This repository contains patches submitted to the Linux Kernel source tree that fix correctness issues related to Persistent Memory (PM) devices.

We have performed a thorough analysis on these bugs and publised our results in our paper "[A Study of Persistent Memory Bugs in the Linux Kernel] (https://dl.acm.org/doi/10.1145/3456727.3463783)" which was accepted and published at the 14th ACM International Systems and Storage Conference (SYSTOR), 2021, and in our journal "Understanding Persistent Memory Related Issues in the Linux Kernel" which was accepted at the ACM Transactions on Storage (TOS), 2023.

The repository consists of five directories, based on the bug pattern observed. Each directories has sub-directories that contain patches related to the bug pattern sub-types. The description of the five directories is as follows:

• Hardware Dependent: Improper PM device usage (e.g.: ambiguous specification, alignment)
• Semantic: Improper design or workflow (e.g.: wrong design, inconsistent state)
• Concurrency: Mismanagement of concurrent PM access (e.g.: race, deadlock)
• Memory: Mismanagement of PM/DRAM components (e.g.: null pointer, resource leak)
• Error Code: Improper handling of error codes (e.g.: error code not returned or checked)
• Dext: Contains an extended set of PM-related bug patches collected between Jan' 2021 - Dec' 2021. We perform similar analysis as the patches in our previous dataset, and published our results to ACM Transactions on Storage Journal 2022. Sections 4 and 5 in our journal submission provide additional details regarding this dataset.
• PM-Reproducibility: This folder contains workloads and scripts to reproduce a subset of bug patches. Additional details can be found in Section 6 of our journal submission.
• DrChecker+: Contains our ported version of Dr. Checker to perform static analysis of PM Kernel Modules. Section 7.3 in our journal provides additional details.

We hope that our dataset is helpful and could contribute to the development of effective PM bug detectors and building robust PM-based systems.

Please feel free to direct your concerns or feedback to: Duo Zhang (duozhang@iastate.edu) Om Rameshwar Gatla (ogatla@iastate.edu)