NVIDIA / NVIDIA/cudf

[FEA] Fix Parquet reader when using `device_read_async` while the destination device buffers are not ready

Open
#17,077 0 comments 0 reactions 0 assignees View on GitHub
cuIO feature request libcudf
Dominant language
C++
Stars
9.8k
Forks
1.1k
Avg merge
3d 6m
Merged PRs (30d)
278

Description

Similar to the issue with ORC reader with is fixed by https://github.com/rapidsai/cudf/pull/17074, Parquet reader also has similar problem with using `device_read_async`, when using `device_read_async` while the destination device buffers are not ready:https://github.com/rapidsai/cudf/blob/e41dea933f044183ccbfe26875a2b6c3ff383814/cpp/src/io/parquet/reader_impl_preprocess.cu#L253-L256

We need to fix it ASAP otherwise there is a risk of data corruption.

From the implementation of datasources, I see that this bug only affects reading data from files. In such situations, the host-2-device memcpy ops do not use user-provided stream. Copying data from other host/device buffers to device buffer is not affected.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.