x86/topology/amd: Evaluate SMT in CPUID leaf 0x8000001e only on family 0x17 and greater
authorThomas Gleixner <tglx@linutronix.de>
Tue, 28 May 2024 20:21:31 +0000 (22:21 +0200)
committerBorislav Petkov (AMD) <bp@alien8.de>
Thu, 30 May 2024 13:58:55 +0000 (15:58 +0200)
commit34bf6bae3286a58762711cfbce2cf74ecd42e1b5
tree6b4140c145f436968fe8fdc84609024aceebd16a
parent1613e604df0cd359cf2a7fbd9be7a0bcfacfabd0
x86/topology/amd: Evaluate SMT in CPUID leaf 0x8000001e only on family 0x17 and greater

The new AMD/HYGON topology parser evaluates the SMT information in CPUID leaf
0x8000001e unconditionally while the original code restricted it to CPUs with
family 0x17 and greater.

This breaks family 0x15 CPUs which advertise that leaf and have a non-zero
value in the SMT section. The machine boots, but the scheduler complains loudly
about the mismatch of the core IDs:

  WARNING: CPU: 1 PID: 0 at kernel/sched/core.c:6482 sched_cpu_starting+0x183/0x250
  WARNING: CPU: 0 PID: 1 at kernel/sched/topology.c:2408 build_sched_domains+0x76b/0x12b0

Add the condition back to cure it.

  [ bp: Make it actually build because grandpa is not concerned with
    trivial stuff. :-P ]

Fixes: f7fb3b2dd92c ("x86/cpu: Provide an AMD/HYGON specific topology parser")
Closes: https://gitlab.archlinux.org/archlinux/packaging/packages/linux/-/issues/56
Reported-by: Tim Teichmann <teichmanntim@outlook.de>
Reported-by: Christian Heusel <christian@heusel.eu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Tested-by: Tim Teichmann <teichmanntim@outlook.de>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/7skhx6mwe4hxiul64v6azhlxnokheorksqsdbp7qw6g2jduf6c@7b5pvomauugk
arch/x86/kernel/cpu/topology_amd.c