[SRU][J:linux-gcp][PATCH v2 0/1] Fix bugs preventing boot on Intel TDX-enabled hosts

[Ian Whitfield]

BugLink: https://bugs.launchpad.net/bugs/2097811

SRU Justification:

[Impact]

Google has requested an upstream fix to enable the boot of 5.15 kernel
instances on their Intel TDX enabled infrastructure. The fix is
described as preventing the recalculation of physical address width of
the cpu during boot in a few select places, presumably using a more
consistent and accurate value throughout the boot process.

[Fix]

There were a large number of unrelated conflicts resolved by adjusting
the context around the changes in the patch. One conflict did have a
direct impact on the patch, but it was resolved by tracing where a
function call had been moved, and making the original changes there.

This thread applies to the Jammy linux-gcp kernel for the sake of
getting the patch applied sooner to a smaller number of kernels where
this issue is more critical. Another thread will be sent to include this
fix in the generic Jammy kernel where the risk is higher. The patch is
unchanged.

[Test Plan]

Google reported inability to boot Focal images (which use a backport of
this kernel) on a specific configuration in a deployment zone where
Intel TDX was enabled. This patchset was tested by booting a Jammy image
on one such machine (which uses the 6.8 kernel), installing this patched
kernel, and booting into it. Before this patch is applied, the installed
kernel doesn't finish the boot process, and after the patch is applied,
it boots as normal.

[Where problems could occur]

As these changes affect booting and the kernel's understanding of the
cpu, an error in the backporting of this patch could cause the user
to be unable to boot the kernel. The patch was primarily modified in
the MTRR cleanup feature, which can be disabled with a kernel parameter.
If the fixes don't work, we would see the kernel continue to not be
bootable on TDX-enabled hosts.

v2: Removed additional cherry-picked commit that was not necessary and
required more of a backport effort.

Juergen Gross (1):
  x86/mtrr: Remove physical address size calculation

 arch/x86/kernel/cpu/mtrr/cleanup.c | 16 ++++----
 arch/x86/kernel/cpu/mtrr/generic.c | 12 +++++-
 arch/x86/kernel/cpu/mtrr/mtrr.c    | 61 ++++--------------------------
 arch/x86/kernel/cpu/mtrr/mtrr.h    |  4 +-
 4 files changed, 29 insertions(+), 64 deletions(-)

-- 
2.43.0