migration: Fix use-after-free during process exit

Message ID	20190408113343.2370-1-yury-kotov@yandex-team.ru (mailing list archive)
State	New, archived
Headers	show Return-Path: <qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org> From: Yury Kotov <yury-kotov@yandex-team.ru> To: "Dr. David Alan Gilbert" <dgilbert@redhat.com>, Juan Quintela <quintela@redhat.com> Date: Mon, 8 Apr 2019 14:33:43 +0300 Message-Id: <20190408113343.2370-1-yury-kotov@yandex-team.ru> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Subject: [Qemu-devel] [PATCH] migration: Fix use-after-free during process exit Precedence: list Cc: =?utf-8?q?Alex_Benn=C3=A9e?= <alex.bennee@linaro.org>, qemu-devel@nongnu.org, Evgeny Yakovlev <wrfsh@yandex-team.ru> Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" <qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org>
Series	migration: Fix use-after-free during process exit \| expand migration: Fix use-after-free during process exit

Yury Kotov April 8, 2019, 11:33 a.m. UTC

It fixes heap-use-after-free which was found by clang's ASAN.

Control flow of this use-after-free:
main_thread:
    * Got SIGTERM and completes main loop
    * Calls migration_shutdown
      - migrate_fd_cancel (so, migration_thread begins to complete)
      - object_unref(OBJECT(current_migration));

migration_thread:
    * migration_iteration_finish -> schedule cleanup bh
    * object_unref(OBJECT(s)); (Now, current_migration is freed)
    * exits

main_thread:
    * Calls vm_shutdown -> drain bdrvs -> main loop
      -> cleanup_bh -> use after free

If you want to reproduce, these couple of sleeps will help:
vl.c:4613:
     migration_shutdown();
+    sleep(2);
migration.c:3269:
+    sleep(1);
     trace_migration_thread_after_loop();
     migration_iteration_finish(s);

Original output:
qemu-system-x86_64: terminating on signal 15 from pid 31980 (<unknown process>)

Yury Kotov April 17, 2019, 12:43 p.m. UTC | #1

Ping

08.04.2019, 14:34, "Yury Kotov" <yury-kotov@yandex-team.ru>:
> It fixes heap-use-after-free which was found by clang's ASAN.
>
> Control flow of this use-after-free:
> main_thread:
>     * Got SIGTERM and completes main loop
>     * Calls migration_shutdown
>       - migrate_fd_cancel (so, migration_thread begins to complete)
>       - object_unref(OBJECT(current_migration));
>
> migration_thread:
>     * migration_iteration_finish -> schedule cleanup bh
>     * object_unref(OBJECT(s)); (Now, current_migration is freed)
>     * exits
>
> main_thread:
>     * Calls vm_shutdown -> drain bdrvs -> main loop
>       -> cleanup_bh -> use after free
>
> If you want to reproduce, these couple of sleeps will help:
> vl.c:4613:
>      migration_shutdown();
> + sleep(2);
> migration.c:3269:
> + sleep(1);
>      trace_migration_thread_after_loop();
>      migration_iteration_finish(s);
>
> Original output:
> qemu-system-x86_64: terminating on signal 15 from pid 31980 (<unknown process>)
> =================================================================
> ==31958==ERROR: AddressSanitizer: heap-use-after-free on address 0x61900001d210
>   at pc 0x555558a535ca bp 0x7fffffffb190 sp 0x7fffffffb188
> READ of size 8 at 0x61900001d210 thread T0 (qemu-vm-0)
>     #0 0x555558a535c9 in migrate_fd_cleanup migration/migration.c:1502:23
>     #1 0x5555594fde0a in aio_bh_call util/async.c:90:5
>     #2 0x5555594fe522 in aio_bh_poll util/async.c:118:13
>     #3 0x555559524783 in aio_poll util/aio-posix.c:725:17
>     #4 0x555559504fb3 in aio_wait_bh_oneshot util/aio-wait.c:71:5
>     #5 0x5555573bddf6 in virtio_blk_data_plane_stop
>       hw/block/dataplane/virtio-blk.c:282:5
>     #6 0x5555589d5c09 in virtio_bus_stop_ioeventfd hw/virtio/virtio-bus.c:246:9
>     #7 0x5555589e9917 in virtio_pci_stop_ioeventfd hw/virtio/virtio-pci.c:287:5
>     #8 0x5555589e22bf in virtio_pci_vmstate_change hw/virtio/virtio-pci.c:1072:9
>     #9 0x555557628931 in virtio_vmstate_change hw/virtio/virtio.c:2257:9
>     #10 0x555557c36713 in vm_state_notify vl.c:1605:9
>     #11 0x55555716ef53 in do_vm_stop cpus.c:1074:9
>     #12 0x55555716eeff in vm_shutdown cpus.c:1092:12
>     #13 0x555557c4283e in main vl.c:4617:5
>     #14 0x7fffdfdb482f in __libc_start_main
>       (/lib/x86_64-linux-gnu/libc.so.6+0x2082f)
>     #15 0x555556ecb118 in _start (x86_64-softmmu/qemu-system-x86_64+0x1977118)
>
> 0x61900001d210 is located 144 bytes inside of 952-byte region
>   [0x61900001d180,0x61900001d538)
> freed by thread T6 (live_migration) here:
>     #0 0x555556f76782 in __interceptor_free
>       /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:124:3
>     #1 0x555558d5fa94 in object_finalize qom/object.c:618:9
>     #2 0x555558d57651 in object_unref qom/object.c:1068:9
>     #3 0x555558a55588 in migration_thread migration/migration.c:3272:5
>     #4 0x5555595393f2 in qemu_thread_start util/qemu-thread-posix.c:502:9
>     #5 0x7fffe057f6b9 in start_thread (/lib/x86_64-linux-gnu/libpthread.so.0+0x76b9)
>
> previously allocated by thread T0 (qemu-vm-0) here:
>     #0 0x555556f76b03 in __interceptor_malloc
>       /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:146:3
>     #1 0x7ffff6ee37b8 in g_malloc (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x4f7b8)
>     #2 0x555558d58031 in object_new qom/object.c:640:12
>     #3 0x555558a31f21 in migration_object_init migration/migration.c:139:25
>     #4 0x555557c41398 in main vl.c:4320:5
>     #5 0x7fffdfdb482f in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x2082f)
>
> Thread T6 (live_migration) created by T0 (qemu-vm-0) here:
>     #0 0x555556f5f0dd in pthread_create
>       /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_interceptors.cc:210:3
>     #1 0x555559538cf9 in qemu_thread_create util/qemu-thread-posix.c:539:11
>     #2 0x555558a53304 in migrate_fd_connect migration/migration.c:3332:5
>     #3 0x555558a72bd8 in migration_channel_connect migration/channel.c:92:5
>     #4 0x555558a6ef87 in exec_start_outgoing_migration migration/exec.c:42:5
>     #5 0x555558a4f3c2 in qmp_migrate migration/migration.c:1922:9
>     #6 0x555558bb4f6a in qmp_marshal_migrate qapi/qapi-commands-migration.c:607:5
>     #7 0x555559363738 in do_qmp_dispatch qapi/qmp-dispatch.c:131:5
>     #8 0x555559362a15 in qmp_dispatch qapi/qmp-dispatch.c:174:11
>     #9 0x5555571bac15 in monitor_qmp_dispatch monitor.c:4124:11
>     #10 0x55555719a22d in monitor_qmp_bh_dispatcher monitor.c:4207:9
>     #11 0x5555594fde0a in aio_bh_call util/async.c:90:5
>     #12 0x5555594fe522 in aio_bh_poll util/async.c:118:13
>     #13 0x5555595201e0 in aio_dispatch util/aio-posix.c:460:5
>     #14 0x555559503553 in aio_ctx_dispatch util/async.c:261:5
>     #15 0x7ffff6ede196 in g_main_context_dispatch
>       (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x4a196)
>
> SUMMARY: AddressSanitizer: heap-use-after-free migration/migration.c:1502:23
>   in migrate_fd_cleanup
> Shadow bytes around the buggy address:
>   0x0c327fffb9f0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba00: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba10: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba20: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba30: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
> =>0x0c327fffba40: fd fd[fd]fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba50: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba60: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba70: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba80: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba90: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
> Shadow byte legend (one shadow byte represents 8 application bytes):
>   Addressable: 00
>   Partially addressable: 01 02 03 04 05 06 07
>   Heap left redzone: fa
>   Freed heap region: fd
>   Stack left redzone: f1
>   Stack mid redzone: f2
>   Stack right redzone: f3
>   Stack after return: f5
>   Stack use after scope: f8
>   Global redzone: f9
>   Global init order: f6
>   Poisoned by user: f7
>   Container overflow: fc
>   Array cookie: ac
>   Intra object redzone: bb
>   ASan internal: fe
>   Left alloca redzone: ca
>   Right alloca redzone: cb
>   Shadow gap: cc
> ==31958==ABORTING
>
> Signed-off-by: Yury Kotov <yury-kotov@yandex-team.ru>
> ---
>  migration/migration.c | 23 ++++++++++++++++++-----
>  1 file changed, 18 insertions(+), 5 deletions(-)
>
> diff --git a/migration/migration.c b/migration/migration.c
> index 609e0df5d0..b9848d1030 100644
> --- a/migration/migration.c
> +++ b/migration/migration.c
> @@ -1495,10 +1495,8 @@ static void block_cleanup_parameters(MigrationState *s)
>      }
>  }
>
> -static void migrate_fd_cleanup(void *opaque)
> +static void migrate_fd_cleanup(MigrationState *s)
>  {
> - MigrationState *s = opaque;
> -
>      qemu_bh_delete(s->cleanup_bh);
>      s->cleanup_bh = NULL;
>
> @@ -1543,6 +1541,21 @@ static void migrate_fd_cleanup(void *opaque)
>      block_cleanup_parameters(s);
>  }
>
> +static void migrate_fd_cleanup_schedule(MigrationState *s)
> +{
> + /* Ref the state for bh, because it may be called when
> + * there're already no other refs */
> + object_ref(OBJECT(s));
> + qemu_bh_schedule(s->cleanup_bh);
> +}
> +
> +static void migrate_fd_cleanup_bh(void *opaque)
> +{
> + MigrationState *s = opaque;
> + migrate_fd_cleanup(s);
> + object_unref(OBJECT(s));
> +}
> +
>  void migrate_set_error(MigrationState *s, const Error *error)
>  {
>      qemu_mutex_lock(&s->error_mutex);
> @@ -3144,7 +3157,7 @@ static void migration_iteration_finish(MigrationState *s)
>          error_report("%s: Unknown ending state %d", __func__, s->state);
>          break;
>      }
> - qemu_bh_schedule(s->cleanup_bh);
> + migrate_fd_cleanup_schedule(s);
>      qemu_mutex_unlock_iothread();
>  }
>
> @@ -3279,7 +3292,7 @@ void migrate_fd_connect(MigrationState *s, Error *error_in)
>      bool resume = s->state == MIGRATION_STATUS_POSTCOPY_PAUSED;
>
>      s->expected_downtime = s->parameters.downtime_limit;
> - s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup, s);
> + s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup_bh, s);
>      if (error_in) {
>          migrate_fd_error(s, error_in);
>          migrate_fd_cleanup(s);
> --
> 2.21.0

Yury Kotov May 14, 2019, 9:40 a.m. UTC | #2

Ping ping

17.04.2019, 15:44, "Yury Kotov" <yury-kotov@yandex-team.ru>:
> Ping
>
> 08.04.2019, 14:34, "Yury Kotov" <yury-kotov@yandex-team.ru>:
>>  It fixes heap-use-after-free which was found by clang's ASAN.
>>
>>  Control flow of this use-after-free:
>>  main_thread:
>>      * Got SIGTERM and completes main loop
>>      * Calls migration_shutdown
>>        - migrate_fd_cancel (so, migration_thread begins to complete)
>>        - object_unref(OBJECT(current_migration));
>>
>>  migration_thread:
>>      * migration_iteration_finish -> schedule cleanup bh
>>      * object_unref(OBJECT(s)); (Now, current_migration is freed)
>>      * exits
>>
>>  main_thread:
>>      * Calls vm_shutdown -> drain bdrvs -> main loop
>>        -> cleanup_bh -> use after free
>>
>>  If you want to reproduce, these couple of sleeps will help:
>>  vl.c:4613:
>>       migration_shutdown();
>>  + sleep(2);
>>  migration.c:3269:
>>  + sleep(1);
>>       trace_migration_thread_after_loop();
>>       migration_iteration_finish(s);
>>
>>  Original output:
>>  qemu-system-x86_64: terminating on signal 15 from pid 31980 (<unknown process>)
>>  =================================================================
>>  ==31958==ERROR: AddressSanitizer: heap-use-after-free on address 0x61900001d210
>>    at pc 0x555558a535ca bp 0x7fffffffb190 sp 0x7fffffffb188
>>  READ of size 8 at 0x61900001d210 thread T0 (qemu-vm-0)
>>      #0 0x555558a535c9 in migrate_fd_cleanup migration/migration.c:1502:23
>>      #1 0x5555594fde0a in aio_bh_call util/async.c:90:5
>>      #2 0x5555594fe522 in aio_bh_poll util/async.c:118:13
>>      #3 0x555559524783 in aio_poll util/aio-posix.c:725:17
>>      #4 0x555559504fb3 in aio_wait_bh_oneshot util/aio-wait.c:71:5
>>      #5 0x5555573bddf6 in virtio_blk_data_plane_stop
>>        hw/block/dataplane/virtio-blk.c:282:5
>>      #6 0x5555589d5c09 in virtio_bus_stop_ioeventfd hw/virtio/virtio-bus.c:246:9
>>      #7 0x5555589e9917 in virtio_pci_stop_ioeventfd hw/virtio/virtio-pci.c:287:5
>>      #8 0x5555589e22bf in virtio_pci_vmstate_change hw/virtio/virtio-pci.c:1072:9
>>      #9 0x555557628931 in virtio_vmstate_change hw/virtio/virtio.c:2257:9
>>      #10 0x555557c36713 in vm_state_notify vl.c:1605:9
>>      #11 0x55555716ef53 in do_vm_stop cpus.c:1074:9
>>      #12 0x55555716eeff in vm_shutdown cpus.c:1092:12
>>      #13 0x555557c4283e in main vl.c:4617:5
>>      #14 0x7fffdfdb482f in __libc_start_main
>>        (/lib/x86_64-linux-gnu/libc.so.6+0x2082f)
>>      #15 0x555556ecb118 in _start (x86_64-softmmu/qemu-system-x86_64+0x1977118)
>>
>>  0x61900001d210 is located 144 bytes inside of 952-byte region
>>    [0x61900001d180,0x61900001d538)
>>  freed by thread T6 (live_migration) here:
>>      #0 0x555556f76782 in __interceptor_free
>>        /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:124:3
>>      #1 0x555558d5fa94 in object_finalize qom/object.c:618:9
>>      #2 0x555558d57651 in object_unref qom/object.c:1068:9
>>      #3 0x555558a55588 in migration_thread migration/migration.c:3272:5
>>      #4 0x5555595393f2 in qemu_thread_start util/qemu-thread-posix.c:502:9
>>      #5 0x7fffe057f6b9 in start_thread (/lib/x86_64-linux-gnu/libpthread.so.0+0x76b9)
>>
>>  previously allocated by thread T0 (qemu-vm-0) here:
>>      #0 0x555556f76b03 in __interceptor_malloc
>>        /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:146:3
>>      #1 0x7ffff6ee37b8 in g_malloc (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x4f7b8)
>>      #2 0x555558d58031 in object_new qom/object.c:640:12
>>      #3 0x555558a31f21 in migration_object_init migration/migration.c:139:25
>>      #4 0x555557c41398 in main vl.c:4320:5
>>      #5 0x7fffdfdb482f in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x2082f)
>>
>>  Thread T6 (live_migration) created by T0 (qemu-vm-0) here:
>>      #0 0x555556f5f0dd in pthread_create
>>        /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_interceptors.cc:210:3
>>      #1 0x555559538cf9 in qemu_thread_create util/qemu-thread-posix.c:539:11
>>      #2 0x555558a53304 in migrate_fd_connect migration/migration.c:3332:5
>>      #3 0x555558a72bd8 in migration_channel_connect migration/channel.c:92:5
>>      #4 0x555558a6ef87 in exec_start_outgoing_migration migration/exec.c:42:5
>>      #5 0x555558a4f3c2 in qmp_migrate migration/migration.c:1922:9
>>      #6 0x555558bb4f6a in qmp_marshal_migrate qapi/qapi-commands-migration.c:607:5
>>      #7 0x555559363738 in do_qmp_dispatch qapi/qmp-dispatch.c:131:5
>>      #8 0x555559362a15 in qmp_dispatch qapi/qmp-dispatch.c:174:11
>>      #9 0x5555571bac15 in monitor_qmp_dispatch monitor.c:4124:11
>>      #10 0x55555719a22d in monitor_qmp_bh_dispatcher monitor.c:4207:9
>>      #11 0x5555594fde0a in aio_bh_call util/async.c:90:5
>>      #12 0x5555594fe522 in aio_bh_poll util/async.c:118:13
>>      #13 0x5555595201e0 in aio_dispatch util/aio-posix.c:460:5
>>      #14 0x555559503553 in aio_ctx_dispatch util/async.c:261:5
>>      #15 0x7ffff6ede196 in g_main_context_dispatch
>>        (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x4a196)
>>
>>  SUMMARY: AddressSanitizer: heap-use-after-free migration/migration.c:1502:23
>>    in migrate_fd_cleanup
>>  Shadow bytes around the buggy address:
>>    0x0c327fffb9f0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>>    0x0c327fffba00: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>>    0x0c327fffba10: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>>    0x0c327fffba20: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>>    0x0c327fffba30: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>>  =>0x0c327fffba40: fd fd[fd]fd fd fd fd fd fd fd fd fd fd fd fd fd
>>    0x0c327fffba50: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>>    0x0c327fffba60: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>>    0x0c327fffba70: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>>    0x0c327fffba80: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>>    0x0c327fffba90: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>>  Shadow byte legend (one shadow byte represents 8 application bytes):
>>    Addressable: 00
>>    Partially addressable: 01 02 03 04 05 06 07
>>    Heap left redzone: fa
>>    Freed heap region: fd
>>    Stack left redzone: f1
>>    Stack mid redzone: f2
>>    Stack right redzone: f3
>>    Stack after return: f5
>>    Stack use after scope: f8
>>    Global redzone: f9
>>    Global init order: f6
>>    Poisoned by user: f7
>>    Container overflow: fc
>>    Array cookie: ac
>>    Intra object redzone: bb
>>    ASan internal: fe
>>    Left alloca redzone: ca
>>    Right alloca redzone: cb
>>    Shadow gap: cc
>>  ==31958==ABORTING
>>
>>  Signed-off-by: Yury Kotov <yury-kotov@yandex-team.ru>
>>  ---
>>   migration/migration.c | 23 ++++++++++++++++++-----
>>   1 file changed, 18 insertions(+), 5 deletions(-)
>>
>>  diff --git a/migration/migration.c b/migration/migration.c
>>  index 609e0df5d0..b9848d1030 100644
>>  --- a/migration/migration.c
>>  +++ b/migration/migration.c
>>  @@ -1495,10 +1495,8 @@ static void block_cleanup_parameters(MigrationState *s)
>>       }
>>   }
>>
>>  -static void migrate_fd_cleanup(void *opaque)
>>  +static void migrate_fd_cleanup(MigrationState *s)
>>   {
>>  - MigrationState *s = opaque;
>>  -
>>       qemu_bh_delete(s->cleanup_bh);
>>       s->cleanup_bh = NULL;
>>
>>  @@ -1543,6 +1541,21 @@ static void migrate_fd_cleanup(void *opaque)
>>       block_cleanup_parameters(s);
>>   }
>>
>>  +static void migrate_fd_cleanup_schedule(MigrationState *s)
>>  +{
>>  + /* Ref the state for bh, because it may be called when
>>  + * there're already no other refs */
>>  + object_ref(OBJECT(s));
>>  + qemu_bh_schedule(s->cleanup_bh);
>>  +}
>>  +
>>  +static void migrate_fd_cleanup_bh(void *opaque)
>>  +{
>>  + MigrationState *s = opaque;
>>  + migrate_fd_cleanup(s);
>>  + object_unref(OBJECT(s));
>>  +}
>>  +
>>   void migrate_set_error(MigrationState *s, const Error *error)
>>   {
>>       qemu_mutex_lock(&s->error_mutex);
>>  @@ -3144,7 +3157,7 @@ static void migration_iteration_finish(MigrationState *s)
>>           error_report("%s: Unknown ending state %d", __func__, s->state);
>>           break;
>>       }
>>  - qemu_bh_schedule(s->cleanup_bh);
>>  + migrate_fd_cleanup_schedule(s);
>>       qemu_mutex_unlock_iothread();
>>   }
>>
>>  @@ -3279,7 +3292,7 @@ void migrate_fd_connect(MigrationState *s, Error *error_in)
>>       bool resume = s->state == MIGRATION_STATUS_POSTCOPY_PAUSED;
>>
>>       s->expected_downtime = s->parameters.downtime_limit;
>>  - s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup, s);
>>  + s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup_bh, s);
>>       if (error_in) {
>>           migrate_fd_error(s, error_in);
>>           migrate_fd_cleanup(s);
>>  --
>>  2.21.0

Dr. David Alan Gilbert May 14, 2019, 4:04 p.m. UTC | #3

* Yury Kotov (yury-kotov@yandex-team.ru) wrote:
> It fixes heap-use-after-free which was found by clang's ASAN.

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

and queued.
(cc'ing in Stefan since aio crashes often get to him).


> Control flow of this use-after-free:
> main_thread:
>     * Got SIGTERM and completes main loop
>     * Calls migration_shutdown
>       - migrate_fd_cancel (so, migration_thread begins to complete)
>       - object_unref(OBJECT(current_migration));
> 
> migration_thread:
>     * migration_iteration_finish -> schedule cleanup bh
>     * object_unref(OBJECT(s)); (Now, current_migration is freed)
>     * exits
> 
> main_thread:
>     * Calls vm_shutdown -> drain bdrvs -> main loop
>       -> cleanup_bh -> use after free
> 
> If you want to reproduce, these couple of sleeps will help:
> vl.c:4613:
>      migration_shutdown();
> +    sleep(2);
> migration.c:3269:
> +    sleep(1);
>      trace_migration_thread_after_loop();
>      migration_iteration_finish(s);
> 
> Original output:
> qemu-system-x86_64: terminating on signal 15 from pid 31980 (<unknown process>)
> =================================================================
> ==31958==ERROR: AddressSanitizer: heap-use-after-free on address 0x61900001d210
>   at pc 0x555558a535ca bp 0x7fffffffb190 sp 0x7fffffffb188
> READ of size 8 at 0x61900001d210 thread T0 (qemu-vm-0)
>     #0 0x555558a535c9 in migrate_fd_cleanup migration/migration.c:1502:23
>     #1 0x5555594fde0a in aio_bh_call util/async.c:90:5
>     #2 0x5555594fe522 in aio_bh_poll util/async.c:118:13
>     #3 0x555559524783 in aio_poll util/aio-posix.c:725:17
>     #4 0x555559504fb3 in aio_wait_bh_oneshot util/aio-wait.c:71:5
>     #5 0x5555573bddf6 in virtio_blk_data_plane_stop
>       hw/block/dataplane/virtio-blk.c:282:5
>     #6 0x5555589d5c09 in virtio_bus_stop_ioeventfd hw/virtio/virtio-bus.c:246:9
>     #7 0x5555589e9917 in virtio_pci_stop_ioeventfd hw/virtio/virtio-pci.c:287:5
>     #8 0x5555589e22bf in virtio_pci_vmstate_change hw/virtio/virtio-pci.c:1072:9
>     #9 0x555557628931 in virtio_vmstate_change hw/virtio/virtio.c:2257:9
>     #10 0x555557c36713 in vm_state_notify vl.c:1605:9
>     #11 0x55555716ef53 in do_vm_stop cpus.c:1074:9
>     #12 0x55555716eeff in vm_shutdown cpus.c:1092:12
>     #13 0x555557c4283e in main vl.c:4617:5
>     #14 0x7fffdfdb482f in __libc_start_main
>       (/lib/x86_64-linux-gnu/libc.so.6+0x2082f)
>     #15 0x555556ecb118 in _start (x86_64-softmmu/qemu-system-x86_64+0x1977118)
> 
> 0x61900001d210 is located 144 bytes inside of 952-byte region
>   [0x61900001d180,0x61900001d538)
> freed by thread T6 (live_migration) here:
>     #0 0x555556f76782 in __interceptor_free
>       /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:124:3
>     #1 0x555558d5fa94 in object_finalize qom/object.c:618:9
>     #2 0x555558d57651 in object_unref qom/object.c:1068:9
>     #3 0x555558a55588 in migration_thread migration/migration.c:3272:5
>     #4 0x5555595393f2 in qemu_thread_start util/qemu-thread-posix.c:502:9
>     #5 0x7fffe057f6b9 in start_thread (/lib/x86_64-linux-gnu/libpthread.so.0+0x76b9)
> 
> previously allocated by thread T0 (qemu-vm-0) here:
>     #0 0x555556f76b03 in __interceptor_malloc
>       /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_malloc_linux.cc:146:3
>     #1 0x7ffff6ee37b8 in g_malloc (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x4f7b8)
>     #2 0x555558d58031 in object_new qom/object.c:640:12
>     #3 0x555558a31f21 in migration_object_init migration/migration.c:139:25
>     #4 0x555557c41398 in main vl.c:4320:5
>     #5 0x7fffdfdb482f in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x2082f)
> 
> Thread T6 (live_migration) created by T0 (qemu-vm-0) here:
>     #0 0x555556f5f0dd in pthread_create
>       /tmp/final/llvm.src/projects/compiler-rt/lib/asan/asan_interceptors.cc:210:3
>     #1 0x555559538cf9 in qemu_thread_create util/qemu-thread-posix.c:539:11
>     #2 0x555558a53304 in migrate_fd_connect migration/migration.c:3332:5
>     #3 0x555558a72bd8 in migration_channel_connect migration/channel.c:92:5
>     #4 0x555558a6ef87 in exec_start_outgoing_migration migration/exec.c:42:5
>     #5 0x555558a4f3c2 in qmp_migrate migration/migration.c:1922:9
>     #6 0x555558bb4f6a in qmp_marshal_migrate qapi/qapi-commands-migration.c:607:5
>     #7 0x555559363738 in do_qmp_dispatch qapi/qmp-dispatch.c:131:5
>     #8 0x555559362a15 in qmp_dispatch qapi/qmp-dispatch.c:174:11
>     #9 0x5555571bac15 in monitor_qmp_dispatch monitor.c:4124:11
>     #10 0x55555719a22d in monitor_qmp_bh_dispatcher monitor.c:4207:9
>     #11 0x5555594fde0a in aio_bh_call util/async.c:90:5
>     #12 0x5555594fe522 in aio_bh_poll util/async.c:118:13
>     #13 0x5555595201e0 in aio_dispatch util/aio-posix.c:460:5
>     #14 0x555559503553 in aio_ctx_dispatch util/async.c:261:5
>     #15 0x7ffff6ede196 in g_main_context_dispatch
>       (/lib/x86_64-linux-gnu/libglib-2.0.so.0+0x4a196)
> 
> SUMMARY: AddressSanitizer: heap-use-after-free migration/migration.c:1502:23
>   in migrate_fd_cleanup
> Shadow bytes around the buggy address:
>   0x0c327fffb9f0: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba00: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba10: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba20: fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa fa
>   0x0c327fffba30: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
> =>0x0c327fffba40: fd fd[fd]fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba50: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba60: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba70: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba80: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
>   0x0c327fffba90: fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd fd
> Shadow byte legend (one shadow byte represents 8 application bytes):
>   Addressable: 00
>   Partially addressable: 01 02 03 04 05 06 07
>   Heap left redzone: fa
>   Freed heap region: fd
>   Stack left redzone: f1
>   Stack mid redzone: f2
>   Stack right redzone: f3
>   Stack after return: f5
>   Stack use after scope: f8
>   Global redzone: f9
>   Global init order: f6
>   Poisoned by user: f7
>   Container overflow: fc
>   Array cookie: ac
>   Intra object redzone: bb
>   ASan internal: fe
>   Left alloca redzone: ca
>   Right alloca redzone: cb
>   Shadow gap: cc
> ==31958==ABORTING
> 
> Signed-off-by: Yury Kotov <yury-kotov@yandex-team.ru>
> ---
>  migration/migration.c | 23 ++++++++++++++++++-----
>  1 file changed, 18 insertions(+), 5 deletions(-)
> 
> diff --git a/migration/migration.c b/migration/migration.c
> index 609e0df5d0..b9848d1030 100644
> --- a/migration/migration.c
> +++ b/migration/migration.c
> @@ -1495,10 +1495,8 @@ static void block_cleanup_parameters(MigrationState *s)
>      }
>  }
>  
> -static void migrate_fd_cleanup(void *opaque)
> +static void migrate_fd_cleanup(MigrationState *s)
>  {
> -    MigrationState *s = opaque;
> -
>      qemu_bh_delete(s->cleanup_bh);
>      s->cleanup_bh = NULL;
>  
> @@ -1543,6 +1541,21 @@ static void migrate_fd_cleanup(void *opaque)
>      block_cleanup_parameters(s);
>  }
>  
> +static void migrate_fd_cleanup_schedule(MigrationState *s)
> +{
> +    /* Ref the state for bh, because it may be called when
> +     * there're already no other refs */
> +    object_ref(OBJECT(s));
> +    qemu_bh_schedule(s->cleanup_bh);
> +}
> +
> +static void migrate_fd_cleanup_bh(void *opaque)
> +{
> +    MigrationState *s = opaque;
> +    migrate_fd_cleanup(s);
> +    object_unref(OBJECT(s));
> +}
> +
>  void migrate_set_error(MigrationState *s, const Error *error)
>  {
>      qemu_mutex_lock(&s->error_mutex);
> @@ -3144,7 +3157,7 @@ static void migration_iteration_finish(MigrationState *s)
>          error_report("%s: Unknown ending state %d", __func__, s->state);
>          break;
>      }
> -    qemu_bh_schedule(s->cleanup_bh);
> +    migrate_fd_cleanup_schedule(s);
>      qemu_mutex_unlock_iothread();
>  }
>  
> @@ -3279,7 +3292,7 @@ void migrate_fd_connect(MigrationState *s, Error *error_in)
>      bool resume = s->state == MIGRATION_STATUS_POSTCOPY_PAUSED;
>  
>      s->expected_downtime = s->parameters.downtime_limit;
> -    s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup, s);
> +    s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup_bh, s);
>      if (error_in) {
>          migrate_fd_error(s, error_in);
>          migrate_fd_cleanup(s);
> -- 
> 2.21.0
> 
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK

Vladimir Sementsov-Ogievskiy Sept. 13, 2019, 1:43 p.m. UTC | #4

Hi!

08.04.2019 14:33, Yury Kotov wrote:
> It fixes heap-use-after-free which was found by clang's ASAN.
> 
> Control flow of this use-after-free:
> main_thread:
>      * Got SIGTERM and completes main loop
>      * Calls migration_shutdown
>        - migrate_fd_cancel (so, migration_thread begins to complete)
>        - object_unref(OBJECT(current_migration));
> 
> migration_thread:
>      * migration_iteration_finish -> schedule cleanup bh
>      * object_unref(OBJECT(s)); (Now, current_migration is freed)
>      * exits
> 
> main_thread:
>      * Calls vm_shutdown -> drain bdrvs -> main loop
>        -> cleanup_bh -> use after free
> 


[..]

>   
> +static void migrate_fd_cleanup_schedule(MigrationState *s)
> +{
> +    /* Ref the state for bh, because it may be called when
> +     * there're already no other refs */
> +    object_ref(OBJECT(s));
> +    qemu_bh_schedule(s->cleanup_bh);
> +}

so you ref on scheduling

> +
> +static void migrate_fd_cleanup_bh(void *opaque)
> +{
> +    MigrationState *s = opaque;
> +    migrate_fd_cleanup(s);
> +    object_unref(OBJECT(s));
> +}

and unref after execution of bh.

but migration code has also call to qemu_bh_delete(s->cleanup_bh) from
migrate_fd_cleanup(), in migrate_fd_cleanup() is called not only from
migrate_fd_cleanup_bh..

I mean, that if we call qemu_bh_delete after scheduling, we will not
unref our new reference.

Still, seems it's impossible, as all other calls to migrate_fd_cleanup
are done before migration thread creation.

Don't know, should something done around it or not, but decided to mention it.


> +
>   void migrate_set_error(MigrationState *s, const Error *error)
>   {
>       qemu_mutex_lock(&s->error_mutex);
> @@ -3144,7 +3157,7 @@ static void migration_iteration_finish(MigrationState *s)
>           error_report("%s: Unknown ending state %d", __func__, s->state);
>           break;
>       }
> -    qemu_bh_schedule(s->cleanup_bh);
> +    migrate_fd_cleanup_schedule(s);
>       qemu_mutex_unlock_iothread();
>   }
>   
> @@ -3279,7 +3292,7 @@ void migrate_fd_connect(MigrationState *s, Error *error_in)
>       bool resume = s->state == MIGRATION_STATUS_POSTCOPY_PAUSED;
>   
>       s->expected_downtime = s->parameters.downtime_limit;
> -    s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup, s);
> +    s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup_bh, s);
>       if (error_in) {
>           migrate_fd_error(s, error_in);
>           migrate_fd_cleanup(s);
>

Yury Kotov Sept. 13, 2019, 2:42 p.m. UTC | #5

Hi Vladimir!

13.09.2019, 16:43, "Vladimir Sementsov-Ogievskiy" <vsementsov@virtuozzo.com>:
> Hi!
>
> 08.04.2019 14:33, Yury Kotov wrote:
>>  It fixes heap-use-after-free which was found by clang's ASAN.
>>
>>  Control flow of this use-after-free:
>>  main_thread:
>>       * Got SIGTERM and completes main loop
>>       * Calls migration_shutdown
>>         - migrate_fd_cancel (so, migration_thread begins to complete)
>>         - object_unref(OBJECT(current_migration));
>>
>>  migration_thread:
>>       * migration_iteration_finish -> schedule cleanup bh
>>       * object_unref(OBJECT(s)); (Now, current_migration is freed)
>>       * exits
>>
>>  main_thread:
>>       * Calls vm_shutdown -> drain bdrvs -> main loop
>>         -> cleanup_bh -> use after free
>
> [..]
>
>>  +static void migrate_fd_cleanup_schedule(MigrationState *s)
>>  +{
>>  + /* Ref the state for bh, because it may be called when
>>  + * there're already no other refs */
>>  + object_ref(OBJECT(s));
>>  + qemu_bh_schedule(s->cleanup_bh);
>>  +}
>
> so you ref on scheduling
>
>>  +
>>  +static void migrate_fd_cleanup_bh(void *opaque)
>>  +{
>>  + MigrationState *s = opaque;
>>  + migrate_fd_cleanup(s);
>>  + object_unref(OBJECT(s));
>>  +}
>
> and unref after execution of bh.
>
> but migration code has also call to qemu_bh_delete(s->cleanup_bh) from
> migrate_fd_cleanup(), in migrate_fd_cleanup() is called not only from
> migrate_fd_cleanup_bh..
>
> I mean, that if we call qemu_bh_delete after scheduling, we will not
> unref our new reference.
>
> Still, seems it's impossible, as all other calls to migrate_fd_cleanup
> are done before migration thread creation.
>
> Don't know, should something done around it or not, but decided to mention it.
>

Hmm.. It's very interesting, thanks! I need more time to understand
is there a problem or not, but after quick look I see one possibility
to achieve a leak: qmp_migrate after cleanup bh schedule and before bh call...

>>  +
>>    void migrate_set_error(MigrationState *s, const Error *error)
>>    {
>>        qemu_mutex_lock(&s->error_mutex);
>>  @@ -3144,7 +3157,7 @@ static void migration_iteration_finish(MigrationState *s)
>>            error_report("%s: Unknown ending state %d", __func__, s->state);
>>            break;
>>        }
>>  - qemu_bh_schedule(s->cleanup_bh);
>>  + migrate_fd_cleanup_schedule(s);
>>        qemu_mutex_unlock_iothread();
>>    }
>>
>>  @@ -3279,7 +3292,7 @@ void migrate_fd_connect(MigrationState *s, Error *error_in)
>>        bool resume = s->state == MIGRATION_STATUS_POSTCOPY_PAUSED;
>>
>>        s->expected_downtime = s->parameters.downtime_limit;
>>  - s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup, s);
>>  + s->cleanup_bh = qemu_bh_new(migrate_fd_cleanup_bh, s);
>>        if (error_in) {
>>            migrate_fd_error(s, error_in);
>>            migrate_fd_cleanup(s);
>
> --
> Best regards,
> Vladimir

Regards,
Yury

migration: Fix use-after-free during process exit

Commit Message

Comments

Patch