Hírolvasó

Az Epic kontra Google perben rövidesen megszülethetnek a jóvátételi előírások.

The LWN.net Weekly Edition for August 15, 2024 is available.

Version 0.102 of Game of Trees has been released and the port updated.

Read more…

Three new stable kernels have been released: 6.10.5, 6.6.46, and 6.1.105. As usual, they contain important fixes all over the kernel tree.

Rust is intended to let programmers write safer code. But compilers are not omniscient, and writing Rust code that interfaces with hardware (or that works with memory outside of Rust's lifetime paradigm) requires, at some point, the programmer's assurance that some operations are permissible. Benno Lossin suggested adding some more documentation to the Rust-for-Linux project clarifying the standards for commenting uses of unsafe in kernel code. There's general agreement that such standards are necessary, but less agreement on exactly when it is appropriate to use unsafe.

Security updates have been issued by AlmaLinux (389-ds-base), Debian (ffmpeg), Fedora (chromium), Red Hat (.NET 8.0, container-tools:rhel8, edk2, firefox, gnome-shell, grafana, jose, kernel, kernel-rt, krb5, open-vm-tools, orc, pcs, poppler, python-urllib3, and wget), SUSE (gtk2, gtk3, kernel, python-setuptools, python310-setuptools, python312-setuptools, python39-setuptools, and webkit2gtk3), and Ubuntu (dotnet8, libcroco, linux-azure, linux-lowlatency, linux-raspi, and linux-oracle).

De még így is maradt jócskán Google hardver, ami nem érhető el hivatalosan itthon.

Az Ascend 910C már felveszi a versenyt az Nvidia H100-zal, csak éppen mennyiséget nem tudnak gyártani belőle.

A múlt heti ítélet egyik legsúlyosabb folyománya lehet a keresőcég egyes részlegeinek kényszerített leválasztása.

A részvénypiaci manőver alapvetően jól illik az átstrukturálási folyamatba.

Már hivatalos, hogy hazánkban is indul a legújabb csúcsmodell értékesítése.

As an update to a earlier post…

The BSDCan 2024 video playlist is now complete and available on both YouTube and Peertube.

Markdown editors are a dime a dozen. Cheaper than that, actually, since many of them are open‑source software. Despite the sheer number of options, finding an editor that has all of the features that one might want can be tricky. For some users, Zettlr might be the right tool. It is a What You See is What You Mean (WYSIWYM) editor that stores its work locally as plain Markdown files. The project is billed as a "one-stop publication workbench", and is suitable for writing anything from blog posts to academic papers, maintaining a personal journal, or keeping notes in a Zettlekasten. It is simple to get started with, but rewards deeper exploration and customization.

With the imminent release of kmod 33, I thought it’d be good to have a post about the different types of module dependencies that we have in the Linux kernel and kmod. The new version adds another type, weak dependency, and as the name implies, is the weakest of all. But let’s revisit what are the other types first.

Hard (symbol) dependency

This is the first dependency that every appeared in kmod (and module-init-tools). A hard (or as some call, “symbol”) dependency occurs when your module calls or uses an exported symbol of another module. The most common way is by calling a function that is exported in another module. Example: the xe.ko calls a function ttm_bo_pin() that is provided and exported by another module, ttm.ko. Looking to the source:

$ nm build/drivers/gpu/drm/ttm/ttm.ko | grep -e "\bttm_bo_pin\b" 0000000000000bc0 T ttm_bo_pin $ nm build64/drivers/gpu/drm/xe/xe.ko | grep -e "\bttm_bo_pin\b" U ttm_bo_pin

It is not possible to insert the xe.ko module before ttm.ko and if you try via insmod (that doesn’t handle dependencies) it will fail with the kernel complaining that the ttm_bo_pin symbol is undefined.

The manual invocations to nm illustrates what the depmod tool does: it opens the modules and reads the ELF headers. Then it takes note of all the symbols required and provided by each module, creating a graph of symbol dependencies. Ultimately that leads to module dependencies: xe ➛ ttm. This is recorded in the modules.dep file and its sibling modules.dep.bin. The former is human-readable and the latter is used by libkmod, but they contain the same information: all dependencies for all the modules. Also note that each line reflects indirect dependencies: module A calls symbol from B and B calls symbol from C will lead to A depending on both B and C. Real world example:

$ cat /lib/modules/$(uname -r)/modules.dep | grep kernel/drivers/gpu/drm/xe/xe.ko.zst: kernel/drivers/gpu/drm/xe/xe.ko.zst: kernel/drivers/gpu/drm/drm_gpuvm.ko.zst kernel/drivers/gpu/drm/drm_exec.ko.zst kernel/drivers/gpu/drm/scheduler/gpu-sched.ko.zst kernel/drivers/gpu/drm/drm_buddy.ko.zst kernel/drivers/i2c/algos/i2c-algo-bit.ko.zst kernel/drivers/gpu/drm/drm_suballoc_helper.ko.zst kernel/drivers/gpu/drm/drm_ttm_helper.ko.zst kernel/drivers/gpu/drm/ttm/ttm.ko.zst kernel/drivers/gpu/drm/display/drm_display_helper.ko.zst kernel/drivers/media/cec/core/cec.ko.zst kernel/drivers/acpi/video.ko.zst kernel/drivers/platform/x86/wmi.ko.zst

So the xe.ko (with .zst extension since it’s compressed) directly or indirectly depends on drm_gpuvm.ko, drm_exec.ko, gpu-sched.ko, drm_buddy.ko, i2c-algo-bit.ko, drm_suballoc_helper.ko, drm_ttm_helper.ko, ttm.ko, drm_display_helper.ko, cec.ko, video.ko, wmi.ko. Same information, but using the kmod tools rather than looking at the raw index:

$ modinfo -F depends xe drm_display_helper,ttm,drm_gpuvm,drm_suballoc_helper,video,drm_buddy,drm_exec,drm_ttm_helper,gpu-sched,cec,i2c-algo-bit

Soft dependency

There are situations in the kernel where it’s not possible or desired to use a symbol directly from another module - they may interact by registering in a subsystem, scanning a bus etc.. In this case depmod doesn’t have enough information from the ELF file about that. Yet the user would have a more complete support if both modules were available - it may even cause failures visible to the end user and not only “partial support for features”.

The softdep implementation contains 2 parts: pre and post dependencies. The post dependencies are not very much used in practice: they instruct kmod to load another module after loading the target one.

They can come from a configuration file like e.g. /etc/modprobe.d/foo.conf or from the kernel itself by embedding that info in the module. From the kernel source this is achieved by using the macro MODULE_SOFTDEP(). Example:

lib/libcrc32c.c:MODULE_SOFTDEP("pre: crc32c");

When libkmod is loading a module it will first load, in order:

1. hard dependencies
2. soft pre dependencies
3. target module
4. soft post dependencies

Historically softdeps were also a way to (mostly) get rid of install rules, in which the configuration instructs libkmod to execute something instead of loading the module - people would add an install rule to execute something and then call modprobe again with --ignore-install to fake a dependency. That could easily lead to a runtime loop which is avoided with softdep since kmod can (and does) check for loops.

Weak dependency

After explaining the other types of dependencies, back to the new addition in kmod 33. These are very similar to pre softdep: they come either from a configuration file or embedded in the module and they express a dependency that wouldn’t cause the target module to fail to load, but that may cause the initialization to export less features or fail while initializing. There is one important difference: weak dependencies don’t cause libkmod to actually load the module. Rather the dependency information may be used by tools like dracut and other tools responsible for assembling an initrd to make the module available since it may or may not be used. Why are they called “weak”? This was a borrowed terminology from “weak symbols”: a weak symbol is there, waiting to be used, but it may or may not be, with the final decision happening in the final link stage.

With weak dependencies, hopefully some of the pre softdep embedded in the kernel may be replaced: if the target module is already doing a request_module() or in some way getting the other module to be loaded, it doesn’t need a softdep that would serialize the module load order and possibly load more modules than required.

The PostgreSQL project has released beta versions of PostgreSQL 17 containing several interesting security and usability improvements, alongside the usual performance improvements and bug fixes. If the release proceeds according to the usual timeline, the full release of version 17 is expected in September or October. The most important changes are in what PostgreSQL does when a database supervisor has their credentials revoked, and added support for incremental database backups.

Lix, the fork of Nix that LWN covered in July, has made its second release since forking. This one includes substantial changes to the backend code, including removing a dependency on Bison, and getting a change to the Nix language back upstream.

The general theme of Lix 2.91 is to perform another wave of refactorings and design improvements in preparation for our evolution plans.

Nevertheless, there are a few exciting user facing changes[.]

Version 6.4 of the Incus container manager is out.

This release builds upon the recently added OCI support from Incus 6.3, making it even easier to run application containers. It also adds a number of useful new features for clustered and larger environments with more control on the virtual CPU used when live migrating VMs and finer grained resource constraints within projects.

See this announcement for details.

Security updates have been issued by Debian (kernel and roundcube), Fedora (microcode_ctl, pypy, python2.7, and python3.6), Oracle (389-ds-base, httpd, kernel, kernel-container, and linux-firmware), Red Hat (kernel-rt), SUSE (firefox, kubernetes1.23, libqt5-qtbase, openssl-1_1, python-gunicorn, python-Twisted, python-urllib3, and qt6-base), and Ubuntu (linux-aws-5.15, linux-gkeop-5.15, linux-ibm, linux-ibm-5.15, linux-raspi, linux-azure, linux-azure-5.15, linux-azure-fde, linux-azure-fde-5.15, linux-oem-6.8, linux-oracle-5.15, and qemu).

A harmadik munkabeszüntetést szintén nemzeti ünnep idejére időzítik az érdekvédők, ami tényleg fájhat a cégnek.

Búcsút int a Microsoft a mindössze 8 éve bemutatott programnak, ami nem váltotta be a hozzá fűzött reményeket.