KVM: SVM: Hide SEV migration lockdep goo behind CONFIG_PROVE_LOCKING
authorSean Christopherson <seanjc@google.com>
Mon, 13 Jun 2022 21:42:37 +0000 (21:42 +0000)
committerPaolo Bonzini <pbonzini@redhat.com>
Wed, 15 Jun 2022 12:07:22 +0000 (08:07 -0400)
commite5380f6d7586ea3ef3a55d8cf19ceffacea31392
tree4fc03286865c8ccaffba5a60d485b95385248258
parent5bdae49fc2f689b5f896b54bd9230425d3643dab
KVM: SVM: Hide SEV migration lockdep goo behind CONFIG_PROVE_LOCKING

Wrap the manipulation of @role and the manual mutex_{release,acquire}()
invocations in CONFIG_PROVE_LOCKING=y to squash a clang-15 warning.  When
building with -Wunused-but-set-parameter and CONFIG_DEBUG_LOCK_ALLOC=n,
clang-15 seees there's no usage of @role in mutex_lock_killable_nested()
and yells.  PROVE_LOCKING selects DEBUG_LOCK_ALLOC, and the only reason
KVM manipulates @role is to make PROVE_LOCKING happy.

To avoid true ugliness, use "i" and "j" to detect the first pass in the
loops; the "idx" field that's used by kvm_for_each_vcpu() is guaranteed
to be '0' on the first pass as it's simply the first entry in the vCPUs
XArray, which is fully KVM controlled.  kvm_for_each_vcpu() passes '0'
for xa_for_each_range()'s "start", and xa_for_each_range() will not enter
the loop if there's no entry at '0'.

Fixes: 0c2c7c069285 ("KVM: SEV: Mark nested locking of vcpu->lock")
Reported-by: kernel test robot <lkp@intel.com>
Cc: Peter Gonda <pgonda@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220613214237.2538266-1-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
arch/x86/kvm/svm/sev.c