Add DiscardableMemoryAllocator to work around FD limit issue.

base/memory/discardable_memory_allocator_android.cc

Index: base/memory/discardable_memory_allocator_android.cc
diff --git a/base/memory/discardable_memory_allocator_android.cc b/base/memory/discardable_memory_allocator_android.cc
new file mode 100644
index 0000000000000000000000000000000000000000..c17cc947a0a1d4c8562b4b6e13dca58bb1498dda
--- /dev/null
+++ b/base/memory/discardable_memory_allocator_android.cc
@@ -0,0 +1,416 @@
+// Copyright 2013 The Chromium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include "base/memory/discardable_memory_allocator_android.h"
+
+#include <algorithm>
+#include <cmath>
+#include <set>
+#include <utility>
+
+#include "base/basictypes.h"
+#include "base/containers/hash_tables.h"
+#include "base/logging.h"
+#include "base/memory/discardable_memory.h"
+#include "base/memory/discardable_memory_android.h"
+#include "base/memory/scoped_vector.h"
+#include "base/synchronization/lock.h"
+#include "base/threading/thread_checker.h"
+
+// The allocator consists of three parts (classes):
+// - DiscardableMemoryAllocator: entry point of all allocations (through its
+// Allocate() method) that are dispatched to the AshmemRegion instances (which
+// it owns).
+// - AshmemRegion: manages allocations and destructions inside a single large
+// (e.g. 32 MBytes) ashmem region.
+// - DiscardableAshmemChunk: class implementing the DiscardableMemory interface
+// whose instances are returned to the client. DiscardableAshmemChunk lets the
+// client seamlessly operate on a subrange of the ashmem region managed by
+// AshmemRegion.
+
+namespace base {
+namespace {
+
+// Only tolerate fragmentation in used chunks *caused by the client* (as opposed
+// to the allocator when a free chunk is reused). The client can cause such
+// fragmentation by e.g. requesting 4097 bytes. This size would be rounded up to
+// 8192 by the allocator which would cause 4095 bytes of fragmentation (which is
+// currently the maximum allowed). If the client requests 4096 bytes and a free
+// chunk of 8192 bytes is available then the free chunk gets splitted into two
+// pieces to minimize fragmentation (since 8192 - 4096 = 4096 which is greater
+// than 4095).
+// TODO(pliard): tune this if splitting chunks too often leads to performance
+// issues.
+const size_t kMaxChunkFragmentationBytes = 4096 - 1;
+
+}  // namespace
+
+namespace internal {
+
+class DiscardableMemoryAllocator::DiscardableAshmemChunk
+    : public DiscardableMemory {
+ public:
+  // Note that |ashmem_region| must outlive |this|.
+  DiscardableAshmemChunk(AshmemRegion* ashmem_region,
+                         int fd,
+                         void* address,
+                         size_t offset,
+                         size_t size)
+      : ashmem_region_(ashmem_region),
+        fd_(fd),
+        address_(address),
+        offset_(offset),
+        size_(size),
+        locked_(true) {
+  }
+
+  // Implemented below AshmemRegion since this requires the full definition of
+  // AshmemRegion.
+  virtual ~DiscardableAshmemChunk();
+
+  // DiscardableMemory:
+  virtual LockDiscardableMemoryStatus Lock() OVERRIDE {
+    DCHECK(!locked_);
+    locked_ = true;
+    return internal::LockAshmemRegion(fd_, offset_, size_, address_);
+  }
+
+  virtual void Unlock() OVERRIDE {
+    DCHECK(locked_);
+    locked_ = false;
+    internal::UnlockAshmemRegion(fd_, offset_, size_, address_);
+  }
+
+  virtual void* Memory() const OVERRIDE {
+    return address_;
+  }
+
+ private:
+  AshmemRegion* const ashmem_region_;
+  const int fd_;
+  void* const address_;
+  const size_t offset_;
+  const size_t size_;
+  bool locked_;
+
+  DISALLOW_COPY_AND_ASSIGN(DiscardableAshmemChunk);
+};
+
+class DiscardableMemoryAllocator::AshmemRegion {
+ public:
+  // Note that |allocator| must outlive |this|.
+  static scoped_ptr<AshmemRegion> Create(
+      size_t size,
+      const std::string& name,
+      DiscardableMemoryAllocator* allocator) {
+    int fd;
+    void* base;
+    if (!internal::CreateAshmemRegion(name.c_str(), size, &fd, &base))
+      return scoped_ptr<AshmemRegion>();
+    return make_scoped_ptr(new AshmemRegion(fd, size, base, allocator));
+  }
+
+  virtual ~AshmemRegion() {
+    const bool result = internal::CloseAshmemRegion(fd_, size_, base_);
+    DCHECK(result);
+  }
+
+  // Returns a new instance of DiscardableMemory whose size is greater or equal
+  // than |actual_size| (which is expected to be greater or equal than
+  // |client_requested_size|).
+  // Allocation works as follows:
+  // 1) Reuse a previously freed chunk and return it if it succeeded. See
+  // ReuseFreeChunk_Locked() below for more information.
+  // 2) If no free chunk could be reused and the region is not big enough for
+  // the requested size then NULL is returned.
+  // 3) If there is enough room in the ashmem region then a new chunk is
+  // returned. This new chunk starts at |offset_| which is the end of the
+  // previously highest chunk in the region.
+  scoped_ptr<DiscardableMemory> Allocate_Locked(size_t client_requested_size,
+                                                size_t actual_size) {
+    DCHECK_LE(client_requested_size, actual_size);
+    allocator_->lock_.AssertAcquired();
+    scoped_ptr<DiscardableMemory> memory = ReuseFreeChunk_Locked(
+        client_requested_size, actual_size);
+    if (memory)
+      return memory.Pass();
+    if (size_ - offset_ < actual_size) {
+      // This region does not have enough space left to hold the requested size.
+      return scoped_ptr<DiscardableMemory>();
+    }
+    void* const address = static_cast<char*>(base_) + offset_;
+    memory.reset(
+        new DiscardableAshmemChunk(this, fd_, address, offset_, actual_size));
+    used_to_previous_chunk_map_.insert(
+        std::make_pair(address, highest_allocated_chunk_));
+    highest_allocated_chunk_ = address;
+    offset_ += actual_size;
+    return memory.Pass();
+  }
+
+  void OnChunkDeletion(void* chunk, size_t size) OVERRIDE {
+    base::AutoLock auto_lock(allocator_->lock_);

[willchan no longer on Chromium, 2013/12/01 00:58:56]

Not sure if you care, but base:: is unnecessary, since this code is all in
base:: already.

[Philippe, 2013/12/02 10:56:39]

Yes, good point.

+    MergeAndAddFreeChunk_Locked(chunk, size);

[willchan no longer on Chromium, 2013/12/01 00:58:56]

Nit: I like to add comments after this indicating |this| may be deleted beyond
this point. Just so no one tries to dereference |this| somehow, like referencing
a member variable. This is a not uncommon bug.

Ditto for the end of MergeAndAddFreeChunk_Locked.

[Philippe, 2013/12/02 10:56:39]

Yes, good idea.

+  }
+
+ private:
+  struct FreeChunk {
+    FreeChunk(void* previous_chunk, void* start, size_t size)
+        : previous_chunk(previous_chunk),
+          start(start),
+          size(size) {
+    }
+
+    void* const previous_chunk;
+    void* const start;
+    const size_t size;
+
+    bool is_null() const { return !start; }
+
+    bool operator<(const FreeChunk& other) const {
+      return size < other.size;
+    }
+  };
+
+  // Note that |allocator| must outlive |this|.
+  AshmemRegion(int fd,
+               size_t size,
+               void* base,
+               DiscardableMemoryAllocator* allocator)
+      : fd_(fd),
+        size_(size),
+        base_(base),
+        allocator_(allocator),
+        highest_allocated_chunk_(NULL),
+        offset_(0) {
+    DCHECK_GE(fd_, 0);
+    DCHECK_GE(size, kMinAshmemRegionSize);
+    DCHECK(base);
+    DCHECK(allocator);
+  }
+
+  // Tries to reuse a previously freed chunk by doing a closest size match.
+  scoped_ptr<DiscardableMemory> ReuseFreeChunk_Locked(
+      size_t client_requested_size,
+      size_t actual_size) {
+    allocator_->lock_.AssertAcquired();
+    const FreeChunk reused_chunk = RemoveFreeChunkFromIterator_Locked(
+        free_chunks_.lower_bound(FreeChunk(NULL, NULL, actual_size)));
+    if (reused_chunk.is_null())
+      return scoped_ptr<DiscardableMemory>();
+
+    used_to_previous_chunk_map_.insert(
+        std::make_pair(reused_chunk.start, reused_chunk.previous_chunk));
+    size_t reused_chunk_size = reused_chunk.size;
+    // |client_requested_size| is used below rather than |actual_size| to
+    // reflect the amount of bytes that would not be usable by the client (i.e.
+    // wasted). Using |actual_size| instead would not allow us to detect
+    // fragmentation caused by the client if he did misaligned allocations.
+    const size_t fragmentation_bytes =
+        reused_chunk.size - client_requested_size;

[willchan no longer on Chromium, 2013/12/01 00:58:56]

Nit: add a DCHECK_GE(reused_chunk.size, client_requested_size)?

[Philippe, 2013/12/02 10:56:39]

Yeah, good idea.

+    if (fragmentation_bytes > kMaxChunkFragmentationBytes) {
+      // Split the free chunk being recycled so that its unused tail doesn't get
+      // reused (i.e. locked) which would prevent it from being evicted under
+      // memory pressure.
+      reused_chunk_size = actual_size;
+      void* const previous_chunk = reused_chunk.start;

[willchan no longer on Chromium, 2013/12/01 00:58:56]

I think it'd be clearer if you deleted this. Since the previous chunk is the
reused_chunk.

[Philippe, 2013/12/02 10:56:39]

Done.

+      void* const new_chunk_start =
+          static_cast<char*>(reused_chunk.start) + actual_size;
+      DCHECK_GT(reused_chunk.size, actual_size);
+      const size_t new_chunk_size = reused_chunk.size - actual_size;
+      // Note that merging is not needed here since there can't be contiguous
+      // free chunks at this point.
+      AddFreeChunk_Locked(
+          FreeChunk(previous_chunk, new_chunk_start, new_chunk_size));
+    }
+    const size_t offset =
+        static_cast<char*>(reused_chunk.start) - static_cast<char*>(base_);
+    internal::LockAshmemRegion(
+        fd_, offset, reused_chunk_size, reused_chunk.start);
+    scoped_ptr<DiscardableMemory> memory(
+        new DiscardableAshmemChunk(this, fd_, reused_chunk.start, offset,
+                                   reused_chunk_size));
+    return memory.Pass();
+  }
+
+  // Makes the chunk identified with the provided arguments free and possibly
+  // merges this chunk with the previous and next contiguous ones.
+  // If the provided chunk is the only one used (and going to be freed) in the
+  // region then the internal ashmem region is closed so that the underlying
+  // physical pages are immediately released.
+  // Note that free chunks are unlocked therefore they can be reclaimed by the
+  // kernel if needed (under memory pressure) but they are not immediately
+  // released unfortunately since madvise(MADV_REMOVE) and
+  // fallocate(FALLOC_FL_PUNCH_HOLE) don't seem to work on ashmem. This might
+  // change in versions of kernel >=3.5 though. The fact that free chunks are
+  // not immediately released is the reason why we are trying to minimize
+  // fragmentation in order not to cause "artificial" memory pressure.
+  void MergeAndAddFreeChunk_Locked(void* chunk, size_t size) {
+    allocator_->lock_.AssertAcquired();
+    size_t new_free_chunk_size = size;
+    // Merge with the previous chunk.
+    void* first_free_chunk = chunk;
+    DCHECK_NE(0U, used_to_previous_chunk_map_.size());

[willchan no longer on Chromium, 2013/12/01 00:58:56]

DCHECK(!used_to_previous_chunk_map_.empty())

Some STL containers have a non-constant .size() impl.

[Philippe, 2013/12/02 10:56:39]

I used this pattern to have more detailed assertion messages but that's true
that we don't really care about the specific size of the container here in case
the assertion fails.

+    const base::hash_map<void*, void*>::iterator previous_chunk_it =
+        used_to_previous_chunk_map_.find(chunk);
+    DCHECK(previous_chunk_it != used_to_previous_chunk_map_.end());
+    void* previous_chunk = previous_chunk_it->second;
+    used_to_previous_chunk_map_.erase(previous_chunk_it);
+    if (previous_chunk) {
+      const FreeChunk free_chunk = RemoveFreeChunk_Locked(previous_chunk);
+      if (!free_chunk.is_null()) {
+        new_free_chunk_size += free_chunk.size;
+        first_free_chunk = previous_chunk;
+        // There should not be more contiguous previous free chunks.
+        DCHECK(!address_to_free_chunk_map_.count(free_chunk.previous_chunk));
+      }
+    }
+    // Merge with the next chunk if free and present.
+    void* next_chunk = static_cast<char*>(chunk) + size;
+    const FreeChunk next_free_chunk = RemoveFreeChunk_Locked(next_chunk);
+    if (!next_free_chunk.is_null()) {
+      new_free_chunk_size += next_free_chunk.size;
+      // Same as above.
+      DCHECK(!address_to_free_chunk_map_.count(static_cast<char*>(next_chunk) +
+                                               next_free_chunk.size));
+    }
+    const bool whole_ashmem_region_is_free =
+        used_to_previous_chunk_map_.empty();
+    if (!whole_ashmem_region_is_free) {
+      AddFreeChunk_Locked(
+          FreeChunk(previous_chunk, first_free_chunk, new_free_chunk_size));
+      return;
+    }
+    // The whole ashmem region is free thus it can be deleted.
+    DCHECK_EQ(base_, first_free_chunk);
+    DCHECK_EQ(0U, free_chunks_.size());

[willchan no longer on Chromium, 2013/12/01 00:58:56]

I'd use .empty() for these guys too.

[Philippe, 2013/12/02 10:56:39]

Done.

+    DCHECK_EQ(0U, address_to_free_chunk_map_.size());
+    DCHECK_EQ(0U, used_to_previous_chunk_map_.size());
+    allocator_->DeleteAshmemRegion_Locked(this);
+  }
+
+  void AddFreeChunk_Locked(const FreeChunk& free_chunk) {
+    allocator_->lock_.AssertAcquired();
+    const std::multiset<FreeChunk>::iterator it = free_chunks_.insert(
+        free_chunk);
+    address_to_free_chunk_map_.insert(std::make_pair(free_chunk.start, it));
+    // Update the next used contiguous chunk, if any, since its previous chunk
+    // may have changed due to free chunks merging/splitting.
+    void* const next_used_contiguous_chunk =
+        static_cast<char*>(free_chunk.start) + free_chunk.size;
+    base::hash_map<void*, void*>::iterator previous_it =
+        used_to_previous_chunk_map_.find(next_used_contiguous_chunk);
+    if (previous_it != used_to_previous_chunk_map_.end())
+      previous_it->second = free_chunk.start;
+  }
+
+  // Finds and removes the free chunk, if any, whose start address is
+  // |chunk_start|. Returns a copy of the unlinked free chunk or a free chunk
+  // whose content is null if it was not found.
+  FreeChunk RemoveFreeChunk_Locked(void* chunk_start) {
+    allocator_->lock_.AssertAcquired();
+    const base::hash_map<
+        void*, std::multiset<FreeChunk>::iterator>::iterator it =
+            address_to_free_chunk_map_.find(chunk_start);
+    if (it == address_to_free_chunk_map_.end())
+      return FreeChunk(NULL, NULL, 0U);
+    return RemoveFreeChunkFromIterator_Locked(it->second);
+  }
+
+  // Same as above but takes an iterator in.
+  FreeChunk RemoveFreeChunkFromIterator_Locked(
+      std::multiset<FreeChunk>::iterator free_chunk_it) {
+    allocator_->lock_.AssertAcquired();
+    if (free_chunk_it == free_chunks_.end())
+      return FreeChunk(NULL, NULL, 0U);
+    DCHECK(free_chunk_it != free_chunks_.end());
+    const FreeChunk free_chunk(*free_chunk_it);
+    address_to_free_chunk_map_.erase(free_chunk_it->start);
+    free_chunks_.erase(free_chunk_it);
+    return free_chunk;
+  }
+
+  const int fd_;
+  const size_t size_;
+  void* const base_;
+  DiscardableMemoryAllocator* const allocator_;
+  void* highest_allocated_chunk_;
+  // Points to the end of |highest_allocated_chunk_|.
+  size_t offset_;
+  // Allows free chunks recycling (lookup, insertion and removal) in O(log N).
+  // Note that FreeChunk values are indexed by their size and also note that
+  // multiple free chunks can have the same size (which is why multiset<> is
+  // used instead of e.g. set<>).
+  std::multiset<FreeChunk> free_chunks_;
+  // Used while merging free contiguous chunks to erase free chunks (from their
+  // start address) in constant time. Note that multiset<>::{insert,erase}()
+  // don't invalidate iterators (except the one for the element being removed
+  // obviously).
+  base::hash_map<
+      void*, std::multiset<FreeChunk>::iterator> address_to_free_chunk_map_;
+  // Maps the address of *used* chunks to the address of their previous
+  // contiguous chunk.
+  base::hash_map<void*, void*> used_to_previous_chunk_map_;
+
+  DISALLOW_COPY_AND_ASSIGN(AshmemRegion);
+};
+
+DiscardableMemoryAllocator::DiscardableAshmemChunk::~DiscardableAshmemChunk() {
+  if (locked_)
+    internal::UnlockAshmemRegion(fd_, offset_, size_, address_);
+  ashmem_region_->OnChunkDeletion(address_, size_);
+}
+
+DiscardableMemoryAllocator::DiscardableMemoryAllocator(const std::string& name)
+    : name_(name) {
+}
+
+DiscardableMemoryAllocator::~DiscardableMemoryAllocator() {
+  DCHECK(thread_checker_.CalledOnValidThread());
+  DCHECK(ashmem_regions_.empty());
+}
+
+scoped_ptr<DiscardableMemory> DiscardableMemoryAllocator::Allocate(
+    size_t size) {
+  const size_t aligned_size = internal::AlignToNextPage(size);
+  // TODO(pliard): make this function less naive by e.g. moving the free chunks
+  // multiset to the allocator itself in order to decrease even more
+  // fragmentation/speedup allocation. Note that there should not be more than a
+  // couple (=5) of AshmemRegion instances in practice though.
+  base::AutoLock auto_lock(lock_);
+  DCHECK_LE(ashmem_regions_.size(), 5U);
+  for (ScopedVector<AshmemRegion>::iterator it = ashmem_regions_.begin();
+       it != ashmem_regions_.end(); ++it) {
+    scoped_ptr<DiscardableMemory> memory(
+        (*it)->Allocate_Locked(size, aligned_size));
+    if (memory)
+      return memory.Pass();
+  }
+  scoped_ptr<AshmemRegion> new_region(
+      AshmemRegion::Create(
+          std::max(static_cast<size_t>(kMinAshmemRegionSize), aligned_size),
+          name_.c_str(), this));
+  if (!new_region) {
+    // TODO(pliard): consider adding an histogram to see how often this happens.
+    return scoped_ptr<DiscardableMemory>();
+  }
+  ashmem_regions_.push_back(new_region.release());
+  return ashmem_regions_.back()->Allocate_Locked(size, aligned_size);
+}
+
+void DiscardableMemoryAllocator::DeleteAshmemRegion_Locked(
+    AshmemRegion* region) {
+  lock_.AssertAcquired();
+  // Note that there should not be more than a couple of ashmem region instances
+  // in |ashmem_regions_|.
+  DCHECK_LE(ashmem_regions_.size(), 5U);
+  const ScopedVector<AshmemRegion>::iterator it = std::find(
+      ashmem_regions_.begin(), ashmem_regions_.end(), region);
+  DCHECK_NE(ashmem_regions_.end(), it);
+  std::swap(*it, ashmem_regions_.back());
+  ashmem_regions_.pop_back();
+}
+
+}  // namespace internal
+}  // namespace base