Files
TheRock/rocm-systems/projects/rccl/test/DdaIpcEligibilityTests.cpp
T
derek.holloway 7502bf64a4
CodeQL / Analyze (actions) (push) Canceled after 0s
CodeQL / Analyze (python) (push) Canceled after 0s
pre-commit / pre-commit (push) Canceled after 0s
Multi-Arch CI / setup (push) Canceled after 0s
Publish build_manylinux_x86_64 images / publish_build_manylinux_x86_64 (push) Canceled after 0s
Setup (Multi-Arch) / setup (push) Canceled after 0s
Multi-Arch CI / Linux::${{ fromJSON(needs.setup.outputs.linux_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubi10 image / publish_no_rocm_image_ubi10 (push) Canceled after 0s
Multi-Arch CI / Windows::${{ fromJSON(needs.setup.outputs.windows_build_config || '{}').build_variant_label || 'skip' }} (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Multi-Arch CI / Manifest Diff (push) Canceled after 0s
TheRock Manifest Diff Report / Generate Manifest Diff Report (push) Canceled after 0s
Multi-Arch CI / CI Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_ocl_rt images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04_rocgdb images / publish_no_rocm_image_ubuntu24_04_rocgdb (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_ocl_rt (push) Canceled after 0s
Publish TheRock Docker image / build-and-push-image (push) Canceled after 0s
Unit Tests / Unit Tests :: ubuntu-24.04 (push) Canceled after 0s
Unit Tests / Unit Tests :: windows-2022 (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_media (push) Canceled after 0s
Unit Tests / Unit Tests Summary (push) Canceled after 0s
Publish no_rocm_image_ubuntu24_04 images / publish_no_rocm_image_ubuntu24_04_openmpi (push) Canceled after 0s
Flatten subprojects into a single base monorepo
2026-09-02 16:34:47 -07:00

212 lines
6.4 KiB
C++

/*************************************************************************
* Copyright (c) 2026 Advanced Micro Devices, Inc. All rights reserved.
*
* See LICENSE.txt for license information
************************************************************************/
#include "common/DdaAlltoAllTestHelpers.hpp"
#include "common/DdaIpcTestHelpers.hpp"
#include "dda_all_gather.h"
#include "dda_alltoall.h"
#include "dda_reduce_scatter.h"
#include "gtest/gtest.h"
#include "dda_init_detail.h"
namespace RcclUnitTesting
{
class DdaIpcEligibilityTest : public ::testing::Test
{
protected:
DdaIpcMockComm mockComm_;
void* sendbuff_{reinterpret_cast<void*>(0x10)};
void* recvbuff_{reinterpret_cast<void*>(0x20)};
};
TEST_F(DdaIpcEligibilityTest, AllGather_NullComm)
{
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
nullptr, sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_MissingBootstrap)
{
mockComm_.comm.bootstrap = nullptr;
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_MissingIpcResources)
{
mockComm_.setIpcResourcesPresent(false);
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_ZeroCount)
{
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 0, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_MultiNode)
{
mockComm_.comm.nNodes = 2;
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_WrongRankCount)
{
mockComm_.comm.nRanks = 4;
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_UnsupportedDatatype)
{
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclInt32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_ScratchTooSmall)
{
mockComm_.comm.ddaScratchBytes = 8;
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_UnalignedCount)
{
EXPECT_FALSE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 3, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_EligibleFloat32)
{
EXPECT_TRUE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllGather_EligibleFloat16)
{
EXPECT_TRUE(ncclAllGatherDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 8, ncclFloat16));
}
TEST_F(DdaIpcEligibilityTest, AllGather_InvalidDatatypeDispatch)
{
EXPECT_EQ(ncclAllGatherDdaIpc(
sendbuff_, recvbuff_, 4, ncclInt32, mockComm_.get(), nullptr),
ncclInvalidArgument);
}
TEST_F(DdaIpcEligibilityTest, AllToAll_EligibleFloat32)
{
EXPECT_TRUE(ncclAllToAllDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllToAll_ScratchTooSmallForTotal)
{
mockComm_.comm.ddaScratchBytes = 64;
EXPECT_FALSE(ncclAllToAllDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllToAll_UnalignedCount)
{
EXPECT_FALSE(ncclAllToAllDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 3, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllToAll_InvalidDatatypeDispatch)
{
EXPECT_EQ(ncclAllToAllDdaIpc(
sendbuff_, recvbuff_, 4, ncclInt32, mockComm_.get(), nullptr),
ncclInvalidArgument);
}
TEST_F(DdaIpcEligibilityTest, AllToAll_CountAt4MbTotal_Eligible)
{
EXPECT_TRUE(ncclAllToAllDdaIpcEligible(
mockComm_.get(),
sendbuff_,
recvbuff_,
kAlltoAllFloat32CountAt4MbThreshold,
ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllToAll_StagingBytesAtThresholdFitsScratch)
{
const size_t stagingBytes = testAlltoAllDdaIpcStagingBytes(
kAlltoAllFloat32CountAt4MbThreshold,
mockComm_.comm.nRanks,
sizeof(float));
EXPECT_EQ(stagingBytes, kDdaAlltoAllGfx950ThresholdBytes);
EXPECT_LE(stagingBytes, mockComm_.comm.ddaScratchBytes);
}
TEST_F(DdaIpcEligibilityTest, AllToAll_StagingBytesOneCountOverThresholdStillEligible)
{
// Eligibility is independent of the 4 MiB dispatch cap enforced in collectives.cc.
const size_t count = kAlltoAllFloat32CountAt4MbThreshold + 1;
const size_t stagingBytes = testAlltoAllDdaIpcStagingBytes(
count, mockComm_.comm.nRanks, sizeof(float));
EXPECT_GT(stagingBytes, kDdaAlltoAllGfx950ThresholdBytes);
EXPECT_TRUE(ncclAllToAllDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, count, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllToAll_MissingBootstrap)
{
mockComm_.comm.bootstrap = nullptr;
EXPECT_FALSE(ncclAllToAllDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, AllToAll_ZeroCount)
{
EXPECT_FALSE(ncclAllToAllDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 0, ncclFloat32));
}
TEST_F(DdaIpcEligibilityTest, ReduceScatter_EligibleFloat32)
{
EXPECT_TRUE(ncclReduceScatterDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32, ncclSum));
}
TEST_F(DdaIpcEligibilityTest, ReduceScatter_UnsupportedOp)
{
EXPECT_FALSE(ncclReduceScatterDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 4, ncclFloat32, ncclMax));
}
TEST_F(DdaIpcEligibilityTest, ReduceScatter_NonAlignedRecvcount)
{
EXPECT_FALSE(ncclReduceScatterDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 3, ncclFloat32, ncclSum));
}
TEST_F(DdaIpcEligibilityTest, ReduceScatter_PerRankUnaligned)
{
EXPECT_FALSE(ncclReduceScatterDdaIpcEligible(
mockComm_.get(), sendbuff_, recvbuff_, 2, ncclFloat32, ncclSum));
}
TEST_F(DdaIpcEligibilityTest, ReduceScatter_InvalidDatatypeDispatch)
{
EXPECT_EQ(ncclReduceScatterDdaIpc(sendbuff_,
recvbuff_,
4,
ncclInt32,
ncclSum,
mockComm_.get(),
nullptr),
ncclInvalidArgument);
}
} // namespace RcclUnitTesting