)]}'
{
  "commit": "58c95070da3a504fbeca7939435bbb062cb96ea3",
  "tree": "a22ad61caee779b2f20fca6b87a5589a10507685",
  "parents": [
    "6fec21449a62702a582cecbb0b351363e039c95e"
  ],
  "author": {
    "name": "Paolo Bonzini",
    "email": "pbonzini@redhat.com",
    "time": "Tue Sep 22 10:15:59 2015 +0200"
  },
  "committer": {
    "name": "Paolo Bonzini",
    "email": "pbonzini@redhat.com",
    "time": "Fri Sep 25 10:31:29 2015 +0200"
  },
  "message": "KVM: x86: fix off-by-one in reserved bits check\n\n29ecd6601904 (\"KVM: x86: avoid uninitialized variable warning\",\n2015-09-06) introduced a not-so-subtle problem, which probably\nescaped review because it was not part of the patch context.\n\nBefore the patch, leaf was always equal to iterator.level.  After,\nit is equal to iterator.level - 1 in the call to is_shadow_zero_bits_set,\nand when is_shadow_zero_bits_set does another \"-1\" the check on\nreserved bits becomes incorrect.  Using \"iterator.level\" in the call\nfixes this call trace:\n\nWARNING: CPU: 2 PID: 17000 at arch/x86/kvm/mmu.c:3385 handle_mmio_page_fault.part.93+0x1a/0x20 [kvm]()\nModules linked in: tun sha256_ssse3 sha256_generic drbg binfmt_misc ipv6 vfat fat fuse dm_crypt dm_mod kvm_amd kvm crc32_pclmul aesni_intel aes_x86_64 lrw gf128mul glue_helper ablk_helper cryptd fam15h_power amd64_edac_mod k10temp edac_core amdkfd amd_iommu_v2 radeon acpi_cpufreq\n[...]\nCall Trace:\n  dump_stack+0x4e/0x84\n  warn_slowpath_common+0x95/0xe0\n  warn_slowpath_null+0x1a/0x20\n  handle_mmio_page_fault.part.93+0x1a/0x20 [kvm]\n  tdp_page_fault+0x231/0x290 [kvm]\n  ? emulator_pio_in_out+0x6e/0xf0 [kvm]\n  kvm_mmu_page_fault+0x36/0x240 [kvm]\n  ? svm_set_cr0+0x95/0xc0 [kvm_amd]\n  pf_interception+0xde/0x1d0 [kvm_amd]\n  handle_exit+0x181/0xa70 [kvm_amd]\n  ? kvm_arch_vcpu_ioctl_run+0x68b/0x1730 [kvm]\n  kvm_arch_vcpu_ioctl_run+0x6f6/0x1730 [kvm]\n  ? kvm_arch_vcpu_ioctl_run+0x68b/0x1730 [kvm]\n  ? preempt_count_sub+0x9b/0xf0\n  ? mutex_lock_killable_nested+0x26f/0x490\n  ? preempt_count_sub+0x9b/0xf0\n  kvm_vcpu_ioctl+0x358/0x710 [kvm]\n  ? __fget+0x5/0x210\n  ? __fget+0x101/0x210\n  do_vfs_ioctl+0x2f4/0x560\n  ? __fget_light+0x29/0x90\n  SyS_ioctl+0x4c/0x90\n  entry_SYSCALL_64_fastpath+0x16/0x73\n---[ end trace 37901c8686d84de6 ]---\n\nReported-by: Borislav Petkov \u003cbp@alien8.de\u003e\nTested-by: Borislav Petkov \u003cbp@alien8.de\u003e\nSigned-off-by: Paolo Bonzini \u003cpbonzini@redhat.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "6f207637e43736502585cd96ef2dcb5f018010c0",
      "old_mode": 33188,
      "old_path": "arch/x86/kvm/mmu.c",
      "new_id": "ff606f5079137736e6827c3b3f33dec808854974",
      "new_mode": 33188,
      "new_path": "arch/x86/kvm/mmu.c"
    }
  ]
}