1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
|
/*******************************************************************************
*
* MIT License
*
* Copyright (C) 2019-2022 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 <cstring>
#include <Tensile/AMDGPU.hpp>
#include <Tensile/hip/HipHardware.hpp>
#include <Tensile/hip/HipUtils.hpp>
namespace Tensile
{
namespace hip
{
HipAMDGPU::HipAMDGPU(hipDeviceProp_t const& prop)
: AMDGPU(static_cast<AMDGPU::Processor>(prop.gcnArch),
prop.multiProcessorCount,
std::string(prop.name))
, properties(prop)
{
}
std::string HipAMDGPU::archName() const
{
return properties.gcnArchName;
}
std::shared_ptr<Hardware> GetCurrentDevice()
{
int deviceId = 0;
HIP_CHECK_EXC(hipGetDevice(&deviceId));
return GetDevice(deviceId);
}
std::shared_ptr<Hardware> GetDevice(int deviceId)
{
hipDeviceProp_t prop;
HIP_CHECK_EXC(hipGetDeviceProperties(&prop, deviceId));
// coerce to compatible arch
switch(prop.gcnArch)
{
case 902:
case 909:
case 912:
prop.gcnArch = 900;
std::strcpy(prop.gcnArchName, "gfx900");
break;
case 1011:
case 1012:
case 1013:
prop.gcnArch = 1010;
std::strcpy(prop.gcnArchName, "gfx1010");
break;
case 1031:
case 1032:
case 1033:
case 1034:
case 1035:
case 1036:
prop.gcnArch = 1030;
std::strcpy(prop.gcnArchName, "gfx1030");
break;
}
#if HIP_VERSION >= 50220730
int hip_version;
HIP_CHECK_EXC(hipRuntimeGetVersion(&hip_version));
if(hip_version >= 50220730)
{
HIP_CHECK_EXC(hipDeviceGetAttribute(&prop.multiProcessorCount,
hipDeviceAttributePhysicalMultiProcessorCount,
deviceId));
}
#endif
return GetDevice(prop);
}
std::shared_ptr<Hardware> GetDevice(hipDeviceProp_t const& prop)
{
return std::make_shared<HipAMDGPU>(prop);
}
} // namespace hip
} // namespace Tensile
|