The wall of warning text from process group in 2.4 is too much

[awaelchli]

🐛 Describe the bug

In PyTorch 2.4 nightly, users get this wall of warning text, sometimes multiple messages per GPU process (e.g. 16 messages on a 8xGPU machine):

[rank0]:[W515 07:06:42.829549331 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank0]:[W515 07:06:42.830097390 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank5]:[W515 07:06:42.894453420 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank5]:[W515 07:06:42.895003560 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank1]:[W515 07:06:42.907063571 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank1]:[W515 07:06:42.907682540 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank4]:[W515 07:06:42.909104688 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank4]:[W515 07:06:42.909512137 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank2]:[W515 07:06:42.912527702 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank2]:[W515 07:06:42.912946502 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank3]:[W515 07:06:42.915317848 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank3]:[W515 07:06:42.915740178 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank6]:[W515 07:06:42.932886561 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank6]:[W515 07:06:42.933299590 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank7]:[W515 07:06:42.942563506 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4
[rank7]:[W515 07:06:42.942917845 ProcessGroupNCCL.cpp:1117] WARNING: process group has NOT been destroyed before it is being destructed. On normal program exit, the application should call destroy_process_group to ensure that any pending NCCL data transfers have finished in this process. In rare cases this process can exit before this point and block the progress of another member of the process group. This constraint has always been present,  but this warning has only been added since PyTorch 2.4

This gets printed regardless of whether the program exited normally or not. Are users now expected to wrap their main program into a try-except to destroy the process group?

import os

import torch
import torch.distributed
import torch.multiprocessing as mp


def run(rank):
    torch.distributed.init_process_group(
        "nccl", init_method="tcp://127.0.0.1:50001", rank=rank, world_size=4
    )
    torch.cuda.set_device(rank)
    tensor = torch.tensor([rank], device="cuda")
    torch.distributed.broadcast(tensor, src=0)
    
    raise Exception("Exception happened somewhere in user code!")
    

if __name__ == "__main__":
    mp.spawn(run, nprocs=4)

Versions

Collecting environment information...
PyTorch version: 2.4.0.dev20240513+cu121
Is debug build: False
CUDA used to build PyTorch: 12.1
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.3 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.26.0
Libc version: glibc-2.35

Python version: 3.10.9 (main, Jan 11 2023, 15:21:40) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-94-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration:
GPU 0: NVIDIA A100-SXM4-40GB
GPU 1: NVIDIA A100-SXM4-40GB
GPU 2: NVIDIA A100-SXM4-40GB
GPU 3: NVIDIA A100-SXM4-40GB
GPU 4: NVIDIA A100-SXM4-40GB
GPU 5: NVIDIA A100-SXM4-40GB
GPU 6: NVIDIA A100-SXM4-40GB
GPU 7: NVIDIA A100-SXM4-40GB

Nvidia driver version: 545.23.08
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.0.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.7
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.7
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 43 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 256
On-line CPU(s) list: 0-254
Off-line CPU(s) list: 255
Vendor ID: AuthenticAMD
Model name: AMD EPYC 7742 64-Core Processor
CPU family: 23
Model: 49
Thread(s) per core: 2
Core(s) per socket: 64
Socket(s): 2
Stepping: 0
Frequency boost: enabled
CPU max MHz: 2250.0000
CPU min MHz: 0.0000
BogoMIPS: 4500.20
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca sme sev sev_es
Virtualization: AMD-V
L1d cache: 4 MiB (128 instances)
L1i cache: 4 MiB (128 instances)
L2 cache: 64 MiB (128 instances)
L3 cache: 512 MiB (32 instances)
NUMA node(s): 2
NUMA node0 CPU(s): 0-63,128-191
NUMA node1 CPU(s): 64-127,192-254
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Retbleed: Mitigation; untrained return thunk; SMT enabled with STIBP protection
Vulnerability Spec rstack overflow: Mitigation; safe RET
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Retpolines, IBPB conditional, STIBP always-on, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected

Versions of relevant libraries:
[pip3] mypy==1.4.1
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.24.2
[pip3] onnx==1.12.0
[pip3] onnxruntime==1.14.1
[pip3] pytorch-lightning==2.0.6
[pip3] pytorch-triton==3.0.0+45fff310c8
[pip3] torch==2.4.0.dev20240513+cu121
[pip3] torch-tb-profiler==0.4.3
[pip3] torchmetrics==1.3.0.post0
[pip3] triton==2.3.0
[conda] numpy 1.24.2 pypi_0 pypi
[conda] pytorch-lightning 2.0.6 pypi_0 pypi
[conda] pytorch-triton 3.0.0+45fff310c8 pypi_0 pypi
[conda] torch 2.4.0.dev20240513+cu121 pypi_0 pypi
[conda] torch-tb-profiler 0.4.3 pypi_0 pypi
[conda] torchmetrics 1.3.0.post0 pypi_0 pypi
[conda] triton 2.3.0 pypi_0 pypi

cc @mrshenli @pritamdamania87 @zhaojuanmao @satgera @gqchen @aazzolini @osalpekar @jiayisuse @H-Huang @kwen2501 @awgu @penguinwu @fegin @XilunWu @wanchaol @fduwjj @wz337 @tianyu-l @wconstab @yf225 @chauhang @d4l3k