Use a leaky singleton for CublasHandlePool. #60987

pritamdamania87 · 2021-06-29T20:39:53Z

Stack from ghstack:

Use a leaky singleton for CublasHandlePool. #60987

We were seeing deadlocks as follows during shutdown:

Thread 1 (LWP 2432101):
#0  0x00007efca470190b in __pause_nocancel () from /lib64/libc.so.6
#1  0x00007efca49de485 in __pthread_mutex_lock_full () from /lib64/libpthread.so.0
#2  0x00007ef91d4c42c6 in __cuda_CallJitEntryPoint () from /lib64/libnvidia-ptxjitcompiler.so.1
#3  0x00007efc651ac8f1 in ?? () from /lib64/libcuda.so
#4  0x00007efc651aee03 in ?? () from /lib64/libcuda.so
#5  0x00007efc64f76b84 in ?? () from /lib64/libcuda.so
#6  0x00007efc64f77f5d in ?? () from /lib64/libcuda.so
#7  0x00007efc64eac858 in ?? () from /lib64/libcuda.so
#8  0x00007efc64eacfbc in ?? () from /lib64/libcuda.so
#9  0x00007efc7810a924 in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#10 0x00007efc780fa2be in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#11 0x00007efc78111044 in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#12 0x00007efc7811580a in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#13 0x00007efc78115aa4 in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#14 0x00007efc781079ec in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#15 0x00007efc780e6a7a in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#16 0x00007efc7811cfa5 in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#17 0x00007efc777ea98c in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#18 0x00007efc777ebd80 in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#19 0x00007efc777ea2c9 in ?? () from /usr/local/cuda/lib64/libcublas.so.11
#20 0x00007efc778c2e2d in cublasDestroy_v2 () from /usr/local/cuda/lib64/libcublas.so.11
#21 0x00007efc51a3fb56 in std::_Sp_counted_ptr_inplace<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle>, std::allocator<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >, (__gnu_cxx::_Lock_policy)2>::_M_dispose() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so
#22 0x00007efc51a3fc5f in std::shared_ptr<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >::~shared_ptr() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so
#23 0x00007efca4648b0c in __run_exit_handlers () from /lib64/libc.so.6
#24 0x00007efca4648c40 in exit () from /lib64/libc.so.6
#25 0x0000558c8852e5f9 in Py_Exit (sts=0) at /tmp/build/80754af9/python_1614362349910/work/Python/pylifecycle.c:2292
#26 0x0000558c8852e6a7 in handle_system_exit () at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:636
#27 0x0000558c8852e742 in PyErr_PrintEx (set_sys_last_vars=<optimized out>, set_sys_last_vars=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:646
#28 0x0000558c88540dd6 in PyRun_SimpleStringFlags (command=0x7efca4dc9050 "from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=9, pipe_handle=13)\n", flags=0x7ffe3a986110) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:457
#29 0x0000558c88540ead in pymain_run_command (cf=0x7ffe3a986110, command=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:420
#30 pymain_run_python (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:2907
#31 pymain_main (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3460
#32 0x0000558c8854122c in _Py_UnixMain (argc=<optimized out>, argv=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3495
#33 0x00007efca4632493 in __libc_start_main () from /lib64/libc.so.6
#34 0x0000558c884e5e90 in _start () at ../sysdeps/x86_64/elf/start.S:103

This was likely caused due to a static singleton that wasn't leaky. Following
the guidance in https://isocpp.org/wiki/faq/ctors#construct-on-first-use-v2 to
use a leaky singleton instead.

Differential Revision: D29468866

We were seeing deadlocks as follows during shutdown: ``` Thread 1 (LWP 2432101): #0 0x00007efca470190b in __pause_nocancel () from /lib64/libc.so.6 #1 0x00007efca49de485 in __pthread_mutex_lock_full () from /lib64/libpthread.so.0 #2 0x00007ef91d4c42c6 in __cuda_CallJitEntryPoint () from /lib64/libnvidia-ptxjitcompiler.so.1 #3 0x00007efc651ac8f1 in ?? () from /lib64/libcuda.so #4 0x00007efc651aee03 in ?? () from /lib64/libcuda.so #5 0x00007efc64f76b84 in ?? () from /lib64/libcuda.so #6 0x00007efc64f77f5d in ?? () from /lib64/libcuda.so #7 0x00007efc64eac858 in ?? () from /lib64/libcuda.so #8 0x00007efc64eacfbc in ?? () from /lib64/libcuda.so #9 0x00007efc7810a924 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #10 0x00007efc780fa2be in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #11 0x00007efc78111044 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #12 0x00007efc7811580a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #13 0x00007efc78115aa4 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #14 0x00007efc781079ec in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #15 0x00007efc780e6a7a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #16 0x00007efc7811cfa5 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #17 0x00007efc777ea98c in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #18 0x00007efc777ebd80 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #19 0x00007efc777ea2c9 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #20 0x00007efc778c2e2d in cublasDestroy_v2 () from /usr/local/cuda/lib64/libcublas.so.11 #21 0x00007efc51a3fb56 in std::_Sp_counted_ptr_inplace<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle>, std::allocator<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >, (__gnu_cxx::_Lock_policy)2>::_M_dispose() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #22 0x00007efc51a3fc5f in std::shared_ptr<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >::~shared_ptr() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #23 0x00007efca4648b0c in __run_exit_handlers () from /lib64/libc.so.6 #24 0x00007efca4648c40 in exit () from /lib64/libc.so.6 #25 0x0000558c8852e5f9 in Py_Exit (sts=0) at /tmp/build/80754af9/python_1614362349910/work/Python/pylifecycle.c:2292 #26 0x0000558c8852e6a7 in handle_system_exit () at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:636 #27 0x0000558c8852e742 in PyErr_PrintEx (set_sys_last_vars=<optimized out>, set_sys_last_vars=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:646 #28 0x0000558c88540dd6 in PyRun_SimpleStringFlags (command=0x7efca4dc9050 "from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=9, pipe_handle=13)\n", flags=0x7ffe3a986110) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:457 #29 0x0000558c88540ead in pymain_run_command (cf=0x7ffe3a986110, command=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:420 #30 pymain_run_python (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:2907 #31 pymain_main (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3460 #32 0x0000558c8854122c in _Py_UnixMain (argc=<optimized out>, argv=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3495 #33 0x00007efca4632493 in __libc_start_main () from /lib64/libc.so.6 #34 0x0000558c884e5e90 in _start () at ../sysdeps/x86_64/elf/start.S:103 ``` This was likely caused due to a static singleton that wasn't leaky. Following the guidance in https://isocpp.org/wiki/faq/ctors#construct-on-first-use-v2 to use a leaky singleton instead. Differential Revision: [D29468866](https://our.internmc.facebook.com/intern/diff/D29468866/) [ghstack-poisoned]

facebook-github-bot · 2021-06-29T20:39:57Z

💊 CI failures summary and remediations

As of commit f729894 (more details on the Dr. CI page and at hud.pytorch.org/pr/60987):

1/1 failures possibly* introduced in this PR
- 1/1 non-scanned failure(s)

Preview docs built from this PR

This comment was automatically generated by Dr. CI (expand for details).

Follow this link to opt-out of these comments for your Pull Requests.

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

We were seeing deadlocks as follows during shutdown: ``` Thread 1 (LWP 2432101): #0 0x00007efca470190b in __pause_nocancel () from /lib64/libc.so.6 #1 0x00007efca49de485 in __pthread_mutex_lock_full () from /lib64/libpthread.so.0 #2 0x00007ef91d4c42c6 in __cuda_CallJitEntryPoint () from /lib64/libnvidia-ptxjitcompiler.so.1 #3 0x00007efc651ac8f1 in ?? () from /lib64/libcuda.so #4 0x00007efc651aee03 in ?? () from /lib64/libcuda.so #5 0x00007efc64f76b84 in ?? () from /lib64/libcuda.so #6 0x00007efc64f77f5d in ?? () from /lib64/libcuda.so #7 0x00007efc64eac858 in ?? () from /lib64/libcuda.so #8 0x00007efc64eacfbc in ?? () from /lib64/libcuda.so #9 0x00007efc7810a924 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #10 0x00007efc780fa2be in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #11 0x00007efc78111044 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #12 0x00007efc7811580a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #13 0x00007efc78115aa4 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #14 0x00007efc781079ec in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #15 0x00007efc780e6a7a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #16 0x00007efc7811cfa5 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #17 0x00007efc777ea98c in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #18 0x00007efc777ebd80 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #19 0x00007efc777ea2c9 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #20 0x00007efc778c2e2d in cublasDestroy_v2 () from /usr/local/cuda/lib64/libcublas.so.11 #21 0x00007efc51a3fb56 in std::_Sp_counted_ptr_inplace<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle>, std::allocator<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >, (__gnu_cxx::_Lock_policy)2>::_M_dispose() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #22 0x00007efc51a3fc5f in std::shared_ptr<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >::~shared_ptr() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #23 0x00007efca4648b0c in __run_exit_handlers () from /lib64/libc.so.6 #24 0x00007efca4648c40 in exit () from /lib64/libc.so.6 #25 0x0000558c8852e5f9 in Py_Exit (sts=0) at /tmp/build/80754af9/python_1614362349910/work/Python/pylifecycle.c:2292 #26 0x0000558c8852e6a7 in handle_system_exit () at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:636 #27 0x0000558c8852e742 in PyErr_PrintEx (set_sys_last_vars=<optimized out>, set_sys_last_vars=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:646 #28 0x0000558c88540dd6 in PyRun_SimpleStringFlags (command=0x7efca4dc9050 "from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=9, pipe_handle=13)\n", flags=0x7ffe3a986110) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:457 #29 0x0000558c88540ead in pymain_run_command (cf=0x7ffe3a986110, command=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:420 #30 pymain_run_python (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:2907 #31 pymain_main (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3460 #32 0x0000558c8854122c in _Py_UnixMain (argc=<optimized out>, argv=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3495 #33 0x00007efca4632493 in __libc_start_main () from /lib64/libc.so.6 #34 0x0000558c884e5e90 in _start () at ../sysdeps/x86_64/elf/start.S:103 ``` This was likely caused due to a static singleton that wasn't leaky. Following the guidance in https://isocpp.org/wiki/faq/ctors#construct-on-first-use-v2 to use a leaky singleton instead. Differential Revision: [D29468866](https://our.internmc.facebook.com/intern/diff/D29468866/) ghstack-source-id: 132706558 Pull Request resolved: #60987

malfet

What's the benefit of leaky shared_ptr vs good-old raw pointer?

malfet · 2021-06-29T21:57:32Z

aten/src/ATen/cuda/CublasHandlePool.cpp

+// Used for following the leaky singleton pattern for shared_ptrs:
+// https://isocpp.org/wiki/faq/ctors#construct-on-first-use-v2
+struct LeakyDeleter {
+    void operator()(CuBlasPoolType* p) const {
+      // Leak the memory.
+    }
+};
+


Can you please turn it a generic template(called something like make_leaky_signleton) and move it to some non-cuda specific header?

aten/src/ATen/cuda/CublasHandlePool.cpp

codecov · 2021-06-30T00:12:12Z

Codecov Report

Merging #60987 (f729894) into gh/pritamdamania87/245/base (5b6818f) will increase coverage by 0.09%.
The diff coverage is n/a.

@@                       Coverage Diff                       @@
##           gh/pritamdamania87/245/base   #60987      +/-   ##
===============================================================
+ Coverage                        76.12%   76.22%   +0.09%     
===============================================================
  Files                             2062     2062              
  Lines                           205577   205577              
===============================================================
+ Hits                            156501   156694     +193     
+ Misses                           49076    48883     -193

ezyang · 2021-06-30T04:32:33Z

absconding responsibility to malfet

pritamdamania87 · 2021-06-30T16:59:17Z

What's the benefit of leaky shared_ptr vs good-old raw pointer?

The reason to have a leaky shared_ptr instead of raw pointer is that this object is used in APIs that expect a shared_ptr.

aten/src/ATen/cuda/CublasHandlePool.cpp

We were seeing deadlocks as follows during shutdown: ``` Thread 1 (LWP 2432101): #0 0x00007efca470190b in __pause_nocancel () from /lib64/libc.so.6 #1 0x00007efca49de485 in __pthread_mutex_lock_full () from /lib64/libpthread.so.0 #2 0x00007ef91d4c42c6 in __cuda_CallJitEntryPoint () from /lib64/libnvidia-ptxjitcompiler.so.1 #3 0x00007efc651ac8f1 in ?? () from /lib64/libcuda.so #4 0x00007efc651aee03 in ?? () from /lib64/libcuda.so #5 0x00007efc64f76b84 in ?? () from /lib64/libcuda.so #6 0x00007efc64f77f5d in ?? () from /lib64/libcuda.so #7 0x00007efc64eac858 in ?? () from /lib64/libcuda.so #8 0x00007efc64eacfbc in ?? () from /lib64/libcuda.so #9 0x00007efc7810a924 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #10 0x00007efc780fa2be in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #11 0x00007efc78111044 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #12 0x00007efc7811580a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #13 0x00007efc78115aa4 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #14 0x00007efc781079ec in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #15 0x00007efc780e6a7a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #16 0x00007efc7811cfa5 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #17 0x00007efc777ea98c in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #18 0x00007efc777ebd80 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #19 0x00007efc777ea2c9 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #20 0x00007efc778c2e2d in cublasDestroy_v2 () from /usr/local/cuda/lib64/libcublas.so.11 #21 0x00007efc51a3fb56 in std::_Sp_counted_ptr_inplace<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle>, std::allocator<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >, (__gnu_cxx::_Lock_policy)2>::_M_dispose() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #22 0x00007efc51a3fc5f in std::shared_ptr<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >::~shared_ptr() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #23 0x00007efca4648b0c in __run_exit_handlers () from /lib64/libc.so.6 #24 0x00007efca4648c40 in exit () from /lib64/libc.so.6 #25 0x0000558c8852e5f9 in Py_Exit (sts=0) at /tmp/build/80754af9/python_1614362349910/work/Python/pylifecycle.c:2292 #26 0x0000558c8852e6a7 in handle_system_exit () at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:636 #27 0x0000558c8852e742 in PyErr_PrintEx (set_sys_last_vars=<optimized out>, set_sys_last_vars=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:646 #28 0x0000558c88540dd6 in PyRun_SimpleStringFlags (command=0x7efca4dc9050 "from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=9, pipe_handle=13)\n", flags=0x7ffe3a986110) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:457 #29 0x0000558c88540ead in pymain_run_command (cf=0x7ffe3a986110, command=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:420 #30 pymain_run_python (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:2907 #31 pymain_main (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3460 #32 0x0000558c8854122c in _Py_UnixMain (argc=<optimized out>, argv=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3495 #33 0x00007efca4632493 in __libc_start_main () from /lib64/libc.so.6 #34 0x0000558c884e5e90 in _start () at ../sysdeps/x86_64/elf/start.S:103 ``` This was likely caused due to a static singleton that wasn't leaky. Following the guidance in https://isocpp.org/wiki/faq/ctors#construct-on-first-use-v2 to use a leaky singleton instead. Differential Revision: [D29468866](https://our.internmc.facebook.com/intern/diff/D29468866/) [ghstack-poisoned]

Pull Request resolved: #60987 We were seeing deadlocks as follows during shutdown: ``` Thread 1 (LWP 2432101): #0 0x00007efca470190b in __pause_nocancel () from /lib64/libc.so.6 #1 0x00007efca49de485 in __pthread_mutex_lock_full () from /lib64/libpthread.so.0 #2 0x00007ef91d4c42c6 in __cuda_CallJitEntryPoint () from /lib64/libnvidia-ptxjitcompiler.so.1 #3 0x00007efc651ac8f1 in ?? () from /lib64/libcuda.so #4 0x00007efc651aee03 in ?? () from /lib64/libcuda.so #5 0x00007efc64f76b84 in ?? () from /lib64/libcuda.so #6 0x00007efc64f77f5d in ?? () from /lib64/libcuda.so #7 0x00007efc64eac858 in ?? () from /lib64/libcuda.so #8 0x00007efc64eacfbc in ?? () from /lib64/libcuda.so #9 0x00007efc7810a924 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #10 0x00007efc780fa2be in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #11 0x00007efc78111044 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #12 0x00007efc7811580a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #13 0x00007efc78115aa4 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #14 0x00007efc781079ec in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #15 0x00007efc780e6a7a in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #16 0x00007efc7811cfa5 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #17 0x00007efc777ea98c in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #18 0x00007efc777ebd80 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #19 0x00007efc777ea2c9 in ?? () from /usr/local/cuda/lib64/libcublas.so.11 #20 0x00007efc778c2e2d in cublasDestroy_v2 () from /usr/local/cuda/lib64/libcublas.so.11 #21 0x00007efc51a3fb56 in std::_Sp_counted_ptr_inplace<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle>, std::allocator<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >, (__gnu_cxx::_Lock_policy)2>::_M_dispose() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #22 0x00007efc51a3fc5f in std::shared_ptr<at::cuda::(anonymous namespace)::DeviceThreadHandlePool<cublasContext*, &at::cuda::(anonymous namespace)::createCublasHandle, &at::cuda::(anonymous namespace)::destroyCublasHandle> >::~shared_ptr() () from /data/users/pritam/pytorch/torch/lib/libtorch_cuda.so #23 0x00007efca4648b0c in __run_exit_handlers () from /lib64/libc.so.6 #24 0x00007efca4648c40 in exit () from /lib64/libc.so.6 #25 0x0000558c8852e5f9 in Py_Exit (sts=0) at /tmp/build/80754af9/python_1614362349910/work/Python/pylifecycle.c:2292 #26 0x0000558c8852e6a7 in handle_system_exit () at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:636 #27 0x0000558c8852e742 in PyErr_PrintEx (set_sys_last_vars=<optimized out>, set_sys_last_vars=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:646 #28 0x0000558c88540dd6 in PyRun_SimpleStringFlags (command=0x7efca4dc9050 "from multiprocessing.spawn import spawn_main; spawn_main(tracker_fd=9, pipe_handle=13)\n", flags=0x7ffe3a986110) at /tmp/build/80754af9/python_1614362349910/work/Python/pythonrun.c:457 #29 0x0000558c88540ead in pymain_run_command (cf=0x7ffe3a986110, command=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:420 #30 pymain_run_python (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:2907 #31 pymain_main (pymain=0x7ffe3a986220) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3460 #32 0x0000558c8854122c in _Py_UnixMain (argc=<optimized out>, argv=<optimized out>) at /tmp/build/80754af9/python_1614362349910/work/Modules/main.c:3495 #33 0x00007efca4632493 in __libc_start_main () from /lib64/libc.so.6 #34 0x0000558c884e5e90 in _start () at ../sysdeps/x86_64/elf/start.S:103 ``` This was likely caused due to a static singleton that wasn't leaky. Following the guidance in https://isocpp.org/wiki/faq/ctors#construct-on-first-use-v2 to use a leaky singleton instead. ghstack-source-id: 132847448 Differential Revision: [D29468866](https://our.internmc.facebook.com/intern/diff/D29468866/)

facebook-github-bot · 2021-07-01T17:24:47Z

This pull request has been merged in 6f92f10.

mruberry · 2021-07-02T00:04:38Z

fyi @mcarilli @ngimel @zasdfgbnm

ngimel · 2021-07-02T01:32:33Z

It would make sense to preemptively make all pools (cudnn, cusparse, cusolver) leaky.

facebook-github-bot added the cla signed label Jun 29, 2021

pritamdamania87 requested a review from ezyang June 29, 2021 20:40

malfet reviewed Jun 29, 2021

View reviewed changes

ezyang removed their request for review June 30, 2021 04:32

malfet reviewed Jun 30, 2021

View reviewed changes

aten/src/ATen/cuda/CublasHandlePool.cpp Outdated Show resolved Hide resolved

pritamdamania87 requested a review from malfet June 30, 2021 23:44

malfet approved these changes Jul 1, 2021

View reviewed changes

facebook-github-bot closed this in 6f92f10 Jul 1, 2021

facebook-github-bot added the Merged label Jul 1, 2021

facebook-github-bot deleted the gh/pritamdamania87/245/head branch July 5, 2021 14:16

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Use a leaky singleton for CublasHandlePool. #60987

Use a leaky singleton for CublasHandlePool. #60987

pritamdamania87 commented Jun 29, 2021 •

edited

Loading

Uh oh!

facebook-github-bot commented Jun 29, 2021 •

edited

Loading

Uh oh!

malfet left a comment

Uh oh!

malfet Jun 29, 2021 •

edited

Loading

Uh oh!

Uh oh!

codecov bot commented Jun 30, 2021 •

edited

Loading

Uh oh!

ezyang commented Jun 30, 2021

Uh oh!

pritamdamania87 commented Jun 30, 2021

Uh oh!

Uh oh!

facebook-github-bot commented Jul 1, 2021

Uh oh!

mruberry commented Jul 2, 2021

Uh oh!

ngimel commented Jul 2, 2021

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

Use a leaky singleton for CublasHandlePool. #60987

Use a leaky singleton for CublasHandlePool. #60987

Conversation

pritamdamania87 commented Jun 29, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

facebook-github-bot commented Jun 29, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

💊 CI failures summary and remediations

Uh oh!

malfet left a comment

Choose a reason for hiding this comment

Uh oh!

malfet Jun 29, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

codecov bot commented Jun 30, 2021 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

ezyang commented Jun 30, 2021

Uh oh!

pritamdamania87 commented Jun 30, 2021

Uh oh!

Uh oh!

facebook-github-bot commented Jul 1, 2021

Uh oh!

mruberry commented Jul 2, 2021

Uh oh!

ngimel commented Jul 2, 2021

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

8 participants

pritamdamania87 commented Jun 29, 2021 •

edited

Loading

facebook-github-bot commented Jun 29, 2021 •

edited

Loading

malfet Jun 29, 2021 •

edited

Loading

codecov bot commented Jun 30, 2021 •

edited

Loading