News: 0001496393

  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)

More Intel Diamond Rapids Enablement Landing For Linux 6.12

([Intel] 9 Minutes Ago Intel Diamond Rapids)


In addition to Intel's Linux engineers being busy [1]preparing hardware enablement support for next-gen Panther Lake client processors , they are also busy beginning to plumb Linux driver support for next-generation Xeon " [2]Diamond Rapids " support as the successor to Xeon 6 Granite Rapids. With Linux 6.12 some new bits are now set to land for Diamond Rapids.

Merged during the Linux 6.12 merge window was [3]the initial model IDs for Panther Lake and Diamond Rapids . With Diamond Rapids is where Intel is [4]ending the long-used "Family 6" era and introducing Family 19 .

This week meanwhile Intel engineer Srinivas Pandruvada sent out the patches for enabling Diamond Rapids support within the SST and TPMI power domain mapping code. This is for the Intel Speed Select Technology (SST) and Intel Topology Aware Register and PM Capsule Interface (TPMI) code as part of [5]the Intel CPU power management bring-up.

With the TPMI and SST enablement for Diamond Rapids just needing the new model IDs added to the driver (the Family 19 / Panther Cove X ID), it's safe for adding the support to the ongoing Linux 6.12 kernel rather than waiting for the Linux v6.13 cycle since it doesn't risk regressing existing hardware support.

So this week's platform-drivers-x86 [6]pull request sent out this morning goes ahead and has those Diamond Rapids patches. Also on the Intel side this week is a PMC fix for suspend/resume issues experienced on some aging Skylake and Kabylake laptops.

This platform drivers x86 pull also updates the Intel IFS maintainers entry since [7]the prior maintainer has left Intel as part of the recent cost cutting.

With Granite Rapids having just launched, there's still another year to go until Diamond Rapids debuts but expect more Linux kernel patch activity to come in the weeks ahead given Intel's typically good early hardware enablement approach for ensuring nice Linux support and optimizations ahead of product launches -- especially when it comes to their server platforms.

The platform-drivers-x86 patches should be merged to Linux 6.12 Git today ahead of the Linux 6.12-rc2 release.



[1] https://www.phoronix.com/news/Intel-Panther-Lake-5th-Gen-NPU

[2] https://www.phoronix.com/search/Diamond+Rapids

[3] https://www.phoronix.com/news/Intel-PNL-DNR-Model-IDs

[4] https://www.phoronix.com/news/Intel-Diamond-Rapids-Family-19

[5] https://www.phoronix.com/news/Intel-TPMI-Linux-Driver

[6] https://lore.kernel.org/platform-driver-x86/280a792b-ec54-419d-8cca-17b020a38d3f@redhat.com/

[7] https://www.phoronix.com/news/Intel-Maintainers-Linux-Depart


phoronix

While the engineer developed his thesis, the director leaned over to
his assistant and whispered, "Did you ever hear of why the sea is salt?"
"Why the sea is salt?" whispered back the assistant. "What do you
mean?"
The director continued: "When I was a little kid, I heard the story of
`Why the sea is salt' many times, but I never thought it important until just
a moment ago. It's something like this: Formerly the sea was fresh water and
salt was rare and expensive. A miller received from a wizard a wonderful
machine that just ground salt out of itself all day long. At first the miller
thought himself the most fortunate man in the world, but soon all the villages
had salt to last them for centuries and still the machine kept on grinding
more salt. The miller had to move out of his house, he had to move off his
acres. At last he determined that he would sink the machine in the sea and
be rid of it. But the mill ground so fast that boat and miller and machine
were sunk together, and down below, the mill still went on grinding and that's
why the sea is salt."
"I don't get you," said the assistant.
-- Guy Endore, "Men of Iron"