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.

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

part framework not sessions friendly in certain cases #12887

Open
hppritcha opened this issue Oct 28, 2024 · 0 comments
Open

part framework not sessions friendly in certain cases #12887

hppritcha opened this issue Oct 28, 2024 · 0 comments
Assignees

Comments

@hppritcha
Copy link
Member

I've been using --enable-mca-dso config option with main to help debug some problems and discovered in the process that the framework doesn't work in a sessions multi init/finalize use case:

[er-head:1037963] *** Process received signal ***
[er-head:1037963] Signal: Segmentation fault (11)
[er-head:1037963] Signal code: Address not mapped (1)
[er-head:1037963] Failing at address: 0x7fffebdfa260
[er-head:1037963] [ 0] /lib64/libpthread.so.0(+0x12cf0)[0x7ffff7765cf0]
[er-head:1037963] [ 1] /lib64/libc.so.6(+0xcbe67)[0x7ffff7459e67]
[er-head:1037963] [ 2] /home/hpritchard/ompi-er2/install_main/lib/libmpi.so.0(mca_part_base_select+0x172)[0x7ffff7ad7289]
[er-head:1037963] [ 3] /home/hpritchard/ompi-er2/install_main/lib/libmpi.so.0(+0x863d9)[0x7ffff79f93d9]
[er-head:1037963] [ 4] /home/hpritchard/ompi-er2/install_main/lib/libmpi.so.0(ompi_mpi_instance_init+0x72)[0x7ffff79f99bc]
[er-head:1037963] [ 5] /home/hpritchard/ompi-er2/install_main/lib/libmpi.so.0(MPI_Session_init+0x1dd)[0x7ffff7a76982]
[er-head:1037963] [ 6] ./hello_sessions_c[0x40078d]
[er-head:1037963] [ 7] /lib64/libc.so.6(__libc_start_main+0xe5)[0x7ffff73c8d85]
[er-head:1037963] [ 8] ./hello_sessions_c[0x40068e]
[er-head:1037963] *** End of error message ***
@hppritcha hppritcha self-assigned this Oct 28, 2024
hppritcha added a commit to hppritcha/ompi that referenced this issue Oct 28, 2024
even when configuring with --enable-mca-dso.

The partition framework open/close was not being properly managed
prior to this patch, leading to segfaults upon reinitialization if
the app starts another session after closing a previous one and the
Open MPI was configured with --enable-mca-dso.

Related to open-mpi#12887

Signed-off-by: Howard Pritchard <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant