为什么直到现在KernelSU Next的开发者还是不知道这场风波的起因到底是什么？还是说，他在刻意地转移重点并逃避？

Why does the developer of KernelSU Next still not know what caused this incident? Or is he deliberately trying to divert the focus and evade the issue?

在他为他的冒犯性言论道歉之前，我仍然不建议任何人使用KernelSU Next。什么？他把那条言论删了？删了就能解决问题了？删了就可以当作无事发生了？互联网是有记忆的哥们。

Until he apologizes for his offensive comments, I still don't recommend anyone to use KernelSU Next. What? He deleted that comment? Deleting it will solve the problem? Deleting it will just be like nothing happened? The Internet has a memory, man.

After my personal testing, Melt Kernel v3.6 works basically normally on AOSPA Beta 8.

Next I will look for other ROMs based on the OSS kernel for testing.

KernelSU Next的开发者已为他的言论道歉，因此我将恢复对KernelSU Next的评价。

The developer of KernelSU Next has apologized for his comments, so I will resume my review of KernelSU Next.

Moto Widget (ported by @meoify) Update
2025-02-15
#MotoWidget

Build version:
4.12.41

Build date:
2025-02-14 16:52:15

MD5:
3d317f690790bb3c3c26043643f9e69f

Download:
Moto-Widget-4.12.41.apk
Pling | Direct

Size:
40.52 MB

Google Clang Prebuilt Update
#clang

Commit

Download tar.gz:
clang 20.0.0 (based on r547379) from build 13065274

Xiaomi.eu inject Module Update
2025-02-15
#XiaomiEU

Build date:
Sat, 15 Feb 2025 06:38:42 UT

MD5:
6bac50c995efb29bf0c266e63563c995

Download:
XiaomiEUModule_2025.02.15.apk

Size:
12.8 KB

Since this is a poll about the compatibility of Melt Kernel with various ROMs, I will add some information that users may be interested in.

Starting from version v3.3, Melt Kernel will replace all kernel-related things (kernel image, kernel modules, dtb, dtbo), so in theory Melt Kernel can be booted on any ROM. The bugs encountered by users usually come from the various modifications made by the LineageOS team to the OSS kernel, such as:

1. The LineageOS team switched to another goodix touchscreen driver, resulting in the screen-off gesture not working properly after installing Melt Kernel.
2. The LineageOS team switched to a completely rewritten xiaomi_touch module, and the behavior is different from MIUI/HyperOS, which may also cause the screen-off gesture not to work properly.
3. The LineageOS team switched to OSS IR HAL, resulting in the infrared remote controller not working properly after installing Melt Kernel.

AOSPA is also based on OSS kernel. For Melt Kernel, AOSPA is the easiest to adapt because the maintainer have high technical capabilities and their own ideas, and many things are basically fixed and will not change easily. But for other OSS kernel AOSP roms, some of them will actively keep synchronization with the LineageOS team, but some choose to maintain the status quo, making it difficult for Melt Kernel to adapt them. However, as I mentioned above, the bugs encountered by users are nothing more than screen-off gestures and infrared remote control. Everything else basically works as expected.

Is Melt Kernel suitable for OSS kernel-based AOSP roms? My answer is: maybe not. What you have to know is that Melt Kernel is still using Xiaomi's pre-compiled (most of) kernel modules, dtb, and dtbo, which is necessary for compatibility with MIUI14/HyperOS. Some of Xiaomi's pre-compiled kernel modules are only suitable for MIUI14/HyperOS, not for AOSP roms, such as: millet, migt, etc.

Therefore, if you find that the power consumption increases after installing Melt Kernel on an AOSP rom based on OSS kernel, this is normal.

But in any case, you have the right to choose to install Melt Kernel. Unlike the weird OSS kernel, Melt Kernel complies with Google's GKI 2.0 specification. Therefore, for AOSP rom based on OSS kernel, after installing Melt Kernel, it will be compatible with GKI 2.0. You can install any KernelSU you like (such as KernelSU Next, MKSU) through LKM, and installing APatch has become easier. You can even use other custom GKIs you like (such as Pandora, ztc1997, WildPlus) instead of Melt Kernel. And this is exactly the goal of Melt Kernel.

New "Faux Bus" API Merged For Linux 6.14 - Including Both Rust & C Bindings
16 February 2025 - Faux Bus

A few weeks back the Linux kernel "Faux Bus" was proposed by Greg Kroah-Hartman as a "fake" bus solution for simple devices. Today ahead of the Linux 6.14-rc3 tagging, the faux bus code was merged and comes at the same time both with C and Rust language bindings.

Comments

#Phoronix #LinuxKernel

Linux 6.14-rc3 Released With Faux Bus & Various Fixes
16 February 2025 - Linux 6.14

Linus Torvalds just released Linux 6.14-rc3 as the newest weekly release candidate for Linux 6.14 that will debuting as stable before the end of March.

Comments

#Phoronix #LinuxKernel

After fixing compatibility with xiaomi_touch rewritten by the LineageOS team, Melt Kernel can now work properly on the latest LineageOS rom almost perfectly.

The beta version of Melt Kernel, which will be released to the public today, will remove the installation restrictions set in v3.6. For roms that have followed the LineageOS team and replaced xiaomi_touch (for example: CrDroid A15), it can also be installed.

Single-touch gestures, DT2W, increased touch sampling rate, all work fine.

Linus Torvalds Would Reportedly Merge Rust Kernel Code Over Maintainer Objections
18 February 2025 - Torvalds Override

The drama surrounding Rust code within the Linux kernel continues... Christoph Hellwig is the maintainer of the DMA mapping helpers and several other areas of the kernel has been an outspoken critic of Rust code or secondary programming languages within the Linux kernel kernel. Hellwig has been critical of Rust code for the Linux kernel and its long-term maintainability. Today he's out with another mailing list post where he notes that Linus Torvalds mentioned in private he would override maintainer vetoes on Rust code within the kernel.

Comments

#Phoronix #LinuxKernel

Greg Kroah-Hartman Makes A Compelling Case For New Linux Kernel Drivers To Be Written In Rust
19 February 2025 - New Rust Code

The debate over the Linux kernel's Rust programming language policy continues... While some kernel maintainers are against it, Linus Torvalds has reportedly said he would override maintainers that may be against honoring Rust code. Linux's second-in-command Greg Kroah-Hartman has also been a big proponent of Rust kernel code. He's crafted another Linux kernel mailing list post today outlining the benefits of Rust and encouraging new kernel code/drivers to be in Rust rather than C.

Comments

#Phoronix #LinuxKernel

New Patches Would Make All Kernel Encryption/Decryption Faster On x86/x86_64 Hardware
20 February 2025 - Widespread Gains

On top of all the recent x86/x86_64 Linux kernel crypto improvements made recently by Google engineer Eric Biggers to better laverage AVX-512 and other modern x86 ISA features, a new patch-set posted today by Biggers would help make all x86/x86_64 kernel encryption/decryption at least slightly faster.

Comments

#Phoronix #LinuxKernel

Linux Finally Introducing A Standardized Way Of Informing User-Space Over Hung GPUs
20 February 2025 - Wedged GPUs

The upcoming Linux 6.15 kernel is set to finally introduce a standardized way of informing user-space of GPUs becoming hung or otherwise unresponsive. This is initially wired up for AMD and Intel graphics drivers on Linux so the user can be properly notified of problems and/or user-space software taking steps to address the hung/unresponsive graphics processor.

Comments

#Phoronix #LinuxKernel

Linux Lazy Unmap Flush "LUF" Reducing TLB Shootdowns By 97%, Faster AI LLM Performance
20 February 2025 - Linux LUF

SK has been working on a Linux kernel feature dubbed Lazy Unmap Flush "LUF" to defer TLB flushes until folios have been unmapped and freed are eventually allocated again.

Comments

#Phoronix #LinuxKernel