]> git.ipfire.org Git - thirdparty/kernel/stable-queue.git/commitdiff
4.14-stable patches
authorGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 10 Jan 2018 12:03:45 +0000 (13:03 +0100)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 10 Jan 2018 12:03:45 +0000 (13:03 +0100)
added patches:
kvm-fix-stack-out-of-bounds-read-in-write_mmio.patch

queue-4.14/kvm-fix-stack-out-of-bounds-read-in-write_mmio.patch [new file with mode: 0644]
queue-4.14/series

diff --git a/queue-4.14/kvm-fix-stack-out-of-bounds-read-in-write_mmio.patch b/queue-4.14/kvm-fix-stack-out-of-bounds-read-in-write_mmio.patch
new file mode 100644 (file)
index 0000000..605e778
--- /dev/null
@@ -0,0 +1,159 @@
+From e39d200fa5bf5b94a0948db0dae44c1b73b84a56 Mon Sep 17 00:00:00 2001
+From: Wanpeng Li <wanpeng.li@hotmail.com>
+Date: Thu, 14 Dec 2017 17:40:50 -0800
+Subject: KVM: Fix stack-out-of-bounds read in write_mmio
+MIME-Version: 1.0
+Content-Type: text/plain; charset=UTF-8
+Content-Transfer-Encoding: 8bit
+
+From: Wanpeng Li <wanpeng.li@hotmail.com>
+
+commit e39d200fa5bf5b94a0948db0dae44c1b73b84a56 upstream.
+
+Reported by syzkaller:
+
+  BUG: KASAN: stack-out-of-bounds in write_mmio+0x11e/0x270 [kvm]
+  Read of size 8 at addr ffff8803259df7f8 by task syz-executor/32298
+
+  CPU: 6 PID: 32298 Comm: syz-executor Tainted: G           OE    4.15.0-rc2+ #18
+  Hardware name: LENOVO ThinkCentre M8500t-N000/SHARKBAY, BIOS FBKTC1AUS 02/16/2016
+  Call Trace:
+   dump_stack+0xab/0xe1
+   print_address_description+0x6b/0x290
+   kasan_report+0x28a/0x370
+   write_mmio+0x11e/0x270 [kvm]
+   emulator_read_write_onepage+0x311/0x600 [kvm]
+   emulator_read_write+0xef/0x240 [kvm]
+   emulator_fix_hypercall+0x105/0x150 [kvm]
+   em_hypercall+0x2b/0x80 [kvm]
+   x86_emulate_insn+0x2b1/0x1640 [kvm]
+   x86_emulate_instruction+0x39a/0xb90 [kvm]
+   handle_exception+0x1b4/0x4d0 [kvm_intel]
+   vcpu_enter_guest+0x15a0/0x2640 [kvm]
+   kvm_arch_vcpu_ioctl_run+0x549/0x7d0 [kvm]
+   kvm_vcpu_ioctl+0x479/0x880 [kvm]
+   do_vfs_ioctl+0x142/0x9a0
+   SyS_ioctl+0x74/0x80
+   entry_SYSCALL_64_fastpath+0x23/0x9a
+
+The path of patched vmmcall will patch 3 bytes opcode 0F 01 C1(vmcall)
+to the guest memory, however, write_mmio tracepoint always prints 8 bytes
+through *(u64 *)val since kvm splits the mmio access into 8 bytes. This
+leaks 5 bytes from the kernel stack (CVE-2017-17741).  This patch fixes
+it by just accessing the bytes which we operate on.
+
+Before patch:
+
+syz-executor-5567  [007] .... 51370.561696: kvm_mmio: mmio write len 3 gpa 0x10 val 0x1ffff10077c1010f
+
+After patch:
+
+syz-executor-13416 [002] .... 51302.299573: kvm_mmio: mmio write len 3 gpa 0x10 val 0xc1010f
+
+Reported-by: Dmitry Vyukov <dvyukov@google.com>
+Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
+Reviewed-by: Marc Zyngier <marc.zyngier@arm.com>
+Tested-by: Marc Zyngier <marc.zyngier@arm.com>
+Cc: Paolo Bonzini <pbonzini@redhat.com>
+Cc: Radim Krčmář <rkrcmar@redhat.com>
+Cc: Marc Zyngier <marc.zyngier@arm.com>
+Cc: Christoffer Dall <christoffer.dall@linaro.org>
+Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
+Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
+Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
+Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+
+---
+ arch/x86/kvm/x86.c         |    8 ++++----
+ include/trace/events/kvm.h |    7 +++++--
+ virt/kvm/arm/mmio.c        |    6 +++---
+ 3 files changed, 12 insertions(+), 9 deletions(-)
+
+--- a/arch/x86/kvm/x86.c
++++ b/arch/x86/kvm/x86.c
+@@ -4362,7 +4362,7 @@ static int vcpu_mmio_read(struct kvm_vcp
+                                        addr, n, v))
+                   && kvm_io_bus_read(vcpu, KVM_MMIO_BUS, addr, n, v))
+                       break;
+-              trace_kvm_mmio(KVM_TRACE_MMIO_READ, n, addr, *(u64 *)v);
++              trace_kvm_mmio(KVM_TRACE_MMIO_READ, n, addr, v);
+               handled += n;
+               addr += n;
+               len -= n;
+@@ -4621,7 +4621,7 @@ static int read_prepare(struct kvm_vcpu
+ {
+       if (vcpu->mmio_read_completed) {
+               trace_kvm_mmio(KVM_TRACE_MMIO_READ, bytes,
+-                             vcpu->mmio_fragments[0].gpa, *(u64 *)val);
++                             vcpu->mmio_fragments[0].gpa, val);
+               vcpu->mmio_read_completed = 0;
+               return 1;
+       }
+@@ -4643,14 +4643,14 @@ static int write_emulate(struct kvm_vcpu
+ static int write_mmio(struct kvm_vcpu *vcpu, gpa_t gpa, int bytes, void *val)
+ {
+-      trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, bytes, gpa, *(u64 *)val);
++      trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, bytes, gpa, val);
+       return vcpu_mmio_write(vcpu, gpa, bytes, val);
+ }
+ static int read_exit_mmio(struct kvm_vcpu *vcpu, gpa_t gpa,
+                         void *val, int bytes)
+ {
+-      trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, bytes, gpa, 0);
++      trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, bytes, gpa, NULL);
+       return X86EMUL_IO_NEEDED;
+ }
+--- a/include/trace/events/kvm.h
++++ b/include/trace/events/kvm.h
+@@ -211,7 +211,7 @@ TRACE_EVENT(kvm_ack_irq,
+       { KVM_TRACE_MMIO_WRITE, "write" }
+ TRACE_EVENT(kvm_mmio,
+-      TP_PROTO(int type, int len, u64 gpa, u64 val),
++      TP_PROTO(int type, int len, u64 gpa, void *val),
+       TP_ARGS(type, len, gpa, val),
+       TP_STRUCT__entry(
+@@ -225,7 +225,10 @@ TRACE_EVENT(kvm_mmio,
+               __entry->type           = type;
+               __entry->len            = len;
+               __entry->gpa            = gpa;
+-              __entry->val            = val;
++              __entry->val            = 0;
++              if (val)
++                      memcpy(&__entry->val, val,
++                             min_t(u32, sizeof(__entry->val), len));
+       ),
+       TP_printk("mmio %s len %u gpa 0x%llx val 0x%llx",
+--- a/virt/kvm/arm/mmio.c
++++ b/virt/kvm/arm/mmio.c
+@@ -112,7 +112,7 @@ int kvm_handle_mmio_return(struct kvm_vc
+               }
+               trace_kvm_mmio(KVM_TRACE_MMIO_READ, len, run->mmio.phys_addr,
+-                             data);
++                             &data);
+               data = vcpu_data_host_to_guest(vcpu, data, len);
+               vcpu_set_reg(vcpu, vcpu->arch.mmio_decode.rt, data);
+       }
+@@ -182,14 +182,14 @@ int io_mem_abort(struct kvm_vcpu *vcpu,
+               data = vcpu_data_guest_to_host(vcpu, vcpu_get_reg(vcpu, rt),
+                                              len);
+-              trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, len, fault_ipa, data);
++              trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, len, fault_ipa, &data);
+               kvm_mmio_write_buf(data_buf, len, data);
+               ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, fault_ipa, len,
+                                      data_buf);
+       } else {
+               trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, len,
+-                             fault_ipa, 0);
++                             fault_ipa, NULL);
+               ret = kvm_io_bus_read(vcpu, KVM_MMIO_BUS, fault_ipa, len,
+                                     data_buf);
index f77b2e574a3635722366215972ca4cb79b58894f..7819b994c5cc72405f8f066a43e2602754c1122d 100644 (file)
@@ -1 +1,2 @@
 dm-bufio-fix-shrinker-scans-when-nr_to_scan-retain_target.patch
+kvm-fix-stack-out-of-bounds-read-in-write_mmio.patch