# git rev-parse -q --verify 19d767f0429ab26cb3de6c1f7a805a973c5dbd26^{commit}
19d767f0429ab26cb3de6c1f7a805a973c5dbd26
already have revision, skipping fetch
# git checkout -q -f -B kisskb 19d767f0429ab26cb3de6c1f7a805a973c5dbd26
# git clean -qxdf
# < git log -1
# commit 19d767f0429ab26cb3de6c1f7a805a973c5dbd26
# Author: Michael Ellerman <mpe@ellerman.id.au>
# Date:   Mon Mar 1 14:32:14 2021 +1100
# 
#     KVM: PPC: Book3S HV: Convert tbacct/stoltb_lock to raw spinlocks
#     
#     With CONFIG_PROVE_RAW_LOCK_NESTING=y we see lockdep splats when
#     running KVM:
#     
#       =============================
#       [ BUG: Invalid wait context ]
#       5.11.0-11811-gb49a419bbf28 #1 Not tainted
#       -----------------------------
#       qemu-system-ppc/14068 is trying to lock:
#       c000000134282100 (&vcpu->arch.tbacct_lock){....}-{3:3}, at: kvmppc_core_vcpu_put_hv+0x44/0xd0 [kvm_hv]
#       other info that might help us debug this:
#       context-{5:5}
#       2 locks held by qemu-system-ppc/14068:
#        #0: c0000001342800c8 (&vcpu->mutex){+.+.}-{4:4}, at: kvm_vcpu_ioctl+0xdc/0x8c0 [kvm]
#        #1: c000001e612cfa98 (&rq->lock){-.-.}-{2:2}, at: __schedule+0x1a4/0xdd0
#       stack backtrace:
#       CPU: 10 PID: 14068 Comm: qemu-system-ppc Not tainted 5.11.0-11811-gb49a419bbf28 #1
#       Call Trace:
#         dump_stack+0xec/0x144 (unreliable)
#         __lock_acquire+0xab0/0x2e70
#         lock_acquire+0x128/0x600
#         _raw_spin_lock_irqsave+0x6c/0xc0
#         kvmppc_core_vcpu_put_hv+0x44/0xd0 [kvm_hv]
#         kvmppc_core_vcpu_put+0x34/0x50 [kvm]
#         kvm_arch_vcpu_put+0x20/0x40 [kvm]
#         kvm_sched_out+0x44/0x80 [kvm]
#         __schedule+0x684/0xdd0
#         preempt_schedule_common+0x30/0x60
#         __cond_resched+0x60/0x80
#         kvmhv_run_single_vcpu+0xe2c/0x1220 [kvm_hv]
#         kvmppc_vcpu_run_hv+0x984/0xcd0 [kvm_hv]
#         kvmppc_vcpu_run+0x34/0x48 [kvm]
#         kvm_arch_vcpu_ioctl_run+0x340/0x450 [kvm]
#         kvm_vcpu_ioctl+0x2ac/0x8c0 [kvm]
#         sys_ioctl+0x300/0x1050
#         system_call_exception+0x160/0x230
#         system_call_common+0xec/0x278
#     
#     Which is complaining that we're taking a regular
#     spinlock (tbacct_lock) while holding a raw spinlock (rq->lock).
#     
#     To fix it, convert tbacct_lock to a raw spinlock.
#     
#     We see a similar splat for stoltb_lock, the analysis & fix is the
#     same.
#     
#     Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
# < /opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-gcc --version
# < /opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-ld --version
# < git log --format=%s --max-count=1 19d767f0429ab26cb3de6c1f7a805a973c5dbd26
# < make -s -j 48 ARCH=powerpc O=/kisskb/build/powerpc-next_ppc64_defconfig+NO_KVM_powerpc-gcc5 CROSS_COMPILE=/opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-  ppc64_defconfig
# Added to kconfig CONFIG_KVM_BOOK3S_64=n
# Added to kconfig CONFIG_KVM=n
# < make -s -j 48 ARCH=powerpc O=/kisskb/build/powerpc-next_ppc64_defconfig+NO_KVM_powerpc-gcc5 CROSS_COMPILE=/opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-  help
# make -s -j 48 ARCH=powerpc O=/kisskb/build/powerpc-next_ppc64_defconfig+NO_KVM_powerpc-gcc5 CROSS_COMPILE=/opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-  olddefconfig
# make -s -j 48 ARCH=powerpc O=/kisskb/build/powerpc-next_ppc64_defconfig+NO_KVM_powerpc-gcc5 CROSS_COMPILE=/opt/cross/kisskb/korg/gcc-5.5.0-nolibc/powerpc64-linux/bin/powerpc64-linux-  
Completed OK
# rm -rf /kisskb/build/powerpc-next_ppc64_defconfig+NO_KVM_powerpc-gcc5
# Build took: 0:03:33.277721