[Mvapich-discuss] Using MVAPICH2 is Singularity container
vru.inbri at yahoo.co.uk
vru.inbri at yahoo.co.uk
Thu Feb 18 14:14:18 EST 2021
I get that after setting MV2_SHOW_ENV_INFO=3
MVAPICH2-2.3.5 Parameters--------------------------------------------------------------------- PROCESSOR ARCH NAME : MV2_ARCH_AMD_EPYC_7742_128 PROCESSOR FAMILY NAME : MV2_CPU_FAMILY_AMD PROCESSOR MODEL NUMBER : 49 HCA NAME : MV2_HCA_MLX_CX_EDR HETEROGENEOUS HCA : NO MV2_VBUF_TOTAL_SIZE : 16384 MV2_IBA_EAGER_THRESHOLD : 16384 MV2_RDMA_FAST_PATH_BUF_SIZE : 5120 MV2_PUT_FALLBACK_THRESHOLD : 8192 MV2_GET_FALLBACK_THRESHOLD : 0 MV2_EAGERSIZE_1SC : 8192 MV2_SMP_EAGERSIZE : 16385 MV2_SMP_QUEUE_LENGTH : 262144 MV2_SMP_NUM_SEND_BUFFER : 32 MV2_SMP_BATCH_SIZE : 8 Tuning Table: : MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR---------------------------------------------------------------------
MVAPICH2 All Parameters MPIRUN_RSH_LAUNCH : 0 MV2_SHMEM_BACKED_UD_CM : 0 MV2_3DTORUS_SUPPORT : 0 MV2_NUM_SA_QUERY_RETRIES : 20 MV2_NUM_SLS : 8 MV2_DEFAULT_SERVICE_LEVEL : 0 MV2_PATH_SL_QUERY : 0 MV2_USE_QOS : 0 MV2_USE_MCAST : 0 MV2_USE_RDMA_CM_MCAST : 0 MV2_MCAST_BCAST_MIN_MSG : 1 MV2_MCAST_BCAST_MAX_MSG : 524288 MV2_ALLGATHER_BRUCK_THRESHOLD : 524288 MV2_ALLGATHER_RD_THRESHOLD : 81920 MV2_ALLGATHER_REVERSE_RANKING : 1 MV2_ALLGATHERV_RD_THRESHOLD : 0 MV2_ALLREDUCE_2LEVEL_MSG : 262144 MV2_ALLREDUCE_SHORT_MSG : 2048 MV2_ALLTOALL_MEDIUM_MSG : 16384 MV2_ALLTOALL_SMALL_MSG : 2048 MV2_ALLTOALL_THROTTLE_FACTOR : 32 MV2_BCAST_TWO_LEVEL_SYSTEM_SIZE : 64 MV2_GATHER_SWITCH_PT : 0 MV2_INTRA_SHMEM_REDUCE_MSG : 2048 MV2_KNOMIAL_2LEVEL_BCAST_MESSAGE_SIZE_THRESHOLD : 2048 MV2_KNOMIAL_2LEVEL_BCAST_SYSTEM_SIZE_THRESHOLD : 64 MV2_KNOMIAL_INTER_LEADER_THRESHOLD : 65536 MV2_KNOMIAL_INTER_NODE_FACTOR : 4 MV2_KNOMIAL_INTRA_NODE_FACTOR : 4 MV2_KNOMIAL_INTRA_NODE_THRESHOLD : 131072 MV2_RED_SCAT_LARGE_MSG : 524288 MV2_RED_SCAT_SHORT_MSG : 64 MV2_REDUCE_2LEVEL_MSG : 16384 MV2_REDUCE_SHORT_MSG : 8192 MV2_SCATTER_MEDIUM_MSG : 0 MV2_SCATTER_SMALL_MSG : 0 MV2_SHMEM_ALLREDUCE_MSG : 32768 MV2_SHMEM_COLL_MAX_MSG_SIZE : 131072 MV2_SHMEM_COLL_NUM_COMM : 32 MV2_SHMEM_COLL_NUM_PROCS : 128 MV2_SHMEM_COLL_SPIN_COUNT : 5 MV2_SHMEM_REDUCE_MSG : 4096 MV2_USE_BCAST_SHORT_MSG : 16384 MV2_USE_DIRECT_GATHER : 1 MV2_USE_DIRECT_GATHER_SYSTEM_SIZE_MEDIUM : 1024 MV2_USE_DIRECT_GATHER_SYSTEM_SIZE_SMALL : 384 MV2_USE_DIRECT_SCATTER : 1 MV2_USE_OSU_COLLECTIVES : 1 MV2_USE_OSU_NB_COLLECTIVES : 1 MV2_USE_KNOMIAL_2LEVEL_BCAST : 1 MV2_USE_KNOMIAL_INTER_LEADER_BCAST : 1 MV2_USE_SCATTER_RD_INTER_LEADER_BCAST : 1 MV2_USE_SCATTER_RING_INTER_LEADER_BCAST : 1 MV2_USE_SHMEM_ALLREDUCE : 1 MV2_USE_SHMEM_BARRIER : 1 MV2_USE_SHMEM_BCAST : 1 MV2_USE_SHMEM_COLL : 1 MV2_USE_SHMEM_REDUCE : 1 MV2_USE_TWO_LEVEL_GATHER : 1 MV2_USE_TWO_LEVEL_SCATTER : 1 MV2_USE_XOR_ALLTOALL : 1 MV2_ENABLE_SOCKET_AWARE_COLLECTIVES : 1 MV2_USE_SOCKET_AWARE_ALLREDUCE : 1 MV2_USE_SOCKET_AWARE_BARRIER : 1 MV2_USE_SOCKET_AWARE_SHARP_ALLREDUCE : 0 MV2_SOCKET_AWARE_ALLREDUCE_MAX_MSG : 2048 MV2_SOCKET_AWARE_ALLREDUCE_MIN_MSG : 1 MV2_DEFAULT_SRC_PATH_BITS : 0 MV2_DEFAULT_STATIC_RATE : 0 MV2_DEFAULT_TIME_OUT : 330772 MV2_DEFAULT_MTU : 5 MV2_DEFAULT_PKEY : 0 MV2_DEFAULT_QKEY : 0 MV2_DEFAULT_PORT : 1 MV2_DEFAULT_GID_INDEX : 0 MV2_DEFAULT_PSN : 0 MV2_DEFAULT_MAX_RECV_WQE : 128 MV2_DEFAULT_MAX_SEND_WQE : 64 MV2_DEFAULT_MAX_SG_LIST : 1 MV2_DEFAULT_MIN_RNR_TIMER : 12 MV2_DEFAULT_QP_OUS_RD_ATOM : 272 MV2_DEFAULT_RETRY_COUNT : 84677639 MV2_DEFAULT_RNR_RETRY : 202639111 MV2_DEFAULT_MAX_CQ_SIZE : 40000 MV2_DEFAULT_MAX_RDMA_DST_OPS : 4 MV2_INITIAL_PREPOST_DEPTH : 10 MV2_IWARP_MULTIPLE_CQ_THRESHOLD : 32 MV2_NUM_HCAS : 1 MV2_NUM_PORTS : 1 MV2_NUM_QP_PER_PORT : 1 MV2_MAX_RDMA_CONNECT_ATTEMPTS : 20 MV2_ON_DEMAND_UD_INFO_EXCHANGE : 0 MV2_PREPOST_DEPTH : 64 MV2_HOMOGENEOUS_CLUSTER : 0 MV2_NUM_CQES_PER_POLL : 96ENDWINDOW_SIZE : 400
MV2_UD_VBUF_POOL_SIZE : 8192 MV2_UD_ZCOPY_RQ_SIZE : 4096 MV2_UD_ZCOPY_THRESHOLD : 16384 MV2_UD_ZCOPY_NUM_RETRY : 50000 MV2_USE_UD_ZCOPY : 1 MV2_USE_UD_HYBRID : 0 MV2_USE_ONLY_UD : 0 MV2_HYBRID_ENABLE_THRESHOLD : 1024 MV2_HYBRID_MAX_RC_CONN : 32 MV2_ASYNC_THREAD_STACK_SIZE : 1048576 MV2_THREAD_YIELD_SPIN_THRESHOLD : 5 MV2_SUPPORT_DPM : 0 MV2_USE_HUGEPAGES : 1---------------------------------------------------------------------
Collective Tuning Tables Collective Architecture Interconnect Allgather MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Allreduce MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Alltoall MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Alltoallv MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Broadcast MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Gather MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Reduce MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Scatter MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR
---------------------------------------------------------------------
MV2_DREG_CACHE_LIMIT : 0 MV2_IBA_EAGER_THRESHOLD : 16384 MV2_MAX_INLINE_SIZE : 168 MV2_MAX_R3_PENDING_DATA : 524288 MV2_MED_MSG_RAIL_SHARING_POLICY : 0 MV2_NDREG_ENTRIES : 8704 MV2_NUM_RDMA_BUFFER : 16 MV2_NUM_SPINS_BEFORE_LOCK : 2000 MV2_POLLING_LEVEL : 1 MV2_POLLING_SET_LIMIT : 64 MV2_POLLING_SET_THRESHOLD : 256 MV2_R3_NOCACHE_THRESHOLD : 32768 MV2_R3_THRESHOLD : 4096 MV2_RAIL_SHARING_LARGE_MSG_THRESHOLD : 16384 MV2_RAIL_SHARING_MED_MSG_THRESHOLD : 2048 MV2_RAIL_SHARING_POLICY : 4 MV2_RDMA_EAGER_LIMIT : 32 MV2_RDMA_FAST_PATH_BUF_SIZE : 5120 MV2_RDMA_NUM_EXTRA_POLLS : 1 MV2_RNDV_EXT_SENDQ_SIZE : 5 MV2_RNDV_PROTOCOL : 4 MV2_SMP_RNDV_PROTOCOL : 4 MV2_SMALL_MSG_RAIL_SHARING_POLICY : 0 MV2_SPIN_COUNT : 5000 MV2_SRQ_LIMIT : 10 MV2_SRQ_MAX_SIZE : 32767 MV2_SRQ_SIZE : 80 MV2_STRIPING_THRESHOLD : 16384 MV2_USE_COALESCE : 1 MV2_USE_XRC : 0 MV2_VBUF_MAX : -1 MV2_VBUF_POOL_SIZE : 80 MV2_VBUF_SECONDARY_POOL_SIZE : 16 MV2_VBUF_TOTAL_SIZE : 16384 MV2_USE_IWARP_MODE : 0 MV2_CPU_BINDING_POLICY : hybrid MV2_USE_HWLOC_CPU_BINDING : 1 MV2_ENABLE_AFFINITY : 1 MV2_ENABLE_LEASTLOAD : 0 MV2_SMP_BATCH_SIZE : 8 MV2_SMP_EAGERSIZE : 16385 MV2_SMP_QUEUE_LENGTH : 262144 MV2_SMP_NUM_SEND_BUFFER : 32 MV2_SMP_SEND_BUF_SIZE : 16384 MV2_USE_SHARED_MEM : 1 MV2_SMP_CMA_MAX_SIZE : 4194304 MV2_SMP_LIMIC2_MAX_SIZE : 0 MV2_SHOW_ENV_INFO : 3 MV2_DEFAULT_PUT_GET_LIST_SIZE : 200 MV2_EAGERSIZE_1SC : 8192 MV2_GET_FALLBACK_THRESHOLD : 0 MV2_PIN_POOL_SIZE : 2097152 MV2_PUT_FALLBACK_THRESHOLD : 8192 MV2_USE_RDMA_CM : 0 MV2_UD_MAX_ACK_PENDING : 100 MV2_UD_MAX_RECV_WQE : 4096 MV2_UD_MAX_RETRY_TIMEOUT : 20000000 MV2_UD_MAX_SEND_WQE : 2048 MV2_UD_MTU : 4096 MV2_UD_NUM_MSG_LIMIT : 512 MV2_UD_NUM_ZCOPY_RNDV_QPS : 64 MV2_UD_PROGRESS_SPIN : 1200 MV2_UD_PROGRESS_TIMEOUT : 48000 MV2_UD_RECVWINDOW_SIZE : 2501 MV2_UD_RETRY_COUNT : 1024 MV2_UD_RETRY_TIMEOUT : 500000 MV2_UD_SENDWINDOW_SIZE : 400 MV2_UD_VBUF_POOL_SIZE : 8192 MV2_UD_ZCOPY_RQ_SIZE : 4096 MV2_UD_ZCOPY_THRESHOLD : 16384 MV2_UD_ZCOPY_NUM_RETRY : 50000 MV2_USE_UD_ZCOPY : 1 MV2_USE_UD_HYBRID : 0 MV2_USE_ONLY_UD : 0 MV2_HYBRID_ENABLE_THRESHOLD : 1024 MV2_HYBRID_MAX_RC_CONN : 32 MV2_ASYNC_THREAD_STACK_SIZE : 1048576 MV2_THREAD_YIELD_SPIN_THRESHOLD : 5 MV2_SUPPORT_DPM : 0 MV2_USE_HUGEPAGES : 1---------------------------------------------------------------------
Collective Tuning Tables Collective Architecture Interconnect Allgather MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Allreduce MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Alltoall MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Alltoallv MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Broadcast MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Gather MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Reduce MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR Scatter MV2_ARCH_AMD_EPYC_7742_128 MV2_HCA_MLX_CX_EDR
---------------------------------------------------------------------
On Thursday, 18 February 2021, 20:00:18 CET, vru.inbri--- via Mvapich-discuss <mvapich-discuss at lists.osu.edu> wrote:
When using a different number of processors the error becomes:
Program received signal SIGSEGV: Segmentation fault - invalid memory reference.
Does it help?
On Thursday, 18 February 2021, 19:38:48 CET, vru.inbri--- via Mvapich-discuss <mvapich-discuss at lists.osu.edu> wrote:
Hi
I built a Singularity container with Ubuntu, GNU compilers and MVAPICH2 2.3.5
When trying to run it on our cluster it fails with errors like:
Fatal error in PMPI_Waitall:Other MPI error, error stack:PMPI_Waitall(419)..................: MPI_Waitall(count=7, req_array=0x55d7f03d0290, status_array=0x55d7f03b4e50) failedMPIR_Waitall_impl(248).............:MPIDI_CH3I_Progress(285)...........:handle_read(1350)..................:handle_read_individual(1408).......:MPIDI_CH3I_MRAIL_Parse_header(1502): Control shouldn't reach here in prototype, header %d (errno 71)
As a test I also installed the same OS, compilers and libraries in an empty virtual machine (directly, without using singularity) and everything works without problem
Does this make any sense for you?
Vru
_______________________________________________
Mvapich-discuss mailing list
Mvapich-discuss at lists.osu.edu
https://lists.osu.edu/mailman/listinfo/mvapich-discuss
_______________________________________________
Mvapich-discuss mailing list
Mvapich-discuss at lists.osu.edu
https://lists.osu.edu/mailman/listinfo/mvapich-discuss
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osu.edu/pipermail/mvapich-discuss/attachments/20210218/08f1ab13/attachment-0022.html>
More information about the Mvapich-discuss
mailing list