On Mon, Dec 11, 2017 at 07:00:03PM +0800, Wanpeng Li wrote: > 2017-12-09 19:39 GMT+08:00 Tianyu Lan <lantianyu1986@xxxxxxxxx>: > > 2017-12-09 17:15 GMT+08:00 syzbot > > <bot+b19026412c772c86fca485fc8342e3490da9c269@xxxxxxxxxxxxxxxxxxxxxxxxx>: > >> syzkaller has found reproducer for the following crash on > >> ad4dac17f9d563b9e34aab78a34293b10993e9b5 > >> git://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/master > >> compiler: gcc (GCC) 7.1.1 20170620 > >> .config is attached > >> Raw console output is attached. > >> C reproducer is attached > >> syzkaller reproducer is attached. See https://goo.gl/kgGztJ > >> for information about syzkaller reproducers > >> > >> > >> kvm: KVM_SET_TSS_ADDR need to be called before entering vcpu > >> ================================================================== > >> BUG: KASAN: stack-out-of-bounds in write_mmio+0x560/0x600 > >> arch/x86/kvm/x86.c:4669 > >> Read of size 8 at addr ffff8801c4587220 by task syzkaller601256/3159 > >> > >> CPU: 1 PID: 3159 Comm: syzkaller601256 Not tainted 4.15.0-rc2-next-20171208+ > >> #63 > >> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS > >> Google 01/01/2011 > >> Call Trace: > >> __dump_stack lib/dump_stack.c:17 [inline] > >> dump_stack+0x194/0x257 lib/dump_stack.c:53 > >> print_address_description+0x73/0x250 mm/kasan/report.c:252 > >> kasan_report_error mm/kasan/report.c:351 [inline] > >> kasan_report+0x25b/0x340 mm/kasan/report.c:409 > >> __asan_report_load8_noabort+0x14/0x20 mm/kasan/report.c:430 > >> write_mmio+0x560/0x600 arch/x86/kvm/x86.c:4669 > >> emulator_read_write_onepage+0x45a/0xea0 arch/x86/kvm/x86.c:4739 > >> emulator_read_write+0xe7/0x540 arch/x86/kvm/x86.c:4788 > >> emulator_write_emulated arch/x86/kvm/x86.c:4825 [inline] > >> emulator_fix_hypercall+0x14d/0x1b0 arch/x86/kvm/x86.c:6349 > >> em_hypercall+0x5d/0x120 arch/x86/kvm/emulate.c:3705 > >> x86_emulate_insn+0x55d/0x3c20 arch/x86/kvm/emulate.c:5498 > >> x86_emulate_instruction+0x411/0x1ad0 arch/x86/kvm/x86.c:5756 > >> emulate_instruction arch/x86/include/asm/kvm_host.h:1177 [inline] > >> handle_exception+0x3d5/0xa20 arch/x86/kvm/vmx.c:5921 > >> vmx_handle_exit+0x25d/0x1ce0 arch/x86/kvm/vmx.c:8887 > >> vcpu_enter_guest arch/x86/kvm/x86.c:7068 [inline] > >> vcpu_run arch/x86/kvm/x86.c:7130 [inline] > >> kvm_arch_vcpu_ioctl_run+0x1836/0x5be0 arch/x86/kvm/x86.c:7300 > >> kvm_vcpu_ioctl+0x64c/0x1010 arch/x86/kvm/../../../virt/kvm/kvm_main.c:2555 > >> vfs_ioctl fs/ioctl.c:46 [inline] > >> do_vfs_ioctl+0x1b1/0x1530 fs/ioctl.c:686 > >> SYSC_ioctl fs/ioctl.c:701 [inline] > >> SyS_ioctl+0x8f/0xc0 fs/ioctl.c:692 > >> entry_SYSCALL_64_fastpath+0x1f/0x96 > >> RIP: 0033:0x4435c9 > >> RSP: 002b:00007fffdf34ced8 EFLAGS: 00000202 ORIG_RAX: 0000000000000010 > >> RAX: ffffffffffffffda RBX: 00000000205b3000 RCX: 00000000004435c9 > >> RDX: 0000000000000000 RSI: 000000000000ae80 RDI: 0000000000000005 > >> RBP: 000000000000000b R08: 0000000000000000 R09: 0000000000000002 > >> R10: 0000000000000012 R11: 0000000000000202 R12: 00000000205b7e00 > >> R13: 00000000205b6e00 R14: 00000000205b7a00 R15: 00000000205b3000 > >> > >> The buggy address belongs to the page: > >> page:000000002e827842 count:0 mapcount:0 mapping: (null) index:0x0 > >> flags: 0x2fffc0000000000() > >> raw: 02fffc0000000000 0000000000000000 0000000000000000 00000000ffffffff > >> raw: 0000000000000000 0000000100000001 0000000000000000 0000000000000000 > >> page dumped because: kasan: bad access detected > >> > >> Memory state around the buggy address: > >> ffff8801c4587100: f2 f2 f2 f2 00 f2 f2 f2 f3 f3 f3 f3 00 00 00 00 > >> ffff8801c4587180: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > >>> > >>> ffff8801c4587200: f1 f1 f1 f1 03 f2 f2 f2 f3 f3 f3 f3 00 00 00 00 > >> > >> ^ > >> ffff8801c4587280: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > >> ffff8801c4587300: 00 f1 f1 f1 f1 02 f2 f2 f2 f2 f2 f2 f2 00 f2 f2 > >> ================================================================== > >> > > > > This issue seems to be as same as previous pop issue. I will have a look. > > It is the *(u64 *)val in trace_kvm_mmio() causes the > stack-out-of-bounds, I will cook a patch soon. > No longer occurring, presumably fixed by: #syz fix: KVM: Fix stack-out-of-bounds read in write_mmio
