Issue Station
Sign In
Tools hereToolbox

pytorch - 💡(How to fix) Fix AOTAutograd backward crash with view and graph break

Official PRs (…)
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.

Helpful · Quick feedback

Loading…

Error Message

""" Minimal repro: torch.compile (AOTAutograd) backward crashes on a bare .expand() tangent. torch 2.12 regression (the same code path worked on 2.11).

RuntimeError: unsupported operation: more than one element of the written-to tensor refers to a single memory location. Please clone() the tensor before performing the operation. from AOTAutograd: process_runtime_tangent -> coerce_to_expected_memory_format -> restrided.copy_(x) (torch/_functorch/_aot_autograd/runtime_wrappers.py:2325)

All of these are needed to trip it (drop any one and it passes):

  1. two differently-broadcast tensors cat'd (establishes the stride pattern),
  2. a bare .expand() over a fresh dim -- the offending 0-stride view,
  3. a graph break, so the expand becomes a fwd/bwd boundary tensor whose tangent keeps the 0-stride,
  4. a matmul consumer after the break + a backward.

python3 repro.py # crashes python3 repro.py --fix # OK: backward completed """

import sys

import torch import torch.nn as nn

FIX = "--fix" in sys.argv print("torch", torch.version, "| fix =", FIX)

class M(nn.Module): def init(self, f=32, c=16, h=8, w=8): super().init() self.ah = nn.Parameter(torch.randn(h, f)) self.aw = nn.Parameter(torch.randn(w, f)) self.lin = nn.Linear(2 * f, c, bias=False)

def forward(self, x):  # x: (B, H, W, c)
    b, h, w, _ = x.shape
    pe = torch.cat([self.ah.unsqueeze(1).expand(h, w, -1), self.aw.unsqueeze(0).expand(h, w, -1)], dim=-1)  # (H, W, 2F)
    pe = pe.unsqueeze(0).expand(b, h, w, -1)  # bare expand over batch -> 0-stride on dim 0
    if FIX:
        pe = pe.contiguous()
    torch._dynamo.graph_break()  # makes pe a fwd/bwd boundary tensor
    return x + self.lin(pe)

m = M().cuda() x = torch.randn(4, 8, 8, 16, device="cuda", requires_grad=True) torch.compile(m)(x).pow(2).sum().backward() print("OK: backward completed")

Root Cause

Root cause

A forward op returns a bare .expand() (a 0-stride view, never materialized) that sits on a graph boundary. The tangent for that boundary tensor is recorded with the expanded (0-stride) layout; at runtime torch 2.12 tries to materialize a buffer with that layout (empty_strided + copy_) and rejects the self-overlap. Materializing the forward output with .contiguous() makes the tangent dense and fixes it.

All four ingredients are required — drop any one and it passes (verified):

  1. two differently-broadcast tensors cat'd together (establishes the stride pattern),
  2. a bare .expand() over a fresh dim — the offending 0-stride view,
  3. a graph break so the expand becomes a fwd/bwd boundary tensor (here an explicit torch._dynamo.graph_break(); in the real code it was a device.type check + a print),
  4. a matmul consumer after the break, plus a backward.

Fix Action

Fix / Workaround

kevin@kevin-b200-0:/mnt/clusterstorage/workspace/kevin/expand-tangent-repro$ python repro.py 
torch 2.12.0 | fix = False
/usr/local/lib/python3.12/dist-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.
  warnings.warn(
Traceback (most recent call last):
  File "/mnt/clusterstorage/workspace/kevin/expand-tangent-repro/repro.py", line 48, in <module>
    torch.compile(m)(x).pow(2).sum().backward()
  File "/usr/local/lib/python3.12/dist-packages/torch/_tensor.py", line 631, in backward
    torch.autograd.backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/__init__.py", line 379, in backward
    _engine_run_backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/graph.py", line 882, in _engine_run_backward
    return Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/function.py", line 317, in apply
    return user_fn(self, *args)
           ^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2856, in backward
    all_args = _backward_prologue_functional(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2170, in _backward_prologue_functional
    AOTDispatchAutograd.process_runtime_tangent(
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 3112, in process_runtime_tangent
    x = coerce_to_expected_memory_format(x, meta.memory_format)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2325, in coerce_to_expected_memory_format
    restrided.copy_(x)
RuntimeError: unsupported operation: more than one element of the written-to tensor refers to a single memory location. Please clone() the tensor before performing the operation.

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 8562Y+ CPU family: 6 Model: 207 Thread(s) per core: 2 Core(s) per socket: 32 Socket(s): 2 Stepping: 2 CPU(s) scaling MHz: 98% 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 intel_ppin 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 user_shstk 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): 0-31,64-95 NUMA node1 CPU(s): 32-63,96-127 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 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 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

"""
Minimal repro: torch.compile (AOTAutograd) backward crashes on a bare .expand() tangent.
torch 2.12 regression (the same code path worked on 2.11).

  RuntimeError: unsupported operation: more than one element of the written-to tensor refers
  to a single memory location. Please clone() the tensor before performing the operation.
    from AOTAutograd: process_runtime_tangent -> coerce_to_expected_memory_format -> restrided.copy_(x)
    (torch/_functorch/_aot_autograd/runtime_wrappers.py:2325)

All of these are needed to trip it (drop any one and it passes):
  1. two differently-broadcast tensors cat'd (establishes the stride pattern),
  2. a bare .expand() over a fresh dim -- the offending 0-stride view,
  3. a graph break, so the expand becomes a fwd/bwd boundary tensor whose tangent keeps the 0-stride,
  4. a matmul consumer after the break + a backward.

  python3 repro.py        # crashes
  python3 repro.py --fix  # OK: backward completed
"""

import sys

import torch
import torch.nn as nn

FIX = "--fix" in sys.argv
print("torch", torch.__version__, "| fix =", FIX)


class M(nn.Module):
    def __init__(self, f=32, c=16, h=8, w=8):
        super().__init__()
        self.ah = nn.Parameter(torch.randn(h, f))
        self.aw = nn.Parameter(torch.randn(w, f))
        self.lin = nn.Linear(2 * f, c, bias=False)

    def forward(self, x):  # x: (B, H, W, c)
        b, h, w, _ = x.shape
        pe = torch.cat([self.ah.unsqueeze(1).expand(h, w, -1), self.aw.unsqueeze(0).expand(h, w, -1)], dim=-1)  # (H, W, 2F)
        pe = pe.unsqueeze(0).expand(b, h, w, -1)  # bare expand over batch -> 0-stride on dim 0
        if FIX:
            pe = pe.contiguous()
        torch._dynamo.graph_break()  # makes pe a fwd/bwd boundary tensor
        return x + self.lin(pe)


m = M().cuda()
x = torch.randn(4, 8, 8, 16, device="cuda", requires_grad=True)
torch.compile(m)(x).pow(2).sum().backward()
print("OK: backward completed")

---

kevin@kevin-b200-0:/mnt/clusterstorage/workspace/kevin/expand-tangent-repro$ python repro.py 
torch 2.12.0 | fix = False
/usr/local/lib/python3.12/dist-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.
  warnings.warn(
Traceback (most recent call last):
  File "/mnt/clusterstorage/workspace/kevin/expand-tangent-repro/repro.py", line 48, in <module>
    torch.compile(m)(x).pow(2).sum().backward()
  File "/usr/local/lib/python3.12/dist-packages/torch/_tensor.py", line 631, in backward
    torch.autograd.backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/__init__.py", line 379, in backward
    _engine_run_backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/graph.py", line 882, in _engine_run_backward
    return Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/function.py", line 317, in apply
    return user_fn(self, *args)
           ^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2856, in backward
    all_args = _backward_prologue_functional(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2170, in _backward_prologue_functional
    AOTDispatchAutograd.process_runtime_tangent(
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 3112, in process_runtime_tangent
    x = coerce_to_expected_memory_format(x, meta.memory_format)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2325, in coerce_to_expected_memory_format
    restrided.copy_(x)
RuntimeError: unsupported operation: more than one element of the written-to tensor refers to a single memory location. Please clone() the tensor before performing the operation.
RAW_BUFFERClick to expand / collapse

🐛 Describe the bug

Claude says (I haven't checked):

Root cause

A forward op returns a bare .expand() (a 0-stride view, never materialized) that sits on a graph boundary. The tangent for that boundary tensor is recorded with the expanded (0-stride) layout; at runtime torch 2.12 tries to materialize a buffer with that layout (empty_strided + copy_) and rejects the self-overlap. Materializing the forward output with .contiguous() makes the tangent dense and fixes it.

All four ingredients are required — drop any one and it passes (verified):

  1. two differently-broadcast tensors cat'd together (establishes the stride pattern),
  2. a bare .expand() over a fresh dim — the offending 0-stride view,
  3. a graph break so the expand becomes a fwd/bwd boundary tensor (here an explicit torch._dynamo.graph_break(); in the real code it was a device.type check + a print),
  4. a matmul consumer after the break, plus a backward.
"""
Minimal repro: torch.compile (AOTAutograd) backward crashes on a bare .expand() tangent.
torch 2.12 regression (the same code path worked on 2.11).

  RuntimeError: unsupported operation: more than one element of the written-to tensor refers
  to a single memory location. Please clone() the tensor before performing the operation.
    from AOTAutograd: process_runtime_tangent -> coerce_to_expected_memory_format -> restrided.copy_(x)
    (torch/_functorch/_aot_autograd/runtime_wrappers.py:2325)

All of these are needed to trip it (drop any one and it passes):
  1. two differently-broadcast tensors cat'd (establishes the stride pattern),
  2. a bare .expand() over a fresh dim -- the offending 0-stride view,
  3. a graph break, so the expand becomes a fwd/bwd boundary tensor whose tangent keeps the 0-stride,
  4. a matmul consumer after the break + a backward.

  python3 repro.py        # crashes
  python3 repro.py --fix  # OK: backward completed
"""

import sys

import torch
import torch.nn as nn

FIX = "--fix" in sys.argv
print("torch", torch.__version__, "| fix =", FIX)


class M(nn.Module):
    def __init__(self, f=32, c=16, h=8, w=8):
        super().__init__()
        self.ah = nn.Parameter(torch.randn(h, f))
        self.aw = nn.Parameter(torch.randn(w, f))
        self.lin = nn.Linear(2 * f, c, bias=False)

    def forward(self, x):  # x: (B, H, W, c)
        b, h, w, _ = x.shape
        pe = torch.cat([self.ah.unsqueeze(1).expand(h, w, -1), self.aw.unsqueeze(0).expand(h, w, -1)], dim=-1)  # (H, W, 2F)
        pe = pe.unsqueeze(0).expand(b, h, w, -1)  # bare expand over batch -> 0-stride on dim 0
        if FIX:
            pe = pe.contiguous()
        torch._dynamo.graph_break()  # makes pe a fwd/bwd boundary tensor
        return x + self.lin(pe)


m = M().cuda()
x = torch.randn(4, 8, 8, 16, device="cuda", requires_grad=True)
torch.compile(m)(x).pow(2).sum().backward()
print("OK: backward completed")

Error logs

kevin@kevin-b200-0:/mnt/clusterstorage/workspace/kevin/expand-tangent-repro$ python repro.py 
torch 2.12.0 | fix = False
/usr/local/lib/python3.12/dist-packages/torch/_inductor/compile_fx.py:320: UserWarning: TensorFloat32 tensor cores for float32 matrix multiplication available but not enabled. Consider setting `torch.set_float32_matmul_precision('high')` for better performance.
  warnings.warn(
Traceback (most recent call last):
  File "/mnt/clusterstorage/workspace/kevin/expand-tangent-repro/repro.py", line 48, in <module>
    torch.compile(m)(x).pow(2).sum().backward()
  File "/usr/local/lib/python3.12/dist-packages/torch/_tensor.py", line 631, in backward
    torch.autograd.backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/__init__.py", line 379, in backward
    _engine_run_backward(
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/graph.py", line 882, in _engine_run_backward
    return Variable._execution_engine.run_backward(  # Calls into the C++ engine to run the backward pass
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/autograd/function.py", line 317, in apply
    return user_fn(self, *args)
           ^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2856, in backward
    all_args = _backward_prologue_functional(
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2170, in _backward_prologue_functional
    AOTDispatchAutograd.process_runtime_tangent(
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 3112, in process_runtime_tangent
    x = coerce_to_expected_memory_format(x, meta.memory_format)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.12/dist-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 2325, in coerce_to_expected_memory_format
    restrided.copy_(x)
RuntimeError: unsupported operation: more than one element of the written-to tensor refers to a single memory location. Please clone() the tensor before performing the operation.

Versions

PyTorch version: 2.12.0 Is debug build: False CUDA used to build PyTorch: 13.2 ROCM used to build PyTorch: N/A

OS: Ubuntu 24.04.4 LTS (x86_64) GCC version: (Ubuntu 13.3.0-6ubuntu2~24.04.1) 13.3.0 Clang version: Could not collect CMake version: version 4.3.2 Libc version: glibc-2.39

Python version: 3.12.3 (main, Mar 23 2026, 19:04:32) [GCC 13.3.0] (64-bit runtime) Python platform: Linux-6.8.12-680-6063-coreweave-amd64-f81899c8-x86_64-with-glibc2.39 Is CUDA available: True CUDA runtime version: 13.2.78 CUDA_MODULE_LOADING set to: GPU models and configuration: GPU 0: NVIDIA B200 Nvidia driver version: 595.45.04 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 Is XPU available: False HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True Caching allocator config: N/A

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 8562Y+ CPU family: 6 Model: 207 Thread(s) per core: 2 Core(s) per socket: 32 Socket(s): 2 Stepping: 2 CPU(s) scaling MHz: 98% 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 intel_ppin 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 user_shstk 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): 0-31,64-95 NUMA node1 CPU(s): 32-63,96-127 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 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 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] lovely-numpy==0.2.24 [pip3] mypy_extensions==1.1.0 [pip3] numpy==1.26.4 [pip3] onnx==1.21.0 [pip3] onnx-ir==0.2.1 [pip3] onnxscript==0.7.0 [pip3] torch==2.12.0 [pip3] torch_c_dlpack_ext==0.1.5 [pip3] torchaudio==2.11.0 [pip3] torchdata==0.11.0 [pip3] torchdiffeq==0.2.5 [pip3] torchsde==0.2.6 [pip3] torchtitan==0.2.2 [pip3] torchvision==0.27.0 [pip3] triton==3.7.0+gitb4e20bbe [pip3] welford-torch==0.2.5 [conda] Could not collect

cc @jamesr66a @chauhang @penguinwu @bdhirsh @bobrenjc93 @aorenste

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?
Loading…

Still need to ship something?

×6

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

Back to top recommendations

TRENDING

pytorch - 💡(How to fix) Fix AOTAutograd backward crash with view and graph break