🚀 The feature, motivation and pitch

torch.linalg.qr is needed by PEFT LoRA EVA(here) for weight initialization and update, but currently we can find it fallback to CPU, and slow, as below log.

Pls plan the XPU backend implementation for this op, thx.

UserWarning: Aten Op fallback from XPU to CPU happends. This may have performance implications. If need debug the fallback ops please set environment variable PYTORCH_DEBUG_XPU_FALLBACK=1 (Triggered internally at /pytorch/build/xpu/ATen/RegisterXPU_0.cpp:54525.)
Q = torch.linalg.qr(X).Q

Alternatives

No response

Additional context

No response