News: 0001650665

  ARM Give a man a fire and he's warm for a day, but set fire to him and he's warm for the rest of his life (Terry Pratchett, Jingo)

Initial Apple M3 Pro / Max / Ultra Support Being Upstreamed For Linux 7.3

([Apple] 4 Hours Ago Apple M3)


Upstreamed for the Linux 7.2 kernel was the Device Tree (DT) support for [1]booting the mainline Linux kernel on the base Apple M3 SoC . Though it's [2]not yet useful for end-users with many features remaining to be supported and upstreamed . Now for the upcoming Linux 7.3 kernel, the M3 Pro, Max, and Ultra variants will also have initial support for booting on the mainline kernel.

This M3 Pro / M3 Max / M3 Ultra support is the basic Device Tree files for the mainline Linux kernel which take it up to around the same level of support of the basic M3 SoC... Again, not yet useful for end-users due to various features not yet being supported, such as GPU acceleration. But at least it's a start and begins laying the foundation for better mainline kernel support in future kernel versions.

The Apple M3 Pro and M3 Max [3]debuted in late 2023 while the M3 Ultra arrived in March 2025 with the Mac Studio. The timing isn't yet clear when the mainline Linux kernel will have usable day-to-day support for the M3 series or newer. Prior to that it will be easier for users to run the downstream Asahi Linux distribution but even there many M3 features remain a work-in-progress by their small volunteer team as shown with the [4]M3 feature matrix .

[5]The Apple DT pull intended for the upcoming Linux 7.3 merge window gets the DTs laid out for all Apple devices with the M3 Pro/Max/Ultra SoCs. That pull also has the HWMON support for Apple M1 and M2 series in dealing with [6]Apple's wild mess of sensors .



[1] https://www.phoronix.com/news/Linux-7.2-SoC-Updates

[2] https://www.phoronix.com/news/Linux-7.2-Boots-Apple-M3

[3] https://www.phoronix.com/news/Apple-Announces-M3

[4] https://asahilinux.org/docs/platform/feature-support/m3/

[5] https://lore.kernel.org/lkml/20260807161840.6011-2-sven@kernel.org/

[6] https://www.phoronix.com/news/Apple-Wild-Mess-Sensors-Linux



<Reed> It is important to note that the primary reason the Roman Empire
fail is that they had no concept of zero... thus they could not
test the success or failure of their C programs.