# Silent kernel death

**URL:** <https://discuss.pytorch.org/t/silent-kernel-death/225653>\
**Category:** nlp\
**Created:** [October 26, 2026, 4:38pm UTC](https://discuss.pytorch.org/t/silent-kernel-death/225653 "2026-10-26T16:38:31Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![hunter](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/h/ba9def/32.png) [@hunter](https://discuss.pytorch.org/u/hunter)\
**Post date:** [October 26, 2026, 4:38pm UTC](https://discuss.pytorch.org/t/silent-kernel-death/225653/1 "2026-10-26T16:38:31Z")

</div>

Environment:  
OS : Linux  
Python : 3.13.6  
PyTorch : 2.6.0+cu124  
torchvision : 0.21.0+cu124  
transformers : 5.19.0  
CUDA : 12.4 (cuDNN 9.1)  
GPU : 8× Tesla V100-PCIE-32GB (compute capability 7.0)  
What I’m doing:  
I load a Qwen3.5-style model (contains GatedDeltaNet linear attention layers) with device\_map=“balanced” across 8 V100 GPUs (dtype float16), then run model.generate(…).  
The code worked, but printed a warning recommending to install flash-linear-attention. So I installed it via uv pip install flash-linear-attention. During generate(), the whole Jupyter kernel process died with no Python traceback (log: “kernel process died, ExitCode undefined”). At that point I had fla + Triton 3.2.0 installed.

what I did:

1. I upgraded Triton to 3.3.0 — didn’t help.
2. I uninstalled flash-linear-attention alone — no change.
3. I then uninstalled both flash-linear-attention and triton. No more crashes, but now model.generate() is extremely slow: ~8 minutes for a simple generation.

What is the probable cause of the original silent kernel death? Is it Triton 3.3.0 incompatible with torch 2.6.0 (which expects Triton 3.2.0), or an sm70 issue — i.e. do the fla/GatedDeltaNet Triton kernels support V100 (sm70) at all?  
Thanks!

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [October 26, 2026, 6:40pm UTC](https://discuss.pytorch.org/t/silent-kernel-death/225653/2 "2026-10-26T18:40:32Z")

</div>

Jupyter notebooks are unfortunately crashing sometimes without properly report error messages. Try to run the code in a terminal (you can export the notebook as a script) and check if proper errors are displayed there.

---

<div class="post-metadata">

**Author:** ![hunter](https://discuss.pytorch.org/letter_avatar_proxy/v4/letter/h/ba9def/32.png) [@hunter](https://discuss.pytorch.org/u/hunter)\
**Post date:** [October 27, 2026, 4:44pm UTC](https://discuss.pytorch.org/t/silent-kernel-death/225653/3 "2026-10-27T16:44:40Z")

</div>

I run the code .py file but at that moment I got this error

LLVM **ERROR** : Failed to compute parent layout for slice layout.

---

<div class="post-metadata">

**Author:** ![ptrblck](https://discuss.pytorch.org/user_avatar/discuss.pytorch.org/ptrblck/32/1823_2.png) [@ptrblck](https://discuss.pytorch.org/u/ptrblck)\
**Post date:** [October 27, 2026, 9:12pm UTC](https://discuss.pytorch.org/t/silent-kernel-death/225653/4 "2026-10-27T21:12:50Z")

</div>

Thanks you! Indeed I think your initial guess was right:

> [@hunter](#):
>
> or an sm70 issue — i.e. do the fla/GatedDeltaNet Triton kernels support V100 (sm70) at all?

Triton does not support GPU architectures older than Ampere (sm\_80). From the [repo](https://github.com/triton-lang/triton/tree/v3.3.0#compatibility):

> Supported Hardware:
> 
> - NVIDIA GPUs (Compute Capability 8.0+)