Linus Torvalds Slams Intel’s Linear Address Masking (LAM) in Linux 6.2
Linus Torvalds sharply criticized Intel’s new Linear Address Masking feature, deeming its kernel implementation fundamentally broken and refusing to merge it into Linux 6.2, while Intel plans to rewrite the code for a future release.
Linus Torvalds, known for his blunt criticism of hardware vendors, recently attacked Intel’s new Linear Address Masking (LAM) feature, which Intel wants to merge into the Linux 6.2 kernel.
LAM, introduced by Intel in 2020, is a mechanism that uses unused high‑order bits of user‑space addresses as metadata, similar to AMD’s UAI and ARM’s TBI, and is expected to appear in the upcoming Sapphire Rapids CPUs.
Torvalds argued that the proposed implementation is fundamentally broken, poorly named, and adds unnecessary complexity to the kernel’s mm subsystem. In the Linux‑kernel mailing list he wrote that Intel should call the feature “Top‑Bits‑Ignore” (TBI) instead of inventing a new acronym.
He further claimed that the changes to untagged_addr() are destructive and that the LAM code would become a persistent bug, even on x86‑64, because it is not specific to the architecture.
After heated discussion, the LAM patch was removed from the 6.2 merge window. Intel’s engineers plan to rewrite the code and submit a revised version for Linux 6.3, hoping to address Torvalds’ concerns.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
21CTO
21CTO (21CTO.com) offers developers community, training, and services, making it your go‑to learning and service platform.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.
