So it sounds like coalescing is not your issue and that the problem has something to do with the queue sizes.  It would be helpful if we could detect the hdr->tag == 0 issue on the sending side and get at least a stack trace.  There is something really odd going on here.

--td

Eloi Gaudry wrote:
Hi Terry,

I'm sorry to say that I might have missed a point here.

I've lately been relaunching all previously failing computations with the message coalescing feature being switched off, and I saw the same hdr->tag=0 error several times, always during a collective call (MPI_Comm_create, MPI_Allreduce and MPI_Broadcast, so far). And as soon as I switched to the peer queue option I was previously using (--mca btl_openib_receive_queues P,65536,256,192,128 instead of using --mca btl_openib_use_message_coalescing 0), all computations ran flawlessly.

As for the reproducer, I've already tried to write something but I haven't succeeded so far at reproducing the hdr->tag=0 issue with it.

Eloi

On 24/09/2010 18:37, Terry Dontje wrote:
Eloi Gaudry wrote:
Terry,

You were right, the error indeed seems to come from the message coalescing feature.
If I turn it off using the "--mca btl_openib_use_message_coalescing 0", I'm not able to observe the "hdr->tag=0" error.

There are some trac requests associated to very similar error (https://svn.open-mpi.org/trac/ompi/search?q=coalescing) but they are all closed (except https://svn.open-mpi.org/trac/ompi/ticket/2352 
that might be related), aren't they ? What would you suggest Terry ?

  
Interesting, though it looks to me like the segv in ticket 2352 would have happened on the send side instead of the receive side like you have.  As to what to do next it would be really nice to have some sort of reproducer that we can try and debug what is really going on.  The only other thing to do without a reproducer is to inspect the code on the send side to figure out what might make it generate at 0 hdr->tag.  Or maybe instrument the send side to stop when it is about ready to send a 0 hdr->tag and see if we can see how the code got there.

I might have some cycles to look at this Monday.

--td
Eloi


On Friday 24 September 2010 16:00:26 Terry Dontje wrote:
  
Eloi Gaudry wrote:
    
Terry,

No, I haven't tried any other values than P,65536,256,192,128 yet.

The reason why is quite simple. I've been reading and reading again this
thread to understand the btl_openib_receive_queues meaning and I can't
figure out why the default values seem to induce the hdr-

      
tag=0 issue
(http://www.open-mpi.org/community/lists/users/2009/01/7808.php).
        
Yeah, the size of the fragments and number of them really should not
cause this issue.  So I too am a little perplexed about it.

    
Do you think that the default shared received queue parameters are
erroneous for this specific Mellanox card ? Any help on finding the
proper parameters would actually be much appreciated.
      
I don't necessarily think it is the queue size for a specific card but
more so the handling of the queues by the BTL when using certain sizes.
At least that is one gut feel I have.

In my mind the tag being 0 is either something below OMPI is polluting
the data fragment or OMPI's internal protocol is some how getting messed
up.  I can imagine (no empirical data here) the queue sizes could change
how the OMPI protocol sets things up.  Another thing may be the
coalescing feature in the openib BTL which tries to gang multiple
messages into one packet when resources are running low.   I can see
where changing the queue sizes might affect the coalescing.  So, it
might be interesting to turn off the coalescing.  You can do that by
setting "--mca btl_openib_use_message_coalescing 0" in your mpirun line.

If that doesn't solve the issue then obviously there must be something
else going on :-).

Note, the reason I am interested in this is I am seeing a similar error
condition (hdr->tag == 0) on a development system.  Though my failing
case fails with np=8 using the connectivity test program which is mainly
point to point and there are not a significant amount of data transfers
going on either.

--td

    
Eloi

On Friday 24 September 2010 14:27:07 you wrote:
      
That is interesting.  So does the number of processes affect your runs
any.  The times I've seen hdr->tag be 0 usually has been due to protocol
issues.  The tag should never be 0.  Have you tried to do other
receive_queue settings other than the default and the one you mention.

I wonder if you did a combination of the two receive queues causes a
failure or not.  Something like

P,128,256,192,128:P,65536,256,192,128

I am wondering if it is the first queuing definition causing the issue
or possibly the SRQ defined in the default.

--td

Eloi Gaudry wrote:
        
Hi Terry,

The messages being send/received can be of any size, but the error
seems to happen more often with small messages (as an int being
broadcasted or allreduced). The failing communication differs from one
run to another, but some spots are more likely to be failing than
another. And as far as I know, there are always located next to a
small message (an int being broadcasted for instance) communication.
Other typical messages size are

          
10k but can be very much larger.
            
I've been checking the hca being used, its' from mellanox (with
vendor_part_id=26428). There is no receive_queues parameters associated
to it.

 $ cat share/openmpi/mca-btl-openib-device-params.ini as well:
[...]

  # A.k.a. ConnectX
  [Mellanox Hermon]
  vendor_id = 0x2c9,0x5ad,0x66a,0x8f1,0x1708,0x03ba,0x15b3
  vendor_part_id =
  25408,25418,25428,26418,26428,25448,26438,26448,26468,26478,26488
  use_eager_rdma = 1
  mtu = 2048
  max_inline_data = 128

[..]

$ ompi_info --param btl openib --parsable | grep receive_queues

 mca:btl:openib:param:btl_openib_receive_queues:value:P,128,256,192,128
 :S ,2048,256,128,32:S,12288,256,128,32:S,65536,256,128,32
 mca:btl:openib:param:btl_openib_receive_queues:data_source:default
 value mca:btl:openib:param:btl_openib_receive_queues:status:writable
 mca:btl:openib:param:btl_openib_receive_queues:help:Colon-delimited,
 comma delimited list of receive queues: P,4096,8,6,4:P,32768,8,6,4
 mca:btl:openib:param:btl_openib_receive_queues:deprecated:no

I was wondering if these parameters (automatically computed at openib
btl init for what I understood) were not incorrect in some way and I
plugged some others values: "P,65536,256,192,128" (someone on the list
used that values when encountering a different issue) . Since that, I
haven't been able to observe the segfault (occuring as hrd->tag = 0 in
btl_openib_component.c:2881) yet.

Eloi


/home/pp_fr/st03230/EG/Softs/openmpi-custom-1.4.2/bin/

On Thursday 23 September 2010 23:33:48 Terry Dontje wrote:
          
Eloi, I am curious about your problem.  Can you tell me what size of
job it is?  Does it always fail on the same bcast,  or same process?

Eloi Gaudry wrote:
            
Hi Nysal,

Thanks for your suggestions.

I'm now able to get the checksum computed and redirected to stdout,
thanks (I forgot the  "-mca pml_base_verbose 5" option, you were
right). I haven't been able to observe the segmentation fault (with
hdr->tag=0) so far (when using pml csum) but I 'll let you know when
I am.

I've got two others question, which may be related to the error
observed:

1/ does the maximum number of MPI_Comm that can be handled by OpenMPI
somehow depends on the btl being used (i.e. if I'm using openib, may
I use the same number of MPI_Comm object as with tcp) ? Is there
something as MPI_COMM_MAX in OpenMPI ?

2/ the segfaults only appears during a mpi collective call, with very
small message (one int is being broadcast, for instance) ; i followed
the guidelines given at http://icl.cs.utk.edu/open-
mpi/faq/?category=openfabrics#ib-small-message-rdma but the
debug-build of OpenMPI asserts if I use a different min-size that
255. Anyway, if I deactivate eager_rdma, the segfaults remains. Does
the openib btl handle very small message differently (even with
eager_rdma
deactivated) than tcp ?
              
Others on the list does coalescing happen with non-eager_rdma?  If so
then that would possibly be one difference between the openib btl and
tcp aside from the actual protocol used.

            
 is there a way to make sure that large messages and small messages
 are handled the same way ?
              
Do you mean so they all look like eager messages?  How large of
messages are we talking about here 1K, 1M or 10M?

--td

            
Regards,
Eloi

On Friday 17 September 2010 17:57:17 Nysal Jan wrote:
              
Hi Eloi,
Create a debug build of OpenMPI (--enable-debug) and while running
with the csum PML add "-mca pml_base_verbose 5" to the command line.
This will print the checksum details for each fragment sent over the
wire. I'm guessing it didnt catch anything because the BTL failed.
The checksum verification is done in the PML, which the BTL calls
via a callback function. In your case the PML callback is never
called because the hdr->tag is invalid. So enabling checksum
tracing also might not be of much use. Is it the first Bcast that
fails or the nth Bcast and what is the message size? I'm not sure
what could be the problem at this moment. I'm afraid you will have
to debug the BTL to find out more.

--Nysal

On Fri, Sep 17, 2010 at 4:39 PM, Eloi Gaudry <eg@fft.be> wrote:
                
Hi Nysal,

thanks for your response.

I've been unable so far to write a test case that could illustrate
the hdr->tag=0 error.
Actually, I'm only observing this issue when running an internode
computation involving infiniband hardware from Mellanox (MT25418,
ConnectX IB DDR, PCIe 2.0
2.5GT/s, rev a0) with our time-domain software.

I checked, double-checked, and rechecked again every MPI use
performed during a parallel computation and I couldn't find any
error so far. The fact that the very
same parallel computation run flawlessly when using tcp (and
disabling openib support) might seem to indicate that the issue is
somewhere located inside the
openib btl or at the hardware/driver level.

I've just used the "-mca pml csum" option and I haven't seen any
related messages (when hdr->tag=0 and the segfaults occurs).
Any suggestion ?

Regards,
Eloi

On Friday 17 September 2010 16:03:34 Nysal Jan wrote:
                  
Hi Eloi,
Sorry for the delay in response. I haven't read the entire email
thread, but do you have a test case which can reproduce this
error? Without that it will be difficult to nail down the cause.
Just to clarify, I do not work for an iwarp vendor. I can
certainly try to reproduce it on an IB system. There is also a
PML called csum, you can use it via "-mca pml csum", which will
checksum the MPI messages and verify it at the receiver side for
any data
corruption. You can try using it to see if it is able
                    
to

                  
catch anything.

Regards
--Nysal

On Thu, Sep 16, 2010 at 3:48 PM, Eloi Gaudry <eg@fft.be> wrote:
                    
Hi Nysal,

I'm sorry to intrrupt, but I was wondering if you had a chance to
look
                      
at

                  
this error.

Regards,
Eloi



--


Eloi Gaudry

Free Field Technologies
Company Website: http://www.fft.be
Company Phone:   +32 10 487 959


---------- Forwarded message ----------
From: Eloi Gaudry <eg@fft.be>
To: Open MPI Users <users@open-mpi.org>
Date: Wed, 15 Sep 2010 16:27:43 +0200
Subject: Re: [OMPI users] [openib] segfault when using openib btl
Hi,

I was wondering if anybody got a chance to have a look at this
issue.

Regards,
Eloi

On Wednesday 18 August 2010 09:16:26 Eloi Gaudry wrote:
                      
Hi Jeff,

Please find enclosed the output (valgrind.out.gz) from
/opt/openmpi-debug-1.4.2/bin/orterun -np 2 --host pbn11,pbn10
--mca
                        
btl

                  
openib,self --display-map --verbose --mca mpi_warn_on_fork 0
--mca btl_openib_want_fork_support 0 -tag-output
/opt/valgrind-3.5.0/bin/valgrind --tool=memcheck
--suppressions=/opt/openmpi-debug-1.4.2/share/openmpi/openmpi-
valgrind.supp --suppressions=./suppressions.python.supp
/opt/actran/bin/actranpy_mp ...

Thanks,
Eloi

On Tuesday 17 August 2010 09:32:53 Eloi Gaudry wrote:
                        
On Monday 16 August 2010 19:14:47 Jeff Squyres wrote:
                          
On Aug 16, 2010, at 10:05 AM, Eloi Gaudry wrote:
                            
I did run our application through valgrind but it couldn't
find any "Invalid write": there is a bunch of "Invalid read"
(I'm using
                              
1.4.2

                      
with the suppression file), "Use of uninitialized bytes" and
"Conditional jump depending on uninitialized bytes" in
                              
different

                  
ompi

                      
routines. Some of them are located in btl_openib_component.c.
I'll send you an output of valgrind shortly.
                              
A lot of them in btl_openib_* are to be expected --
OpenFabrics uses OS-bypass methods for some of its memory,
and therefore valgrind is unaware of them (and therefore
incorrectly marks them as
uninitialized).
                            
would it  help if i use the upcoming 1.5 version of openmpi ? i
                          
read

                  
that

                      
a huge effort has been done to clean-up the valgrind output ?
but maybe that this doesn't concern this btl (for the reasons
you mentionned).

                          
Another question, you said that the callback function pointer
                              
should

                      
never be 0. But can the tag be null (hdr->tag) ?
                              
The tag is not a pointer -- it's just an integer.
                            
I was worrying that its value could not be null.

I'll send a valgrind output soon (i need to build libpython
without pymalloc first).

Thanks,
Eloi

                          
Thanks for your help,
Eloi

On 16/08/2010 18:22, Jeff Squyres wrote:
                              
Sorry for the delay in replying.

Odd; the values of the callback function pointer should
never
                                
be

                  
0.

                      
This seems to suggest some kind of memory corruption is
occurring.

I don't know if it's possible, because the stack trace looks
like you're calling through python, but can you run this
application through valgrind, or some other memory-checking
debugger?

On Aug 10, 2010, at 7:15 AM, Eloi Gaudry wrote:
                                
Hi,

sorry, i just forgot to add the values of the function
                                  
parameters:
                      
(gdb) print reg->cbdata
$1 = (void *) 0x0
(gdb) print openib_btl->super
$2 = {btl_component = 0x2b341edd7380, btl_eager_limit =
                                  
12288,

                  
btl_rndv_eager_limit = 12288, btl_max_send_size = 65536,
btl_rdma_pipeline_send_length = 1048576,

  btl_rdma_pipeline_frag_size = 1048576,
                                  
btl_min_rdma_pipeline_size

                      
  = 1060864, btl_exclusivity = 1024, btl_latency = 10,
  btl_bandwidth = 800, btl_flags = 310, btl_add_procs =
  0x2b341eb8ee47<mca_btl_openib_add_procs>, btl_del_procs =
  0x2b341eb90156<mca_btl_openib_del_procs>, btl_register =
  0, btl_finalize =
  0x2b341eb93186<mca_btl_openib_finalize>,
                                  
btl_alloc

                      
  = 0x2b341eb90a3e<mca_btl_openib_alloc>, btl_free =
  0x2b341eb91400<mca_btl_openib_free>, btl_prepare_src =
  0x2b341eb91813<mca_btl_openib_prepare_src>,
  btl_prepare_dst
                                  
=

                  
  0x2b341eb91f2e<mca_btl_openib_prepare_dst>, btl_send =
  0x2b341eb94517<mca_btl_openib_send>, btl_sendi =
  0x2b341eb9340d<mca_btl_openib_sendi>, btl_put =
  0x2b341eb94660<mca_btl_openib_put>, btl_get =
  0x2b341eb94c4e<mca_btl_openib_get>, btl_dump =
  0x2b341acd45cb<mca_btl_base_dump>, btl_mpool = 0xf3f4110,
  btl_register_error =
  0x2b341eb90565<mca_btl_openib_register_error_cb>,
  btl_ft_event
                                  
=

                      
  0x2b341eb952e7<mca_btl_openib_ft_event>}

(gdb) print hdr->tag
$3 = 0 '\0'
(gdb) print des
$4 = (mca_btl_base_descriptor_t *) 0xf4a6700
(gdb) print reg->cbfunc
$5 = (mca_btl_base_module_recv_cb_fn_t) 0

Eloi

On Tuesday 10 August 2010 16:04:08 Eloi Gaudry wrote:
                                  
Hi,

Here is the output of a core file generated during a
                                    
segmentation

                      
fault observed during a collective call (using openib):

#0  0x0000000000000000 in ?? ()
(gdb) where
#0  0x0000000000000000 in ?? ()
#1  0x00002aedbc4e05f4 in btl_openib_handle_incoming
(openib_btl=0x1902f9b0, ep=0x1908a1c0, frag=0x190d9700,
byte_len=18) at btl_openib_component.c:2881 #2
0x00002aedbc4e25e2 in handle_wc (device=0x19024ac0, cq=0,
wc=0x7ffff279ce90) at
btl_openib_component.c:3178 #3  0x00002aedbc4e2e9d in
                                    
poll_device

                      
(device=0x19024ac0, count=2) at
btl_openib_component.c:3318
                                    
#4

                  
0x00002aedbc4e34b8 in progress_one_device
                                    
(device=0x19024ac0)

                  
at btl_openib_component.c:3426 #5  0x00002aedbc4e3561 in
btl_openib_component_progress () at
btl_openib_component.c:3451
                                    
#6

                      
0x00002aedb8b22ab8 in opal_progress () at
runtime/opal_progress.c:207 #7 0x00002aedb859f497 in
opal_condition_wait (c=0x2aedb888ccc0, m=0x2aedb888cd20)
at ../opal/threads/condition.h:99 #8
0x00002aedb859fa31 in ompi_request_default_wait_all
                                    
(count=2,

                  
requests=0x7ffff279d0e0, statuses=0x0) at
request/req_wait.c:262 #9 0x00002aedbd7559ad in
ompi_coll_tuned_allreduce_intra_recursivedoubling
(sbuf=0x7ffff279d444, rbuf=0x7ffff279d440, count=1,
dtype=0x6788220, op=0x6787a20,
comm=0x19d81ff0, module=0x19d82b20) at
                                    
coll_tuned_allreduce.c:223

                      
#10 0x00002aedbd7514f7 in
ompi_coll_tuned_allreduce_intra_dec_fixed
(sbuf=0x7ffff279d444, rbuf=0x7ffff279d440, count=1,
dtype=0x6788220, op=0x6787a20, comm=0x19d81ff0,
module=0x19d82b20) at
coll_tuned_decision_fixed.c:63
#11 0x00002aedb85c7792 in PMPI_Allreduce
                                    
(sendbuf=0x7ffff279d444,

                      
recvbuf=0x7ffff279d440, count=1, datatype=0x6788220,
                                    
op=0x6787a20,

                      
comm=0x19d81ff0) at pallreduce.c:102 #12
0x0000000004387dbf
                                    
in

                  
FEMTown::MPI::Allreduce (sendbuf=0x7ffff279d444,
recvbuf=0x7ffff279d440, count=1, datatype=0x6788220,
                                    
op=0x6787a20,

                      
comm=0x19d81ff0) at stubs.cpp:626 #13 0x0000000004058be8
in FEMTown::Domain::align (itf=
                                    
{<FEMTown::Boost::shared_base_ptr<FEMTown::Domain::Int

                      
            er fa ce>>

= {_vptr.shared_base_ptr = 0x7ffff279d620, ptr_ = {px =
0x199942a4, pn = {pi_ = 0x6}}},<No data fields>}) at
interface.cpp:371 #14 0x00000000040cb858 in
FEMTown::Field::detail::align_itfs_and_neighbhors (dim=2,
                                    
set={px

                      
= 0x7ffff279d780, pn = {pi_ = 0x2f279d640}},
check_info=@0x7ffff279d7f0) at check.cpp:63 #15
                                    
0x00000000040cbfa8

                      
in FEMTown::Field::align_elements (set={px =
0x7ffff279d950, pn
                                    
=

                      
{pi_ = 0x66e08d0}}, check_info=@0x7ffff279d7f0) at
check.cpp:159 #16 0x00000000039acdd4 in
PyField_align_elements (self=0x0, args=0x2aaab0765050,
kwds=0x19d2e950) at check.cpp:31 #17 0x0000000001fbf76d in
FEMTown::Main::ExErrCatch<_object* (*)(_object*, _object*,
_object*)>::exec<_object>
(this=0x7ffff279dc20, s=0x0, po1=0x2aaab0765050,
po2=0x19d2e950) at
/home/qa/svntop/femtown/modules/main/py/exception.hpp:463
                                    
#18

                  
0x00000000039acc82 in PyField_align_elements_ewrap
                                    
(self=0x0,

                  
args=0x2aaab0765050, kwds=0x19d2e950) at check.cpp:39 #19
0x00000000044093a0 in PyEval_EvalFrameEx (f=0x19b52e90,
throwflag=<value optimized out>) at Python/ceval.c:3921
#20 0x000000000440aae9 in PyEval_EvalCodeEx
(co=0x2aaab754ad50, globals=<value optimized out>,
locals=<value optimized out>, args=0x3, argcount=1,
kws=0x19ace4a0, kwcount=2,
defs=0x2aaab75e4800, defcount=2, closure=0x0) at
Python/ceval.c:2968
#21 0x0000000004408f58 in PyEval_EvalFrameEx
(f=0x19ace2d0, throwflag=<value optimized out>) at
Python/ceval.c:3802 #22 0x000000000440aae9 in
PyEval_EvalCodeEx (co=0x2aaab7550120, globals=<value
optimized out>, locals=<value optimized out>, args=0x7,
argcount=1, kws=0x19acc418, kwcount=3,
defs=0x2aaab759e958, defcount=6, closure=0x0) at
Python/ceval.c:2968
#23 0x0000000004408f58 in PyEval_EvalFrameEx
(f=0x19acc1c0, throwflag=<value optimized out>) at
Python/ceval.c:3802 #24 0x000000000440aae9 in
PyEval_EvalCodeEx (co=0x2aaab8b5e738, globals=<value
optimized out>, locals=<value optimized out>, args=0x6,
argcount=1, kws=0x19abd328, kwcount=5,
defs=0x2aaab891b7e8, defcount=3, closure=0x0) at
Python/ceval.c:2968
#25 0x0000000004408f58 in PyEval_EvalFrameEx
(f=0x19abcea0, throwflag=<value optimized out>) at
Python/ceval.c:3802 #26 0x000000000440aae9 in
PyEval_EvalCodeEx (co=0x2aaab3eb4198, globals=<value
optimized out>, locals=<value optimized out>, args=0xb,
argcount=1, kws=0x19a89df0, kwcount=10, defs=0x0,
defcount=0, closure=0x0) at Python/ceval.c:2968
#27 0x0000000004408f58 in PyEval_EvalFrameEx
(f=0x19a89c40, throwflag=<value optimized out>) at
Python/ceval.c:3802 #28 0x000000000440aae9 in
PyEval_EvalCodeEx (co=0x2aaab3eb4288, globals=<value
optimized out>, locals=<value optimized out>, args=0x1,
argcount=0, kws=0x19a89330, kwcount=0,
defs=0x2aaab8b66668, defcount=1, closure=0x0) at
Python/ceval.c:2968
#29 0x0000000004408f58 in PyEval_EvalFrameEx
(f=0x19a891b0, throwflag=<value optimized out>) at
Python/ceval.c:3802 #30 0x000000000440aae9 in
PyEval_EvalCodeEx (co=0x2aaab8b6a738, globals=<value
optimized out>, locals=<value optimized out>, args=0x0,
argcount=0, kws=0x0, kwcount=0, defs=0x0, defcount=0,
closure=0x0) at
Python/ceval.c:2968
#31 0x000000000440ac02 in PyEval_EvalCode (co=0x1902f9b0,
globals=0x0, locals=0x190d9700) at Python/ceval.c:522 #32
0x000000000442853c in PyRun_StringFlags (str=0x192fd3d8
"DIRECT.Actran.main()", start=<value optimized out>,
globals=0x192213d0, locals=0x192213d0, flags=0x0) at
Python/pythonrun.c:1335 #33 0x0000000004429690 in
PyRun_SimpleStringFlags (command=0x192fd3d8
"DIRECT.Actran.main()", flags=0x0) at
Python/pythonrun.c:957 #34 0x0000000001fa1cf9 in
FEMTown::Python::FEMPy::run_application
                                    
(this=0x7ffff279f650)

                  
at fempy.cpp:873 #35 0x000000000434ce99 in
                                    
FEMTown::Main::Batch::run

                      
(this=0x7ffff279f650) at batch.cpp:374 #36
                                    
0x0000000001f9aa25

                  
in main (argc=8, argv=0x7ffff279fa48) at main.cpp:10 (gdb)
f 1 #1  0x00002aedbc4e05f4 in btl_openib_handle_incoming
(openib_btl=0x1902f9b0, ep=0x1908a1c0, frag=0x190d9700,
byte_len=18) at btl_openib_component.c:2881 2881
reg->cbfunc( &openib_btl->super, hdr->tag, des,
reg->cbdata
                                    
);

                  
Current language: auto; currently c
(gdb)
#1  0x00002aedbc4e05f4 in btl_openib_handle_incoming
(openib_btl=0x1902f9b0, ep=0x1908a1c0, frag=0x190d9700,
byte_len=18) at btl_openib_component.c:2881 2881
reg->cbfunc( &openib_btl->super, hdr->tag, des,
reg->cbdata
                                    
);

                  
(gdb) l 2876
2877        if(OPAL_LIKELY(!(is_credit_msg =
is_credit_message(frag)))) { 2878            /* call
registered callback */
2879            mca_btl_active_message_callback_t* reg;
2880            reg = mca_btl_base_active_message_trigger
+ hdr->tag; 2881
reg->cbfunc(&openib_btl->super, hdr->tag, des,
reg->cbdata ); 2882
if(MCA_BTL_OPENIB_RDMA_FRAG(frag)) { 2883
cqp
                                    
=

                  
(hdr->credits>>  11)&  0x0f;
2884                hdr->credits&= 0x87ff;
2885            } else {

Regards,
Eloi

On Friday 16 July 2010 16:01:02 Eloi Gaudry wrote:
                                    
Hi Edgar,

The only difference I could observed was that the
segmentation fault appeared sometimes later during the
parallel computation.

I'm running out of idea here. I wish I could use the
"--mca
                                      
coll

                      
tuned" with "--mca self,sm,tcp" so that I could check
that the issue is not somehow limited to the tuned
collective routines.

Thanks,
Eloi

On Thursday 15 July 2010 17:24:24 Edgar Gabriel wrote:
                                      
On 7/15/2010 10:18 AM, Eloi Gaudry wrote:
                                        
hi edgar,

thanks for the tips, I'm gonna try this option as well.
                                          
the

                  
segmentation fault i'm observing always happened during
a collective communication indeed... does it basically
                                          
switch

                  
all

                      
collective communication to basic mode, right ?

sorry for my ignorance, but what's a NCA ?
                                          
sorry, I meant to type HCA (InifinBand networking card)

Thanks
Edgar

                                        
thanks,
éloi

On Thursday 15 July 2010 16:20:54 Edgar Gabriel wrote:
                                          
you could try first to use the algorithms in the basic
                                            
module,

                      
e.g.

mpirun -np x --mca coll basic ./mytest

and see whether this makes a difference. I used to
                                            
observe

                  
sometimes a (similar ?) problem in the openib btl
triggered from the tuned collective component, in
cases where the ofed libraries were installed but no
NCA was found on a node. It used to work however with
the basic component.

Thanks
Edgar

On 7/15/2010 3:08 AM, Eloi Gaudry wrote:
                                            
hi Rolf,

unfortunately, i couldn't get rid of that annoying
segmentation fault when selecting another bcast
algorithm. i'm now going to replace MPI_Bcast with a
naive
implementation (using MPI_Send and MPI_Recv) and see
if
                                              
that

                      
helps.

regards,
éloi

On Wednesday 14 July 2010 10:59:53 Eloi Gaudry wrote:
                                              
Hi Rolf,

thanks for your input. You're right, I miss the
coll_tuned_use_dynamic_rules option.

I'll check if I the segmentation fault disappears
when
                                                
using

                      
the basic bcast linear algorithm using the proper
command line you provided.

Regards,
Eloi

On Tuesday 13 July 2010 20:39:59 Rolf vandeVaart
                                                
wrote:
                  
Hi Eloi:
To select the different bcast algorithms, you need
to add an extra mca parameter that tells the
library to use dynamic selection. --mca
coll_tuned_use_dynamic_rules 1

One way to make sure you are typing this in
correctly is
                                                  
to

                      
use it with ompi_info.  Do the following:
ompi_info -mca coll_tuned_use_dynamic_rules 1
--param
                                                  
coll

                      
You should see lots of output with all the
different algorithms that can be selected for the
various collectives. Therefore, you need this:

--mca coll_tuned_use_dynamic_rules 1 --mca
coll_tuned_bcast_algorithm 1

Rolf

On 07/13/10 11:28, Eloi Gaudry wrote:
                                                  
Hi,

I've found that "--mca coll_tuned_bcast_algorithm
1" allowed to switch to the basic linear
algorithm. Anyway whatever the algorithm used,
the segmentation fault remains.

Does anyone could give some advice on ways to
                                                    
diagnose

                  
the

                      
issue I'm facing ?

Regards,
Eloi

On Monday 12 July 2010 10:53:58 Eloi Gaudry wrote:
                                                    
Hi,

I'm focusing on the MPI_Bcast routine that seems
to randomly segfault when using the openib btl.
I'd
                                                      
like

                  
to

                      
know if there is any way to make OpenMPI switch
to
                                                      
a

                  
different algorithm than the default one being
selected for MPI_Bcast.

Thanks for your help,
Eloi

On Friday 02 July 2010 11:06:52 Eloi Gaudry wrote:
                                                      
Hi,

I'm observing a random segmentation fault during
                                                        
an

                  
internode parallel computation involving the
                                                        
openib

                  
btl

                      
and OpenMPI-1.4.2 (the same issue can be
observed with OpenMPI-1.3.3).

   mpirun (Open MPI) 1.4.2
   Report bugs to
   http://www.open-mpi.org/community/help/
   [pbn08:02624] *** Process received signal ***
   [pbn08:02624] Signal: Segmentation fault (11)
   [pbn08:02624] Signal code: Address not mapped
                                                        
(1)

                  
   [pbn08:02624] Failing at address: (nil)
   [pbn08:02624] [ 0] /lib64/libpthread.so.0
   [0x349540e4c0] [pbn08:02624] *** End of error
                                                        
message

                      
   ***
   sh: line 1:  2624 Segmentation fault
                                                        
\/share\/hpc3\/actran_suite\/Actran_11\.0\.rc2\.41872\/R

                      
ed Ha tE L\ -5 \/ x 86 _6 4\ /bin\/actranpy_mp
                                                        
'--apl=/share/hpc3/actran_suite/Actran_11.0.rc2.41872/Re

                      
dH at EL -5 /x 86 _ 64 /A c tran_11.0.rc2.41872'
                                                        
'--inputfile=/work/st25652/LSF_130073_0_47696_0/Case1_3D

                      
re al _m 4_ n2 .d a t'
                                                        
'--scratch=/scratch/st25652/LSF_130073_0_47696_0/scratch

                      
' '--mem=3200' '--threads=1'
'--errorlevel=FATAL' '--t_max=0.1'
'--parallel=domain'

If I choose not to use the openib btl (by using
--mca btl self,sm,tcp on the command line, for
instance), I don't encounter any problem and the
parallel computation runs flawlessly.

I would like to get some help to be able:
- to diagnose the issue I'm facing with the
openib btl - understand why this issue is
observed only when
                                                        
using

                      
the openib btl and not when using self,sm,tcp

Any help would be very much appreciated.

The outputs of ompi_info and the configure
scripts of OpenMPI are enclosed to this email,
and some
                                                        
information

                      
on the infiniband drivers as well.

Here is the command line used when launching a
                                                        
parallel

                      
computation

using infiniband:
   path_to_openmpi/bin/mpirun -np $NPROCESS
   --hostfile host.list --mca

btl openib,sm,self,tcp  --display-map --verbose
--version --mca mpi_warn_on_fork 0 --mca
btl_openib_want_fork_support 0 [...]

and the command line used if not using infiniband:
   path_to_openmpi/bin/mpirun -np $NPROCESS
   --hostfile host.list --mca

btl self,sm,tcp  --display-map --verbose
--version
                                                        
--mca

                      
mpi_warn_on_fork 0 --mca
btl_openib_want_fork_support
                                                        
0

                      
[...]

Thanks,
Eloi
                                                        
_______________________________________________
                                                    
_______________________________________________
users mailing list
users@open-mpi.org
http://www.open-mpi.org/mailman/listinfo.cgi/users
              

  



--
Oracle
Terry D. Dontje | Principal Software Engineer
Developer Tools Engineering | +1.781.442.2631
Oracle - Performance Technologies
95 Network Drive, Burlington, MA 01803
Email terry.dontje@oracle.com