BUG: `config.change_flags` is not thread-safe: concurrent graph construction can permanently clear `config.cxx`
维护者通常 1 天内回复
还没有人认领这个 Issue。
评估
调研方向
Start by reading _ChangeFlagsDecorator in pytensor/configparser.py, then inspect the config.change_flags uses in pytensor/tensor/basic.py and the concurrent reproduction in the issue. Identify how overlapping or recursive scopes affect restoration, and add regression coverage for both cases. Done means concurrent graph construction leaves config.cxx and optdb__max_use_ratio at their original values.
由索引模型根据 Issue 内容生成。
描述
Describe the issue:
config.change_flags saves the current global values on entry and writes them back on exit (pytensor/configparser.py, _ChangeFlagsDecorator). When two threads' scopes over the same flag interleave, the scope that exits last restores the temporary value that the other scope had set:
A enters: saves "g++", sets cxx=""
B enters: saves "", sets cxx=""
A exits: restores "g++"
B exits: restores "" <- process-wide, permanently
Every later scope then saves and restores the corrupted value.
Graph construction enters such a scope. Static-shape inference (pytensor/tensor/basic.py, with config.change_flags(optdb__max_use_ratio=10, cxx="")) runs whenever a shape is symbolic, for example in pt.zeros((n,)). PyMC's resolve_shapes (pymc/pytensorf.py) does the same. Building graphs on several threads at once therefore leaves cxx == "" and optdb__max_use_ratio == 10 for the rest of the process.
A shared decorator instance has the same problem without threads. For example, @config.change_flags(warn_float64="ignore") in pytensor/tensor/basic.py is a single object, and a recursive or concurrent call overwrites the values it saved.
Reproducible code example:
from concurrent.futures import ThreadPoolExecutor
import pytensor
import pytensor.tensor as pt
before = pytensor.config.cxx, pytensor.config.optdb__max_use_ratio
def build(_):
n = pt.lscalar("n")
for _ in range(500):
pt.zeros((n + 1,)) # static-shape inference enters change_flags(cxx="")
with ThreadPoolExecutor(8) as executor:
list(executor.map(build, range(8)))
print(before, (pytensor.config.cxx, pytensor.config.optdb__max_use_ratio))
# ('/usr/bin/g++', 8.0) ('', 10.0)
xs = pt.dvectors(*(f"x{i}" for i in range(33)))
f = pytensor.function(xs, pt.exp(sum(xs)), mode="CVM")
f(*[[0.0]] * 33)
Error message:
NotImplementedError: Elemwise.perform (Python mode) does not support more than 32 operands. Use the C backend or disable Elemwise fusion.
PyTensor version information:
PyTensor 3.3.1 (pip), Python 3.14, Linux x86_64, with g++ as cxx. The same code is on main (332ea8fb7).
Context for the issue:
- Every C-linked function compiled afterwards in the process silently falls back to Python implementations. Already compiled functions are unaffected.
- Fused elemwise graphs with more than 32 operands then raise the error above.
- Python and C implementations differ at rounding level, so seeded results can change without an error, depending on thread timing earlier in the process. I measured differences of about 1e-14 in a downstream PyMC model.
I hit this in a library that fits independent models on worker threads with PyMC. A long-running process lost C compilation after one threaded fit, and an unrelated later computation failed with the error above.
- 主要语言
- Python
- 星标
- 654
- 派生
- 258
- 平均合并
- 2 天 7 小时
- 30 天内合并 PR
- 39
环境准备
- 没有 Dockerfile 或 Docker Compose 文件
- 有 Pull Request 模板
- 阅读贡献指南
从这里开始
- 先读完整个 Issue,再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库,在一个分支上完成修改。
- 提交 Pull Request,并在描述里引用这个 Issue 编号。
pymc-devs/pytensor 的其他 Issue
-
Docs of mode still mentions C/CVM as the default with FAST_RUN可能已有人在做 @ADITYASINGH1206 于 8 天前认领。 未关闭docs
难度 2/5 1-3 小时 新手友好度 62/100
pymc-devs/pytensor#2641 · 1 条评论 ·
维护者通常 1 天内回复
-
docs
难度 2/5 1-3 小时 新手友好度 62/100
维护者通常 1 天内回复
-
BUG: geomspace returns wrong sign for the endpoints when start and stop are negative可能已有人在做 @mayuriphad 于 27 天前认领。 未关闭
难度 2/5 1 小时以内 新手友好度 85/100
维护者通常 1 天内回复
-
Reciprocal.impl computes np.float32(1.0) / x: float32 precision for float64 scalars under NEP 50 (Python linker)可能重新可做 关联的 PR 已关闭且未合并。 未关闭
难度 2/5 1-3 小时 新手友好度 70/100
pymc-devs/pytensor#2430 · 1 条评论 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 78/100
pymc-devs/pytensor#2428 · 5 条评论 ·
维护者通常 1 天内回复
查看 pymc-devs/pytensor 的全部 Issue
相似的 Issue
-
难度 2/5 1-3 小时 新手友好度 62/100
维护者通常 1 天内回复
-
enhancement
难度 2/5 1-3 小时 新手友好度 68/100
dgunning/edgartools#2019 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 62/100
topoteretes/cognee#6818 · 1 条评论 ·
维护者通常 1 天内回复
-
source_verification_agent: no outcome when the publisher confirms a work but its DOI resolves to a different record可能已有人在做 @pinjun99 今天认领。 未关闭
难度 2/5 1-3 小时 新手友好度 68/100
Imbad0202/academic-research-skills#1159 · 1 条评论 ·
维护者通常 1 天内回复
-
[Bug]: DoubleMLData.from_arrays checks the treatment d with force_all_x_finite instead of force_all_d_finite可能已有人在做 @leninworld 今天认领。 未关闭
难度 1/5 1 小时以内 新手友好度 88/100
DoubleML/doubleml-for-py#471 ·
维护者通常 1 天内回复