maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

[CUDA/SPIR-V] Force passing aggregate type byval 

This patch forces copying aggregate type in kernel arguments by value when
compiling CUDA targeting SPIR-V. The original behavior is not passing by value
when there is any of destructor, copy constructor and move constructor defined
by user. This patch makes the behavior of SPIR-V generated from CUDA follow
the CUDA spec
(https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#global-function-argument-processing),
and matches the NVPTX
implementation (
41958f76d8/clang/lib/CodeGen/TargetInfo.cpp (L7241)).

Differential Revision: https://reviews.llvm.org/D130387
This commit is contained in:
Shangwu Yao 2022-07-22 19:22:08 +00:00
parent 2f46983981
commit 31d8dbd1e5
2 changed files with 34 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
clang/lib/CodeGen/TargetInfo.cpp
View File

@ -10449,6 +10449,15 @@ ABIArgInfo SPIRVABIInfo::classifyKernelArgumentType(QualType Ty) const {
LTy = llvm::PointerType::getWithSamePointeeType(PtrTy, GlobalAS);
return ABIArgInfo::getDirect(LTy, 0, nullptr, false);
}
// Force copying aggregate type in kernel arguments by value when
// compiling CUDA targeting SPIR-V. This is required for the object
// copied to be valid on the device.
// This behavior follows the CUDA spec
// https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#global-function-argument-processing,
// and matches the NVPTX implementation.
if (isAggregateTypeForABI(Ty))
return getNaturalAlignIndirect(Ty, /* byval */ true);
}
return classifyArgumentType(Ty);
}

25
clang/test/CodeGenCUDASPIRV/copy-aggregate-byval.cu Normal file
View File

@ -0,0 +1,25 @@
// Tests CUDA kernel arguments get copied by value when targeting SPIR-V, even with
// destructor, copy constructor or move constructor defined by user.
// RUN: %clang -Xclang -no-opaque-pointers -emit-llvm --cuda-device-only --offload=spirv32 \
// RUN: -nocudalib -nocudainc %s -o %t.bc -c 2>&1
// RUN: llvm-dis %t.bc -o %t.ll
// RUN: FileCheck %s --input-file=%t.ll
// RUN: %clang -Xclang -no-opaque-pointers -emit-llvm --cuda-device-only --offload=spirv64 \
// RUN: -nocudalib -nocudainc %s -o %t.bc -c 2>&1
// RUN: llvm-dis %t.bc -o %t.ll
// RUN: FileCheck %s --input-file=%t.ll
class GpuData {
public:
__attribute__((host)) __attribute__((device)) GpuData(int* src) {}
__attribute__((host)) __attribute__((device)) ~GpuData() {}
__attribute__((host)) __attribute__((device)) GpuData(const GpuData& other) {}
__attribute__((host)) __attribute__((device)) GpuData(GpuData&& other) {}
};
// CHECK: define
// CHECK-SAME: spir_kernel void @_Z6kernel7GpuData(%class.GpuData* noundef byval(%class.GpuData) align
__attribute__((global)) void kernel(GpuData output) {}