August 2024 - discuss - mailman3.cels.anl.gov

Re: [mpich-discuss] How to read --pmi-process-mapping
by Zhou, Hui 07 Aug '24

07 Aug '24

Hi Niyaz, The triplet meanings are (start node id, number of nodes, processes per node), and processes are assigned round-robin fashion. So "(0, 2, 1)" with 2 processes on 2 nodes expands into "0, 1" - first process on node 0 and 2nd process on node 1. Hope that helps. -- Hui ________________________________ From: Niyaz Murshed via discuss <discuss(a)mpich.org> Sent: Wednesday, August 7, 2024 12:18 PM To: discuss(a)mpich.org <discuss(a)mpich.org> Cc: Niyaz Murshed <Niyaz.Murshed(a)arm.com>; nd <nd(a)arm.com> Subject: [mpich-discuss] How to read --pmi-process-mapping Hello, I am launching 2 processes on 2 nodes. The processes get rank 0 and 1 --version 4. 3. 0a1 --iface-ip-env-name PMI_HOSTNAME --hostname 192. 168. 2. 100 --global-core-map 0,1,2 --pmi-id-map 0,0 --global-process-count 2 --auto-cleanup 1 --pmi-kvsname ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hello, I am launching 2 processes on 2 nodes. The processes get rank 0 and 1 --version 4.3.0a1 --iface-ip-env-name PMI_HOSTNAME --hostname 192.168.2.100 --global-core-map 0,1,2 --pmi-id-map 0,0 --global-process-count 2 --auto-cleanup 1 --pmi-kvsname kvs_31321_0_150440369_ampere-altra-2-1 --pmi-process-mapping (vector,(0,2,1)) I see once rankmap is generated, its converted to a string “(vector,(0,2,1))”. How do we read this as? Regards, Niyaz

1 0

How to read --pmi-process-mapping
by Niyaz Murshed 07 Aug '24

07 Aug '24

Hello, I am launching 2 processes on 2 nodes. The processes get rank 0 and 1 --version 4.3.0a1 --iface-ip-env-name PMI_HOSTNAME --hostname 192.168.2.100 --global-core-map 0,1,2 --pmi-id-map 0,0 --global-process-count 2 --auto-cleanup 1 --pmi-kvsname kvs_31321_0_150440369_ampere-altra-2-1 --pmi-process-mapping (vector,(0,2,1)) I see once rankmap is generated, its converted to a string “(vector,(0,2,1))”. How do we read this as? Regards, Niyaz

1 0

Re: [mpich-discuss] Error in mpiech compilation
by Zhou, Hui 06 Aug '24

06 Aug '24

Did you configure and build libpmi separately? Anyway, try again with the clean source tree and let us know if the issue persists. Hui ________________________________ From: Amine CHADIL via discuss <discuss(a)mpich.org> Sent: Tuesday, August 6, 2024 10:44 AM To: discuss(a)mpich.org <discuss(a)mpich.org> Cc: Amine CHADIL <amine.chadil(a)cnrs.fr> Subject: [mpich-discuss] Error in mpiech compilation Hello I followed the instructions given in the MPICH Installer’s Guide, and thank you for sharing it. The compilation didn’t get through; you can find here attached, as requested in the document, the log files. Could you please tell me if you can see why it’s not working. Thank you in advance for your help Best regards Amine CHADIL CNRS, MSME

1 0

Error in mpiech compilation
by Amine CHADIL 06 Aug '24

06 Aug '24

Hello I followed the instructions given in the MPICH Installer’s Guide, and thank you for sharing it. The compilation didn’t get through; you can find here attached, as requested in the document, the log files. Could you please tell me if you can see why it’s not working. Thank you in advance for your help Best regards Amine CHADIL CNRS, MSME

1 0

Re: [mpich-discuss] OFI poll failed error if using more than one cluster node
by Zhou, Hui 05 Aug '24

05 Aug '24

Hi Stephen, No, the PMI choice should not affect the init error you are seeing. Sorry for the lack of responses. I didn't find any clues in the debug log. Do you have the issue running on the local node or only when launching on remote nodes? What kind of systems do you have? -- Hui ________________________________ From: Stephen Wong <stephen.photond(a)gmail.com> Sent: Monday, August 5, 2024 4:41 AM To: Zhou, Hui <zhouh(a)anl.gov> Cc: discuss(a)mpich.org <discuss(a)mpich.org> Subject: Re: [mpich-discuss] OFI poll failed error if using more than one cluster node Since there is no update to this after posting the Debug messages, I have since got a new suggestion from someone. I was told that MPICH was only built with PMIx support whereas mpiexec. hydra only provides a PMI server. Could that be the source ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Since there is no update to this after posting the Debug messages, I have since got a new suggestion from someone. I was told that MPICH was only built with PMIx support whereas mpiexec.hydra only provides a PMI server. Could that be the source of the problem? On Mon, 29 Jul 2024 at 11:15, Stephen Wong <stephen.photond(a)gmail.com<mailto:[email protected]>> wrote: For the build configured with --with-device=ch4:ofi, I got ==== Various sizes and limits ==== sizeof(MPIDI_per_vci_t): 192 Required minimum FI_VERSION: 0, current version: 10014 provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN [16] 192.168.1.5 provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN [16] 127.0.0.1 provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN6 [28] ::1 provider: shm, score = 4, pref = -2, FI_ADDR_STR [14] - fi_shm://4595 provider: shm, score = 4, pref = -2, FI_ADDR_STR [14] - fi_shm://4595 provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] fe80::1ff:fe23:4567:890a provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] ::1 provider: sm2, score = 3, pref = 0, FI_ADDR_STR [13] - fi_sm2://4595 provider: sm2, score = 3, pref = 0, FI_ADDR_STR [13] - fi_sm2://4595 Required minimum FI_VERSION: 10005, current version: 10014 ==== Capability set configuration ==== libfabric provider: sockets - 192.168.1.0/24<https://urldefense.us/v3/__http://192.168.1.0/24__;!!G_uCfscf7eWS!ek__1tWG5…> MPIDI_OFI_ENABLE_DATA: 1 MPIDI_OFI_ENABLE_AV_TABLE: 1 MPIDI_OFI_ENABLE_SCALABLE_ENDPOINTS: 1 MPIDI_OFI_ENABLE_SHARED_CONTEXTS: 0 MPIDI_OFI_ENABLE_MR_VIRT_ADDRESS: 0 MPIDI_OFI_ENABLE_MR_ALLOCATED: 0 MPIDI_OFI_ENABLE_MR_REGISTER_NULL: 1 MPIDI_OFI_ENABLE_MR_PROV_KEY: 0 MPIDI_OFI_ENABLE_TAGGED: 1 MPIDI_OFI_ENABLE_AM: 1 MPIDI_OFI_ENABLE_RMA: 1 MPIDI_OFI_ENABLE_ATOMICS: 1 MPIDI_OFI_FETCH_ATOMIC_IOVECS: 1 MPIDI_OFI_ENABLE_DATA_AUTO_PROGRESS: 0 MPIDI_OFI_ENABLE_CONTROL_AUTO_PROGRESS: 0 MPIDI_OFI_ENABLE_PT2PT_NOPACK: 1 MPIDI_OFI_ENABLE_TRIGGERED: 0 MPIDI_OFI_ENABLE_HMEM: 0 MPIDI_OFI_NUM_AM_BUFFERS: 8 MPIDI_OFI_NUM_OPTIMIZED_MEMORY_REGIONS: 0 MPIDI_OFI_CONTEXT_BITS: 20 MPIDI_OFI_SOURCE_BITS: 0 MPIDI_OFI_TAG_BITS: 31 MPIDI_OFI_VNI_USE_DOMAIN: 1 MAXIMUM SUPPORTED RANKS: 4294967296 MAXIMUM TAG: 2147483648 ==== Provider global thresholds ==== max_buffered_send: 255 max_buffered_write: 255 max_msg_size: 9223372036854775807 max_order_raw: -1 max_order_war: -1 max_order_waw: -1 tx_iov_limit: 8 rx_iov_limit: 8 rma_iov_limit: 8 max_mr_key_size: 8 ==== Various sizes and limits ==== MPIDI_OFI_AM_MSG_HEADER_SIZE: 24 MPIDI_OFI_MAX_AM_HDR_SIZE: 255 sizeof(MPIDI_OFI_am_request_header_t): 416 sizeof(MPIDI_OFI_per_vci_t): 52480 MPIDI_OFI_AM_HDR_POOL_CELL_SIZE: 1024 MPIDI_OFI_DEFAULT_SHORT_SEND_SIZE: 16384 ==== OFI dynamic settings ==== num_vcis: 1 num_nics: 1 ====================================== mpiexec -host host1,host2 -n 2 cpi Abort(883025295) on node 1: Fatal error in internal_Init: Other MPI error, error stack: internal_Init(48306).............: MPI_Init(argc=0x7ffcb8c7aedc, argv=0x7ffcb8c7aed0) failed MPII_Init_thread(265)............: MPIR_init_comm_world(34).........: MPIR_Comm_commit(823)............: MPID_Comm_commit_post_hook(222)..: MPIDI_world_post_init(660).......: MPIDI_OFI_init_vcis(842).........: check_num_nics(891)..............: MPIR_Allreduce_allcomm_auto(4726): MPIC_Sendrecv(306)...............: MPIC_Wait(91)....................: MPIR_Wait(780)...................: MPIR_Wait_state(737).............: MPIDI_progress_test(134).........: MPIDI_OFI_handle_cq_error(791)...: OFI poll failed (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) Abort(883025295) on node 0: Fatal error in internal_Init: Other MPI error, error stack: internal_Init(48306).............: MPI_Init(argc=0x7fff18beee1c, argv=0x7fff18beee10) failed MPII_Init_thread(265)............: MPIR_init_comm_world(34).........: MPIR_Comm_commit(823)............: MPID_Comm_commit_post_hook(222)..: MPIDI_world_post_init(660).......: MPIDI_OFI_init_vcis(842).........: check_num_nics(891)..............: MPIR_Allreduce_allcomm_auto(4726): MPIC_Sendrecv(306)...............: MPIC_Wait(91)....................: MPIR_Wait(780)...................: MPIR_Wait_state(737).............: MPIDI_progress_test(134).........: MPIDI_OFI_handle_cq_error(791)...: OFI poll failed (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) On Fri, 26 Jul 2024 at 19:04, Zhou, Hui <zhouh(a)anl.gov<mailto:[email protected]>> wrote: Could you run cpi with MPIR_CVAR_DEBUG_SUMMARY=1 and post the output? Hui Zhou ________________________________ From: Stephen Wong via discuss <discuss(a)mpich.org<mailto:[email protected]>> Sent: Friday, July 26, 2024 4:46 AM To: discuss(a)mpich.org<mailto:[email protected]> <discuss(a)mpich.org<mailto:[email protected]>> Cc: Stephen Wong <stephen.photond(a)gmail.com<mailto:[email protected]>> Subject: [mpich-discuss] OFI poll failed error if using more than one cluster node Hi, (I sent this previously without a subject line. ) I am using MPICH 4. 2. 2 on Ubuntu 24. 04 testing with the small program cpi that calculates the value of pi using MPI. I can start on host1 to run cpi on either host1 or host2 alone and I can ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd Hi, (I sent this previously without a subject line.) I am using MPICH 4.2.2 on Ubuntu 24.04 testing with the small program cpi that calculates the value of pi using MPI. I can start on host1 to run cpi on either host1 or host2 alone and I can start on host2 to run cpi on either host2 or host1 alone. The problem occurs only if I try to use both host1 and host2 together. This is done using, for example, the command mpiexec -host host1,host2 -n 2 cpi then it ends with the error Abort(77718927) on node 1: Fatal error in internal_Init: Other MPI error, error stack: internal_Init(48306).............: MPI_Init(argc=0x7ffdb68e7fec, argv=0x7ffdb68e7fe0) failed MPII_Init_thread(265)............: MPIR_init_comm_world(34).........: MPIR_Comm_commit(823)............: MPID_Comm_commit_post_hook(222)..: MPIDI_world_post_init(660).......: MPIDI_OFI_init_vcis(842).........: check_num_nics(891)..............: MPIR_Allreduce_allcomm_auto(4726): MPIC_Sendrecv(306)...............: MPIC_Wait(91)....................: MPIR_Wait(780)...................: MPIR_Wait_state(737).............: MPIDI_progress_test(134).........: MPIDI_OFI_handle_cq_error(791)...: OFI poll failed (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) Abort(77718927) on node 0: Fatal error in internal_Init: Other MPI error, error stack: internal_Init(48306).............: MPI_Init(argc=0x7ffcb5b28adc, argv=0x7ffcb5b28ad0) failed MPII_Init_thread(265)............: MPIR_init_comm_world(34).........: MPIR_Comm_commit(823)............: MPID_Comm_commit_post_hook(222)..: MPIDI_world_post_init(660).......: MPIDI_OFI_init_vcis(842).........: check_num_nics(891)..............: MPIR_Allreduce_allcomm_auto(4726): MPIC_Sendrecv(306)...............: MPIC_Wait(91)....................: MPIR_Wait(780)...................: MPIR_Wait_state(737).............: MPIDI_progress_test(134).........: MPIDI_OFI_handle_cq_error(791)...: OFI poll failed (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) I searched through the archive of this mailing list and there is only one thread that has this OFI poll failed error. In the thread, it mentioned this has something to do with the device configuration of ch4:ofi ? I checked my configure log and it has device : ch4:ofi (embedded libfabric) in the configuration when I built the MPI. So I am wondering if I should switch this option to something else? If this will fix it. I am not too sure what other option I could substitute for ch4:ofi. ***************************************************** Next I tried running configure for a build with the --enable-device = ch3:nemesis option. Now again I can run the cpi on any of host1 or host2 alone. If I run it on host1 and host2 together, it just crashed with a core dump. Using the --enable-device = ch3:sock configure option resulted in more or less the same problem but now it just quits silently when running on host1 and host2 together. Any ideas? Thanks! Stephen.

1 0

Re: [mpich-discuss] OFI poll failed error if using more than one cluster node
by Stephen Wong 05 Aug '24

05 Aug '24

Since there is no update to this after posting the Debug messages, I have since got a new suggestion from someone. I was told that MPICH was only built with PMIx support whereas mpiexec.hydra only provides a PMI server. Could that be the source of the problem? On Mon, 29 Jul 2024 at 11:15, Stephen Wong <stephen.photond(a)gmail.com> wrote: > For the build configured with --with-device=ch4:ofi, I got > > ==== Various sizes and limits ==== > sizeof(MPIDI_per_vci_t): 192 > Required minimum FI_VERSION: 0, current version: 10014 > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: tcp;ofi_rxm, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN [16] > 192.168.1.5 > provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: udp;ofi_rxd, score = 5, pref = -2, FI_SOCKADDR_IN6 [28] ::1 > provider: shm, score = 4, pref = -2, FI_ADDR_STR [14] - fi_shm://4595 > provider: shm, score = 4, pref = -2, FI_ADDR_STR [14] - fi_shm://4595 > provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: udp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: tcp, score = 0, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: tcp, score = 4, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: sockets, score = 3, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 192.168.1.5 > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] > fe80::1ff:fe23:4567:890a > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN [16] 127.0.0.1 > provider: sockets, score = 5, pref = 0, FI_SOCKADDR_IN6 [28] ::1 > provider: sm2, score = 3, pref = 0, FI_ADDR_STR [13] - fi_sm2://4595 > provider: sm2, score = 3, pref = 0, FI_ADDR_STR [13] - fi_sm2://4595 > Required minimum FI_VERSION: 10005, current version: 10014 > ==== Capability set configuration ==== > libfabric provider: sockets - 192.168.1.0/24 > MPIDI_OFI_ENABLE_DATA: 1 > MPIDI_OFI_ENABLE_AV_TABLE: 1 > MPIDI_OFI_ENABLE_SCALABLE_ENDPOINTS: 1 > MPIDI_OFI_ENABLE_SHARED_CONTEXTS: 0 > MPIDI_OFI_ENABLE_MR_VIRT_ADDRESS: 0 > MPIDI_OFI_ENABLE_MR_ALLOCATED: 0 > MPIDI_OFI_ENABLE_MR_REGISTER_NULL: 1 > MPIDI_OFI_ENABLE_MR_PROV_KEY: 0 > MPIDI_OFI_ENABLE_TAGGED: 1 > MPIDI_OFI_ENABLE_AM: 1 > MPIDI_OFI_ENABLE_RMA: 1 > MPIDI_OFI_ENABLE_ATOMICS: 1 > MPIDI_OFI_FETCH_ATOMIC_IOVECS: 1 > MPIDI_OFI_ENABLE_DATA_AUTO_PROGRESS: 0 > MPIDI_OFI_ENABLE_CONTROL_AUTO_PROGRESS: 0 > MPIDI_OFI_ENABLE_PT2PT_NOPACK: 1 > MPIDI_OFI_ENABLE_TRIGGERED: 0 > MPIDI_OFI_ENABLE_HMEM: 0 > MPIDI_OFI_NUM_AM_BUFFERS: 8 > MPIDI_OFI_NUM_OPTIMIZED_MEMORY_REGIONS: 0 > MPIDI_OFI_CONTEXT_BITS: 20 > MPIDI_OFI_SOURCE_BITS: 0 > MPIDI_OFI_TAG_BITS: 31 > MPIDI_OFI_VNI_USE_DOMAIN: 1 > MAXIMUM SUPPORTED RANKS: 4294967296 > MAXIMUM TAG: 2147483648 > ==== Provider global thresholds ==== > max_buffered_send: 255 > max_buffered_write: 255 > max_msg_size: 9223372036854775807 > max_order_raw: -1 > max_order_war: -1 > max_order_waw: -1 > tx_iov_limit: 8 > rx_iov_limit: 8 > rma_iov_limit: 8 > max_mr_key_size: 8 > ==== Various sizes and limits ==== > MPIDI_OFI_AM_MSG_HEADER_SIZE: 24 > MPIDI_OFI_MAX_AM_HDR_SIZE: 255 > sizeof(MPIDI_OFI_am_request_header_t): 416 > sizeof(MPIDI_OFI_per_vci_t): 52480 > MPIDI_OFI_AM_HDR_POOL_CELL_SIZE: 1024 > MPIDI_OFI_DEFAULT_SHORT_SEND_SIZE: 16384 > ==== OFI dynamic settings ==== > num_vcis: 1 > num_nics: 1 > ====================================== > > *mpiexec -host host1,host2 -n 2 cpi* > > Abort(883025295) on node 1: Fatal error in internal_Init: Other MPI error, > error stack: > internal_Init(48306).............: MPI_Init(argc=0x7ffcb8c7aedc, > argv=0x7ffcb8c7aed0) failed > MPII_Init_thread(265)............: > MPIR_init_comm_world(34).........: > MPIR_Comm_commit(823)............: > MPID_Comm_commit_post_hook(222)..: > MPIDI_world_post_init(660).......: > MPIDI_OFI_init_vcis(842).........: > check_num_nics(891)..............: > MPIR_Allreduce_allcomm_auto(4726): > MPIC_Sendrecv(306)...............: > MPIC_Wait(91)....................: > MPIR_Wait(780)...................: > MPIR_Wait_state(737).............: > MPIDI_progress_test(134).........: > MPIDI_OFI_handle_cq_error(791)...: OFI poll failed > (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) > Abort(883025295) on node 0: Fatal error in internal_Init: Other MPI error, > error stack: > internal_Init(48306).............: MPI_Init(argc=0x7fff18beee1c, > argv=0x7fff18beee10) failed > MPII_Init_thread(265)............: > MPIR_init_comm_world(34).........: > MPIR_Comm_commit(823)............: > MPID_Comm_commit_post_hook(222)..: > MPIDI_world_post_init(660).......: > MPIDI_OFI_init_vcis(842).........: > check_num_nics(891)..............: > MPIR_Allreduce_allcomm_auto(4726): > MPIC_Sendrecv(306)...............: > MPIC_Wait(91)....................: > MPIR_Wait(780)...................: > MPIR_Wait_state(737).............: > MPIDI_progress_test(134).........: > MPIDI_OFI_handle_cq_error(791)...: OFI poll failed > (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) > > > On Fri, 26 Jul 2024 at 19:04, Zhou, Hui <zhouh(a)anl.gov> wrote: > >> Could you run cpi with MPIR_CVAR_DEBUG_SUMMARY=1 and post the output? >> >> Hui Zhou >> ------------------------------ >> *From:* Stephen Wong via discuss <discuss(a)mpich.org> >> *Sent:* Friday, July 26, 2024 4:46 AM >> *To:* discuss(a)mpich.org <discuss(a)mpich.org> >> *Cc:* Stephen Wong <stephen.photond(a)gmail.com> >> *Subject:* [mpich-discuss] OFI poll failed error if using more than one >> cluster node >> >> Hi, (I sent this previously without a subject line. ) I am using MPICH 4. >> 2. 2 on Ubuntu 24. 04 testing with the small program cpi that calculates >> the value of pi using MPI. I can start on host1 to run cpi on either host1 >> or host2 alone and I can >> ZjQcmQRYFpfptBannerStart >> This Message Is From an External Sender >> This message came from outside your organization. >> >> ZjQcmQRYFpfptBannerEnd >> Hi, >> >> (I sent this previously without a subject line.) >> >> I am using MPICH 4.2.2 on Ubuntu 24.04 testing with the small program cpi >> that calculates the value of pi using MPI. I can start on host1 to run cpi >> on either host1 or host2 alone and I can start on host2 to run cpi on >> either host2 or host1 alone. The problem occurs only if I try to use both >> host1 and host2 together. >> >> This is done using, for example, the command >> *mpiexec -host host1,host2 -n 2 cpi* >> then it ends with the error >> >> Abort(77718927) on node 1: Fatal error in internal_Init: Other MPI error, >> error stack: >> internal_Init(48306).............: MPI_Init(argc=0x7ffdb68e7fec, >> argv=0x7ffdb68e7fe0) failed >> MPII_Init_thread(265)............: >> MPIR_init_comm_world(34).........: >> MPIR_Comm_commit(823)............: >> MPID_Comm_commit_post_hook(222)..: >> MPIDI_world_post_init(660).......: >> MPIDI_OFI_init_vcis(842).........: >> check_num_nics(891)..............: >> MPIR_Allreduce_allcomm_auto(4726): >> MPIC_Sendrecv(306)...............: >> MPIC_Wait(91)....................: >> MPIR_Wait(780)...................: >> MPIR_Wait_state(737).............: >> MPIDI_progress_test(134).........: >> MPIDI_OFI_handle_cq_error(791)...: OFI poll failed >> (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) >> Abort(77718927) on node 0: Fatal error in internal_Init: Other MPI error, >> error stack: >> internal_Init(48306).............: MPI_Init(argc=0x7ffcb5b28adc, >> argv=0x7ffcb5b28ad0) failed >> MPII_Init_thread(265)............: >> MPIR_init_comm_world(34).........: >> MPIR_Comm_commit(823)............: >> MPID_Comm_commit_post_hook(222)..: >> MPIDI_world_post_init(660).......: >> MPIDI_OFI_init_vcis(842).........: >> check_num_nics(891)..............: >> MPIR_Allreduce_allcomm_auto(4726): >> MPIC_Sendrecv(306)...............: >> MPIC_Wait(91)....................: >> MPIR_Wait(780)...................: >> MPIR_Wait_state(737).............: >> MPIDI_progress_test(134).........: >> MPIDI_OFI_handle_cq_error(791)...: OFI poll failed >> (ofi_events.c:793:MPIDI_OFI_handle_cq_error:Input/output error) >> I searched through the archive of this mailing list and there is only one >> thread that has this OFI poll failed error. >> In the thread, it mentioned this has something to do with the device >> configuration of ch4:ofi ? >> I checked my configure log and it has >> device : ch4:ofi (embedded libfabric) >> in the configuration when I built the MPI. So I am wondering if I should >> switch this option to something else? If this will fix it. I am not too >> sure what other option I could substitute for ch4:ofi. >> >> ***************************************************** >> >> Next I tried running configure for a build with the --enable-device = >> ch3:nemesis option. >> Now again I can run the cpi on any of host1 or host2 alone. If I run it >> on host1 and host2 together, it just crashed with a core dump. >> >> Using the --enable-device = ch3:sock configure option resulted in more or >> less the same problem but now it just quits silently when running on host1 >> and host2 together. >> >> Any ideas? >> Thanks! >> Stephen. >> >>

1 0

Re: [mpich-discuss] mpich-4.2.2 and ABI compatibility
by Zhou, Hui 01 Aug '24

01 Aug '24

I agree to all your guesses 🙂 -- Hui ________________________________ From: Edric Ellis via discuss <discuss(a)mpich.org> Sent: Wednesday, July 31, 2024 4:39 AM To: discuss(a)mpich.org <discuss(a)mpich.org> Cc: Edric Ellis <eellis(a)mathworks.com> Subject: [mpich-discuss] mpich-4.2.2 and ABI compatibility In trying to upgrade our build of mpich to 4. 2. 2 from 4. 1. 2, I notice that the libraries are not completely binary compatible - at least, not in the way that our system presumes will be the case. We currently support users bringing their own ZjQcmQRYFpfptBannerStart This Message Is From an External Sender This message came from outside your organization. ZjQcmQRYFpfptBannerEnd In trying to upgrade our build of mpich to 4.2.2 from 4.1.2, I notice that the libraries are not completely binary compatible - at least, not in the way that our system presumes will be the case. We currently support users bringing their own MPI, providing it is ABI-compatible with our "primary" MPI, which is whatever current release of mpich we're using. In the past, this has meant that in practice, because nothing we do needs the very latest functionality provided by mpich, even very old mpich-derivative MPIs would work. With mpich-4.2.2, I see from the CHANGES file that many of the functions that were implemented as macros (e.g. MPI_Comm_f2c) are now "real" functions – to support the requirements of the MPI-4.1 standard. I notice further that the implementation of those functions is basically the same as the old macro. For now, I'm planning to work around this by effectively using the old macro definitions for cases where a user-supplied MPI library doesn't provide the necessary entry points. Does that seem reasonable? In fact, I note this PR https://urldefense.us/v3/__https://github.com/pmodels/mpich/pull/6711__;!!G… <https://urldefense.us/v3/__https://github.com/pmodels/mpich/pull/6711__;!!G…> in which Ken Raffenetti says: “Just thinking from the ABI perspective, I don't think there is any backwards compatibility concern. Apps compiled when these were macros will continue to work with newer MPICH libraries.” So, I infer from that comment that the approach I’m contemplating should be fine. I imagine the other ABI-compatible libraries listed here https://urldefense.us/v3/__https://www.mpich.org/abi/__;!!G_uCfscf7eWS!dh3e… <https://urldefense.us/v3/__https://www.mpich.org/abi/__;!!G_uCfscf7eWS!Z3Ge…> will eventually adopt this change too – I don’t suppose there is any known timeline when this might happen? Cheers, Edric.

1 0