Files
pytorch/torch/csrc/jit/passes/utils/subgraph_utils.cpp
cyy c2f28d1c1d fix missing-prototypes warnings in torch_cpu (Part 4) (#100849)
This PR fixes more missing-prototypes violations in the torch_cpu source following PRs #100053, #100147 and #100245

Pull Request resolved: https://github.com/pytorch/pytorch/pull/100849
Approved by: https://github.com/albanD
2023-05-18 03:49:45 +00:00

640 lines
20 KiB
C++

#include <torch/csrc/jit/passes/utils/subgraph_utils.h>
#include <torch/csrc/jit/passes/canonicalize.h>
#include <ATen/core/symbol.h>
#include <c10/util/StringUtil.h>
#include <c10/util/irange.h>
#include <torch/csrc/jit/jit_log.h>
#include <utility>
namespace torch {
namespace jit {
namespace SubgraphUtils {
namespace {
bool hasSubgraph(Node* n) {
return n->hasAttribute(attr::Subgraph);
}
std::vector<c10::optional<const Use>> gatherLastUses(
at::ArrayRef<Value*> values) {
return fmap(values, [&](Value* v) -> c10::optional<const Use> {
return firstOrLastUse(v, /*find_first*/ false);
});
}
// When merging a node into a subgraph, we wish to preserve all of the
// aliasing properties of the node's outputs. It is difficult to track
// the node or its contained nodes through all of the ir manipulation
// involved in merging; it is pretty easy to uniquely identify the value
// based on its uses. We can identify the value by its last use in the graph.
// Values which do not have uses or which do not have a last use
// outside of the subgraph to be merged into we do not need to track.
struct ValueMapper {
// `to_merge` is the node we're merginginto a subgraph, `existing_subgraph` is
// the subgraph node that we're merging into if it exists
ValueMapper(
Node* to_merge,
AliasDb& db,
c10::optional<Node*> existing_subgraph) {
last_uses_ = gatherLastUses(to_merge->outputs());
if (existing_subgraph) {
existing_last_uses_ = gatherLastUses((*existing_subgraph)->outputs());
}
WithInsertPoint guard(to_merge);
auto g = to_merge->owningGraph();
// temporary node to put the aliasing properties of the node before its
// merged and destroyed
placeholder_node_ = g->insertNode(g->create(prim::Uninitialized, 0));
for (size_t i = 0; i < to_merge->outputs().size(); ++i) {
Value* existing = to_merge->outputs().at(i);
Value* new_value = placeholder_node_->insertOutput(i)->copyMetadata(
to_merge->outputs().at(i));
db.replaceWithNewValue(existing, new_value);
}
}
bool usesEqual(const Use& a, const Use& b) {
return a.user == b.user && a.offset == b.offset;
}
void copyAliasing(Node* merged_node, AliasDb& db) {
auto new_outputs = merged_node->outputs();
for (Value* v : new_outputs) {
auto maybe_last_use = firstOrLastUse(v, /*find_first*/ false);
// if it doesnt have a use it shouldnt have been added as output
TORCH_INTERNAL_ASSERT(maybe_last_use);
const Use last_use = *maybe_last_use;
// existing outputs of the subgraph do not need to have alias db mappings
// updated
bool is_existing_value = false;
for (size_t i = 0; i < existing_last_uses_.size() && !is_existing_value;
++i) {
is_existing_value = existing_last_uses_[i].has_value() &&
usesEqual(*existing_last_uses_[i], last_use);
}
if (is_existing_value) {
continue;
}
size_t i = 0;
while (i < last_uses_.size() && last_uses_.at(i).has_value() &&
!usesEqual(*last_uses_.at(i), last_use)) {
++i;
}
TORCH_INTERNAL_ASSERT(i != last_uses_.size());
db.replaceWithNewValue(placeholder_node_->outputs().at(i), v);
}
placeholder_node_->destroy();
}
std::vector<c10::optional<const Use>> last_uses_;
std::vector<c10::optional<const Use>> existing_last_uses_;
Node* placeholder_node_;
};
Node* executeSubgraphMergeAndUpdateAliasing(
Node* to_merge,
c10::optional<Node*> existing,
AliasDb& db,
const std::function<Node*(void)>& merge_fn) {
// When we merge a node into a subgraph, the new subgraph outputs
// have the same aliasing properties as the original node's outputs.
// Here we create a placeholder node, transfer the aliasing properties
// to the placeholder, execute the merge, and transfer the aliasing
// properties to the appropriate fusion group outputs
ValueMapper vm(to_merge, db, existing);
Node* fusion_group = merge_fn();
vm.copyAliasing(fusion_group, db);
return fusion_group;
}
// Combine the nodes in two subgraph together. The nodes will end up in
// `mergeTo`, and `mergeFrom` is destroyed.
void mergeSubgraph(Node* mergeTo, Node* mergeFrom) {
bool merge_from_is_after = mergeFrom->isAfter(mergeTo);
Node* nodeBeforeMergeFrom = mergeFrom->prev();
Node* nodeAfterMergeFrom = mergeFrom->next();
unmergeSubgraph(mergeFrom);
graph_node_list_iterator end_it;
graph_node_list_iterator it;
if (merge_from_is_after) {
it = nodeBeforeMergeFrom->iterator();
end_it = nodeAfterMergeFrom->iterator();
} else {
end_it = nodeBeforeMergeFrom->reverseIterator();
it = nodeAfterMergeFrom->reverseIterator();
}
++it;
std::vector<Node*> merged_nodes;
while (it != end_it) {
Node* node = *it;
++it;
mergeNodeIntoSubgraph(node, mergeTo);
}
}
struct topo_cmp_value {
bool operator()(Value* a, Value* b) const {
if (a->node() == b->node()) {
return a->unique() < b->unique();
}
return a->node()->isBefore(b->node());
}
};
struct topo_cmp_node {
bool operator()(Node* a, Node* b) const {
return a->isBefore(b);
}
};
void collectNodesToUnfuse(Node* start, std::set<Node*, topo_cmp_node>& s) {
if (start->kind() == prim::Return || start->kind() == prim::Param) {
GRAPH_DEBUG("reached the param or return node", getHeader(start));
return;
}
if (s.count(start) != 0) {
// already visited, no need to visit descendants
return;
}
GRAPH_DEBUG("collectNodesToUnfuse: inserting node ", getHeader(start));
s.insert(start);
for (auto o : start->outputs()) {
for (auto use : o->uses()) {
collectNodesToUnfuse(use.user, s);
}
}
}
std::vector<std::set<Value*, topo_cmp_value>> buildAliasedSets(
std::shared_ptr<Graph> subgraph) {
auto outputs = subgraph->outputs();
AliasDb alias_db(std::move(subgraph));
TORCH_INTERNAL_ASSERT(outputs.size() > 1);
std::vector<std::set<Value*, topo_cmp_value>> res;
for (auto o : outputs) {
auto grouped = false;
for (auto& s : res) {
auto os = *s.begin();
auto aliased = alias_db.mayContainAlias(os, o);
GRAPH_DEBUG(
"comparing %",
o->debugName(),
" with %",
os->debugName(),
" result ",
aliased);
if (aliased) {
s.insert(o);
GRAPH_DEBUG("Grouping %", o->debugName(), " with %", os->debugName());
grouped = true;
}
}
if (!grouped) {
res.push_back({o});
}
}
return res;
}
} // namespace
std::shared_ptr<Graph> getSubgraph(Node* n) {
return n->g(attr::Subgraph);
}
void unmergeSubgraph(Node* subgraphNode) {
// Inline the graph, replace uses of node outputs and destroy the node
auto outerGraph = subgraphNode->owningGraph();
WithInsertPoint guard(subgraphNode);
const auto subgraphOutputs = insertGraph(
*outerGraph, *getSubgraph(subgraphNode), subgraphNode->inputs());
AT_ASSERT(subgraphOutputs.size() >= subgraphNode->outputs().size());
for (size_t i = 0; i < subgraphNode->outputs().size(); ++i) {
subgraphNode->outputs()[i]->replaceAllUsesWith(subgraphOutputs[i]);
}
subgraphNode->destroy();
}
static void collectNestedUses(
std::unordered_set<Value*>& closed_over_values,
std::unordered_set<Value*>& new_values,
std::unordered_map<Value*, Value*>& externalValuesMap,
Node* input_node) {
for (auto input : input_node->inputs()) {
if (externalValuesMap.count(input) == 0 && new_values.count(input) == 0) {
closed_over_values.insert(input);
}
}
if (input_node->kind() == prim::If) {
for (Block* block : input_node->blocks()) {
for (Node* node : block->nodes()) {
collectNestedUses(
closed_over_values, new_values, externalValuesMap, node);
}
for (Value* v : block->outputs()) {
if (externalValuesMap.count(v) == 0 && new_values.count(v) == 0) {
closed_over_values.insert(v);
}
}
}
} else if (input_node->kind() == prim::Loop) {
for (Value* v : input_node->inputs()) {
if (externalValuesMap.count(v) == 0 && new_values.count(v) == 0) {
closed_over_values.insert(v);
}
}
Block* block = input_node->blocks().at(0);
for (Value* v : block->inputs()) {
new_values.insert(v);
}
for (Node* node : block->nodes()) {
collectNestedUses(
closed_over_values, new_values, externalValuesMap, node);
}
} else if (!input_node->blocks().empty()) {
TORCH_INTERNAL_ASSERT(false, input_node, " kind not handled yet");
}
for (Value* output : input_node->outputs()) {
new_values.insert(output);
}
}
static std::unordered_set<Value*> closedOverValues(
Node* toMerge,
std::unordered_map<Value*, Value*>& externalValuesMap) {
std::unordered_set<Value*> closed_over_values;
std::unordered_set<Value*> new_values;
collectNestedUses(closed_over_values, new_values, externalValuesMap, toMerge);
return closed_over_values;
}
void mergeNodeIntoSubgraph(
Node* toMerge,
Node* subgraphNode,
bool destroyNode) {
AT_ASSERT(hasSubgraph(subgraphNode) && toMerge != subgraphNode);
if (hasSubgraph(toMerge)) {
return mergeSubgraph(subgraphNode, toMerge);
}
auto subgraph = getSubgraph(subgraphNode);
// Map from values in the surrounding graph to inputs/outputs in the subgraph
std::unordered_map<Value*, Value*> externalValuesMap;
AT_ASSERT(subgraphNode->inputs().size() == subgraph->inputs().size());
size_t idx = 0;
for (auto input : subgraphNode->inputs()) {
externalValuesMap[input] = subgraph->inputs()[idx];
idx++;
}
for (size_t i = 0; i < subgraphNode->outputs().size(); ++i) {
externalValuesMap[subgraphNode->outputs().at(i)] =
subgraph->outputs().at(i);
}
// Add n's inputs to the group's input list if we don't already have them
bool merging_node_after_subgraph = toMerge->isAfter(subgraphNode);
Node* guard_node = merging_node_after_subgraph ? *subgraph->nodes().end()
: *subgraph->nodes().begin();
WithInsertPoint guard(guard_node);
std::unordered_set<Value*> closedValues =
closedOverValues(toMerge, externalValuesMap);
// There are currently downstream usage that relies on a fixed ordering
// of graph inputs. TODO: remove
std::vector<Value*> orderedClosedValues;
std::unordered_set<Value*> orderedSeenValues;
for (Value* input : toMerge->inputs()) {
orderedClosedValues.push_back(input);
orderedSeenValues.insert(input);
}
for (Value* closedValue : closedValues) {
if (!orderedSeenValues.count(closedValue)) {
orderedClosedValues.push_back(closedValue);
orderedSeenValues.insert(closedValue);
}
}
for (auto input : orderedClosedValues) {
if (externalValuesMap.count(input) == 0) {
// Clone constants inside the subgraph instead of referencing them, to
// enable more optimizations
if (auto value = toIValue(input)) {
auto nv = subgraph->insertConstant(*value);
nv->copyMetadata(input);
externalValuesMap[input] = nv;
} else {
// The common case: this is a regular input, so just register it with
// the group node and inner subgraph
subgraphNode->addInput(input);
auto inputToGraph = subgraph->addInput();
inputToGraph->copyMetadata(input);
externalValuesMap[input] = inputToGraph;
}
}
}
// Merge the node into the graph
auto mergedNode = subgraph->insertNode(subgraph->createClone(
toMerge, [&](Value* v) { return externalValuesMap[v]; }));
if (!merging_node_after_subgraph) {
// If n's outputs were inputs to `group`, remove them since we just merged
// n in.
//
// i.e.,
// x = f(w); group(x, y, z) becomes group(w, y, z).
// x, y, z = f(w); group(x, y, z) becomes group(w).
auto inputs = subgraphNode->inputs();
for (size_t i = 0; i < toMerge->outputs().size(); ++i) {
auto it = std::find(inputs.begin(), inputs.end(), toMerge->outputs()[i]);
if (it != inputs.end()) {
size_t p = it - inputs.begin();
subgraphNode->removeInput(p);
subgraph->inputs()[p]->replaceAllUsesWith(mergedNode->outputs()[i]);
subgraph->eraseInput(p);
}
}
}
// Add n's outputs to the group node and inner subgraph outputs.
for (const auto i : c10::irange(toMerge->outputs().size())) {
auto oldOutput = toMerge->outputs()[i];
auto newOutput = mergedNode->outputs()[i];
subgraph->registerOutput(newOutput);
auto groupOutput = subgraphNode->addOutput();
groupOutput->copyMetadata(oldOutput);
oldOutput->replaceAllUsesWith(groupOutput);
}
// Remove the original node now that the merge is complete
if (destroyNode) {
toMerge->destroy();
}
// We wait till destroying `toMerge` before pruning subgraph outputs,
// since destroying `toMerge` could cause a subgraph output to no longer
// have any uses
const auto hasUsesOutsideSubgraph = [&](Value* v) {
return std::any_of(
v->uses().cbegin(), v->uses().cend(), [&](const Use& use) {
return use.user->isAfter(subgraphNode);
});
};
for (int64_t i = subgraphNode->outputs().size() - 1; i >= 0; i--) {
if (!hasUsesOutsideSubgraph(subgraphNode->outputs().at(i))) {
subgraphNode->eraseOutput(i);
subgraph->eraseOutput(i);
}
}
}
Node* createSingletonSubgraph(Node* n, Symbol subgraphKind) {
auto graph = n->owningGraph();
auto subgraph = graph->create(subgraphKind, 0);
subgraph->g_(attr::Subgraph, std::make_shared<Graph>(graph->current_scope()));
subgraph->insertBefore(n);
mergeNodeIntoSubgraph(n, subgraph);
return subgraph;
}
void mergeNodeIntoSubgraphAndUpdateAliasing(
Node* to_merge,
Node* subgraphNode,
AliasDb& db) {
executeSubgraphMergeAndUpdateAliasing(to_merge, subgraphNode, db, [&]() {
mergeNodeIntoSubgraph(to_merge, subgraphNode);
return subgraphNode;
});
}
Node* createSingletonSubgraphAndUpdateAliasing(
Node* to_merge,
Symbol subgraphKind,
AliasDb& db) {
return executeSubgraphMergeAndUpdateAliasing(
to_merge, c10::nullopt, db, [&]() {
return createSingletonSubgraph(to_merge, subgraphKind);
});
}
bool unmergeOutputsAlisingInputs(Node* subgraphNode) {
GRAPH_DEBUG("unfuseOutputsAlisingInputs on ", getHeader(subgraphNode));
auto subgraph = subgraphNode->g(attr::Subgraph);
AliasDb alias_db(subgraph);
std::set<Node*, topo_cmp_node> nodes;
for (auto o : subgraph->outputs()) {
if (alias_db.mayContainAlias(o, subgraph->inputs())) {
collectNodesToUnfuse(o->node(), nodes);
}
}
// unfuse in the reverse topo order
for (auto it = nodes.rbegin(); it != nodes.rend(); it++) {
SubgraphUtils::unmergeNode(*it, subgraphNode);
}
return !nodes.empty();
}
bool unmergeAliasedOutputs(Node* subgraphNode) {
GRAPH_DEBUG("unfuseAliasedOutputs on ", getHeader(subgraphNode));
if (subgraphNode->outputs().size() < 2) {
return false;
}
auto subgraph = subgraphNode->g(attr::Subgraph);
GRAPH_DUMP("unfuseAliasedOutputs Subgraph ", subgraph);
auto sets = buildAliasedSets(std::move(subgraph));
GRAPH_DEBUG("buildAliasedSets sets.size() = ", sets.size());
std::set<Node*, topo_cmp_node> nodes;
for (auto i : c10::irange(sets.size())) {
if (sets[i].size() <= 1) {
GRAPH_DEBUG(
"Set ",
i,
" with leader ",
(*(sets[i].begin()))->debugName(),
" size = ",
sets[i].size());
continue;
}
// we have at least two aliased outputs
// we skip the earliest node w.r.t. the topo order
// NB. after some nodes are unfused, the outputs of some other nodes
// may become the outputs of the subgraph and alias the remaining ones
// so we have to re-run this function until there are no more changes
auto it = ++sets[i].begin();
while (it != sets[i].end()) {
GRAPH_DEBUG(
"root aliased value ", (*it)->debugName(), " node ", *(*it)->node());
collectNodesToUnfuse((*it)->node(), nodes);
it++;
}
}
// unfuse in the reverse topo order
for (auto it = nodes.rbegin(); it != nodes.rend(); it++) {
unmergeNode(*it, subgraphNode);
}
return !nodes.empty();
}
void unmergeNode(Node* n, Node* subgraphNode) {
// collect output indices
GRAPH_DEBUG("unfuseNode node ", getHeader(n));
auto subgraph = n->owningGraph();
std::set<Value*> node_outputs(n->outputs().begin(), n->outputs().end());
std::set<size_t> output_indices;
std::set<Value*> node_inputs(n->inputs().begin(), n->inputs().end());
std::unordered_map<Value*, Value*> local_map;
auto env = [&](Value* v) {
auto it = local_map.find(v);
if (it != local_map.end()) {
return it->second;
}
TORCH_INTERNAL_ASSERT(
false,
"all inputs should've been mapped. Couldn't map %",
v->debugName());
return v;
};
for (auto i : c10::irange(subgraph->outputs().size())) {
if (node_outputs.count(subgraph->outputs().at(i)) != 0) {
output_indices.insert(i);
}
if (node_inputs.count(subgraph->outputs().at(i)) != 0) {
GRAPH_DEBUG(
"output %",
subgraph->outputs().at(i)->debugName(),
" is already subgraph's output");
GRAPH_DEBUG(
"Mapping %",
subgraph->outputs().at(i)->debugName(),
" to %",
subgraphNode->outputs().at(i)->debugName());
local_map[subgraph->outputs().at(i)] = subgraphNode->outputs().at(i);
node_inputs.erase(subgraph->outputs().at(i));
}
}
WithInsertPoint wip(subgraphNode->next());
// these node inputs need to be added to subgraph's outputs
// put them in vmap
for (auto ni : node_inputs) {
if (local_map.count(ni) != 0) {
// this could happen if `n` uses two or more outputs
// of a constant node and we already cloned the constant
// into the outer graph and mapped its outputs
continue;
}
Value* sno = nullptr;
if (ni->node()->kind() == prim::Constant) {
auto copy = subgraphNode->owningGraph()->createClone(ni->node(), env);
subgraphNode->owningGraph()->insertNode(copy);
// in case we have a multi-output const, map the rest of the outputs
// so when we get to clone `n`, `n`'s clone will use the outputs of this
// constant clone
for (auto i : c10::irange(n->outputs().size())) {
GRAPH_DEBUG(
"Mapping %",
ni->node()->output(i)->debugName(),
" to %",
copy->output(i)->debugName());
local_map[ni->node()->output(i)] = copy->output(i);
}
} else {
subgraph->registerOutput(ni);
sno = subgraphNode->addOutput();
sno->setType(ni->type());
GRAPH_DEBUG("Mapping %", ni->debugName(), " to %", sno->debugName());
local_map[ni] = sno;
}
}
auto copy = subgraphNode->owningGraph()->createClone(n, env);
GRAPH_DEBUG("copy ", *copy);
for (auto i : c10::irange(n->outputs().size())) {
auto oo = n->outputs()[i];
auto no = copy->outputs()[i];
no->copyMetadata(oo);
GRAPH_DEBUG("Mapping %", oo->debugName(), " to %", no->debugName());
local_map[oo] = no;
}
subgraphNode->owningGraph()->insertNode(copy);
for (auto it = output_indices.rbegin(); it != output_indices.rend(); it++) {
auto replace_val = local_map[subgraph->outputs().at(*it)];
subgraphNode->outputs().at(*it)->replaceAllUsesWith(replace_val);
subgraphNode->eraseOutput(*it);
subgraph->eraseOutput(*it);
}
n->destroy();
}
static std::string truncateStrWithHash(const std::string& s, size_t maxlen) {
if (s.size() <= maxlen) {
return s;
}
std::string hash_str = c10::to_string(c10::hash<std::string>{}(s));
// If hash-string plus '_' can fit into maxlen, then truncate the original
// string correspondingly so that the final string with the hash included fits
// into maxlen. If that's not possible, at least truncate the original string
// to maxlen (and append the hash to it).
size_t trunc_len =
(maxlen > hash_str.size() + 1) ? (maxlen - hash_str.size() - 1) : maxlen;
std::stringstream truncated;
truncated << s.substr(0, trunc_len);
truncated << "_" << hash_str;
return truncated.str();
}
std::string generateNameForGraph(
const std::shared_ptr<Graph>& graph,
size_t maxlen,
const std::string& prefix) {
std::stringstream graph_name;
graph_name << prefix;
for (Node* node : graph->nodes()) {
if (!node->kind().is_aten()) {
continue;
}
graph_name << "_" << node->kind().toUnqualString();
}
return truncateStrWithHash(graph_name.str(), maxlen);
}
} // namespace SubgraphUtils
} // namespace jit
} // namespace torch