[pytorch] Migrating index_add cuda to ATen (pytorch#30573)

Summary: Pull Request resolved: pytorch#30573 Mostly just moved code. Index dim and number of indices checks are added to make checks idential to index_add_cpu_ ghstack-source-id: 98010129 Test Plan: existing tests Differential Revision: D18749922 fbshipit-source-id: d243be43a3b6a9b9591caf0c35ef2fb6ec0d3ead
orionr · Feb 11, 2020 · 330d051 · 330d051
1 parent 9857d9b
commit 330d051
Show file tree

Hide file tree

Showing 6 changed files with 326 additions and 243 deletions.
diff --git a/aten/src/ATen/Declarations.cwrap b/aten/src/ATen/Declarations.cwrap
@@ -190,19 +190,6 @@
     - THTensor* source
     - bool accumulate
 ]]
-[[
-  name: _th_index_add_
-  cname: indexAdd
-  variants: function
-  backends:
-    - CUDA
-  return: argument 0
-  arguments:
-    - THTensor* self
-    - long dim
-    - THIndexTensor* index
-    - THTensor* source
-]]
 [[
   name: _th_index_fill_
   cpu_bool: True