tight_layout gives segfault when MKL is used

Reproduziere die beiden Beispiele zur Importreihenfolge mit Julia 1.7, MKL 0.4.2, PyPlot 2.10.0 und PyCall 1.92.5. Beginne bei src/PyPlot.jl rund um tight_layout (Zeilen 174–177) und folge dem im Trace gezeigten PyCall-Aufruf. Erledigt ist die Aufgabe, wenn tight_layout nicht mehr mit einem Segmentation Fault abbricht, wenn MKL vor PyPlot geladen wird, und das gemeldete Verhalten weiterhin durch einen Regressionstest abgedeckt ist.

This gives a segfault:

using MKL
using PyPlot
fig = PyPlot.figure()
plot(rand(10), rand(10))
tight_layout()

signal (11): Segmentation fault
in expression starting at REPL[5]:1
mkl_blas_avx2_xdcopy at /home/jmlim/.julia/artifacts/72d4adc3ef9236a92f4fefeb0291cb6e8aaae2d7/lib/libmkl_avx2.so.1 (unknown line)
mkl_blas_dcopy at /home/jmlim/.julia/artifacts/72d4adc3ef9236a92f4fefeb0291cb6e8aaae2d7/lib/libmkl_intel_thread.so.1 (unknown line)
dcopy_ at /home/jmlim/.julia/artifacts/72d4adc3ef9236a92f4fefeb0291cb6e8aaae2d7/lib/libmkl_intel_ilp64.so.1 (unknown line)
DOUBLE_inv at /home/jmlim/appl/miniconda3/lib/python3.8/site-packages/numpy/linalg/_umath_linalg.cpython-38-x86_64-linux-gnu.so (unknown line)
PyUFunc_GenericFunction_int at /home/jmlim/appl/miniconda3/lib/python3.8/site-packages/numpy/core/_multiarray_umath.cpython-38-x86_64-linux-gnu.so (unknown line)
ufunc_generic_call at /home/jmlim/appl/miniconda3/lib/python3.8/site-packages/numpy/core/_multiarray_umath.cpython-38-x86_64-linux-gnu.so (unknown line)
_PyObject_MakeTpCall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
call_function at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
function_code_fastcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
PyVectorcall_Call at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
array_implement_array_function at /home/jmlim/appl/miniconda3/lib/python3.8/site-packages/numpy/core/_multiarray_umath.cpython-38-x86_64-linux-gnu.so (unknown line)
PyCFunction_Call at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyObject_MakeTpCall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
call_function at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalCodeWithName at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyFunction_Vectorcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
call_function at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
function_code_fastcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
call_function at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalCodeWithName at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyFunction_Vectorcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
call_function at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalCodeWithName at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyFunction_Vectorcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
call_function at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalCodeWithName at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyFunction_Vectorcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
PyVectorcall_Call at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalCodeWithName at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyFunction_Vectorcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
method_vectorcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
call_function at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
function_code_fastcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
PyVectorcall_Call at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalFrameDefault at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyEval_EvalCodeWithName at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
_PyFunction_Vectorcall at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
PyVectorcall_Call at /home/jmlim/appl/miniconda3/lib/libpython3.8.so.1.0 (unknown line)
macro expansion at /home/jmlim/.julia/packages/PyCall/3fwVL/src/exception.jl:95 [inlined]
#107 at /home/jmlim/.julia/packages/PyCall/3fwVL/src/pyfncall.jl:43 [inlined]
disable_sigint at ./c.jl:458 [inlined]
__pycall! at /home/jmlim/.julia/packages/PyCall/3fwVL/src/pyfncall.jl:42 [inlined]
_pycall! at /home/jmlim/.julia/packages/PyCall/3fwVL/src/pyfncall.jl:29
_pycall! at /home/jmlim/.julia/packages/PyCall/3fwVL/src/pyfncall.jl:11 [inlined]
#pycall#113 at /home/jmlim/.julia/packages/PyCall/3fwVL/src/pyfncall.jl:83 [inlined]
pycall at /home/jmlim/.julia/packages/PyCall/3fwVL/src/pyfncall.jl:83 [inlined]
#tight_layout#125 at /home/jmlim/.julia/packages/PyPlot/XaELc/src/PyPlot.jl:177
tight_layout at /home/jmlim/.julia/packages/PyPlot/XaELc/src/PyPlot.jl:174
unknown function (ip: 0x2b0f574bbebf)
_jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2247 [inlined]
jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2429
jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1788 [inlined]
do_call at /buildworker/worker/package_linux64/build/src/interpreter.c:126
eval_value at /buildworker/worker/package_linux64/build/src/interpreter.c:215
eval_stmt_value at /buildworker/worker/package_linux64/build/src/interpreter.c:166 [inlined]
eval_body at /buildworker/worker/package_linux64/build/src/interpreter.c:587
jl_interpret_toplevel_thunk at /buildworker/worker/package_linux64/build/src/interpreter.c:731
jl_toplevel_eval_flex at /buildworker/worker/package_linux64/build/src/toplevel.c:885
jl_toplevel_eval_flex at /buildworker/worker/package_linux64/build/src/toplevel.c:830
jl_toplevel_eval_in at /buildworker/worker/package_linux64/build/src/toplevel.c:944
eval at ./boot.jl:373 [inlined]
eval_user_input at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.7/REPL/src/REPL.jl:150
repl_backend_loop at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.7/REPL/src/REPL.jl:244
start_repl_backend at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.7/REPL/src/REPL.jl:229
#run_repl#47 at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.7/REPL/src/REPL.jl:362
run_repl at /buildworker/worker/package_linux64/build/usr/share/julia/stdlib/v1.7/REPL/src/REPL.jl:349
_jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2247 [inlined]
jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2429
#930 at ./client.jl:394
jfptr_YY.930_32578.clone_1 at /home/jmlim/appl/julia-1.7.0/lib/julia/sys.so (unknown line)
_jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2247 [inlined]
jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2429
jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1788 [inlined]
jl_f__call_latest at /buildworker/worker/package_linux64/build/src/builtins.c:757
#invokelatest#2 at ./essentials.jl:716 [inlined]
invokelatest at ./essentials.jl:714 [inlined]
run_main_repl at ./client.jl:379
exec_options at ./client.jl:309
_start at ./client.jl:495
jfptr__start_43127.clone_1 at /home/jmlim/appl/julia-1.7.0/lib/julia/sys.so (unknown line)
_jl_invoke at /buildworker/worker/package_linux64/build/src/gf.c:2247 [inlined]
jl_apply_generic at /buildworker/worker/package_linux64/build/src/gf.c:2429
jl_apply at /buildworker/worker/package_linux64/build/src/julia.h:1788 [inlined]
true_main at /buildworker/worker/package_linux64/build/src/jlapi.c:559
jl_repl_entrypoint at /buildworker/worker/package_linux64/build/src/jlapi.c:701
main at julia (unknown line)
__libc_start_main at /lib64/libc.so.6 (unknown line)
unknown function (ip: 0x400808)
Allocations: 7008166 (Pool: 7005366; Big: 2800); GC: 8
Segmentation fault (core dumped)

But this doesn't (MKL is used after PyPlot):

using PyPlot
using MKL
fig = PyPlot.figure()
plot(rand(10), rand(10))
tight_layout()

julia> versioninfo()
Julia Version 1.7.0
Commit 3bf9d17731 (2021-11-30 12:12 UTC)
Platform Info:
  OS: Linux (x86_64-pc-linux-gnu)
  CPU: Intel(R) Xeon(R) CPU E5-2630 v3 @ 2.40GHz
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-12.0.1 (ORCJIT, haswell)
Environment:
  JULIA = /home/jmlim/appl/julia-1.7.0/bin/julia

(@v1.7) pkg> status MKL
      Status `~/.julia/environments/v1.7/Project.toml`
  [33e6dc65] MKL v0.4.2

(@v1.7) pkg> status PyPlot
      Status `~/.julia/environments/v1.7/Project.toml`
  [d330b81b] PyPlot v2.10.0

(@v1.7) pkg> status PyCall
      Status `~/.julia/environments/v1.7/Project.toml`
  [438e738f] PyCall v1.92.5