New Intel Linux Patches Continue Working To Improve Hybrid CPU Task Placement

[phoronix]

Phoronix: New Intel Linux Patches Continue Working To Improve Hybrid CPU Task Placement

Building off some "request for comments" patches sent out in April, a new set of patches appeared on Friday for the Intel P-State Linux driver for setting the asymmetric CPU capacity on hybrid systems. This is another attempt at helping to improve the Linux kernel scheduler behavior in ensuring optimal task placement between Intel Core processors having a mix of P and E cores. This patch series in particular helps when SMT / Hyper Threading support is disabled or like with upcoming Lunar Lake processors where there is no HT support...

https://www.phoronix.com/news/Intel-Linux-Mid-2024-Hybrid

[pWe00Iri3e7Z9lHOX2Qx]

At least with all the work Intel and AMD are putting into shedulers dealing with their non-homogeneous CPUs, this area of the kernel should end up being quite robust for whatever crazy designs people come up with. It's too bad it has taken this long, but at least it keeps getting better.

[NotMine999]

At least with all the work Intel and AMD are putting into shedulers dealing with their non-homogeneous CPUs, this area of the kernel should end up being quite robust for whatever crazy designs people come up with. It's too bad it has taken this long, but at least it keeps getting better.

IMHO Linux kernel code development has mostly been "reactive" rather than "proactive".

And why? The Linux community either: (1) doesn't have the resources (time, people, money, etc.) to commit to pre-release development efforts like Micro$haft; or, (2) doesn't have the "access" needed to develop code based on pre-release parts.

So that means Linux code development tends to get done after a product is released by open source developers that are willing to invest the time, effort, and expense to poke around these products to figure out how they internally work since lots of needed development information is not available in public documention and signing NDAs, if they are available, seems to "go against the grain" of many open source developers.

We have all heard the stories of limited Linux development resources at various manufacturers (for whatever reason those limits exist). We can also surmise all of the internal legal reviews of Linux code that might occur at these various manufacturers so that valuable "intellectual property" is not revealed.