pytorch - 💡(How to fix) Fix torch.compile + view_as_complex: RuntimeError: Tensor must have a last dimension with stride 1 [3 comments, 2 participants]

pytorch2026-04-04 06:12:25

ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

GitHub issue URL

Helpful · Quick feedback

GitHub stats

pytorch/pytorch#179368•Fetched 2026-04-08 02:43:30

View on GitHub

Comments

Participants

Timeline

193

Reactions

Author

ad8e

Participants

ad8e

Arths17

Timeline (top)

mentioned ×91subscribed ×91labeled ×6commented ×3

Error Message

Eager: OK Compiled FAILS: RuntimeError: Tensor must have a last dimension with stride 1

While executing %view_as_complex_2 : [num_users=1] = call_function[target=torch.ops.aten.view_as_complex.default](args = (%view_9,), kwargs = {}) Original traceback: File "/mnt/clusterstorage/workspace/kevin/centrifuge/reproducer_view_as_complex.py", line 46, in <lambda> torch.compile(lambda m, x: m(x).sum())(model, x).backward() File "/mnt/clusterstorage/workspace/kevin/centrifuge/reproducer_view_as_complex.py", line 32, in forward x = torch.view_as_real(x_ * freqs).flatten(-2)

Use tlparse to see full graph. (https://github.com/pytorch/tlparse?tab=readme-ov-file#tlparse-parse-structured-pt2-logs)

Fix Action

Fix / Workaround

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): 128 On-line CPU(s) list: 0-127 Vendor ID: GenuineIntel Model name: Intel(R) Xeon(R) Platinum 8462Y+ CPU family: 6 Model: 143 Thread(s) per core: 2 Core(s) per socket: 32 Socket(s): 2 Stepping: 8 CPU max MHz: 4100.0000 CPU min MHz: 800.0000 BogoMIPS: 5600.00 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 cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow 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 hfi vnmi 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 ibt amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities Virtualization: VT-x L1d cache: 3 MiB (64 instances) L1i cache: 2 MiB (64 instances) L2 cache: 128 MiB (64 instances) L3 cache: 120 MiB (2 instances) NUMA node(s): 2 NUMA node0 CPU(s): 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78,80,82,84,86,88,90,92,94,96,98,100,102,104,106,108,110,112,114,116,118,120,122,124,126 NUMA node1 CPU(s): 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79,81,83,85,87,89,91,93,95,97,99,101,103,105,107,109,111,113,115,117,119,121,123,125,127 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: Not affected Vulnerability Spec rstack overflow: Not affected Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl 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

Code Example

"""torch.compile backward: view_as_complex gets non-unit last-dim stride

When a Conv2d follows SDPA + view_as_real/view_as_complex in the compiled
graph, inductor's layout planner assigns channels-last strides to tensors
saved for the backward. This causes view_as_complex (backward of view_as_real)
to receive a tensor with last-dim stride != 1.

Eager mode works fine. All 5 components are needed to trigger:
  matmul → view_as_complex/view_as_real → .bfloat16() → SDPA → Conv2d

PyTorch 2.10.0
"""
import torch
import torch.nn as nn
import torch.nn.functional as F

B, H, W, D, heads = 1, 32, 32, 128, 2


class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.w = nn.Parameter(torch.randn(D, D))
        self.register_buffer("freqs", torch.view_as_real(
            torch.randn(1, H, W, 1, D // heads // 2, dtype=torch.cfloat)))
        self.conv = nn.Conv2d(D, D, 1, bias=False)

    def forward(self, x):
        x = (x @ self.w).view(B, H, W, heads, D // heads)
        freqs = torch.view_as_complex(self.freqs)
        x_ = torch.view_as_complex(x.reshape(*x.shape[:-1], -1, 2))
        x = torch.view_as_real(x_ * freqs).flatten(-2)
        x = x.reshape(B, -1, heads, D // heads).permute(0, 2, 1, 3).bfloat16()
        x = F.scaled_dot_product_attention(x, x, x)
        x = x.permute(0, 2, 1, 3).reshape(B, H, W, D).float()
        return self.conv(x.permute(0, 3, 1, 2)).permute(0, 2, 3, 1)


model = Model().cuda()
x = torch.randn(B, H, W, D, device="cuda")

model(x).sum().backward()
print("Eager: OK")

try:
    torch.compile(lambda m, x: m(x).sum())(model, x).backward()
    print("Compiled: OK")
except RuntimeError as e:
    print(f"Compiled FAILS: {e}")

---

Eager: OK
Compiled FAILS: RuntimeError: Tensor must have a last dimension with stride 1

While executing %view_as_complex_2 : [num_users=1] = call_function[target=torch.ops.aten.view_as_complex.default](args = (%view_9,), kwargs = {})
Original traceback:
  File "/mnt/clusterstorage/workspace/kevin/centrifuge/reproducer_view_as_complex.py", line 46, in <lambda>
    torch.compile(lambda m, x: m(x).sum())(model, x).backward()
  File "/mnt/clusterstorage/workspace/kevin/centrifuge/reproducer_view_as_complex.py", line 32, in forward
    x = torch.view_as_real(x_ * freqs).flatten(-2)

Use tlparse to see full graph. (https://github.com/pytorch/tlparse?tab=readme-ov-file#tlparse-parse-structured-pt2-logs)

---

PyTorch version: 2.10.0
Is debug build: False
CUDA used to build PyTorch: 13.2
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04.3) 11.4.0
Clang version: Could not collect
CMake version: version 3.22.1
Libc version: glibc-2.35

Python version: 3.10.12 (main, Mar  3 2026, 11:56:32) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-6.5.13-65-650-4141-22041-coreweave-amd64-85c45edc-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 13.2.51
CUDA_MODULE_LOADING set to: 
GPU models and configuration: GPU 0: NVIDIA H100 80GB HBM3
Nvidia driver version: 570.172.08
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.20.0
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):                             128
On-line CPU(s) list:                0-127
Vendor ID:                          GenuineIntel
Model name:                         Intel(R) Xeon(R) Platinum 8462Y+
CPU family:                         6
Model:                              143
Thread(s) per core:                 2
Core(s) per socket:                 32
Socket(s):                          2
Stepping:                           8
CPU max MHz:                        4100.0000
CPU min MHz:                        800.0000
BogoMIPS:                           5600.00
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 cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow 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 hfi vnmi 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 ibt amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities
Virtualization:                     VT-x
L1d cache:                          3 MiB (64 instances)
L1i cache:                          2 MiB (64 instances)
L2 cache:                           128 MiB (64 instances)
L3 cache:                           120 MiB (2 instances)
NUMA node(s):                       2
NUMA node0 CPU(s):                  0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78,80,82,84,86,88,90,92,94,96,98,100,102,104,106,108,110,112,114,116,118,120,122,124,126
NUMA node1 CPU(s):                  1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79,81,83,85,87,89,91,93,95,97,99,101,103,105,107,109,111,113,115,117,119,121,123,125,127
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:             Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass disabled via prctl
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] clip-anytorch==2.6.0
[pip3] dctorch==0.1.2
[pip3] DISTS_pytorch==0.1
[pip3] gpytorch==1.15.2
[pip3] lovely-numpy==0.2.20
[pip3] mypy_extensions==1.1.0
[pip3] NATTEN==0.21.5+torch2100cu130
[pip3] numpy==1.24.4
[pip3] onnx==1.20.1
[pip3] onnx-ir==0.2.0
[pip3] onnxscript==0.3.1
[pip3] torch==2.10.0
[pip3] torchaudio==2.10.0
[pip3] torchdata==0.11.0
[pip3] torchdiffeq==0.2.5
[pip3] torchsde==0.2.6
[pip3] torchtitan==0.1.0
[pip3] torchvision==0.25.0
[pip3] triton==3.6.0+git8fedd49b
[pip3] welford-torch==0.2.5
[conda] Could not collect

RAW_BUFFERClick to expand / collapse

🐛 Describe the bug

Might be the same underlying cause as https://github.com/pytorch/pytorch/issues/176986.

Claude's repro:

"""torch.compile backward: view_as_complex gets non-unit last-dim stride

When a Conv2d follows SDPA + view_as_real/view_as_complex in the compiled
graph, inductor's layout planner assigns channels-last strides to tensors
saved for the backward. This causes view_as_complex (backward of view_as_real)
to receive a tensor with last-dim stride != 1.

Eager mode works fine. All 5 components are needed to trigger:
  matmul → view_as_complex/view_as_real → .bfloat16() → SDPA → Conv2d

PyTorch 2.10.0
"""
import torch
import torch.nn as nn
import torch.nn.functional as F

B, H, W, D, heads = 1, 32, 32, 128, 2


class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.w = nn.Parameter(torch.randn(D, D))
        self.register_buffer("freqs", torch.view_as_real(
            torch.randn(1, H, W, 1, D // heads // 2, dtype=torch.cfloat)))
        self.conv = nn.Conv2d(D, D, 1, bias=False)

    def forward(self, x):
        x = (x @ self.w).view(B, H, W, heads, D // heads)
        freqs = torch.view_as_complex(self.freqs)
        x_ = torch.view_as_complex(x.reshape(*x.shape[:-1], -1, 2))
        x = torch.view_as_real(x_ * freqs).flatten(-2)
        x = x.reshape(B, -1, heads, D // heads).permute(0, 2, 1, 3).bfloat16()
        x = F.scaled_dot_product_attention(x, x, x)
        x = x.permute(0, 2, 1, 3).reshape(B, H, W, D).float()
        return self.conv(x.permute(0, 3, 1, 2)).permute(0, 2, 3, 1)


model = Model().cuda()
x = torch.randn(B, H, W, D, device="cuda")

model(x).sum().backward()
print("Eager: OK")

try:
    torch.compile(lambda m, x: m(x).sum())(model, x).backward()
    print("Compiled: OK")
except RuntimeError as e:
    print(f"Compiled FAILS: {e}")

Error logs

Eager: OK
Compiled FAILS: RuntimeError: Tensor must have a last dimension with stride 1

While executing %view_as_complex_2 : [num_users=1] = call_function[target=torch.ops.aten.view_as_complex.default](args = (%view_9,), kwargs = {})
Original traceback:
  File "/mnt/clusterstorage/workspace/kevin/centrifuge/reproducer_view_as_complex.py", line 46, in <lambda>
    torch.compile(lambda m, x: m(x).sum())(model, x).backward()
  File "/mnt/clusterstorage/workspace/kevin/centrifuge/reproducer_view_as_complex.py", line 32, in forward
    x = torch.view_as_real(x_ * freqs).flatten(-2)

Use tlparse to see full graph. (https://github.com/pytorch/tlparse?tab=readme-ov-file#tlparse-parse-structured-pt2-logs)

Versions

PyTorch version: 2.10.0
Is debug build: False
CUDA used to build PyTorch: 13.2
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.5 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04.3) 11.4.0
Clang version: Could not collect
CMake version: version 3.22.1
Libc version: glibc-2.35

Python version: 3.10.12 (main, Mar  3 2026, 11:56:32) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-6.5.13-65-650-4141-22041-coreweave-amd64-85c45edc-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 13.2.51
CUDA_MODULE_LOADING set to: 
GPU models and configuration: GPU 0: NVIDIA H100 80GB HBM3
Nvidia driver version: 570.172.08
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_adv.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_cnn.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_precompiled.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_engines_runtime_compiled.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_graph.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_heuristic.so.9.20.0
/usr/lib/x86_64-linux-gnu/libcudnn_ops.so.9.20.0
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):                             128
On-line CPU(s) list:                0-127
Vendor ID:                          GenuineIntel
Model name:                         Intel(R) Xeon(R) Platinum 8462Y+
CPU family:                         6
Model:                              143
Thread(s) per core:                 2
Core(s) per socket:                 32
Socket(s):                          2
Stepping:                           8
CPU max MHz:                        4100.0000
CPU min MHz:                        800.0000
BogoMIPS:                           5600.00
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 cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow 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 hfi vnmi 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 ibt amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities
Virtualization:                     VT-x
L1d cache:                          3 MiB (64 instances)
L1i cache:                          2 MiB (64 instances)
L2 cache:                           128 MiB (64 instances)
L3 cache:                           120 MiB (2 instances)
NUMA node(s):                       2
NUMA node0 CPU(s):                  0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78,80,82,84,86,88,90,92,94,96,98,100,102,104,106,108,110,112,114,116,118,120,122,124,126
NUMA node1 CPU(s):                  1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79,81,83,85,87,89,91,93,95,97,99,101,103,105,107,109,111,113,115,117,119,121,123,125,127
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:             Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass:    Mitigation; Speculative Store Bypass disabled via prctl
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] clip-anytorch==2.6.0
[pip3] dctorch==0.1.2
[pip3] DISTS_pytorch==0.1
[pip3] gpytorch==1.15.2
[pip3] lovely-numpy==0.2.20
[pip3] mypy_extensions==1.1.0
[pip3] NATTEN==0.21.5+torch2100cu130
[pip3] numpy==1.24.4
[pip3] onnx==1.20.1
[pip3] onnx-ir==0.2.0
[pip3] onnxscript==0.3.1
[pip3] torch==2.10.0
[pip3] torchaudio==2.10.0
[pip3] torchdata==0.11.0
[pip3] torchdiffeq==0.2.5
[pip3] torchsde==0.2.6
[pip3] torchtitan==0.1.0
[pip3] torchvision==0.25.0
[pip3] triton==3.6.0+git8fedd49b
[pip3] welford-torch==0.2.5
[conda] Could not collect

cc @ezyang @anjali411 @dylanbespalko @mruberry @nikitaved @amjames @jamesr66a @chauhang @penguinwu @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @aakhundov @coconutruben @jataylo

extent analysis

TL;DR

The issue can be resolved by modifying the model to avoid using view_as_complex and view_as_real with tensors that have a last dimension with stride not equal to 1, potentially by reordering or reorganizing the tensor operations.

Guidance

Identify the specific tensor operations that are causing the issue, potentially by using the tlparse tool to visualize the graph.
Consider reordering the tensor operations to avoid using view_as_complex and view_as_real with tensors that have a last dimension with stride not equal to 1.
Verify that the modified model produces the correct results in both eager and compiled modes.
If the issue persists, try to isolate the problem by creating a minimal reproducible example that demonstrates the issue.

Example

No specific code example can be provided without modifying the original code, but the general approach would be to reorganize the tensor operations to avoid the problematic view_as_complex and view_as_real calls.

Notes

The issue appears to be related to the interaction between the torch.compile mode and the specific tensor operations used in the model. The fact that the issue only occurs in compiled mode suggests that the problem may be related to the optimization or transformation of the tensor operations.

Recommendation

Apply a workaround by modifying the model to avoid the problematic tensor operations, as this is likely to be the most effective way to resolve the issue in the short term.

Vote matrix · Quick signals

Works

Did the solution work? Tap to confirm.

Easy Fix

Was it a quick fix?

Time Saver

Did it save you time?

Blocking

Was it severely blocking?

Common Issue

Are others likely hitting this too?

Flaky / Intermittent

Is it intermittent?

Verified / Reproducible

Can you reproduce it reliably?

#api #container setup #orchestration issue #cache issue #memory leak

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Data

Security

Network

Code

UI/UX

Text

System

Multimedia

Protocol

API

Engineering

pytorch - 💡(How to fix) Fix torch.compile + view_as_complex: RuntimeError: Tensor must have a last dimension with stride 1 [3 comments, 2 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Fix Action

Fix / Workaround

Code Example

🐛 Describe the bug

Error logs

Versions

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

TRENDING

pytorch - 💡(How to fix) Fix torch.compile + view_as_complex: RuntimeError: Tensor must have a last dimension with stride 1 [3 comments, 2 participants]

Recommended Tools

GitHub issue graph ai analysis

Error Message

Fix Action

Fix / Workaround

Code Example

🐛 Describe the bug

Error logs

Versions

extent analysis

TL;DR

Guidance

Example

Notes

Recommendation

Still need to ship something?

RELATED_DISCOVERY

TRENDING