Benchmarking Bcachefs 1.38.6: The First Release No Longer "Experimental"

Written by Michael Larabel in Software on 22 June 2026 at 11:30 AM EDT. Page 3 of 4. 105 Comments.
Flexible IO Tester benchmark with settings of Type: Random Read, Engine: IO_uring, Direct: No, Block Size: 4KB, Job Count: 1, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Read, Engine: IO_uring, Direct: No, Block Size: 4KB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Read, Engine: IO_uring, Direct: No, Block Size: 4KB, Job Count: 32, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Read, Engine: IO_uring, Direct: Yes, Block Size: 4KB, Job Count: 1, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Read, Engine: IO_uring, Direct: Yes, Block Size: 4KB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Read, Engine: IO_uring, Direct: Yes, Block Size: 4KB, Job Count: 32, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.

But for FIO 4K random read performance, the outcome was mixed with some of the configurations showing worse performance than Bcachefs 1.38.5.

Flexible IO Tester benchmark with settings of Type: Random Write, Engine: IO_uring, Direct: No, Block Size: 4KB, Job Count: 1, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Write, Engine: IO_uring, Direct: No, Block Size: 4KB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Write, Engine: IO_uring, Direct: No, Block Size: 4KB, Job Count: 32, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Write, Engine: IO_uring, Direct: Yes, Block Size: 4KB, Job Count: 1, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Random Write, Engine: IO_uring, Direct: Yes, Block Size: 4KB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.

Similarly, some of the FIO random write tests also regressed on Bcachefs 1.38.6.

Flexible IO Tester benchmark with settings of Type: Sequential Read, Engine: IO_uring, Direct: No, Block Size: 2MB, Job Count: 1, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.
Flexible IO Tester benchmark with settings of Type: Sequential Read, Engine: IO_uring, Direct: No, Block Size: 2MB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.
Flexible IO Tester benchmark with settings of Type: Sequential Read, Engine: IO_uring, Direct: No, Block Size: 2MB, Job Count: 32, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.
Flexible IO Tester benchmark with settings of Type: Sequential Read, Engine: IO_uring, Direct: Yes, Block Size: 2MB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.

The FIO sequential read performance at least was showing consistent and very significant read improvements using the new Bcachefs 1.38.6.

Flexible IO Tester benchmark with settings of Type: Sequential Write, Engine: IO_uring, Direct: No, Block Size: 2MB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.
Flexible IO Tester benchmark with settings of Type: Sequential Write, Engine: IO_uring, Direct: No, Block Size: 2MB, Job Count: 32, Disk Target: Default Test Directory. Bcachefs 1.38.6 was the fastest.
Flexible IO Tester benchmark with settings of Type: Sequential Write, Engine: IO_uring, Direct: Yes, Block Size: 2MB, Job Count: 16, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.
Flexible IO Tester benchmark with settings of Type: Sequential Write, Engine: IO_uring, Direct: Yes, Block Size: 2MB, Job Count: 32, Disk Target: Default Test Directory. Bcachefs 1.38.5 was the fastest.

With the sequential write performance there were also some improvements using the new Bcachefs 1.38.6 but also some regressions.

Related Articles
Approaching A 10 Second Linux Kernel Build
Benchmarking Java Performance Of JDK 8 Through OpenJDK 27
AMD EPYC Server Performance Up ~8% Geo Mean For 2026 With Linux Kernel Improvements
Linux 7.3 Delivering Some Performance Gains On Intel Panther Lake / Framework Laptop 13 Pro
Early Benchmarks Of AMD EPYC On Linux 7.3 Show Some Performance Gains On The Horizon
Linux 7.3 Features Many Exciting Improvements, New Hardware Support & Faster Btrfs