3j&SSKrS/r\rSr\"\R S5(a\R RrSr"SS\5rg)N GreenContextF_CUDAGreenContextTc ^\rSrSrSr\SSSSS.S\S-S\S-S\S-S\S-S \4 S jj5r \SS\S-S \4S jj5r SU4S jjr SU4S jjr SU4Sjjr SrU=r$)rzmWrapper around a CUDA green context. .. warning:: This API is in beta and may change in future releases. Nnum_smsworkqueue_scopeworkqueue_concurrency_limit device_idrr r r returncZ[(d [S5e[RUUUUS9$)aCreate a CUDA green context. At least one of ``num_sms`` or ``workqueue_scope`` must be specified. Both can be combined to partition SMs and configure workqueues in the same green context. Arguments: num_sms (int, optional): The number of SMs to use in the green context. When ``None``, SMs are not partitioned. workqueue_scope (str, optional): Workqueue sharing scope. One of ``"device_ctx"`` (shared across all contexts, default driver behaviour) or ``"balanced"`` (non-overlapping workqueues with other balanced green contexts). When ``None``, no workqueue configuration is applied. workqueue_concurrency_limit (int, optional): Maximum number of concurrent stream-ordered workloads for the workqueue. Requires ``workqueue_scope`` to be set. device_id (int, optional): The device index of green context. When ``None``, the current device is used. 1PyTorch was not built with Green Context support!)r rr r ) SUPPORTED RuntimeError _GreenContextcreaters S/home/wildlama/miniconda3/lib/python3.13/site-packages/torch/cuda/green_contexts.pyrGreenContext.creates98yRS S##+(C $  cT[(d [S5e[RUS9$)a?Return the maximum workqueue concurrency limit for the device. This queries the device for the default number of concurrent stream-ordered workloads supported by workqueue configuration resources. Arguments: device_id (int, optional): The device index to query. When ``None``, the current device is used. rr )rrrmax_workqueue_concurrencyrs rr&GreenContext.max_workqueue_concurrency>s)yRS S666KKrc >[TU]5$)z+Make the green context the current context.)super set_contextself __class__s rrGreenContext.set_contextPsw"$$rc >[TU]5$)zsAssuming the green context is the current context, pop it from the context stack and restore the previous context. )r pop_contextrs rr"GreenContext.pop_contextTsw"$$rc >[TU]5$)z1Return the CUDA Stream used by the green context.)rStreamrs rr%GreenContext.StreamZsw~r)N)r N)r ztorch.cuda.Stream)__name__ __module__ __qualname____firstlineno____doc__ staticmethodintstrrrrrr"r%__static_attributes__ __classcell__)rs@rrrs #&*26 $ " t" t" &)4Z " : "  " " H LS4Z L3 L L"%%   r) torch__all__objectrrhasattr_Crrr'rrr7sT     588())HH..MI J =J r