Chapter 2. Kernel


System Call Tracepoints

The following tracepoints for system call events have been added:

  • sys_enter
  • sys_exit
The system call enter and exit tracepoints are only supported on architectures that have the HAVE_SYSCALL_TRACEPOINTS configuration option enabled.
IPv6 UDP Hardware Checksum

Red Hat Enterprise Linux 5.9 adds hardware checksum support for UDP running over IPv6.

Per-process Resource Limits

The prlimit64() system call has been added to allow users to dynamically change the limits of a running process via the /proc/<PID>/limits file (this file is now writable).

VLAN Support Added to pktgen

VLAN support has been added to the pktgen module. The pktgen module is now able to produce 802.1Q tagged frames.

Restricting Access to /proc/<PID>/

The hidepid= and gid= mount options have been added to procfs to allow restricting of access to /proc/<PID>/ directories.

DSCP Field Mangling

In Red Hat Enterprise Linux 5.9, the netfilter module now supports mangling of the DSCP field.

Comparison of Booted System and Dumped System

This feature allows you to compare a booted system with a dumped system to efficiently analyze changes that might be introduced by image migration. To identify a guest, stsi and stfle data is used. A new function, lgr_info_log() compares the current data (lgr_info_cur) with the last recorded one (lgr_info_last).

FICON HyperPAV Enablement

The Parallel access volume (PAV) storage server feature is intended to achieve better I/O performance through alias devices. With HyperPAV, you no longer need to configure dedicated aliases for each device for which you want to use PAV with. HyperPAV supports an alias pool from which alias devices are automatically assigned when required. This results in reduced configuration effort and dynamic self-optimization of the PAV configuration.

VDSO Support

Applications often use system calls that are related to system time. The virtual dynamic shared object (vdso) support is a kernel-provided shared library that holds the required data for the correct system time. The calculation is done in user space, and accelerates the operations. This results in increased performance, especially for applications with high amounts of time-related system calls.

Red Hat logoGithubRedditYoutubeTwitter

Learn

Try, buy, & sell

Communities

About Red Hat Documentation

We help Red Hat users innovate and achieve their goals with our products and services with content they can trust.

Making open source more inclusive

Red Hat is committed to replacing problematic language in our code, documentation, and web properties. For more details, see the Red Hat Blog.

About Red Hat

We deliver hardened solutions that make it easier for enterprises to work across platforms and environments, from the core datacenter to the network edge.

© 2024 Red Hat, Inc.