Skip to content

Commit

Permalink
Fix some minor issues, still have FP16 issue.
Browse files Browse the repository at this point in the history
@liuliu
liuliu committed Aug 13, 2024
1 parent 1843cb5 commit db87da6
Showing 2 changed files with 66 additions and 15 deletions.
79 changes: 65 additions & 14 deletions bin/nnc/laplacian_test.cpp
View file
Original file line number Diff line number Diff line change
@@ -21,14 +21,15 @@ std::pair<int, int> profileProblemSize(GEMMDescriptor descriptor)
float* A = (float*)ccmalloc(sizeof(float) * problemSize * problemSize);
float* B = (float*)ccmalloc(sizeof(float) * problemSize * problemSize);
float* C = (float*)ccmalloc(sizeof(float) * problemSize * problemSize);
float* previousOperandC = (float*)ccmalloc(sizeof(float) * problemSize * problemSize);
float* bias = (float*)ccmalloc(sizeof(float) * problemSize);

// Initialize A as the 2nd-order periodic Laplacian.
int i, j;
for (i = 0; i < problemSize; i++)
{
for (j = 0; j < problemSize; j++)
A[i * problemSize + j] = 0;
for (i = 0; i < problemSize; i++)
{
const int diagonalAddress = i * problemSize + i;
A[diagonalAddress] = -2;

@@ -56,11 +57,34 @@ std::pair<int, int> profileProblemSize(GEMMDescriptor descriptor)
// Initialize C to random numbers.
for (int rowID = 0; rowID < problemSize; rowID++)
{
for (int columnID = 0; columnID < problemSize; columnID++)
{
const int address = rowID * problemSize + columnID;
previousOperandC[address] = dsfmt_genrand_open_close(&dsfmt);
}
bias[rowID] = dsfmt_genrand_open_close(&dsfmt);
}
void* A_storage = nullptr;
if (descriptor.memoryPrecisions.A == GEMMOperandPrecision::FP16)
{
A_storage = (uint16_t*)ccmalloc(sizeof(uint16_t) * problemSize * problemSize);
ccv_float_to_half_precision(A, (uint16_t*)A_storage, problemSize * problemSize);
void* t = A_storage;
A_storage = A;
A = (float*)t;
}
void* B_storage = nullptr;
if (descriptor.memoryPrecisions.B == GEMMOperandPrecision::FP16)
{
B_storage = (uint16_t*)ccmalloc(sizeof(uint16_t) * problemSize * problemSize);
ccv_float_to_half_precision(B, (uint16_t*)B_storage, problemSize * problemSize);
void* t = B_storage;
B_storage = B;
B = (float*)t;
}
void* bias_storage = nullptr;
if (descriptor.memoryPrecisions.bias == GEMMOperandPrecision::FP16)
{
bias_storage = (uint16_t*)ccmalloc(sizeof(uint16_t) * problemSize);
ccv_float_to_half_precision(bias, (uint16_t*)bias_storage, problemSize);
void* t = bias_storage;
bias_storage = bias;
bias = (float*)t;
}

// Since the Laplacian is symmetric, we swap roles of the matrices to test
@@ -88,9 +112,10 @@ std::pair<int, int> profileProblemSize(GEMMDescriptor descriptor)
// Generate the kernel.
auto pipelineValue = shaderCache.findKernel<GEMMKernel, GEMMDescriptor, GEMMKernelDescriptor>(descriptor, device.get(), dprops);
occupancy = pipelineValue->pipeline->maxTotalThreadsPerThreadgroup();
NS::SharedPtr<MTL::Buffer> bufferA = NS::TransferPtr(device->newBuffer(A, sizeof(float) * problemSize * problemSize, MTL::ResourceStorageModeShared | MTL::ResourceHazardTrackingModeTracked));
NS::SharedPtr<MTL::Buffer> bufferB = NS::TransferPtr(device->newBuffer(B, sizeof(float) * problemSize * problemSize, MTL::ResourceStorageModeShared | MTL::ResourceHazardTrackingModeTracked));
NS::SharedPtr<MTL::Buffer> bufferC = NS::TransferPtr(device->newBuffer(previousOperandC, sizeof(float) * problemSize * problemSize, MTL::ResourceStorageModeShared | MTL::ResourceHazardTrackingModeTracked));
NS::SharedPtr<MTL::Buffer> bufferA = NS::TransferPtr(device->newBuffer(A, descriptor.memoryPrecisions.A.size() * problemSize * problemSize, MTL::ResourceStorageModeShared | MTL::ResourceHazardTrackingModeTracked));
NS::SharedPtr<MTL::Buffer> bufferB = NS::TransferPtr(device->newBuffer(B, descriptor.memoryPrecisions.B.size() * problemSize * problemSize, MTL::ResourceStorageModeShared | MTL::ResourceHazardTrackingModeTracked));
NS::SharedPtr<MTL::Buffer> bufferC = NS::TransferPtr(device->newBuffer(C, descriptor.memoryPrecisions.C.size() * problemSize * problemSize, MTL::ResourceStorageModeShared | MTL::ResourceHazardTrackingModeTracked));
NS::SharedPtr<MTL::Buffer> bufferBias = NS::TransferPtr(device->newBuffer(bias, descriptor.memoryPrecisions.bias.size() * problemSize, MTL::ResourceStorageModeShared | MTL::ResourceHazardTrackingModeTracked));

// load = directAccessCondition,
// store = false
@@ -176,6 +201,14 @@ std::pair<int, int> profileProblemSize(GEMMDescriptor descriptor)
encoder->setBuffer(bufferA.get(), 0, 0);
encoder->setBuffer(bufferB.get(), 0, 1);
encoder->setBuffer(bufferC.get(), 0, 2);
encoder->useResource(bufferA.get(), MTL::ResourceUsageRead);
encoder->useResource(bufferB.get(), MTL::ResourceUsageRead);
encoder->useResource(bufferC.get(), MTL::ResourceUsageWrite);
if (descriptor.useBias)
{
encoder->setBuffer(bufferBias.get(), 0, 3);
encoder->useResource(bufferBias.get(), MTL::ResourceUsageRead);
}
for (int j = 0; j < duplicatedCommandCount; j++)
{
auto ceilDivide =
@@ -253,6 +286,18 @@ std::pair<int, int> profileProblemSize(GEMMDescriptor descriptor)
}
// Check the results.
int errorCount = 0;
if (A_storage != nullptr)
{
void* t = A_storage;
A_storage = A;
A = (float*)t;
}
if (B_storage != nullptr)
{
void* t = B_storage;
B_storage = B;
B = (float*)t;
}
for (int m = 0; m < problemSize; m++)
{
for (int n = 0; n < problemSize; n++)
@@ -308,7 +353,13 @@ std::pair<int, int> profileProblemSize(GEMMDescriptor descriptor)
ccfree(A);
ccfree(B);
ccfree(C);
ccfree(previousOperandC);
ccfree(bias);
if (A_storage != nullptr)
ccfree(A_storage);
if (B_storage != nullptr)
ccfree(B_storage);
if (bias_storage != nullptr)
ccfree(bias_storage);
return std::make_pair(maxGFLOPS, occupancy);
}

@@ -326,7 +377,7 @@ void runTest(TestDescriptor descriptor)
unsigned int n = (unsigned int)descriptor.problemSize;
gemmDesc.matrixDimensions = simd::uint3 { n, n, n };
gemmDesc.memoryPrecisions = {
.A = precision, .B = precision, .C = precision
.A = precision, .B = precision, .C = precision, .bias = precision
};
gemmDesc.transposeState = simd::uchar3 { descriptor.transposeState[0], descriptor.transposeState[1] };
gemmDesc.useBias = false;
@@ -403,9 +454,9 @@ int main(int argc, char** argv)
for (int j = 0; j < sizeof(transposeStates) / (sizeof(bool) * 2); j++)
{
TestDescriptor testDescriptor = TestDescriptor();
testDescriptor.precision = GEMMOperandPrecision::FP32;
testDescriptor.precision = GEMMOperandPrecision::FP16;
testDescriptor.problemSize = problemSize;
testDescriptor.transposeState[2] = transposeStates[j * 2];
testDescriptor.transposeState[0] = transposeStates[j * 2];
testDescriptor.transposeState[1] = transposeStates[j * 2 + 1];
runTest(testDescriptor);
}
2 changes: 1 addition & 1 deletion lib/nnc/mfa/v2/GEMMKernelDescriptor.hpp
View file
Original file line number Diff line number Diff line change
@@ -203,7 +203,7 @@ struct GEMMKernelDescriptor {
/// Mapping from the Swift implementation:
/// - A -> transposeState[0]
/// - B -> transposeState[1]
/// - bias -> transposeState[3]
/// - bias -> transposeState[2]
simd::uchar3 transposeState;

/// Required. Whether it contains the bias.

0 comments on commit db87da6

Please sign in to comment.