|
| pure logical function, public | accel_is_enabled () |
| |
| pure logical function, public | accel_allow_cpu_only () |
| |
| subroutine, public | accel_init (base_grp, namespace) |
| |
| subroutine, public | accel_end (namespace) |
| |
| integer(int64) function | accel_padded_size_i8 (nn) |
| |
| integer(int32) function | accel_padded_size_i4 (nn) |
| |
| subroutine | accel_create_buffer_4 (this, flags, type, size, set_zero, async) |
| |
| subroutine | accel_create_buffer_8 (this, flags, type, size, set_zero, async) |
| |
| subroutine, public | accel_release_buffer (this, async) |
| |
| subroutine, public | accel_ensure_buffer_size (buffer, flags, type, required_size, set_zero, async) |
| |
| logical pure function, public | accel_buffer_is_allocated (this) |
| |
| subroutine, public | accel_finish () |
| |
| subroutine | accel_set_kernel_arg_buffer (kernel, narg, buffer) |
| |
| subroutine | accel_set_kernel_arg_local (kernel, narg, type, size) |
| |
| subroutine | accel_kernel_run_8 (kernel, globalsizes, localsizes) |
| |
| subroutine | accel_kernel_run_4 (kernel, globalsizes, localsizes) |
| |
| integer pure function, public | accel_max_workgroup_size () |
| |
| integer function, public | accel_kernel_workgroup_size (kernel) |
| |
| subroutine | accel_set_buffer_to (buffer, type, val, nval, offset, async) |
| |
| subroutine | accel_set_buffer_to_zero_i8 (buffer, type, nval, offset, async) |
| |
| subroutine | accel_set_buffer_to_zero_i4 (buffer, type, nval, offset, async) |
| |
| subroutine | accel_check_bandwidth () |
| |
| logical pure function, public | accel_use_shared_mem () |
| |
| subroutine | accel_kernel_global_init () |
| |
| subroutine | accel_kernel_global_end () |
| |
| subroutine, public | accel_kernel_build (this, file_name, kernel_name, flags) |
| |
| subroutine | accel_kernel_end (this) |
| |
| subroutine, public | accel_kernel_start_call (this, file_name, kernel_name, flags) |
| |
| integer(int64) pure function, public | accel_global_memory_size () |
| |
| integer(int64) pure function, public | accel_local_memory_size () |
| |
| integer pure function, public | accel_max_size_per_dim (dim) |
| |
| subroutine, public | accel_set_stream (stream_number) |
| |
| subroutine, public | accel_get_stream (stream_number) |
| |
| subroutine, public | accel_synchronize_all_streams () |
| |
| type(c_ptr) function, public | daccel_get_pointer_with_offset (buffer, offset) |
| |
| type(c_ptr) function, public | zaccel_get_pointer_with_offset (buffer, offset) |
| |
| subroutine, public | accel_clean_pointer (buffer) |
| |
| subroutine, public | accel_get_unfolded_size (size, grid_size, thread_block_size) |
| | Get unfolded size: some kernels (e.g. projectors) unfold the array across warps as an optimization. This is, however, only benefitial on Nvidia GPUs. More...
|
| |
| subroutine | daccel_write_buffer_single (this, data, async) |
| |
| subroutine | daccel_write_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | daccel_write_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | daccel_write_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | daccel_write_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | daccel_write_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | daccel_write_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | daccel_write_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | daccel_write_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | daccel_write_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | daccel_write_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | daccel_write_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | daccel_write_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | daccel_write_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | daccel_write_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | daccel_read_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | daccel_read_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | daccel_read_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | daccel_read_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | daccel_read_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | daccel_read_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | daccel_read_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | daccel_read_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | daccel_read_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | daccel_read_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | daccel_read_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | daccel_read_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | daccel_read_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | daccel_read_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | daccel_set_kernel_arg_data (kernel, narg, data) |
| |
| subroutine | daccel_get_device_pointer_1 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | daccel_get_device_pointer_2 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | daccel_get_device_pointer_3 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | daccel_get_device_pointer_1l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | daccel_get_device_pointer_2l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | daccel_get_device_pointer_3l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | daccel_create_blas_alpha_beta_buffer (this, data, async) |
| |
| subroutine | daccel_release_blas_alpha_beta_buffer (this, data, async) |
| |
| subroutine | zaccel_write_buffer_single (this, data, async) |
| |
| subroutine | zaccel_write_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | zaccel_write_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | zaccel_read_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | zaccel_set_kernel_arg_data (kernel, narg, data) |
| |
| subroutine | zaccel_get_device_pointer_1 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | zaccel_get_device_pointer_2 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | zaccel_get_device_pointer_3 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | zaccel_get_device_pointer_1l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | zaccel_get_device_pointer_2l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | zaccel_get_device_pointer_3l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | zaccel_create_blas_alpha_beta_buffer (this, data, async) |
| |
| subroutine | zaccel_release_blas_alpha_beta_buffer (this, data, async) |
| |
| subroutine | iaccel_write_buffer_single (this, data, async) |
| |
| subroutine | iaccel_write_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | iaccel_write_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | iaccel_read_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | iaccel_set_kernel_arg_data (kernel, narg, data) |
| |
| subroutine | iaccel_get_device_pointer_1 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | iaccel_get_device_pointer_2 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | iaccel_get_device_pointer_3 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | iaccel_get_device_pointer_1l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | iaccel_get_device_pointer_2l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | iaccel_get_device_pointer_3l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | iaccel_create_blas_alpha_beta_buffer (this, data, async) |
| |
| subroutine | iaccel_release_blas_alpha_beta_buffer (this, data, async) |
| |
| subroutine | laccel_write_buffer_single (this, data, async) |
| |
| subroutine | laccel_write_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | laccel_write_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | laccel_write_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | laccel_write_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | laccel_write_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | laccel_write_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | laccel_write_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | laccel_write_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | laccel_write_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | laccel_write_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | laccel_write_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | laccel_write_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | laccel_write_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | laccel_write_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | laccel_read_buffer_0 (this, n1, data, offset, async) |
| |
| subroutine | laccel_read_buffer_1 (this, n1, data, offset, async) |
| |
| subroutine | laccel_read_buffer_2 (this, n1, n2, data, offset, async) |
| |
| subroutine | laccel_read_buffer_3 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | laccel_read_buffer_4 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | laccel_read_buffer_5 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | laccel_read_buffer_6 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | laccel_read_buffer_0_int32 (this, n1, data, offset, async) |
| |
| subroutine | laccel_read_buffer_1_int32 (this, n1, data, offset, async) |
| |
| subroutine | laccel_read_buffer_2_int32 (this, n1, n2, data, offset, async) |
| |
| subroutine | laccel_read_buffer_3_int32 (this, n1, n2, n3, data, offset, async) |
| |
| subroutine | laccel_read_buffer_4_int32 (this, n1, n2, n3, n4, data, offset, async) |
| |
| subroutine | laccel_read_buffer_5_int32 (this, n1, n2, n3, n4, n5, data, offset, async) |
| |
| subroutine | laccel_read_buffer_6_int32 (this, n1, n2, n3, n4, n5, n6, data, offset, async) |
| |
| subroutine | laccel_set_kernel_arg_data (kernel, narg, data) |
| |
| subroutine | laccel_get_device_pointer_1 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | laccel_get_device_pointer_2 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | laccel_get_device_pointer_3 (host_pointer, device_pointer, dimensions) |
| |
| subroutine | laccel_get_device_pointer_1l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | laccel_get_device_pointer_2l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | laccel_get_device_pointer_3l (host_pointer, device_pointer, dimensions) |
| |
| subroutine | laccel_create_blas_alpha_beta_buffer (this, data, async) |
| |
| subroutine | laccel_release_blas_alpha_beta_buffer (this, data, async) |
| |