Remembering Li Shaohua: Insights from a Linux Kernel Pioneer

This tribute honors the late Linux kernel developer Li Shaohua, highlighting his pivotal contributions to MD RAID, block I/O, NVMe, and storage performance, while sharing personal anecdotes and technical lessons that continue to inspire kernel engineers worldwide.

21CTO
21CTO
21CTO
Remembering Li Shaohua: Insights from a Linux Kernel Pioneer
On January 2, 2019, the Linux kernel mailing list announced the sad news that kernel developer Li Shaohua had passed away from cancer during the 2018 Christmas period.

Li Shaohua was one of the most outstanding Chinese contributors to the Linux kernel, maintaining the MD (software RAID) subsystem and making significant advances in block devices, I/O schedulers, NVMe, memory management, and power management.

He was known for his pure passion for system software, often discussing new ideas and code optimizations even in his final days.

One memorable collaboration involved improving MD RAID1 performance on NVMe SSDs. After extensive mailing‑list discussions, Li confirmed that the I/O path would not deadlock and accepted a patch that boosted read performance severalfold, while also identifying and fixing additional issues.

Another effort addressed the excessive time taken by MD RAID0 discard (TRIM) operations on large NVMe arrays. Li explained that the discard bio was being split into millions of small bios, causing severe slowdown. The solution was to recombine these split bios per component device, dramatically reducing the number of bios and speeding up XFS formatting from over 300 seconds to about 20 seconds.

raid0_handle_discard core code snippet
raid0_handle_discard core code snippet

Li later moved to work on the bcache subsystem, but his influence persisted; his code remains widely used in industry, including high‑performance I/O stacks for large‑scale data analysis in biotech.

Colleagues remember him as a brilliant engineer whose work continues to illuminate the kernel world, inspiring future developers to pursue excellence.

Source: 程序员那些事 CLSF[1]: China Linux Storage, Memory management & File system Workshop
Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

Performance Optimizationopen sourceLinux kernelNVMekernel-developmentMD RAID
21CTO
Written by

21CTO

21CTO (21CTO.com) offers developers community, training, and services, making it your go‑to learning and service platform.

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.