Disk cache: Reference CL for the implementation of file format version 3.

net/disk_cache/v3/disk_format_v3.h

+// Copyright (c) 2011 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.
+
+// The cache is stored on disk as a collection of block-files, plus an index
+// plus a collection of external files.
+//
+// Any data blob bigger than kMaxBlockSize (disk_cache/addr.h) will be stored in
+// a separate file named f_xxx where x is a hexadecimal number. Shorter data
+// will be stored as a series of blocks on a block-file. In any case, CacheAddr
+// represents the address of the data inside the cache.
+//
+// The index is actually a collection of four files that store a hash table with
+// allocation bitmaps and backup data. Hash collisions are handled directly by
+// the table, which from some point of view behaves like a 4-way associative
+// cache with overflow buckets (so not really open addressing).
+//
+// Basically the hash table is a collection of buckets. The first part of the
+// table has a fixed number of buckets and it is directly addressed by the hash,
+// while the second part of the table (stored on a second file) has a variable
+// number of buckets. Each bucket stores up to four cells (each cell represents
+// a possibl entry). The index bitmap tracks the state of individual cells.
+//
+// The last element of the cache is the block-file. A block file is a file
+// designed to store blocks of data of a given size. For more details see
+// disk_cache/disk_format_base.h
+//
+// A new cache is initialized with a set of block files (named data_0 through
+// data_6), each one dedicated to store blocks of a given size or function. The
+// number at the end of the file name is the block file number (in decimal).
+//
+// There are three "special" types of blocks: normal entries, evicted entries
+// and control data for external files.
+//
+// The files that store internal information for the cache (blocks and index)
+// are memory mapped. They have a location that is signaled every time the
+// internal structures are modified, so it is possible to detect (most of the
+// time) when the process dies in the middle of an update. There are dedicated
+// backup files for cache bitmaps, used to detect entries out of date.
+//
+// Although cache files are to be consumed on the same machine that creates
+// them, if files are to be moved accross machines, little endian storage is
+// assumed.
+
+#ifndef NET_DISK_CACHE_V3_DISK_FORMAT_V3_H_
+#define NET_DISK_CACHE_V3_DISK_FORMAT_V3_H_
+
+#include "base/basictypes.h"
+#include "net/disk_cache/disk_format_base.h"
+
+namespace disk_cache {
+
+const int kBaseTableLen = 0x400;
+const uint32 kIndexMagicV3 = 0xC103CAC3;
+const uint32 kVersion3 = 0x30000;  // Version 3.0.
+
+// Flags for a given cache.
+enum CacheFlags {
+  SMALL_CACHE = 1 << 0,       // See IndexCell.
+  CACHE_EVICTION_2 = 1 << 1,  // Keep multiple lists for eviction.
+  CACHE_EVICTED = 1 << 2      // Already evicted at least one entry.
+};
+
+// Header for the master index file.
+struct IndexHeaderV3 {
+  uint32      magic;
+  uint32      version;
+  int32       num_entries;   // Number of entries currently stored.
+  int32       num_bytes;     // Total size of the stored data.
+  int32       last_file;     // Last external file created.
+  int32       reserved1;
+  CacheAddr   stats;         // Storage for usage data.
+  int32       table_len;     // Actual size of the table.
+  int32       crash;         // Signals a previous crash.
+  int32       experiment;    // Id of an ongoing test.
+  int32       max_bytes;     // Total maximum size of the stored data.
+  uint32      flags;
+  int32       used_cells;
+  int32       max_bucket;
+  uint64      create_time;   // Creation time for this set of files.
+  uint64      base_time;     // Current base for timestamps.
+  uint64      old_time;      // Previous time used for timestamps.
+  int32       max_block_file;
+  int32       num_no_use_entries;
+  int32       num_low_use_entries;
+  int32       num_high_use_entries;
+  int32       reserved;
+  int32       num_evicted_entries;
+  int32       pad[6];
+};
+
+const int kBaseBitmapBytes = 3968;
+// The IndexBitmap is directly saved to a file named index. The file grows in
+// page increments (4096 bytes), but all bits don't have to be in use at any
+// given time. The required file size can be computed from header.table_len.
+struct IndexBitmap {
+  IndexHeaderV3   header;
+  uint32          bitmap[kBaseBitmapBytes / 4];  // First page of the bitmap.
+};
+COMPILE_ASSERT(sizeof(IndexBitmap) == 4096, bad_IndexHeader);
+
+// Possible states for a given entry.
+enum EntryState {
+  ENTRY_FREE = 0,   // Available slot.
+  ENTRY_NEW,        // The entry is being created.
+  ENTRY_OPEN,       // The entry is being accessed.
+  ENTRY_MODIFIED,   // The entry is being modified.
+  ENTRY_DELETED,    // The entry is being deleted.
+  ENTRY_FIXING,     // Inconsistent state. The entry is being verified.
+  ENTRY_USED        // The slot is in use (entry is present).
+};
+COMPILE_ASSERT(ENTRY_USED <= 7, state_uses_3_bits);
+
+enum EntryGroup {
+  ENTRY_NO_USE = 0,   // The entry has not been reused.
+  ENTRY_LOW_USE,      // The entry has low reuse.
+  ENTRY_HIGH_USE,     // The entry has high reuse.
+  ENTRY_RESERVED,     // Reserved for future use.
+  ENTRY_EVICTED       // The entry was deleted.
+};
+COMPILE_ASSERT(ENTRY_USED <= 7, group_uses_3_bits);
+
+#pragma pack(push, 1)
+struct IndexCell {
+  void Clear() { memset(this, 0, sizeof(*this)); }
+
+  // A cell is a 9 byte bit-field that stores 7 values:
+  //   address : 22 bits
+  //   hash : 18 bits
+  //   timestamp : 20 bits
+  //   reuse : 4 bits
+  //   state : 3 bits
+  //   group : 3 bits
+  //   sum : 2 bits
+  // The actual layout is as follows:
+  //
+  // first_part (low order 32 bits):
+  //   0000 0000 0011 1111 1111 1111 1111 1111 : address
+  //   1111 1111 1100 0000 0000 0000 0000 0000 : hash
+  //
+  // first_part (high order 32 bits):
+  //   0000 0000 0000 0000 0000 0000 1111 1111 : hash
+  //   0000 1111 1111 1111 1111 1111 0000 0000 : timestamp
+  //   1111 0000 0000 0000 0000 0000 0000 0000 : reuse
+  //
+  // last_part:
+  //   0000 0111 : state
+  //   0011 1000 : group
+  //   1100 0000 : sum
+  //
+  // The small-cache version of the format moves some bits from the address to
+  // the hash fileds, like so:
+  //   address : 16 bits
+  //   hash : 24 bits
+  //
+  // first_part (low order 32 bits):
+  //   0000 0000 0000 0000 1111 1111 1111 1111 : address
+  //   1111 1111 1111 1111 0000 0000 0000 0000 : hash
+  //
+  // The actual bit distribution between address and hash is determined by the
+  // table size (IndexHeaderV3.table_len). Tables smaller than 65536 entries
+  // use the small-cache version; after that size, caches should have the
+  // SMALL_CACHE flag cleared.
+  //
+  // To locate a given entry after recovering the address from the cell, the
+  // file type and file number are appended (see disk_cache/addr.h). For a large
+  // table only the file type is implied; for a small table, the file number
+  // is also implied, and it should be the first file for that type of entry,
+  // as determined by the EntryGroup (two files in total, one for active entries
+  // and another one for evicted entries).
+  //
+  // For example, a small table may store something like 0x1234 as the address
+  // field. That means it stores the entry number 0x1234. If that record belongs
+  // to a deleted entry, the regular cache address may look something like
+  //     BLOCK_EVICTED + 1 block + file number 6 + entry number 0x1234
+  //     so Addr = 0xf0061234
+  //
+  // If that same Addr is stored on a large table, the address field would be
+  // 0x61234
+
+  uint64      first_part;
+  uint8       last_part;
+};
+COMPILE_ASSERT(sizeof(IndexCell) == 9, bad_IndexCell);
+
+const int kCellsPerBucket = 4;
+struct IndexBucket {
+  IndexCell   cells[kCellsPerBucket];
+  int32       next;
+  uint32      hash;  // The high order byte is reserved (should be zero).
+};
+COMPILE_ASSERT(sizeof(IndexBucket) == 44, bad_IndexBucket);
+const int kBytesPerCell = 44 / kCellsPerBucket;
+
+// The main cache index. Backed by a file named index_tb1.
+// The extra table (index_tb2) has a similar format, but different size.
+struct Index {
+  // Default size. Actual size controlled by header.table_len.
+  IndexBucket table[kBaseTableLen / kCellsPerBucket];
+};
+#pragma pack(pop)
+
+// Flags that can be applied to an entry.
+enum EntryFlags {
+  PARENT_ENTRY = 1,         // This entry has children (sparse) entries.
+  CHILD_ENTRY = 1 << 1      // Child entry that stores sparse data.
+};
+
+struct EntryRecord {
+  uint32      hash;
+  uint32      pad1;
+  uint8       reuse_count;
+  uint8       refetch_count;
+  int8        state;              // Current EntryState.
+  uint8       flags;              // Any combination of EntryFlags.
+  int32       key_len;
+  int32       data_size[4];       // We can store up to 4 data streams for each
+  CacheAddr   data_addr[4];       // entry.
+  uint32      data_hash[4];
+  uint64      creation_time;
+  uint64      last_modified_time;
+  uint64      last_access_time;
+  int32       pad[3];
+  uint32      self_hash;
+};
+COMPILE_ASSERT(sizeof(EntryRecord) == 104, bad_EntryRecord);
+
+struct ShortEntryRecord {
+  uint32      hash;
+  uint32      pad1;
+  uint8       reuse_count;
+  uint8       refetch_count;
+  int8        state;              // Current EntryState.
+  uint8       flags;
+  int32       key_len;
+  uint64      last_access_time;
+  uint32      long_hash[5];
+  uint32      self_hash;
+};
+COMPILE_ASSERT(sizeof(ShortEntryRecord) == 48, bad_ShortEntryRecord);
+
+}  // namespace disk_cache
+
+#endif  // NET_DISK_CACHE_V3_DISK_FORMAT_V3_H_