Skip to content

[SYCL] Specialize atomic fetch_min/fetch_max for FP types #3297

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 4, 2021
Merged

[SYCL] Specialize atomic fetch_min/fetch_max for FP types #3297

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 23 additions & 4 deletions sycl/include/CL/__spirv/spirv_ops.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,9 +10,9 @@
#include <CL/__spirv/spirv_types.hpp>
#include <CL/sycl/detail/defines.hpp>
#include <CL/sycl/detail/export.hpp>
#include <CL/sycl/detail/stl_type_traits.hpp>
#include <cstddef>
#include <cstdint>
#include <type_traits>

// Convergent attribute
#ifdef __SYCL_DEVICE_ONLY__
Expand Down Expand Up @@ -91,6 +91,10 @@ extern SYCL_EXTERNAL TempRetT __spirv_ImageSampleExplicitLod(SampledType,
extern SYCL_EXTERNAL Type __spirv_AtomicUMin( \
AS Type *P, __spv::Scope::Flag S, __spv::MemorySemanticsMask::Flag O, \
Type V);
#define __SPIRV_ATOMIC_FMIN(AS, Type) \
extern SYCL_EXTERNAL Type __spirv_AtomicFMinEXT( \
AS Type *P, __spv::Scope::Flag S, __spv::MemorySemanticsMask::Flag O, \
Type V);
#define __SPIRV_ATOMIC_SMAX(AS, Type) \
extern SYCL_EXTERNAL Type __spirv_AtomicSMax( \
AS Type *P, __spv::Scope::Flag S, __spv::MemorySemanticsMask::Flag O, \
Expand All @@ -99,6 +103,10 @@ extern SYCL_EXTERNAL TempRetT __spirv_ImageSampleExplicitLod(SampledType,
extern SYCL_EXTERNAL Type __spirv_AtomicUMax( \
AS Type *P, __spv::Scope::Flag S, __spv::MemorySemanticsMask::Flag O, \
Type V);
#define __SPIRV_ATOMIC_FMAX(AS, Type) \
extern SYCL_EXTERNAL Type __spirv_AtomicFMaxEXT( \
AS Type *P, __spv::Scope::Flag S, __spv::MemorySemanticsMask::Flag O, \
Type V);
#define __SPIRV_ATOMIC_AND(AS, Type) \
extern SYCL_EXTERNAL Type __spirv_AtomicAnd( \
AS Type *P, __spv::Scope::Flag S, __spv::MemorySemanticsMask::Flag O, \
Expand All @@ -114,6 +122,8 @@ extern SYCL_EXTERNAL TempRetT __spirv_ImageSampleExplicitLod(SampledType,

#define __SPIRV_ATOMIC_FLOAT(AS, Type) \
__SPIRV_ATOMIC_FADD(AS, Type) \
__SPIRV_ATOMIC_FMIN(AS, Type) \
__SPIRV_ATOMIC_FMAX(AS, Type) \
__SPIRV_ATOMIC_LOAD(AS, Type) \
__SPIRV_ATOMIC_STORE(AS, Type) \
__SPIRV_ATOMIC_EXCHANGE(AS, Type)
Expand All @@ -138,21 +148,30 @@ extern SYCL_EXTERNAL TempRetT __spirv_ImageSampleExplicitLod(SampledType,
__SPIRV_ATOMIC_UMAX(AS, Type)

// Helper atomic operations which select correct signed/unsigned version
// of atomic min/max based on the signed-ness of the type
// of atomic min/max based on the type
#define __SPIRV_ATOMIC_MINMAX(AS, Op) \
template <typename T> \
typename std::enable_if<std::is_signed<T>::value, T>::type \
typename cl::sycl::detail::enable_if_t< \
std::is_integral<T>::value && std::is_signed<T>::value, T> \
__spirv_Atomic##Op(AS T *Ptr, __spv::Scope::Flag Memory, \
__spv::MemorySemanticsMask::Flag Semantics, \
T Value) { \
return __spirv_AtomicS##Op(Ptr, Memory, Semantics, Value); \
} \
template <typename T> \
typename std::enable_if<!std::is_signed<T>::value, T>::type \
typename cl::sycl::detail::enable_if_t< \
std::is_integral<T>::value && !std::is_signed<T>::value, T> \
__spirv_Atomic##Op(AS T *Ptr, __spv::Scope::Flag Memory, \
__spv::MemorySemanticsMask::Flag Semantics, \
T Value) { \
return __spirv_AtomicU##Op(Ptr, Memory, Semantics, Value); \
} \
template <typename T> \
typename cl::sycl::detail::enable_if_t<std::is_floating_point<T>::value, T> \
__spirv_Atomic##Op(AS T *Ptr, __spv::Scope::Flag Memory, \
__spv::MemorySemanticsMask::Flag Semantics, \
T Value) { \
return __spirv_AtomicF##Op##EXT(Ptr, Memory, Semantics, Value); \
}

#define __SPIRV_ATOMICS(macro, Arg) \
Expand Down
13 changes: 12 additions & 1 deletion sycl/include/CL/sycl/ONEAPI/atomic_ref.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -413,7 +413,6 @@ class atomic_ref_impl<T, DefaultOrder, DefaultScope, AddressSpace,
};

// Partial specialization for floating-point types
// TODO: Leverage floating-point SPIR-V atomics instead of emulation
template <typename T, memory_order DefaultOrder, memory_scope DefaultScope,
access::address_space AddressSpace>
class atomic_ref_impl<
Expand Down Expand Up @@ -486,22 +485,34 @@ class atomic_ref_impl<

T fetch_min(T operand, memory_order order = default_read_modify_write_order,
memory_scope scope = default_scope) const noexcept {
// TODO: Remove the "native atomics" macro check once implemented for all
// backends
#if defined(__SYCL_DEVICE_ONLY__) && defined(SYCL_USE_NATIVE_FP_ATOMICS)
return detail::spirv::AtomicMin(ptr, scope, order, operand);
#else
auto load_order = detail::getLoadOrder(order);
T old = load(load_order, scope);
while (operand < old &&
!compare_exchange_weak(old, operand, order, scope)) {
}
return old;
#endif
}

T fetch_max(T operand, memory_order order = default_read_modify_write_order,
memory_scope scope = default_scope) const noexcept {
// TODO: Remove the "native atomics" macro check once implemented for all
// backends
#if defined(__SYCL_DEVICE_ONLY__) && defined(SYCL_USE_NATIVE_FP_ATOMICS)
return detail::spirv::AtomicMax(ptr, scope, order, operand);
#else
auto load_order = detail::getLoadOrder(order);
T old = load(load_order, scope);
while (operand > old &&
!compare_exchange_weak(old, operand, order, scope)) {
}
return old;
#endif
}

private:
Expand Down
20 changes: 20 additions & 0 deletions sycl/include/CL/sycl/detail/spirv.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -435,6 +435,16 @@ AtomicMin(multi_ptr<T, AddressSpace> MPtr, ONEAPI::memory_scope Scope,
return __spirv_AtomicMin(Ptr, SPIRVScope, SPIRVOrder, Value);
}

template <typename T, access::address_space AddressSpace>
inline typename detail::enable_if_t<std::is_floating_point<T>::value, T>
AtomicMin(multi_ptr<T, AddressSpace> MPtr, ONEAPI::memory_scope Scope,
ONEAPI::memory_order Order, T Value) {
auto *Ptr = MPtr.get();
auto SPIRVOrder = getMemorySemanticsMask(Order);
auto SPIRVScope = getScope(Scope);
return __spirv_AtomicMin(Ptr, SPIRVScope, SPIRVOrder, Value);
}

template <typename T, access::address_space AddressSpace>
inline typename detail::enable_if_t<std::is_integral<T>::value, T>
AtomicMax(multi_ptr<T, AddressSpace> MPtr, ONEAPI::memory_scope Scope,
Expand All @@ -445,6 +455,16 @@ AtomicMax(multi_ptr<T, AddressSpace> MPtr, ONEAPI::memory_scope Scope,
return __spirv_AtomicMax(Ptr, SPIRVScope, SPIRVOrder, Value);
}

template <typename T, access::address_space AddressSpace>
inline typename detail::enable_if_t<std::is_floating_point<T>::value, T>
AtomicMax(multi_ptr<T, AddressSpace> MPtr, ONEAPI::memory_scope Scope,
ONEAPI::memory_order Order, T Value) {
auto *Ptr = MPtr.get();
auto SPIRVOrder = getMemorySemanticsMask(Order);
auto SPIRVScope = getScope(Scope);
return __spirv_AtomicMax(Ptr, SPIRVScope, SPIRVOrder, Value);
}

// Native shuffles map directly to a shuffle intrinsic:
// - The Intel SPIR-V extension natively supports all arithmetic types
// - The CUDA shfl intrinsics do not support vectors, and we use the _i32
Expand Down
30 changes: 17 additions & 13 deletions sycl/test/atomic_ref/max.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,7 @@
// RUN: %clangxx -fsycl -fsycl-unnamed-lambda -DSYCL_USE_NATIVE_FP_ATOMICS \
// RUN: -fsycl-device-only -S %s -o - | FileCheck %s --check-prefix=CHECK-LLVM
// RUN: %clangxx -fsycl -fsycl-unnamed-lambda -fsycl-device-only -S %s -o - \
// RUN: | FileCheck %s --check-prefix=CHECK-LLVM
// RUN: | FileCheck %s --check-prefix=CHECK-LLVM-EMU
// RUN: %clangxx -fsycl -fsycl-unnamed-lambda -fsycl-targets=%sycl_triple %s -o %t.out
// RUN: %RUN_ON_HOST %t.out

Expand Down Expand Up @@ -83,19 +85,21 @@ int main() {
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicUMax
// CHECK-LLVM-SAME: (i64 addrspace(1)*, i32, i32, i64)
max_test<unsigned long long>(q, N);
// CHECK-LLVM: declare dso_local spir_func i32
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicLoad
// CHECK-LLVM-SAME: (i32 addrspace(1)*, i32, i32)
// CHECK-LLVM: declare dso_local spir_func i32
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicCompareExchange
// CHECK-LLVM-SAME: (i32 addrspace(1)*, i32, i32, i32, i32, i32)
// CHECK-LLVM: declare dso_local spir_func float
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicFMaxEXT
// CHECK-LLVM-SAME: (float addrspace(1)*, i32, i32, float)
// CHECK-LLVM-EMU: declare {{.*}} i32 @{{.*}}__spirv_AtomicLoad
// CHECK-LLVM-EMU-SAME: (i32 addrspace(1)*, i32, i32)
// CHECK-LLVM-EMU: declare {{.*}} i32 @{{.*}}__spirv_AtomicCompareExchange
// CHECK-LLVM-EMU-SAME: (i32 addrspace(1)*, i32, i32, i32, i32, i32)
max_test<float>(q, N);
// CHECK-LLVM: declare dso_local spir_func i64
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicLoad
// CHECK-LLVM-SAME: (i64 addrspace(1)*, i32, i32)
// CHECK-LLVM: declare dso_local spir_func i64
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicCompareExchange
// CHECK-LLVM-SAME: (i64 addrspace(1)*, i32, i32, i32, i64, i64)
// CHECK-LLVM: declare dso_local spir_func double
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicFMaxEXT
// CHECK-LLVM-SAME: (double addrspace(1)*, i32, i32, double)
// CHECK-LLVM-EMU: declare {{.*}} i64 @{{.*}}__spirv_AtomicLoad
// CHECK-LLVM-EMU-SAME: (i64 addrspace(1)*, i32, i32)
// CHECK-LLVM-EMU: declare {{.*}} i64 @{{.*}}__spirv_AtomicCompareExchange
// CHECK-LLVM-EMU-SAME: (i64 addrspace(1)*, i32, i32, i32, i64, i64)
max_test<double>(q, N);

std::cout << "Test passed." << std::endl;
Expand Down
30 changes: 17 additions & 13 deletions sycl/test/atomic_ref/min.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,7 @@
// RUN: %clangxx -fsycl -fsycl-unnamed-lambda -DSYCL_USE_NATIVE_FP_ATOMICS \
// RUN: -fsycl-device-only -S %s -o - | FileCheck %s --check-prefix=CHECK-LLVM
// RUN: %clangxx -fsycl -fsycl-unnamed-lambda -fsycl-device-only -S %s -o - \
// RUN: | FileCheck %s --check-prefix=CHECK-LLVM
// RUN: | FileCheck %s --check-prefix=CHECK-LLVM-EMU
// RUN: %clangxx -fsycl -fsycl-unnamed-lambda -fsycl-targets=%sycl_triple %s -o %t.out
// RUN: %RUN_ON_HOST %t.out

Expand Down Expand Up @@ -81,19 +83,21 @@ int main() {
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicUMin
// CHECK-LLVM-SAME: (i64 addrspace(1)*, i32, i32, i64)
min_test<unsigned long long>(q, N);
// CHECK-LLVM: declare dso_local spir_func i32
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicLoad
// CHECK-LLVM-SAME: (i32 addrspace(1)*, i32, i32)
// CHECK-LLVM: declare dso_local spir_func i32
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicCompareExchange
// CHECK-LLVM-SAME: (i32 addrspace(1)*, i32, i32, i32, i32, i32)
// CHECK-LLVM: declare dso_local spir_func float
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicFMinEXT
// CHECK-LLVM-SAME: (float addrspace(1)*, i32, i32, float)
// CHECK-LLVM-EMU: declare {{.*}} i32 @{{.*}}__spirv_AtomicLoad
// CHECK-LLVM-EMU-SAME: (i32 addrspace(1)*, i32, i32)
// CHECK-LLVM-EMU: declare {{.*}} i32 @{{.*}}__spirv_AtomicCompareExchange
// CHECK-LLVM-EMU-SAME: (i32 addrspace(1)*, i32, i32, i32, i32, i32)
min_test<float>(q, N);
// CHECK-LLVM: declare dso_local spir_func i64
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicLoad
// CHECK-LLVM-SAME: (i64 addrspace(1)*, i32, i32)
// CHECK-LLVM: declare dso_local spir_func i64
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicCompareExchange
// CHECK-LLVM-SAME: (i64 addrspace(1)*, i32, i32, i32, i64, i64)
// CHECK-LLVM: declare dso_local spir_func double
// CHECK-LLVM-SAME: @_Z{{[0-9]+}}__spirv_AtomicFMinEXT
// CHECK-LLVM-SAME: (double addrspace(1)*, i32, i32, double)
// CHECK-LLVM-EMU: declare {{.*}} i64 @{{.*}}__spirv_AtomicLoad
// CHECK-LLVM-EMU-SAME: (i64 addrspace(1)*, i32, i32)
// CHECK-LLVM-EMU: declare {{.*}} i64 @{{.*}}__spirv_AtomicCompareExchange
// CHECK-LLVM-EMU-SAME: (i64 addrspace(1)*, i32, i32, i32, i64, i64)
min_test<double>(q, N);

std::cout << "Test passed." << std::endl;
Expand Down