Issues

Empty-array argmax/min/max inside prange returns 0 instead of raising

numba/numba#10933 · opened

on Oct 21, 2026

1

Fix compile-time standard_indexing in inline stencils

#10913 opened

Oct 11, 2026

· Updated

Oct 24, 2026

5

Track aliases through build_tuple and exhaust_iter in parfor (#10907)

#10912 opened

Oct 11, 2026

· Updated

Oct 20, 2026

3

Fix parfor crash when inlined helper returns a view used by np.argmax

#10909 opened

Oct 9, 2026

· Updated

Oct 13, 2026

1

parfor + inline="always" + np.argmax on structured dtype view: "Illegal IR, del found" in native parfor lowering

numba/numba#10908 · opened

on Oct 9, 2026

1

Parallel changes made to arrays do not persist when aliased from a tuple unpack with len >= 4

numba/numba#10907 · opened

on Oct 8, 2026

1

`nb.prange(limit)` is significant slower than `nb.prange(1, limit)`

numba/numba#10641 · opened

on Jun 9, 2026

4

Cannot assign to NamedTuple array when @njit(parallel=True)

numba/numba#10338 · opened

on Nov 19, 2025

5

Poor performance for stencils in parallel

numba/numba#10182 · opened

on Aug 15, 2025

7

prange step size support

numba/numba#10083 · opened

on May 27, 2025

2

The process terminated with exit code -1073741819 (0xC0000005) during the second invocation.

numba/numba#9998 · opened

on Mar 18, 2025

4

Object mode jitted function does not interact with global variable correctly when parallelised

numba/numba#9976 · opened

on Mar 13, 2025

1

Unrecognized reduction pattern.

numba/numba#9949 · opened

on Feb 28, 2025

parallel code with prange returns a wrong results with two successive loops

numba/numba#9829 · opened

on Dec 6, 2024

6

Redefining numpy array in parallel mode leads to wrong shapes

numba/numba#9769 · opened

on Oct 24, 2024

Hangs when printing `get_thread_id` inside a cached jit function

numba/numba#9720 · opened

on Sep 6, 2024

1

NotImplementedError When creating an empty structured array in parallel

numba/numba#9696 · opened

on Aug 11, 2024

3

Fix aliasing issues with fusion.

#9537 opened

Apr 23, 2024

· Updated

Apr 24, 2024

2

Wrong list result with parallel=True despite no resizing and no cross-thread access of the list

numba/numba#9529 · opened

on Apr 14, 2024

3

Elapsed running time of parallel jit for the logistic regression documentation example is dozens of seconds

numba/numba#9395 · opened

on Jan 16, 2024

3

Regression in parallel code between 0.55 and 0.56

numba/numba#9333 · opened

on Dec 5, 2023

3

Python abort vs exception in `target="parallel"` with workqueue

numba/numba#9327 · opened

on Dec 2, 2023

[BUG] use np.newaxis in parallel mode cause the process terminated unexpectedly.

numba/numba#9317 · opened

on Nov 30, 2023

2

Parfors not exporting variable sometimes

numba/numba#9265 · opened

on Nov 2, 2023

2

print failed in parallel=True, cache=True

numba/numba#9233 · opened

on Oct 10, 2023

2

Navigation Menu

Repository navigation

Issues

Empty-array argmax/min/max inside prange returns 0 instead of raising

Fix compile-time standard_indexing in inline stencils

Track aliases through build_tuple and exhaust_iter in parfor (#10907)

Fix parfor crash when inlined helper returns a view used by np.argmax

parfor + inline="always" + np.argmax on structured dtype view: "Illegal IR, del found" in native parfor lowering

Parallel changes made to arrays do not persist when aliased from a tuple unpack with len >= 4

`nb.prange(limit)` is significant slower than `nb.prange(1, limit)`

Cannot assign to NamedTuple array when @njit(parallel=True)

Poor performance for stencils in parallel

prange step size support

The process terminated with exit code -1073741819 (0xC0000005) during the second invocation.

Object mode jitted function does not interact with global variable correctly when parallelised

Unrecognized reduction pattern.

parallel code with prange returns a wrong results with two successive loops

Redefining numpy array in parallel mode leads to wrong shapes

Hangs when printing `get_thread_id` inside a cached jit function

NotImplementedError When creating an empty structured array in parallel

Fix aliasing issues with fusion.

Wrong list result with parallel=True despite no resizing and no cross-thread access of the list

Elapsed running time of parallel jit for the logistic regression documentation example is dozens of seconds

Regression in parallel code between 0.55 and 0.56

Python abort vs exception in `target="parallel"` with workqueue

[BUG] use np.newaxis in parallel mode cause the process terminated unexpectedly.

Parfors not exporting variable sometimes

print failed in parallel=True, cache=True