MachineTom
4a31affb76
* SWDEV-550626 - Refactor atomics header and tests
1. Introduce __HIP_ATOMIC_BACKWARD_COMPAT.
By default we define __HIP_ATOMIC_BACKWARD_COMPAT=1 to
let hip atomic functions maintain old assumptions. if
users want to adopt the new behavior, that is , by default
assume no-fine-grained no-remote-memory, then they can
define __HIP_ATOMIC_BACKWARD_COMPAT=0 and get the new
behaviour.
2. Use __HIP_ATOMIC_BACKWARD_COMPAT_MEMORY to replace
original __HIP_FINE_GRAINED_MEMORY in atomic header.
And apply __HIP_FINE_GRAINED_MEMORY onto all
atomicXXX_system() functions to prevent failure on memory
allocated by hipHostMalloc().
3. Replace HIP_TEST_FINE_GRAINED_MEMORY with
HIP_TEST_ATOMIC_BACKWARD_COMPAT_MEMORY in hip-tests.
4. Fix negative test errors.
Fix managed memory test error on memory order.
some other minor changes.
As a result all originally disabled tests are enabled.
5. Add more atomics tests in some cases.
6. Reduce test time in each case.
Reduce iteration number to 1 for tests that cost too much time.
8. Put common codes into hip_test_common.hh
130 wiersze
4.5 KiB
C++
130 wiersze
4.5 KiB
C++
/*
|
|
Copyright (c) 2023 Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
Permission is hereby granted, free of charge, to any person obtaining a copy
|
|
of this software and associated documentation files (the "Software"), to deal
|
|
in the Software without restriction, including without limitation the rights
|
|
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
|
|
copies of the Software, and to permit persons to whom the Software is
|
|
furnished to do so, subject to the following conditions:
|
|
|
|
The above copyright notice and this permission notice shall be included in
|
|
all copies or substantial portions of the Software.
|
|
|
|
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
|
|
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
|
|
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
|
|
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
|
|
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
|
|
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
|
|
THE SOFTWARE.
|
|
*/
|
|
|
|
#include "min_max_common.hh"
|
|
|
|
#include <hip_test_common.hh>
|
|
|
|
/**
|
|
* @addtogroup atomicMax_system atomicMax_system
|
|
* @{
|
|
* @ingroup AtomicsTest
|
|
* `atomicMax_system(TestType* address, TestType* val)` -
|
|
* performs system-wide atomic maximum between address and val, returns old value.
|
|
*/
|
|
|
|
/**
|
|
* Test Description
|
|
* ------------------------
|
|
* - Performs atomicMax_system from multiple threads on the same address.
|
|
* - Uses multiple devices and launches multiple kernels.
|
|
* Test source
|
|
* ------------------------
|
|
* - unit/atomics/atomicMax_system.cc
|
|
* Test requirements
|
|
* ------------------------
|
|
* - Multi-device
|
|
* - HIP_VERSION >= 5.2
|
|
*/
|
|
#if HT_AMD
|
|
TEMPLATE_TEST_CASE("Unit_atomicMax_system_Positive_Peer_GPUs_Same_Address", "", int, unsigned int,
|
|
unsigned long, unsigned long long, float, double) {
|
|
#else
|
|
TEMPLATE_TEST_CASE("Unit_atomicMax_system_Positive_Peer_GPUs_Same_Address", "", int, unsigned int,
|
|
unsigned long, unsigned long long) {
|
|
#endif
|
|
for (auto current = 0; current < 1; ++current) {
|
|
DYNAMIC_SECTION("Same address " << current) {
|
|
MinMax::MultipleDeviceMultipleKernelTest<TestType, MinMax::AtomicOperation::kMaxSystem>(
|
|
2, 2, 1, sizeof(TestType));
|
|
}
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Test Description
|
|
* ------------------------
|
|
* - Performs atomicMax_system from multiple threads on adjacent addresses.
|
|
* - Uses multiple devices and launches multiple kernels.
|
|
* Test source
|
|
* ------------------------
|
|
* - unit/atomics/atomicMax_system.cc
|
|
* Test requirements
|
|
* ------------------------
|
|
* - Multi-device
|
|
* - HIP_VERSION >= 5.2
|
|
*/
|
|
#if HT_AMD
|
|
TEMPLATE_TEST_CASE("Unit_atomicMax_system_Positive_Peer_GPUs_Adjacent_Addresses", "", int,
|
|
unsigned int, unsigned long, unsigned long long, float, double) {
|
|
#else
|
|
TEMPLATE_TEST_CASE("Unit_atomicMax_system_Positive_Peer_GPUs_Adjacent_Addresses", "", int,
|
|
unsigned int, unsigned long, unsigned long long) {
|
|
#endif
|
|
int warp_size = 0;
|
|
HIP_CHECK(hipDeviceGetAttribute(&warp_size, hipDeviceAttributeWarpSize, 0));
|
|
|
|
for (auto current = 0; current < 1; ++current) {
|
|
DYNAMIC_SECTION("Adjacent address " << current) {
|
|
MinMax::MultipleDeviceMultipleKernelTest<TestType, MinMax::AtomicOperation::kMaxSystem>(
|
|
2, 2, warp_size, sizeof(TestType));
|
|
}
|
|
}
|
|
}
|
|
|
|
/**
|
|
* Test Description
|
|
* ------------------------
|
|
* - Performs atomicMax_system from multiple threads on scaterred addresses.
|
|
* - Uses multiple devices and launches multiple kernels.
|
|
* Test source
|
|
* ------------------------
|
|
* - unit/atomics/atomicMax_system.cc
|
|
* Test requirements
|
|
* ------------------------
|
|
* - Multi-device
|
|
* - HIP_VERSION >= 5.2
|
|
*/
|
|
#if HT_AMD
|
|
TEMPLATE_TEST_CASE("Unit_atomicMax_system_Positive_Peer_GPUs_Scattered_Addresses", "", int,
|
|
unsigned int, unsigned long, unsigned long long, float, double) {
|
|
#else
|
|
TEMPLATE_TEST_CASE("Unit_atomicMax_system_Positive_Peer_GPUs_Scattered_Addresses", "", int,
|
|
unsigned int, unsigned long, unsigned long long) {
|
|
#endif
|
|
int warp_size = 0;
|
|
HIP_CHECK(hipDeviceGetAttribute(&warp_size, hipDeviceAttributeWarpSize, 0));
|
|
const auto cache_line_size = 128u;
|
|
|
|
for (auto current = 0; current < 1; ++current) {
|
|
DYNAMIC_SECTION("Scattered address " << current) {
|
|
MinMax::MultipleDeviceMultipleKernelTest<TestType, MinMax::AtomicOperation::kMaxSystem>(
|
|
2, 2, warp_size, cache_line_size);
|
|
}
|
|
}
|
|
}
|
|
|
|
/**
|
|
* End doxygen group AtomicsTest.
|
|
* @}
|
|
*/
|