{"hash": "772fd94eeb223063f118d1a7299139dcf8d929a63036b8576173946187d0ae7d", "target": {"backend": "cuda", "arch": 89, "warp_size": 32}, "num_warps": 16, "num_ctas": 1, "num_stages": 3, "warp_size": 32, "maxnreg": null, "ptx_version": null, "ptx_options": null, "ir_override": null, "enable_fp_fusion": true, "enable_reflect_ftz": true, "launch_cooperative_grid": false, "launch_pdl": false, "supported_fp8_dtypes": ["fp8e4b15", "fp8e4nv", "fp8e5"], "deprecated_fp8_dot_operand_dtypes": [], "default_dot_input_precision": "tf32", "allowed_dot_input_precisions": ["tf32", "tf32x3", "ieee", "bf16x3", "bf16x6"], "max_num_imprecise_acc_default": 0, "extern_libs": [["libdevice", "D:\\Comfy\\App\\ComfyUI_windows_portable\\python_embeded\\Lib\\site-packages\\triton\\backends\\nvidia\\lib\\libdevice.10.bc"]], "debug": false, "backend_name": "cuda", "sanitize_overflow": true, "arch": "sm89", "instrumentation_mode": "", "triton_version": "3.8.0", "tensordesc_meta": [], "shared": 64, "tmem_size": 0, "global_scratch_size": 0, "global_scratch_align": 1, "profile_scratch_size": 0, "profile_scratch_align": 1, "name": "_rms_norm_scale_shift_kernel"}