor guide

Research direction
Study how prange works internally and implement a wrapper that converts iterables to integer indices for OpenMP parallel loops.
Tech stack
python
Domain
compilersperformance
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
CythonOpenMPPython
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30