[Parallel_libs-commits] [parallel-libs] r281424 - [SE] Pack global dev handle addresses
Jason Henline via Parallel_libs-commits
parallel_libs-commits at lists.llvm.org
Tue Sep 13 16:59:10 PDT 2016
Author: jhen
Date: Tue Sep 13 18:59:10 2016
New Revision: 281424
URL: http://llvm.org/viewvc/llvm-project?rev=281424&view=rev
Log:
[SE] Pack global dev handle addresses
Summary:
We were packing global device memory handles in
`PackedKernelArgumentArray`, but as I was implementing the CUDA
platform, I realized that CUDA wants the address of the handle, not the
handle itself. So this patch switches to packing the address of the
handle.
Reviewers: jlebar
Subscribers: jprice, jlebar, parallel_libs-commits
Differential Revision: https://reviews.llvm.org/D24528
Modified:
parallel-libs/trunk/streamexecutor/examples/HostSaxpy.cpp
parallel-libs/trunk/streamexecutor/include/streamexecutor/DeviceMemory.h
parallel-libs/trunk/streamexecutor/include/streamexecutor/PackedKernelArgumentArray.h
parallel-libs/trunk/streamexecutor/unittests/CoreTests/PackedKernelArgumentArrayTest.cpp
Modified: parallel-libs/trunk/streamexecutor/examples/HostSaxpy.cpp
URL: http://llvm.org/viewvc/llvm-project/parallel-libs/trunk/streamexecutor/examples/HostSaxpy.cpp?rev=281424&r1=281423&r2=281424&view=diff
==============================================================================
--- parallel-libs/trunk/streamexecutor/examples/HostSaxpy.cpp (original)
+++ parallel-libs/trunk/streamexecutor/examples/HostSaxpy.cpp Tue Sep 13 18:59:10 2016
@@ -33,8 +33,8 @@ using SaxpyKernel =
// Wrapper function converts argument addresses to arguments.
void SaxpyWrapper(const void *const *ArgumentAddresses) {
Saxpy(*static_cast<const float *>(ArgumentAddresses[0]),
- static_cast<float *>(const_cast<void *>(ArgumentAddresses[1])),
- static_cast<float *>(const_cast<void *>(ArgumentAddresses[2])),
+ *static_cast<float **>(const_cast<void *>(ArgumentAddresses[1])),
+ *static_cast<float **>(const_cast<void *>(ArgumentAddresses[2])),
*static_cast<const size_t *>(ArgumentAddresses[3]));
}
Modified: parallel-libs/trunk/streamexecutor/include/streamexecutor/DeviceMemory.h
URL: http://llvm.org/viewvc/llvm-project/parallel-libs/trunk/streamexecutor/include/streamexecutor/DeviceMemory.h?rev=281424&r1=281423&r2=281424&view=diff
==============================================================================
--- parallel-libs/trunk/streamexecutor/include/streamexecutor/DeviceMemory.h (original)
+++ parallel-libs/trunk/streamexecutor/include/streamexecutor/DeviceMemory.h Tue Sep 13 18:59:10 2016
@@ -133,6 +133,9 @@ public:
/// Returns an opaque handle to the underlying memory.
const void *getHandle() const { return Handle; }
+ /// Returns the address of the opaque handle as stored by this object.
+ const void *const *getHandleAddress() const { return &Handle; }
+
// Cannot copy because the handle must be owned by a single object.
GlobalDeviceMemoryBase(const GlobalDeviceMemoryBase &) = delete;
GlobalDeviceMemoryBase &operator=(const GlobalDeviceMemoryBase &) = delete;
Modified: parallel-libs/trunk/streamexecutor/include/streamexecutor/PackedKernelArgumentArray.h
URL: http://llvm.org/viewvc/llvm-project/parallel-libs/trunk/streamexecutor/include/streamexecutor/PackedKernelArgumentArray.h?rev=281424&r1=281423&r2=281424&view=diff
==============================================================================
--- parallel-libs/trunk/streamexecutor/include/streamexecutor/PackedKernelArgumentArray.h (original)
+++ parallel-libs/trunk/streamexecutor/include/streamexecutor/PackedKernelArgumentArray.h Tue Sep 13 18:59:10 2016
@@ -164,31 +164,10 @@ private:
Types[Index] = KernelArgumentType::VALUE;
}
- // Pack a GlobalDeviceMemoryBase argument.
- void PackOneArgument(size_t Index, const GlobalDeviceMemoryBase &Argument) {
- Addresses[Index] = Argument.getHandle();
- Sizes[Index] = sizeof(void *);
- Types[Index] = KernelArgumentType::GLOBAL_DEVICE_MEMORY;
- }
-
- // Pack a GlobalDeviceMemoryBase pointer argument.
- void PackOneArgument(size_t Index, GlobalDeviceMemoryBase *Argument) {
- Addresses[Index] = Argument->getHandle();
- Sizes[Index] = sizeof(void *);
- Types[Index] = KernelArgumentType::GLOBAL_DEVICE_MEMORY;
- }
-
- // Pack a const GlobalDeviceMemoryBase pointer argument.
- void PackOneArgument(size_t Index, const GlobalDeviceMemoryBase *Argument) {
- Addresses[Index] = Argument->getHandle();
- Sizes[Index] = sizeof(void *);
- Types[Index] = KernelArgumentType::GLOBAL_DEVICE_MEMORY;
- }
-
// Pack a GlobalDeviceMemory<T> argument.
template <typename T>
void PackOneArgument(size_t Index, const GlobalDeviceMemory<T> &Argument) {
- Addresses[Index] = Argument.getHandle();
+ Addresses[Index] = Argument.getHandleAddress();
Sizes[Index] = sizeof(void *);
Types[Index] = KernelArgumentType::GLOBAL_DEVICE_MEMORY;
}
@@ -196,7 +175,7 @@ private:
// Pack a GlobalDeviceMemory<T> pointer argument.
template <typename T>
void PackOneArgument(size_t Index, GlobalDeviceMemory<T> *Argument) {
- Addresses[Index] = Argument->getHandle();
+ Addresses[Index] = Argument->getHandleAddress();
Sizes[Index] = sizeof(void *);
Types[Index] = KernelArgumentType::GLOBAL_DEVICE_MEMORY;
}
@@ -204,7 +183,7 @@ private:
// Pack a const GlobalDeviceMemory<T> pointer argument.
template <typename T>
void PackOneArgument(size_t Index, const GlobalDeviceMemory<T> *Argument) {
- Addresses[Index] = Argument->getHandle();
+ Addresses[Index] = Argument->getHandleAddress();
Sizes[Index] = sizeof(void *);
Types[Index] = KernelArgumentType::GLOBAL_DEVICE_MEMORY;
}
Modified: parallel-libs/trunk/streamexecutor/unittests/CoreTests/PackedKernelArgumentArrayTest.cpp
URL: http://llvm.org/viewvc/llvm-project/parallel-libs/trunk/streamexecutor/unittests/CoreTests/PackedKernelArgumentArrayTest.cpp?rev=281424&r1=281423&r2=281424&view=diff
==============================================================================
--- parallel-libs/trunk/streamexecutor/unittests/CoreTests/PackedKernelArgumentArrayTest.cpp (original)
+++ parallel-libs/trunk/streamexecutor/unittests/CoreTests/PackedKernelArgumentArrayTest.cpp Tue Sep 13 18:59:10 2016
@@ -76,7 +76,7 @@ TEST_F(DeviceMemoryPackingTest, SingleVa
TEST_F(DeviceMemoryPackingTest, SingleTypedGlobal) {
auto Array = se::make_kernel_argument_pack(TypedGlobal);
- ExpectEqual(TypedGlobal.getHandle(), sizeof(void *),
+ ExpectEqual(TypedGlobal.getHandleAddress(), sizeof(void *),
Type::GLOBAL_DEVICE_MEMORY, Array, 0);
EXPECT_EQ(1u, Array.getArgumentCount());
EXPECT_EQ(0u, Array.getSharedCount());
@@ -84,7 +84,7 @@ TEST_F(DeviceMemoryPackingTest, SingleTy
TEST_F(DeviceMemoryPackingTest, SingleTypedGlobalPointer) {
auto Array = se::make_kernel_argument_pack(&TypedGlobal);
- ExpectEqual(TypedGlobal.getHandle(), sizeof(void *),
+ ExpectEqual(TypedGlobal.getHandleAddress(), sizeof(void *),
Type::GLOBAL_DEVICE_MEMORY, Array, 0);
EXPECT_EQ(1u, Array.getArgumentCount());
EXPECT_EQ(0u, Array.getSharedCount());
@@ -93,7 +93,7 @@ TEST_F(DeviceMemoryPackingTest, SingleTy
TEST_F(DeviceMemoryPackingTest, SingleConstTypedGlobalPointer) {
const se::GlobalDeviceMemory<int> *ArgumentPointer = &TypedGlobal;
auto Array = se::make_kernel_argument_pack(ArgumentPointer);
- ExpectEqual(TypedGlobal.getHandle(), sizeof(void *),
+ ExpectEqual(TypedGlobal.getHandleAddress(), sizeof(void *),
Type::GLOBAL_DEVICE_MEMORY, Array, 0);
EXPECT_EQ(1u, Array.getArgumentCount());
EXPECT_EQ(0u, Array.getSharedCount());
@@ -131,11 +131,11 @@ TEST_F(DeviceMemoryPackingTest, PackSeve
TypedGlobalPointer, TypedShared,
&TypedShared, TypedSharedPointer);
ExpectEqual(&Value, sizeof(Value), Type::VALUE, Array, 0);
- ExpectEqual(TypedGlobal.getHandle(), sizeof(void *),
+ ExpectEqual(TypedGlobal.getHandleAddress(), sizeof(void *),
Type::GLOBAL_DEVICE_MEMORY, Array, 1);
- ExpectEqual(TypedGlobal.getHandle(), sizeof(void *),
+ ExpectEqual(TypedGlobal.getHandleAddress(), sizeof(void *),
Type::GLOBAL_DEVICE_MEMORY, Array, 2);
- ExpectEqual(TypedGlobal.getHandle(), sizeof(void *),
+ ExpectEqual(TypedGlobal.getHandleAddress(), sizeof(void *),
Type::GLOBAL_DEVICE_MEMORY, Array, 3);
ExpectEqual(nullptr, TypedShared.getByteCount(), Type::SHARED_DEVICE_MEMORY,
Array, 4);
More information about the Parallel_libs-commits
mailing list