[85] | 1 | |
---|
| 2 | Hi Ole, |
---|
| 3 | |
---|
| 4 | Yes, that appears to work - no more segmentation faults: |
---|
| 5 | |
---|
| 6 | mpirun -np 2 python demo.py |
---|
| 7 | MPI initialised OK with 2 processors |
---|
| 8 | I am proc 0 of 2 on node fourier |
---|
| 9 | I am proc 1 of 2 on node hopper |
---|
| 10 | Processor 0 sending message "MSGP0" to processor 1 |
---|
| 11 | Processor 1 received message "MSGP0" from processor 0 |
---|
| 12 | Size of msg was 5 bytes |
---|
| 13 | Processor 1 sending msg "MSGP0->P1" to 0 |
---|
| 14 | Processor 0 received message "MSGP0->P1" from processor 1 |
---|
| 15 | Size of msg was 9 bytes |
---|
| 16 | |
---|
| 17 | Thanks! |
---|
| 18 | |
---|
| 19 | Dave |
---|
| 20 | |
---|
| 21 | |
---|
| 22 | |
---|
| 23 | > Date: Sat, 30 Aug 2003 19:47:39 +1000 |
---|
| 24 | > Reply-To: Ole.Nielsen@anu.edu.au |
---|
| 25 | > Content-Type: text/plain; charset=us-ascii |
---|
| 26 | > Content-Disposition: inline |
---|
| 27 | > User-Agent: Mutt/1.5.4i |
---|
| 28 | > From: ole@cuttlefish.anu.edu.au (Ole Moller Nielsen) |
---|
| 29 | > Content-Length: 9270 |
---|
| 30 | > |
---|
| 31 | > |
---|
| 32 | > Hello Dave |
---|
| 33 | > |
---|
| 34 | > Regarding your problem I may have the solution (thanks to input from Doug Orr). |
---|
| 35 | > It seems that newer versions of LAM require the last commandline argument to be NULL |
---|
| 36 | > so try to modify mpiext.c as follows around line 472 and let me know if it works for you. |
---|
| 37 | > The symbol + means add the line and - means remove the line. |
---|
| 38 | > |
---|
| 39 | > Cheers |
---|
| 40 | > Ole |
---|
| 41 | > |
---|
| 42 | > ----PATCH for mpiext.c |
---|
| 43 | > |
---|
| 44 | > /* Reconstruct C-commandline */ |
---|
| 45 | > /* */ |
---|
| 46 | > argc = PyList_Size(input); //Number of commandline arguments |
---|
| 47 | > - argv = (char**) malloc((argc)*sizeof(char*)); |
---|
| 48 | > + argv = (char**) malloc((argc+1)*sizeof(char*)); |
---|
| 49 | > |
---|
| 50 | > for (i=0; i<argc; i++) |
---|
| 51 | > argv[i] = PyString_AsString( PyList_GetItem(input, i) ); |
---|
| 52 | > |
---|
| 53 | > + argv[i] = NULL; |
---|
| 54 | > |
---|
| 55 | > ------------------------------- |
---|
| 56 | > |
---|
| 57 | > |
---|
| 58 | > |
---|
| 59 | > |
---|
| 60 | > On Tue, Aug 05, 2003 at 11:14:20AM -0400, Dave Reed wrote: |
---|
| 61 | > > |
---|
| 62 | > > Hi, |
---|
| 63 | > > |
---|
| 64 | > > I'm trying to install the lam-7.0 version of MPI and pypar on a |
---|
| 65 | > > Solaris 9 system for one of my colleagues. I've got lam-7.0 working |
---|
| 66 | > > fine (successfully ran one of its test cases). I then installed pypar |
---|
| 67 | > > 1.6.4 using: |
---|
| 68 | > > |
---|
| 69 | > > python setup.py build |
---|
| 70 | > > python setup.py install |
---|
| 71 | > > |
---|
| 72 | > > and that worked without any errors; however, I can't seem to get any |
---|
| 73 | > > of the examples to work. |
---|
| 74 | > > |
---|
| 75 | > > Details of the system are: |
---|
| 76 | > > relevant details of the system are: |
---|
| 77 | > > Solaris 9, gcc 3.2.3, python 2.3, Numeric-23.0 |
---|
| 78 | > > |
---|
| 79 | > > When I try: |
---|
| 80 | > > |
---|
| 81 | > > mpirun -np 2 python demo.py |
---|
| 82 | > > ----------------------------------------------------------------------------- |
---|
| 83 | > > It seems that [at least] one of the processes that was started with |
---|
| 84 | > > mpirun did not invoke MPI_INIT before quitting (it is possible that |
---|
| 85 | > > more than one process did not invoke MPI_INIT -- mpirun was only |
---|
| 86 | > > notified of the first one, which was on node n0). |
---|
| 87 | > > |
---|
| 88 | > > mpirun can *only* be used with MPI programs (i.e., programs that |
---|
| 89 | > > invoke MPI_INIT and MPI_FINALIZE). You can use the "lamexec" program |
---|
| 90 | > > to run non-MPI programs over the lambooted nodes. |
---|
| 91 | > > ----------------------------------------------------------------------------- |
---|
| 92 | > > |
---|
| 93 | > > mpirun C python demo.py |
---|
| 94 | > > |
---|
| 95 | > > produces the same results. |
---|
| 96 | > > |
---|
| 97 | > > mpirun C hello |
---|
| 98 | > > |
---|
| 99 | > > runs the hello demo that comes with lam-7.0 fine. |
---|
| 100 | > > |
---|
| 101 | > > when I execute: |
---|
| 102 | > > python |
---|
| 103 | > > import pypar |
---|
| 104 | > > |
---|
| 105 | > > I get a segmentation fault. Here's the backtrace |
---|
| 106 | > > |
---|
| 107 | > > #0 0xfef33400 in strlen () from /usr/lib/libc.so.1 |
---|
| 108 | > > #1 0xfece5c40 in sfh_argv_dup () |
---|
| 109 | > > from /usr/local/lib/python2.3/site-packages/pypar/mpiext.so |
---|
| 110 | > > #2 0xfece67b0 in ao_parse () |
---|
| 111 | > > from /usr/local/lib/python2.3/site-packages/pypar/mpiext.so |
---|
| 112 | > > #3 0xfecb7be0 in lam_linit () |
---|
| 113 | > > from /usr/local/lib/python2.3/site-packages/pypar/mpiext.so |
---|
| 114 | > > #4 0xfecb82fc in lam_mpi_init () |
---|
| 115 | > > from /usr/local/lib/python2.3/site-packages/pypar/mpiext.so |
---|
| 116 | > > #5 0xfecb5a70 in MPI_Init () |
---|
| 117 | > > from /usr/local/lib/python2.3/site-packages/pypar/mpiext.so |
---|
| 118 | > > #6 0xfecb4a08 in init (self=0x1, args=0xffbfc7e0) at lib/pypar/mpiext.c:480 |
---|
| 119 | > > #7 0xff25b340 in PyCFunction_Call (func=0xc5620, arg=0xc6650, kw=0x0) |
---|
| 120 | > > at Objects/methodobject.c:93 |
---|
| 121 | > > #8 0xff29a39c in call_function (pp_stack=0xffbfc968, oparg=1) |
---|
| 122 | > > at Python/ceval.c:3439 |
---|
| 123 | > > #9 0xff297694 in eval_frame (f=0x9a8f0) at Python/ceval.c:2116 |
---|
| 124 | > > #10 0xff2987a8 in PyEval_EvalCodeEx (co=0xbf9a0, globals=0x1, locals=0x0, |
---|
| 125 | > > args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0, |
---|
| 126 | > > closure=0x0) at Python/ceval.c:2663 |
---|
| 127 | > > #11 0xff294c34 in PyEval_EvalCode (co=0xbf9a0, globals=0x0, locals=0xbc5d0) |
---|
| 128 | > > at Python/ceval.c:537 |
---|
| 129 | > > #12 0xff2bd2fc in PyImport_ExecCodeModuleEx (name=0xffbfd558 "pypar.pypar", |
---|
| 130 | > > co=0xbf9a0, |
---|
| 131 | > > pathname=0xffbfcb70 "/usr/local/lib/python2.3/site-packages/pypar/pypar.pyc") at Python/import.c:621 |
---|
| 132 | > > #13 0xff2c016c in load_source_module (name=0xffbfd558 "pypar.pypar", |
---|
| 133 | > > pathname=0xffbfcb70 "/usr/local/lib/python2.3/site-packages/pypar/pypar.pyc", fp=0x22e8) at Python/import.c:894 |
---|
| 134 | > > #14 0xff2bdf64 in load_module (name=0xffbfd558 "pypar.pypar", fp=0xfefc02b4, |
---|
| 135 | > > buf=0xffbfd068 "/usr/local/lib/python2.3/site-packages/pypar/pypar.py", |
---|
| 136 | > > type=1, loader=0x0) at Python/import.c:1699 |
---|
| 137 | > > #15 0xff2bf104 in import_submodule (mod=0xbb1b0, subname=0xffbfd55e "pypar", |
---|
| 138 | > > fullname=0xffbfd558 "pypar.pypar") at Python/import.c:2290 |
---|
| 139 | > > #16 0xff2bebd4 in load_next (mod=0xbb1b0, altmod=0xff318028, |
---|
| 140 | > > p_name=0xffbfd55e, buf=0xffbfd558 "pypar.pypar", p_buflen=0xffbfd554) |
---|
| 141 | > > at Python/import.c:2111 |
---|
| 142 | > > #17 0xff2c0890 in import_module_ex (name=0x0, globals=0xbb1b0, locals=0x48d20, |
---|
| 143 | > > fromlist=0x45590) at Python/import.c:1957 |
---|
| 144 | > > #18 0xff2be7e8 in PyImport_ImportModuleEx (name=0xbb1f4 "pypar", |
---|
| 145 | > > globals=0x48d20, locals=0x48d20, fromlist=0x45590) at Python/import.c:1998 |
---|
| 146 | > > #19 0xff28cda0 in builtin___import__ (self=0x0, args=0xbb1f4) |
---|
| 147 | > > at Python/bltinmodule.c:45 |
---|
| 148 | > > #20 0xff25b340 in PyCFunction_Call (func=0x29e68, arg=0xb3f60, kw=0x0) |
---|
| 149 | > > at Objects/methodobject.c:93 |
---|
| 150 | > > #21 0xff22e090 in PyObject_Call (func=0x29e68, arg=0xb3f60, kw=0x0) |
---|
| 151 | > > at Objects/abstract.c:1755 |
---|
| 152 | > > #22 0xff299f84 in PyEval_CallObjectWithKeywords (func=0x29e68, arg=0xb3f60, |
---|
| 153 | > > kw=0x0) at Python/ceval.c:3346 |
---|
| 154 | > > #23 0xff29740c in eval_frame (f=0xf71d0) at Python/ceval.c:1996 |
---|
| 155 | > > #24 0xff2987a8 in PyEval_EvalCodeEx (co=0xacee0, globals=0x1, locals=0x0, |
---|
| 156 | > > args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0, |
---|
| 157 | > > closure=0x0) at Python/ceval.c:2663 |
---|
| 158 | > > #25 0xff294c34 in PyEval_EvalCode (co=0xacee0, globals=0x0, locals=0x48d20) |
---|
| 159 | > > at Python/ceval.c:537 |
---|
| 160 | > > #26 0xff2bd2fc in PyImport_ExecCodeModuleEx (name=0xffbfed28 "pypar", |
---|
| 161 | > > co=0xacee0, |
---|
| 162 | > > pathname=0xffbfde48 "/usr/local/lib/python2.3/site-packages/pypar/__init__.pyc") at Python/import.c:621 |
---|
| 163 | > > #27 0xff2c016c in load_source_module (name=0xffbfed28 "pypar", |
---|
| 164 | > > pathname=0xffbfde48 "/usr/local/lib/python2.3/site-packages/pypar/__init__.pyc", fp=0x22e8) at Python/import.c:894 |
---|
| 165 | > > #28 0xff2bdf64 in load_module (name=0xffbfed28 "pypar", fp=0xfefc02a4, |
---|
| 166 | > > buf=0xffbfe340 "/usr/local/lib/python2.3/site-packages/pypar/__init__.py", |
---|
| 167 | > > type=1, loader=0x0) at Python/import.c:1699 |
---|
| 168 | > > #29 0xff2c0624 in load_package (name=0xffbfed28 "pypar", pathname=0xb2338 "") |
---|
| 169 | > > at Python/import.c:951 |
---|
| 170 | > > #30 0xff2be11c in load_module (name=0xffbfed28 "pypar", fp=0x0, |
---|
| 171 | > > buf=0xffbfe838 "/usr/local/lib/python2.3/site-packages/pypar", type=5, |
---|
| 172 | > > loader=0x0) at Python/import.c:1722 |
---|
| 173 | > > #31 0xff2bf104 in import_submodule (mod=0xff318028, |
---|
| 174 | > > subname=0xffbfed28 "pypar", fullname=0xffbfed28 "pypar") |
---|
| 175 | > > at Python/import.c:2290 |
---|
| 176 | > > #32 0xff2bebd4 in load_next (mod=0xff318028, altmod=0xff318028, |
---|
| 177 | > > p_name=0xffbfed28, buf=0xffbfed28 "pypar", p_buflen=0xffbfed24) |
---|
| 178 | > > at Python/import.c:2111 |
---|
| 179 | > > #33 0xff2c0890 in import_module_ex (name=0x0, globals=0xff318028, |
---|
| 180 | > > locals=0x3a9c0, fromlist=0xff318028) at Python/import.c:1957 |
---|
| 181 | > > #34 0xff2be7e8 in PyImport_ImportModuleEx (name=0xbb1f4 "pypar", |
---|
| 182 | > > globals=0x3a9c0, locals=0x3a9c0, fromlist=0xff318028) |
---|
| 183 | > > at Python/import.c:1998 |
---|
| 184 | > > #35 0xff28cda0 in builtin___import__ (self=0x0, args=0xbb1f4) |
---|
| 185 | > > at Python/bltinmodule.c:45 |
---|
| 186 | > > #36 0xff25b340 in PyCFunction_Call (func=0x29e68, arg=0x39900, kw=0x0) |
---|
| 187 | > > at Objects/methodobject.c:93 |
---|
| 188 | > > #37 0xff22e090 in PyObject_Call (func=0x29e68, arg=0x39900, kw=0x0) |
---|
| 189 | > > at Objects/abstract.c:1755 |
---|
| 190 | > > #38 0xff299f84 in PyEval_CallObjectWithKeywords (func=0x29e68, arg=0x39900, |
---|
| 191 | > > kw=0x0) at Python/ceval.c:3346 |
---|
| 192 | > > #39 0xff29740c in eval_frame (f=0xedb78) at Python/ceval.c:1996 |
---|
| 193 | > > #40 0xff2987a8 in PyEval_EvalCodeEx (co=0xb0ce0, globals=0x0, locals=0x0, |
---|
| 194 | > > args=0x0, argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0, |
---|
| 195 | > > closure=0x0) at Python/ceval.c:2663 |
---|
| 196 | > > #41 0xff294c34 in PyEval_EvalCode (co=0xb0ce0, globals=0x0, locals=0x3a9c0) |
---|
| 197 | > > at Python/ceval.c:537 |
---|
| 198 | > > #42 0xff2caf3c in run_node (n=0x2a140, filename=0x3a9c0 "", globals=0x3a9c0, |
---|
| 199 | > > locals=0x3a9c0, flags=0x3a9c0) at Python/pythonrun.c:1205 |
---|
| 200 | > > #43 0xff2ca000 in PyRun_InteractiveOneFlags (fp=0xffffffff, |
---|
| 201 | > > filename=0xff2fed08 "<stdin>", flags=0xffbff79c) at Python/pythonrun.c:697 |
---|
| 202 | > > #44 0xff2c9d60 in PyRun_InteractiveLoopFlags (fp=0xfefc0274, |
---|
| 203 | > > filename=0xff2fed08 "<stdin>", flags=0xffbff79c) at Python/pythonrun.c:630 |
---|
| 204 | > > #45 0xff2c9cac in PyRun_AnyFileExFlags (fp=0xfefc0274, |
---|
| 205 | > > filename=0xff2fed08 "<stdin>", closeit=0, flags=0xffbff79c) |
---|
| 206 | > > at Python/pythonrun.c:593 |
---|
| 207 | > > #46 0xff2d2a08 in Py_Main (argc=5120, argv=0x16bc) at Modules/main.c:415 |
---|
| 208 | > > |
---|
| 209 | > > |
---|
| 210 | > > Any suggestions or ideas for getting this working? |
---|
| 211 | > > |
---|
| 212 | > > Thanks, |
---|
| 213 | > > Dave |
---|
| 214 | > > |
---|
| 215 | > > -- |
---|
| 216 | > > David Reed, Ph.D. e-mail: dreed@capital.edu |
---|
| 217 | > > Math/Computer Science Dept. http://capital2.capital.edu/faculty/dreed |
---|
| 218 | > > Capital University phone: 614-236-6133 |
---|
| 219 | > > 2199 E. Main St. fax: 614-236-6518 |
---|
| 220 | > > Columbus, OH 43209 |
---|
| 221 | > |
---|
| 222 | > -- |
---|
| 223 | > ------------------------------------------------------------------- |
---|
| 224 | > Ole Nielsen | Email: Ole.Nielsen@anu.edu.au |
---|
| 225 | > ------------------------------------------------------------------- |
---|
| 226 | > Mathematical Sciences Institute | Phone: +61 2 6125 3873 (Direct) |
---|
| 227 | > Australian National University | Fax: +61 2 6125 5549 |
---|
| 228 | > Canberra ACT 0200 | |
---|
| 229 | > Australia | |
---|
| 230 | > ------------------------------------------------------------------- |
---|
| 231 | > ANU CRICOS # 00120C |
---|
| 232 | > ------------------------------------------------------------------- |
---|
| 233 | > |
---|