Linux 3271 Published by

The latest release of Linux Kernel 6.17.4 has introduced several improvements to make it more robust and secure against potential crashes. The main updates include better management of null pointers, stricter checks for extensible ioctls, and improved error checking for file input/output and iomaps, which should help stop security problems and unexpected system crashes. Also, the writeback process has been improved, and several fixes have been made for specific problems with ACPI, irqchip, device property management, media drivers, and mm/ksm. The updated kernel is now available for download from the Linux kernel website, along with a summary of changes and patch information.



Linux Kernel 6.17.4 released

The latest release of Linux Kernel 6.17.4 has brought several improvements to various aspects of the kernel. One key change is related to handling null pointers in certain functions, making it more robust against potential crashes.

Kernel

The kernel may pass a null pointer to mnt_ns_release() when calling listmount(). To handle this situation, the kernel will now gracefully manage the case where the pointer is null. This ensures that the system does not crash unexpectedly due to a null pointer dereference.

Another significant improvement is related to validating extensible ioctls in various file systems. The kernel has been made more stringent in its validation of these ioctls, which should help prevent potential security vulnerabilities.

Additionally, there have been several changes aimed at improving the handling of errors and preventing crashes in specific scenarios. For instance, when dealing with file IO and iomaps, the kernel now checks if an iomap_begin implementation returns an IOMAP_INLINE buffer but fails to set the inline data pointer. If this occurs, the kernel will return an error instead of crashing.

In another significant improvement, the writeback mechanism has been optimized to avoid excessively long inode switching times when the lazytime mount option is enabled. This should help prevent workers from becoming pegged for hours and consuming 100% CPU.

Furthermore, several patches have addressed specific bugs and issues. For example, a patch fixes a soft lockup issue when switching many inodes by adding the possibility of rescheduling after processing each inode. Another patch adds a judgment to check if ns is null in pid_nr_ns, preventing a kernel panic from occurring when null is returned for task_active_pid_ns.

There have also been several patches addressing issues related to ACPI, irqchip, and device property handling. For instance, one patch fixes an issue where the ACPI handle of a data-only node may be NULL, preventing attempts to attach that node to an ACPI namespace object. Another patch adds code comments explaining what is going on in certain places in the ACPI device properties handling code.

In addition, several patches have addressed issues related to media drivers and mm/ksm. For example, one patch fixes a bug where the device's minor number should not be cleared after it is released. Another patch fixes an issue with incorrect accounting of KSM counters during a fork, ensuring that KSM statistics remain accurate and reflect the activity of each process correctly.

Linux kernel 6.17.4 released

Linux kernel version 6.17.4 is now available:

Full source: https://cdn.kernel.org/pub/linux/kernel/v6.x/linux-6.17.4.tar.xz
Patch: https://cdn.kernel.org/pub/linux/kernel/v6.x/patch-6.17.4.xz
PGP Signature: https://cdn.kernel.org/pub/linux/kernel/v6.x/linux-6.17.4.tar.sign

You can view the summary of the changes at the following URL:
https://git.kernel.org/stable/ds/v6.17.4/v6.17.3