Re: [O-MPI devel] [OMPI svn] svn:open-mpi r8692
Hi all,
whatever this fixed/changed, I no longer get corrupted memory in the
tuned data segment hung off each communicator... ! I'm still testing to
see if I get TimPs error.
G
On Sat, 14 Jan 2006 bosi...@osl.iu.edu wrote:
Author: bosilca
Date: 2006-01-14 15:21:44 -0500 (Sat, 14 Jan 2006)
New Revision: 8692
Modified:
trunk/ompi/mca/btl/tcp/btl_tcp_endpoint.c
trunk/ompi/mca/btl/tcp/btl_tcp_endpoint.h
trunk/ompi/mca/btl/tcp/btl_tcp_frag.c
trunk/ompi/mca/btl/tcp/btl_tcp_frag.h
Log:
A better implementation for the TCP endpoint cache + few comments.
Modified: trunk/ompi/mca/btl/tcp/btl_tcp_endpoint.c
===
--- trunk/ompi/mca/btl/tcp/btl_tcp_endpoint.c 2006-01-14 20:19:01 UTC (rev
8691)
+++ trunk/ompi/mca/btl/tcp/btl_tcp_endpoint.c 2006-01-14 20:21:44 UTC (rev
8692)
@@ -79,7 +79,7 @@
endpoint->endpoint_nbo = false;
#if MCA_BTL_TCP_ENDPOINT_CACHE
endpoint->endpoint_cache= NULL;
-endpoint->endpoint_cache_pos= 0;
+endpoint->endpoint_cache_pos= NULL;
endpoint->endpoint_cache_length = 0;
#endif /* MCA_BTL_TCP_ENDPOINT_CACHE */
OBJ_CONSTRUCT(&endpoint->endpoint_frags, opal_list_t);
@@ -187,21 +187,20 @@
static inline void mca_btl_tcp_endpoint_event_init(mca_btl_base_endpoint_t*
btl_endpoint, int sd)
{
#if MCA_BTL_TCP_ENDPOINT_CACHE
-btl_endpoint->endpoint_cache =
(char*)malloc(mca_btl_tcp_component.tcp_endpoint_cache);
+btl_endpoint->endpoint_cache =
(char*)malloc(mca_btl_tcp_component.tcp_endpoint_cache);
+btl_endpoint->endpoint_cache_pos = btl_endpoint->endpoint_cache;
#endif /* MCA_BTL_TCP_ENDPOINT_CACHE */
-opal_event_set(
-&btl_endpoint->endpoint_recv_event,
-btl_endpoint->endpoint_sd,
-OPAL_EV_READ|OPAL_EV_PERSIST,
-mca_btl_tcp_endpoint_recv_handler,
-btl_endpoint);
-opal_event_set(
-&btl_endpoint->endpoint_send_event,
-btl_endpoint->endpoint_sd,
-OPAL_EV_WRITE|OPAL_EV_PERSIST,
-mca_btl_tcp_endpoint_send_handler,
-btl_endpoint);
+opal_event_set( &btl_endpoint->endpoint_recv_event,
+ btl_endpoint->endpoint_sd,
+ OPAL_EV_READ|OPAL_EV_PERSIST,
+ mca_btl_tcp_endpoint_recv_handler,
+ btl_endpoint );
+opal_event_set( &btl_endpoint->endpoint_send_event,
+ btl_endpoint->endpoint_sd,
+ OPAL_EV_WRITE|OPAL_EV_PERSIST,
+ mca_btl_tcp_endpoint_send_handler,
+ btl_endpoint);
}
@@ -357,7 +356,9 @@
btl_endpoint->endpoint_sd = -1;
#if MCA_BTL_TCP_ENDPOINT_CACHE
free( btl_endpoint->endpoint_cache );
-btl_endpoint->endpoint_cache = NULL;
+btl_endpoint->endpoint_cache= NULL;
+btl_endpoint->endpoint_cache_pos= NULL;
+btl_endpoint->endpoint_cache_length = 0;
#endif /* MCA_BTL_TCP_ENDPOINT_CACHE */
}
btl_endpoint->endpoint_state = MCA_BTL_TCP_CLOSED;
@@ -619,13 +620,12 @@
}
#if MCA_BTL_TCP_ENDPOINT_CACHE
-btl_endpoint->endpoint_cache_pos = 0;
+assert( 0 == btl_endpoint->endpoint_cache_length );
data_still_pending_on_endpoint:
#endif /* MCA_BTL_TCP_ENDPOINT_CACHE */
/* check for completion of non-blocking recv on the current
fragment */
if(mca_btl_tcp_frag_recv(frag, sd) == false) {
btl_endpoint->endpoint_recv_frag = frag;
-OPAL_THREAD_UNLOCK(&btl_endpoint->endpoint_recv_lock);
} else {
btl_endpoint->endpoint_recv_frag = NULL;
switch(frag->hdr.type) {
@@ -636,39 +636,37 @@
break;
}
default:
-{
-break;
-}
+break;
}
#if MCA_BTL_TCP_ENDPOINT_CACHE
if( 0 != btl_endpoint->endpoint_cache_length ) {
+ /* If the cache still contain some data we can reuse the
same fragment
+* until we flush it completly.
+*/
MCA_BTL_TCP_FRAG_INIT_DST(frag, btl_endpoint);
goto data_still_pending_on_endpoint;
}
#endif /* MCA_BTL_TCP_ENDPOINT_CACHE */
MCA_BTL_TCP_FRAG_RETURN_MAX(frag);
-OPAL_THREAD_UNLOCK(&btl_endpoint->endpoint_recv_lock);
}
-break;
-}
-case MCA_BTL_TCP_SHUTDOWN:
-{
OPAL_THREAD_UNLOCK(&btl_endpoint->endpoint_recv_lock);
+assert( 0 == btl_endpoint->endpoint_cache_length );
break;
}
+case MCA_BTL_TCP_SHUTDOWN:
+OPAL_THREAD_UNLOCK(&btl_endpoint->endpoint_recv_lock);
+break;
default:
-{
-OPAL_THREAD_UNLOCK(&btl_endpoint->endpoint_recv_lock);
-BTL_ERROR(("invalid socket state(%d)",
btl_endpoint->e
Re: [O-MPI devel] Intel tests
Graham, with the trunk r8695 I no longer get these errors. Tim Quoting Graham E Fagg : > Hi Tim > I can get an error but not quite the same as yours. In my case I > get a > segfault as someone corrupts the memory attached to a communicator > (data > segment). Looks like a possible inplace error. Expect a fix shortly. > > G. > > On Tue, 10 Jan 2006, Tim Prins wrote: > > > Graham, > > > > It works properly if I select the basic coll component. Anyways, > here is > > the output you requested. The full output is about 140MB, so I > killed it > > before it finished... > > > > Tim > > > > Quoting Graham E Fagg : > > > >> Hi Tim > >> nope, can you rerun with mpirun -np 4 -mca coll_base_verbose 1 > >> > >> and email me the output? > >> Thanks > >> G > >> On Tue, 10 Jan 2006, Tim Prins wrote: > >> > >>> Hi everyone, > >>> > >>> I have been playing around with Open-MPI, using it as a test bed > >> for > >>> another project I am working on, and have found that on the > intel > >> test > >>> suite, ompi is failing the MPI_Allreduce_user_c, > >>> MPI_Reduce_scatter_user_c, and MPI_Reduce_user_c tests (it > prints > >>> something like MPITEST error (2): i=0, int value=4, expected 1, > >> etc). > >>> Are these known error? > >>> > >>> BTW, this is on a x86_64 linux box running 4 processes locally, > >> running > >>> the trunk svn version 8667, with no additional mca parameters > set. > >>> > >>> Thanks, > >>> > >>> TIm > >>> ___ > >>> devel mailing list > >>> de...@open-mpi.org > >>> http://www.open-mpi.org/mailman/listinfo.cgi/devel > >>> > >> > >> > >> Thanks, > >>Graham. > >> > -- > >> Dr Graham E. Fagg | Distributed, Parallel and > Meta-Computing > >> Innovative Computing Lab. PVM3.4, HARNESS, FT-MPI, SNIPE & Open > MPI > >> Computer Science Dept | Suite 203, 1122 Volunteer Blvd, > >> University of Tennessee | Knoxville, Tennessee, USA. TN > 37996-3450 > >> Email: f...@cs.utk.edu | Phone:+1(865)974-5790 | > >> Fax:+1(865)974-8296 > >> Broken complex systems are always derived from working simple > >> systems > >> > -- > >> ___ > >> devel mailing list > >> de...@open-mpi.org > >> http://www.open-mpi.org/mailman/listinfo.cgi/devel > >> > > > > > > > > > Thanks, > Graham. > -- > Dr Graham E. Fagg | Distributed, Parallel and Meta-Computing > Innovative Computing Lab. PVM3.4, HARNESS, FT-MPI, SNIPE & Open MPI > Computer Science Dept | Suite 203, 1122 Volunteer Blvd, > University of Tennessee | Knoxville, Tennessee, USA. TN 37996-3450 > Email: f...@cs.utk.edu | Phone:+1(865)974-5790 | > Fax:+1(865)974-8296 > Broken complex systems are always derived from working simple > systems > -- >
Re: [O-MPI devel] Intel tests
Hi Graham, On Jan 14, 2006, at 2:07 PM, Graham E Fagg wrote: Hi all, whatever this fixed/changed, I no longer get corrupted memory in the tuned data segment hung off each communicator... ! I'm still testing to see if I get TimPs error. G On Sat, 14 Jan 2006 bosi...@osl.iu.edu wrote: Author: bosilca Date: 2006-01-14 15:21:44 -0500 (Sat, 14 Jan 2006) New Revision: 8692 Modified: trunk/ompi/mca/btl/tcp/btl_tcp_endpoint.c trunk/ompi/mca/btl/tcp/btl_tcp_endpoint.h trunk/ompi/mca/btl/tcp/btl_tcp_frag.c trunk/ompi/mca/btl/tcp/btl_tcp_frag.h Log: A better implementation for the TCP endpoint cache + few comments. On a 64-bit bproc/myrinet system I'm seeing Tim P's problem with the current head of the trunk. See attached output. David $ ompi_info | head Open MPI: 1.1a1svn01142006 Open MPI SVN revision: svn01142006 Open RTE: 1.1a1svn01142006 Open RTE SVN revision: svn01142006 OPAL: 1.1a1svn01142006 OPAL SVN revision: svn01142006 Prefix: /scratch/modules/opt/openmpi-trunk- nofortran-bproc64 Configured architecture: x86_64-unknown-linux-gnu Configured by: ddd Configured on: Sat Jan 14 17:22:16 MST 2006 $ make MPIRUN='mpirun -mca coll basic' MPI_Allreduce_user_c (cd src ; make MPI_Allreduce_user_c) make[1]: Entering directory `/home/ddd/intel_tests/src' mpicc -g -Isrc -c -o libmpitest.o libmpitest.c mpicc -g -Isrc -o MPI_Allreduce_user_c MPI_Allreduce_user_c.c libmpitest.o -lm make[1]: Leaving directory `/home/ddd/intel_tests/src' mpirun -mca coll basic -n 4 -- `pwd`/src/MPI_Allreduce_user_c MPITEST info (0): Starting MPI_Allreduce_user() test MPITEST_results: MPI_Allreduce_user() all tests PASSED (7076) $ make MPIRUN='mpirun' MPI_Allreduce_user_c (cd src ; make MPI_Allreduce_user_c) make[1]: Entering directory `/home/ddd/intel_tests/src' make[1]: `MPI_Allreduce_user_c' is up to date. make[1]: Leaving directory `/home/ddd/intel_tests/src' mpirun -n 4 -- `pwd`/src/MPI_Allreduce_user_c MPITEST info (0): Starting MPI_Allreduce_user() test MPITEST error (0): i=0, int value=4, expected 1 MPITEST error (0): i=1, int value=4, expected 1 MPITEST error (0): i=2, int value=4, expected 1 MPITEST error (0): i=3, int value=4, expected 1 ...
