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
119 lines
4.0 KiB
C++
119 lines
4.0 KiB
C++
// Copyright Advanced Micro Devices, Inc., or its affiliates.
|
|
// SPDX-License-Identifier: MIT
|
|
|
|
#include <iterator>
|
|
|
|
#include <rocRoller/KernelGraph/ControlGraph/LastRWTracer.hpp>
|
|
#include <rocRoller/KernelGraph/KernelGraph.hpp>
|
|
#include <rocRoller/KernelGraph/Transforms/InlineIncrements.hpp>
|
|
#include <rocRoller/KernelGraph/Utils.hpp>
|
|
|
|
namespace rocRoller
|
|
{
|
|
namespace KernelGraph
|
|
{
|
|
using namespace ControlGraph;
|
|
using namespace CoordinateGraph;
|
|
|
|
using GD = rocRoller::Graph::Direction;
|
|
|
|
struct InlineIncrementStage
|
|
{
|
|
int assign;
|
|
int increment;
|
|
int forLoop;
|
|
std::set<int> lastRWOps;
|
|
};
|
|
|
|
struct InlineIncrementer
|
|
{
|
|
std::vector<InlineIncrementStage> inlineIncrement;
|
|
|
|
void stageForLoop(KernelGraph const& graph, int forLoop)
|
|
{
|
|
auto tracer = LastRWTracer(graph, forLoop, true);
|
|
|
|
auto rw = tracer.lastRWLocations();
|
|
|
|
for(auto increment : graph.control.getNeighbours<GD::Downstream>(forLoop))
|
|
{
|
|
auto maybeIncrement = graph.control.get<ForLoopIncrement>(increment);
|
|
if(!maybeIncrement)
|
|
continue;
|
|
|
|
for(auto assign : graph.control.getNeighbours<GD::Downstream>(increment))
|
|
{
|
|
auto maybeAssign = graph.control.get<Assign>(assign);
|
|
AssertFatal(maybeAssign, "Invalid ForLoopIncrement operation.");
|
|
|
|
auto dstTag = graph.mapper.get(assign, NaryArgument::DEST);
|
|
|
|
// Include operations that modify the
|
|
// increment.
|
|
auto ops = rw[dstTag];
|
|
|
|
// Include operations that modify any
|
|
// coordinates connected by a DataFlow edge.
|
|
if(graph.coordinates.getElementType(dstTag) == Graph::ElementType::Node)
|
|
{
|
|
for(auto df :
|
|
graph.coordinates.getOutputNodeIndices<DataFlowEdge>(dstTag))
|
|
{
|
|
std::copy(rw[df].cbegin(),
|
|
rw[df].cend(),
|
|
std::inserter(ops, ops.begin()));
|
|
}
|
|
}
|
|
|
|
inlineIncrement.push_back({assign, increment, forLoop, ops});
|
|
}
|
|
}
|
|
}
|
|
|
|
void stage(KernelGraph const& graph)
|
|
{
|
|
for(auto forLoop : graph.control.getNodes<ForLoopOp>())
|
|
{
|
|
stageForLoop(graph, forLoop);
|
|
}
|
|
}
|
|
|
|
void commit(KernelGraph& graph)
|
|
{
|
|
// Commit staged changes
|
|
for(auto const& s : inlineIncrement)
|
|
{
|
|
graph.control.deleteElement(s.increment);
|
|
if(s.lastRWOps.empty())
|
|
{
|
|
graph.control.addElement(Body(), {s.forLoop}, {s.assign});
|
|
}
|
|
else
|
|
{
|
|
for(auto op : s.lastRWOps)
|
|
{
|
|
graph.control.addElement(Sequence(), {op}, {s.assign});
|
|
}
|
|
}
|
|
}
|
|
|
|
// Sanity checks
|
|
for(auto const& s : inlineIncrement)
|
|
{
|
|
auto forLoop = findContainingOperation<ForLoopOp>(s.assign, graph);
|
|
AssertFatal(forLoop.value_or(-1) == s.forLoop);
|
|
}
|
|
}
|
|
};
|
|
|
|
KernelGraph InlineIncrements::apply(KernelGraph const& original)
|
|
{
|
|
auto graph = original;
|
|
auto transform = InlineIncrementer();
|
|
transform.stage(graph);
|
|
transform.commit(graph);
|
|
return graph;
|
|
}
|
|
}
|
|
}
|