Re: itaps-parallel [Fwd: Re: problem with data]

14 Nov 2008


      I don't get that behavior.  Are you sure it's something wrong with the 
iBase.h and not the application?  I've compiled all the Example 
applications with C using the original enum definition...

- tim

Mark Miller wrote:
...
I found this in working with FMDB but then thought it might be important
to people in general...
------------------------------------------------------------------------
Subject:
Re: problem with data
From:
Mark Miller <[email protected]>
Date:
Fri, 14 Nov 2008 06:07:29 -0800
To:
[email protected]
To:
[email protected]
CC:
Onkar Sahni <[email protected]>
enum def'n in iBase.h are somehow incompatibile with C compiler.
I modified hello_iMeshP.c with iMeshP_getAllVtxCoords and needed
iBase_StorageOrder. However, when I tried to compile with C compiler on
mesh machine, it would fail...
typedef enum {
    iBase_BLOCKED,
    iBase_INTERLEAVED,
    iBase_UNDETERMINED
  } iBase_StorageOrder;
hello_iMeshP.c: In function ï¿½mainï¿½:
hello_iMeshP.c:54: error: ï¿½iBase_StorageOrderï¿½ undeclared (first use in
this function)
hello_iMeshP.c:54: error: (Each undeclared identifier is reported only
once
hello_iMeshP.c:54: error: for each function it appears in.)
hello_iMeshP.c:54: error: expected ï¿½;ï¿½ before ï¿½storageOrder2ï¿½
hello_iMeshP.c:56: error: ï¿½storageOrder2ï¿½ undeclared (first use in this
function)
To correct it, I think all the enums in iBase.h need to be re-formatted
like so...
typedef enum {
    iBase_BLOCKED,
    iBase_INTERLEAVED,
    iBase_UNDETERMINED
  } iBase_StorageOrder;
On Fri, 2008-11-14 at 06:04, [email protected] wrote:
...
Hi, Mark,
Currently we assume that one part per process in the FMDB_iMeshP
implementation. And imesh_load can be used in more than one process.
Thanks,
Ting
...
imesh_load appears to enter inf. loop if you happen to start on a number
of processors less than number of parts.
Mark
On Thu, 2008-11-13 at 22:44, Mark Miller wrote:
...
Onkar,
I am really puzzled by the data I am getting back from parallel FMDB via
iMesh interface for the mesh_cube_660_4proc test case you gave me.
On proc 0, the call to getallvtxcoords returns 64 coordinate values
which is what I expect. However, the call to imesh_getvtxcoordindex
returns 'connectivity' index offsets that are much larger than 64.
I wonder if somehow I am getting back 'global' coordinate index values
from this method?
Mark
avtITAPS_FMDBFileFormat::GetMesh
Made it past call to "imesh_getentities_" at line 1244 in file
avtITAPS_FMDBFileFormat.C
entRegnHdls_size = 165
Made it past call to "imesh_getentities_" at line 1251 in file
avtITAPS_FMDBFileFormat.C
entFaceHdls_size = 379
Made it past call to "imesh_getentities_" at line 1258 in file
avtITAPS_FMDBFileFormat.C
entCurvHdls_size = 277
Made it past call to "imesh_getentities_" at line 1265 in file
avtITAPS_FMDBFileFormat.C
entVertHdls_size = 64
Made it past call to "imesh_getallvtxcoords_" at line 1274 in file
avtITAPS_FMDBFileFormat.C
1, 1, 0
-1, -1, 0
1, -1, -1
1, 1, 0
1, 1, 1
1, -1, 1
1, -0.333054, 0.661111
1, 0.333892, 1
0.338889, 1, 1
-0.333888, -1, 1.32222
0.332224, -1, 1
0.522005, -1, -1
0.519968, -1, 0.667783
-0.636653, -1, 1
0.00498673, -1, -0.333054
-1, -0.579238, 0
0.580276, 1, 1
0.476439, -0.476542, 0.333892
-0.554779, -0.558122, 0
0.562053, 0.556558, 0.338889
0.194421, -0.703906, 1
0.753433, -0.744433, 0
0.67532, -0.254162, -0.333888
-0.100118, -0.518252, -1
0.536317, 0.0201365, 0
-0.525001, -0.0705034, 0.332224
0.00463185, -0.0108681, -1
1, -0.562141, 0
1, 0.551738, 0.522005
1, 0.553414, -1
1, -0.46401, 0.48233
1, 0.438439, 0.519968
1, -0.00404706, -1
1, -0.01364, 1.03979
-0.413533, -0.358763, -0.636653
0.0333995, -0.382789, -1
-0.00137382, 0.0753535, 0.696981
-0.414604, 0.0216768, 0.00498673
-0.406887, -0.370834, -1
0.0920263, -0.433486, 0.773063
0, 0, -1
-0.472829, 0.0315931, -0.579238
-0.0108454, 0.00129541, 0.47684
0.408, -0.408, 0.580276
0.448358, 0.00914508, 1
0.501977, -0.452777, 0.65203
0.41117, -0.0437623, 0.476439
0.408, -0.408, -0.476542
0.432157, -0.0151216, 0
0, 0.408, -0.554779
0.387662, 0.41085, -0.558122
0.422741, 0.350308, 0
0.368157, 0.402968, 0.562053
0.395943, 0.390392, 0.556558
0.00183323, -0.709481, 0
-0.195426, -0.641512, 0.194421
0.677074, -0.247432, -0.703906
0.669767, 0.24477, 0.378452
0.713723, 0.257543, 0.753433
0.620712, 0.197067, -0.744433
0.245164, 0.667283, 0.720594
0.639625, 0.250531, 0.67532
0.693725, 0.686982, -0.254162
0.203988, -0.204012, 0.390573
Made it past call to "imesh_getvtxcoordindex_" at line 1334 in file
avtITAPS_FMDBFileFormat.C
zoneType = 10: 86, 81, 85, 84,
zoneType = 10: 86, 80, 82, 81,
zoneType = 10: 84, 81, 80, 86,
zoneType = 10: 86, 80, 84, 83,
zoneType = 10: 80, 83, 86, 82,
zoneType = 10: 90, 84, 86, 85,
zoneType = 10: 84, 87, 90, 86,
zoneType = 10: 87, 86, 84, 83,
zoneType = 10: 94, 91, 93, 85,
zoneType = 10: 85, 91, 81, 94,
zoneType = 10: 94, 81, 85, 86,
zoneType = 10: 96, 85, 94, 93,
zoneType = 10: 85, 90, 96, 94,
zoneType = 10: 90, 94, 85, 86,
zoneType = 10: 104, 92, 94, 86,
zoneType = 10: 104, 82, 103, 92,
zoneType = 10: 86, 92, 82, 104,
zoneType = 10: 82, 100, 104, 103,
zoneType = 10: 105, 94, 96, 90,
zoneType = 10: 105, 86, 104, 94,
zoneType = 10: 90, 94, 86, 105,
zoneType = 10: 122, 90, 105, 96,
zoneType = 10: 6, 30, 63, 46,
zoneType = 10: 48, 6, 63, 46,
zoneType = 10: 66, 63, 30, 80,
zoneType = 10: 66, 63, 80, 68,
zoneType = 10: 66, 70, 68, 80,
zoneType = 10: 82, 81, 80, 70,
zoneType = 10: 66, 70, 80, 81,
zoneType = 10: 30, 63, 46, 80,
zoneType = 10: 63, 48, 46, 80,
zoneType = 10: 46, 48, 84, 80,
zoneType = 10: 41, 7, 31, 62,
zoneType = 10: 41, 7, 62, 71,
zoneType = 10: 31, 62, 66, 91,
zoneType = 10: 66, 81, 31, 91,
zoneType = 10: 31, 62, 91, 41,
zoneType = 10: 82, 92, 70, 103,
zoneType = 10: 67, 62, 24, 91,
zoneType = 10: 91, 67, 92, 24,
zoneType = 10: 67, 92, 24, 78,
zoneType = 10: 7, 24, 62, 71,
zoneType = 10: 24, 62, 71, 91,
zoneType = 10: 47, 44, 41, 93,
zoneType = 10: 41, 22, 7, 71,
zoneType = 10: 85, 91, 93, 41,
zoneType = 10: 62, 41, 71, 91,
zoneType = 10: 91, 93, 41, 71,
zoneType = 10: 93, 71, 78, 75,
zoneType = 10: 41, 47, 93, 85,
zoneType = 10: 94, 93, 78, 96,
zoneType = 10: 75, 78, 96, 77,
zoneType = 10: 78, 25, 24, 67,
zoneType = 10: 67, 70, 92, 103,
zoneType = 10: 25, 64, 67, 103,
zoneType = 10: 92, 67, 103, 25,
zoneType = 10: 92, 67, 25, 78,
zoneType = 10: 25, 64, 103, 73,
zoneType = 10: 25, 4, 64, 73,
zoneType = 10: 4, 64, 73, 60,
zoneType = 10: 4, 26, 64, 60,
zoneType = 10: 64, 103, 73, 60,
zoneType = 10: 104, 73, 103, 60,
zoneType = 10: 104, 73, 60, 76,
zoneType = 10: 96, 78, 76, 77,
zoneType = 10: 20, 73, 76, 60,
zoneType = 10: 76, 20, 60, 21,
zoneType = 10: 4, 20, 60, 73,
zoneType = 10: 122, 76, 74, 77,
zoneType = 10: 60, 104, 76, 105,
zoneType = 10: 30, 126, 80, 46,
zoneType = 10: 85, 126, 84, 81,
zoneType = 10: 125, 126, 85, 81,
zoneType = 10: 66, 125, 31, 81,
zoneType = 10: 80, 126, 84, 46,
--
Mark C. Miller, Lawrence Livermore National Laboratory
email: mailto:[email protected]
(M/T/W) (925)-423-5901 (!!LLNL BUSINESS ONLY!!)
(Th/F)  (530)-753-8511 (!!LLNL BUSINESS ONLY!!)
-- 
================================================================
"You will keep in perfect peace him whose mind is
   steadfast, because he trusts in you."               Isaiah 26:3

              Tim Tautges            Argonne National Laboratory
          ([email protected])      (telecommuting from UW-Madison)
          phone: (608) 263-8485      1500 Engineering Dr.
            fax: (608) 263-4499      Madison, WI 53706

Tim Tautges

tags

participants (1)