[distributed] ValueError: Cannot use ReduceOp.PREMUL_SUM with XCCL #1571
Description
🐛 Describe the bug
PYTORCH_TEST_WITH_SLOW=1 python test/distributed/_composable/fsdp/test_fully_shard_comm.py TestFullyShardCommunication.test_set_reduce_scatter_divide_factor
RuntimeError: Process 0 exited with error code 10 and exception:
Traceback (most recent call last):
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/testing/_internal/common_distributed.py", line 761, in run_test
getattr(self, test_name)()
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/testing/_internal/common_distributed.py", line 634, in wrapper
fn()
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 3157, in wrapper
method(*args, **kwargs)
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/testing/_internal/common_distributed.py", line 209, in wrapper
return func(*args, **kwargs)
File "/home/sdp/daisyden/pytorch/test/distributed/_composable/fsdp/test_fully_shard_comm.py", line 390, in test_set_reduce_scatter_divide_factor
self.run_subtests(
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/testing/_internal/common_fsdp.py", line 1188, in run_subtests
return run_subtests(self, *args, **kwargs)
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/testing/_internal/common_distributed.py", line 1030, in run_subtests
test_fn(*test_args, **test_kwargs, **subtest_kwargs)
File "/home/sdp/daisyden/pytorch/test/distributed/_composable/fsdp/test_fully_shard_comm.py", line 418, in _test_set_reduce_scatter_divide_factor
loss.backward()
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/_tensor.py", line 648, in backward
torch.autograd.backward(
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/autograd/init.py", line 353, in backward
_engine_run_backward(
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/autograd/graph.py", line 824, in _engine_run_backward
return Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/distributed/fsdp/_fully_shard/_fsdp_state.py", line 291, in _root_post_backward_final_callback
fsdp_param_group.post_backward()
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/distributed/fsdp/_fully_shard/_fsdp_param_group.py", line 445, in post_backward
) = foreach_reduce(
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/distributed/fsdp/_fully_shard/_fsdp_collectives.py", line 420, in foreach_reduce
dist.reduce_scatter_tensor(
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/distributed/c10d_logger.py", line 81, in wrapper
return func(*args, **kwargs)
File "/home/sdp/.conda/envs/daisy_weekly/lib/python3.10/site-packages/torch/distributed/distributed_c10d.py", line 4382, in reduce_scatter_tensor
work = group._reduce_scatter_base(output, input, opts)
ValueError: Cannot use ReduceOp.PREMUL_SUM with XCCL
Versions
oneccl: 2021.15
oneapi 2025.1
GPU 0/0 GPU 1/0 GPU 2/0 GPU 3/0 GPU 4/0 GPU 5/0 GPU 6/0 GPU 7/0 CPU Affinity
GPU 0/0 S XL8 XL8 XL8 SYS SYS SYS SYS 0-47,96-143
GPU 1/0 XL8 S XL8 XL8 SYS SYS SYS SYS 0-47,96-143
GPU 2/0 XL8 XL8 S XL8 SYS SYS SYS SYS 0-47,96-143
GPU 3/0 XL8 XL8 XL8 S SYS SYS SYS SYS 0-47,96-143
GPU 4/0 SYS SYS SYS SYS S XL8 XL8 XL8 48-95,144-191
GPU 5/0 SYS SYS SYS SYS XL8 S XL8 XL8 48-95,144-191
GPU 6/0 SYS SYS SYS SYS XL8 XL8 S XL8 48-95,144-191
GPU 7/0 SYS SYS SYS SYS XL8 XL8 XL8 S 48-95,144-191
ZE_AFFINITY_MASK=0,1,2,3
Collecting environment information...
PyTorch version: 2.8.0a0+git624be3a
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A
OS: Ubuntu 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 4.0.0
Libc version: glibc-2.35
Python version: 3.10.16 | packaged by conda-forge | (main, Apr 8 2025, 20:53:32) [GCC 13.3.0] (64-bit runtime)
Python platform: Linux-5.15.0-131-generic-x86_64-with-glibc2.35
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
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: 46 bits physical, 57 bits virtual
Byte Order: Little Endian
CPU(s): 192
On-line CPU(s) list: 0-191
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) Platinum 8468V
CPU family: 6
Model: 143
Thread(s) per core: 2
Core(s) per socket: 48
Socket(s): 2
Stepping: 8
Frequency boost: enabled
CPU max MHz: 2401.0000
CPU min MHz: 800.0000
BogoMIPS: 4800.00
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cat_l2 cdp_l3 invpcid_single intel_ppin cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect avx_vnni avx512_bf16 wbnoinvd dtherm ida arat pln pts avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid bus_lock_detect cldemote movdiri movdir64b enqcmd fsrm md_clear serialize tsxldtrk pconfig arch_lbr amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities
Virtualization: VT-x
L1d cache: 4.5 MiB (96 instances)
L1i cache: 3 MiB (96 instances)
L2 cache: 192 MiB (96 instances)
L3 cache: 195 MiB (2 instances)
NUMA node(s): 2
NUMA node0 CPU(s): 0-47,96-143
NUMA node1 CPU(s): 48-95,144-191
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 Reg file data sampling: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
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; Enhanced / Automatic IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
Versions of relevant libraries:
[pip3] mypy==1.14.0
[pip3] mypy-extensions==1.0.0
[pip3] numpy==1.22.4
[pip3] onnx==1.17.0
[pip3] onnxscript==0.2.2
[pip3] optree==0.13.0
[pip3] torch==2.8.0a0+git624be3a
[conda] mkl-include 2025.1.0 pypi_0 pypi
[conda] mkl-static 2025.1.0 pypi_0 pypi
[conda] numpy 1.22.4 pypi_0 pypi
[conda] optree 0.13.0 pypi_0 pypi
[conda] torch 2.8.0a0+git624be3a pypi_0 pypi