[GitHub] [incubator-tvm] petrex commented on a change in pull request #4342: Add workgroup size attribute to AMDGPU functions in codegen

[GitBox <gi...@apache.org>]

petrex commented on a change in pull request #4342: Add workgroup size attribute to AMDGPU functions in codegen
URL: https://github.com/apache/incubator-tvm/pull/4342#discussion_r346592319
 
 

 ##########
 File path: src/codegen/llvm/codegen_amdgpu.cc
 ##########
 @@ -36,13 +36,39 @@
 namespace tvm {
 namespace codegen {
 
+namespace {
+
+// calls the device api to get the max threads per block
+static inline int DetectROCMmaxThreadsPerBlock() {
+  TVMContext tvm_ctx;
+  tvm_ctx.device_type = kDLROCM;
+  tvm_ctx.device_id = 0;
+  tvm::runtime::DeviceAPI* api = tvm::runtime::DeviceAPI::Get(tvm_ctx, true);
+  if (api != nullptr) {
+    TVMRetValue val;
+    api->GetAttr(tvm_ctx, tvm::runtime::kExist, &val);
+    if (val.operator int() == 1) {
+      tvm::runtime::DeviceAPI::Get(tvm_ctx)->
+        GetAttr(tvm_ctx, tvm::runtime::kMaxThreadsPerBlock, &val);
+      return val.operator int();
+    }
+  }
+  LOG(WARNING) << "Cannot get maximum number of threads for AMD codegen";
+  return 1024;
 
 Review comment:
   Can we set this value to 256?
   
   In general 1024 is fine as long as the kernel did not use a lot of sgpr/vgpr.
   however, we also see kernels with high vgpr usage (> 128), generate incorrect results with workgroup size == 1024.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services