Re: [PATCH v4 10/25] migration: Add Error** argument to qemu_savevm_state_setup()

[Fabiano Rosas <farosas@suse.de>]

Peter Xu <peterx@redhat.com> writes:

> On Wed, Mar 06, 2024 at 02:34:25PM +0100, Cédric Le Goater wrote:
>> This prepares ground for the changes coming next which add an Error**
>> argument to the .save_setup() handler. Callers of qemu_savevm_state_setup()
>> now handle the error and fail earlier setting the migration state from
>> MIGRATION_STATUS_SETUP to MIGRATION_STATUS_FAILED.
>> 
>> In qemu_savevm_state(), move the cleanup to preserve the error
>> reported by .save_setup() handlers.
>> 
>> Since the previous behavior was to ignore errors at this step of
>> migration, this change should be examined closely to check that
>> cleanups are still correctly done.
>> 
>> Signed-off-by: Cédric Le Goater <clg@redhat.com>
>> ---
>> 
>>  Changes in v4:
>>  
>>  - Merged cleanup change in qemu_savevm_state()
>>    
>>  Changes in v3:
>>  
>>  - Set migration state to MIGRATION_STATUS_FAILED 
>>  - Fixed error handling to be done under lock in bg_migration_thread()
>>  - Made sure an error is always set in case of failure in
>>    qemu_savevm_state_setup()
>>    
>>  migration/savevm.h    |  2 +-
>>  migration/migration.c | 27 ++++++++++++++++++++++++---
>>  migration/savevm.c    | 26 +++++++++++++++-----------
>>  3 files changed, 40 insertions(+), 15 deletions(-)
>> 
>> diff --git a/migration/savevm.h b/migration/savevm.h
>> index 74669733dd63a080b765866c703234a5c4939223..9ec96a995c93a42aad621595f0ed58596c532328 100644
>> --- a/migration/savevm.h
>> +++ b/migration/savevm.h
>> @@ -32,7 +32,7 @@
>>  bool qemu_savevm_state_blocked(Error **errp);
>>  void qemu_savevm_non_migratable_list(strList **reasons);
>>  int qemu_savevm_state_prepare(Error **errp);
>> -void qemu_savevm_state_setup(QEMUFile *f);
>> +int qemu_savevm_state_setup(QEMUFile *f, Error **errp);
>>  bool qemu_savevm_state_guest_unplug_pending(void);
>>  int qemu_savevm_state_resume_prepare(MigrationState *s);
>>  void qemu_savevm_state_header(QEMUFile *f);
>> diff --git a/migration/migration.c b/migration/migration.c
>> index a49fcd53ee19df1ce0182bc99d7e064968f0317b..6d1544224e96f5edfe56939a9c8395d88ef29581 100644
>> --- a/migration/migration.c
>> +++ b/migration/migration.c
>> @@ -3408,6 +3408,8 @@ static void *migration_thread(void *opaque)
>>      int64_t setup_start = qemu_clock_get_ms(QEMU_CLOCK_HOST);
>>      MigThrError thr_error;
>>      bool urgent = false;
>> +    Error *local_err = NULL;
>> +    int ret;
>>  
>>      thread = migration_threads_add("live_migration", qemu_get_thread_id());
>>  
>> @@ -3451,9 +3453,17 @@ static void *migration_thread(void *opaque)
>>      }
>>  
>>      bql_lock();
>> -    qemu_savevm_state_setup(s->to_dst_file);
>> +    ret = qemu_savevm_state_setup(s->to_dst_file, &local_err);
>>      bql_unlock();
>>  
>> +    if (ret) {
>> +        migrate_set_error(s, local_err);
>> +        error_free(local_err);
>> +        migrate_set_state(&s->state, MIGRATION_STATUS_SETUP,
>> +                          MIGRATION_STATUS_FAILED);
>> +        goto out;
>> +     }
>
> There's a small indent issue, I can fix it.
>
> The bigger problem is I _think_ this will trigger a ci failure in the
> virtio-net-failover test:
>
> ▶ 121/464 ERROR:../tests/qtest/virtio-net-failover.c:1203:test_migrate_abort_wait_unplug: assertion failed (status == "cancelling"): ("cancelled" == "cancelling") ERROR         
> 121/464 qemu:qtest+qtest-x86_64 / qtest-x86_64/virtio-net-failover    ERROR            4.77s   killed by signal 6 SIGABRT
>>>> PYTHON=/builds/peterx/qemu/build/pyvenv/bin/python3.8 G_TEST_DBUS_DAEMON=/builds/peterx/qemu/tests/dbus-vmstate-daemon.sh MALLOC_PERTURB_=161 QTEST_QEMU_IMG=./qemu-img QTEST_QEMU_STORAGE_DAEMON_BINARY=./storage-daemon/qemu-storage-daemon QTEST_QEMU_BINARY=./qemu-system-x86_64 /builds/peterx/qemu/build/tests/qtest/virtio-net-failover --tap -k
> ――――――――――――――――――――――――――――――――――――― ✀  ―――――――――――――――――――――――――――――――――――――
> stderr:
> qemu-system-x86_64: ram_save_setup failed: Input/output error
> **
> ERROR:../tests/qtest/virtio-net-failover.c:1203:test_migrate_abort_wait_unplug: assertion failed (status == "cancelling"): ("cancelled" == "cancelling")

I would say testing for the CANCELLING state is unreliable, there's even
code a few lines below in the test that does the proper thing of testing
for CANCELLED in a loop.

However, the comment: 

 /* while the card is not ejected, we must be in "cancelling" state */

seems to imply that after migrate_fd_cancel (state==CANCELLING), the
migrate_fd_cleanup (state==CANCELLED) would only be executed after
"unplugging the card". So there must be "some logic" that has the effect
of preventing cleanup.

> (test program exited with status code -6)
> ――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――
>
> I am not familiar enough with the failover code, and may not have time
> today to follow this up, copy Laurent.  Cedric, if you have time, please
> have a look.  I'll give it a shot on Monday to find a solution, otherwise
> we may need to postpone some of the patches to 9.1.
>
> Thanks,
>
>> +
>>      qemu_savevm_wait_unplug(s, MIGRATION_STATUS_SETUP,
>>                                 MIGRATION_STATUS_ACTIVE);
>>  
>> @@ -3530,6 +3540,9 @@ static void *bg_migration_thread(void *opaque)
>>      MigThrError thr_error;
>>      QEMUFile *fb;
>>      bool early_fail = true;
>> +    bool setup_fail = true;
>> +    Error *local_err = NULL;
>> +    int ret;
>>  
>>      rcu_register_thread();
>>      object_ref(OBJECT(s));
>> @@ -3563,9 +3576,16 @@ static void *bg_migration_thread(void *opaque)
>>  
>>      bql_lock();
>>      qemu_savevm_state_header(s->to_dst_file);
>> -    qemu_savevm_state_setup(s->to_dst_file);
>> +    ret = qemu_savevm_state_setup(s->to_dst_file, &local_err);
>> +    if (ret) {
>> +        migrate_set_error(s, local_err);
>> +        error_free(local_err);
>> +        goto fail;
>> +    }
>>      bql_unlock();
>>  
>> +    setup_fail = false;
>> +
>>      qemu_savevm_wait_unplug(s, MIGRATION_STATUS_SETUP,
>>                                 MIGRATION_STATUS_ACTIVE);
>>  
>> @@ -3632,7 +3652,8 @@ static void *bg_migration_thread(void *opaque)
>>  
>>  fail:
>>      if (early_fail) {
>> -        migrate_set_state(&s->state, MIGRATION_STATUS_ACTIVE,
>> +        migrate_set_state(&s->state,
>> +                setup_fail ? MIGRATION_STATUS_SETUP : MIGRATION_STATUS_ACTIVE,
>>                  MIGRATION_STATUS_FAILED);
>>          bql_unlock();
>>      }
>> diff --git a/migration/savevm.c b/migration/savevm.c
>> index ee31ffb5e88cea723039c754c30ce2c8a0ef35f3..63fdbb5ad7d4dbfaef1d2094350bf302cc677602 100644
>> --- a/migration/savevm.c
>> +++ b/migration/savevm.c
>> @@ -1310,11 +1310,11 @@ int qemu_savevm_state_prepare(Error **errp)
>>      return 0;
>>  }
>>  
>> -void qemu_savevm_state_setup(QEMUFile *f)
>> +int qemu_savevm_state_setup(QEMUFile *f, Error **errp)
>>  {
>> +    ERRP_GUARD();
>>      MigrationState *ms = migrate_get_current();
>>      SaveStateEntry *se;
>> -    Error *local_err = NULL;
>>      int ret = 0;
>>  
>>      json_writer_int64(ms->vmdesc, "page_size", qemu_target_page_size());
>> @@ -1323,10 +1323,9 @@ void qemu_savevm_state_setup(QEMUFile *f)
>>      trace_savevm_state_setup();
>>      QTAILQ_FOREACH(se, &savevm_state.handlers, entry) {
>>          if (se->vmsd && se->vmsd->early_setup) {
>> -            ret = vmstate_save(f, se, ms->vmdesc, &local_err);
>> +            ret = vmstate_save(f, se, ms->vmdesc, errp);
>>              if (ret) {
>> -                migrate_set_error(ms, local_err);
>> -                error_report_err(local_err);
>> +                migrate_set_error(ms, *errp);
>>                  qemu_file_set_error(f, ret);
>>                  break;
>>              }
>> @@ -1346,18 +1345,19 @@ void qemu_savevm_state_setup(QEMUFile *f)
>>          ret = se->ops->save_setup(f, se->opaque);
>>          save_section_footer(f, se);
>>          if (ret < 0) {
>> +            error_setg(errp, "failed to setup SaveStateEntry with id(name): "
>> +                       "%d(%s): %d", se->section_id, se->idstr, ret);
>>              qemu_file_set_error(f, ret);
>>              break;
>>          }
>>      }
>>  
>>      if (ret) {
>> -        return;
>> +        return ret;
>>      }
>>  
>> -    if (precopy_notify(PRECOPY_NOTIFY_SETUP, &local_err)) {
>> -        error_report_err(local_err);
>> -    }
>> +    /* TODO: Should we check that errp is set in case of failure ? */
>> +    return precopy_notify(PRECOPY_NOTIFY_SETUP, errp);
>>  }
>>  
>>  int qemu_savevm_state_resume_prepare(MigrationState *s)
>> @@ -1728,7 +1728,10 @@ static int qemu_savevm_state(QEMUFile *f, Error **errp)
>>      ms->to_dst_file = f;
>>  
>>      qemu_savevm_state_header(f);
>> -    qemu_savevm_state_setup(f);
>> +    ret = qemu_savevm_state_setup(f, errp);
>> +    if (ret) {
>> +        goto cleanup;
>> +    }
>>  
>>      while (qemu_file_get_error(f) == 0) {
>>          if (qemu_savevm_state_iterate(f, false) > 0) {
>> @@ -1741,10 +1744,11 @@ static int qemu_savevm_state(QEMUFile *f, Error **errp)
>>          qemu_savevm_state_complete_precopy(f, false, false);
>>          ret = qemu_file_get_error(f);
>>      }
>> -    qemu_savevm_state_cleanup();
>>      if (ret != 0) {
>>          error_setg_errno(errp, -ret, "Error while writing VM state");
>>      }
>> +cleanup:
>> +    qemu_savevm_state_cleanup();
>>  
>>      if (ret != 0) {
>>          status = MIGRATION_STATUS_FAILED;
>> -- 
>> 2.44.0
>>