maria-developers team mailing list archive

[Sergei Golubchik <serg@xxxxxxxxxxx>]

Hi, Jan!

On Oct 04, Jan Lindström wrote:
> Hi Sergei,
> > > +/* This is wrapper for wsrep_break_lock in thr_lock.c */
> > > +static int wsrep_thr_abort_thd(void *bf_thd_ptr, void *victim_thd_ptr, my_bool signal)
> > > +{
> > > +  THD* victim_thd= (THD *) victim_thd_ptr;
> > > +  /* We need to lock THD::LOCK_thd_data to protect victim
> > > +  from concurrent usage or disconnect or delete. */
> >
> > How do you know victim_thd wasn't deleted before you locked
> > LOCK_thd_data below?
> 
> I must say the thr_lock code is not familiar to me but there are
> mysql_mutex_lock() calls to lock->mutex. After code review  it is not
> clear to me what that mutex is.

where are mysql_mutex_lock() calls to lock->mutex?

> > >          if (victim_trx) {
> > > +                wsrep_thd_UNLOCK(victim_thd);
> >
> > what keeps victim_trx from disappearing here?
> 
> Nothing. Do you have suggestions ?

A couple of thoughts:

* Why do you unlock LOCK_thd_data here at all? I believed the whole
  point of using TOI was to make sure that even if you lock mutexes in a
  different order, it will not cause a deadlock.

* You cannot pass a THD safely over a gap where no locks keep it in
  existence. You can pass an integer, thread id, and use
  find_thread_by_id later.

By the way, how long can WSREP_TO_ISOLATION_BEGIN() take?
Does it need to wait for everything else being replicated?

Regards,
Sergei
VP of MariaDB Server Engineering
and security@xxxxxxxxxxx