// Causes the calling thread to block till all ops dispatched in this executor
// have been executed. Note that "execution" here refers to kernel execution /
// scheduling of copies, etc. Similar to sync execution, it doesn't guarantee
// that lower level device queues (like GPU streams) have been flushed.
//
// This call may not block for execution of ops enqueued concurrently with this
// call.

// have all gradients available. This set usually is a subset of the set of
// targets (not all since targets which have outputs in the tape will not have
// gradients available initially).
//
// Then we repeatedly pop an entry from the stack, run its backprop, and update
// the gradients of its inputs. Once we have computed all gradients for a single

                                                     TF_Status* status);

// Gets the `PJRT_Buffer*` stored in the tensor. The status will contain error
// if the tensor does not have a `PjRtCApiBuffer`.
TF_CAPI_EXPORT extern PJRT_Buffer* TF_GetPjRtCBuffer(TF_Tensor* c_tensor,
                                                     TF_Status* status);