Rename alltoall -> alltoall_single (#3546)

Summary:
Pull Request resolved: #3546

X-link: facebookresearch/FBGEMM#631

current naming is confusing as other alltoall interface takes in list of tensors, which are more generic.

Differential Revision: D67814911

fbshipit-source-id: a46f855bf9275968a862577e4f3e12a87bfde7d4

fbgemm_gpu/experimental/gen_ai/src/comm/car.cpp

@@ -156,7 +156,7 @@ void nccl_allgather(at::Tensor dst, at::Tensor src, int64_t comm_idx) {
       "ncclAllGather");
 }
 
-void nccl_alltoall(
+void nccl_alltoall_single(
     at::Tensor dst,
     at::Tensor src,
     int64_t world_size,
@@ -271,7 +271,7 @@ TORCH_LIBRARY_FRAGMENT(fbgemm, m) {
   m.def("nccl_allgather(Tensor(a!) dst, Tensor src, int comm_idx=0) -> ()");
 
   m.def(
-      "nccl_alltoall(Tensor(a!) dst, Tensor src, int world_size, int comm_idx=0) -> ()");
+      "nccl_alltoall_single(Tensor(a!) dst, Tensor src, int world_size, int comm_idx=0) -> ()");
 
   m.def("nccl_reducescatter(Tensor(a!) dst, Tensor src, int comm_idx=0) -> ()");
 
@@ -298,7 +298,7 @@ TORCH_LIBRARY_FRAGMENT(fbgemm, m) {
 TORCH_LIBRARY_IMPL(fbgemm, CUDA, m) {
   m.impl("nccl_allreduce", nccl_allreduce);
   m.impl("nccl_allgather", nccl_allgather);
-  m.impl("nccl_alltoall", nccl_alltoall);
+  m.impl("nccl_alltoall_single", nccl_alltoall_single);
   m.impl("nccl_reducescatter", nccl_reducescatter);
   m.impl("one_shot_car_allreduce", one_shot_car_allreduce);
   m.impl("two_shot_car_allreduce", two_shot_car_allreduce);
@@ -309,7 +309,7 @@ TORCH_LIBRARY_IMPL(fbgemm, CUDA, m) {
 TORCH_LIBRARY_IMPL(fbgemm, CPU, m) {
   m.impl("nccl_allreduce", nccl_allreduce);
   m.impl("nccl_allgather", nccl_allgather);
-  m.impl("nccl_alltoall", nccl_alltoall);
+  m.impl("nccl_alltoall_single", nccl_alltoall_single);
   m.impl("nccl_reducescatter", nccl_reducescatter);
   m.impl("one_shot_car_allreduce", one_shot_car_allreduce);
   m.impl("two_shot_car_allreduce", two_shot_car_allreduce);
@@ -331,7 +331,7 @@ void nccl_allgather_meta(
   return;
 }
 
-void nccl_alltoall_meta(
+void nccl_alltoall_single_meta(
     at::Tensor /* dst */,
     at::Tensor /* src */,
     int64_t /* world_size */,
@@ -365,7 +365,7 @@ void two_shot_car_allreduce_meta(
 TORCH_LIBRARY_IMPL(fbgemm, Meta, m) {
   m.impl("nccl_allreduce", nccl_allreduce_meta);
   m.impl("nccl_allgather", nccl_allgather_meta);
-  m.impl("nccl_alltoall", nccl_alltoall_meta);
+  m.impl("nccl_alltoall_single", nccl_alltoall_single_meta);
   m.impl("nccl_reducescatter", nccl_reducescatter_meta);
   m.impl("one_shot_car_allreduce", one_shot_car_allreduce_meta);
   m.impl("two_shot_car_allreduce", two_shot_car_allreduce_meta);