Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Multiple failures running collective-big-count tests with OMPI main branch and 'han' collective component #10187

Open
drwootton opened this issue Mar 29, 2022 · 0 comments
Open

Multiple failures running collective-big-count tests with OMPI main branch and 'han' collective component #10187

drwootton opened this issue Mar 29, 2022 · 0 comments
Labels
Target: main

Comments

@drwootton
Copy link
Contributor

Thank you for taking the time to submit an issue!

Background information

What version of Open MPI are you using? (e.g., v3.0.5, v4.0.2, git branch name and hash, etc.)

OpenMPI main branch

Describe how Open MPI was installed (e.g., from a source/distribution tarball, from a git clone, from an operating system distribution package, etc.)

Built from current main branch (3/22/22)

If you are building/installing from a git clone, please copy-n-paste the output from git submodule status.

 git submodule status
 1b86a35db2816ee9c0f3a41988005a2ba7d29adb 3rd-party/openpmix (v1.1.3-3481-g1b86a35d)
 91f791e209ccbdfb4b8647900d292ef51d52f37d 3rd-party/prrte (psrvr-v2.0.0rc1-4319-g91f791e2)

Please describe the system on which you are running

  • Operating system/version:
  • RHEL 8.4
  • Computer hardware:
  • Single Power8 node
  • Network type:
  • Localhost

Details of the problem

I ran the set of self-checking tests from ompi-tests-public/collective-big-count with collective components specified as --mca coll_han_priority 100 --mca coll han,basic,sm,self,inter,libnbc

The following testcases had failures. The remaining testcases were successful:

  • test_allgather_uniform_count
  • test-alltoall-uniform_count
  • test-gather-uniform-count
  • test-scatter-uniform-count

The tests were compiled by running make in the directory containing the source files

The following environment variables were set for all tests:

BIGCOUNT_HOSTS          : -np 3
BIGCOUNT_MEMORY_PERCENT : 70
BIGCOUNT_MEMORY_DIFF    : 10

The following command fails with a self-check error message indicating invalid data, followed by a SIGSEGV.

mpirun ${BIGCOUNT_HOSTS} -x BIGCOUNT_MEMORY_PERCENT=${BIGCOUNT_MEMORY_PERCENT} -x BIGCOUNT_MEMORY_DIFF=${BIGCOUNT_MEMORY_DIFF} --mca btl ^openib --mca coll_han_priority 100 --mca coll han,basic,sm,self,inter,libnbc /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_allgather_uniform_count

The error and traceback is

Results from MPI_Iallgather(int x 6442450941 = 25769803764 or 24.0 GB):  MPI_IN_PLACE
Rank  1: ERROR: DI in     2147483647 of     6442450941 slots (  33.3 % wrong)
Rank  0: ERROR: DI in     2147483647 of     6442450941 slots (  33.3 % wrong)
Rank  2: PASSED
_____________________ Adjust count to fit in memory: 2147483647 x  60.0% = 1288490188
Root  : payload    61847529024  57.6 GB =  16 dt x 1288490188 count x   3 peers x   1.0 inflation
Peer  : payload    61847529024  57.6 GB =  16 dt x 1288490188 count x   3 peers x   1.0 inflation
Total : payload   185542587072 172.8 GB =  57.6 GB root +  57.6 GB x   2 local peers
_____________________
Results from MPI_Iallgather(double _Complex x 3865470564 = 61847529024 or 57.6 GB):  MPI_IN_PLACE
[c656f6n01:1747159] *** Process received signal ***
[c656f6n01:1747159] Signal: Segmentation fault (11)
[c656f6n01:1747159] Signal code: Address not mapped (1)
[c656f6n01:1747159] Failing at address: 0x1ff9a2999990
[c656f6n01:1747159] [ 0] linux-vdso64.so.1(__kernel_sigtramp_rt64+0x0)[0x2000000604d8]
[c656f6n01:1747159] [ 1] /usr/lib64/libc.so.6(+0xb083c)[0x20000074083c]
[c656f6n01:1747159] [ 2] /u/dwootton/ompi-master/lib/libmpi.so.0(+0x3f85e8)[0x2000004785e8]
[c656f6n01:1747159] [ 3] /u/dwootton/ompi-master/lib/libmpi.so.0(mca_pml_ob1_send_request_schedule_once+0x2a4)[0x20000047c034]
[c656f6n01:1747159] [ 4] /u/dwootton/ompi-master/lib/libmpi.so.0(+0x3ea700)[0x20000046a700]
[c656f6n01:1747159] [ 5] /u/dwootton/ompi-master/lib/libmpi.so.0(+0x3ea7c0)[0x20000046a7c0]
[c656f6n01:1747159] [ 6] /u/dwootton/ompi-master/lib/libmpi.so.0(mca_pml_ob1_recv_frag_callback_ack+0x2e4)[0x20000046cca8]
[c656f6n01:1747159] [ 7] /u/dwootton/ompi-master/lib/libopen-pal.so.0(mca_btl_sm_poll_handle_frag+0x200)[0x2000009f076c]
[c656f6n01:1747159] [ 8] /u/dwootton/ompi-master/lib/libopen-pal.so.0(+0xf0890)[0x2000009f0890]
[c656f6n01:1747159] [ 9] /u/dwootton/ompi-master/lib/libopen-pal.so.0(+0xf0c08)[0x2000009f0c08]
[c656f6n01:1747159] [10] /u/dwootton/ompi-master/lib/libopen-pal.so.0(opal_progress+0x5c)[0x20000093d5b0]
[c656f6n01:1747159] [11] /u/dwootton/ompi-master/lib/libmpi.so.0(+0xded50)[0x20000015ed50]
[c656f6n01:1747159] [12] /u/dwootton/ompi-master/lib/libmpi.so.0(ompi_request_default_wait+0x38)[0x20000015edfc]
[c656f6n01:1747159] [13] /u/dwootton/ompi-master/lib/libmpi.so.0(MPI_Wait+0x194)[0x200000258e38]
[c656f6n01:1747159] [14] /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_allgather_uniform_count[0x100030d8]
[c656f6n01:1747159] [15] /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_allgather_uniform_count[0x100029d0]
[c656f6n01:1747159] [16] /usr/lib64/libc.so.6(+0x24c78)[0x2000006b4c78]
[c656f6n01:1747159] [17] /usr/lib64/libc.so.6(__libc_start_main+0xb4)[0x2000006b4e64]
[c656f6n01:1747159] *** End of error message ***

The following command fails with a self-check error message indicating invalid data, followed by a SIGSEGV.

mpirun ${BIGCOUNT_HOSTS} -x BIGCOUNT_MEMORY_PERCENT=${BIGCOUNT_MEMORY_PERCENT} -x BIGCOUNT_MEMORY_DIFF=${BIGCOUNT_MEMORY_DIFF} --mca btl ^openib --mca coll_han_priority 100 --mca coll han,basic,sm,self,inter,libnbc /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_alltoall_uniform_count

The error message and traceback is

Results from MPI_Alltoall(int x 6442450941 = 25769803764 or 24.0 GB): MPI_IN_PLACE
Rank  1: ERROR: DI in     2147483647 of     2147483647 slots ( 100.0 % wrong)
Rank  2: ERROR: DI in     4294967294 of     2147483647 slots ( 200.0 % wrong)
Rank  0: ERROR: DI in     5637111799 of     2147483647 slots ( 262.5 % wrong)
_____________________ Adjust count to fit in memory: 2147483647 x  60.0% = 1288490188
Root  : payload    61847529024  57.6 GB =  16 dt x 1288490188 count x   3 peers x   1.0 inflation
Peer  : payload    61847529024  57.6 GB =  16 dt x 1288490188 count x   3 peers x   1.0 inflation
Total : payload   185542587072 172.8 GB =  57.6 GB root +  57.6 GB x   2 local peers
[c656f6n01:1748510] *** Process received signal ***
[c656f6n01:1748510] Signal: Segmentation fault (11)
[c656f6n01:1748510] Signal code: Address not mapped (1)
[c656f6n01:1748510] Failing at address: 0x1ff9a2999990
[c656f6n01:1748510] [ 0] linux-vdso64.so.1(__kernel_sigtramp_rt64+0x0)[0x2000000604d8]
[c656f6n01:1748510] [ 1] /usr/lib64/libc.so.6(+0xb083c)[0x20000074083c]
[c656f6n01:1748510] [ 2] /u/dwootton/ompi-master/lib/libmpi.so.0(mca_coll_base_alltoall_intra_basic_inplace+0x22c)[0x2000002b3c94]
[c656f6n01:1748510] [ 3] /u/dwootton/ompi-master/lib/libmpi.so.0(ompi_coll_base_alltoall_intra_basic_linear+0x8c)[0x2000002b5684]
[c656f6n01:1748510] [ 4] /u/dwootton/ompi-master/lib/libmpi.so.0(PMPI_Alltoall+0x538)[0x200000193cd4]
[c656f6n01:1748510] [ 5] /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_alltoall_uniform_count[0x10002dd0]
[c656f6n01:1748510] [ 6] /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_alltoall_uniform_count[0x1000289c]
[c656f6n01:1748510] [ 7] /usr/lib64/libc.so.6(+0x24c78)[0x2000006b4c78]
[c656f6n01:1748510] [ 8] /usr/lib64/libc.so.6(__libc_start_main+0xb4)[0x2000006b4e64]
[c656f6n01:1748510] *** End of error message ***

The following command fails with a self-check error message indicating invalid data, followed by a double
free or storage corruption

mpirun ${BIGCOUNT_HOSTS} -x BIGCOUNT_MEMORY_PERCENT=${BIGCOUNT_MEMORY_PERCENT} -x BIGCOUNT_MEMORY_DIFF=${BIGCOUNT_MEMORY_DIFF} --mca btl ^openib --mca coll_han_priority 100 --mca coll han,basic,sm,self,inter,libnbc /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_gather_uniform_count

The error and traceback is

_____________________
Results from MPI_Igather(int x 6442450941 = 25769803764 or 24.0 GB):
Rank  0: ERROR: DI in     4294967292 of     6442450941 slots (  66.7 % wrong)
_____________________
Results from MPI_Igather(double _Complex x 6442450941 = 103079215056 or 96.0 GB):
Rank  0: ERROR: DI in     4294967292 of     6442450941 slots (  66.7 % wrong)
double free or corruption (out)
[c656f6n01:1748925] *** Process received signal ***
[c656f6n01:1748925] Signal: Aborted (6)
[c656f6n01:1748925] Signal code:  (-6)
[c656f6n01:1748925] [ 0] linux-vdso64.so.1(__kernel_sigtramp_rt64+0x0)[0x2000000604d8]
[c656f6n01:1748925] [ 1] /usr/lib64/libc.so.6(gsignal+0xd8)[0x2000006d44d8]
[c656f6n01:1748925] [ 2] /usr/lib64/libc.so.6(abort+0x164)[0x2000006b462c]
[c656f6n01:1748925] [ 3] /usr/lib64/libc.so.6(+0x908bc)[0x2000007208bc]
[c656f6n01:1748925] [ 4] /usr/lib64/libc.so.6(+0x9b828)[0x20000072b828]
[c656f6n01:1748925] [ 5] /usr/lib64/libc.so.6(+0x9e0ec)[0x20000072e0ec]
[c656f6n01:1748925] [ 6] ./test_gather_uniform_count[0x100030b0]
[c656f6n01:1748925] [ 7] ./test_gather_uniform_count[0x10002920]
[c656f6n01:1748925] [ 8] /usr/lib64/libc.so.6(+0x24c78)[0x2000006b4c78]
[c656f6n01:1748925] [ 9] /usr/lib64/libc.so.6(__libc_start_main+0xb4)[0x2000006b4e64]
[c656f6n01:1748925] *** End of error message ***

The following command fails with a self-check error message indicating invalid data

mpirun ${BIGCOUNT_HOSTS} -x BIGCOUNT_MEMORY_PERCENT=${BIGCOUNT_MEMORY_PERCENT} -x BIGCOUNT_MEMORY_DIFF=${BIGCOUNT_MEMORY_DIFF} --mca btl ^openib --mca coll_han_priority 100 --mca coll han,basic,sm,self,inter,libnbc /u/dwootton/bigcount-master/BigCountUpstream/ompi-tests-public/collective-big-count/./test_scatter_uniform_count

The error message is

____________________
Results from MPI_Iscatter(int x 6442450941 = 25769803764 or 24.0 GB):
Rank  2: ERROR: DI in     2147483647 of     2147483647 slots ( 100.0 % wrong)
Rank  1: PASSED
Rank  0: PASSED
____________________
Results from MPI_Iscatter(double _Complex x 6442450941 = 103079215056 or 96.0 GB):
Rank  2: ERROR: DI in     2147483647 of     2147483647 slots ( 100.0 % wrong)
Rank  1: PASSED
Rank  0: PASSED
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
Target: main
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@drwootton @awlauria @wenduwan