BUG #15887: An extra "break" instruction causes a bug - Mailing list pgsql-bugs

From PG Bug reporting form
Subject BUG #15887: An extra "break" instruction causes a bug
Date
Msg-id 15887-e9320033d84cf7ea@postgresql.org
Whole thread Raw
List pgsql-bugs
The following bug has been logged on the website:

Bug reference:      15887
Logged by:          Rui Hai Jiang
Email address:      ruihaij@gmail.com
PostgreSQL version: 12beta2
Operating system:   all
Description:

In function TopoSort(), an extra "break" instruction make it impossible to
set beforeConstraints to -1 for other 
processes in the same group.  If the waiter's group has more than one
members in the wait for queue, the sorting couldn't get the right order. 

The issue exists in all V10,V11,V12 releases.

static bool
TopoSort(LOCK *lock,
                 EDGE *constraints,
                 int nConstraints,
                 PGPROC **ordering)             /* output argument */
{
     MemSet(beforeConstraints, 0, queue_size * sizeof(int));
        MemSet(afterConstraints, 0, queue_size * sizeof(int));
        for (i = 0; i < nConstraints; i++)
        {
                /*
                 * Find a representative process that is on the lock queue
and part of
                 * the waiting lock group.  This may or may not be the
leader, which
                 * may or may not be waiting at all.  If there are any other
processes
                 * in the same lock group on the queue, set their number
of
                 * beforeConstraints to -1 to indicate that they should be
emitted
                 * with their groupmates rather than considered
separately.
                 */
                proc = constraints[i].waiter;
                Assert(proc != NULL);
                jj = -1;
                for (j = queue_size; --j >= 0;)
                {
                        PGPROC     *waiter = topoProcs[j];

                        if (waiter == proc || waiter->lockGroupLeader ==
proc)
                        {
                                Assert(waiter->waitLock == lock);
                                if (jj == -1)
                                        jj = j;
                                else
                                {
                                        Assert(beforeConstraints[j] <= 0);
                                        beforeConstraints[j] = -1;
                                }
                                break;
                        }
                }
        }
}


A proposed patch is also sent to the hacker' list.

https://www.postgresql.org/message-id/CAEri+mLd3bpHLyW+a9pSe1y=aEkeuJpwBSwvo-+m4n7-ceRmXw@mail.gmail.com


pgsql-bugs by date:

Previous
From: Andrew Gierth
Date:
Subject: Re: SELECT results in "ERROR: index key does not match expected index column"
Next
From: PG Bug reporting form
Date:
Subject: BUG #15888: Bogus "idle in transaction" state for logical decoding client after creating a slot