]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
x86/mce/amd: Filter bogus hardware errors on Zen3 clients
authorYazen Ghannam <yazen.ghannam@amd.com>
Sat, 28 Feb 2026 14:08:14 +0000 (09:08 -0500)
committerBorislav Petkov (AMD) <bp@alien8.de>
Sun, 5 Apr 2026 10:42:22 +0000 (12:42 +0200)
commit0422b07bc4c296b736e240d95d21fbfebbfaa2ca
tree5fada9996d49fb8bc45062f098d8843272873c88
parent7aaa8047eafd0bd628065b15757d9b48c5f9c07d
x86/mce/amd: Filter bogus hardware errors on Zen3 clients

Users have been observing multiple L3 cache deferred errors after recent
kernel rework of deferred error handling.¹ ⁴

The errors are bogus due to inconsistent status values. Also, user verified
that bogus MCA_DESTAT values are present on the system even with an older
kernel.²

The errors seem to be garbage values present in the MCA_DESTAT of some L3
cache banks. These were implicitly ignored before the recent kernel rework
because these do not generate a deferred error interrupt.

A later revision of the rework patch was merged for v6.19. This naturally
filtered out most of the bogus error logs. However, a few signatures still
remain.³

Minimize the scope of the filter to the reported CPU
family/model/stepping and only for errors which don't have the Enabled
bit in the MCi status MSR.

¹ https://lore.kernel.org/20250915010010.3547-1-spasswolf@web.de
² https://lore.kernel.org/6e1eda7dd55f6fa30405edf7b0f75695cf55b237.camel@web.de
³ https://lore.kernel.org/21ba47fa8893b33b94370c2a42e5084cf0d2e975.camel@web.de
⁴ https://lore.kernel.org/r/CAKFB093B2k3sKsGJ_QNX1jVQsaXVFyy=wNwpzCGLOXa_vSDwXw@mail.gmail.com

  [ bp: Generalize the condition according to which errors are bogus. ]

Fixes: 7cb735d7c0cb ("x86/mce: Unify AMD DFR handler with MCA Polling")
Closes: https://lore.kernel.org/20250915010010.3547-1-spasswolf@web.de
Reported-by: Bert Karwatzki <spasswolf@web.de>
Signed-off-by: Yazen Ghannam <yazen.ghannam@amd.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Tested-By: Bert Karwatzki <spasswolf@web.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/20250915010010.3547-1-spasswolf@web.de
arch/x86/kernel/cpu/mce/amd.c