Cuda.ModuleA CUDA module type represents CUDA code that's ready to execute, i.e. is loaded. See: Module Management.
Compute device classes. See enum CUjit_target.
val sexp_of_jit_target : jit_target -> Sexplib0.Sexp.tCubin matching fallback strategies. See CUjit_fallback.
val sexp_of_jit_fallback : jit_fallback -> Sexplib0.Sexp.tCaching modes for dlcm. See CUjit_cacheMode.
val sexp_of_jit_cache_mode : jit_cache_mode -> Sexplib0.Sexp.ttype jit_option = | MAX_REGISTERS of intMax number of registers that a thread may use.
*)| THREADS_PER_BLOCK of intSpecifies minimum number of threads per block to target compilation for or returns the number of threads the compiler actually targeted. Cannot be combined with TARGET.
| WALL_TIME of {}| INFO_LOG_BUFFER of bigstring| ERROR_LOG_BUFFER of bigstring| OPTIMIZATION_LEVEL of int0 to 4, with 4 being the default and highest level of optimizations.
*)| TARGET_FROM_CUCONTEXT| TARGET of jit_target| FALLBACK_STRATEGY of jit_fallback| GENERATE_DEBUG_INFO of boolHelpful for cuda-gdb.
*)| LOG_VERBOSE of bool| GENERATE_LINE_INFO of boolHelpful for cuda-gdb.
*)| CACHE_MODE of jit_cache_mode| POSITION_INDEPENDENT_CODE of boolSee CUjit_option.
val sexp_of_jit_option : jit_option -> Sexplib0.Sexp.tSee CUfunction.
See CUmodule.
val load_data_ex : Nvrtc.compile_to_ptx_result -> jit_option list -> tCurrently, the image passed via this call is the PTX source. See cuModuleLoadDataEx.
The module is finalized using cuModuleUnload. The finalizer captures the context when load_data_ex is called to temporarily push it on the stack for unloading.
See cuModuleGetFunction.
val get_global : t -> name:string -> Deviceptr.t * Unsigned.size_tSee cuModuleGetGlobal.