[Isolates] Merge crankshaft (r5922 from bleeding_edge).

src/runtime-profiler.cc

+// Copyright 2010 the V8 project authors. All rights reserved.
+// Redistribution and use in source and binary forms, with or without
+// modification, are permitted provided that the following conditions are
+// met:
+//
+//     * Redistributions of source code must retain the above copyright
+//       notice, this list of conditions and the following disclaimer.
+//     * Redistributions in binary form must reproduce the above
+//       copyright notice, this list of conditions and the following
+//       disclaimer in the documentation and/or other materials provided
+//       with the distribution.
+//     * Neither the name of Google Inc. nor the names of its
+//       contributors may be used to endorse or promote products derived
+//       from this software without specific prior written permission.
+//
+// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+#include "v8.h"
+
+#include "runtime-profiler.h"
+
+#include "assembler.h"
+#include "code-stubs.h"
+#include "compilation-cache.h"
+#include "deoptimizer.h"
+#include "execution.h"
+#include "global-handles.h"
+#include "platform.h"
+#include "scopeinfo.h"
+
+namespace v8 {
+namespace internal {
+
+
+class PendingListNode : public Malloced {
+ public:
+  explicit PendingListNode(JSFunction* function);
+  ~PendingListNode() { Destroy(); }
+
+  PendingListNode* next() const { return next_; }
+  void set_next(PendingListNode* node) { next_ = node; }
+  Handle<JSFunction> function() { return Handle<JSFunction>::cast(function_); }
+
+  // If the function is garbage collected before we've had the chance
+  // to optimize it the weak handle will be null.
+  bool IsValid() { return !function_.is_null(); }
+
+  // Returns the number of microseconds this node has been pending.
+  int Delay() const { return static_cast<int>(OS::Ticks() - start_); }
+
+ private:
+  void Destroy();
+  static void WeakCallback(v8::Persistent<v8::Value> object, void* data);
+
+  PendingListNode* next_;
+  Handle<Object> function_;  // Weak handle.
+  int64_t start_;
+};
+
+
+// Optimization sampler constants.
+static const int kSamplerFrameCount = 2;
+static const int kSamplerFrameWeight[kSamplerFrameCount] = { 2, 1 };
+
+static const int kSamplerTicksDelta = 32;
+
+static const int kSamplerThresholdInit = 3;
+static const int kSamplerThresholdMin = 1;
+static const int kSamplerThresholdDelta = 1;
+
+static const int kSamplerThresholdSizeFactorInit = 3;
+static const int kSamplerThresholdSizeFactorMin = 1;
+static const int kSamplerThresholdSizeFactorDelta = 1;
+
+static const int kSizeLimit = 1500;
+
+
+PendingListNode::PendingListNode(JSFunction* function) : next_(NULL) {
+  GlobalHandles* global_handles = Isolate::Current()->global_handles();
+  function_ = global_handles->Create(function);
+  start_ = OS::Ticks();
+  global_handles->MakeWeak(function_.location(), this, &WeakCallback);
+}
+
+
+void PendingListNode::Destroy() {
+  if (!IsValid()) return;
+  GlobalHandles* global_handles = Isolate::Current()->global_handles();
+  global_handles->Destroy(function_.location());
+  function_= Handle<Object>::null();
+}
+
+
+void PendingListNode::WeakCallback(v8::Persistent<v8::Value>, void* data) {
+  reinterpret_cast<PendingListNode*>(data)->Destroy();
+}
+
+
+static bool IsOptimizable(JSFunction* function) {
+  Code* code = function->code();
+  return code->kind() == Code::FUNCTION && code->optimizable();
+}
+
+
+Atomic32 RuntimeProfiler::state_ = 0;
+// TODO(isolates): Create the semaphore lazily and clean it up when no
+// longer required.
+Semaphore* RuntimeProfiler::semaphore_ = OS::CreateSemaphore(0);
+
+
+RuntimeProfiler::RuntimeProfiler(Isolate* isolate)
+    : isolate_(isolate),
+      sampler_threshold_(kSamplerThresholdInit),
+      sampler_threshold_size_factor_(kSamplerThresholdSizeFactorInit),
+      sampler_window_position_(0),
+      optimize_soon_list_(NULL) {
+  ClearSampleBuffer();
+}
+
+
+bool RuntimeProfiler::IsEnabled() {
+  return V8::UseCrankshaft() && FLAG_opt;
+}
+
+
+void RuntimeProfiler::Optimize(JSFunction* function, bool eager, int delay) {
+  ASSERT(IsOptimizable(function));
+  if (FLAG_trace_opt) {
+    PrintF("[marking (%s) ", eager ? "eagerly" : "lazily");
+    function->PrintName();
+    PrintF(" for recompilation");
+    if (delay > 0) {
+      PrintF(" (delayed %0.3f ms)", static_cast<double>(delay) / 1000);
+    }
+    PrintF("]\n");
+  }
+
+  // The next call to the function will trigger optimization.
+  function->MarkForLazyRecompilation();
+}
+
+
+void RuntimeProfiler::AttemptOnStackReplacement(JSFunction* function) {
+  // See AlwaysFullCompiler (in compiler.cc) comment on why we need
+  // Debug::has_break_points().
+  ASSERT(function->IsMarkedForLazyRecompilation());
+  if (!FLAG_use_osr ||
+      isolate_->debug()->has_break_points() ||
+      function->IsBuiltin()) {
+    return;
+  }
+
+  SharedFunctionInfo* shared = function->shared();
+  // If the code is not optimizable, don't try OSR.
+  if (!shared->code()->optimizable()) return;
+
+  // We are not prepared to do OSR for a function that already has an
+  // allocated arguments object.  The optimized code would bypass it for
+  // arguments accesses, which is unsound.  Don't try OSR.
+  if (shared->scope_info()->HasArgumentsShadow()) return;
+
+  // We're using on-stack replacement: patch the unoptimized code so that
+  // any back edge in any unoptimized frame will trigger on-stack
+  // replacement for that frame.
+  if (FLAG_trace_osr) {
+    PrintF("[patching stack checks in ");
+    function->PrintName();
+    PrintF(" for on-stack replacement]\n");
+  }
+
+  // Get the stack check stub code object to match against.  We aren't
+  // prepared to generate it, but we don't expect to have to.
+  StackCheckStub check_stub;
+  Object* check_code;
+  MaybeObject* maybe_check_code = check_stub.TryGetCode();
+  if (maybe_check_code->ToObject(&check_code)) {
+    Code* replacement_code =
+        isolate_->builtins()->builtin(Builtins::OnStackReplacement);
+    Code* unoptimized_code = shared->code();
+    // Iterate the unoptimized code and patch every stack check except at
+    // the function entry.  This code assumes the function entry stack
+    // check appears first i.e., is not deferred or otherwise reordered.
+    bool first = true;
+    for (RelocIterator it(unoptimized_code, RelocInfo::kCodeTargetMask);
+         !it.done();
+         it.next()) {
+      RelocInfo* rinfo = it.rinfo();
+      if (rinfo->target_address() == Code::cast(check_code)->entry()) {
+        if (first) {
+          first = false;
+        } else {
+          Deoptimizer::PatchStackCheckCode(rinfo, replacement_code);
+        }
+      }
+    }
+  }
+}
+
+
+void RuntimeProfiler::ClearSampleBuffer() {
+  memset(sampler_window_, 0, sizeof(sampler_window_));
+  memset(sampler_window_weight_, 0, sizeof(sampler_window_weight_));
+}
+
+
+void RuntimeProfiler::ClearSampleBufferNewSpaceEntries() {
+  for (int i = 0; i < kSamplerWindowSize; i++) {
+    if (isolate_->heap()->InNewSpace(sampler_window_[i])) {
+      sampler_window_[i] = NULL;
+      sampler_window_weight_[i] = 0;
+    }
+  }
+}
+
+
+int RuntimeProfiler::LookupSample(JSFunction* function) {
+  int weight = 0;
+  for (int i = 0; i < kSamplerWindowSize; i++) {
+    JSFunction* sample = sampler_window_[i];
+    if (sample != NULL) {
+      if (function == sample) {
+        weight += sampler_window_weight_[i];
+      }
+    }
+  }
+  return weight;
+}
+
+
+void RuntimeProfiler::AddSample(JSFunction* function, int weight) {
+  ASSERT(IsPowerOf2(kSamplerWindowSize));
+  sampler_window_[sampler_window_position_] = function;
+  sampler_window_weight_[sampler_window_position_] = weight;
+  sampler_window_position_ = (sampler_window_position_ + 1) &
+      (kSamplerWindowSize - 1);
+}
+
+
+void RuntimeProfiler::OptimizeNow() {
+  HandleScope scope(isolate_);
+  PendingListNode* current = optimize_soon_list_;
+  while (current != NULL) {
+    PendingListNode* next = current->next();
+    if (current->IsValid()) {
+      Handle<JSFunction> function = current->function();
+      int delay = current->Delay();
+      if (IsOptimizable(*function)) {
+        Optimize(*function, true, delay);
+      }
+    }
+    delete current;
+    current = next;
+  }
+  optimize_soon_list_ = NULL;
+
+  // Run through the JavaScript frames and collect them. If we already
+  // have a sample of the function, we mark it for optimizations
+  // (eagerly or lazily).
+  JSFunction* samples[kSamplerFrameCount];
+  int count = 0;
+  for (JavaScriptFrameIterator it;
+       count < kSamplerFrameCount && !it.done();
+       it.Advance()) {
+    JavaScriptFrame* frame = it.frame();
+    JSFunction* function = JSFunction::cast(frame->function());
+    int function_size = function->shared()->SourceSize();
+    int threshold_size_factor;
+    if (function_size > kSizeLimit) {
+      threshold_size_factor = sampler_threshold_size_factor_;
+    } else {
+      threshold_size_factor = 1;
+    }
+
+    int threshold = sampler_threshold_ * threshold_size_factor;
+    samples[count++] = function;
+    if (function->IsMarkedForLazyRecompilation()) {
+      Code* unoptimized = function->shared()->code();
+      int nesting = unoptimized->allow_osr_at_loop_nesting_level();
+      if (nesting == 0) AttemptOnStackReplacement(function);
+      int new_nesting = Min(nesting + 1, Code::kMaxLoopNestingMarker);
+      unoptimized->set_allow_osr_at_loop_nesting_level(new_nesting);
+    } else if (LookupSample(function) >= threshold) {
+      if (IsOptimizable(function)) {
+        Optimize(function, false, 0);
+        isolate_->compilation_cache()->MarkForEagerOptimizing(
+            Handle<JSFunction>(function, isolate_));
+      }
+    }
+  }
+
+  // Add the collected functions as samples. It's important not to do
+  // this as part of collecting them because this will interfere with
+  // the sample lookup in case of recursive functions.
+  for (int i = 0; i < count; i++) {
+    AddSample(samples[i], kSamplerFrameWeight[i]);
+  }
+}
+
+
+void RuntimeProfiler::OptimizeSoon(JSFunction* function) {
+  if (!IsOptimizable(function)) return;
+  PendingListNode* node = new PendingListNode(function);
+  node->set_next(optimize_soon_list_);
+  optimize_soon_list_ = node;
+}
+
+
+void RuntimeProfiler::NotifyTick() {
+  isolate_->stack_guard()->RequestRuntimeProfilerTick();
+}
+
+
+void RuntimeProfiler::MarkCompactPrologue(bool is_compacting) {
+  if (is_compacting) {
+    // Clear all samples before mark-sweep-compact because every
+    // function might move.
+    ClearSampleBuffer();
+  } else {
+    // Clear only new space entries on mark-sweep since none of the
+    // old-space functions will move.
+    ClearSampleBufferNewSpaceEntries();
+  }
+}
+
+
+void RuntimeProfiler::Setup() {
+  ClearSampleBuffer();
+  // If the ticker hasn't already started, make sure to do so to get
+  // the ticks for the runtime profiler.
+  if (IsEnabled()) isolate_->logger()->EnsureTickerStarted();
+}
+
+
+void RuntimeProfiler::Reset() {
+  sampler_threshold_ = kSamplerThresholdInit;
+  sampler_threshold_size_factor_ = kSamplerThresholdSizeFactorInit;
+}
+
+
+void RuntimeProfiler::TearDown() {
+  // Nothing to do.
+}
+
+
+Object** RuntimeProfiler::SamplerWindowAddress() {
+  return reinterpret_cast<Object**>(sampler_window_);
+}
+
+
+int RuntimeProfiler::SamplerWindowSize() {
+  return kSamplerWindowSize;
+}
+
+
+void RuntimeProfiler::HandleWakeUp(Isolate* isolate) {
+  // The profiler thread must still be waiting.
+  ASSERT(NoBarrier_Load(&state_) >= 0);
+  // In IsolateEnteredJS we have already incremented the counter and
+  // undid the decrement done by the profiler thread. Increment again
+  // to get the right count of active isolates.
+  NoBarrier_AtomicIncrement(&state_, 1);
+  semaphore_->Signal();
+  isolate->ResetEagerOptimizingData();
+}
+
+
+bool RuntimeProfiler::IsSomeIsolateInJS() {
+  return NoBarrier_Load(&state_) > 0;
+}
+
+
+bool RuntimeProfiler::WaitForSomeIsolateToEnterJS() {
+  Atomic32 old_state = NoBarrier_CompareAndSwap(&state_, 0, -1);
+  ASSERT(old_state >= -1);
+  if (old_state != 0) return false;
+  semaphore_->Wait();
+  return true;
+}
+
+
+void RuntimeProfiler::WakeUpRuntimeProfilerThreadBeforeShutdown() {
+  semaphore_->Signal();
+}
+
+
+bool RuntimeProfilerRateLimiter::SuspendIfNecessary() {
+  if (!RuntimeProfiler::IsEnabled()) return false;
+  static const int kNonJSTicksThreshold = 100;
+  if (RuntimeProfiler::IsSomeIsolateInJS()) {
+    non_js_ticks_ = 0;
+  } else {
+    if (non_js_ticks_ < kNonJSTicksThreshold) {
+      ++non_js_ticks_;
+    } else {
+      return RuntimeProfiler::WaitForSomeIsolateToEnterJS();
+    }
+  }
+  return false;
+}
+
+
+} }  // namespace v8::internal