Linux 6.11-rc3 Released - Boosts Intel EMR Performance, Fixes 32-bit PTI For Meltdown
Linus Torvalds released Linux 6.11-rc3 a few minutes ago as the latest weekly release candidate.
The Linux 6.11 kernel continues tracking for a release in mid-September. Linus Torvalds wrote in today's 6.11-rc3 announcement:
Among the notable fixes this week are more AMD Zen 5 CPU IDs added, an ASUS ROG Ally X gaming handheld quirk, and an Intel P-State EPP update for helping to boost the performance and power efficiency of Xeon Scalable "Emerald Rapids" CPUs.
Also being merged today for Linux Git and making it just in-time for Linux 6.11-rc3 is fixing recently discovered breakage in the Linux kernel 32-bit x86 Page Table Isolation (PTI) code. The Kernel Page Table Isolation code is important for mitigating the Meltdown security bug. It turns out the 32-bit x86 PTI code wasn't quite up to par:
Onward to more Linux 6.11 kernel testing and benchmarking.
The Linux 6.11 kernel continues tracking for a release in mid-September. Linus Torvalds wrote in today's 6.11-rc3 announcement:
"Nothing particularly strange or interesting going on, things look normal.
Half the changes are to drivers (all the usual suspects: sound, gpu, and networking are big, but we have usb and other misc driver fixes there too).
The rest is the usual random mix, with filesystem fixes (bcachefs, btrfs, smb), architecture fixes (arm64, loongarch, x86), documentation, and networking. And some core kernel and tooling fixlets too."
Among the notable fixes this week are more AMD Zen 5 CPU IDs added, an ASUS ROG Ally X gaming handheld quirk, and an Intel P-State EPP update for helping to boost the performance and power efficiency of Xeon Scalable "Emerald Rapids" CPUs.
Also being merged today for Linux Git and making it just in-time for Linux 6.11-rc3 is fixing recently discovered breakage in the Linux kernel 32-bit x86 Page Table Isolation (PTI) code. The Kernel Page Table Isolation code is important for mitigating the Meltdown security bug. It turns out the 32-bit x86 PTI code wasn't quite up to par:
"Fix 32-bit PTI for real.
pti_clone_entry_text() is called twice, once before initcalls so that initcalls can use the user-mode helper and then again after text is set read only. Setting read only on 32-bit might break up the PMD mapping, which makes the second invocation of pti_clone_entry_text() find the mappings out of sync and failing.
Allow the second call to split the existing PMDs in the user mapping and synchronize with the kernel mapping."
Onward to more Linux 6.11 kernel testing and benchmarking.
Add A Comment