To: Kevin.Grittner@wicourts.gov
Cc: pgsql-hackers@postgresql.org
Subject: Re: SSI bug?
In-Reply-To: Your message of "Fri, 11 Feb 2011 09:48:18 -0600"
	<4D5505E2020000250003A83C@gw.wicourts.gov>
References: <4D5505E2020000250003A83C@gw.wicourts.gov>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; boundary="Boundary-20110212160545-1048700"
Message-Id: <20110212070822.46BE319D15D@mail.netbsd.org>
Date: Sat, 12 Feb 2011 07:08:21 +0000 (UTC)
From: yamt@mwd.biglobe.ne.jp (YAMAMOTO Takashi)

--Boundary-20110212160545-1048700
Content-Type: Text/Plain; charset=us-ascii

hi,

> YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp> wrote:
>  
>> it seems that PredicateLockTupleRowVersionLink sometimes create
>> a loop of targets (it founds an existing 'newtarget' whose
>> nextVersionOfRow chain points to the 'oldtarget') and it later
>> causes CheckTargetForConflictsIn loop forever.
>  
> Is this a hypothetical risk based on looking at the code, or have
> you seen this actually happen?  Either way, could you provide more
> details?  (A reproducible test case would be ideal.)

i have seen this actually happen.  i've confirmed the creation of the loop
with the attached patch.  it's easily reproducable with my application.
i can provide the full source code of my application if you want.
(but it isn't easy to run unless you are familiar with the recent
version of NetBSD)
i haven't found a smaller reproducible test case yet.

YAMAMOTO Takashi

>  
> This being the newest part of the code, I'll grant that it is the
> most likely to have an unidentified bug; but given that the pointers
> are from one predicate lock target structure identified by a tuple
> ID to one identified by the tuple ID of the next version of the row,
> it isn't obvious to me how a cycle could develop.
>  
> -Kevin

--Boundary-20110212160545-1048700
Content-Type: Text/Plain; charset=us-ascii
Content-Disposition: attachment; filename="a.diff"

diff --git a/src/backend/storage/lmgr/predicate.c b/src/backend/storage/lmgr/predicate.c
index 722d0f8..3e1a3e2 100644
--- a/src/backend/storage/lmgr/predicate.c
+++ b/src/backend/storage/lmgr/predicate.c
@@ -2350,7 +2350,25 @@ PredicateLockTupleRowVersionLink(const Relation relation,
 			newtarget->nextVersionOfRow = NULL;
 		}
 		else
+		{
 			Assert(newtarget->priorVersionOfRow == NULL);
+#if 0
+			Assert(newtarget->nextVersionOfRow == NULL);
+#endif
+			if (newtarget->nextVersionOfRow != NULL) {
+				PREDICATELOCKTARGET *t;
+
+				elog(WARNING, "new %p has next %p\n",
+					newtarget, newtarget->nextVersionOfRow);
+				for (t = newtarget->nextVersionOfRow; t != NULL;
+					t = t->nextVersionOfRow) {
+					if (oldtarget != t) {
+						elog(WARNING, "creating a loop new=%p old=%p\n",
+							newtarget, oldtarget);
+					}
+				}
+			}
+		}
 
 		newtarget->priorVersionOfRow = oldtarget;
 		oldtarget->nextVersionOfRow = newtarget;

--Boundary-20110212160545-1048700--