With the release of the Linux kernel version 6.1, Linux users will soon have a new approach to detect defective CPUs.
With the new capability, users will be able to identify whether a specific CPU or core is frequently creating issues by printing the socket and core that are most likely to be at fault when a segmentation fault occurs.
Even though this may not mean much to casual Linux users, system administrators that manage a large number of Linux servers stand to gain from it.
How does the function operate?
By adding "print()" to "show signal msg()," which prints the CPU, core, and socket at the moment of the segfault, you can activate the new feature.
To control this specific printk, navigate to /proc/sys/debug/exception-trace. For additional details, consult the official documentation(opens in new tab).
The task "may get rescheduled on another CPU between when the fault hit, and when the notice is written," admit the authors of the Kernal documentation, Rik van Riel and Borislav Petkov, who state that the functionality "is not flawless."
They did, however, add, "In practise, this has been good enough to enable users discover many faulty CPU cores."
You're not likely to contemplate employing this feature alone if you're a developer who already uses your Linux distribution for demanding, high-performance use cases. Additionally, you should use well-known CPU stress assessments like Prime95 or Aida64.
The update will be released as part of the Linux 6.1 merge window in October. Version 6.0 of the preceding Linux update has just been available for testing.
Version 6.0 included a number of improvements, 60% of which were driver upgrades, but also addressed GPU, networking, and sound, despite Linus Torvalds' assertion that "there's nothing fundamentally changing about" it.
Are you considering trying Linux? Visit Phoronix to see our list of the top Linux laptops.
