The development of Linux 7.0 It has entered a decisive and, at the same time, more turbulent phase than is usually the case for the kernel. The third release candidate, Linux 7.0-rc3It comes with such a large volume of changes that even its creator, Linus Torvalds, has acknowledged some concern about the size of the patch at such an advanced stage of the cycle.
Although this branch aims to become the basis for some of the major Linux distributions While Ubuntu and Fedora are expected to be released in the coming months, the reality is that it remains a beta version, with a lot of development activity, a large amount of new code, and an integration pace that is forcing those responsible for the project to exercise extreme caution.
A larger and more turbulent development cycle than expected
At this stage of kernel design, it is normal for RCs to focus on Minor corrections and detail polishingwith relatively contained patches. However, both rc2 and rc3 are breaking that trend, presenting a number of commits clearly higher than in other recent cycles.
Torvalds has admitted that the magnitude of Linux 7.0-rc3 It's unusual, but at the same time, he insists that, after reviewing the changes, he sees nothing particularly alarming. Much of the growth is explained by code cleanups, the addition of automated tests (self-tests), and internal adjustments that, in theory, shouldn't compromise stability.
The concern stems more from the quantity than the severity: Too much new code in just a few weeks The stable release increases the risk of difficult-to-detect regressions slipping through. This is especially noticeable because many distributions plan to rely on this kernel almost immediately.
The situation is reminiscent of what happened in other major releases, such as series 6.19 and Linux 6.18where the backlog of work ended up being concentrated in just a few weeks of the development schedule, forcing an extra effort in review and testing.
Key changes in Linux 7.0-rc3: performance, memory, and networking
Among the most relevant modifications in this third Release Candidate are those relating to memory and network performance, two critical areas in servers, data centers and environments with intensive and real-time workloads (Preempt RT).
On the one hand, a serious regression in the SLAB systemThe memory allocator responsible for managing objects within the kernel. This regression was affecting performance in certain scenarios, so fixing it was a priority to avoid surprises in the final version.
In the networking section, the kernel incorporates a small but interesting optimization in the handling of input and output events. The Google developer Eric Dumazet has adjusted the function epoll_put_uevent() to take advantage of the technique of scoped user access, originally introduced in Linux 6.19.
This modification replaces several calls and instructions stac/clac linked to access control between user space and the kernel, reducing its impact on certain CPUs. In network stress tests focused on packets per second, the improvement is around 1,5% additional performance on AMD Zen 2 processors, a modest but significant figure when millions of packets are handled every second.
The epoll adjustment and the SLAB correction are good examples of how seemingly small changes These benefits can be translated into measurable advantages in European servers and data centers where Linux is the foundation of the infrastructure.
Security, virtualization, and servers: focus on AMD and Intel
Linux 7.0 also reinforces its commitment to the security and isolation in virtualized environments, a particularly relevant area for cloud providers and hosting companies in Europe.
One of the most notable new features is the inclusion of support for IBPB-On-Entry in virtual machines with AMD SEV-SNPThis feature, designed primarily for servers with next-generation AMD EPYC processors, helps to better protect execution contexts between guest and host, limiting potential attack vectors based on speculation.
On the Intel side, the kernel incorporates adjustments in the detection of Sub-NUMA Clustering (SNC) topologies In some recent models, correcting identification problems that could affect the optimal allocation of memory and the localization of processes in multiprocessor systems.
Under the hood, Linux 7.0 also integrates optimizations related to Intel TSX on chips that support it, with the goal of recovering some of the performance lost after security patches in recent years. Although these types of changes are carefully validated to avoid reopening vulnerabilities, they can provide a significant boost for concurrency-intensive applications.
Taken together, these improvements position kernel 7.0 as a particularly attractive option for server and cloud environmentswhich are precisely the ones that benefit most from advanced security features and new CPU architectures.
More supported hardware and improved compatibility
As with virtually every kernel version, one of the pillars of Linux 7.0-rc3 is the expansion of catalog of supported devices and the polishing of existing controllers.
The changelog includes new identifiers and settings for equipment from manufacturers such as ASUS, Dell, HP, Lenovo and OneXPlayer, and Raspberry PiThis is especially important so that future distributions can easily recognize laptops and desktops that will arrive on the European market in the coming months.
Among the interesting details, a bug that affected the Apple Magic Trackpad 2When connected via USB, the kernel did not correctly report the battery level. With the new code, this peripheral should integrate more seamlessly into Linux desktops.
In addition, the 7.0 branch continues to incorporate preliminary support and compatibility improvements for CPU architectures that have not yet reached the market, such as Intel Nova Lake, Diamond Rapids or AMD Zen 6This anticipation means that, when the hardware goes on sale in Europe, many distributions will already be ready to take advantage of it without the need for additional patches.
Ultimately, much of the patch's appeal lies in its desire to offer a a kernel better prepared for future hardware, while also correcting minor details in devices already present in offices and homes.
File systems and storage: testing with Linux 7.0
The performance of the File System This is another area where the impact of Linux 7.0 is being measured. Taking advantage of the latest optimizations, the following has been carried out comparative performance tests with some of the most widely used storage technologies in servers and workstations.
The benchmarks were performed using the Latest Linux 7.0 code under developmentThis study compares four well-known file systems within the free ecosystem: Btrfs, EXT4, F2FS, and XFS. Each was tested with its default configuration, aiming to reflect what any administrator would encounter when installing a distribution without touching the advanced settings.
In the case of BtrfsAdditional tests have also been performed by disabling the function Copy-on-Write (COW)This allows us to observe how this feature affects speed in scenarios with many writes.
The test environment has been based on a server with AMD EPYC 9745 processor and storage NVMe PCIe 5.0Specifically, a Solidigm D7-PS1010 drive (model SB5PH27X038T). Before each benchmark, the file systems were formatted from scratch to ensure comparable results.
With this type of high-end hardware, very close to what is beginning to be deployed in European data centers, the tests provide a useful reference on how each file system behaves when combined with Linux 7.0 and next-generation NVMe SSD.
Competition between XFS, EXT4, Btrfs and F2FS
The four systems tested represent quite distinct approaches within the Linux world, and the results show a competitive and constantly evolving landscape.
EXT4 It maintains its reputation as a balanced, stable, and predictable option. It's the file system that many distributions still use by default, and tests confirm its good overall performance and maturity in mixed read and write workloads.
Btrfs It positions itself as a modern alternative with advanced features such as snapshots, transparent compression, and flexible storage management. Tests show that these features come at a cost in certain scenarios, although disabling COW can improve performance in write-intensive tasks at the expense of some advantages.
F2FSDesigned with flash memory and SSDs in mind, it offers interesting performance in operations geared towards solid-state devices, but it remains a niche system that requires careful consideration of the type of workload before adopting it in production.
In this round of benchmarks, the one that has performed best in various scenarios has been XFSespecially in high-performance environments with large volumes of data. This is no surprise: it has been optimized for servers and data centers for years, and with Linux 7.0 it further solidifies its position as a A very solid option for business infrastructures.
File systems still pending testing with Linux 7.0
The initial tests also considered including other large file systems, such as OpenZFS o Bcachefs (in its unstable branch). However, its current versions are not yet They are fully compatible with the Linux 7.0 state in Git, which has forced them to postpone their analysis.
The expectation is that, as integrations progress and support for kernel 7.0 stabilizes, these systems will be eligible for future benchmarks. Their inclusion will be relevant for system administrators in Europe who already use OpenZFS in storage arrays or are considering Bcachefs as a modern alternative.
Until that compatibility is confirmed, the current picture of storage performance with Linux 7.0 focuses on the most established options, where XFS and EXT4 They continue to lead the way in many productive deployments.
Rust in the kernel and cleaning up obsolete technologies
Beyond the visible improvements in performance or hardware support, Linux 7.0 marks a further step in a fundamental transformation: the Gradual introduction of Rust into the kernel and the removal of components considered obsolete.
The first steps toward Rust code acceptance in the kernel began about three years ago, and since then, this language has become a key tool in the kernel developers' arsenal. Its main objective is to facilitate the writing of safer components by reducing memory errors that are easy to make in C.
In parallel, Torvalds has decided to move forward with eliminating legacy technologies that no longer make sense in current systems. A symbolic example is the retirement of the protocol HIPPI, a standard from the 90s that has little relevance in modern computers and only added complexity and potential problem vectors.
These combined moves — introducing Rust and clean old code— they point to a somewhat lighter kernel and, over time, potentially more secure. Although the immediate impact may not be obvious to the end user, it does mark a clear direction for the project's evolution.
The fact that Linux 7.0 is a particularly large version is also explained by this desire to reorganize internal parts of the kernel, laying the foundations for future developments that will reach European distributions over the next few years.
Calendar, distributions and risks of installing Linux 7.0-rc3
The current work plan places the Linux 7.0 stable release around mid-Aprilprovided that the next Release Candidates manage to reduce the volume of changes and no last-minute errors appear.
This schedule is especially critical because several major distributors already have release dates set. Canonical, for example, is preparing Ubuntu LTS 26.04, an extended support version with a strong presence in European companies and public administrations, and is planned to be based on kernel 7.0.
Something similar happens with Fedora 44which also aims to include this branch as the default kernel. In both cases, the distribution teams need the code to be sufficiently mature to guarantee a Stable support from day one.
For now, the message from the kernel project is clear: Linux 7.0-rc3 and the rest of the RC are intended for developers, testers, and advanced users who want to help polish the final version. Installing these builds on a primary workstation, or on production servers, is considered risky.
The general recommendation is to limit testing to virtual machines or equipment dedicated exclusively to testingwhere a critical failure does not cause a serious disruption to daily life.
How to try Linux 7.0-rc3 on your system today
Anyone wanting to experiment with Linux 7.0-rc3 has several options, including tools such as Live Update Orchestrator, although all of them with the warning that they are experimental software and therefore potentially unstable.
One of the easiest ways for users of Debian or Ubuntu-based distributions is to use the tool Mainline, an open-source program that allows you to download and install recent kernels without waiting for them to arrive in the official repositories.
On an Ubuntu system, simply add the corresponding repository and let the package manager handle the rest. The typical process involves running a command that adds the tool's PPA, updates the package list, and installs the application. Then, when you open Mainline, a list of available kernel versions is displayed, from which you can select the one you need. Linux 7.0-rc3 for download and installation.
Once the process is complete and the reboot has begun, the distribution's boot manager will allow you to choose the new kernel version. If something goes wrong, you can always revert to a previous kernel from that same menu, keeping the operating system intact.
For those who prefer a more classic approach, the possibility still exists to Download the source code and compile the kernel manuallyThis is somewhat more laborious but offers total control over the configuration and is usually better suited to very specific hardware.
Manual compilation: basic steps and common problems
The traditional procedure begins by downloading the code file, for example linux-7.0-rc3.tar.xzFrom the official kernel website. Then, the contents are extracted and the newly created directory, where the entire source structure is located, is accessed.
Before compiling, it is advisable to run a configuration tool such as make menuconfigThis allows you to adjust which modules and options are enabled in the kernel. This step is key to adapting the kernel to the existing hardware, although you can also keep the default configuration to avoid complications.
In addition, it is necessary to install the development packages and dependencies Required for compilation: build tools, encryption libraries, ncurses support, module utilities, among others. Without them, the process of make It will fail with errors that are difficult for inexperienced users to interpret.
Once the dependencies are met, you can start the build process with a command that utilizes all CPU cores, and then install the modules and the kernel itself using the usual system commands. This process can take several minutes, and even more than an hour on less powerful machines.
Among the most frequent problems when compiling are the following: absent dependenciesfailures in make menuconfig due to a lack of text interface libraries, errors from poorly cleaned previous installations, and very often, Insufficient space in the /boot partitionIt is advisable to have at least a few hundred megabytes of free space to avoid surprises when installing the new kernel image.
If everything goes well, after the next reboot the system will display Linux 7.0-rc3 as one of the boot options, preserving the possibility of reverting to a previous kernel if something doesn't work as expected.
Does it make sense to install Linux 7.0 on a computer used daily?
The short answer, for most home and professional users, is that It is not recommended to use Linux 7.0-rc3 as the main kernelAlthough this is the third Release Candidate and the code has already passed several rounds of testing, it is still a version intended for bug detection, not for production.
The kernel developers assume that there are still bugs, regressions, and compatibility issues These technologies are still undiscovered, and they need the community to find them in controlled testing environments. Installing them on a work computer or a critical server means accepting the risk of failures that are difficult to anticipate.
The situation is similar for European companies and organizations that rely on Linux in critical infrastructures: the prudent course of action is to wait for the official release of the stable version and, in many cases, to the distributions integrating the kernel with their own patches and support tools.
For technical profiles who want to get ahead and learn firsthand about the new features of Linux 7.0, the best option remains to install virtual machines in test environments or dedicate a secondary team to experimenting with these RCs.
Meanwhile, current kernel versions in the 6.x branch continue to offer a more than reasonable balance between stability, performance, and compatibility for almost any real-world use case.
The evolution of Linux 7.0 and its third Release Candidate This makes it clear that the project is going through a period of intense activity: the kernel is growing with new security features, expanded compatibility, performance tweaks in memory, networking, and file systems, and profound changes such as the addition of Rust and the removal of legacy technologies. All of this makes this version a turning point for future distributions that we will see in Spain and the rest of Europe, but it also means that its adoption should be approached cautiously and that we should closely monitor how the code settles in the coming weeks.
