Core dump in regression tests. - Mailing list pgsql-hackers

If anyone has been watching my trials and tribulations building
and running the latest CVS snapshot under S/Linux on a SUN
SPARCstation IPX.

The latest position is:-

If I compile with optimization turned off (-O0 instead of -O2)
I get an almost clean run of the regressin tests. Only the
"create_function" tests fail unexpectedly.

[postgres@sparclinux regress]$ psql regression
Welcome to the POSTGRESQL interactive sql monitor:
  Please read the file COPYRIGHT for copyright terms of POSTGRESQL

   type \? for help on slash commands
   type \q to quit
   type \g or terminate with semicolon to execute query
 You are currently connected to the database: regression

regression=>  CREATE FUNCTION widget_in(opaque)
regression->    RETURNS widget
regression->    AS '/usr/local/pgsql/src/test/regress/input/../regress.so'
regression->    LANGUAGE 'c';
NOTICE: ProcedureCreate: type 'widget' is not yet defined
pqReadData() -- backend closed the channel unexpectedly.
        This probably means the backend terminated abnormally before or while
processing the request.
We have lost the connection to the backend, so further processing is impossible.
 Terminating.
[postgres@sparclinux regress]$

The backtrace shows:-

Program received signal SIGSEGV, Segmentation fault.
0x44744 in GetIndexValue (tuple=0x25e210, hTupDesc=0x25e26c, attOff=0,
attrNums=0x261944, fInfo=0x0,
    attNull=0xefffcbcf "") at indexam.c:404
404                     returnVal = heap_getattr(tuple, attrNums[attOff],
(gdb) bt
#0  0x44744 in GetIndexValue (tuple=0x25e210, hTupDesc=0x25e26c, attOff=0,
attrNums=0x261944, fInfo=0x0,
    attNull=0xefffcbcf "") at indexam.c:404
#1  0x68e9c in FormIndexDatum (numberOfAttributes=1, attributeNumber=0x261944,
heapTuple=0x25e210,
    heapDescriptor=0x25e26c, datum=0xefffccb8, nullv=0xefffcc58 "   ",
fInfo=0x0) at index.c:1284
#2  0x69c38 in CatalogIndexInsert (idescs=0xefffcd30, nIndices=3,
heapRelation=0x213b90, heapTuple=0x25e210)
    at indexing.c:154
#3  0x6f344 in ProcedureCreate (procedureName=0x208bb0 "widget_in", returnsSet=0
'\000',
    returnTypeName=0x208b30 "widget", languageName=0xefffcec8 "C",
prosrc=0x18d7f8 "-",
    probin=0x251b10 "/usr/local/pgsql/src/test/regress/input/../regress.so",
canCache=24 '\030',
    trusted=1 '\001', byte_pct=100, perbyte_cpu=0, percall_cpu=0,
outin_ratio=100, argList=0x208b50,
    dest=Remote) at pg_proc.c:275
#4  0x786c8 in CreateFunction (stmt=0x207650, dest=Remote) at define.c:329
#5  0x131694 in ProcessUtility (parsetree=0x207650, dest=Remote) at
utility.c:392
#6  0x12db18 in pg_exec_query_dest (
    query_string=0xefffd130 " CREATE FUNCTION widget_in(opaque)\n   RETURNS
widget\n   AS '/usr/local/pgsql/src/test/regress/input/../regress.so'\n
LANGUAGE 'c';", dest=Remote, aclOverride=0 '\000') at postgres.c:720
#7  0x12d98c in pg_exec_query (
    query_string=0xefffd130 " CREATE FUNCTION widget_in(opaque)\n   RETURNS
widget\n   AS '/usr/local/pgsql/src/test/regress/input/../regress.so'\n
LANGUAGE 'c';") at postgres.c:658
#8  0x12f81c in PostgresMain (argc=9, argv=0xeffff210, real_argc=10,
real_argv=0xeffffd84) at postgres.c:1578
#9  0x10183c in DoBackend (port=0x209000) at postmaster.c:1519
#10 0x100ffc in BackendStartup (port=0x209000) at postmaster.c:1291
#11 0xffed8 in ServerLoop () at postmaster.c:750
#12 0xff860 in PostmasterMain (argc=10, argv=0xeffffd84) at postmaster.c:556
#13 0xb02c0 in main (argc=10, argv=0xeffffd84) at main.c:93
(gdb)

Has anyone else seen anything similar?

Keith.

PS: Bruce, I still need to find which file breaks an -O2 compile but
    will spend some time playing over this long weekend.