[HIP] Change default --gpu-max-threads-per-block value to 1024

Differential Revision: https://reviews.llvm.org/D76795

Author: yxsamliu

clang/include/clang/Basic/LangOptions.def

@@ -238,7 +238,7 @@ LANGOPT(CUDAHostDeviceConstexpr, 1, 1, "treating unattributed constexpr function
 LANGOPT(CUDADeviceApproxTranscendentals, 1, 0, "using approximate transcendental functions")
 LANGOPT(GPURelocatableDeviceCode, 1, 0, "generate relocatable device code")
 LANGOPT(GPUAllowDeviceInit, 1, 0, "allowing device side global init functions for HIP")
-LANGOPT(GPUMaxThreadsPerBlock, 32, 256, "default max threads per block for kernel launch bounds for HIP")
+LANGOPT(GPUMaxThreadsPerBlock, 32, 1024, "default max threads per block for kernel launch bounds for HIP")
 
 LANGOPT(SYCL              , 1, 0, "SYCL")
 LANGOPT(SYCLIsDevice      , 1, 0, "Generate code for SYCL device")

clang/lib/CodeGen/TargetInfo.cpp

@@ -8815,9 +8815,13 @@ void AMDGPUTargetCodeGenInfo::setTargetAttributes(
       assert(Max == 0 && "Max must be zero");
   } else if (IsOpenCLKernel || IsHIPKernel) {
     // By default, restrict the maximum size to a value specified by
-    // --gpu-max-threads-per-block=n or its default value.
+    // --gpu-max-threads-per-block=n or its default value for HIP.
+    const unsigned OpenCLDefaultMaxWorkGroupSize = 256;
+    const unsigned DefaultMaxWorkGroupSize =
+        IsOpenCLKernel ? OpenCLDefaultMaxWorkGroupSize
+                       : M.getLangOpts().GPUMaxThreadsPerBlock;
     std::string AttrVal =
-        std::string("1,") + llvm::utostr(M.getLangOpts().GPUMaxThreadsPerBlock);
+        std::string("1,") + llvm::utostr(DefaultMaxWorkGroupSize);
     F->addFnAttr("amdgpu-flat-work-group-size", AttrVal);
   }
 

clang/test/CodeGenCUDA/amdgpu-kernel-attrs.cu

@@ -39,7 +39,7 @@ __global__ void num_vgpr_64() {
 // NAMD-NOT: "amdgpu-num-vgpr"
 // NAMD-NOT: "amdgpu-num-sgpr"
 
-// DEFAULT-DAG: attributes [[FLAT_WORK_GROUP_SIZE_DEFAULT]] = {{.*}}"amdgpu-flat-work-group-size"="1,256"{{.*}}"uniform-work-group-size"="true"
+// DEFAULT-DAG: attributes [[FLAT_WORK_GROUP_SIZE_DEFAULT]] = {{.*}}"amdgpu-flat-work-group-size"="1,1024"{{.*}}"uniform-work-group-size"="true"
 // MAX1024-DAG: attributes [[FLAT_WORK_GROUP_SIZE_DEFAULT]] = {{.*}}"amdgpu-flat-work-group-size"="1,1024"
 // CHECK-DAG: attributes [[FLAT_WORK_GROUP_SIZE_32_64]] = {{.*}}"amdgpu-flat-work-group-size"="32,64"
 // CHECK-DAG: attributes [[WAVES_PER_EU_2]] = {{.*}}"amdgpu-waves-per-eu"="2"

clang/test/CodeGenCUDA/kernel-amdgcn.cu

@@ -39,4 +39,4 @@ int main() {
   launch((void*)D.Empty());
   return 0;
 }
-// CHECK: attributes #[[ATTR]] = {{.*}}"amdgpu-flat-work-group-size"="1,256"
+// CHECK: attributes #[[ATTR]] = {{.*}}"amdgpu-flat-work-group-size"="1,1024"