Fix double-lookups into Scope and map (#9062)

In several places we were checking something was in a scope, and if so,
looking it up again to actually do something with it. We should use
::find (or ::shallow_find for a mutable reference into a Scope)

No functional impact.

Author: abadams

src/CodeGen_D3D12Compute_Dev.cpp

@@ -684,8 +684,9 @@ void CodeGen_D3D12Compute_Dev::CodeGen_D3D12Compute_C::visit(const Store *op) {
     bool shared_promotion_required = false;
     string promotion_str = "";
     if (groupshared_allocations.contains(op->name)) {
-        internal_assert(allocations.contains(op->name));
-        Type promoted_type = allocations.get(op->name).type;
+        const auto *alloc = allocations.find(op->name);
+        internal_assert(alloc);
+        Type promoted_type = alloc->type;
         if (promoted_type != op->value.type()) {
             shared_promotion_required = true;
             // NOTE(marcos): might need to resort to StoragePackUnpack::pack_store() here

src/CompilerLogger.cpp

@@ -271,11 +271,13 @@ std::ostream &JSONCompilerLogger::emit_to_stream(std::ostream &o) {
     }
 
     // If these are present, emit them, even if value is zero
-    if (compilation_time.count(Phase::HalideLowering)) {
-        emit_key_value(o, indent, "compilation_time_halide_lowering", compilation_time[Phase::HalideLowering]);
+    if (auto it = compilation_time.find(Phase::HalideLowering);
+        it != compilation_time.end()) {
+        emit_key_value(o, indent, "compilation_time_halide_lowering", it->second);
     }
-    if (compilation_time.count(Phase::LLVM)) {
-        emit_key_value(o, indent, "compilation_time_llvm", compilation_time[Phase::LLVM]);
+    if (auto it = compilation_time.find(Phase::LLVM);
+        it != compilation_time.end()) {
+        emit_key_value(o, indent, "compilation_time_llvm", it->second);
     }
 
     if (!matched_simplifier_rules.empty()) {

src/HexagonOptimize.cpp

@@ -1521,8 +1521,8 @@ class EliminateInterleaves : public IRMutator {
         }
 
         if (const Load *load = x.as<Load>()) {
-            if (buffers.contains(load->name)) {
-                BufferState &state = buffers.ref(load->name);
+            if (auto *state_ptr = buffers.shallow_find(load->name)) {
+                BufferState &state = *state_ptr;
                 if (state != BufferState::NotInterleaved) {
                     state = BufferState::Interleaved;
                     return x;
@@ -1814,23 +1814,27 @@ class EliminateInterleaves : public IRMutator {
                                    op->func, op->value_index, op->image, op->param);
             // Add the interleave back to the result of the call.
             return native_interleave(expr);
-        } else if (deinterleaving_alts.find(op->name) != deinterleaving_alts.end() && hvx_target >= deinterleaving_alts[op->name].first &&
-
+        } else if (auto it = deinterleaving_alts.find(op->name);
+                   it != deinterleaving_alts.end() &&
+                   hvx_target >= it->second.first &&
                    yields_removable_interleave(args)) {
             // This call has a deinterleaving alternative, and the
             // arguments are interleaved, so we should use the
             // alternative instead.
             for (Expr &i : args) {
                 i = remove_interleave(i);
             }
-            return Call::make(op->type, deinterleaving_alts[op->name].second, args, op->call_type);
-        } else if (interleaving_alts.count(op->name) && hvx_target >= interleaving_alts[op->name].first && is_native_deinterleave(args[0])) {
+            return Call::make(op->type, it->second.second, args, op->call_type);
+        } else if (auto it = interleaving_alts.find(op->name);
+                   it != interleaving_alts.end() &&
+                   hvx_target >= it->second.first &&
+                   is_native_deinterleave(args[0])) {
             // This is an interleaving alternative with a
             // deinterleave, which can be generated when we
             // deinterleave storage. Revert back to the interleaving
             // op so we can remove the deinterleave.
             Expr arg = args[0].as<Call>()->args[0];
-            return Call::make(op->type, interleaving_alts[op->name].second, {arg}, op->call_type,
+            return Call::make(op->type, it->second.second, {arg}, op->call_type,
                               op->func, op->value_index, op->image, op->param);
         } else if (changed) {
             return Call::make(op->type, op->name, args, op->call_type,
@@ -1937,7 +1941,7 @@ class EliminateInterleaves : public IRMutator {
     }
 
     Expr visit(const Load *op) override {
-        if (buffers.contains(op->name)) {
+        if (auto *buf_state = buffers.shallow_find(op->name)) {
             if ((op->type.lanes() * op->type.bits()) % (native_vector_bits * 2) == 0) {
                 // This is a double vector load, we might be able to
                 // deinterleave the storage of this buffer.
@@ -1959,8 +1963,7 @@ class EliminateInterleaves : public IRMutator {
             } else {
                 // This is not a double vector load, so we can't
                 // deinterleave the storage of this buffer.
-                BufferState &state = buffers.ref(op->name);
-                state = BufferState::NotInterleaved;
+                *buf_state = BufferState::NotInterleaved;
             }
         }
         Expr expr = IRMutator::visit(op);
@@ -2224,16 +2227,18 @@ class SyncronizationBarriers : public IRMutator {
     // Creates entry in sync map for the stmt requiring a
     // scatter-release instruction before it.
     void check_hazard(const string &name) {
-        if (in_flight.find(name) == in_flight.end()) {
+        auto it = in_flight.find(name);
+        if (it == in_flight.end()) {
             return;
         }
         // Sync Needed. Add the scatter-release before the first different For
         // loop lock between the curr_path and the hazard src location.
-        size_t min_size = std::min(in_flight[name].size(), curr_path.size());
+        const auto &flight_path = it->second;
+        size_t min_size = std::min(flight_path.size(), curr_path.size());
         size_t i = 0;
         // Find the first different For loop block.
         for (; i < min_size; i++) {
-            if (in_flight[name][i] != curr_path[i]) {
+            if (flight_path[i] != curr_path[i]) {
                 break;
             }
         }
@@ -2266,9 +2271,9 @@ class SyncronizationBarriers : public IRMutator {
         curr = &s;
         Stmt new_s = IRMutator::mutate(s);
         // Wrap the stmt with scatter-release if any hazard was detected.
-        if (sync.find(&s) != sync.end()) {
+        if (auto it = sync.find(&s); it != sync.end()) {
             Stmt scatter_sync =
-                Evaluate::make(Call::make(Int(32), Call::hvx_scatter_release, {sync[&s]}, Call::Intrinsic));
+                Evaluate::make(Call::make(Int(32), Call::hvx_scatter_release, {it->second}, Call::Intrinsic));
             return Block::make(scatter_sync, new_s);
         }
         return new_s;

src/SlidingWindow.cpp

@@ -270,9 +270,11 @@ class SlidingWindowOnFunctionAndLoop : public IRMutator {
             for (int i = 0; i < func.dimensions(); i++) {
                 // Look up the region required of this function's last stage
                 string var = prefix + func_args[i];
-                internal_assert(scope.contains(var + ".min") && scope.contains(var + ".max"));
-                Expr min_req = scope.get(var + ".min");
-                Expr max_req = scope.get(var + ".max");
+                const auto *min_val = scope.find(var + ".min");
+                const auto *max_val = scope.find(var + ".max");
+                internal_assert(min_val && max_val);
+                Expr min_req = *min_val;
+                Expr max_req = *max_val;
                 min_req = expand_expr(min_req, scope);
                 max_req = expand_expr(max_req, scope);
 

src/SplitTuples.cpp

@@ -60,8 +60,9 @@ class SplitTuples : public IRMutator {
         if (op->types.size() > 1) {
             // If there is a corresponding HoistedStorage node record the new number of
             // realizes.
-            if (hoisted_tuple_count.count(op->name)) {
-                hoisted_tuple_count[op->name] = op->types.size();
+            if (auto it = hoisted_tuple_count.find(op->name);
+                it != hoisted_tuple_count.end()) {
+                it->second = op->types.size();
             }
             // Make a nested set of realize nodes for each tuple element
             Stmt body = mutate(op->body);

src/StorageFlattening.cpp

@@ -504,8 +504,9 @@ class HoistStorage : public IRMutator {
     }
 
     Stmt visit(const Allocate *op) override {
-        if (hoisted_storages_map.count(op->name) > 0) {
-            HoistedStorageData &hoisted_storage_data = hoisted_storages[hoisted_storages_map[op->name]];
+        if (auto it = hoisted_storages_map.find(op->name);
+            it != hoisted_storages_map.end()) {
+            HoistedStorageData &hoisted_storage_data = hoisted_storages[it->second];
 
             auto expand_and_bound = [&](Expr e) {
                 // Iterate from innermost outwards

src/VectorizeLoops.cpp

@@ -541,8 +541,8 @@ class VectorSubs : public IRMutator {
     }
 
     Expr visit(const Variable *op) override {
-        if (replacements.count(op->name) > 0) {
-            return replacements[op->name];
+        if (auto it = replacements.find(op->name); it != replacements.end()) {
+            return it->second;
         } else if (scope.contains(op->name)) {
             string widened_name = get_widened_var_name(op->name);
             return Variable::make(vector_scope.get(widened_name).type(), widened_name);
@@ -994,11 +994,12 @@ class VectorSubs : public IRMutator {
             // them according to the current loop level.
             for (const auto &[var, val] : containing_lets) {
                 // Skip if this var wasn't vectorized.
-                if (!scope.contains(var)) {
+                const auto *scope_val = scope.find(var);
+                if (!scope_val) {
                     continue;
                 }
                 string vectorized_name = get_widened_var_name(var);
-                Expr vectorized_value = mutate(scope.get(var));
+                Expr vectorized_value = mutate(*scope_val);
                 vector_scope.push(vectorized_name, vectorized_value);
             }