/*************************************************************************
 * Copyright (c) 2015-2020, NVIDIA CORPORATION. All rights reserved.
 * Modifications Copyright (c) Microsoft Corporation. Licensed under the MIT License.
 *
 * See LICENSE.txt for license information
 ************************************************************************/

#include "enqueue.h"
#include "collectives.h"

#include "msccl/msccl_lifecycle.h"
#include "hipprof/hip_prof_rccl_api.h"

NCCL_API(ncclResult_t, ncclAllGather, const void* sendbuff, void* recvbuff, size_t sendcount,
    ncclDataType_t datatype, ncclComm_t comm, cudaStream_t stream);
ncclResult_t ncclAllGather(const void* sendbuff, void* recvbuff, size_t sendcount,
    ncclDataType_t datatype, ncclComm_t comm, cudaStream_t stream) {
  // Just pass the size of one message and not the total bytes sent/received.
  constexpr nvtxPayloadSchemaEntry_t AllGatherSchema[] = {
    {0, NVTX_PAYLOAD_ENTRY_TYPE_SIZE, "Message size [bytes]"}
  };
  size_t msgsize = sendcount * ncclTypeSize(datatype);
  NVTX3_FUNC_WITH_PARAMS(AllGather, AllGatherSchema, msgsize)

  if (mscclAvailable() && !mscclIsCaller()) {
    return mscclEnqueueCheck(
      sendbuff, nullptr, nullptr, recvbuff, nullptr, nullptr,
      sendcount, datatype, 0, 0, ncclSum, mscclFuncAllGather, comm, stream);
  }

  struct ncclInfo info = { ncclFuncAllGather, "AllGather",
    sendbuff, recvbuff, sendcount, datatype, ncclSum, 0, comm, stream, /* Args */
    ALLGATHER_CHUNKSTEPS, ALLGATHER_SLICESTEPS };

  RCCL_CB_SPAWNER_OBJECT(ncclAllGather, &info);

  return ncclEnqueueCheck(&info);
}