偶尔触发CUDA error: device-side assert triggered，频率还挺高

[LongMarch7]

cosyvoice/cli/model.py", line 108, in token2wav token_len=torch.tensor([token.shape[1]], dtype=torch.int32).to(self.device), RuntimeError: CUDA error: device-side assert triggered Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

[aluminumbox]

没遇到过

[matrixloong]

RuntimeError: CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

[github-actions[bot]]

This issue is stale because it has been open for 30 days with no activity.

[DongmingShenDS]

后来解决了吗？

[wang-TJ-20]

后来解决了吗？

@matrixloong @LongMarch7 @DongmingShenDS @passerbya hi，你们有解决吗，我发现flow推理，在压测时会出现这个问题

[duanyu]

+1，同遇到这个问题！大家有解决方案不？

[DongmingShenDS]

我也是推理的时候遇到的这个问题。