On Tue, Oct 31, 2017 at 2:59 PM, Dmitry Vyukov <dvyukov@xxxxxxxxxx> wrote: > On Tue, Oct 31, 2017 at 4:51 PM, syzbot > <bot+53e9b911da5806836ed78f23e2b8ead1c905b469@xxxxxxxxxxxxxxxxxxxxxxxxx> > wrote: >> Hello, >> >> syzkaller hit the following crash on >> 0b5477d9dabd96ded4c5ef7a5f08b00188fc1dec >> git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/master >> compiler: gcc (GCC) 7.1.1 20170620 >> .config is attached >> Raw console output is attached. >> >> syzkaller reproducer is attached. See https://goo.gl/kgGztJ >> for information about syzkaller reproducers > > For the provided repro crashes in native_write_cr4, which seems to be > a separate unrelated issue (reported today separately). > > Here is another repro associated with this crash: > https://gist.githubusercontent.com/dvyukov/7bdbcdb30b084516e7c5785fa82db838/raw/98fe8a27f899d6d8ea9d1dd3092e6c917570e1b6/gistfile1.txt > I've just tried it on upstream > 5f479447d983111c039f1d6d958553c1ad1b2ff1 (Oct 30); > "./syz-execprog -procs=8 -repeat repro.txt" it crashes as: > > ------------[ cut here ]------------ > kernel BUG at arch/x86/kvm/mmu.c:1194! > invalid opcode: 0000 [#1] SMP KASAN > Modules linked in: > CPU: 1 PID: 3091 Comm: syz-executor Not tainted 4.14.0-rc7+ #13 > kvm [3107]: vcpu0, guest rIP: 0x9111 Hyper-V uhandled wrmsr: > 0x40000020 data 0xf0047 > Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011 > task: ffff880064f206c0 task.stack: ffff88005eb40000 > RIP: 0010:pte_list_remove+0x3ae/0x3c0 arch/x86/kvm/mmu.c:1193 > RSP: 0018:ffff88005eb469d8 EFLAGS: 00010282 > RAX: 0000000000000028 RBX: ffff88006af40d80 RCX: 0000000000000000 > RDX: 0000000000000028 RSI: 1ffff1000bd68cfb RDI: ffffed000bd68d2f > RBP: ffff88005eb46a18 R08: 0000000000000001 R09: 0000000000000000 > R10: ffff88005eb46dc0 R11: 0000000000000000 R12: ffff8800687a0078 > R13: 0000000000000000 R14: ffff88006a2c7910 R15: ffff88006a2c7938 > FS: 0000000000000000(0000) GS:ffff88006ca80000(0000) knlGS:0000000000000000 > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > CR2: 0000000000000000 CR3: 0000000005a22003 CR4: 00000000001626e0 > Call Trace: > rmap_remove arch/x86/kvm/mmu.c:1270 [inline] > drop_spte+0x15a/0x250 arch/x86/kvm/mmu.c:1352 > mmu_page_zap_pte+0x224/0x340 arch/x86/kvm/mmu.c:2484 > kvm_mmu_page_unlink_children arch/x86/kvm/mmu.c:2506 [inline] > kvm_mmu_prepare_zap_page+0x1c5/0x1310 arch/x86/kvm/mmu.c:2550 > kvm_zap_obsolete_pages arch/x86/kvm/mmu.c:5330 [inline] > kvm_mmu_invalidate_zap_all_pages+0x4a0/0x680 arch/x86/kvm/mmu.c:5371 > kvm_arch_flush_shadow_all+0x15/0x20 arch/x86/kvm/x86.c:8444 > kvm_mmu_notifier_release+0x59/0x90 > arch/x86/kvm/../../../virt/kvm/kvm_main.c:467 > __mmu_notifier_release+0x1d5/0x690 mm/mmu_notifier.c:75 > mmu_notifier_release include/linux/mmu_notifier.h:222 [inline] > exit_mmap+0x42d/0x530 mm/mmap.c:2981 > __mmput kernel/fork.c:928 [inline] > mmput+0x223/0x6d0 kernel/fork.c:949 > exit_mm kernel/exit.c:544 [inline] > do_exit+0x904/0x1ad0 kernel/exit.c:852 > do_group_exit+0x149/0x400 kernel/exit.c:968 > get_signal+0x73f/0x16d0 kernel/signal.c:2334 > do_signal+0x94/0x1ee0 arch/x86/kernel/signal.c:808 > exit_to_usermode_loop+0x214/0x310 arch/x86/entry/common.c:158 > prepare_exit_to_usermode arch/x86/entry/common.c:197 [inline] > syscall_return_slowpath+0x42f/0x510 arch/x86/entry/common.c:266 > kvm [3097]: vcpu0, guest rIP: 0x9111 Hyper-V uhandled wrmsr: > 0x40000020 data 0xf0047 > entry_SYSCALL_64_fastpath+0xbc/0xbe > RIP: 0033:0x447ad7 > RSP: 002b:00007f4d993c2058 EFLAGS: 00000207 ORIG_RAX: 0000000000000010 > RAX: 0000000000000000 RBX: 00007f4d993c27b0 RCX: 0000000000447ad7 > RDX: 00007f4d993c27b0 RSI: 00000000c008ae05 RDI: 0000000000000018 > RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 > R10: 0000000000000015 R11: 0000000000000207 R12: 0000000000000018 > R13: 0000000000008340 R14: 00000000006ec3e0 R15: 00007f4d993c4700 > Code: f9 c9 5e 00 48 8b 75 d0 48 c7 c7 40 60 e2 84 e8 8e 80 49 00 0f > 0b e8 e2 c9 5e 00 48 8b 75 d0 48 c7 c7 00 60 e2 84 e8 77 80 49 00 <0f> > 0b 4c 89 ef e8 48 8f 93 00 e9 01 fe ff ff 0f 1f 00 55 48 89 > RIP: pte_list_remove+0x3ae/0x3c0 arch/x86/kvm/mmu.c:1193 RSP: ffff88005eb469d8 > ---[ end trace a329c3bead6aac1c ]--- > > > >> kasan: CONFIG_KASAN_INLINE enabled >> kasan: GPF could be caused by NULL-ptr deref or user memory access >> general protection fault: 0000 [#1] SMP KASAN >> Dumping ftrace buffer: >> (ftrace buffer empty) >> Modules linked in: >> CPU: 0 PID: 4407 Comm: syz-executor5 Not tainted 4.13.0-rc2+ #9 >> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS >> Google 01/01/2011 >> task: ffff8801d661c0c0 task.stack: ffff8801ca9b0000 >> RIP: 0010:__kvm_memslots include/linux/kvm_host.h:571 [inline] >> RIP: 0010:gfn_to_rmap+0x57f/0x6b0 arch/x86/kvm/mmu.c:1235 >> RSP: 0018:ffff8801ca9b6820 EFLAGS: 00010297 >> RAX: dffffc0000000000 RBX: ffff8801cc3e1078 RCX: 1ffff1003987c204 >> RDX: 0000000000000000 RSI: ffff8801cc3ec3c8 RDI: ffff8801cc3e1080 >> RBP: ffff8801ca9b6960 R08: 0000000000000002 R09: 0000000000000004 >> R10: ffff8801ca9b6cc8 R11: ffffffff81120ef1 R12: 0000000000000002 >> R13: 000000000000000f R14: 000000000000000e R15: dffffc0000000000 >> FS: 0000000000000000(0000) GS:ffff8801dc000000(0000) knlGS:0000000000000000 >> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 >> CR2: 0000000000000000 CR3: 00000001d28ce000 CR4: 00000000001426f0 >> Call Trace: >> rmap_remove arch/x86/kvm/mmu.c:1267 [inline] >> drop_spte+0x161/0x270 arch/x86/kvm/mmu.c:1350 >> mmu_page_zap_pte+0x224/0x350 arch/x86/kvm/mmu.c:2482 >> kvm_mmu_page_unlink_children arch/x86/kvm/mmu.c:2504 [inline] >> kvm_mmu_prepare_zap_page+0x1b7/0x1260 arch/x86/kvm/mmu.c:2548 >> kvm_zap_obsolete_pages arch/x86/kvm/mmu.c:5223 [inline] >> kvm_mmu_invalidate_zap_all_pages+0x4a0/0x680 arch/x86/kvm/mmu.c:5264 >> kvm_arch_flush_shadow_all+0x15/0x20 arch/x86/kvm/x86.c:8394 >> kvm_mmu_notifier_release+0x59/0x90 >> arch/x86/kvm/../../../virt/kvm/kvm_main.c:508 >> __mmu_notifier_release+0x1d5/0x690 mm/mmu_notifier.c:75 >> mmu_notifier_release include/linux/mmu_notifier.h:235 [inline] >> exit_mmap+0x3a3/0x470 mm/mmap.c:2972 >> __mmput kernel/fork.c:903 [inline] >> mmput+0x223/0x6e0 kernel/fork.c:925 >> exit_mm kernel/exit.c:544 [inline] >> do_exit+0x981/0x1b10 kernel/exit.c:852 >> do_group_exit+0x149/0x400 kernel/exit.c:969 >> get_signal+0x7e8/0x17e0 kernel/signal.c:2330 >> do_signal+0x94/0x1ee0 arch/x86/kernel/signal.c:808 >> exit_to_usermode_loop+0x21c/0x2d0 arch/x86/entry/common.c:157 >> prepare_exit_to_usermode arch/x86/entry/common.c:194 [inline] >> syscall_return_slowpath+0x3a7/0x450 arch/x86/entry/common.c:263 >> entry_SYSCALL_64_fastpath+0xbc/0xbe >> RIP: 0033:0x4512c9 >> RSP: 002b:00007f5548015c18 EFLAGS: 00000206 ORIG_RAX: 00000000000000ca >> RAX: 0000000000000001 RBX: 00000000007180a8 RCX: 00000000004512c9 >> RDX: 0000000000000000 RSI: 0000000000000001 RDI: 00000000007180cc >> RBP: 0000000000001fe0 R08: 0000000000000000 R09: 0000000000000000 >> R10: 0000000000000000 R11: 0000000000000206 R12: 00000000004b7f09 >> R13: 00000000ffffffff R14: 0000000000000007 R15: 000000000000ae80 >> Code: 3c 02 00 00 48 c7 c7 20 63 62 84 c6 05 e0 42 f8 03 01 e8 25 de 44 00 >> e9 da fb ff ff e8 6b 7f 5d 00 48 b8 00 00 00 00 00 fc ff df <80> 38 00 0f 85 >> 1b 01 00 00 4c 8b 24 25 00 00 00 00 31 db e9 83 >> RIP: __kvm_memslots include/linux/kvm_host.h:571 [inline] RSP: >> ffff8801ca9b6820 >> RIP: gfn_to_rmap+0x57f/0x6b0 arch/x86/kvm/mmu.c:1235 RSP: ffff8801ca9b6820 >> ---[ end trace 0a16957bec756c04 ]--- Fix bisection says that this is fixed by: commit b7e31be385584afe7f073130e8e570d53c95f7fe Author: Radim Krčmář <rkrcmar@xxxxxxxxxx> Date: Thu Mar 1 15:24:25 2018 +0100 KVM: x86: fix vcpu initialization with userspace lapic #syz fix: KVM: x86: fix vcpu initialization with userspace lapic
