Tom Lane <tgl@sss.pgh.pa.us> writes:
> Greg Stark <gsstark@mit.edu> writes:
>
> Two things: one, backends dump gmon.out into $PGDATA/base/yourdb/,
> whereas the postmaster dumps into wherever you started it (it never does
> chdir IIRC). Two, if you're on Linux you must compile postmaster.c with
> -DLINUX_PROFILE, or you won't get any profiling timer interrupts in
> child processes.
>
> Your sample looks like the second problem...
Actually I had both problems, I didn't realize the backends did a chdir.
I cancelled it partway through.
% cumulative self self total
time seconds seconds calls s/call s/call name
16.59 97.88 97.88 69608411 0.00 0.00 ExecMakeFunctionResult
7.08 139.65 41.77 79472258 0.00 0.00 comparetup_heap
4.50 166.17 26.52 192211935 0.00 0.00 ExecEvalExpr
3.94 189.41 23.24 155884536 0.00 0.00 AllocSetAlloc
2.57 204.59 15.18 69608411 0.00 0.00 ExecEvalFuncArgs
2.57 219.76 15.17 17360796 0.00 0.00 heap_formtuple
2.25 233.03 13.27 248222443 0.00 0.00 AllocSetFreeIndex
2.22 246.11 13.08 17360724 0.00 0.00 ExecTargetList
2.17 258.92 12.81 45871623 0.00 0.00 ExecEvalParam
1.99 270.65 11.73 17360796 0.00 0.00 DataFill
1.82 281.37 10.72 23020517 0.00 0.00 AllocSetReset
1.57 290.62 9.25 29962044 0.00 0.00 ExecEvalVar
1.29 298.20 7.58 17360796 0.00 0.00 ComputeDataSize
1.17 305.09 6.89 92337151 0.00 0.00 AllocSetFree
1.13 311.74 6.64 97183879 0.00 0.00 myFunctionCall2
1.11 318.27 6.54 155883526 0.00 0.00 MemoryContextAlloc
1.07 324.57 6.30 97183879 0.00 0.00 inlineApplySortFunction
1.05 330.75 6.17 17285775 0.00 0.00 newNode
--
greg
