Hello I found a bug in latest upstream 6.7-rc7 titled "KMSAN: uninit-value in em_ret_far” and maybe is realted with kvm. If you fix this issue, please add the following tag to the commit: Reported-by: xingwei lee <xrivendell7@xxxxxxxxx> kernel: mainline 861deac3b092f37b2c5e6871732f3e11486f7082 kernel config: https://syzkaller.appspot.com/text?tag=KernelConfig&x=4a65fa9f077ead01 with KMSAN enabled compiler: Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40 run the repro.c for about 3minus and it crashed! TITLE: KMSAN: uninit-value in em_ret_far ===================================================== BUG: KMSAN: uninit-value in emulator_recalc_and_set_mode arch/x86/kvm/emulate.c:797 [inline] BUG: KMSAN: uninit-value in assign_eip_far arch/x86/kvm/emulate.c:833 [inline] BUG: KMSAN: uninit-value in em_ret_far+0x348/0x350 arch/x86/kvm/emulate.c:2258 emulator_recalc_and_set_mode arch/x86/kvm/emulate.c:797 [inline] assign_eip_far arch/x86/kvm/emulate.c:833 [inline] em_ret_far+0x348/0x350 arch/x86/kvm/emulate.c:2258 em_ret_far_imm arch/x86/kvm/emulate.c:2273 [inline] em_ret_far_imm+0x37/0x510 arch/x86/kvm/emulate.c:2266 string_registers_quirk arch/x86/kvm/emulate.c:2647 [inline] x86_emulate_insn+0x1d59/0x5790 arch/x86/kvm/emulate.c:5229 kvm_vcpu_check_hw_bp arch/x86/kvm/x86.c:8813 [inline] kvm_vcpu_check_code_breakpoint arch/x86/kvm/x86.c:8919 [inline] x86_emulate_instruction+0x182b/0x3070 arch/x86/kvm/x86.c:9017 complete_emulated_mmio+0x6ed/0x890 arch/x86/kvm/x86.c:11199 kvm_queue_exception_vmexit arch/x86/kvm/x86.c:634 [inline] kvm_arch_vcpu_ioctl_run+0x190c/0xc5f0 arch/x86/kvm/x86.c:11293 copy_from_user include/linux/uaccess.h:183 [inline] kvm_vcpu_ioctl+0xc6c/0x1800 arch/x86/kvm/../../../virt/kvm/kvm_main.c:4274 __se_sys_ioctl+0x211/0x400 __do_sys_ioctl fs/ioctl.c:862 [inline] __se_sys_ioctl fs/ioctl.c:857 [inline] __x64_sys_ioctl+0x97/0xe0 fs/ioctl.c:857 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0x44/0x110 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x63/0x6b Local variable eip created at: em_ret_far+0x3a/0x350 arch/x86/kvm/emulate.c:2241 em_ret_far_imm arch/x86/kvm/emulate.c:2273 [inline] em_ret_far_imm+0x37/0x510 arch/x86/kvm/emulate.c:2266 CPU: 3 PID: 15635 Comm: be9 Not tainted 6.7.0-rc7-dirty #9 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.2-1.fc38 04/01/2014 ===================================================== =* repro.c =* #define _GNU_SOURCE #include <dirent.h> #include <endian.h> #include <errno.h> #include <fcntl.h> #include <linux/futex.h> #include <linux/kvm.h> #include <pthread.h> #include <setjmp.h> #include <signal.h> #include <stdarg.h> #include <stdbool.h> #include <stddef.h> #include <stdint.h> #include <stdio.h> #include <stdlib.h> #include <string.h> #include <sys/ioctl.h> #include <sys/prctl.h> #include <sys/stat.h> #include <sys/syscall.h> #include <sys/types.h> #include <sys/wait.h> #include <time.h> #include <unistd.h> #ifndef __NR_memfd_create #define __NR_memfd_create 319 #endif static unsigned long long procid; static __thread int clone_ongoing; static __thread int skip_segv; static __thread jmp_buf segv_env; static void segv_handler(int sig, siginfo_t* info, void* ctx) { if (__atomic_load_n(&clone_ongoing, __ATOMIC_RELAXED) != 0) { exit(sig); } uintptr_t addr = (uintptr_t)info->si_addr; const uintptr_t prog_start = 1 << 20; const uintptr_t prog_end = 100 << 20; int skip = __atomic_load_n(&skip_segv, __ATOMIC_RELAXED) != 0; int valid = addr < prog_start || addr > prog_end; if (skip && valid) { _longjmp(segv_env, 1); } exit(sig); } static void install_segv_handler(void) { struct sigaction sa; memset(&sa, 0, sizeof(sa)); sa.sa_handler = SIG_IGN; syscall(SYS_rt_sigaction, 0x20, &sa, NULL, 8); syscall(SYS_rt_sigaction, 0x21, &sa, NULL, 8); memset(&sa, 0, sizeof(sa)); sa.sa_sigaction = segv_handler; sa.sa_flags = SA_NODEFER | SA_SIGINFO; sigaction(SIGSEGV, &sa, NULL); sigaction(SIGBUS, &sa, NULL); } #define NONFAILING(...) \ ({ \ int ok = 1; \ __atomic_fetch_add(&skip_segv, 1, __ATOMIC_SEQ_CST); \ if (_setjmp(segv_env) == 0) { \ __VA_ARGS__; \ } else \ ok = 0; \ __atomic_fetch_sub(&skip_segv, 1, __ATOMIC_SEQ_CST); \ ok; \ }) static void sleep_ms(uint64_t ms) { usleep(ms * 1000); } static uint64_t current_time_ms(void) { struct timespec ts; if (clock_gettime(CLOCK_MONOTONIC, &ts)) exit(1); return (uint64_t)ts.tv_sec * 1000 + (uint64_t)ts.tv_nsec / 1000000; } static void thread_start(void* (*fn)(void*), void* arg) { pthread_t th; pthread_attr_t attr; pthread_attr_init(&attr); pthread_attr_setstacksize(&attr, 128 << 10); int i = 0; for (; i < 100; i++) { if (pthread_create(&th, &attr, fn, arg) == 0) { pthread_attr_destroy(&attr); return; } if (errno == EAGAIN) { usleep(50); continue; } break; } exit(1); } typedef struct { int state; } event_t; static void event_init(event_t* ev) { ev->state = 0; } static void event_reset(event_t* ev) { ev->state = 0; } static void event_set(event_t* ev) { if (ev->state) exit(1); __atomic_store_n(&ev->state, 1, __ATOMIC_RELEASE); syscall(SYS_futex, &ev->state, FUTEX_WAKE | FUTEX_PRIVATE_FLAG, 1000000); } static void event_wait(event_t* ev) { while (!__atomic_load_n(&ev->state, __ATOMIC_ACQUIRE)) syscall(SYS_futex, &ev->state, FUTEX_WAIT | FUTEX_PRIVATE_FLAG, 0, 0); } static int event_isset(event_t* ev) { return __atomic_load_n(&ev->state, __ATOMIC_ACQUIRE); } static int event_timedwait(event_t* ev, uint64_t timeout) { uint64_t start = current_time_ms(); uint64_t now = start; for (;;) { uint64_t remain = timeout - (now - start); struct timespec ts; ts.tv_sec = remain / 1000; ts.tv_nsec = (remain % 1000) * 1000 * 1000; syscall(SYS_futex, &ev->state, FUTEX_WAIT | FUTEX_PRIVATE_FLAG, 0, &ts); if (__atomic_load_n(&ev->state, __ATOMIC_ACQUIRE)) return 1; now = current_time_ms(); if (now - start > timeout) return 0; } } static bool write_file(const char* file, const char* what, ...) { char buf[1024]; va_list args; va_start(args, what); vsnprintf(buf, sizeof(buf), what, args); va_end(args); buf[sizeof(buf) - 1] = 0; int len = strlen(buf); int fd = open(file, O_WRONLY | O_CLOEXEC); if (fd == -1) return false; if (write(fd, buf, len) != len) { int err = errno; close(fd); errno = err; return false; } close(fd); return true; } #define ADDR_TEXT 0x0000 #define ADDR_GDT 0x1000 #define ADDR_LDT 0x1800 #define ADDR_PML4 0x2000 #define ADDR_PDP 0x3000 #define ADDR_PD 0x4000 #define ADDR_STACK0 0x0f80 #define ADDR_VAR_HLT 0x2800 #define ADDR_VAR_SYSRET 0x2808 #define ADDR_VAR_SYSEXIT 0x2810 #define ADDR_VAR_IDT 0x3800 #define ADDR_VAR_TSS64 0x3a00 #define ADDR_VAR_TSS64_CPL3 0x3c00 #define ADDR_VAR_TSS16 0x3d00 #define ADDR_VAR_TSS16_2 0x3e00 #define ADDR_VAR_TSS16_CPL3 0x3f00 #define ADDR_VAR_TSS32 0x4800 #define ADDR_VAR_TSS32_2 0x4a00 #define ADDR_VAR_TSS32_CPL3 0x4c00 #define ADDR_VAR_TSS32_VM86 0x4e00 #define ADDR_VAR_VMXON_PTR 0x5f00 #define ADDR_VAR_VMCS_PTR 0x5f08 #define ADDR_VAR_VMEXIT_PTR 0x5f10 #define ADDR_VAR_VMWRITE_FLD 0x5f18 #define ADDR_VAR_VMWRITE_VAL 0x5f20 #define ADDR_VAR_VMXON 0x6000 #define ADDR_VAR_VMCS 0x7000 #define ADDR_VAR_VMEXIT_CODE 0x9000 #define ADDR_VAR_USER_CODE 0x9100 #define ADDR_VAR_USER_CODE2 0x9120 #define SEL_LDT (1 << 3) #define SEL_CS16 (2 << 3) #define SEL_DS16 (3 << 3) #define SEL_CS16_CPL3 ((4 << 3) + 3) #define SEL_DS16_CPL3 ((5 << 3) + 3) #define SEL_CS32 (6 << 3) #define SEL_DS32 (7 << 3) #define SEL_CS32_CPL3 ((8 << 3) + 3) #define SEL_DS32_CPL3 ((9 << 3) + 3) #define SEL_CS64 (10 << 3) #define SEL_DS64 (11 << 3) #define SEL_CS64_CPL3 ((12 << 3) + 3) #define SEL_DS64_CPL3 ((13 << 3) + 3) #define SEL_CGATE16 (14 << 3) #define SEL_TGATE16 (15 << 3) #define SEL_CGATE32 (16 << 3) #define SEL_TGATE32 (17 << 3) #define SEL_CGATE64 (18 << 3) #define SEL_CGATE64_HI (19 << 3) #define SEL_TSS16 (20 << 3) #define SEL_TSS16_2 (21 << 3) #define SEL_TSS16_CPL3 ((22 << 3) + 3) #define SEL_TSS32 (23 << 3) #define SEL_TSS32_2 (24 << 3) #define SEL_TSS32_CPL3 ((25 << 3) + 3) #define SEL_TSS32_VM86 (26 << 3) #define SEL_TSS64 (27 << 3) #define SEL_TSS64_HI (28 << 3) #define SEL_TSS64_CPL3 ((29 << 3) + 3) #define SEL_TSS64_CPL3_HI (30 << 3) #define MSR_IA32_FEATURE_CONTROL 0x3a #define MSR_IA32_VMX_BASIC 0x480 #define MSR_IA32_SMBASE 0x9e #define MSR_IA32_SYSENTER_CS 0x174 #define MSR_IA32_SYSENTER_ESP 0x175 #define MSR_IA32_SYSENTER_EIP 0x176 #define MSR_IA32_STAR 0xC0000081 #define MSR_IA32_LSTAR 0xC0000082 #define MSR_IA32_VMX_PROCBASED_CTLS2 0x48B #define NEXT_INSN $0xbadc0de #define PREFIX_SIZE 0xba1d const char kvm_asm16_cpl3[] = "\x0f\x20\xc0\x66\x83\xc8\x01\x0f\x22\xc0\xb8\xa0\x00\x0f\x00\xd8\xb8\x2b" "\x00\x8e\xd8\x8e\xc0\x8e\xe0\x8e\xe8\xbc\x00\x01\xc7\x06\x00\x01\x1d\xba" "\xc7\x06\x02\x01\x23\x00\xc7\x06\x04\x01\x00\x01\xc7\x06\x06\x01\x2b\x00" "\xcb"; const char kvm_asm32_paged[] = "\x0f\x20\xc0\x0d\x00\x00\x00\x80\x0f\x22\xc0"; const char kvm_asm32_vm86[] = "\x66\xb8\xb8\x00\x0f\x00\xd8\xea\x00\x00\x00\x00\xd0\x00"; const char kvm_asm32_paged_vm86[] = "\x0f\x20\xc0\x0d\x00\x00\x00\x80\x0f\x22\xc0\x66\xb8\xb8\x00\x0f\x00\xd8" "\xea\x00\x00\x00\x00\xd0\x00"; const char kvm_asm64_enable_long[] = "\x0f\x20\xc0\x0d\x00\x00\x00\x80\x0f\x22\xc0\xea\xde\xc0\xad\x0b\x50\x00" "\x48\xc7\xc0\xd8\x00\x00\x00\x0f\x00\xd8"; const char kvm_asm64_init_vm[] = "\x0f\x20\xc0\x0d\x00\x00\x00\x80\x0f\x22\xc0\xea\xde\xc0\xad\x0b\x50\x00" "\x48\xc7\xc0\xd8\x00\x00\x00\x0f\x00\xd8\x48\xc7\xc1\x3a\x00\x00\x00\x0f" "\x32\x48\x83\xc8\x05\x0f\x30\x0f\x20\xe0\x48\x0d\x00\x20\x00\x00\x0f\x22" "\xe0\x48\xc7\xc1\x80\x04\x00\x00\x0f\x32\x48\xc7\xc2\x00\x60\x00\x00\x89" "\x02\x48\xc7\xc2\x00\x70\x00\x00\x89\x02\x48\xc7\xc0\x00\x5f\x00\x00\xf3" "\x0f\xc7\x30\x48\xc7\xc0\x08\x5f\x00\x00\x66\x0f\xc7\x30\x0f\xc7\x30\x48" "\xc7\xc1\x81\x04\x00\x00\x0f\x32\x48\x83\xc8\x00\x48\x21\xd0\x48\xc7\xc2" "\x00\x40\x00\x00\x0f\x79\xd0\x48\xc7\xc1\x82\x04\x00\x00\x0f\x32\x48\x83" "\xc8\x00\x48\x21\xd0\x48\xc7\xc2\x02\x40\x00\x00\x0f\x79\xd0\x48\xc7\xc2" "\x1e\x40\x00\x00\x48\xc7\xc0\x81\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc1\x83" "\x04\x00\x00\x0f\x32\x48\x0d\xff\x6f\x03\x00\x48\x21\xd0\x48\xc7\xc2\x0c" "\x40\x00\x00\x0f\x79\xd0\x48\xc7\xc1\x84\x04\x00\x00\x0f\x32\x48\x0d\xff" "\x17\x00\x00\x48\x21\xd0\x48\xc7\xc2\x12\x40\x00\x00\x0f\x79\xd0\x48\xc7" "\xc2\x04\x2c\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2" "\x00\x28\x00\x00\x48\xc7\xc0\xff\xff\xff\xff\x0f\x79\xd0\x48\xc7\xc2\x02" "\x0c\x00\x00\x48\xc7\xc0\x50\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc0\x58\x00" "\x00\x00\x48\xc7\xc2\x00\x0c\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x04\x0c\x00" "\x00\x0f\x79\xd0\x48\xc7\xc2\x06\x0c\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x08" "\x0c\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x0a\x0c\x00\x00\x0f\x79\xd0\x48\xc7" "\xc0\xd8\x00\x00\x00\x48\xc7\xc2\x0c\x0c\x00\x00\x0f\x79\xd0\x48\xc7\xc2" "\x02\x2c\x00\x00\x48\xc7\xc0\x00\x05\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x00" "\x4c\x00\x00\x48\xc7\xc0\x50\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x10\x6c" "\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x12\x6c\x00" "\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x0f\x20\xc0\x48\xc7\xc2\x00" "\x6c\x00\x00\x48\x89\xc0\x0f\x79\xd0\x0f\x20\xd8\x48\xc7\xc2\x02\x6c\x00" "\x00\x48\x89\xc0\x0f\x79\xd0\x0f\x20\xe0\x48\xc7\xc2\x04\x6c\x00\x00\x48" "\x89\xc0\x0f\x79\xd0\x48\xc7\xc2\x06\x6c\x00\x00\x48\xc7\xc0\x00\x00\x00" "\x00\x0f\x79\xd0\x48\xc7\xc2\x08\x6c\x00\x00\x48\xc7\xc0\x00\x00\x00\x00" "\x0f\x79\xd0\x48\xc7\xc2\x0a\x6c\x00\x00\x48\xc7\xc0\x00\x3a\x00\x00\x0f" "\x79\xd0\x48\xc7\xc2\x0c\x6c\x00\x00\x48\xc7\xc0\x00\x10\x00\x00\x0f\x79" "\xd0\x48\xc7\xc2\x0e\x6c\x00\x00\x48\xc7\xc0\x00\x38\x00\x00\x0f\x79\xd0" "\x48\xc7\xc2\x14\x6c\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48" "\xc7\xc2\x16\x6c\x00\x00\x48\x8b\x04\x25\x10\x5f\x00\x00\x0f\x79\xd0\x48" "\xc7\xc2\x00\x00\x00\x00\x48\xc7\xc0\x01\x00\x00\x00\x0f\x79\xd0\x48\xc7" "\xc2\x02\x00\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2" "\x00\x20\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x02" "\x20\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x04\x20" "\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x06\x20\x00" "\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc1\x77\x02\x00\x00" "\x0f\x32\x48\xc1\xe2\x20\x48\x09\xd0\x48\xc7\xc2\x00\x2c\x00\x00\x48\x89" "\xc0\x0f\x79\xd0\x48\xc7\xc2\x04\x40\x00\x00\x48\xc7\xc0\x00\x00\x00\x00" "\x0f\x79\xd0\x48\xc7\xc2\x0a\x40\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f" "\x79\xd0\x48\xc7\xc2\x0e\x40\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79" "\xd0\x48\xc7\xc2\x10\x40\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0" "\x48\xc7\xc2\x16\x40\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48" "\xc7\xc2\x14\x40\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7" "\xc2\x00\x60\x00\x00\x48\xc7\xc0\xff\xff\xff\xff\x0f\x79\xd0\x48\xc7\xc2" "\x02\x60\x00\x00\x48\xc7\xc0\xff\xff\xff\xff\x0f\x79\xd0\x48\xc7\xc2\x1c" "\x20\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x1e\x20" "\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x20\x20\x00" "\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x22\x20\x00\x00" "\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x00\x08\x00\x00\x48" "\xc7\xc0\x58\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x02\x08\x00\x00\x48\xc7" "\xc0\x50\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x04\x08\x00\x00\x48\xc7\xc0" "\x58\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x06\x08\x00\x00\x48\xc7\xc0\x58" "\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x08\x08\x00\x00\x48\xc7\xc0\x58\x00" "\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x0a\x08\x00\x00\x48\xc7\xc0\x58\x00\x00" "\x00\x0f\x79\xd0\x48\xc7\xc2\x0c\x08\x00\x00\x48\xc7\xc0\x00\x00\x00\x00" "\x0f\x79\xd0\x48\xc7\xc2\x0e\x08\x00\x00\x48\xc7\xc0\xd8\x00\x00\x00\x0f" "\x79\xd0\x48\xc7\xc2\x12\x68\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79" "\xd0\x48\xc7\xc2\x14\x68\x00\x00\x48\xc7\xc0\x00\x3a\x00\x00\x0f\x79\xd0" "\x48\xc7\xc2\x16\x68\x00\x00\x48\xc7\xc0\x00\x10\x00\x00\x0f\x79\xd0\x48" "\xc7\xc2\x18\x68\x00\x00\x48\xc7\xc0\x00\x38\x00\x00\x0f\x79\xd0\x48\xc7" "\xc2\x00\x48\x00\x00\x48\xc7\xc0\xff\xff\x0f\x00\x0f\x79\xd0\x48\xc7\xc2" "\x02\x48\x00\x00\x48\xc7\xc0\xff\xff\x0f\x00\x0f\x79\xd0\x48\xc7\xc2\x04" "\x48\x00\x00\x48\xc7\xc0\xff\xff\x0f\x00\x0f\x79\xd0\x48\xc7\xc2\x06\x48" "\x00\x00\x48\xc7\xc0\xff\xff\x0f\x00\x0f\x79\xd0\x48\xc7\xc2\x08\x48\x00" "\x00\x48\xc7\xc0\xff\xff\x0f\x00\x0f\x79\xd0\x48\xc7\xc2\x0a\x48\x00\x00" "\x48\xc7\xc0\xff\xff\x0f\x00\x0f\x79\xd0\x48\xc7\xc2\x0c\x48\x00\x00\x48" "\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x0e\x48\x00\x00\x48\xc7" "\xc0\xff\x1f\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x10\x48\x00\x00\x48\xc7\xc0" "\xff\x1f\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x12\x48\x00\x00\x48\xc7\xc0\xff" "\x1f\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x14\x48\x00\x00\x48\xc7\xc0\x93\x40" "\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x16\x48\x00\x00\x48\xc7\xc0\x9b\x20\x00" "\x00\x0f\x79\xd0\x48\xc7\xc2\x18\x48\x00\x00\x48\xc7\xc0\x93\x40\x00\x00" "\x0f\x79\xd0\x48\xc7\xc2\x1a\x48\x00\x00\x48\xc7\xc0\x93\x40\x00\x00\x0f" "\x79\xd0\x48\xc7\xc2\x1c\x48\x00\x00\x48\xc7\xc0\x93\x40\x00\x00\x0f\x79" "\xd0\x48\xc7\xc2\x1e\x48\x00\x00\x48\xc7\xc0\x93\x40\x00\x00\x0f\x79\xd0" "\x48\xc7\xc2\x20\x48\x00\x00\x48\xc7\xc0\x82\x00\x00\x00\x0f\x79\xd0\x48" "\xc7\xc2\x22\x48\x00\x00\x48\xc7\xc0\x8b\x00\x00\x00\x0f\x79\xd0\x48\xc7" "\xc2\x1c\x68\x00\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2" "\x1e\x68\x00\x00\x48\xc7\xc0\x00\x91\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x20" "\x68\x00\x00\x48\xc7\xc0\x02\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x06\x28" "\x00\x00\x48\xc7\xc0\x00\x05\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x0a\x28\x00" "\x00\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x0c\x28\x00\x00" "\x48\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x0e\x28\x00\x00\x48" "\xc7\xc0\x00\x00\x00\x00\x0f\x79\xd0\x48\xc7\xc2\x10\x28\x00\x00\x48\xc7" "\xc0\x00\x00\x00\x00\x0f\x79\xd0\x0f\x20\xc0\x48\xc7\xc2\x00\x68\x00\x00" "\x48\x89\xc0\x0f\x79\xd0\x0f\x20\xd8\x48\xc7\xc2\x02\x68\x00\x00\x48\x89" "\xc0\x0f\x79\xd0\x0f\x20\xe0\x48\xc7\xc2\x04\x68\x00\x00\x48\x89\xc0\x0f" "\x79\xd0\x48\xc7\xc0\x18\x5f\x00\x00\x48\x8b\x10\x48\xc7\xc0\x20\x5f\x00" "\x00\x48\x8b\x08\x48\x31\xc0\x0f\x78\xd0\x48\x31\xc8\x0f\x79\xd0\x0f\x01" "\xc2\x48\xc7\xc2\x00\x44\x00\x00\x0f\x78\xd0\xf4"; const char kvm_asm64_vm_exit[] = "\x48\xc7\xc3\x00\x44\x00\x00\x0f\x78\xda\x48\xc7\xc3\x02\x44\x00\x00\x0f" "\x78\xd9\x48\xc7\xc0\x00\x64\x00\x00\x0f\x78\xc0\x48\xc7\xc3\x1e\x68\x00" "\x00\x0f\x78\xdb\xf4"; const char kvm_asm64_cpl3[] = "\x0f\x20\xc0\x0d\x00\x00\x00\x80\x0f\x22\xc0\xea\xde\xc0\xad\x0b\x50\x00" "\x48\xc7\xc0\xd8\x00\x00\x00\x0f\x00\xd8\x48\xc7\xc0\x6b\x00\x00\x00\x8e" "\xd8\x8e\xc0\x8e\xe0\x8e\xe8\x48\xc7\xc4\x80\x0f\x00\x00\x48\xc7\x04\x24" "\x1d\xba\x00\x00\x48\xc7\x44\x24\x04\x63\x00\x00\x00\x48\xc7\x44\x24\x08" "\x80\x0f\x00\x00\x48\xc7\x44\x24\x0c\x6b\x00\x00\x00\xcb"; #define KVM_SMI _IO(KVMIO, 0xb7) #define CR0_PE 1 #define CR0_MP (1 << 1) #define CR0_EM (1 << 2) #define CR0_TS (1 << 3) #define CR0_ET (1 << 4) #define CR0_NE (1 << 5) #define CR0_WP (1 << 16) #define CR0_AM (1 << 18) #define CR0_NW (1 << 29) #define CR0_CD (1 << 30) #define CR0_PG (1 << 31) #define CR4_VME 1 #define CR4_PVI (1 << 1) #define CR4_TSD (1 << 2) #define CR4_DE (1 << 3) #define CR4_PSE (1 << 4) #define CR4_PAE (1 << 5) #define CR4_MCE (1 << 6) #define CR4_PGE (1 << 7) #define CR4_PCE (1 << 8) #define CR4_OSFXSR (1 << 8) #define CR4_OSXMMEXCPT (1 << 10) #define CR4_UMIP (1 << 11) #define CR4_VMXE (1 << 13) #define CR4_SMXE (1 << 14) #define CR4_FSGSBASE (1 << 16) #define CR4_PCIDE (1 << 17) #define CR4_OSXSAVE (1 << 18) #define CR4_SMEP (1 << 20) #define CR4_SMAP (1 << 21) #define CR4_PKE (1 << 22) #define EFER_SCE 1 #define EFER_LME (1 << 8) #define EFER_LMA (1 << 10) #define EFER_NXE (1 << 11) #define EFER_SVME (1 << 12) #define EFER_LMSLE (1 << 13) #define EFER_FFXSR (1 << 14) #define EFER_TCE (1 << 15) #define PDE32_PRESENT 1 #define PDE32_RW (1 << 1) #define PDE32_USER (1 << 2) #define PDE32_PS (1 << 7) #define PDE64_PRESENT 1 #define PDE64_RW (1 << 1) #define PDE64_USER (1 << 2) #define PDE64_ACCESSED (1 << 5) #define PDE64_DIRTY (1 << 6) #define PDE64_PS (1 << 7) #define PDE64_G (1 << 8) struct tss16 { uint16_t prev; uint16_t sp0; uint16_t ss0; uint16_t sp1; uint16_t ss1; uint16_t sp2; uint16_t ss2; uint16_t ip; uint16_t flags; uint16_t ax; uint16_t cx; uint16_t dx; uint16_t bx; uint16_t sp; uint16_t bp; uint16_t si; uint16_t di; uint16_t es; uint16_t cs; uint16_t ss; uint16_t ds; uint16_t ldt; } __attribute__((packed)); struct tss32 { uint16_t prev, prevh; uint32_t sp0; uint16_t ss0, ss0h; uint32_t sp1; uint16_t ss1, ss1h; uint32_t sp2; uint16_t ss2, ss2h; uint32_t cr3; uint32_t ip; uint32_t flags; uint32_t ax; uint32_t cx; uint32_t dx; uint32_t bx; uint32_t sp; uint32_t bp; uint32_t si; uint32_t di; uint16_t es, esh; uint16_t cs, csh; uint16_t ss, ssh; uint16_t ds, dsh; uint16_t fs, fsh; uint16_t gs, gsh; uint16_t ldt, ldth; uint16_t trace; uint16_t io_bitmap; } __attribute__((packed)); struct tss64 { uint32_t reserved0; uint64_t rsp[3]; uint64_t reserved1; uint64_t ist[7]; uint64_t reserved2; uint32_t reserved3; uint32_t io_bitmap; } __attribute__((packed)); static void fill_segment_descriptor(uint64_t* dt, uint64_t* lt, struct kvm_segment* seg) { uint16_t index = seg->selector >> 3; uint64_t limit = seg->g ? seg->limit >> 12 : seg->limit; uint64_t sd = (limit & 0xffff) | (seg->base & 0xffffff) << 16 | (uint64_t)seg->type << 40 | (uint64_t)seg->s << 44 | (uint64_t)seg->dpl << 45 | (uint64_t)seg->present << 47 | (limit & 0xf0000ULL) << 48 | (uint64_t)seg->avl << 52 | (uint64_t)seg->l << 53 | (uint64_t)seg->db << 54 | (uint64_t)seg->g << 55 | (seg->base & 0xff000000ULL) << 56; dt[index] = sd; lt[index] = sd; } static void fill_segment_descriptor_dword(uint64_t* dt, uint64_t* lt, struct kvm_segment* seg) { fill_segment_descriptor(dt, lt, seg); uint16_t index = seg->selector >> 3; dt[index + 1] = 0; lt[index + 1] = 0; } static void setup_syscall_msrs(int cpufd, uint16_t sel_cs, uint16_t sel_cs_cpl3) { char buf[sizeof(struct kvm_msrs) + 5 * sizeof(struct kvm_msr_entry)]; memset(buf, 0, sizeof(buf)); struct kvm_msrs* msrs = (struct kvm_msrs*)buf; struct kvm_msr_entry* entries = msrs->entries; msrs->nmsrs = 5; entries[0].index = MSR_IA32_SYSENTER_CS; entries[0].data = sel_cs; entries[1].index = MSR_IA32_SYSENTER_ESP; entries[1].data = ADDR_STACK0; entries[2].index = MSR_IA32_SYSENTER_EIP; entries[2].data = ADDR_VAR_SYSEXIT; entries[3].index = MSR_IA32_STAR; entries[3].data = ((uint64_t)sel_cs << 32) | ((uint64_t)sel_cs_cpl3 << 48); entries[4].index = MSR_IA32_LSTAR; entries[4].data = ADDR_VAR_SYSRET; ioctl(cpufd, KVM_SET_MSRS, msrs); } static void setup_32bit_idt(struct kvm_sregs* sregs, char* host_mem, uintptr_t guest_mem) { sregs->idt.base = guest_mem + ADDR_VAR_IDT; sregs->idt.limit = 0x1ff; uint64_t* idt = (uint64_t*)(host_mem + sregs->idt.base); for (int i = 0; i < 32; i++) { struct kvm_segment gate; gate.selector = i << 3; switch (i % 6) { case 0: gate.type = 6; gate.base = SEL_CS16; break; case 1: gate.type = 7; gate.base = SEL_CS16; break; case 2: gate.type = 3; gate.base = SEL_TGATE16; break; case 3: gate.type = 14; gate.base = SEL_CS32; break; case 4: gate.type = 15; gate.base = SEL_CS32; break; case 5: gate.type = 11; gate.base = SEL_TGATE32; break; } gate.limit = guest_mem + ADDR_VAR_USER_CODE2; gate.present = 1; gate.dpl = 0; gate.s = 0; gate.g = 0; gate.db = 0; gate.l = 0; gate.avl = 0; fill_segment_descriptor(idt, idt, &gate); } } static void setup_64bit_idt(struct kvm_sregs* sregs, char* host_mem, uintptr_t guest_mem) { sregs->idt.base = guest_mem + ADDR_VAR_IDT; sregs->idt.limit = 0x1ff; uint64_t* idt = (uint64_t*)(host_mem + sregs->idt.base); for (int i = 0; i < 32; i++) { struct kvm_segment gate; gate.selector = (i * 2) << 3; gate.type = (i & 1) ? 14 : 15; gate.base = SEL_CS64; gate.limit = guest_mem + ADDR_VAR_USER_CODE2; gate.present = 1; gate.dpl = 0; gate.s = 0; gate.g = 0; gate.db = 0; gate.l = 0; gate.avl = 0; fill_segment_descriptor_dword(idt, idt, &gate); } } struct kvm_text { uintptr_t typ; const void* text; uintptr_t size; }; struct kvm_opt { uint64_t typ; uint64_t val; }; #define KVM_SETUP_PAGING (1 << 0) #define KVM_SETUP_PAE (1 << 1) #define KVM_SETUP_PROTECTED (1 << 2) #define KVM_SETUP_CPL3 (1 << 3) #define KVM_SETUP_VIRT86 (1 << 4) #define KVM_SETUP_SMM (1 << 5) #define KVM_SETUP_VM (1 << 6) static volatile long syz_kvm_setup_cpu(volatile long a0, volatile long a1, volatile long a2, volatile long a3, volatile long a4, volatile long a5, volatile long a6, volatile long a7) { const int vmfd = a0; const int cpufd = a1; char* const host_mem = (char*)a2; const struct kvm_text* const text_array_ptr = (struct kvm_text*)a3; const uintptr_t text_count = a4; const uintptr_t flags = a5; const struct kvm_opt* const opt_array_ptr = (struct kvm_opt*)a6; uintptr_t opt_count = a7; const uintptr_t page_size = 4 << 10; const uintptr_t ioapic_page = 10; const uintptr_t guest_mem_size = 24 * page_size; const uintptr_t guest_mem = 0; (void)text_count; int text_type = text_array_ptr[0].typ; const void* text = text_array_ptr[0].text; uintptr_t text_size = text_array_ptr[0].size; for (uintptr_t i = 0; i < guest_mem_size / page_size; i++) { struct kvm_userspace_memory_region memreg; memreg.slot = i; memreg.flags = 0; memreg.guest_phys_addr = guest_mem + i * page_size; if (i == ioapic_page) memreg.guest_phys_addr = 0xfec00000; memreg.memory_size = page_size; memreg.userspace_addr = (uintptr_t)host_mem + i * page_size; ioctl(vmfd, KVM_SET_USER_MEMORY_REGION, &memreg); } struct kvm_userspace_memory_region memreg; memreg.slot = 1 + (1 << 16); memreg.flags = 0; memreg.guest_phys_addr = 0x30000; memreg.memory_size = 64 << 10; memreg.userspace_addr = (uintptr_t)host_mem; ioctl(vmfd, KVM_SET_USER_MEMORY_REGION, &memreg); struct kvm_sregs sregs; if (ioctl(cpufd, KVM_GET_SREGS, &sregs)) return -1; struct kvm_regs regs; memset(®s, 0, sizeof(regs)); regs.rip = guest_mem + ADDR_TEXT; regs.rsp = ADDR_STACK0; sregs.gdt.base = guest_mem + ADDR_GDT; sregs.gdt.limit = 256 * sizeof(uint64_t) - 1; uint64_t* gdt = (uint64_t*)(host_mem + sregs.gdt.base); struct kvm_segment seg_ldt; seg_ldt.selector = SEL_LDT; seg_ldt.type = 2; seg_ldt.base = guest_mem + ADDR_LDT; seg_ldt.limit = 256 * sizeof(uint64_t) - 1; seg_ldt.present = 1; seg_ldt.dpl = 0; seg_ldt.s = 0; seg_ldt.g = 0; seg_ldt.db = 1; seg_ldt.l = 0; sregs.ldt = seg_ldt; uint64_t* ldt = (uint64_t*)(host_mem + sregs.ldt.base); struct kvm_segment seg_cs16; seg_cs16.selector = SEL_CS16; seg_cs16.type = 11; seg_cs16.base = 0; seg_cs16.limit = 0xfffff; seg_cs16.present = 1; seg_cs16.dpl = 0; seg_cs16.s = 1; seg_cs16.g = 0; seg_cs16.db = 0; seg_cs16.l = 0; struct kvm_segment seg_ds16 = seg_cs16; seg_ds16.selector = SEL_DS16; seg_ds16.type = 3; struct kvm_segment seg_cs16_cpl3 = seg_cs16; seg_cs16_cpl3.selector = SEL_CS16_CPL3; seg_cs16_cpl3.dpl = 3; struct kvm_segment seg_ds16_cpl3 = seg_ds16; seg_ds16_cpl3.selector = SEL_DS16_CPL3; seg_ds16_cpl3.dpl = 3; struct kvm_segment seg_cs32 = seg_cs16; seg_cs32.selector = SEL_CS32; seg_cs32.db = 1; struct kvm_segment seg_ds32 = seg_ds16; seg_ds32.selector = SEL_DS32; seg_ds32.db = 1; struct kvm_segment seg_cs32_cpl3 = seg_cs32; seg_cs32_cpl3.selector = SEL_CS32_CPL3; seg_cs32_cpl3.dpl = 3; struct kvm_segment seg_ds32_cpl3 = seg_ds32; seg_ds32_cpl3.selector = SEL_DS32_CPL3; seg_ds32_cpl3.dpl = 3; struct kvm_segment seg_cs64 = seg_cs16; seg_cs64.selector = SEL_CS64; seg_cs64.l = 1; struct kvm_segment seg_ds64 = seg_ds32; seg_ds64.selector = SEL_DS64; struct kvm_segment seg_cs64_cpl3 = seg_cs64; seg_cs64_cpl3.selector = SEL_CS64_CPL3; seg_cs64_cpl3.dpl = 3; struct kvm_segment seg_ds64_cpl3 = seg_ds64; seg_ds64_cpl3.selector = SEL_DS64_CPL3; seg_ds64_cpl3.dpl = 3; struct kvm_segment seg_tss32; seg_tss32.selector = SEL_TSS32; seg_tss32.type = 9; seg_tss32.base = ADDR_VAR_TSS32; seg_tss32.limit = 0x1ff; seg_tss32.present = 1; seg_tss32.dpl = 0; seg_tss32.s = 0; seg_tss32.g = 0; seg_tss32.db = 0; seg_tss32.l = 0; struct kvm_segment seg_tss32_2 = seg_tss32; seg_tss32_2.selector = SEL_TSS32_2; seg_tss32_2.base = ADDR_VAR_TSS32_2; struct kvm_segment seg_tss32_cpl3 = seg_tss32; seg_tss32_cpl3.selector = SEL_TSS32_CPL3; seg_tss32_cpl3.base = ADDR_VAR_TSS32_CPL3; struct kvm_segment seg_tss32_vm86 = seg_tss32; seg_tss32_vm86.selector = SEL_TSS32_VM86; seg_tss32_vm86.base = ADDR_VAR_TSS32_VM86; struct kvm_segment seg_tss16 = seg_tss32; seg_tss16.selector = SEL_TSS16; seg_tss16.base = ADDR_VAR_TSS16; seg_tss16.limit = 0xff; seg_tss16.type = 1; struct kvm_segment seg_tss16_2 = seg_tss16; seg_tss16_2.selector = SEL_TSS16_2; seg_tss16_2.base = ADDR_VAR_TSS16_2; seg_tss16_2.dpl = 0; struct kvm_segment seg_tss16_cpl3 = seg_tss16; seg_tss16_cpl3.selector = SEL_TSS16_CPL3; seg_tss16_cpl3.base = ADDR_VAR_TSS16_CPL3; seg_tss16_cpl3.dpl = 3; struct kvm_segment seg_tss64 = seg_tss32; seg_tss64.selector = SEL_TSS64; seg_tss64.base = ADDR_VAR_TSS64; seg_tss64.limit = 0x1ff; struct kvm_segment seg_tss64_cpl3 = seg_tss64; seg_tss64_cpl3.selector = SEL_TSS64_CPL3; seg_tss64_cpl3.base = ADDR_VAR_TSS64_CPL3; seg_tss64_cpl3.dpl = 3; struct kvm_segment seg_cgate16; seg_cgate16.selector = SEL_CGATE16; seg_cgate16.type = 4; seg_cgate16.base = SEL_CS16 | (2 << 16); seg_cgate16.limit = ADDR_VAR_USER_CODE2; seg_cgate16.present = 1; seg_cgate16.dpl = 0; seg_cgate16.s = 0; seg_cgate16.g = 0; seg_cgate16.db = 0; seg_cgate16.l = 0; seg_cgate16.avl = 0; struct kvm_segment seg_tgate16 = seg_cgate16; seg_tgate16.selector = SEL_TGATE16; seg_tgate16.type = 3; seg_cgate16.base = SEL_TSS16_2; seg_tgate16.limit = 0; struct kvm_segment seg_cgate32 = seg_cgate16; seg_cgate32.selector = SEL_CGATE32; seg_cgate32.type = 12; seg_cgate32.base = SEL_CS32 | (2 << 16); struct kvm_segment seg_tgate32 = seg_cgate32; seg_tgate32.selector = SEL_TGATE32; seg_tgate32.type = 11; seg_tgate32.base = SEL_TSS32_2; seg_tgate32.limit = 0; struct kvm_segment seg_cgate64 = seg_cgate16; seg_cgate64.selector = SEL_CGATE64; seg_cgate64.type = 12; seg_cgate64.base = SEL_CS64; int kvmfd = open("/dev/kvm", O_RDWR); char buf[sizeof(struct kvm_cpuid2) + 128 * sizeof(struct kvm_cpuid_entry2)]; memset(buf, 0, sizeof(buf)); struct kvm_cpuid2* cpuid = (struct kvm_cpuid2*)buf; cpuid->nent = 128; ioctl(kvmfd, KVM_GET_SUPPORTED_CPUID, cpuid); ioctl(cpufd, KVM_SET_CPUID2, cpuid); close(kvmfd); const char* text_prefix = 0; int text_prefix_size = 0; char* host_text = host_mem + ADDR_TEXT; if (text_type == 8) { if (flags & KVM_SETUP_SMM) { if (flags & KVM_SETUP_PROTECTED) { sregs.cs = seg_cs16; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds16; sregs.cr0 |= CR0_PE; } else { sregs.cs.selector = 0; sregs.cs.base = 0; } *(host_mem + ADDR_TEXT) = 0xf4; host_text = host_mem + 0x8000; ioctl(cpufd, KVM_SMI, 0); } else if (flags & KVM_SETUP_VIRT86) { sregs.cs = seg_cs32; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds32; sregs.cr0 |= CR0_PE; sregs.efer |= EFER_SCE; setup_syscall_msrs(cpufd, SEL_CS32, SEL_CS32_CPL3); setup_32bit_idt(&sregs, host_mem, guest_mem); if (flags & KVM_SETUP_PAGING) { uint64_t pd_addr = guest_mem + ADDR_PD; uint64_t* pd = (uint64_t*)(host_mem + ADDR_PD); pd[0] = PDE32_PRESENT | PDE32_RW | PDE32_USER | PDE32_PS; sregs.cr3 = pd_addr; sregs.cr4 |= CR4_PSE; text_prefix = kvm_asm32_paged_vm86; text_prefix_size = sizeof(kvm_asm32_paged_vm86) - 1; } else { text_prefix = kvm_asm32_vm86; text_prefix_size = sizeof(kvm_asm32_vm86) - 1; } } else { sregs.cs.selector = 0; sregs.cs.base = 0; } } else if (text_type == 16) { if (flags & KVM_SETUP_CPL3) { sregs.cs = seg_cs16; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds16; text_prefix = kvm_asm16_cpl3; text_prefix_size = sizeof(kvm_asm16_cpl3) - 1; } else { sregs.cr0 |= CR0_PE; sregs.cs = seg_cs16; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds16; } } else if (text_type == 32) { sregs.cr0 |= CR0_PE; sregs.efer |= EFER_SCE; setup_syscall_msrs(cpufd, SEL_CS32, SEL_CS32_CPL3); setup_32bit_idt(&sregs, host_mem, guest_mem); if (flags & KVM_SETUP_SMM) { sregs.cs = seg_cs32; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds32; *(host_mem + ADDR_TEXT) = 0xf4; host_text = host_mem + 0x8000; ioctl(cpufd, KVM_SMI, 0); } else if (flags & KVM_SETUP_PAGING) { sregs.cs = seg_cs32; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds32; uint64_t pd_addr = guest_mem + ADDR_PD; uint64_t* pd = (uint64_t*)(host_mem + ADDR_PD); pd[0] = PDE32_PRESENT | PDE32_RW | PDE32_USER | PDE32_PS; sregs.cr3 = pd_addr; sregs.cr4 |= CR4_PSE; text_prefix = kvm_asm32_paged; text_prefix_size = sizeof(kvm_asm32_paged) - 1; } else if (flags & KVM_SETUP_CPL3) { sregs.cs = seg_cs32_cpl3; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds32_cpl3; } else { sregs.cs = seg_cs32; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds32; } } else { sregs.efer |= EFER_LME | EFER_SCE; sregs.cr0 |= CR0_PE; setup_syscall_msrs(cpufd, SEL_CS64, SEL_CS64_CPL3); setup_64bit_idt(&sregs, host_mem, guest_mem); sregs.cs = seg_cs32; sregs.ds = sregs.es = sregs.fs = sregs.gs = sregs.ss = seg_ds32; uint64_t pml4_addr = guest_mem + ADDR_PML4; uint64_t* pml4 = (uint64_t*)(host_mem + ADDR_PML4); uint64_t pdpt_addr = guest_mem + ADDR_PDP; uint64_t* pdpt = (uint64_t*)(host_mem + ADDR_PDP); uint64_t pd_addr = guest_mem + ADDR_PD; uint64_t* pd = (uint64_t*)(host_mem + ADDR_PD); pml4[0] = PDE64_PRESENT | PDE64_RW | PDE64_USER | pdpt_addr; pdpt[0] = PDE64_PRESENT | PDE64_RW | PDE64_USER | pd_addr; pd[0] = PDE64_PRESENT | PDE64_RW | PDE64_USER | PDE64_PS; sregs.cr3 = pml4_addr; sregs.cr4 |= CR4_PAE; if (flags & KVM_SETUP_VM) { sregs.cr0 |= CR0_NE; *((uint64_t*)(host_mem + ADDR_VAR_VMXON_PTR)) = ADDR_VAR_VMXON; *((uint64_t*)(host_mem + ADDR_VAR_VMCS_PTR)) = ADDR_VAR_VMCS; memcpy(host_mem + ADDR_VAR_VMEXIT_CODE, kvm_asm64_vm_exit, sizeof(kvm_asm64_vm_exit) - 1); *((uint64_t*)(host_mem + ADDR_VAR_VMEXIT_PTR)) = ADDR_VAR_VMEXIT_CODE; text_prefix = kvm_asm64_init_vm; text_prefix_size = sizeof(kvm_asm64_init_vm) - 1; } else if (flags & KVM_SETUP_CPL3) { text_prefix = kvm_asm64_cpl3; text_prefix_size = sizeof(kvm_asm64_cpl3) - 1; } else { text_prefix = kvm_asm64_enable_long; text_prefix_size = sizeof(kvm_asm64_enable_long) - 1; } } struct tss16 tss16; memset(&tss16, 0, sizeof(tss16)); tss16.ss0 = tss16.ss1 = tss16.ss2 = SEL_DS16; tss16.sp0 = tss16.sp1 = tss16.sp2 = ADDR_STACK0; tss16.ip = ADDR_VAR_USER_CODE2; tss16.flags = (1 << 1); tss16.cs = SEL_CS16; tss16.es = tss16.ds = tss16.ss = SEL_DS16; tss16.ldt = SEL_LDT; struct tss16* tss16_addr = (struct tss16*)(host_mem + seg_tss16_2.base); memcpy(tss16_addr, &tss16, sizeof(tss16)); memset(&tss16, 0, sizeof(tss16)); tss16.ss0 = tss16.ss1 = tss16.ss2 = SEL_DS16; tss16.sp0 = tss16.sp1 = tss16.sp2 = ADDR_STACK0; tss16.ip = ADDR_VAR_USER_CODE2; tss16.flags = (1 << 1); tss16.cs = SEL_CS16_CPL3; tss16.es = tss16.ds = tss16.ss = SEL_DS16_CPL3; tss16.ldt = SEL_LDT; struct tss16* tss16_cpl3_addr = (struct tss16*)(host_mem + seg_tss16_cpl3.base); memcpy(tss16_cpl3_addr, &tss16, sizeof(tss16)); struct tss32 tss32; memset(&tss32, 0, sizeof(tss32)); tss32.ss0 = tss32.ss1 = tss32.ss2 = SEL_DS32; tss32.sp0 = tss32.sp1 = tss32.sp2 = ADDR_STACK0; tss32.ip = ADDR_VAR_USER_CODE; tss32.flags = (1 << 1) | (1 << 17); tss32.ldt = SEL_LDT; tss32.cr3 = sregs.cr3; tss32.io_bitmap = offsetof(struct tss32, io_bitmap); struct tss32* tss32_addr = (struct tss32*)(host_mem + seg_tss32_vm86.base); memcpy(tss32_addr, &tss32, sizeof(tss32)); memset(&tss32, 0, sizeof(tss32)); tss32.ss0 = tss32.ss1 = tss32.ss2 = SEL_DS32; tss32.sp0 = tss32.sp1 = tss32.sp2 = ADDR_STACK0; tss32.ip = ADDR_VAR_USER_CODE; tss32.flags = (1 << 1); tss32.cr3 = sregs.cr3; tss32.es = tss32.ds = tss32.ss = tss32.gs = tss32.fs = SEL_DS32; tss32.cs = SEL_CS32; tss32.ldt = SEL_LDT; tss32.cr3 = sregs.cr3; tss32.io_bitmap = offsetof(struct tss32, io_bitmap); struct tss32* tss32_cpl3_addr = (struct tss32*)(host_mem + seg_tss32_2.base); memcpy(tss32_cpl3_addr, &tss32, sizeof(tss32)); struct tss64 tss64; memset(&tss64, 0, sizeof(tss64)); tss64.rsp[0] = ADDR_STACK0; tss64.rsp[1] = ADDR_STACK0; tss64.rsp[2] = ADDR_STACK0; tss64.io_bitmap = offsetof(struct tss64, io_bitmap); struct tss64* tss64_addr = (struct tss64*)(host_mem + seg_tss64.base); memcpy(tss64_addr, &tss64, sizeof(tss64)); memset(&tss64, 0, sizeof(tss64)); tss64.rsp[0] = ADDR_STACK0; tss64.rsp[1] = ADDR_STACK0; tss64.rsp[2] = ADDR_STACK0; tss64.io_bitmap = offsetof(struct tss64, io_bitmap); struct tss64* tss64_cpl3_addr = (struct tss64*)(host_mem + seg_tss64_cpl3.base); memcpy(tss64_cpl3_addr, &tss64, sizeof(tss64)); if (text_size > 1000) text_size = 1000; if (text_prefix) { memcpy(host_text, text_prefix, text_prefix_size); void* patch = memmem(host_text, text_prefix_size, "\xde\xc0\xad\x0b", 4); if (patch) *((uint32_t*)patch) = guest_mem + ADDR_TEXT + ((char*)patch - host_text) + 6; uint16_t magic = PREFIX_SIZE; patch = memmem(host_text, text_prefix_size, &magic, sizeof(magic)); if (patch) *((uint16_t*)patch) = guest_mem + ADDR_TEXT + text_prefix_size; } memcpy((void*)(host_text + text_prefix_size), text, text_size); *(host_text + text_prefix_size + text_size) = 0xf4; memcpy(host_mem + ADDR_VAR_USER_CODE, text, text_size); *(host_mem + ADDR_VAR_USER_CODE + text_size) = 0xf4; *(host_mem + ADDR_VAR_HLT) = 0xf4; memcpy(host_mem + ADDR_VAR_SYSRET, "\x0f\x07\xf4", 3); memcpy(host_mem + ADDR_VAR_SYSEXIT, "\x0f\x35\xf4", 3); *(uint64_t*)(host_mem + ADDR_VAR_VMWRITE_FLD) = 0; *(uint64_t*)(host_mem + ADDR_VAR_VMWRITE_VAL) = 0; if (opt_count > 2) opt_count = 2; for (uintptr_t i = 0; i < opt_count; i++) { uint64_t typ = opt_array_ptr[i].typ; uint64_t val = opt_array_ptr[i].val; switch (typ % 9) { case 0: sregs.cr0 ^= val & (CR0_MP | CR0_EM | CR0_ET | CR0_NE | CR0_WP | CR0_AM | CR0_NW | CR0_CD); break; case 1: sregs.cr4 ^= val & (CR4_VME | CR4_PVI | CR4_TSD | CR4_DE | CR4_MCE | CR4_PGE | CR4_PCE | CR4_OSFXSR | CR4_OSXMMEXCPT | CR4_UMIP | CR4_VMXE | CR4_SMXE | CR4_FSGSBASE | CR4_PCIDE | CR4_OSXSAVE | CR4_SMEP | CR4_SMAP | CR4_PKE); break; case 2: sregs.efer ^= val & (EFER_SCE | EFER_NXE | EFER_SVME | EFER_LMSLE | EFER_FFXSR | EFER_TCE); break; case 3: val &= ((1 << 8) | (1 << 9) | (1 << 10) | (1 << 12) | (1 << 13) | (1 << 14) | (1 << 15) | (1 << 18) | (1 << 19) | (1 << 20) | (1 << 21)); regs.rflags ^= val; tss16_addr->flags ^= val; tss16_cpl3_addr->flags ^= val; tss32_addr->flags ^= val; tss32_cpl3_addr->flags ^= val; break; case 4: seg_cs16.type = val & 0xf; seg_cs32.type = val & 0xf; seg_cs64.type = val & 0xf; break; case 5: seg_cs16_cpl3.type = val & 0xf; seg_cs32_cpl3.type = val & 0xf; seg_cs64_cpl3.type = val & 0xf; break; case 6: seg_ds16.type = val & 0xf; seg_ds32.type = val & 0xf; seg_ds64.type = val & 0xf; break; case 7: seg_ds16_cpl3.type = val & 0xf; seg_ds32_cpl3.type = val & 0xf; seg_ds64_cpl3.type = val & 0xf; break; case 8: *(uint64_t*)(host_mem + ADDR_VAR_VMWRITE_FLD) = (val & 0xffff); *(uint64_t*)(host_mem + ADDR_VAR_VMWRITE_VAL) = (val >> 16); break; default: exit(1); } } regs.rflags |= 2; fill_segment_descriptor(gdt, ldt, &seg_ldt); fill_segment_descriptor(gdt, ldt, &seg_cs16); fill_segment_descriptor(gdt, ldt, &seg_ds16); fill_segment_descriptor(gdt, ldt, &seg_cs16_cpl3); fill_segment_descriptor(gdt, ldt, &seg_ds16_cpl3); fill_segment_descriptor(gdt, ldt, &seg_cs32); fill_segment_descriptor(gdt, ldt, &seg_ds32); fill_segment_descriptor(gdt, ldt, &seg_cs32_cpl3); fill_segment_descriptor(gdt, ldt, &seg_ds32_cpl3); fill_segment_descriptor(gdt, ldt, &seg_cs64); fill_segment_descriptor(gdt, ldt, &seg_ds64); fill_segment_descriptor(gdt, ldt, &seg_cs64_cpl3); fill_segment_descriptor(gdt, ldt, &seg_ds64_cpl3); fill_segment_descriptor(gdt, ldt, &seg_tss32); fill_segment_descriptor(gdt, ldt, &seg_tss32_2); fill_segment_descriptor(gdt, ldt, &seg_tss32_cpl3); fill_segment_descriptor(gdt, ldt, &seg_tss32_vm86); fill_segment_descriptor(gdt, ldt, &seg_tss16); fill_segment_descriptor(gdt, ldt, &seg_tss16_2); fill_segment_descriptor(gdt, ldt, &seg_tss16_cpl3); fill_segment_descriptor_dword(gdt, ldt, &seg_tss64); fill_segment_descriptor_dword(gdt, ldt, &seg_tss64_cpl3); fill_segment_descriptor(gdt, ldt, &seg_cgate16); fill_segment_descriptor(gdt, ldt, &seg_tgate16); fill_segment_descriptor(gdt, ldt, &seg_cgate32); fill_segment_descriptor(gdt, ldt, &seg_tgate32); fill_segment_descriptor_dword(gdt, ldt, &seg_cgate64); if (ioctl(cpufd, KVM_SET_SREGS, &sregs)) return -1; if (ioctl(cpufd, KVM_SET_REGS, ®s)) return -1; return 0; } static void kill_and_wait(int pid, int* status) { kill(-pid, SIGKILL); kill(pid, SIGKILL); for (int i = 0; i < 100; i++) { if (waitpid(-1, status, WNOHANG | __WALL) == pid) return; usleep(1000); } DIR* dir = opendir("/sys/fs/fuse/connections"); if (dir) { for (;;) { struct dirent* ent = readdir(dir); if (!ent) break; if (strcmp(ent->d_name, ".") == 0 || strcmp(ent->d_name, "..") == 0) continue; char abort[300]; snprintf(abort, sizeof(abort), "/sys/fs/fuse/connections/%s/abort", ent->d_name); int fd = open(abort, O_WRONLY); if (fd == -1) { continue; } if (write(fd, abort, 1) < 0) { } close(fd); } closedir(dir); } else { } while (waitpid(-1, status, __WALL) != pid) { } } static void setup_test() { prctl(PR_SET_PDEATHSIG, SIGKILL, 0, 0, 0); setpgrp(); write_file("/proc/self/oom_score_adj", "1000"); } struct thread_t { int created, call; event_t ready, done; }; static struct thread_t threads[16]; static void execute_call(int call); static int running; static void* thr(void* arg) { struct thread_t* th = (struct thread_t*)arg; for (;;) { event_wait(&th->ready); event_reset(&th->ready); execute_call(th->call); __atomic_fetch_sub(&running, 1, __ATOMIC_RELAXED); event_set(&th->done); } return 0; } static void execute_one(void) { int i, call, thread; for (call = 0; call < 10; call++) { for (thread = 0; thread < (int)(sizeof(threads) / sizeof(threads[0])); thread++) { struct thread_t* th = &threads[thread]; if (!th->created) { th->created = 1; event_init(&th->ready); event_init(&th->done); event_set(&th->done); thread_start(thr, th); } if (!event_isset(&th->done)) continue; event_reset(&th->done); th->call = call; __atomic_fetch_add(&running, 1, __ATOMIC_RELAXED); event_set(&th->ready); event_timedwait(&th->done, 50); break; } } for (i = 0; i < 100 && __atomic_load_n(&running, __ATOMIC_RELAXED); i++) sleep_ms(1); } static void execute_one(void); #define WAIT_FLAGS __WALL static void loop(void) { int iter = 0; for (;; iter++) { int pid = fork(); if (pid < 0) exit(1); if (pid == 0) { setup_test(); execute_one(); exit(0); } int status = 0; uint64_t start = current_time_ms(); for (;;) { if (waitpid(-1, &status, WNOHANG | WAIT_FLAGS) == pid) break; sleep_ms(1); if (current_time_ms() - start < 5000) continue; kill_and_wait(pid, &status); break; } } } uint64_t r[4] = {0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff, 0xffffffffffffffff}; void execute_call(int call) { intptr_t res = 0; switch (call) { case 0: NONFAILING(memcpy((void*)0x20000280, "/dev/kvm\000", 9)); res = syscall(__NR_openat, /*fd=*/0xffffffffffffff9cul, /*file=*/0x20000280ul, /*flags=*/0ul, /*mode=*/0ul); if (res != -1) r[0] = res; break; case 1: res = syscall(__NR_ioctl, /*fd=*/r[0], /*cmd=*/0xae01, /*type=*/0ul); if (res != -1) r[1] = res; break; case 2: NONFAILING(memcpy( (void*)0x20000180, "-B\325NI\305j\232ppp\360\b\204\242m\000\v\030\0004\246Ey\333\321\247" "\261S\361:)\000\312\327Uw\000\274\3722\263\273\215\254\254va}knh#" "\317)\017\310\300:\234c\020d\356\251\213\0066\270G\321c\341$" "\377\227k\336\305\3516\335U)\3118M\315\373\314\202n=\177=" "\315Jx\252\217~\2710a\251\262\004K\230\223=" "\253Q\367\005\035\241\316\213\031\352\357\343", 123)); syscall(__NR_memfd_create, /*name=*/0x20000180ul, /*flags=*/0ul); break; case 3: NONFAILING(syz_kvm_setup_cpu( /*fd=*/-1, /*cpufd=*/-1, /*usermem=*/0x2000e000, /*text=*/0, /*ntext=*/0, /*flags=*/0x70, /*opts=*/0, /*nopt=*/0)); break; case 4: res = syscall(__NR_dup, /*oldfd=*/r[1]); if (res != -1) r[2] = res; break; case 5: res = syscall(__NR_ioctl, /*fd=*/r[2], /*cmd=*/0xae41, /*id=*/0ul); if (res != -1) r[3] = res; break; case 6: NONFAILING(*(uint32_t*)0x20000180 = 0); NONFAILING(*(uint32_t*)0x20000184 = 0); NONFAILING(*(uint64_t*)0x20000188 = 0); NONFAILING(*(uint64_t*)0x20000190 = 0x2000); NONFAILING(*(uint64_t*)0x20000198 = 0x20000000); syscall(__NR_ioctl, /*fd=*/r[1], /*cmd=*/0x4020ae46, /*arg=*/0x20000180ul); break; case 7: NONFAILING(*(uint64_t*)0x20000100 = 8); NONFAILING(*(uint64_t*)0x20000108 = 0); NONFAILING(*(uint64_t*)0x20000110 = 0); NONFAILING(syz_kvm_setup_cpu(/*fd=*/-1, /*cpufd=*/r[3], /*usermem=*/0x20000000, /*text=*/0x20000100, /*ntext=*/1, /*flags=*/0, /*opts=*/0, /*nopt=*/0)); { int i; for (i = 0; i < 32; i++) { NONFAILING(syz_kvm_setup_cpu(/*fd=*/-1, /*cpufd=*/r[3], /*usermem=*/0x20000000, /*text=*/0x20000100, /*ntext=*/1, /*flags=*/0, /*opts=*/0, /*nopt=*/0)); } } break; case 8: syscall(__NR_ioctl, /*fd=*/-1, /*cmd=*/0x4010ae67, /*arg=*/0ul); break; case 9: syscall(__NR_ioctl, /*fd=*/r[3], /*cmd=*/0xae80, /*arg=*/0ul); { int i; for (i = 0; i < 32; i++) { syscall(__NR_ioctl, /*fd=*/r[3], /*cmd=*/0xae80, /*arg=*/0ul); } } break; } } int main(void) { syscall(__NR_mmap, /*addr=*/0x1ffff000ul, /*len=*/0x1000ul, /*prot=*/0ul, /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul); syscall(__NR_mmap, /*addr=*/0x20000000ul, /*len=*/0x1000000ul, /*prot=*/7ul, /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul); syscall(__NR_mmap, /*addr=*/0x21000000ul, /*len=*/0x1000ul, /*prot=*/0ul, /*flags=*/0x32ul, /*fd=*/-1, /*offset=*/0ul); install_segv_handler(); for (procid = 0; procid < 4; procid++) { if (fork() == 0) { loop(); } } sleep(1000000); return 0; } =* repro.txt =* r0 = openat$kvm(0xffffffffffffff9c, &(0x7f0000000280), 0x0, 0x0) r1 = ioctl$KVM_CREATE_VM(r0, 0xae01, 0x0) memfd_create(&(0x7f0000000180)='-B\xd5NI\xc5j\x9appp\xf0\b\x84\xa2m\x00\v\x18\x004\xa6Ey\xdb\xd1\xa7\xb1S\xf1:)\x00\xca\xd7Uw\x00\xbc\xfa2\xb3\xbb\x8d\xac\xacva}knh#\xcf)\x0f\xc8\xc0:\x9cc\x10d\xee\xa9\x8b\x066\xb8G\xd1c\xe1$\xff\x97k\xde\xc5\xe96\xddU)\xc98M\xcd\xfb\xcc\x82n=\x7f=\xcdJx\xaa\x8f~\xb90a\xa9\xb2\x04K\x98\x93=\xabQ\xf7\x05\x1d\xa1\xce\x8b\x19\xea\xef\xe3', 0x0) syz_kvm_setup_cpu$x86(0xffffffffffffffff, 0xffffffffffffffff, &(0x7f000000e000/0x18000)=nil, 0x0, 0x0, 0x70, 0x0, 0x0) r2 = dup(r1) r3 = ioctl$KVM_CREATE_VCPU(r2, 0xae41, 0x0) ioctl$KVM_SET_USER_MEMORY_REGION(r1, 0x4020ae46, &(0x7f0000000180)={0x0, 0x0, 0x0, 0x2000, &(0x7f0000000000/0x2000)=nil}) syz_kvm_setup_cpu$x86(0xffffffffffffffff, r3, &(0x7f0000000000/0x18000)=nil, &(0x7f0000000100)=[@textreal={0x8, 0x0}], 0x1, 0x0, 0x0, 0x0) (rerun: 32) ioctl$KVM_REGISTER_COALESCED_MMIO(0xffffffffffffffff, 0x4010ae67, 0x0) ioctl$KVM_RUN(r3, 0xae80, 0x0) (rerun: 32) See aslo https://gist.github.com/xrivendell7/8d293339a1c88fda7301c43aa07dd3b9 I hope it helps. Merry Christmas! xingwei Lee