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

[cmdlineargs] Limit total number of threads in oversubscription test #38633

Merged
merged 1 commit into from
Dec 1, 2020

Conversation

staticfloat
Copy link
Member

This test was exhausting our limit of 1000 PIDs because the machine it
was running on has 128 cores, which ends up trying to launch a Julia
process with 1280 threads.

This test was exhausting our limit of 1000 PIDs because the machine it
was running on has 128 cores, which ends up trying to launch a Julia
process with 1280 threads.
@Keno
Copy link
Member

Keno commented Dec 1, 2020

Fixes #38631 (hopefully).

@vtjnash
Copy link
Member

vtjnash commented Dec 1, 2020

Win32 failure is interesting (and unrelated)

macOS failure looks like a problem with Artifacts / pkg server

@vtjnash vtjnash merged commit c27d5e9 into master Dec 1, 2020
@vtjnash vtjnash deleted the sf/fewer_oversubscribed_threads branch December 1, 2020 09:58
@vtjnash
Copy link
Member

vtjnash commented Dec 11, 2020

This still might be too many for 32-bit machines, as this is 1.6 GB of stacks:

$ /data/vtjnash/julia/build-x86/usr/bin/julia -t 200 -e 'print(Threads.nthreads())'
OpenBLAS blas_thread_init: pthread_create failed for thread 1 of 8: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 1542271 current, 1542271 max
OpenBLAS blas_thread_init: pthread_create failed for thread 2 of 8: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 1542271 current, 1542271 max
OpenBLAS blas_thread_init: pthread_create failed for thread 3 of 8: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 1542271 current, 1542271 max
OpenBLAS blas_thread_init: pthread_create failed for thread 4 of 8: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 1542271 current, 1542271 max
OpenBLAS blas_thread_init: pthread_create failed for thread 5 of 8: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 1542271 current, 1542271 max
OpenBLAS blas_thread_init: pthread_create failed for thread 6 of 8: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 1542271 current, 1542271 max
OpenBLAS blas_thread_init: pthread_create failed for thread 7 of 8: Resource temporarily unavailable
OpenBLAS blas_thread_init: RLIMIT_NPROC 1542271 current, 1542271 max

signal (11): Segmentation fault
in expression starting at none:0
unknown function (ip: (nil))
Allocations: 2649 (Pool: 2644; Big: 5); GC: 0
Segmentation fault

$ /data/vtjnash/julia/build-x86/usr/bin/julia -t 200 -e 'print(Threads.nthreads())'
WARNING: Error during initialization of module LinearAlgebra:
ErrorException("could not load library "libopenblas"
libopenblas.so: failed to map segment from shared object")
fatal: error thrown and no exception handler available.
InitError(mod=:Profile, error=ErrorException("could not allocate space for 10000000 instruction pointers"))
unknown function (ip: 0xe533e49d)
unknown function (ip: 0xe533e354)
unknown function (ip: 0xe533e136)
unknown function (ip: 0xe533e154)
jl_apply_generic at /data/vtjnash/julia/build-x86/usr/bin/../lib/libjulia-internal.so.1.7 (unknown line)
unknown function (ip: 0xf7a208c3)
unknown function (ip: 0xf7a06e34)
julia_init__threading at /data/vtjnash/julia/build-x86/usr/bin/../lib/libjulia-internal.so.1.7 (unknown line)
repl_entrypoint at /data/vtjnash/julia/build-x86/usr/bin/../lib/libjulia-internal.so.1.7 (unknown line)
main at /data/vtjnash/julia/cli/loader_exe.c:51
__libc_start_main at /lib/i386-linux-gnu/libc.so.6 (unknown line)
_start at /data/vtjnash/julia/build-x86/usr/bin/julia (unknown line)

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

Successfully merging this pull request may close these issues.

3 participants