--- sqlite3.c.sqlite 2013-11-09 12:42:08.623520057 +0100 +++ sqlite3.c 2013-11-09 12:41:58.695520274 +0100 @@ -12566,9 +12566,45 @@ #endif /* _SQLITEINT_H_ */ /************** End of sqliteInt.h *******************************************/ -/************** Begin file global.c ******************************************/ +/************** Begin file crypto.c ******************************************/ +/* +** SQLCipher +** http://sqlcipher.net +** +** Copyright (c) 2008 - 2013, ZETETIC LLC +** 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 the ZETETIC LLC 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 ZETETIC LLC ''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 ZETETIC LLC 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. +** +*/ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC + +/* #include */ +/************** Include btreeInt.h in the middle of crypto.c *****************/ +/************** Begin file btreeInt.h ****************************************/ /* -** 2008 June 13 +** 2004 April 6 ** ** The author disclaims copyright to this source code. In place of ** a legal notice, here is a blessing: @@ -12578,246 +12614,3589 @@ ** May you share freely, never taking more than you give. ** ************************************************************************* +** This file implements a external (disk-based) database using BTrees. +** For a detailed discussion of BTrees, refer to ** -** This file contains definitions of global variables and contants. -*/ - -/* An array to map all upper-case characters into their corresponding -** lower-case character. +** Donald E. Knuth, THE ART OF COMPUTER PROGRAMMING, Volume 3: +** "Sorting And Searching", pages 473-480. Addison-Wesley +** Publishing Company, Reading, Massachusetts. ** -** SQLite only considers US-ASCII (or EBCDIC) characters. We do not -** handle case conversions for the UTF character set since the tables -** involved are nearly as big or bigger than SQLite itself. -*/ -SQLITE_PRIVATE const unsigned char sqlite3UpperToLower[] = { -#ifdef SQLITE_ASCII - 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, - 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, - 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, - 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 97, 98, 99,100,101,102,103, - 104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121, - 122, 91, 92, 93, 94, 95, 96, 97, 98, 99,100,101,102,103,104,105,106,107, - 108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125, - 126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143, - 144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161, - 162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179, - 180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197, - 198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215, - 216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232,233, - 234,235,236,237,238,239,240,241,242,243,244,245,246,247,248,249,250,251, - 252,253,254,255 -#endif -#ifdef SQLITE_EBCDIC - 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, /* 0x */ - 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, /* 1x */ - 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, /* 2x */ - 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, /* 3x */ - 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, /* 4x */ - 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, /* 5x */ - 96, 97, 66, 67, 68, 69, 70, 71, 72, 73,106,107,108,109,110,111, /* 6x */ - 112, 81, 82, 83, 84, 85, 86, 87, 88, 89,122,123,124,125,126,127, /* 7x */ - 128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143, /* 8x */ - 144,145,146,147,148,149,150,151,152,153,154,155,156,157,156,159, /* 9x */ - 160,161,162,163,164,165,166,167,168,169,170,171,140,141,142,175, /* Ax */ - 176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191, /* Bx */ - 192,129,130,131,132,133,134,135,136,137,202,203,204,205,206,207, /* Cx */ - 208,145,146,147,148,149,150,151,152,153,218,219,220,221,222,223, /* Dx */ - 224,225,162,163,164,165,166,167,168,169,232,203,204,205,206,207, /* Ex */ - 239,240,241,242,243,244,245,246,247,248,249,219,220,221,222,255, /* Fx */ -#endif -}; - -/* -** The following 256 byte lookup table is used to support SQLites built-in -** equivalents to the following standard library functions: +** The basic idea is that each page of the file contains N database +** entries and N+1 pointers to subpages. ** -** isspace() 0x01 -** isalpha() 0x02 -** isdigit() 0x04 -** isalnum() 0x06 -** isxdigit() 0x08 -** toupper() 0x20 -** SQLite identifier character 0x40 +** ---------------------------------------------------------------- +** | Ptr(0) | Key(0) | Ptr(1) | Key(1) | ... | Key(N-1) | Ptr(N) | +** ---------------------------------------------------------------- ** -** Bit 0x20 is set if the mapped character requires translation to upper -** case. i.e. if the character is a lower-case ASCII character. -** If x is a lower-case ASCII character, then its upper-case equivalent -** is (x - 0x20). Therefore toupper() can be implemented as: +** All of the keys on the page that Ptr(0) points to have values less +** than Key(0). All of the keys on page Ptr(1) and its subpages have +** values greater than Key(0) and less than Key(1). All of the keys +** on Ptr(N) and its subpages have values greater than Key(N-1). And +** so forth. ** -** (x & ~(map[x]&0x20)) +** Finding a particular key requires reading O(log(M)) pages from the +** disk where M is the number of entries in the tree. ** -** Standard function tolower() is implemented using the sqlite3UpperToLower[] -** array. tolower() is used more often than toupper() by SQLite. +** In this implementation, a single file can hold one or more separate +** BTrees. Each BTree is identified by the index of its root page. The +** key and data for any entry are combined to form the "payload". A +** fixed amount of payload can be carried directly on the database +** page. If the payload is larger than the preset amount then surplus +** bytes are stored on overflow pages. The payload for an entry +** and the preceding pointer are combined to form a "Cell". Each +** page has a small header which contains the Ptr(N) pointer and other +** information such as the size of key and data. ** -** Bit 0x40 is set if the character non-alphanumeric and can be used in an -** SQLite identifier. Identifiers are alphanumerics, "_", "$", and any -** non-ASCII UTF character. Hence the test for whether or not a character is -** part of an identifier is 0x46. +** FORMAT DETAILS ** -** SQLite's versions are identical to the standard versions assuming a -** locale of "C". They are implemented as macros in sqliteInt.h. +** The file is divided into pages. The first page is called page 1, +** the second is page 2, and so forth. A page number of zero indicates +** "no such page". The page size can be any power of 2 between 512 and 65536. +** Each page can be either a btree page, a freelist page, an overflow +** page, or a pointer-map page. +** +** The first page is always a btree page. The first 100 bytes of the first +** page contain a special header (the "file header") that describes the file. +** The format of the file header is as follows: +** +** OFFSET SIZE DESCRIPTION +** 0 16 Header string: "SQLite format 3\000" +** 16 2 Page size in bytes. +** 18 1 File format write version +** 19 1 File format read version +** 20 1 Bytes of unused space at the end of each page +** 21 1 Max embedded payload fraction +** 22 1 Min embedded payload fraction +** 23 1 Min leaf payload fraction +** 24 4 File change counter +** 28 4 Reserved for future use +** 32 4 First freelist page +** 36 4 Number of freelist pages in the file +** 40 60 15 4-byte meta values passed to higher layers +** +** 40 4 Schema cookie +** 44 4 File format of schema layer +** 48 4 Size of page cache +** 52 4 Largest root-page (auto/incr_vacuum) +** 56 4 1=UTF-8 2=UTF16le 3=UTF16be +** 60 4 User version +** 64 4 Incremental vacuum mode +** 68 4 unused +** 72 4 unused +** 76 4 unused +** +** All of the integer values are big-endian (most significant byte first). +** +** The file change counter is incremented when the database is changed +** This counter allows other processes to know when the file has changed +** and thus when they need to flush their cache. +** +** The max embedded payload fraction is the amount of the total usable +** space in a page that can be consumed by a single cell for standard +** B-tree (non-LEAFDATA) tables. A value of 255 means 100%. The default +** is to limit the maximum cell size so that at least 4 cells will fit +** on one page. Thus the default max embedded payload fraction is 64. +** +** If the payload for a cell is larger than the max payload, then extra +** payload is spilled to overflow pages. Once an overflow page is allocated, +** as many bytes as possible are moved into the overflow pages without letting +** the cell size drop below the min embedded payload fraction. +** +** The min leaf payload fraction is like the min embedded payload fraction +** except that it applies to leaf nodes in a LEAFDATA tree. The maximum +** payload fraction for a LEAFDATA tree is always 100% (or 255) and it +** not specified in the header. +** +** Each btree pages is divided into three sections: The header, the +** cell pointer array, and the cell content area. Page 1 also has a 100-byte +** file header that occurs before the page header. +** +** |----------------| +** | file header | 100 bytes. Page 1 only. +** |----------------| +** | page header | 8 bytes for leaves. 12 bytes for interior nodes +** |----------------| +** | cell pointer | | 2 bytes per cell. Sorted order. +** | array | | Grows downward +** | | v +** |----------------| +** | unallocated | +** | space | +** |----------------| ^ Grows upwards +** | cell content | | Arbitrary order interspersed with freeblocks. +** | area | | and free space fragments. +** |----------------| +** +** The page headers looks like this: +** +** OFFSET SIZE DESCRIPTION +** 0 1 Flags. 1: intkey, 2: zerodata, 4: leafdata, 8: leaf +** 1 2 byte offset to the first freeblock +** 3 2 number of cells on this page +** 5 2 first byte of the cell content area +** 7 1 number of fragmented free bytes +** 8 4 Right child (the Ptr(N) value). Omitted on leaves. +** +** The flags define the format of this btree page. The leaf flag means that +** this page has no children. The zerodata flag means that this page carries +** only keys and no data. The intkey flag means that the key is a integer +** which is stored in the key size entry of the cell header rather than in +** the payload area. +** +** The cell pointer array begins on the first byte after the page header. +** The cell pointer array contains zero or more 2-byte numbers which are +** offsets from the beginning of the page to the cell content in the cell +** content area. The cell pointers occur in sorted order. The system strives +** to keep free space after the last cell pointer so that new cells can +** be easily added without having to defragment the page. +** +** Cell content is stored at the very end of the page and grows toward the +** beginning of the page. +** +** Unused space within the cell content area is collected into a linked list of +** freeblocks. Each freeblock is at least 4 bytes in size. The byte offset +** to the first freeblock is given in the header. Freeblocks occur in +** increasing order. Because a freeblock must be at least 4 bytes in size, +** any group of 3 or fewer unused bytes in the cell content area cannot +** exist on the freeblock chain. A group of 3 or fewer free bytes is called +** a fragment. The total number of bytes in all fragments is recorded. +** in the page header at offset 7. +** +** SIZE DESCRIPTION +** 2 Byte offset of the next freeblock +** 2 Bytes in this freeblock +** +** Cells are of variable length. Cells are stored in the cell content area at +** the end of the page. Pointers to the cells are in the cell pointer array +** that immediately follows the page header. Cells is not necessarily +** contiguous or in order, but cell pointers are contiguous and in order. +** +** Cell content makes use of variable length integers. A variable +** length integer is 1 to 9 bytes where the lower 7 bits of each +** byte are used. The integer consists of all bytes that have bit 8 set and +** the first byte with bit 8 clear. The most significant byte of the integer +** appears first. A variable-length integer may not be more than 9 bytes long. +** As a special case, all 8 bytes of the 9th byte are used as data. This +** allows a 64-bit integer to be encoded in 9 bytes. +** +** 0x00 becomes 0x00000000 +** 0x7f becomes 0x0000007f +** 0x81 0x00 becomes 0x00000080 +** 0x82 0x00 becomes 0x00000100 +** 0x80 0x7f becomes 0x0000007f +** 0x8a 0x91 0xd1 0xac 0x78 becomes 0x12345678 +** 0x81 0x81 0x81 0x81 0x01 becomes 0x10204081 +** +** Variable length integers are used for rowids and to hold the number of +** bytes of key and data in a btree cell. +** +** The content of a cell looks like this: +** +** SIZE DESCRIPTION +** 4 Page number of the left child. Omitted if leaf flag is set. +** var Number of bytes of data. Omitted if the zerodata flag is set. +** var Number of bytes of key. Or the key itself if intkey flag is set. +** * Payload +** 4 First page of the overflow chain. Omitted if no overflow +** +** Overflow pages form a linked list. Each page except the last is completely +** filled with data (pagesize - 4 bytes). The last page can have as little +** as 1 byte of data. +** +** SIZE DESCRIPTION +** 4 Page number of next overflow page +** * Data +** +** Freelist pages come in two subtypes: trunk pages and leaf pages. The +** file header points to the first in a linked list of trunk page. Each trunk +** page points to multiple leaf pages. The content of a leaf page is +** unspecified. A trunk page looks like this: +** +** SIZE DESCRIPTION +** 4 Page number of next trunk page +** 4 Number of leaf pointers on this page +** * zero or more pages numbers of leaves */ -#ifdef SQLITE_ASCII -SQLITE_PRIVATE const unsigned char sqlite3CtypeMap[256] = { - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 00..07 ........ */ - 0x00, 0x01, 0x01, 0x01, 0x01, 0x01, 0x00, 0x00, /* 08..0f ........ */ - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 10..17 ........ */ - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 18..1f ........ */ - 0x01, 0x00, 0x00, 0x00, 0x40, 0x00, 0x00, 0x00, /* 20..27 !"#$%&' */ - 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 28..2f ()*+,-./ */ - 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, /* 30..37 01234567 */ - 0x0c, 0x0c, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 38..3f 89:;<=>? */ - - 0x00, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x02, /* 40..47 @ABCDEFG */ - 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, /* 48..4f HIJKLMNO */ - 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, /* 50..57 PQRSTUVW */ - 0x02, 0x02, 0x02, 0x00, 0x00, 0x00, 0x00, 0x40, /* 58..5f XYZ[\]^_ */ - 0x00, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x22, /* 60..67 `abcdefg */ - 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, /* 68..6f hijklmno */ - 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, /* 70..77 pqrstuvw */ - 0x22, 0x22, 0x22, 0x00, 0x00, 0x00, 0x00, 0x00, /* 78..7f xyz{|}~. */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 80..87 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 88..8f ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 90..97 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 98..9f ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* a0..a7 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* a8..af ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* b0..b7 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* b8..bf ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* c0..c7 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* c8..cf ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* d0..d7 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* d8..df ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* e0..e7 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* e8..ef ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* f0..f7 ........ */ - 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40 /* f8..ff ........ */ -}; -#endif +/* The following value is the maximum cell size assuming a maximum page +** size give above. +*/ +#define MX_CELL_SIZE(pBt) ((int)(pBt->pageSize-8)) -#ifndef SQLITE_USE_URI -# define SQLITE_USE_URI 0 -#endif +/* The maximum number of cells on a single page of the database. This +** assumes a minimum cell size of 6 bytes (4 bytes for the cell itself +** plus 2 bytes for the index to the cell in the page header). Such +** small cells will be rare, but they are possible. +*/ +#define MX_CELL(pBt) ((pBt->pageSize-8)/6) -#ifndef SQLITE_ALLOW_COVERING_INDEX_SCAN -# define SQLITE_ALLOW_COVERING_INDEX_SCAN 1 -#endif +/* Forward declarations */ +typedef struct MemPage MemPage; +typedef struct BtLock BtLock; /* -** The following singleton contains the global configuration for -** the SQLite library. +** This is a magic string that appears at the beginning of every +** SQLite database in order to identify the file as a real database. +** +** You can change this value at compile-time by specifying a +** -DSQLITE_FILE_HEADER="..." on the compiler command-line. The +** header must be exactly 16 bytes including the zero-terminator so +** the string itself should be 15 characters long. If you change +** the header, then your custom library will not be able to read +** databases generated by the standard tools and the standard tools +** will not be able to read databases created by your custom library. */ -SQLITE_PRIVATE SQLITE_WSD struct Sqlite3Config sqlite3Config = { - SQLITE_DEFAULT_MEMSTATUS, /* bMemstat */ - 1, /* bCoreMutex */ - SQLITE_THREADSAFE==1, /* bFullMutex */ - SQLITE_USE_URI, /* bOpenUri */ - SQLITE_ALLOW_COVERING_INDEX_SCAN, /* bUseCis */ - 0x7ffffffe, /* mxStrlen */ - 128, /* szLookaside */ - 500, /* nLookaside */ - {0,0,0,0,0,0,0,0}, /* m */ - {0,0,0,0,0,0,0,0,0}, /* mutex */ - {0,0,0,0,0,0,0,0,0,0,0,0,0},/* pcache2 */ - (void*)0, /* pHeap */ - 0, /* nHeap */ - 0, 0, /* mnHeap, mxHeap */ - SQLITE_DEFAULT_MMAP_SIZE, /* szMmap */ - SQLITE_MAX_MMAP_SIZE, /* mxMmap */ - (void*)0, /* pScratch */ - 0, /* szScratch */ - 0, /* nScratch */ - (void*)0, /* pPage */ - 0, /* szPage */ - 0, /* nPage */ - 0, /* mxParserStack */ - 0, /* sharedCacheEnabled */ - /* All the rest should always be initialized to zero */ - 0, /* isInit */ - 0, /* inProgress */ - 0, /* isMutexInit */ - 0, /* isMallocInit */ - 0, /* isPCacheInit */ - 0, /* pInitMutex */ - 0, /* nRefInitMutex */ - 0, /* xLog */ - 0, /* pLogArg */ - 0, /* bLocaltimeFault */ -#ifdef SQLITE_ENABLE_SQLLOG - 0, /* xSqllog */ - 0 /* pSqllogArg */ +#ifndef SQLITE_FILE_HEADER /* 123456789 123456 */ +# define SQLITE_FILE_HEADER "SQLite format 3" #endif -}; - /* -** Hash table for global functions - functions common to all -** database connections. After initialization, this table is -** read-only. +** Page type flags. An ORed combination of these flags appear as the +** first byte of on-disk image of every BTree page. */ -SQLITE_PRIVATE SQLITE_WSD FuncDefHash sqlite3GlobalFunctions; +#define PTF_INTKEY 0x01 +#define PTF_ZERODATA 0x02 +#define PTF_LEAFDATA 0x04 +#define PTF_LEAF 0x08 /* -** Constant tokens for values 0 and 1. +** As each page of the file is loaded into memory, an instance of the following +** structure is appended and initialized to zero. This structure stores +** information about the page that is decoded from the raw file page. +** +** The pParent field points back to the parent page. This allows us to +** walk up the BTree from any leaf to the root. Care must be taken to +** unref() the parent page pointer when this page is no longer referenced. +** The pageDestructor() routine handles that chore. +** +** Access to all fields of this structure is controlled by the mutex +** stored in MemPage.pBt->mutex. */ -SQLITE_PRIVATE const Token sqlite3IntTokens[] = { - { "0", 1 }, - { "1", 1 } +struct MemPage { + u8 isInit; /* True if previously initialized. MUST BE FIRST! */ + u8 nOverflow; /* Number of overflow cell bodies in aCell[] */ + u8 intKey; /* True if intkey flag is set */ + u8 leaf; /* True if leaf flag is set */ + u8 hasData; /* True if this page stores data */ + u8 hdrOffset; /* 100 for page 1. 0 otherwise */ + u8 childPtrSize; /* 0 if leaf==1. 4 if leaf==0 */ + u8 max1bytePayload; /* min(maxLocal,127) */ + u16 maxLocal; /* Copy of BtShared.maxLocal or BtShared.maxLeaf */ + u16 minLocal; /* Copy of BtShared.minLocal or BtShared.minLeaf */ + u16 cellOffset; /* Index in aData of first cell pointer */ + u16 nFree; /* Number of free bytes on the page */ + u16 nCell; /* Number of cells on this page, local and ovfl */ + u16 maskPage; /* Mask for page offset */ + u16 aiOvfl[5]; /* Insert the i-th overflow cell before the aiOvfl-th + ** non-overflow cell */ + u8 *apOvfl[5]; /* Pointers to the body of overflow cells */ + BtShared *pBt; /* Pointer to BtShared that this page is part of */ + u8 *aData; /* Pointer to disk image of the page data */ + u8 *aDataEnd; /* One byte past the end of usable data */ + u8 *aCellIdx; /* The cell index area */ + DbPage *pDbPage; /* Pager page handle */ + Pgno pgno; /* Page number for this page */ }; - /* -** The value of the "pending" byte must be 0x40000000 (1 byte past the -** 1-gibabyte boundary) in a compatible database. SQLite never uses -** the database page that contains the pending byte. It never attempts -** to read or write that page. The pending byte page is set assign -** for use by the VFS layers as space for managing file locks. -** -** During testing, it is often desirable to move the pending byte to -** a different position in the file. This allows code that has to -** deal with the pending byte to run on files that are much smaller -** than 1 GiB. The sqlite3_test_control() interface can be used to -** move the pending byte. -** -** IMPORTANT: Changing the pending byte to any value other than -** 0x40000000 results in an incompatible database file format! -** Changing the pending byte during operating results in undefined -** and dileterious behavior. +** The in-memory image of a disk page has the auxiliary information appended +** to the end. EXTRA_SIZE is the number of bytes of space needed to hold +** that extra information. */ -#ifndef SQLITE_OMIT_WSD -SQLITE_PRIVATE int sqlite3PendingByte = 0x40000000; -#endif +#define EXTRA_SIZE sizeof(MemPage) /* -** Properties of opcodes. The OPFLG_INITIALIZER macro is -** created by mkopcodeh.awk during compilation. Data is obtained -** from the comments following the "case OP_xxxx:" statements in -** the vdbe.c file. +** A linked list of the following structures is stored at BtShared.pLock. +** Locks are added (or upgraded from READ_LOCK to WRITE_LOCK) when a cursor +** is opened on the table with root page BtShared.iTable. Locks are removed +** from this list when a transaction is committed or rolled back, or when +** a btree handle is closed. */ -SQLITE_PRIVATE const unsigned char sqlite3OpcodeProperty[] = OPFLG_INITIALIZER; +struct BtLock { + Btree *pBtree; /* Btree handle holding this lock */ + Pgno iTable; /* Root page of table */ + u8 eLock; /* READ_LOCK or WRITE_LOCK */ + BtLock *pNext; /* Next in BtShared.pLock list */ +}; -/************** End of global.c **********************************************/ -/************** Begin file ctime.c *******************************************/ -/* -** 2010 February 23 -** -** The author disclaims copyright to this source code. In place of -** a legal notice, here is a blessing: +/* Candidate values for BtLock.eLock */ +#define READ_LOCK 1 +#define WRITE_LOCK 2 + +/* A Btree handle ** -** May you do good and not evil. -** May you find forgiveness for yourself and forgive others. -** May you share freely, never taking more than you give. +** A database connection contains a pointer to an instance of +** this object for every database file that it has open. This structure +** is opaque to the database connection. The database connection cannot +** see the internals of this structure and only deals with pointers to +** this structure. ** -************************************************************************* +** For some database files, the same underlying database cache might be +** shared between multiple connections. In that case, each connection +** has it own instance of this object. But each instance of this object +** points to the same BtShared object. The database cache and the +** schema associated with the database file are all contained within +** the BtShared object. ** -** This file implements routines used to report what compile-time options -** SQLite was built with. +** All fields in this structure are accessed under sqlite3.mutex. +** The pBt pointer itself may not be changed while there exists cursors +** in the referenced BtShared that point back to this Btree since those +** cursors have to go through this Btree to find their BtShared and +** they often do so without holding sqlite3.mutex. */ +struct Btree { + sqlite3 *db; /* The database connection holding this btree */ + BtShared *pBt; /* Sharable content of this btree */ + u8 inTrans; /* TRANS_NONE, TRANS_READ or TRANS_WRITE */ + u8 sharable; /* True if we can share pBt with another db */ + u8 locked; /* True if db currently has pBt locked */ + int wantToLock; /* Number of nested calls to sqlite3BtreeEnter() */ + int nBackup; /* Number of backup operations reading this btree */ + Btree *pNext; /* List of other sharable Btrees from the same db */ + Btree *pPrev; /* Back pointer of the same list */ +#ifndef SQLITE_OMIT_SHARED_CACHE + BtLock lock; /* Object used to lock page 1 */ +#endif +}; + +/* +** Btree.inTrans may take one of the following values. +** +** If the shared-data extension is enabled, there may be multiple users +** of the Btree structure. At most one of these may open a write transaction, +** but any number may have active read transactions. +*/ +#define TRANS_NONE 0 +#define TRANS_READ 1 +#define TRANS_WRITE 2 + +/* +** An instance of this object represents a single database file. +** +** A single database file can be in use at the same time by two +** or more database connections. When two or more connections are +** sharing the same database file, each connection has it own +** private Btree object for the file and each of those Btrees points +** to this one BtShared object. BtShared.nRef is the number of +** connections currently sharing this database file. +** +** Fields in this structure are accessed under the BtShared.mutex +** mutex, except for nRef and pNext which are accessed under the +** global SQLITE_MUTEX_STATIC_MASTER mutex. The pPager field +** may not be modified once it is initially set as long as nRef>0. +** The pSchema field may be set once under BtShared.mutex and +** thereafter is unchanged as long as nRef>0. +** +** isPending: +** +** If a BtShared client fails to obtain a write-lock on a database +** table (because there exists one or more read-locks on the table), +** the shared-cache enters 'pending-lock' state and isPending is +** set to true. +** +** The shared-cache leaves the 'pending lock' state when either of +** the following occur: +** +** 1) The current writer (BtShared.pWriter) concludes its transaction, OR +** 2) The number of locks held by other connections drops to zero. +** +** while in the 'pending-lock' state, no connection may start a new +** transaction. +** +** This feature is included to help prevent writer-starvation. +*/ +struct BtShared { + Pager *pPager; /* The page cache */ + sqlite3 *db; /* Database connection currently using this Btree */ + BtCursor *pCursor; /* A list of all open cursors */ + MemPage *pPage1; /* First page of the database */ + u8 openFlags; /* Flags to sqlite3BtreeOpen() */ +#ifndef SQLITE_OMIT_AUTOVACUUM + u8 autoVacuum; /* True if auto-vacuum is enabled */ + u8 incrVacuum; /* True if incr-vacuum is enabled */ + u8 bDoTruncate; /* True to truncate db on commit */ +#endif + u8 inTransaction; /* Transaction state */ + u8 max1bytePayload; /* Maximum first byte of cell for a 1-byte payload */ + u16 btsFlags; /* Boolean parameters. See BTS_* macros below */ + u16 maxLocal; /* Maximum local payload in non-LEAFDATA tables */ + u16 minLocal; /* Minimum local payload in non-LEAFDATA tables */ + u16 maxLeaf; /* Maximum local payload in a LEAFDATA table */ + u16 minLeaf; /* Minimum local payload in a LEAFDATA table */ + u32 pageSize; /* Total number of bytes on a page */ + u32 usableSize; /* Number of usable bytes on each page */ + int nTransaction; /* Number of open transactions (read + write) */ + u32 nPage; /* Number of pages in the database */ + void *pSchema; /* Pointer to space allocated by sqlite3BtreeSchema() */ + void (*xFreeSchema)(void*); /* Destructor for BtShared.pSchema */ + sqlite3_mutex *mutex; /* Non-recursive mutex required to access this object */ + Bitvec *pHasContent; /* Set of pages moved to free-list this transaction */ +#ifndef SQLITE_OMIT_SHARED_CACHE + int nRef; /* Number of references to this structure */ + BtShared *pNext; /* Next on a list of sharable BtShared structs */ + BtLock *pLock; /* List of locks held on this shared-btree struct */ + Btree *pWriter; /* Btree with currently open write transaction */ +#endif + u8 *pTmpSpace; /* BtShared.pageSize bytes of space for tmp use */ +}; + +/* +** Allowed values for BtShared.btsFlags +*/ +#define BTS_READ_ONLY 0x0001 /* Underlying file is readonly */ +#define BTS_PAGESIZE_FIXED 0x0002 /* Page size can no longer be changed */ +#define BTS_SECURE_DELETE 0x0004 /* PRAGMA secure_delete is enabled */ +#define BTS_INITIALLY_EMPTY 0x0008 /* Database was empty at trans start */ +#define BTS_NO_WAL 0x0010 /* Do not open write-ahead-log files */ +#define BTS_EXCLUSIVE 0x0020 /* pWriter has an exclusive lock */ +#define BTS_PENDING 0x0040 /* Waiting for read-locks to clear */ + +/* +** An instance of the following structure is used to hold information +** about a cell. The parseCellPtr() function fills in this structure +** based on information extract from the raw disk page. +*/ +typedef struct CellInfo CellInfo; +struct CellInfo { + i64 nKey; /* The key for INTKEY tables, or number of bytes in key */ + u8 *pCell; /* Pointer to the start of cell content */ + u32 nData; /* Number of bytes of data */ + u32 nPayload; /* Total amount of payload */ + u16 nHeader; /* Size of the cell content header in bytes */ + u16 nLocal; /* Amount of payload held locally */ + u16 iOverflow; /* Offset to overflow page number. Zero if no overflow */ + u16 nSize; /* Size of the cell content on the main b-tree page */ +}; + +/* +** Maximum depth of an SQLite B-Tree structure. Any B-Tree deeper than +** this will be declared corrupt. This value is calculated based on a +** maximum database size of 2^31 pages a minimum fanout of 2 for a +** root-node and 3 for all other internal nodes. +** +** If a tree that appears to be taller than this is encountered, it is +** assumed that the database is corrupt. +*/ +#define BTCURSOR_MAX_DEPTH 20 + +/* +** A cursor is a pointer to a particular entry within a particular +** b-tree within a database file. +** +** The entry is identified by its MemPage and the index in +** MemPage.aCell[] of the entry. +** +** A single database file can be shared by two more database connections, +** but cursors cannot be shared. Each cursor is associated with a +** particular database connection identified BtCursor.pBtree.db. +** +** Fields in this structure are accessed under the BtShared.mutex +** found at self->pBt->mutex. +*/ +struct BtCursor { + Btree *pBtree; /* The Btree to which this cursor belongs */ + BtShared *pBt; /* The BtShared this cursor points to */ + BtCursor *pNext, *pPrev; /* Forms a linked list of all cursors */ + struct KeyInfo *pKeyInfo; /* Argument passed to comparison function */ +#ifndef SQLITE_OMIT_INCRBLOB + Pgno *aOverflow; /* Cache of overflow page locations */ +#endif + Pgno pgnoRoot; /* The root page of this tree */ + sqlite3_int64 cachedRowid; /* Next rowid cache. 0 means not valid */ + CellInfo info; /* A parse of the cell we are pointing at */ + i64 nKey; /* Size of pKey, or last integer key */ + void *pKey; /* Saved key that was cursor's last known position */ + int skipNext; /* Prev() is noop if negative. Next() is noop if positive */ + u8 wrFlag; /* True if writable */ + u8 atLast; /* Cursor pointing to the last entry */ + u8 validNKey; /* True if info.nKey is valid */ + u8 eState; /* One of the CURSOR_XXX constants (see below) */ +#ifndef SQLITE_OMIT_INCRBLOB + u8 isIncrblobHandle; /* True if this cursor is an incr. io handle */ +#endif + u8 hints; /* As configured by CursorSetHints() */ + i16 iPage; /* Index of current page in apPage */ + u16 aiIdx[BTCURSOR_MAX_DEPTH]; /* Current index in apPage[i] */ + MemPage *apPage[BTCURSOR_MAX_DEPTH]; /* Pages from root to current page */ +}; + +/* +** Potential values for BtCursor.eState. +** +** CURSOR_INVALID: +** Cursor does not point to a valid entry. This can happen (for example) +** because the table is empty or because BtreeCursorFirst() has not been +** called. +** +** CURSOR_VALID: +** Cursor points to a valid entry. getPayload() etc. may be called. +** +** CURSOR_SKIPNEXT: +** Cursor is valid except that the Cursor.skipNext field is non-zero +** indicating that the next sqlite3BtreeNext() or sqlite3BtreePrevious() +** operation should be a no-op. +** +** CURSOR_REQUIRESEEK: +** The table that this cursor was opened on still exists, but has been +** modified since the cursor was last used. The cursor position is saved +** in variables BtCursor.pKey and BtCursor.nKey. When a cursor is in +** this state, restoreCursorPosition() can be called to attempt to +** seek the cursor to the saved position. +** +** CURSOR_FAULT: +** A unrecoverable error (an I/O error or a malloc failure) has occurred +** on a different connection that shares the BtShared cache with this +** cursor. The error has left the cache in an inconsistent state. +** Do nothing else with this cursor. Any attempt to use the cursor +** should return the error code stored in BtCursor.skip +*/ +#define CURSOR_INVALID 0 +#define CURSOR_VALID 1 +#define CURSOR_SKIPNEXT 2 +#define CURSOR_REQUIRESEEK 3 +#define CURSOR_FAULT 4 + +/* +** The database page the PENDING_BYTE occupies. This page is never used. +*/ +# define PENDING_BYTE_PAGE(pBt) PAGER_MJ_PGNO(pBt) + +/* +** These macros define the location of the pointer-map entry for a +** database page. The first argument to each is the number of usable +** bytes on each page of the database (often 1024). The second is the +** page number to look up in the pointer map. +** +** PTRMAP_PAGENO returns the database page number of the pointer-map +** page that stores the required pointer. PTRMAP_PTROFFSET returns +** the offset of the requested map entry. +** +** If the pgno argument passed to PTRMAP_PAGENO is a pointer-map page, +** then pgno is returned. So (pgno==PTRMAP_PAGENO(pgsz, pgno)) can be +** used to test if pgno is a pointer-map page. PTRMAP_ISPAGE implements +** this test. +*/ +#define PTRMAP_PAGENO(pBt, pgno) ptrmapPageno(pBt, pgno) +#define PTRMAP_PTROFFSET(pgptrmap, pgno) (5*(pgno-pgptrmap-1)) +#define PTRMAP_ISPAGE(pBt, pgno) (PTRMAP_PAGENO((pBt),(pgno))==(pgno)) + +/* +** The pointer map is a lookup table that identifies the parent page for +** each child page in the database file. The parent page is the page that +** contains a pointer to the child. Every page in the database contains +** 0 or 1 parent pages. (In this context 'database page' refers +** to any page that is not part of the pointer map itself.) Each pointer map +** entry consists of a single byte 'type' and a 4 byte parent page number. +** The PTRMAP_XXX identifiers below are the valid types. +** +** The purpose of the pointer map is to facility moving pages from one +** position in the file to another as part of autovacuum. When a page +** is moved, the pointer in its parent must be updated to point to the +** new location. The pointer map is used to locate the parent page quickly. +** +** PTRMAP_ROOTPAGE: The database page is a root-page. The page-number is not +** used in this case. +** +** PTRMAP_FREEPAGE: The database page is an unused (free) page. The page-number +** is not used in this case. +** +** PTRMAP_OVERFLOW1: The database page is the first page in a list of +** overflow pages. The page number identifies the page that +** contains the cell with a pointer to this overflow page. +** +** PTRMAP_OVERFLOW2: The database page is the second or later page in a list of +** overflow pages. The page-number identifies the previous +** page in the overflow page list. +** +** PTRMAP_BTREE: The database page is a non-root btree page. The page number +** identifies the parent page in the btree. +*/ +#define PTRMAP_ROOTPAGE 1 +#define PTRMAP_FREEPAGE 2 +#define PTRMAP_OVERFLOW1 3 +#define PTRMAP_OVERFLOW2 4 +#define PTRMAP_BTREE 5 + +/* A bunch of assert() statements to check the transaction state variables +** of handle p (type Btree*) are internally consistent. +*/ +#define btreeIntegrity(p) \ + assert( p->pBt->inTransaction!=TRANS_NONE || p->pBt->nTransaction==0 ); \ + assert( p->pBt->inTransaction>=p->inTrans ); + + +/* +** The ISAUTOVACUUM macro is used within balance_nonroot() to determine +** if the database supports auto-vacuum or not. Because it is used +** within an expression that is an argument to another macro +** (sqliteMallocRaw), it is not possible to use conditional compilation. +** So, this macro is defined instead. +*/ +#ifndef SQLITE_OMIT_AUTOVACUUM +#define ISAUTOVACUUM (pBt->autoVacuum) +#else +#define ISAUTOVACUUM 0 +#endif + + +/* +** This structure is passed around through all the sanity checking routines +** in order to keep track of some global state information. +** +** The aRef[] array is allocated so that there is 1 bit for each page in +** the database. As the integrity-check proceeds, for each page used in +** the database the corresponding bit is set. This allows integrity-check to +** detect pages that are used twice and orphaned pages (both of which +** indicate corruption). +*/ +typedef struct IntegrityCk IntegrityCk; +struct IntegrityCk { + BtShared *pBt; /* The tree being checked out */ + Pager *pPager; /* The associated pager. Also accessible by pBt->pPager */ + u8 *aPgRef; /* 1 bit per page in the db (see above) */ + Pgno nPage; /* Number of pages in the database */ + int mxErr; /* Stop accumulating errors when this reaches zero */ + int nErr; /* Number of messages written to zErrMsg so far */ + int mallocFailed; /* A memory allocation error has occurred */ + StrAccum errMsg; /* Accumulate the error message text here */ +}; + +/* +** Routines to read or write a two- and four-byte big-endian integer values. +*/ +#define get2byte(x) ((x)[0]<<8 | (x)[1]) +#define put2byte(p,v) ((p)[0] = (u8)((v)>>8), (p)[1] = (u8)(v)) +#define get4byte sqlite3Get4byte +#define put4byte sqlite3Put4byte + +/************** End of btreeInt.h ********************************************/ +/************** Continuing where we left off in crypto.c *********************/ +/************** Include crypto.h in the middle of crypto.c *******************/ +/************** Begin file crypto.h ******************************************/ +/* +** SQLCipher +** crypto.h developed by Stephen Lombardo (Zetetic LLC) +** sjlombardo at zetetic dot net +** http://zetetic.net +** +** Copyright (c) 2008, ZETETIC LLC +** 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 the ZETETIC LLC 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 ZETETIC LLC ''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 ZETETIC LLC 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. +** +*/ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +#ifndef CRYPTO_H +#define CRYPTO_H + +#if !defined (SQLCIPHER_CRYPTO_CC) \ + && !defined (SQLCIPHER_CRYPTO_LIBTOMCRYPT) \ + && !defined (SQLCIPHER_CRYPTO_OPENSSL) +#define SQLCIPHER_CRYPTO_OPENSSL +#endif + +#define FILE_HEADER_SZ 16 + +#ifndef CIPHER_VERSION +#define CIPHER_VERSION "3.0.0" +#endif + +#ifndef CIPHER +#define CIPHER "aes-256-cbc" +#endif + +#define CIPHER_DECRYPT 0 +#define CIPHER_ENCRYPT 1 + +#define CIPHER_READ_CTX 0 +#define CIPHER_WRITE_CTX 1 +#define CIPHER_READWRITE_CTX 2 + +#ifndef PBKDF2_ITER +#define PBKDF2_ITER 64000 +#endif + +/* possible flags for cipher_ctx->flags */ +#define CIPHER_FLAG_HMAC 0x01 +#define CIPHER_FLAG_LE_PGNO 0x02 +#define CIPHER_FLAG_BE_PGNO 0x04 + +#ifndef DEFAULT_CIPHER_FLAGS +#define DEFAULT_CIPHER_FLAGS CIPHER_FLAG_HMAC | CIPHER_FLAG_LE_PGNO +#endif + + +/* by default, sqlcipher will use a reduced number of iterations to generate + the HMAC key / or transform a raw cipher key + */ +#ifndef FAST_PBKDF2_ITER +#define FAST_PBKDF2_ITER 2 +#endif + +/* this if a fixed random array that will be xor'd with the database salt to ensure that the + salt passed to the HMAC key derivation function is not the same as that used to derive + the encryption key. This can be overridden at compile time but it will make the resulting + binary incompatible with the default builds when using HMAC. A future version of SQLcipher + will likely allow this to be defined at runtime via pragma */ +#ifndef HMAC_SALT_MASK +#define HMAC_SALT_MASK 0x3a +#endif + +#ifndef CIPHER_MAX_IV_SZ +#define CIPHER_MAX_IV_SZ 16 +#endif + +#ifndef CIPHER_MAX_KEY_SZ +#define CIPHER_MAX_KEY_SZ 64 +#endif + + +#ifdef CODEC_DEBUG +#define CODEC_TRACE(X) {printf X;fflush(stdout);} +#else +#define CODEC_TRACE(X) +#endif + +#ifdef CODEC_DEBUG_PAGEDATA +#define CODEC_HEXDUMP(DESC,BUFFER,LEN) \ + { \ + int __pctr; \ + printf(DESC); \ + for(__pctr=0; __pctr < LEN; __pctr++) { \ + if(__pctr % 16 == 0) printf("\n%05x: ",__pctr); \ + printf("%02x ",((unsigned char*) BUFFER)[__pctr]); \ + } \ + printf("\n"); \ + fflush(stdout); \ + } +#else +#define CODEC_HEXDUMP(DESC,BUFFER,LEN) +#endif + +/* extensions defined in pager.c */ +SQLITE_PRIVATE void sqlite3pager_get_codec(Pager *pPager, void **ctx); +SQLITE_PRIVATE int sqlite3pager_is_mj_pgno(Pager *pPager, Pgno pgno); +SQLITE_PRIVATE sqlite3_file *sqlite3Pager_get_fd(Pager *pPager); +SQLITE_PRIVATE void sqlite3pager_sqlite3PagerSetCodec( + Pager *pPager, + void *(*xCodec)(void*,void*,Pgno,int), + void (*xCodecSizeChng)(void*,int,int), + void (*xCodecFree)(void*), + void *pCodec +); +SQLITE_PRIVATE void sqlite3pager_sqlite3PagerSetError(Pager *pPager, int error); +/* end extensions defined in pager.c */ + +/* +** Simple shared routines for converting hex char strings to binary data + */ +static int cipher_hex2int(char c) { + return (c>='0' && c<='9') ? (c)-'0' : + (c>='A' && c<='F') ? (c)-'A'+10 : + (c>='a' && c<='f') ? (c)-'a'+10 : 0; +} + +static void cipher_hex2bin(const unsigned char *hex, int sz, unsigned char *out){ + int i; + for(i = 0; i < sz; i += 2){ + out[i/2] = (cipher_hex2int(hex[i])<<4) | cipher_hex2int(hex[i+1]); + } +} + +static void cipher_bin2hex(const unsigned char* in, int sz, char *out) { + int i; + for(i=0; i < sz; i++) { + sqlite3_snprintf(3, out + (i*2), "%02x ", in[i]); + } +} + +/* extensions defined in crypto_impl.c */ +typedef struct codec_ctx codec_ctx; + +/* activation and initialization */ +void sqlcipher_activate(); +void sqlcipher_deactivate(); +int sqlcipher_codec_ctx_init(codec_ctx **, Db *, Pager *, sqlite3_file *, const void *, int); +void sqlcipher_codec_ctx_free(codec_ctx **); +int sqlcipher_codec_key_derive(codec_ctx *); +int sqlcipher_codec_key_copy(codec_ctx *, int); + +/* page cipher implementation */ +int sqlcipher_page_cipher(codec_ctx *, int, Pgno, int, int, unsigned char *, unsigned char *); + +/* context setters & getters */ +void sqlcipher_codec_ctx_set_error(codec_ctx *, int); + +int sqlcipher_codec_ctx_set_pass(codec_ctx *, const void *, int, int); +void sqlcipher_codec_get_keyspec(codec_ctx *, void **zKey, int *nKey); + +int sqlcipher_codec_ctx_set_pagesize(codec_ctx *, int); +int sqlcipher_codec_ctx_get_pagesize(codec_ctx *); +int sqlcipher_codec_ctx_get_reservesize(codec_ctx *); + +void sqlcipher_set_default_kdf_iter(int iter); +int sqlcipher_get_default_kdf_iter(); + +int sqlcipher_codec_ctx_set_kdf_iter(codec_ctx *, int, int); +int sqlcipher_codec_ctx_get_kdf_iter(codec_ctx *ctx, int); + +void* sqlcipher_codec_ctx_get_kdf_salt(codec_ctx *ctx); + +int sqlcipher_codec_ctx_set_fast_kdf_iter(codec_ctx *, int, int); +int sqlcipher_codec_ctx_get_fast_kdf_iter(codec_ctx *, int); + +int sqlcipher_codec_ctx_set_cipher(codec_ctx *, const char *, int); +const char* sqlcipher_codec_ctx_get_cipher(codec_ctx *ctx, int for_ctx); + +void* sqlcipher_codec_ctx_get_data(codec_ctx *); + +void sqlcipher_exportFunc(sqlite3_context *, int, sqlite3_value **); + +void sqlcipher_set_default_use_hmac(int use); +int sqlcipher_get_default_use_hmac(); + +void sqlcipher_set_hmac_salt_mask(unsigned char mask); +unsigned char sqlcipher_get_hmac_salt_mask(); + +int sqlcipher_codec_ctx_set_use_hmac(codec_ctx *ctx, int use); +int sqlcipher_codec_ctx_get_use_hmac(codec_ctx *ctx, int for_ctx); + +int sqlcipher_codec_ctx_set_flag(codec_ctx *ctx, unsigned int flag); +int sqlcipher_codec_ctx_unset_flag(codec_ctx *ctx, unsigned int flag); +int sqlcipher_codec_ctx_get_flag(codec_ctx *ctx, unsigned int flag, int for_ctx); + +const char* sqlcipher_codec_get_cipher_provider(codec_ctx *ctx); +int sqlcipher_codec_ctx_migrate(codec_ctx *ctx); +#endif +#endif +/* END SQLCIPHER */ + +/************** End of crypto.h **********************************************/ +/************** Continuing where we left off in crypto.c *********************/ + +static const char* codec_get_cipher_version() { + return CIPHER_VERSION; +} + +/* Generate code to return a string value */ +static void codec_vdbe_return_static_string(Parse *pParse, const char *zLabel, const char *value){ + Vdbe *v = sqlite3GetVdbe(pParse); + sqlite3VdbeSetNumCols(v, 1); + sqlite3VdbeSetColName(v, 0, COLNAME_NAME, zLabel, SQLITE_STATIC); + sqlite3VdbeAddOp4(v, OP_String8, 0, 1, 0, value, 0); + sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 1); +} + +static int codec_set_btree_to_codec_pagesize(sqlite3 *db, Db *pDb, codec_ctx *ctx) { + int rc, page_sz, reserve_sz; + + page_sz = sqlcipher_codec_ctx_get_pagesize(ctx); + reserve_sz = sqlcipher_codec_ctx_get_reservesize(ctx); + + sqlite3_mutex_enter(db->mutex); + db->nextPagesize = page_sz; + + /* before forcing the page size we need to unset the BTS_PAGESIZE_FIXED flag, else + sqliteBtreeSetPageSize will block the change */ + pDb->pBt->pBt->btsFlags &= ~BTS_PAGESIZE_FIXED; + CODEC_TRACE(("codec_set_btree_to_codec_pagesize: sqlite3BtreeSetPageSize() size=%d reserve=%d\n", page_sz, reserve_sz)); + rc = sqlite3BtreeSetPageSize(pDb->pBt, page_sz, reserve_sz, 0); + sqlite3_mutex_leave(db->mutex); + return rc; +} + +static int codec_set_pass_key(sqlite3* db, int nDb, const void *zKey, int nKey, int for_ctx) { + struct Db *pDb = &db->aDb[nDb]; + CODEC_TRACE(("codec_set_pass_key: entered db=%p nDb=%d zKey=%s nKey=%d for_ctx=%d\n", db, nDb, (char *)zKey, nKey, for_ctx)); + if(pDb->pBt) { + codec_ctx *ctx; + sqlite3pager_get_codec(pDb->pBt->pBt->pPager, (void **) &ctx); + if(ctx) return sqlcipher_codec_ctx_set_pass(ctx, zKey, nKey, for_ctx); + } + return SQLITE_ERROR; +} + +int sqlcipher_codec_pragma(sqlite3* db, int iDb, Parse *pParse, const char *zLeft, const char *zRight) { + struct Db *pDb = &db->aDb[iDb]; + codec_ctx *ctx = NULL; + int rc; + + if(pDb->pBt) { + sqlite3pager_get_codec(pDb->pBt->pBt->pPager, (void **) &ctx); + } + + CODEC_TRACE(("sqlcipher_codec_pragma: entered db=%p iDb=%d pParse=%p zLeft=%s zRight=%s ctx=%p\n", db, iDb, pParse, zLeft, zRight, ctx)); + + if( sqlite3StrICmp(zLeft, "cipher_migrate")==0 && !zRight ){ + if(ctx){ + char *migrate_status = sqlite3_mprintf("%d", sqlcipher_codec_ctx_migrate(ctx)); + codec_vdbe_return_static_string(pParse, "cipher_migrate", migrate_status); + sqlite3_free(migrate_status); + } + } else + if( sqlite3StrICmp(zLeft, "cipher_provider")==0 && !zRight ){ + if(ctx) { codec_vdbe_return_static_string(pParse, "cipher_provider", + sqlcipher_codec_get_cipher_provider(ctx)); + } + } else + if( sqlite3StrICmp(zLeft, "cipher_version")==0 && !zRight ){ + codec_vdbe_return_static_string(pParse, "cipher_version", codec_get_cipher_version()); + }else + if( sqlite3StrICmp(zLeft, "cipher")==0 ){ + if(ctx) { + if( zRight ) { + sqlcipher_codec_ctx_set_cipher(ctx, zRight, 2); // change cipher for both + }else { + codec_vdbe_return_static_string(pParse, "cipher", + sqlcipher_codec_ctx_get_cipher(ctx, 2)); + } + } + }else + if( sqlite3StrICmp(zLeft, "rekey_cipher")==0 && zRight ){ + if(ctx) sqlcipher_codec_ctx_set_cipher(ctx, zRight, 1); // change write cipher only + }else + if( sqlite3StrICmp(zLeft,"cipher_default_kdf_iter")==0 ){ + if( zRight ) { + sqlcipher_set_default_kdf_iter(atoi(zRight)); // change default KDF iterations + } else { + char *kdf_iter = sqlite3_mprintf("%d", sqlcipher_get_default_kdf_iter()); + codec_vdbe_return_static_string(pParse, "cipher_default_kdf_iter", kdf_iter); + sqlite3_free(kdf_iter); + } + }else + if( sqlite3StrICmp(zLeft, "kdf_iter")==0 ){ + if(ctx) { + if( zRight ) { + sqlcipher_codec_ctx_set_kdf_iter(ctx, atoi(zRight), 2); // change of RW PBKDF2 iteration + } else { + char *kdf_iter = sqlite3_mprintf("%d", sqlcipher_codec_ctx_get_kdf_iter(ctx, 2)); + codec_vdbe_return_static_string(pParse, "kdf_iter", kdf_iter); + sqlite3_free(kdf_iter); + } + } + }else + if( sqlite3StrICmp(zLeft, "fast_kdf_iter")==0){ + if(ctx) { + if( zRight ) { + sqlcipher_codec_ctx_set_fast_kdf_iter(ctx, atoi(zRight), 2); // change of RW PBKDF2 iteration + } else { + char *fast_kdf_iter = sqlite3_mprintf("%d", sqlcipher_codec_ctx_get_fast_kdf_iter(ctx, 2)); + codec_vdbe_return_static_string(pParse, "fast_kdf_iter", fast_kdf_iter); + sqlite3_free(fast_kdf_iter); + } + } + }else + if( sqlite3StrICmp(zLeft, "rekey_kdf_iter")==0 && zRight ){ + if(ctx) sqlcipher_codec_ctx_set_kdf_iter(ctx, atoi(zRight), 1); // write iterations only + }else + if( sqlite3StrICmp(zLeft,"cipher_page_size")==0 ){ + if(ctx) { + if( zRight ) { + int size = atoi(zRight); + rc = sqlcipher_codec_ctx_set_pagesize(ctx, size); + if(rc != SQLITE_OK) sqlcipher_codec_ctx_set_error(ctx, rc); + rc = codec_set_btree_to_codec_pagesize(db, pDb, ctx); + if(rc != SQLITE_OK) sqlcipher_codec_ctx_set_error(ctx, rc); + } else { + char * page_size = sqlite3_mprintf("%d", sqlcipher_codec_ctx_get_pagesize(ctx)); + codec_vdbe_return_static_string(pParse, "cipher_page_size", page_size); + sqlite3_free(page_size); + } + } + }else + if( sqlite3StrICmp(zLeft,"cipher_default_use_hmac")==0 ){ + if( zRight ) { + sqlcipher_set_default_use_hmac(sqlite3GetBoolean(zRight,1)); + } else { + char *default_use_hmac = sqlite3_mprintf("%d", sqlcipher_get_default_use_hmac()); + codec_vdbe_return_static_string(pParse, "cipher_default_use_hmac", default_use_hmac); + sqlite3_free(default_use_hmac); + } + }else + if( sqlite3StrICmp(zLeft,"cipher_use_hmac")==0 ){ + if(ctx) { + if( zRight ) { + rc = sqlcipher_codec_ctx_set_use_hmac(ctx, sqlite3GetBoolean(zRight,1)); + if(rc != SQLITE_OK) sqlcipher_codec_ctx_set_error(ctx, rc); + /* since the use of hmac has changed, the page size may also change */ + rc = codec_set_btree_to_codec_pagesize(db, pDb, ctx); + if(rc != SQLITE_OK) sqlcipher_codec_ctx_set_error(ctx, rc); + } else { + char *hmac_flag = sqlite3_mprintf("%d", sqlcipher_codec_ctx_get_use_hmac(ctx, 2)); + codec_vdbe_return_static_string(pParse, "cipher_use_hmac", hmac_flag); + sqlite3_free(hmac_flag); + } + } + }else + if( sqlite3StrICmp(zLeft,"cipher_hmac_pgno")==0 ){ + if(ctx) { + if(zRight) { + // clear both pgno endian flags + if(sqlite3StrICmp(zRight, "le") == 0) { + sqlcipher_codec_ctx_unset_flag(ctx, CIPHER_FLAG_BE_PGNO); + sqlcipher_codec_ctx_set_flag(ctx, CIPHER_FLAG_LE_PGNO); + } else if(sqlite3StrICmp(zRight, "be") == 0) { + sqlcipher_codec_ctx_unset_flag(ctx, CIPHER_FLAG_LE_PGNO); + sqlcipher_codec_ctx_set_flag(ctx, CIPHER_FLAG_BE_PGNO); + } else if(sqlite3StrICmp(zRight, "native") == 0) { + sqlcipher_codec_ctx_unset_flag(ctx, CIPHER_FLAG_LE_PGNO); + sqlcipher_codec_ctx_unset_flag(ctx, CIPHER_FLAG_BE_PGNO); + } + } else { + if(sqlcipher_codec_ctx_get_flag(ctx, CIPHER_FLAG_LE_PGNO, 2)) { + codec_vdbe_return_static_string(pParse, "cipher_hmac_pgno", "le"); + } else if(sqlcipher_codec_ctx_get_flag(ctx, CIPHER_FLAG_BE_PGNO, 2)) { + codec_vdbe_return_static_string(pParse, "cipher_hmac_pgno", "be"); + } else { + codec_vdbe_return_static_string(pParse, "cipher_hmac_pgno", "native"); + } + } + } + }else + if( sqlite3StrICmp(zLeft,"cipher_hmac_salt_mask")==0 ){ + if(ctx) { + if(zRight) { + if (sqlite3StrNICmp(zRight ,"x'", 2) == 0 && sqlite3Strlen30(zRight) == 5) { + unsigned char mask = 0; + const unsigned char *hex = (const unsigned char *)zRight+2; + cipher_hex2bin(hex,2,&mask); + sqlcipher_set_hmac_salt_mask(mask); + } + } else { + char *hmac_salt_mask = sqlite3_mprintf("%02x", sqlcipher_get_hmac_salt_mask()); + codec_vdbe_return_static_string(pParse, "cipher_hmac_salt_mask", hmac_salt_mask); + sqlite3_free(hmac_salt_mask); + } + } + }else { + return 0; + } + return 1; +} + +/* + * sqlite3Codec can be called in multiple modes. + * encrypt mode - expected to return a pointer to the + * encrypted data without altering pData. + * decrypt mode - expected to return a pointer to pData, with + * the data decrypted in the input buffer + */ +void* sqlite3Codec(void *iCtx, void *data, Pgno pgno, int mode) { + codec_ctx *ctx = (codec_ctx *) iCtx; + int offset = 0, rc = 0; + int page_sz = sqlcipher_codec_ctx_get_pagesize(ctx); + unsigned char *pData = (unsigned char *) data; + void *buffer = sqlcipher_codec_ctx_get_data(ctx); + void *kdf_salt = sqlcipher_codec_ctx_get_kdf_salt(ctx); + CODEC_TRACE(("sqlite3Codec: entered pgno=%d, mode=%d, page_sz=%d\n", pgno, mode, page_sz)); + + /* call to derive keys if not present yet */ + if((rc = sqlcipher_codec_key_derive(ctx)) != SQLITE_OK) { + sqlcipher_codec_ctx_set_error(ctx, rc); + return NULL; + } + + if(pgno == 1) offset = FILE_HEADER_SZ; /* adjust starting pointers in data page for header offset on first page*/ + + CODEC_TRACE(("sqlite3Codec: switch mode=%d offset=%d\n", mode, offset)); + switch(mode) { + case 0: /* decrypt */ + case 2: + case 3: + if(pgno == 1) memcpy(buffer, SQLITE_FILE_HEADER, FILE_HEADER_SZ); /* copy file header to the first 16 bytes of the page */ + rc = sqlcipher_page_cipher(ctx, CIPHER_READ_CTX, pgno, CIPHER_DECRYPT, page_sz - offset, pData + offset, (unsigned char*)buffer + offset); + if(rc != SQLITE_OK) sqlcipher_codec_ctx_set_error(ctx, rc); + memcpy(pData, buffer, page_sz); /* copy buffer data back to pData and return */ + return pData; + break; + case 6: /* encrypt */ + if(pgno == 1) memcpy(buffer, kdf_salt, FILE_HEADER_SZ); /* copy salt to output buffer */ + rc = sqlcipher_page_cipher(ctx, CIPHER_WRITE_CTX, pgno, CIPHER_ENCRYPT, page_sz - offset, pData + offset, (unsigned char*)buffer + offset); + if(rc != SQLITE_OK) sqlcipher_codec_ctx_set_error(ctx, rc); + return buffer; /* return persistent buffer data, pData remains intact */ + break; + case 7: + if(pgno == 1) memcpy(buffer, kdf_salt, FILE_HEADER_SZ); /* copy salt to output buffer */ + rc = sqlcipher_page_cipher(ctx, CIPHER_READ_CTX, pgno, CIPHER_ENCRYPT, page_sz - offset, pData + offset, (unsigned char*)buffer + offset); + if(rc != SQLITE_OK) sqlcipher_codec_ctx_set_error(ctx, rc); + return buffer; /* return persistent buffer data, pData remains intact */ + break; + default: + return pData; + break; + } +} + +SQLITE_PRIVATE void sqlite3FreeCodecArg(void *pCodecArg) { + codec_ctx *ctx = (codec_ctx *) pCodecArg; + if(pCodecArg == NULL) return; + sqlcipher_codec_ctx_free(&ctx); // wipe and free allocated memory for the context + sqlcipher_deactivate(); /* cleanup related structures, OpenSSL etc, when codec is detatched */ +} + +SQLITE_PRIVATE int sqlite3CodecAttach(sqlite3* db, int nDb, const void *zKey, int nKey) { + struct Db *pDb = &db->aDb[nDb]; + + CODEC_TRACE(("sqlite3CodecAttach: entered nDb=%d zKey=%s, nKey=%d\n", nDb, (char *)zKey, nKey)); + + + if(nKey && zKey && pDb->pBt) { + int rc; + Pager *pPager = pDb->pBt->pBt->pPager; + sqlite3_file *fd = sqlite3Pager_get_fd(pPager); + codec_ctx *ctx; + + sqlcipher_activate(); /* perform internal initialization for sqlcipher */ + + sqlite3_mutex_enter(db->mutex); + + /* point the internal codec argument against the contet to be prepared */ + rc = sqlcipher_codec_ctx_init(&ctx, pDb, pDb->pBt->pBt->pPager, fd, zKey, nKey); + + if(rc != SQLITE_OK) return rc; /* initialization failed, do not attach potentially corrupted context */ + + sqlite3pager_sqlite3PagerSetCodec(sqlite3BtreePager(pDb->pBt), sqlite3Codec, NULL, sqlite3FreeCodecArg, (void *) ctx); + + codec_set_btree_to_codec_pagesize(db, pDb, ctx); + + /* force secure delete. This has the benefit of wiping internal data when deleted + and also ensures that all pages are written to disk (i.e. not skipped by + sqlite3PagerDontWrite optimizations) */ + sqlite3BtreeSecureDelete(pDb->pBt, 1); + + /* if fd is null, then this is an in-memory database and + we dont' want to overwrite the AutoVacuum settings + if not null, then set to the default */ + if(fd != NULL) { + sqlite3BtreeSetAutoVacuum(pDb->pBt, SQLITE_DEFAULT_AUTOVACUUM); + } + sqlite3_mutex_leave(db->mutex); + } + return SQLITE_OK; +} + +SQLITE_API void sqlite3_activate_see(const char* in) { + /* do nothing, security enhancements are always active */ +} + +static int sqlcipher_find_db_index(sqlite3 *db, const char *zDb) { + int db_index; + if(zDb == NULL){ + return 0; + } + for(db_index = 0; db_index < db->nDb; db_index++) { + struct Db *pDb = &db->aDb[db_index]; + if(strcmp(pDb->zName, zDb) == 0) { + return db_index; + } + } + return 0; +} + +SQLITE_API int sqlite3_key(sqlite3 *db, const void *pKey, int nKey) { + CODEC_TRACE(("sqlite3_key entered: db=%p pKey=%s nKey=%d\n", db, (char *)pKey, nKey)); + return sqlite3_key_v2(db, "main", pKey, nKey); +} + +SQLITE_API int sqlite3_key_v2(sqlite3 *db, const char *zDb, const void *pKey, int nKey) { + CODEC_TRACE(("sqlite3_key_v2: entered db=%p zDb=%s pKey=%s nKey=%d\n", db, zDb, (char *)pKey, nKey)); + /* attach key if db and pKey are not null and nKey is > 0 */ + if(db && pKey && nKey) { + int db_index = sqlcipher_find_db_index(db, zDb); + return sqlite3CodecAttach(db, db_index, pKey, nKey); + } + return SQLITE_ERROR; +} + +SQLITE_API int sqlite3_rekey(sqlite3 *db, const void *pKey, int nKey) { + CODEC_TRACE(("sqlite3_rekey entered: db=%p pKey=%s nKey=%d\n", db, (char *)pKey, nKey)); + return sqlite3_rekey_v2(db, "main", pKey, nKey); +} + +/* sqlite3_rekey_v2 +** Given a database, this will reencrypt the database using a new key. +** There is only one possible modes of operation - to encrypt a database +** that is already encrpyted. If the database is not already encrypted +** this should do nothing +** The proposed logic for this function follows: +** 1. Determine if the database is already encryptped +** 2. If there is NOT already a key present do nothing +** 3. If there is a key present, re-encrypt the database with the new key +*/ +SQLITE_API int sqlite3_rekey_v2(sqlite3 *db, const char *zDb, const void *pKey, int nKey) { + CODEC_TRACE(("sqlite3_rekey_v2: entered db=%p zDb=%s pKey=%s, nKey=%d\n", db, zDb, (char *)pKey, nKey)); + if(db && pKey && nKey) { + int db_index = sqlcipher_find_db_index(db, zDb); + struct Db *pDb = &db->aDb[db_index]; + CODEC_TRACE(("sqlite3_rekey_v2: database pDb=%p db_index:%d\n", pDb, db_index)); + if(pDb->pBt) { + codec_ctx *ctx; + int rc, page_count; + Pgno pgno; + PgHdr *page; + Pager *pPager = pDb->pBt->pBt->pPager; + + sqlite3pager_get_codec(pDb->pBt->pBt->pPager, (void **) &ctx); + + if(ctx == NULL) { + /* there was no codec attached to this database, so this should do nothing! */ + CODEC_TRACE(("sqlite3_rekey_v2: no codec attached to db, exiting\n")); + return SQLITE_OK; + } + + sqlite3_mutex_enter(db->mutex); + + codec_set_pass_key(db, db_index, pKey, nKey, CIPHER_WRITE_CTX); + + /* do stuff here to rewrite the database + ** 1. Create a transaction on the database + ** 2. Iterate through each page, reading it and then writing it. + ** 3. If that goes ok then commit and put ctx->rekey into ctx->key + ** note: don't deallocate rekey since it may be used in a subsequent iteration + */ + rc = sqlite3BtreeBeginTrans(pDb->pBt, 1); /* begin write transaction */ + sqlite3PagerPagecount(pPager, &page_count); + for(pgno = 1; rc == SQLITE_OK && pgno <= (unsigned int)page_count; pgno++) { /* pgno's start at 1 see pager.c:pagerAcquire */ + if(!sqlite3pager_is_mj_pgno(pPager, pgno)) { /* skip this page (see pager.c:pagerAcquire for reasoning) */ + rc = sqlite3PagerGet(pPager, pgno, &page); + if(rc == SQLITE_OK) { /* write page see pager_incr_changecounter for example */ + rc = sqlite3PagerWrite(page); + if(rc == SQLITE_OK) { + sqlite3PagerUnref(page); + } else { + CODEC_TRACE(("sqlite3_rekey_v2: error %d occurred writing page %d\n", rc, pgno)); + } + } else { + CODEC_TRACE(("sqlite3_rekey_v2: error %d occurred getting page %d\n", rc, pgno)); + } + } + } + + /* if commit was successful commit and copy the rekey data to current key, else rollback to release locks */ + if(rc == SQLITE_OK) { + CODEC_TRACE(("sqlite3_rekey_v2: committing\n")); + rc = sqlite3BtreeCommit(pDb->pBt); + sqlcipher_codec_key_copy(ctx, CIPHER_WRITE_CTX); + } else { + CODEC_TRACE(("sqlite3_rekey_v2: rollback\n")); + sqlite3BtreeRollback(pDb->pBt, SQLITE_ABORT_ROLLBACK); + } + + sqlite3_mutex_leave(db->mutex); + } + return SQLITE_OK; + } + return SQLITE_ERROR; +} + +SQLITE_PRIVATE void sqlite3CodecGetKey(sqlite3* db, int nDb, void **zKey, int *nKey) { + struct Db *pDb = &db->aDb[nDb]; + CODEC_TRACE(("sqlite3CodecGetKey: entered db=%p, nDb=%d\n", db, nDb)); + if( pDb->pBt ) { + codec_ctx *ctx; + sqlite3pager_get_codec(pDb->pBt->pBt->pPager, (void **) &ctx); + if(ctx) { /* if the codec has an attached codec_context user the raw key data */ + sqlcipher_codec_get_keyspec(ctx, zKey, nKey); + } else { + *zKey = NULL; + *nKey = 0; + } + } +} + +#ifndef OMIT_EXPORT + +/* + * Implementation of an "export" function that allows a caller + * to duplicate the main database to an attached database. This is intended + * as a conveneince for users who need to: + * + * 1. migrate from an non-encrypted database to an encrypted database + * 2. move from an encrypted database to a non-encrypted database + * 3. convert beween the various flavors of encrypted databases. + * + * This implementation is based heavily on the procedure and code used + * in vacuum.c, but is exposed as a function that allows export to any + * named attached database. + */ + +/* +** Finalize a prepared statement. If there was an error, store the +** text of the error message in *pzErrMsg. Return the result code. +** +** Based on vacuumFinalize from vacuum.c +*/ +static int sqlcipher_finalize(sqlite3 *db, sqlite3_stmt *pStmt, char **pzErrMsg){ + int rc; + rc = sqlite3VdbeFinalize((Vdbe*)pStmt); + if( rc ){ + sqlite3SetString(pzErrMsg, db, sqlite3_errmsg(db)); + } + return rc; +} + +/* +** Execute zSql on database db. Return an error code. +** +** Based on execSql from vacuum.c +*/ +static int sqlcipher_execSql(sqlite3 *db, char **pzErrMsg, const char *zSql){ + sqlite3_stmt *pStmt; + VVA_ONLY( int rc; ) + if( !zSql ){ + return SQLITE_NOMEM; + } + if( SQLITE_OK!=sqlite3_prepare(db, zSql, -1, &pStmt, 0) ){ + sqlite3SetString(pzErrMsg, db, sqlite3_errmsg(db)); + return sqlite3_errcode(db); + } + VVA_ONLY( rc = ) sqlite3_step(pStmt); + assert( rc!=SQLITE_ROW ); + return sqlcipher_finalize(db, pStmt, pzErrMsg); +} + +/* +** Execute zSql on database db. The statement returns exactly +** one column. Execute this as SQL on the same database. +** +** Based on execExecSql from vacuum.c +*/ +static int sqlcipher_execExecSql(sqlite3 *db, char **pzErrMsg, const char *zSql){ + sqlite3_stmt *pStmt; + int rc; + + rc = sqlite3_prepare(db, zSql, -1, &pStmt, 0); + if( rc!=SQLITE_OK ) return rc; + + while( SQLITE_ROW==sqlite3_step(pStmt) ){ + rc = sqlcipher_execSql(db, pzErrMsg, (char*)sqlite3_column_text(pStmt, 0)); + if( rc!=SQLITE_OK ){ + sqlcipher_finalize(db, pStmt, pzErrMsg); + return rc; + } + } + + return sqlcipher_finalize(db, pStmt, pzErrMsg); +} + +/* + * copy database and schema from the main database to an attached database + * + * Based on sqlite3RunVacuum from vacuum.c +*/ +void sqlcipher_exportFunc(sqlite3_context *context, int argc, sqlite3_value **argv) { + sqlite3 *db = sqlite3_context_db_handle(context); + const char* attachedDb = (const char*) sqlite3_value_text(argv[0]); + int saved_flags; /* Saved value of the db->flags */ + int saved_nChange; /* Saved value of db->nChange */ + int saved_nTotalChange; /* Saved value of db->nTotalChange */ + void (*saved_xTrace)(void*,const char*); /* Saved db->xTrace */ + int rc = SQLITE_OK; /* Return code from service routines */ + char *zSql = NULL; /* SQL statements */ + char *pzErrMsg = NULL; + + saved_flags = db->flags; + saved_nChange = db->nChange; + saved_nTotalChange = db->nTotalChange; + saved_xTrace = db->xTrace; + db->flags |= SQLITE_WriteSchema | SQLITE_IgnoreChecks | SQLITE_PreferBuiltin; + db->flags &= ~(SQLITE_ForeignKeys | SQLITE_ReverseOrder); + db->xTrace = 0; + + /* Query the schema of the main database. Create a mirror schema + ** in the temporary database. + */ + zSql = sqlite3_mprintf( + "SELECT 'CREATE TABLE %s.' || substr(sql,14) " + " FROM sqlite_master WHERE type='table' AND name!='sqlite_sequence'" + " AND rootpage>0" + , attachedDb); + rc = (zSql == NULL) ? SQLITE_NOMEM : sqlcipher_execExecSql(db, &pzErrMsg, zSql); + if( rc!=SQLITE_OK ) goto end_of_export; + sqlite3_free(zSql); + + zSql = sqlite3_mprintf( + "SELECT 'CREATE INDEX %s.' || substr(sql,14)" + " FROM sqlite_master WHERE sql LIKE 'CREATE INDEX %%' " + , attachedDb); + rc = (zSql == NULL) ? SQLITE_NOMEM : sqlcipher_execExecSql(db, &pzErrMsg, zSql); + if( rc!=SQLITE_OK ) goto end_of_export; + sqlite3_free(zSql); + + zSql = sqlite3_mprintf( + "SELECT 'CREATE UNIQUE INDEX %s.' || substr(sql,21) " + " FROM sqlite_master WHERE sql LIKE 'CREATE UNIQUE INDEX %%'" + , attachedDb); + rc = (zSql == NULL) ? SQLITE_NOMEM : sqlcipher_execExecSql(db, &pzErrMsg, zSql); + if( rc!=SQLITE_OK ) goto end_of_export; + sqlite3_free(zSql); + + /* Loop through the tables in the main database. For each, do + ** an "INSERT INTO rekey_db.xxx SELECT * FROM main.xxx;" to copy + ** the contents to the temporary database. + */ + zSql = sqlite3_mprintf( + "SELECT 'INSERT INTO %s.' || quote(name) " + "|| ' SELECT * FROM main.' || quote(name) || ';'" + "FROM main.sqlite_master " + "WHERE type = 'table' AND name!='sqlite_sequence' " + " AND rootpage>0" + , attachedDb); + rc = (zSql == NULL) ? SQLITE_NOMEM : sqlcipher_execExecSql(db, &pzErrMsg, zSql); + if( rc!=SQLITE_OK ) goto end_of_export; + sqlite3_free(zSql); + + /* Copy over the sequence table + */ + zSql = sqlite3_mprintf( + "SELECT 'DELETE FROM %s.' || quote(name) || ';' " + "FROM %s.sqlite_master WHERE name='sqlite_sequence' " + , attachedDb, attachedDb); + rc = (zSql == NULL) ? SQLITE_NOMEM : sqlcipher_execExecSql(db, &pzErrMsg, zSql); + if( rc!=SQLITE_OK ) goto end_of_export; + sqlite3_free(zSql); + + zSql = sqlite3_mprintf( + "SELECT 'INSERT INTO %s.' || quote(name) " + "|| ' SELECT * FROM main.' || quote(name) || ';' " + "FROM %s.sqlite_master WHERE name=='sqlite_sequence';" + , attachedDb, attachedDb); + rc = (zSql == NULL) ? SQLITE_NOMEM : sqlcipher_execExecSql(db, &pzErrMsg, zSql); + if( rc!=SQLITE_OK ) goto end_of_export; + sqlite3_free(zSql); + + /* Copy the triggers, views, and virtual tables from the main database + ** over to the temporary database. None of these objects has any + ** associated storage, so all we have to do is copy their entries + ** from the SQLITE_MASTER table. + */ + zSql = sqlite3_mprintf( + "INSERT INTO %s.sqlite_master " + " SELECT type, name, tbl_name, rootpage, sql" + " FROM main.sqlite_master" + " WHERE type='view' OR type='trigger'" + " OR (type='table' AND rootpage=0)" + , attachedDb); + rc = (zSql == NULL) ? SQLITE_NOMEM : sqlcipher_execSql(db, &pzErrMsg, zSql); + if( rc!=SQLITE_OK ) goto end_of_export; + sqlite3_free(zSql); + + zSql = NULL; +end_of_export: + db->flags = saved_flags; + db->nChange = saved_nChange; + db->nTotalChange = saved_nTotalChange; + db->xTrace = saved_xTrace; + + sqlite3_free(zSql); + + if(rc) { + if(pzErrMsg != NULL) { + sqlite3_result_error(context, pzErrMsg, -1); + sqlite3DbFree(db, pzErrMsg); + } else { + sqlite3_result_error(context, sqlite3ErrStr(rc), -1); + } + } +} + +#endif + +/* END SQLCIPHER */ +#endif + +/************** End of crypto.c **********************************************/ +/************** Begin file crypto_impl.c *************************************/ +/* +** SQLCipher +** http://sqlcipher.net +** +** Copyright (c) 2008 - 2013, ZETETIC LLC +** 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 the ZETETIC LLC 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 ZETETIC LLC ''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 ZETETIC LLC 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. +** +*/ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC + +/************** Include sqlcipher.h in the middle of crypto_impl.c ***********/ +/************** Begin file sqlcipher.h ***************************************/ +/* +** SQLCipher +** sqlcipher.h developed by Stephen Lombardo (Zetetic LLC) +** sjlombardo at zetetic dot net +** http://zetetic.net +** +** Copyright (c) 2008, ZETETIC LLC +** 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 the ZETETIC LLC 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 ZETETIC LLC ''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 ZETETIC LLC 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. +** +*/ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +#ifndef SQLCIPHER_H +#define SQLCIPHER_H + + +typedef struct { + int (*activate)(void *ctx); + int (*deactivate)(void *ctx); + const char* (*get_provider_name)(void *ctx); + int (*add_random)(void *ctx, void *buffer, int length); + int (*random)(void *ctx, void *buffer, int length); + int (*hmac)(void *ctx, unsigned char *hmac_key, int key_sz, unsigned char *in, int in_sz, unsigned char *in2, int in2_sz, unsigned char *out); + int (*kdf)(void *ctx, const unsigned char *pass, int pass_sz, unsigned char* salt, int salt_sz, int workfactor, int key_sz, unsigned char *key); + int (*cipher)(void *ctx, int mode, unsigned char *key, int key_sz, unsigned char *iv, unsigned char *in, int in_sz, unsigned char *out); + int (*set_cipher)(void *ctx, const char *cipher_name); + const char* (*get_cipher)(void *ctx); + int (*get_key_sz)(void *ctx); + int (*get_iv_sz)(void *ctx); + int (*get_block_sz)(void *ctx); + int (*get_hmac_sz)(void *ctx); + int (*ctx_copy)(void *target_ctx, void *source_ctx); + int (*ctx_cmp)(void *c1, void *c2); + int (*ctx_init)(void **ctx); + int (*ctx_free)(void **ctx); +} sqlcipher_provider; + +/* utility functions */ +void sqlcipher_free(void *ptr, int sz); +void* sqlcipher_malloc(int sz); +void* sqlcipher_memset(void *v, unsigned char value, int len); +int sqlcipher_ismemset(const void *v, unsigned char value, int len); +int sqlcipher_memcmp(const void *v0, const void *v1, int len); +void sqlcipher_free(void *, int); + +/* provider interfaces */ +int sqlcipher_register_provider(sqlcipher_provider *p); +sqlcipher_provider* sqlcipher_get_provider(); + +#endif +#endif +/* END SQLCIPHER */ + + +/************** End of sqlcipher.h *******************************************/ +/************** Continuing where we left off in crypto_impl.c ****************/ +#ifndef OMIT_MEMLOCK +#if defined(__unix__) || defined(__APPLE__) +#include +#elif defined(_WIN32) +/* # include */ +#endif +#endif + +/* the default implementation of SQLCipher uses a cipher_ctx + to keep track of read / write state separately. The following + struct and associated functions are defined here */ +typedef struct { + int derive_key; + int kdf_iter; + int fast_kdf_iter; + int key_sz; + int iv_sz; + int block_sz; + int pass_sz; + int reserve_sz; + int hmac_sz; + int keyspec_sz; + unsigned int flags; + unsigned char *key; + unsigned char *hmac_key; + unsigned char *pass; + char *keyspec; + sqlcipher_provider *provider; + void *provider_ctx; +} cipher_ctx; + +static unsigned int default_flags = DEFAULT_CIPHER_FLAGS; +static unsigned char hmac_salt_mask = HMAC_SALT_MASK; +static int default_kdf_iter = PBKDF2_ITER; +static unsigned int sqlcipher_activate_count = 0; +static sqlite3_mutex* sqlcipher_provider_mutex = NULL; +static sqlcipher_provider *default_provider = NULL; + +struct codec_ctx { + int kdf_salt_sz; + int page_sz; + unsigned char *kdf_salt; + unsigned char *hmac_kdf_salt; + unsigned char *buffer; + Btree *pBt; + cipher_ctx *read_ctx; + cipher_ctx *write_ctx; + unsigned int skip_read_hmac; +}; + +int sqlcipher_register_provider(sqlcipher_provider *p) { + sqlite3_mutex_enter(sqlcipher_provider_mutex); + if(default_provider != NULL && default_provider != p) { + /* only free the current registerd provider if it has been initialized + and it isn't a pointer to the same provider passed to the function + (i.e. protect against a caller calling register twice for the same provider) */ + sqlcipher_free(default_provider, sizeof(sqlcipher_provider)); + } + default_provider = p; + sqlite3_mutex_leave(sqlcipher_provider_mutex); + return SQLITE_OK; +} + +/* return a pointer to the currently registered provider. This will + allow an application to fetch the current registered provider and + make minor changes to it */ +sqlcipher_provider* sqlcipher_get_provider() { + return default_provider; +} + +void sqlcipher_activate() { + sqlite3_mutex_enter(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); + + if(sqlcipher_provider_mutex == NULL) { + /* allocate a new mutex to guard access to the provider */ + sqlcipher_provider_mutex = sqlite3_mutex_alloc(SQLITE_MUTEX_FAST); + } + + /* check to see if there is a provider registered at this point + if there no provider registered at this point, register the + default provider */ + if(sqlcipher_get_provider() == NULL) { + sqlcipher_provider *p = sqlcipher_malloc(sizeof(sqlcipher_provider)); +#if defined (SQLCIPHER_CRYPTO_CC) + extern int sqlcipher_cc_setup(sqlcipher_provider *p); + sqlcipher_cc_setup(p); +#elif defined (SQLCIPHER_CRYPTO_LIBTOMCRYPT) + extern int sqlcipher_ltc_setup(sqlcipher_provider *p); + sqlcipher_ltc_setup(p); +#elif defined (SQLCIPHER_CRYPTO_OPENSSL) + extern int sqlcipher_openssl_setup(sqlcipher_provider *p); + sqlcipher_openssl_setup(p); +#else +#error "NO DEFAULT SQLCIPHER CRYPTO PROVIDER DEFINED" +#endif + sqlcipher_register_provider(p); + } + + sqlcipher_activate_count++; /* increment activation count */ + + sqlite3_mutex_leave(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); +} + +void sqlcipher_deactivate() { + sqlite3_mutex_enter(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); + sqlcipher_activate_count--; + /* if no connections are using sqlcipher, cleanup globals */ + if(sqlcipher_activate_count < 1) { + sqlite3_mutex_enter(sqlcipher_provider_mutex); + if(default_provider != NULL) { + sqlcipher_free(default_provider, sizeof(sqlcipher_provider)); + default_provider = NULL; + } + sqlite3_mutex_leave(sqlcipher_provider_mutex); + + /* last connection closed, free provider mutex*/ + sqlite3_mutex_free(sqlcipher_provider_mutex); + sqlcipher_provider_mutex = NULL; + + sqlcipher_activate_count = 0; /* reset activation count */ + } + sqlite3_mutex_leave(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); +} + +/* constant time memset using volitile to avoid having the memset + optimized out by the compiler. + Note: As suggested by Joachim Schipper (joachim.schipper@fox-it.com) +*/ +void* sqlcipher_memset(void *v, unsigned char value, int len) { + int i = 0; + volatile unsigned char *a = v; + + if (v == NULL) return v; + + for(i = 0; i < len; i++) { + a[i] = value; + } + + return v; +} + +/* constant time memory check tests every position of a memory segement + matches a single value (i.e. the memory is all zeros) + returns 0 if match, 1 of no match */ +int sqlcipher_ismemset(const void *v, unsigned char value, int len) { + const unsigned char *a = v; + int i = 0, result = 0; + + for(i = 0; i < len; i++) { + result |= a[i] ^ value; + } + + return (result != 0); +} + +/* constant time memory comparison routine. + returns 0 if match, 1 if no match */ +int sqlcipher_memcmp(const void *v0, const void *v1, int len) { + const unsigned char *a0 = v0, *a1 = v1; + int i = 0, result = 0; + + for(i = 0; i < len; i++) { + result |= a0[i] ^ a1[i]; + } + + return (result != 0); +} + +/** + * Free and wipe memory. Uses SQLites internal sqlite3_free so that memory + * can be countend and memory leak detection works in the test suite. + * If ptr is not null memory will be freed. + * If sz is greater than zero, the memory will be overwritten with zero before it is freed + * If sz is > 0, and not compiled with OMIT_MEMLOCK, system will attempt to unlock the + * memory segment so it can be paged + */ +void sqlcipher_free(void *ptr, int sz) { + if(ptr) { + if(sz > 0) { + sqlcipher_memset(ptr, 0, sz); +#ifndef OMIT_MEMLOCK +#if defined(__unix__) || defined(__APPLE__) + munlock(ptr, sz); +#elif defined(_WIN32) +#if !(defined(WINAPI_FAMILY) && (WINAPI_FAMILY == WINAPI_FAMILY_PHONE_APP || WINAPI_FAMILY == WINAPI_FAMILY_APP)) +VirtualUnlock(ptr, sz); +#endif +#endif +#endif + } + sqlite3_free(ptr); + } +} + +/** + * allocate memory. Uses sqlite's internall malloc wrapper so memory can be + * reference counted and leak detection works. Unless compiled with OMIT_MEMLOCK + * attempts to lock the memory pages so sensitive information won't be swapped + */ +void* sqlcipher_malloc(int sz) { + void *ptr = sqlite3Malloc(sz); + sqlcipher_memset(ptr, 0, sz); +#ifndef OMIT_MEMLOCK + if(ptr) { +#if defined(__unix__) || defined(__APPLE__) + mlock(ptr, sz); +#elif defined(_WIN32) +#if !(defined(WINAPI_FAMILY) && (WINAPI_FAMILY == WINAPI_FAMILY_PHONE_APP || WINAPI_FAMILY == WINAPI_FAMILY_APP)) + VirtualLock(ptr, sz); +#endif +#endif + } +#endif + return ptr; +} + + +/** + * Initialize new cipher_ctx struct. This function will allocate memory + * for the cipher context and for the key + * + * returns SQLITE_OK if initialization was successful + * returns SQLITE_NOMEM if an error occured allocating memory + */ +static int sqlcipher_cipher_ctx_init(cipher_ctx **iCtx) { + int rc; + cipher_ctx *ctx; + *iCtx = (cipher_ctx *) sqlcipher_malloc(sizeof(cipher_ctx)); + ctx = *iCtx; + if(ctx == NULL) return SQLITE_NOMEM; + + ctx->provider = (sqlcipher_provider *) sqlcipher_malloc(sizeof(sqlcipher_provider)); + if(ctx->provider == NULL) return SQLITE_NOMEM; + + /* make a copy of the provider to be used for the duration of the context */ + sqlite3_mutex_enter(sqlcipher_provider_mutex); + memcpy(ctx->provider, default_provider, sizeof(sqlcipher_provider)); + sqlite3_mutex_leave(sqlcipher_provider_mutex); + + if((rc = ctx->provider->ctx_init(&ctx->provider_ctx)) != SQLITE_OK) return rc; + ctx->key = (unsigned char *) sqlcipher_malloc(CIPHER_MAX_KEY_SZ); + ctx->hmac_key = (unsigned char *) sqlcipher_malloc(CIPHER_MAX_KEY_SZ); + if(ctx->key == NULL) return SQLITE_NOMEM; + if(ctx->hmac_key == NULL) return SQLITE_NOMEM; + + /* setup default flags */ + ctx->flags = default_flags; + + return SQLITE_OK; +} + +/** + * Free and wipe memory associated with a cipher_ctx + */ +static void sqlcipher_cipher_ctx_free(cipher_ctx **iCtx) { + cipher_ctx *ctx = *iCtx; + CODEC_TRACE(("cipher_ctx_free: entered iCtx=%p\n", iCtx)); + ctx->provider->ctx_free(&ctx->provider_ctx); + sqlcipher_free(ctx->provider, sizeof(sqlcipher_provider)); + sqlcipher_free(ctx->key, ctx->key_sz); + sqlcipher_free(ctx->hmac_key, ctx->key_sz); + sqlcipher_free(ctx->pass, ctx->pass_sz); + sqlcipher_free(ctx->keyspec, ctx->keyspec_sz); + sqlcipher_free(ctx, sizeof(cipher_ctx)); +} + +/** + * Compare one cipher_ctx to another. + * + * returns 0 if all the parameters (except the derived key data) are the same + * returns 1 otherwise + */ +static int sqlcipher_cipher_ctx_cmp(cipher_ctx *c1, cipher_ctx *c2) { + CODEC_TRACE(("sqlcipher_cipher_ctx_cmp: entered c1=%p c2=%p\n", c1, c2)); + + if( + c1->iv_sz == c2->iv_sz + && c1->kdf_iter == c2->kdf_iter + && c1->fast_kdf_iter == c2->fast_kdf_iter + && c1->key_sz == c2->key_sz + && c1->pass_sz == c2->pass_sz + && c1->flags == c2->flags + && c1->hmac_sz == c2->hmac_sz + && c1->provider->ctx_cmp(c1->provider_ctx, c2->provider_ctx) + && ( + c1->pass == c2->pass + || !sqlcipher_memcmp((const unsigned char*)c1->pass, + (const unsigned char*)c2->pass, + c1->pass_sz) + ) + ) return 0; + return 1; +} + +/** + * Copy one cipher_ctx to another. For instance, assuming that read_ctx is a + * fully initialized context, you could copy it to write_ctx and all yet data + * and pass information across + * + * returns SQLITE_OK if initialization was successful + * returns SQLITE_NOMEM if an error occured allocating memory + */ +static int sqlcipher_cipher_ctx_copy(cipher_ctx *target, cipher_ctx *source) { + void *key = target->key; + void *hmac_key = target->hmac_key; + void *provider = target->provider; + void *provider_ctx = target->provider_ctx; + + CODEC_TRACE(("sqlcipher_cipher_ctx_copy: entered target=%p, source=%p\n", target, source)); + sqlcipher_free(target->pass, target->pass_sz); + sqlcipher_free(target->keyspec, target->keyspec_sz); + memcpy(target, source, sizeof(cipher_ctx)); + + target->key = key; //restore pointer to previously allocated key data + memcpy(target->key, source->key, CIPHER_MAX_KEY_SZ); + + target->hmac_key = hmac_key; //restore pointer to previously allocated hmac key data + memcpy(target->hmac_key, source->hmac_key, CIPHER_MAX_KEY_SZ); + + target->provider = provider; // restore pointer to previouly allocated provider; + memcpy(target->provider, source->provider, sizeof(sqlcipher_provider)); + + target->provider_ctx = provider_ctx; // restore pointer to previouly allocated provider context; + target->provider->ctx_copy(target->provider_ctx, source->provider_ctx); + + if(source->pass && source->pass_sz) { + target->pass = sqlcipher_malloc(source->pass_sz); + if(target->pass == NULL) return SQLITE_NOMEM; + memcpy(target->pass, source->pass, source->pass_sz); + } + if(source->keyspec && source->keyspec_sz) { + target->keyspec = sqlcipher_malloc(source->keyspec_sz); + if(target->keyspec == NULL) return SQLITE_NOMEM; + memcpy(target->keyspec, source->keyspec, source->keyspec_sz); + } + return SQLITE_OK; +} + +/** + * Set the keyspec for the cipher_ctx + * + * returns SQLITE_OK if assignment was successfull + * returns SQLITE_NOMEM if an error occured allocating memory + */ +static int sqlcipher_cipher_ctx_set_keyspec(cipher_ctx *ctx, const unsigned char *key, int key_sz, const unsigned char *salt, int salt_sz) { + + /* free, zero existing pointers and size */ + sqlcipher_free(ctx->keyspec, ctx->keyspec_sz); + ctx->keyspec = NULL; + ctx->keyspec_sz = 0; + + /* establic a hex-formated key specification, containing the raw encryption key and + the salt used to generate it */ + ctx->keyspec_sz = ((key_sz + salt_sz) * 2) + 3; + ctx->keyspec = sqlcipher_malloc(ctx->keyspec_sz); + if(ctx->keyspec == NULL) return SQLITE_NOMEM; + + ctx->keyspec[0] = 'x'; + ctx->keyspec[1] = '\''; + cipher_bin2hex(key, key_sz, ctx->keyspec + 2); + cipher_bin2hex(salt, salt_sz, ctx->keyspec + (key_sz * 2) + 2); + ctx->keyspec[ctx->keyspec_sz - 1] = '\''; + return SQLITE_OK; +} + +/** + * Set the passphrase for the cipher_ctx + * + * returns SQLITE_OK if assignment was successfull + * returns SQLITE_NOMEM if an error occured allocating memory + */ +static int sqlcipher_cipher_ctx_set_pass(cipher_ctx *ctx, const void *zKey, int nKey) { + + /* free, zero existing pointers and size */ + sqlcipher_free(ctx->pass, ctx->pass_sz); + ctx->pass = NULL; + ctx->pass_sz = 0; + + if(zKey && nKey) { /* if new password is provided, copy it */ + ctx->pass_sz = nKey; + ctx->pass = sqlcipher_malloc(nKey); + if(ctx->pass == NULL) return SQLITE_NOMEM; + memcpy(ctx->pass, zKey, nKey); + } + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_set_pass(codec_ctx *ctx, const void *zKey, int nKey, int for_ctx) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + int rc; + + if((rc = sqlcipher_cipher_ctx_set_pass(c_ctx, zKey, nKey)) != SQLITE_OK) return rc; + c_ctx->derive_key = 1; + + if(for_ctx == 2) + if((rc = sqlcipher_cipher_ctx_copy( for_ctx ? ctx->read_ctx : ctx->write_ctx, c_ctx)) != SQLITE_OK) + return rc; + + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_set_cipher(codec_ctx *ctx, const char *cipher_name, int for_ctx) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + int rc; + + c_ctx->provider->set_cipher(c_ctx->provider_ctx, cipher_name); + + c_ctx->key_sz = c_ctx->provider->get_key_sz(c_ctx->provider_ctx); + c_ctx->iv_sz = c_ctx->provider->get_iv_sz(c_ctx->provider_ctx); + c_ctx->block_sz = c_ctx->provider->get_block_sz(c_ctx->provider_ctx); + c_ctx->hmac_sz = c_ctx->provider->get_hmac_sz(c_ctx->provider_ctx); + c_ctx->derive_key = 1; + + if(for_ctx == 2) + if((rc = sqlcipher_cipher_ctx_copy( for_ctx ? ctx->read_ctx : ctx->write_ctx, c_ctx)) != SQLITE_OK) + return rc; + + return SQLITE_OK; +} + +const char* sqlcipher_codec_ctx_get_cipher(codec_ctx *ctx, int for_ctx) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + return c_ctx->provider->get_cipher(c_ctx->provider_ctx); +} + +/* set the global default KDF iteration */ +void sqlcipher_set_default_kdf_iter(int iter) { + default_kdf_iter = iter; +} + +int sqlcipher_get_default_kdf_iter() { + return default_kdf_iter; +} + +int sqlcipher_codec_ctx_set_kdf_iter(codec_ctx *ctx, int kdf_iter, int for_ctx) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + int rc; + + c_ctx->kdf_iter = kdf_iter; + c_ctx->derive_key = 1; + + if(for_ctx == 2) + if((rc = sqlcipher_cipher_ctx_copy( for_ctx ? ctx->read_ctx : ctx->write_ctx, c_ctx)) != SQLITE_OK) + return rc; + + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_get_kdf_iter(codec_ctx *ctx, int for_ctx) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + return c_ctx->kdf_iter; +} + +int sqlcipher_codec_ctx_set_fast_kdf_iter(codec_ctx *ctx, int fast_kdf_iter, int for_ctx) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + int rc; + + c_ctx->fast_kdf_iter = fast_kdf_iter; + c_ctx->derive_key = 1; + + if(for_ctx == 2) + if((rc = sqlcipher_cipher_ctx_copy( for_ctx ? ctx->read_ctx : ctx->write_ctx, c_ctx)) != SQLITE_OK) + return rc; + + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_get_fast_kdf_iter(codec_ctx *ctx, int for_ctx) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + return c_ctx->fast_kdf_iter; +} + +/* set the global default flag for HMAC */ +void sqlcipher_set_default_use_hmac(int use) { + if(use) default_flags |= CIPHER_FLAG_HMAC; + else default_flags &= ~CIPHER_FLAG_HMAC; +} + +int sqlcipher_get_default_use_hmac() { + return (default_flags & CIPHER_FLAG_HMAC) != 0; +} + +void sqlcipher_set_hmac_salt_mask(unsigned char mask) { + hmac_salt_mask = mask; +} + +unsigned char sqlcipher_get_hmac_salt_mask() { + return hmac_salt_mask; +} + +/* set the codec flag for whether this individual database should be using hmac */ +int sqlcipher_codec_ctx_set_use_hmac(codec_ctx *ctx, int use) { + int reserve = CIPHER_MAX_IV_SZ; /* base reserve size will be IV only */ + + if(use) reserve += ctx->read_ctx->hmac_sz; /* if reserve will include hmac, update that size */ + + /* calculate the amount of reserve needed in even increments of the cipher block size */ + + reserve = ((reserve % ctx->read_ctx->block_sz) == 0) ? reserve : + ((reserve / ctx->read_ctx->block_sz) + 1) * ctx->read_ctx->block_sz; + + CODEC_TRACE(("sqlcipher_codec_ctx_set_use_hmac: use=%d block_sz=%d md_size=%d reserve=%d\n", + use, ctx->read_ctx->block_sz, ctx->read_ctx->hmac_sz, reserve)); + + + if(use) { + sqlcipher_codec_ctx_set_flag(ctx, CIPHER_FLAG_HMAC); + } else { + sqlcipher_codec_ctx_unset_flag(ctx, CIPHER_FLAG_HMAC); + } + + ctx->write_ctx->reserve_sz = ctx->read_ctx->reserve_sz = reserve; + + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_get_use_hmac(codec_ctx *ctx, int for_ctx) { + cipher_ctx * c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + return (c_ctx->flags & CIPHER_FLAG_HMAC) != 0; +} + +int sqlcipher_codec_ctx_set_flag(codec_ctx *ctx, unsigned int flag) { + ctx->write_ctx->flags |= flag; + ctx->read_ctx->flags |= flag; + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_unset_flag(codec_ctx *ctx, unsigned int flag) { + ctx->write_ctx->flags &= ~flag; + ctx->read_ctx->flags &= ~flag; + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_get_flag(codec_ctx *ctx, unsigned int flag, int for_ctx) { + cipher_ctx * c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + return (c_ctx->flags & flag) != 0; +} + +void sqlcipher_codec_ctx_set_error(codec_ctx *ctx, int error) { + CODEC_TRACE(("sqlcipher_codec_ctx_set_error: ctx=%p, error=%d\n", ctx, error)); + sqlite3pager_sqlite3PagerSetError(ctx->pBt->pBt->pPager, error); + ctx->pBt->pBt->db->errCode = error; +} + +int sqlcipher_codec_ctx_get_reservesize(codec_ctx *ctx) { + return ctx->read_ctx->reserve_sz; +} + +void* sqlcipher_codec_ctx_get_data(codec_ctx *ctx) { + return ctx->buffer; +} + +void* sqlcipher_codec_ctx_get_kdf_salt(codec_ctx *ctx) { + return ctx->kdf_salt; +} + +void sqlcipher_codec_get_keyspec(codec_ctx *ctx, void **zKey, int *nKey) { + *zKey = ctx->read_ctx->keyspec; + *nKey = ctx->read_ctx->keyspec_sz; +} + +int sqlcipher_codec_ctx_set_pagesize(codec_ctx *ctx, int size) { + /* attempt to free the existing page buffer */ + sqlcipher_free(ctx->buffer,ctx->page_sz); + ctx->page_sz = size; + + /* pre-allocate a page buffer of PageSize bytes. This will + be used as a persistent buffer for encryption and decryption + operations to avoid overhead of multiple memory allocations*/ + ctx->buffer = sqlcipher_malloc(size); + if(ctx->buffer == NULL) return SQLITE_NOMEM; + + return SQLITE_OK; +} + +int sqlcipher_codec_ctx_get_pagesize(codec_ctx *ctx) { + return ctx->page_sz; +} + +int sqlcipher_codec_ctx_init(codec_ctx **iCtx, Db *pDb, Pager *pPager, sqlite3_file *fd, const void *zKey, int nKey) { + int rc; + codec_ctx *ctx; + *iCtx = sqlcipher_malloc(sizeof(codec_ctx)); + ctx = *iCtx; + + if(ctx == NULL) return SQLITE_NOMEM; + + ctx->pBt = pDb->pBt; /* assign pointer to database btree structure */ + + /* allocate space for salt data. Then read the first 16 bytes + directly off the database file. This is the salt for the + key derivation function. If we get a short read allocate + a new random salt value */ + ctx->kdf_salt_sz = FILE_HEADER_SZ; + ctx->kdf_salt = sqlcipher_malloc(ctx->kdf_salt_sz); + if(ctx->kdf_salt == NULL) return SQLITE_NOMEM; + + /* allocate space for separate hmac salt data. We want the + HMAC derivation salt to be different than the encryption + key derivation salt */ + ctx->hmac_kdf_salt = sqlcipher_malloc(ctx->kdf_salt_sz); + if(ctx->hmac_kdf_salt == NULL) return SQLITE_NOMEM; + + + /* + Always overwrite page size and set to the default because the first page of the database + in encrypted and thus sqlite can't effectively determine the pagesize. this causes an issue in + cases where bytes 16 & 17 of the page header are a power of 2 as reported by John Lehman + */ + if((rc = sqlcipher_codec_ctx_set_pagesize(ctx, SQLITE_DEFAULT_PAGE_SIZE)) != SQLITE_OK) return rc; + + if((rc = sqlcipher_cipher_ctx_init(&ctx->read_ctx)) != SQLITE_OK) return rc; + if((rc = sqlcipher_cipher_ctx_init(&ctx->write_ctx)) != SQLITE_OK) return rc; + + if(fd == NULL || sqlite3OsRead(fd, ctx->kdf_salt, FILE_HEADER_SZ, 0) != SQLITE_OK) { + /* if unable to read the bytes, generate random salt */ + if(ctx->read_ctx->provider->random(ctx->read_ctx->provider_ctx, ctx->kdf_salt, FILE_HEADER_SZ) != SQLITE_OK) return SQLITE_ERROR; + } + + if((rc = sqlcipher_codec_ctx_set_cipher(ctx, CIPHER, 0)) != SQLITE_OK) return rc; + if((rc = sqlcipher_codec_ctx_set_kdf_iter(ctx, default_kdf_iter, 0)) != SQLITE_OK) return rc; + if((rc = sqlcipher_codec_ctx_set_fast_kdf_iter(ctx, FAST_PBKDF2_ITER, 0)) != SQLITE_OK) return rc; + if((rc = sqlcipher_codec_ctx_set_pass(ctx, zKey, nKey, 0)) != SQLITE_OK) return rc; + + /* Note that use_hmac is a special case that requires recalculation of page size + so we call set_use_hmac to perform setup */ + if((rc = sqlcipher_codec_ctx_set_use_hmac(ctx, default_flags & CIPHER_FLAG_HMAC)) != SQLITE_OK) return rc; + + if((rc = sqlcipher_cipher_ctx_copy(ctx->write_ctx, ctx->read_ctx)) != SQLITE_OK) return rc; + + return SQLITE_OK; +} + +/** + * Free and wipe memory associated with a cipher_ctx, including the allocated + * read_ctx and write_ctx. + */ +void sqlcipher_codec_ctx_free(codec_ctx **iCtx) { + codec_ctx *ctx = *iCtx; + CODEC_TRACE(("codec_ctx_free: entered iCtx=%p\n", iCtx)); + sqlcipher_free(ctx->kdf_salt, ctx->kdf_salt_sz); + sqlcipher_free(ctx->hmac_kdf_salt, ctx->kdf_salt_sz); + sqlcipher_free(ctx->buffer, 0); + sqlcipher_cipher_ctx_free(&ctx->read_ctx); + sqlcipher_cipher_ctx_free(&ctx->write_ctx); + sqlcipher_free(ctx, sizeof(codec_ctx)); +} + +/** convert a 32bit unsigned integer to little endian byte ordering */ +static void sqlcipher_put4byte_le(unsigned char *p, u32 v) { + p[0] = (u8)v; + p[1] = (u8)(v>>8); + p[2] = (u8)(v>>16); + p[3] = (u8)(v>>24); +} + +static int sqlcipher_page_hmac(cipher_ctx *ctx, Pgno pgno, unsigned char *in, int in_sz, unsigned char *out) { + unsigned char pgno_raw[sizeof(pgno)]; + /* we may convert page number to consistent representation before calculating MAC for + compatibility across big-endian and little-endian platforms. + + Note: The public release of sqlcipher 2.0.0 to 2.0.6 had a bug where the bytes of pgno + were used directly in the MAC. SQLCipher convert's to little endian by default to preserve + backwards compatibility on the most popular platforms, but can optionally be configured + to use either big endian or native byte ordering via pragma. */ + + if(ctx->flags & CIPHER_FLAG_LE_PGNO) { /* compute hmac using little endian pgno*/ + sqlcipher_put4byte_le(pgno_raw, pgno); + } else if(ctx->flags & CIPHER_FLAG_BE_PGNO) { /* compute hmac using big endian pgno */ + sqlite3Put4byte(pgno_raw, pgno); /* sqlite3Put4byte converts 32bit uint to big endian */ + } else { /* use native byte ordering */ + memcpy(pgno_raw, &pgno, sizeof(pgno)); + } + + /* include the encrypted page data, initialization vector, and page number in HMAC. This will + prevent both tampering with the ciphertext, manipulation of the IV, or resequencing otherwise + valid pages out of order in a database */ + ctx->provider->hmac( + ctx->provider_ctx, ctx->hmac_key, + ctx->key_sz, in, + in_sz, (unsigned char*) &pgno_raw, + sizeof(pgno), out); + return SQLITE_OK; +} + +/* + * ctx - codec context + * pgno - page number in database + * size - size in bytes of input and output buffers + * mode - 1 to encrypt, 0 to decrypt + * in - pointer to input bytes + * out - pouter to output bytes + */ +int sqlcipher_page_cipher(codec_ctx *ctx, int for_ctx, Pgno pgno, int mode, int page_sz, unsigned char *in, unsigned char *out) { + cipher_ctx *c_ctx = for_ctx ? ctx->write_ctx : ctx->read_ctx; + unsigned char *iv_in, *iv_out, *hmac_in, *hmac_out, *out_start; + int size; + + /* calculate some required positions into various buffers */ + size = page_sz - c_ctx->reserve_sz; /* adjust size to useable size and memset reserve at end of page */ + iv_out = out + size; + iv_in = in + size; + + /* hmac will be written immediately after the initialization vector. the remainder of the page reserve will contain + random bytes. note, these pointers are only valid when using hmac */ + hmac_in = in + size + c_ctx->iv_sz; + hmac_out = out + size + c_ctx->iv_sz; + out_start = out; /* note the original position of the output buffer pointer, as out will be rewritten during encryption */ + + CODEC_TRACE(("codec_cipher:entered pgno=%d, mode=%d, size=%d\n", pgno, mode, size)); + CODEC_HEXDUMP("codec_cipher: input page data", in, page_sz); + + /* the key size should never be zero. If it is, error out. */ + if(c_ctx->key_sz == 0) { + CODEC_TRACE(("codec_cipher: error possible context corruption, key_sz is zero for pgno=%d\n", pgno)); + sqlcipher_memset(out, 0, page_sz); + return SQLITE_ERROR; + } + + if(mode == CIPHER_ENCRYPT) { + /* start at front of the reserve block, write random data to the end */ + if(c_ctx->provider->random(c_ctx->provider_ctx, iv_out, c_ctx->reserve_sz) != SQLITE_OK) return SQLITE_ERROR; + } else { /* CIPHER_DECRYPT */ + memcpy(iv_out, iv_in, c_ctx->iv_sz); /* copy the iv from the input to output buffer */ + } + + if((c_ctx->flags & CIPHER_FLAG_HMAC) && (mode == CIPHER_DECRYPT) && !ctx->skip_read_hmac) { + if(sqlcipher_page_hmac(c_ctx, pgno, in, size + c_ctx->iv_sz, hmac_out) != SQLITE_OK) { + sqlcipher_memset(out, 0, page_sz); + CODEC_TRACE(("codec_cipher: hmac operations failed for pgno=%d\n", pgno)); + return SQLITE_ERROR; + } + + CODEC_TRACE(("codec_cipher: comparing hmac on in=%p out=%p hmac_sz=%d\n", hmac_in, hmac_out, c_ctx->hmac_sz)); + if(sqlcipher_memcmp(hmac_in, hmac_out, c_ctx->hmac_sz) != 0) { /* the hmac check failed */ + if(sqlcipher_ismemset(in, 0, page_sz) == 0) { + /* first check if the entire contents of the page is zeros. If so, this page + resulted from a short read (i.e. sqlite attempted to pull a page after the end of the file. these + short read failures must be ignored for autovaccum mode to work so wipe the output buffer + and return SQLITE_OK to skip the decryption step. */ + CODEC_TRACE(("codec_cipher: zeroed page (short read) for pgno %d, encryption but returning SQLITE_OK\n", pgno)); + sqlcipher_memset(out, 0, page_sz); + return SQLITE_OK; + } else { + /* if the page memory is not all zeros, it means the there was data and a hmac on the page. + since the check failed, the page was either tampered with or corrupted. wipe the output buffer, + and return SQLITE_ERROR to the caller */ + CODEC_TRACE(("codec_cipher: hmac check failed for pgno=%d returning SQLITE_ERROR\n", pgno)); + sqlcipher_memset(out, 0, page_sz); + return SQLITE_ERROR; + } + } + } + + c_ctx->provider->cipher(c_ctx->provider_ctx, mode, c_ctx->key, c_ctx->key_sz, iv_out, in, size, out); + + if((c_ctx->flags & CIPHER_FLAG_HMAC) && (mode == CIPHER_ENCRYPT)) { + sqlcipher_page_hmac(c_ctx, pgno, out_start, size + c_ctx->iv_sz, hmac_out); + } + + CODEC_HEXDUMP("codec_cipher: output page data", out_start, page_sz); + + return SQLITE_OK; +} + +/** + * Derive an encryption key for a cipher contex key based on the raw password. + * + * If the raw key data is formated as x'hex' and there are exactly enough hex chars to fill + * the key (i.e 64 hex chars for a 256 bit key) then the key data will be used directly. + + * Else, if the raw key data is formated as x'hex' and there are exactly enough hex chars to fill + * the key and the salt (i.e 92 hex chars for a 256 bit key and 16 byte salt) then it will be unpacked + * as the key followed by the salt. + * + * Otherwise, a key data will be derived using PBKDF2 + * + * returns SQLITE_OK if initialization was successful + * returns SQLITE_ERROR if the key could't be derived (for instance if pass is NULL or pass_sz is 0) + */ +static int sqlcipher_cipher_ctx_key_derive(codec_ctx *ctx, cipher_ctx *c_ctx) { + int rc; + CODEC_TRACE(("cipher_ctx_key_derive: entered c_ctx->pass=%s, c_ctx->pass_sz=%d \ + ctx->kdf_salt=%p ctx->kdf_salt_sz=%d c_ctx->kdf_iter=%d \ + ctx->hmac_kdf_salt=%p, c_ctx->fast_kdf_iter=%d c_ctx->key_sz=%d\n", + c_ctx->pass, c_ctx->pass_sz, ctx->kdf_salt, ctx->kdf_salt_sz, c_ctx->kdf_iter, + ctx->hmac_kdf_salt, c_ctx->fast_kdf_iter, c_ctx->key_sz)); + + + if(c_ctx->pass && c_ctx->pass_sz) { // if pass is not null + if (c_ctx->pass_sz == ((c_ctx->key_sz * 2) + 3) && sqlite3StrNICmp((const char *)c_ctx->pass ,"x'", 2) == 0) { + int n = c_ctx->pass_sz - 3; /* adjust for leading x' and tailing ' */ + const unsigned char *z = c_ctx->pass + 2; /* adjust lead offset of x' */ + CODEC_TRACE(("cipher_ctx_key_derive: using raw key from hex\n")); + cipher_hex2bin(z, n, c_ctx->key); + } else if (c_ctx->pass_sz == (((c_ctx->key_sz + ctx->kdf_salt_sz) * 2) + 3) && sqlite3StrNICmp((const char *)c_ctx->pass ,"x'", 2) == 0) { + const unsigned char *z = c_ctx->pass + 2; /* adjust lead offset of x' */ + CODEC_TRACE(("cipher_ctx_key_derive: using raw key from hex\n")); + cipher_hex2bin(z, (c_ctx->key_sz * 2), c_ctx->key); + cipher_hex2bin(z + (c_ctx->key_sz * 2), (ctx->kdf_salt_sz * 2), ctx->kdf_salt); + } else { + CODEC_TRACE(("cipher_ctx_key_derive: deriving key using full PBKDF2 with %d iterations\n", c_ctx->kdf_iter)); + c_ctx->provider->kdf(c_ctx->provider_ctx, c_ctx->pass, c_ctx->pass_sz, + ctx->kdf_salt, ctx->kdf_salt_sz, c_ctx->kdf_iter, + c_ctx->key_sz, c_ctx->key); + } + + /* set the context "keyspec" containing the hex-formatted key and salt to be used when attaching databases */ + if((rc = sqlcipher_cipher_ctx_set_keyspec(c_ctx, c_ctx->key, c_ctx->key_sz, ctx->kdf_salt, ctx->kdf_salt_sz)) != SQLITE_OK) return rc; + + /* if this context is setup to use hmac checks, generate a seperate and different + key for HMAC. In this case, we use the output of the previous KDF as the input to + this KDF run. This ensures a distinct but predictable HMAC key. */ + if(c_ctx->flags & CIPHER_FLAG_HMAC) { + int i; + + /* start by copying the kdf key into the hmac salt slot + then XOR it with the fixed hmac salt defined at compile time + this ensures that the salt passed in to derive the hmac key, while + easy to derive and publically known, is not the same as the salt used + to generate the encryption key */ + memcpy(ctx->hmac_kdf_salt, ctx->kdf_salt, ctx->kdf_salt_sz); + for(i = 0; i < ctx->kdf_salt_sz; i++) { + ctx->hmac_kdf_salt[i] ^= hmac_salt_mask; + } + + CODEC_TRACE(("cipher_ctx_key_derive: deriving hmac key from encryption key using PBKDF2 with %d iterations\n", + c_ctx->fast_kdf_iter)); + + + c_ctx->provider->kdf(c_ctx->provider_ctx, c_ctx->key, c_ctx->key_sz, + ctx->hmac_kdf_salt, ctx->kdf_salt_sz, c_ctx->fast_kdf_iter, + c_ctx->key_sz, c_ctx->hmac_key); + } + + c_ctx->derive_key = 0; + return SQLITE_OK; + }; + return SQLITE_ERROR; +} + +int sqlcipher_codec_key_derive(codec_ctx *ctx) { + /* derive key on first use if necessary */ + if(ctx->read_ctx->derive_key) { + if(sqlcipher_cipher_ctx_key_derive(ctx, ctx->read_ctx) != SQLITE_OK) return SQLITE_ERROR; + } + + if(ctx->write_ctx->derive_key) { + if(sqlcipher_cipher_ctx_cmp(ctx->write_ctx, ctx->read_ctx) == 0) { + /* the relevant parameters are the same, just copy read key */ + if(sqlcipher_cipher_ctx_copy(ctx->write_ctx, ctx->read_ctx) != SQLITE_OK) return SQLITE_ERROR; + } else { + if(sqlcipher_cipher_ctx_key_derive(ctx, ctx->write_ctx) != SQLITE_OK) return SQLITE_ERROR; + } + } + + /* TODO: wipe and free passphrase after key derivation */ + sqlcipher_cipher_ctx_set_pass(ctx->read_ctx, NULL, 0); + sqlcipher_cipher_ctx_set_pass(ctx->write_ctx, NULL, 0); + + return SQLITE_OK; +} + +int sqlcipher_codec_key_copy(codec_ctx *ctx, int source) { + if(source == CIPHER_READ_CTX) { + return sqlcipher_cipher_ctx_copy(ctx->write_ctx, ctx->read_ctx); + } else { + return sqlcipher_cipher_ctx_copy(ctx->read_ctx, ctx->write_ctx); + } +} + +const char* sqlcipher_codec_get_cipher_provider(codec_ctx *ctx) { + return ctx->read_ctx->provider->get_provider_name(ctx->read_ctx); +} + + +static int sqlcipher_check_connection(const char *filename, char *key, int key_sz, char *sql, int *user_version) { + int rc; + sqlite3 *db = NULL; + sqlite3_stmt *statement = NULL; + char *query_user_version = "PRAGMA user_version;"; + + rc = sqlite3_open(filename, &db); + if(rc != SQLITE_OK){ + goto cleanup; + } + rc = sqlite3_key(db, key, key_sz); + if(rc != SQLITE_OK){ + goto cleanup; + } + rc = sqlite3_exec(db, sql, NULL, NULL, NULL); + if(rc != SQLITE_OK){ + goto cleanup; + } + rc = sqlite3_prepare(db, query_user_version, -1, &statement, NULL); + if(rc != SQLITE_OK){ + goto cleanup; + } + rc = sqlite3_step(statement); + if(rc == SQLITE_ROW){ + *user_version = sqlite3_column_int(statement, 0); + rc = SQLITE_OK; + } + +cleanup: + if(statement){ + sqlite3_finalize(statement); + } + if(db){ + sqlite3_close(db); + } + return rc; +} + +int sqlcipher_codec_ctx_migrate(codec_ctx *ctx) { + u32 meta; + int rc = 0; + int command_idx = 0; + int password_sz; + int saved_flags; + int saved_nChange; + int saved_nTotalChange; + void (*saved_xTrace)(void*,const char*); + Db *pDb = 0; + sqlite3 *db = ctx->pBt->db; + const char *db_filename = sqlite3_db_filename(db, "main"); + char *migrated_db_filename = sqlite3_mprintf("%s-migrated", db_filename); + char *pragma_hmac_off = "PRAGMA cipher_use_hmac = OFF;"; + char *pragma_4k_kdf_iter = "PRAGMA kdf_iter = 4000;"; + char *pragma_1x_and_4k; + char *set_user_version; + char *key; + int key_sz; + int user_version = 0; + int upgrade_1x_format = 0; + int upgrade_4k_format = 0; + static const unsigned char aCopy[] = { + BTREE_SCHEMA_VERSION, 1, /* Add one to the old schema cookie */ + BTREE_DEFAULT_CACHE_SIZE, 0, /* Preserve the default page cache size */ + BTREE_TEXT_ENCODING, 0, /* Preserve the text encoding */ + BTREE_USER_VERSION, 0, /* Preserve the user version */ + BTREE_APPLICATION_ID, 0, /* Preserve the application id */ + }; + + + key_sz = ctx->read_ctx->pass_sz + 1; + key = sqlcipher_malloc(key_sz); + memset(key, 0, key_sz); + memcpy(key, ctx->read_ctx->pass, ctx->read_ctx->pass_sz); + + if(db_filename){ + const char* commands[5]; + char *attach_command = sqlite3_mprintf("ATTACH DATABASE '%s-migrated' as migrate KEY '%q';", + db_filename, key); + + int rc = sqlcipher_check_connection(db_filename, key, key_sz, "", &user_version); + if(rc == SQLITE_OK){ + CODEC_TRACE(("No upgrade required - exiting\n")); + goto exit; + } + + // Version 2 - check for 4k with hmac format + rc = sqlcipher_check_connection(db_filename, key, key_sz, pragma_4k_kdf_iter, &user_version); + if(rc == SQLITE_OK) { + CODEC_TRACE(("Version 2 format found\n")); + upgrade_4k_format = 1; + } + + // Version 1 - check both no hmac and 4k together + pragma_1x_and_4k = sqlite3_mprintf("%s%s", pragma_hmac_off, + pragma_4k_kdf_iter); + rc = sqlcipher_check_connection(db_filename, key, key_sz, pragma_1x_and_4k, &user_version); + sqlite3_free(pragma_1x_and_4k); + if(rc == SQLITE_OK) { + CODEC_TRACE(("Version 1 format found\n")); + upgrade_1x_format = 1; + upgrade_4k_format = 1; + } + + if(upgrade_1x_format == 0 && upgrade_4k_format == 0) { + CODEC_TRACE(("Upgrade format not determined\n")); + goto handle_error; + } + + set_user_version = sqlite3_mprintf("PRAGMA migrate.user_version = %d;", user_version); + commands[0] = upgrade_4k_format == 1 ? pragma_4k_kdf_iter : ""; + commands[1] = upgrade_1x_format == 1 ? pragma_hmac_off : ""; + commands[2] = attach_command; + commands[3] = "SELECT sqlcipher_export('migrate');"; + commands[4] = set_user_version; + + for(command_idx = 0; command_idx < ArraySize(commands); command_idx++){ + const char *command = commands[command_idx]; + if(strcmp(command, "") == 0){ + continue; + } + rc = sqlite3_exec(db, command, NULL, NULL, NULL); + if(rc != SQLITE_OK){ + break; + } + } + sqlite3_free(attach_command); + sqlite3_free(set_user_version); + sqlcipher_free(key, key_sz); + + if(rc == SQLITE_OK){ + Btree *pDest; + Btree *pSrc; + int i = 0; + + if( !db->autoCommit ){ + CODEC_TRACE(("cannot migrate from within a transaction")); + goto handle_error; + } + if( db->nVdbeActive>1 ){ + CODEC_TRACE(("cannot migrate - SQL statements in progress")); + goto handle_error; + } + + /* Save the current value of the database flags so that it can be + ** restored before returning. Then set the writable-schema flag, and + ** disable CHECK and foreign key constraints. */ + saved_flags = db->flags; + saved_nChange = db->nChange; + saved_nTotalChange = db->nTotalChange; + saved_xTrace = db->xTrace; + db->flags |= SQLITE_WriteSchema | SQLITE_IgnoreChecks | SQLITE_PreferBuiltin; + db->flags &= ~(SQLITE_ForeignKeys | SQLITE_ReverseOrder); + db->xTrace = 0; + + pDest = db->aDb[0].pBt; + pDb = &(db->aDb[db->nDb-1]); + pSrc = pDb->pBt; + + rc = sqlite3_exec(db, "BEGIN;", NULL, NULL, NULL); + rc = sqlite3BtreeBeginTrans(pSrc, 2); + rc = sqlite3BtreeBeginTrans(pDest, 2); + + assert( 1==sqlite3BtreeIsInTrans(pDest) ); + assert( 1==sqlite3BtreeIsInTrans(pSrc) ); + + sqlite3CodecGetKey(db, db->nDb - 1, (void**)&key, &password_sz); + sqlite3CodecAttach(db, 0, key, password_sz); + sqlite3pager_get_codec(pDest->pBt->pPager, (void**)&ctx); + + ctx->skip_read_hmac = 1; + for(i=0; iskip_read_hmac = 0; + if( rc!=SQLITE_OK ) goto handle_error; + rc = sqlite3BtreeCommit(pDest); + + db->flags = saved_flags; + db->nChange = saved_nChange; + db->nTotalChange = saved_nTotalChange; + db->xTrace = saved_xTrace; + db->autoCommit = 1; + if( pDb ){ + sqlite3BtreeClose(pDb->pBt); + pDb->pBt = 0; + pDb->pSchema = 0; + } + sqlite3ResetAllSchemasOfConnection(db); + remove(migrated_db_filename); + sqlite3_free(migrated_db_filename); + } else { + CODEC_TRACE(("*** migration failure** \n\n")); + } + + } + goto exit; + + handle_error: + CODEC_TRACE(("An error occurred attempting to migrate the database\n")); + rc = SQLITE_ERROR; + + exit: + return rc; +} + + +#endif +/* END SQLCIPHER */ + +/************** End of crypto_impl.c *****************************************/ +/************** Begin file crypto_libtomcrypt.c ******************************/ +/* +** SQLCipher +** http://sqlcipher.net +** +** Copyright (c) 2008 - 2013, ZETETIC LLC +** 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 the ZETETIC LLC 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 ZETETIC LLC ''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 ZETETIC LLC 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. +** +*/ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +#ifdef SQLCIPHER_CRYPTO_LIBTOMCRYPT +#include + +typedef struct { + prng_state prng; +} ltc_ctx; + +static unsigned int ltc_init = 0; + +static int sqlcipher_ltc_add_random(void *ctx, void *buffer, int length) { + ltc_ctx *ltc = (ltc_ctx*)ctx; + int rc = fortuna_add_entropy(buffer, length, &(ltc->prng)); + return rc != CRYPT_OK ? SQLITE_ERROR : SQLITE_OK; +} + +static int sqlcipher_ltc_activate(void *ctx) { + ltc_ctx *ltc = (ltc_ctx*)ctx; + int random_buffer_sz = sizeof(char) * 32; + unsigned char *random_buffer = sqlcipher_malloc(random_buffer_sz); + sqlcipher_memset(random_buffer, 0, random_buffer_sz); + + if(ltc_init == 0) { + if(register_prng(&fortuna_desc) != CRYPT_OK) return SQLITE_ERROR; + if(register_cipher(&rijndael_desc) != CRYPT_OK) return SQLITE_ERROR; + if(register_hash(&sha1_desc) != CRYPT_OK) return SQLITE_ERROR; + ltc_init = 1; + } + if(fortuna_start(&(ltc->prng)) != CRYPT_OK) { + return SQLITE_ERROR; + } + sqlite3_randomness(random_buffer_sz, random_buffer); + if(sqlcipher_ltc_add_random(ctx, random_buffer, random_buffer_sz) != SQLITE_OK) { + return SQLITE_ERROR; + } + if(sqlcipher_ltc_add_random(ctx, <c, sizeof(ltc_ctx*)) != SQLITE_OK) { + return SQLITE_ERROR; + } + if(fortuna_ready(&(ltc->prng)) != CRYPT_OK) { + return SQLITE_ERROR; + } + sqlcipher_free(random_buffer, random_buffer_sz); + return SQLITE_OK; +} + +static int sqlcipher_ltc_deactivate(void *ctx) { + ltc_ctx *ltc = (ltc_ctx*)ctx; + fortuna_done(&(ltc->prng)); + return SQLITE_OK; +} + +static const char* sqlcipher_ltc_get_provider_name(void *ctx) { + return "libtomcrypt"; +} + +static int sqlcipher_ltc_random(void *ctx, void *buffer, int length) { + ltc_ctx *ltc = (ltc_ctx*)ctx; + int rc; + + if((rc = fortuna_ready(&(ltc->prng))) != CRYPT_OK) { + return SQLITE_ERROR; + } + fortuna_read(buffer, length, &(ltc->prng)); + return SQLITE_OK; +} + +static int sqlcipher_ltc_hmac(void *ctx, unsigned char *hmac_key, int key_sz, unsigned char *in, int in_sz, unsigned char *in2, int in2_sz, unsigned char *out) { + int rc, hash_idx; + hmac_state hmac; + unsigned long outlen = key_sz; + + hash_idx = find_hash("sha1"); + if((rc = hmac_init(&hmac, hash_idx, hmac_key, key_sz)) != CRYPT_OK) return SQLITE_ERROR; + if((rc = hmac_process(&hmac, in, in_sz)) != CRYPT_OK) return SQLITE_ERROR; + if((rc = hmac_process(&hmac, in2, in2_sz)) != CRYPT_OK) return SQLITE_ERROR; + if((rc = hmac_done(&hmac, out, &outlen)) != CRYPT_OK) return SQLITE_ERROR; + return SQLITE_OK; +} + +static int sqlcipher_ltc_kdf(void *ctx, const unsigned char *pass, int pass_sz, unsigned char* salt, int salt_sz, int workfactor, int key_sz, unsigned char *key) { + int rc, hash_idx; + ltc_ctx *ltc = (ltc_ctx*)ctx; + unsigned long outlen = key_sz; + unsigned long random_buffer_sz = sizeof(char) * 256; + unsigned char *random_buffer = sqlcipher_malloc(random_buffer_sz); + sqlcipher_memset(random_buffer, 0, random_buffer_sz); + + hash_idx = find_hash("sha1"); + if((rc = pkcs_5_alg2(pass, pass_sz, salt, salt_sz, + workfactor, hash_idx, key, &outlen)) != CRYPT_OK) { + return SQLITE_ERROR; + } + if((rc = pkcs_5_alg2(key, key_sz, salt, salt_sz, + 1, hash_idx, random_buffer, &random_buffer_sz)) != CRYPT_OK) { + return SQLITE_ERROR; + } + sqlcipher_ltc_add_random(ctx, random_buffer, random_buffer_sz); + sqlcipher_free(random_buffer, random_buffer_sz); + return SQLITE_OK; +} + +static const char* sqlcipher_ltc_get_cipher(void *ctx) { + return "rijndael"; +} + +static int sqlcipher_ltc_cipher(void *ctx, int mode, unsigned char *key, int key_sz, unsigned char *iv, unsigned char *in, int in_sz, unsigned char *out) { + int rc, cipher_idx; + symmetric_CBC cbc; + + if((cipher_idx = find_cipher(sqlcipher_ltc_get_cipher(ctx))) == -1) return SQLITE_ERROR; + if((rc = cbc_start(cipher_idx, iv, key, key_sz, 0, &cbc)) != CRYPT_OK) return SQLITE_ERROR; + rc = mode == 1 ? cbc_encrypt(in, out, in_sz, &cbc) : cbc_decrypt(in, out, in_sz, &cbc); + if(rc != CRYPT_OK) return SQLITE_ERROR; + cbc_done(&cbc); + return SQLITE_OK; +} + +static int sqlcipher_ltc_set_cipher(void *ctx, const char *cipher_name) { + return SQLITE_OK; +} + +static int sqlcipher_ltc_get_key_sz(void *ctx) { + int cipher_idx = find_cipher(sqlcipher_ltc_get_cipher(ctx)); + return cipher_descriptor[cipher_idx].max_key_length; +} + +static int sqlcipher_ltc_get_iv_sz(void *ctx) { + int cipher_idx = find_cipher(sqlcipher_ltc_get_cipher(ctx)); + return cipher_descriptor[cipher_idx].block_length; +} + +static int sqlcipher_ltc_get_block_sz(void *ctx) { + int cipher_idx = find_cipher(sqlcipher_ltc_get_cipher(ctx)); + return cipher_descriptor[cipher_idx].block_length; +} + +static int sqlcipher_ltc_get_hmac_sz(void *ctx) { + int hash_idx = find_hash("sha1"); + return hash_descriptor[hash_idx].hashsize; +} + +static int sqlcipher_ltc_ctx_copy(void *target_ctx, void *source_ctx) { + memcpy(target_ctx, source_ctx, sizeof(ltc_ctx)); + return SQLITE_OK; +} + +static int sqlcipher_ltc_ctx_cmp(void *c1, void *c2) { + return 1; +} + +static int sqlcipher_ltc_ctx_init(void **ctx) { + *ctx = sqlcipher_malloc(sizeof(ltc_ctx)); + if(*ctx == NULL) return SQLITE_NOMEM; + sqlcipher_ltc_activate(*ctx); + return SQLITE_OK; +} + +static int sqlcipher_ltc_ctx_free(void **ctx) { + sqlcipher_ltc_deactivate(&ctx); + sqlcipher_free(*ctx, sizeof(ltc_ctx)); + return SQLITE_OK; +} + +int sqlcipher_ltc_setup(sqlcipher_provider *p) { + p->activate = sqlcipher_ltc_activate; + p->deactivate = sqlcipher_ltc_deactivate; + p->get_provider_name = sqlcipher_ltc_get_provider_name; + p->random = sqlcipher_ltc_random; + p->hmac = sqlcipher_ltc_hmac; + p->kdf = sqlcipher_ltc_kdf; + p->cipher = sqlcipher_ltc_cipher; + p->set_cipher = sqlcipher_ltc_set_cipher; + p->get_cipher = sqlcipher_ltc_get_cipher; + p->get_key_sz = sqlcipher_ltc_get_key_sz; + p->get_iv_sz = sqlcipher_ltc_get_iv_sz; + p->get_block_sz = sqlcipher_ltc_get_block_sz; + p->get_hmac_sz = sqlcipher_ltc_get_hmac_sz; + p->ctx_copy = sqlcipher_ltc_ctx_copy; + p->ctx_cmp = sqlcipher_ltc_ctx_cmp; + p->ctx_init = sqlcipher_ltc_ctx_init; + p->ctx_free = sqlcipher_ltc_ctx_free; + p->add_random = sqlcipher_ltc_add_random; + return SQLITE_OK; +} + +#endif +#endif +/* END SQLCIPHER */ + +/************** End of crypto_libtomcrypt.c **********************************/ +/************** Begin file crypto_openssl.c **********************************/ +/* +** SQLCipher +** http://sqlcipher.net +** +** Copyright (c) 2008 - 2013, ZETETIC LLC +** 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 the ZETETIC LLC 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 ZETETIC LLC ''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 ZETETIC LLC 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. +** +*/ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +#ifdef SQLCIPHER_CRYPTO_OPENSSL +#include +#include +#include + +typedef struct { + EVP_CIPHER *evp_cipher; +} openssl_ctx; + + +static unsigned int openssl_external_init = 0; +static unsigned int openssl_init_count = 0; +static sqlite3_mutex* openssl_rand_mutex = NULL; + +static int sqlcipher_openssl_add_random(void *ctx, void *buffer, int length) { +#ifndef SQLCIPHER_OPENSSL_NO_MUTEX_RAND + sqlite3_mutex_enter(openssl_rand_mutex); +#endif + RAND_add(buffer, length, 0); +#ifndef SQLCIPHER_OPENSSL_NO_MUTEX_RAND + sqlite3_mutex_leave(openssl_rand_mutex); +#endif + return SQLITE_OK; +} + +/* activate and initialize sqlcipher. Most importantly, this will automatically + intialize OpenSSL's EVP system if it hasn't already be externally. Note that + this function may be called multiple times as new codecs are intiialized. + Thus it performs some basic counting to ensure that only the last and final + sqlcipher_openssl_deactivate() will free the EVP structures. +*/ +static int sqlcipher_openssl_activate(void *ctx) { + /* initialize openssl and increment the internal init counter + but only if it hasn't been initalized outside of SQLCipher by this program + e.g. on startup */ + sqlite3_mutex_enter(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); + + if(openssl_init_count == 0 && EVP_get_cipherbyname(CIPHER) != NULL) { + /* if openssl has not yet been initialized by this library, but + a call to get_cipherbyname works, then the openssl library + has been initialized externally already. */ + openssl_external_init = 1; + } + + if(openssl_init_count == 0 && openssl_external_init == 0) { + /* if the library was not externally initialized, then should be now */ + OpenSSL_add_all_algorithms(); + } + +#ifndef SQLCIPHER_OPENSSL_NO_MUTEX_RAND + if(openssl_rand_mutex == NULL) { + /* allocate a mutex to guard against concurrent calls to RAND_bytes() */ + openssl_rand_mutex = sqlite3_mutex_alloc(SQLITE_MUTEX_FAST); + } +#endif + + openssl_init_count++; + sqlite3_mutex_leave(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); + return SQLITE_OK; +} + +/* deactivate SQLCipher, most imporantly decremeting the activation count and + freeing the EVP structures on the final deactivation to ensure that + OpenSSL memory is cleaned up */ +static int sqlcipher_openssl_deactivate(void *ctx) { + sqlite3_mutex_enter(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); + openssl_init_count--; + + if(openssl_init_count == 0) { + if(openssl_external_init == 0) { + /* if OpenSSL hasn't be initialized externally, and the counter reaches zero + after it's decremented, release EVP memory + Note: this code will only be reached if OpensSSL_add_all_algorithms() + is called by SQLCipher internally. This should prevent SQLCipher from + "cleaning up" openssl when it was initialized externally by the program */ + EVP_cleanup(); + } +#ifndef SQLCIPHER_OPENSSL_NO_MUTEX_RAND + sqlite3_mutex_free(openssl_rand_mutex); + openssl_rand_mutex = NULL; +#endif + } + sqlite3_mutex_leave(sqlite3_mutex_alloc(SQLITE_MUTEX_STATIC_MASTER)); + return SQLITE_OK; +} + +static const char* sqlcipher_openssl_get_provider_name(void *ctx) { + return "openssl"; +} + +/* generate a defined number of random bytes */ +static int sqlcipher_openssl_random (void *ctx, void *buffer, int length) { + int rc = 0; + /* concurrent calls to RAND_bytes can cause a crash under some openssl versions when a + naive application doesn't use CRYPTO_set_locking_callback and + CRYPTO_THREADID_set_callback to ensure openssl thread safety. + This is simple workaround to prevent this common crash + but a more proper solution is that applications setup platform-appropriate + thread saftey in openssl externally */ +#ifndef SQLCIPHER_OPENSSL_NO_MUTEX_RAND + sqlite3_mutex_enter(openssl_rand_mutex); +#endif + rc = RAND_bytes((unsigned char *)buffer, length); +#ifndef SQLCIPHER_OPENSSL_NO_MUTEX_RAND + sqlite3_mutex_leave(openssl_rand_mutex); +#endif + return (rc == 1) ? SQLITE_OK : SQLITE_ERROR; +} + +static int sqlcipher_openssl_hmac(void *ctx, unsigned char *hmac_key, int key_sz, unsigned char *in, int in_sz, unsigned char *in2, int in2_sz, unsigned char *out) { + HMAC_CTX hctx; + unsigned int outlen; + HMAC_CTX_init(&hctx); + HMAC_Init_ex(&hctx, hmac_key, key_sz, EVP_sha1(), NULL); + HMAC_Update(&hctx, in, in_sz); + HMAC_Update(&hctx, in2, in2_sz); + HMAC_Final(&hctx, out, &outlen); + HMAC_CTX_cleanup(&hctx); + return SQLITE_OK; +} + +static int sqlcipher_openssl_kdf(void *ctx, const unsigned char *pass, int pass_sz, unsigned char* salt, int salt_sz, int workfactor, int key_sz, unsigned char *key) { + PKCS5_PBKDF2_HMAC_SHA1((const char *)pass, pass_sz, salt, salt_sz, workfactor, key_sz, key); + return SQLITE_OK; +} + +static int sqlcipher_openssl_cipher(void *ctx, int mode, unsigned char *key, int key_sz, unsigned char *iv, unsigned char *in, int in_sz, unsigned char *out) { + EVP_CIPHER_CTX ectx; + int tmp_csz, csz; + + EVP_CipherInit(&ectx, ((openssl_ctx *)ctx)->evp_cipher, NULL, NULL, mode); + EVP_CIPHER_CTX_set_padding(&ectx, 0); // no padding + EVP_CipherInit(&ectx, NULL, key, iv, mode); + EVP_CipherUpdate(&ectx, out, &tmp_csz, in, in_sz); + csz = tmp_csz; + out += tmp_csz; + EVP_CipherFinal(&ectx, out, &tmp_csz); + csz += tmp_csz; + EVP_CIPHER_CTX_cleanup(&ectx); + assert(in_sz == csz); + return SQLITE_OK; +} + +static int sqlcipher_openssl_set_cipher(void *ctx, const char *cipher_name) { + openssl_ctx *o_ctx = (openssl_ctx *)ctx; + o_ctx->evp_cipher = (EVP_CIPHER *) EVP_get_cipherbyname(cipher_name); + return SQLITE_OK; +} + +static const char* sqlcipher_openssl_get_cipher(void *ctx) { + return EVP_CIPHER_name(((openssl_ctx *)ctx)->evp_cipher); +} + +static int sqlcipher_openssl_get_key_sz(void *ctx) { + return EVP_CIPHER_key_length(((openssl_ctx *)ctx)->evp_cipher); +} + +static int sqlcipher_openssl_get_iv_sz(void *ctx) { + return EVP_CIPHER_iv_length(((openssl_ctx *)ctx)->evp_cipher); +} + +static int sqlcipher_openssl_get_block_sz(void *ctx) { + return EVP_CIPHER_block_size(((openssl_ctx *)ctx)->evp_cipher); +} + +static int sqlcipher_openssl_get_hmac_sz(void *ctx) { + return EVP_MD_size(EVP_sha1()); +} + +static int sqlcipher_openssl_ctx_copy(void *target_ctx, void *source_ctx) { + memcpy(target_ctx, source_ctx, sizeof(openssl_ctx)); + return SQLITE_OK; +} + +static int sqlcipher_openssl_ctx_cmp(void *c1, void *c2) { + return ((openssl_ctx *)c1)->evp_cipher == ((openssl_ctx *)c2)->evp_cipher; +} + +static int sqlcipher_openssl_ctx_init(void **ctx) { + *ctx = sqlcipher_malloc(sizeof(openssl_ctx)); + if(*ctx == NULL) return SQLITE_NOMEM; + sqlcipher_openssl_activate(*ctx); + return SQLITE_OK; +} + +static int sqlcipher_openssl_ctx_free(void **ctx) { + sqlcipher_openssl_deactivate(*ctx); + sqlcipher_free(*ctx, sizeof(openssl_ctx)); + return SQLITE_OK; +} + +int sqlcipher_openssl_setup(sqlcipher_provider *p) { + p->activate = sqlcipher_openssl_activate; + p->deactivate = sqlcipher_openssl_deactivate; + p->get_provider_name = sqlcipher_openssl_get_provider_name; + p->random = sqlcipher_openssl_random; + p->hmac = sqlcipher_openssl_hmac; + p->kdf = sqlcipher_openssl_kdf; + p->cipher = sqlcipher_openssl_cipher; + p->set_cipher = sqlcipher_openssl_set_cipher; + p->get_cipher = sqlcipher_openssl_get_cipher; + p->get_key_sz = sqlcipher_openssl_get_key_sz; + p->get_iv_sz = sqlcipher_openssl_get_iv_sz; + p->get_block_sz = sqlcipher_openssl_get_block_sz; + p->get_hmac_sz = sqlcipher_openssl_get_hmac_sz; + p->ctx_copy = sqlcipher_openssl_ctx_copy; + p->ctx_cmp = sqlcipher_openssl_ctx_cmp; + p->ctx_init = sqlcipher_openssl_ctx_init; + p->ctx_free = sqlcipher_openssl_ctx_free; + p->add_random = sqlcipher_openssl_add_random; + return SQLITE_OK; +} + +#endif +#endif +/* END SQLCIPHER */ + +/************** End of crypto_openssl.c **************************************/ +/************** Begin file crypto_cc.c ***************************************/ +/* +** SQLCipher +** http://sqlcipher.net +** +** Copyright (c) 2008 - 2013, ZETETIC LLC +** 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 the ZETETIC LLC 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 ZETETIC LLC ''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 ZETETIC LLC 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. +** +*/ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +#ifdef SQLCIPHER_CRYPTO_CC +#include +#include + +/* generate a defined number of random bytes */ +static int sqlcipher_cc_random (void *ctx, void *buffer, int length) { + return (SecRandomCopyBytes(kSecRandomDefault, length, (uint8_t *)buffer) == 0) ? SQLITE_OK : SQLITE_ERROR; +} + +static const char* sqlcipher_cc_get_provider_name(void *ctx) { + return "commoncrypto"; +} + +static int sqlcipher_cc_hmac(void *ctx, unsigned char *hmac_key, int key_sz, unsigned char *in, int in_sz, unsigned char *in2, int in2_sz, unsigned char *out) { + CCHmacContext hmac_context; + CCHmacInit(&hmac_context, kCCHmacAlgSHA1, hmac_key, key_sz); + CCHmacUpdate(&hmac_context, in, in_sz); + CCHmacUpdate(&hmac_context, in2, in2_sz); + CCHmacFinal(&hmac_context, out); + return SQLITE_OK; +} + +static int sqlcipher_cc_kdf(void *ctx, const unsigned char *pass, int pass_sz, unsigned char* salt, int salt_sz, int workfactor, int key_sz, unsigned char *key) { + CCKeyDerivationPBKDF(kCCPBKDF2, (const char *)pass, pass_sz, salt, salt_sz, kCCPRFHmacAlgSHA1, workfactor, key, key_sz); + return SQLITE_OK; +} + +static int sqlcipher_cc_cipher(void *ctx, int mode, unsigned char *key, int key_sz, unsigned char *iv, unsigned char *in, int in_sz, unsigned char *out) { + CCCryptorRef cryptor; + size_t tmp_csz, csz; + CCOperation op = mode == CIPHER_ENCRYPT ? kCCEncrypt : kCCDecrypt; + + CCCryptorCreate(op, kCCAlgorithmAES128, 0, key, kCCKeySizeAES256, iv, &cryptor); + CCCryptorUpdate(cryptor, in, in_sz, out, in_sz, &tmp_csz); + csz = tmp_csz; + out += tmp_csz; + CCCryptorFinal(cryptor, out, in_sz - csz, &tmp_csz); + csz += tmp_csz; + CCCryptorRelease(cryptor); + assert(size == csz); + + return SQLITE_OK; +} + +static int sqlcipher_cc_set_cipher(void *ctx, const char *cipher_name) { + return SQLITE_OK; +} + +static const char* sqlcipher_cc_get_cipher(void *ctx) { + return "aes-256-cbc"; +} + +static int sqlcipher_cc_get_key_sz(void *ctx) { + return kCCKeySizeAES256; +} + +static int sqlcipher_cc_get_iv_sz(void *ctx) { + return kCCBlockSizeAES128; +} + +static int sqlcipher_cc_get_block_sz(void *ctx) { + return kCCBlockSizeAES128; +} + +static int sqlcipher_cc_get_hmac_sz(void *ctx) { + return CC_SHA1_DIGEST_LENGTH; +} + +static int sqlcipher_cc_ctx_copy(void *target_ctx, void *source_ctx) { + return SQLITE_OK; +} + +static int sqlcipher_cc_ctx_cmp(void *c1, void *c2) { + return SQLITE_OK; +} + +static int sqlcipher_cc_ctx_init(void **ctx) { + return SQLITE_OK; +} + +static int sqlcipher_cc_ctx_free(void **ctx) { + return SQLITE_OK; +} + +int sqlcipher_cc_setup(sqlcipher_provider *p) { + p->random = sqlcipher_cc_random; + p->get_provider_name = sqlcipher_cc_get_provider_name; + p->hmac = sqlcipher_cc_hmac; + p->kdf = sqlcipher_cc_kdf; + p->cipher = sqlcipher_cc_cipher; + p->set_cipher = sqlcipher_cc_set_cipher; + p->get_cipher = sqlcipher_cc_get_cipher; + p->get_key_sz = sqlcipher_cc_get_key_sz; + p->get_iv_sz = sqlcipher_cc_get_iv_sz; + p->get_block_sz = sqlcipher_cc_get_block_sz; + p->get_hmac_sz = sqlcipher_cc_get_hmac_sz; + p->ctx_copy = sqlcipher_cc_ctx_copy; + p->ctx_cmp = sqlcipher_cc_ctx_cmp; + p->ctx_init = sqlcipher_cc_ctx_init; + p->ctx_free = sqlcipher_cc_ctx_free; + return SQLITE_OK; +} + +#endif +#endif +/* END SQLCIPHER */ + +/************** End of crypto_cc.c *******************************************/ +/************** Begin file global.c ******************************************/ +/* +** 2008 June 13 +** +** The author disclaims copyright to this source code. In place of +** a legal notice, here is a blessing: +** +** May you do good and not evil. +** May you find forgiveness for yourself and forgive others. +** May you share freely, never taking more than you give. +** +************************************************************************* +** +** This file contains definitions of global variables and contants. +*/ + +/* An array to map all upper-case characters into their corresponding +** lower-case character. +** +** SQLite only considers US-ASCII (or EBCDIC) characters. We do not +** handle case conversions for the UTF character set since the tables +** involved are nearly as big or bigger than SQLite itself. +*/ +SQLITE_PRIVATE const unsigned char sqlite3UpperToLower[] = { +#ifdef SQLITE_ASCII + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, + 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, + 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, + 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 97, 98, 99,100,101,102,103, + 104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121, + 122, 91, 92, 93, 94, 95, 96, 97, 98, 99,100,101,102,103,104,105,106,107, + 108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125, + 126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143, + 144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161, + 162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179, + 180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197, + 198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215, + 216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232,233, + 234,235,236,237,238,239,240,241,242,243,244,245,246,247,248,249,250,251, + 252,253,254,255 +#endif +#ifdef SQLITE_EBCDIC + 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, /* 0x */ + 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, /* 1x */ + 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, /* 2x */ + 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, /* 3x */ + 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, /* 4x */ + 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, /* 5x */ + 96, 97, 66, 67, 68, 69, 70, 71, 72, 73,106,107,108,109,110,111, /* 6x */ + 112, 81, 82, 83, 84, 85, 86, 87, 88, 89,122,123,124,125,126,127, /* 7x */ + 128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143, /* 8x */ + 144,145,146,147,148,149,150,151,152,153,154,155,156,157,156,159, /* 9x */ + 160,161,162,163,164,165,166,167,168,169,170,171,140,141,142,175, /* Ax */ + 176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191, /* Bx */ + 192,129,130,131,132,133,134,135,136,137,202,203,204,205,206,207, /* Cx */ + 208,145,146,147,148,149,150,151,152,153,218,219,220,221,222,223, /* Dx */ + 224,225,162,163,164,165,166,167,168,169,232,203,204,205,206,207, /* Ex */ + 239,240,241,242,243,244,245,246,247,248,249,219,220,221,222,255, /* Fx */ +#endif +}; + +/* +** The following 256 byte lookup table is used to support SQLites built-in +** equivalents to the following standard library functions: +** +** isspace() 0x01 +** isalpha() 0x02 +** isdigit() 0x04 +** isalnum() 0x06 +** isxdigit() 0x08 +** toupper() 0x20 +** SQLite identifier character 0x40 +** +** Bit 0x20 is set if the mapped character requires translation to upper +** case. i.e. if the character is a lower-case ASCII character. +** If x is a lower-case ASCII character, then its upper-case equivalent +** is (x - 0x20). Therefore toupper() can be implemented as: +** +** (x & ~(map[x]&0x20)) +** +** Standard function tolower() is implemented using the sqlite3UpperToLower[] +** array. tolower() is used more often than toupper() by SQLite. +** +** Bit 0x40 is set if the character non-alphanumeric and can be used in an +** SQLite identifier. Identifiers are alphanumerics, "_", "$", and any +** non-ASCII UTF character. Hence the test for whether or not a character is +** part of an identifier is 0x46. +** +** SQLite's versions are identical to the standard versions assuming a +** locale of "C". They are implemented as macros in sqliteInt.h. +*/ +#ifdef SQLITE_ASCII +SQLITE_PRIVATE const unsigned char sqlite3CtypeMap[256] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 00..07 ........ */ + 0x00, 0x01, 0x01, 0x01, 0x01, 0x01, 0x00, 0x00, /* 08..0f ........ */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 10..17 ........ */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 18..1f ........ */ + 0x01, 0x00, 0x00, 0x00, 0x40, 0x00, 0x00, 0x00, /* 20..27 !"#$%&' */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 28..2f ()*+,-./ */ + 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, /* 30..37 01234567 */ + 0x0c, 0x0c, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 38..3f 89:;<=>? */ + + 0x00, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x02, /* 40..47 @ABCDEFG */ + 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, /* 48..4f HIJKLMNO */ + 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, /* 50..57 PQRSTUVW */ + 0x02, 0x02, 0x02, 0x00, 0x00, 0x00, 0x00, 0x40, /* 58..5f XYZ[\]^_ */ + 0x00, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x22, /* 60..67 `abcdefg */ + 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, /* 68..6f hijklmno */ + 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, 0x22, /* 70..77 pqrstuvw */ + 0x22, 0x22, 0x22, 0x00, 0x00, 0x00, 0x00, 0x00, /* 78..7f xyz{|}~. */ + + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 80..87 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 88..8f ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 90..97 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* 98..9f ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* a0..a7 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* a8..af ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* b0..b7 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* b8..bf ........ */ + + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* c0..c7 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* c8..cf ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* d0..d7 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* d8..df ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* e0..e7 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* e8..ef ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, /* f0..f7 ........ */ + 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40 /* f8..ff ........ */ +}; +#endif + +#ifndef SQLITE_USE_URI +# define SQLITE_USE_URI 0 +#endif + +#ifndef SQLITE_ALLOW_COVERING_INDEX_SCAN +# define SQLITE_ALLOW_COVERING_INDEX_SCAN 1 +#endif + +/* +** The following singleton contains the global configuration for +** the SQLite library. +*/ +SQLITE_PRIVATE SQLITE_WSD struct Sqlite3Config sqlite3Config = { + SQLITE_DEFAULT_MEMSTATUS, /* bMemstat */ + 1, /* bCoreMutex */ + SQLITE_THREADSAFE==1, /* bFullMutex */ + SQLITE_USE_URI, /* bOpenUri */ + SQLITE_ALLOW_COVERING_INDEX_SCAN, /* bUseCis */ + 0x7ffffffe, /* mxStrlen */ + 128, /* szLookaside */ + 500, /* nLookaside */ + {0,0,0,0,0,0,0,0}, /* m */ + {0,0,0,0,0,0,0,0,0}, /* mutex */ + {0,0,0,0,0,0,0,0,0,0,0,0,0},/* pcache2 */ + (void*)0, /* pHeap */ + 0, /* nHeap */ + 0, 0, /* mnHeap, mxHeap */ + SQLITE_DEFAULT_MMAP_SIZE, /* szMmap */ + SQLITE_MAX_MMAP_SIZE, /* mxMmap */ + (void*)0, /* pScratch */ + 0, /* szScratch */ + 0, /* nScratch */ + (void*)0, /* pPage */ + 0, /* szPage */ + 0, /* nPage */ + 0, /* mxParserStack */ + 0, /* sharedCacheEnabled */ + /* All the rest should always be initialized to zero */ + 0, /* isInit */ + 0, /* inProgress */ + 0, /* isMutexInit */ + 0, /* isMallocInit */ + 0, /* isPCacheInit */ + 0, /* pInitMutex */ + 0, /* nRefInitMutex */ + 0, /* xLog */ + 0, /* pLogArg */ + 0, /* bLocaltimeFault */ +#ifdef SQLITE_ENABLE_SQLLOG + 0, /* xSqllog */ + 0 /* pSqllogArg */ +#endif +}; + + +/* +** Hash table for global functions - functions common to all +** database connections. After initialization, this table is +** read-only. +*/ +SQLITE_PRIVATE SQLITE_WSD FuncDefHash sqlite3GlobalFunctions; + +/* +** Constant tokens for values 0 and 1. +*/ +SQLITE_PRIVATE const Token sqlite3IntTokens[] = { + { "0", 1 }, + { "1", 1 } +}; + + +/* +** The value of the "pending" byte must be 0x40000000 (1 byte past the +** 1-gibabyte boundary) in a compatible database. SQLite never uses +** the database page that contains the pending byte. It never attempts +** to read or write that page. The pending byte page is set assign +** for use by the VFS layers as space for managing file locks. +** +** During testing, it is often desirable to move the pending byte to +** a different position in the file. This allows code that has to +** deal with the pending byte to run on files that are much smaller +** than 1 GiB. The sqlite3_test_control() interface can be used to +** move the pending byte. +** +** IMPORTANT: Changing the pending byte to any value other than +** 0x40000000 results in an incompatible database file format! +** Changing the pending byte during operating results in undefined +** and dileterious behavior. +*/ +#ifndef SQLITE_OMIT_WSD +SQLITE_PRIVATE int sqlite3PendingByte = 0x40000000; +#endif + +/* +** Properties of opcodes. The OPFLG_INITIALIZER macro is +** created by mkopcodeh.awk during compilation. Data is obtained +** from the comments following the "case OP_xxxx:" statements in +** the vdbe.c file. +*/ +SQLITE_PRIVATE const unsigned char sqlite3OpcodeProperty[] = OPFLG_INITIALIZER; + +/************** End of global.c **********************************************/ +/************** Begin file ctime.c *******************************************/ +/* +** 2010 February 23 +** +** The author disclaims copyright to this source code. In place of +** a legal notice, here is a blessing: +** +** May you do good and not evil. +** May you find forgiveness for yourself and forgive others. +** May you share freely, never taking more than you give. +** +************************************************************************* +** +** This file implements routines used to report what compile-time options +** SQLite was built with. +*/ + +#ifndef SQLITE_OMIT_COMPILEOPTION_DIAGS -#ifndef SQLITE_OMIT_COMPILEOPTION_DIAGS - /* ** An array of names of all compile-time options. This array should @@ -22968,7 +26347,7 @@ #include #include #if !defined(SQLITE_OMIT_WAL) || SQLITE_MAX_MMAP_SIZE>0 -#include +/* #include */ #endif @@ -45519,7 +48898,39 @@ } #endif -#endif /* SQLITE_OMIT_DISKIO */ +#endif /* SQLITE_OMIT_DISKIO */ + +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +SQLITE_PRIVATE void sqlite3pager_get_codec(Pager *pPager, void **ctx) { + *ctx = pPager->pCodec; +} + +SQLITE_PRIVATE int sqlite3pager_is_mj_pgno(Pager *pPager, Pgno pgno) { + return (PAGER_MJ_PGNO(pPager) == pgno) ? 1 : 0; +} + +SQLITE_PRIVATE sqlite3_file *sqlite3Pager_get_fd(Pager *pPager) { + return (isOpen(pPager->fd)) ? pPager->fd : NULL; +} + +SQLITE_PRIVATE void sqlite3pager_sqlite3PagerSetCodec( + Pager *pPager, + void *(*xCodec)(void*,void*,Pgno,int), + void (*xCodecSizeChng)(void*,int,int), + void (*xCodecFree)(void*), + void *pCodec +){ + sqlite3PagerSetCodec(pPager, xCodec, xCodecSizeChng, xCodecFree, pCodec); +} + +SQLITE_PRIVATE void sqlite3pager_sqlite3PagerSetError( Pager *pPager, int error) { + pPager->errCode = error; +} + +#endif +/* END SQLCIPHER */ + /************** End of pager.c ***********************************************/ /************** Begin file wal.c *********************************************/ @@ -46046,3267 +49457,2596 @@ if( pWal->nWiData<=iPage ){ int nByte = sizeof(u32*)*(iPage+1); volatile u32 **apNew; - apNew = (volatile u32 **)sqlite3_realloc((void *)pWal->apWiData, nByte); - if( !apNew ){ - *ppPage = 0; - return SQLITE_NOMEM; - } - memset((void*)&apNew[pWal->nWiData], 0, - sizeof(u32*)*(iPage+1-pWal->nWiData)); - pWal->apWiData = apNew; - pWal->nWiData = iPage+1; - } - - /* Request a pointer to the required page from the VFS */ - if( pWal->apWiData[iPage]==0 ){ - if( pWal->exclusiveMode==WAL_HEAPMEMORY_MODE ){ - pWal->apWiData[iPage] = (u32 volatile *)sqlite3MallocZero(WALINDEX_PGSZ); - if( !pWal->apWiData[iPage] ) rc = SQLITE_NOMEM; - }else{ - rc = sqlite3OsShmMap(pWal->pDbFd, iPage, WALINDEX_PGSZ, - pWal->writeLock, (void volatile **)&pWal->apWiData[iPage] - ); - if( rc==SQLITE_READONLY ){ - pWal->readOnly |= WAL_SHM_RDONLY; - rc = SQLITE_OK; - } - } - } - - *ppPage = pWal->apWiData[iPage]; - assert( iPage==0 || *ppPage || rc!=SQLITE_OK ); - return rc; -} - -/* -** Return a pointer to the WalCkptInfo structure in the wal-index. -*/ -static volatile WalCkptInfo *walCkptInfo(Wal *pWal){ - assert( pWal->nWiData>0 && pWal->apWiData[0] ); - return (volatile WalCkptInfo*)&(pWal->apWiData[0][sizeof(WalIndexHdr)/2]); -} - -/* -** Return a pointer to the WalIndexHdr structure in the wal-index. -*/ -static volatile WalIndexHdr *walIndexHdr(Wal *pWal){ - assert( pWal->nWiData>0 && pWal->apWiData[0] ); - return (volatile WalIndexHdr*)pWal->apWiData[0]; -} - -/* -** The argument to this macro must be of type u32. On a little-endian -** architecture, it returns the u32 value that results from interpreting -** the 4 bytes as a big-endian value. On a big-endian architecture, it -** returns the value that would be produced by intepreting the 4 bytes -** of the input value as a little-endian integer. -*/ -#define BYTESWAP32(x) ( \ - (((x)&0x000000FF)<<24) + (((x)&0x0000FF00)<<8) \ - + (((x)&0x00FF0000)>>8) + (((x)&0xFF000000)>>24) \ -) - -/* -** Generate or extend an 8 byte checksum based on the data in -** array aByte[] and the initial values of aIn[0] and aIn[1] (or -** initial values of 0 and 0 if aIn==NULL). -** -** The checksum is written back into aOut[] before returning. -** -** nByte must be a positive multiple of 8. -*/ -static void walChecksumBytes( - int nativeCksum, /* True for native byte-order, false for non-native */ - u8 *a, /* Content to be checksummed */ - int nByte, /* Bytes of content in a[]. Must be a multiple of 8. */ - const u32 *aIn, /* Initial checksum value input */ - u32 *aOut /* OUT: Final checksum value output */ -){ - u32 s1, s2; - u32 *aData = (u32 *)a; - u32 *aEnd = (u32 *)&a[nByte]; - - if( aIn ){ - s1 = aIn[0]; - s2 = aIn[1]; - }else{ - s1 = s2 = 0; - } - - assert( nByte>=8 ); - assert( (nByte&0x00000007)==0 ); - - if( nativeCksum ){ - do { - s1 += *aData++ + s2; - s2 += *aData++ + s1; - }while( aDataexclusiveMode!=WAL_HEAPMEMORY_MODE ){ - sqlite3OsShmBarrier(pWal->pDbFd); - } -} - -/* -** Write the header information in pWal->hdr into the wal-index. -** -** The checksum on pWal->hdr is updated before it is written. -*/ -static void walIndexWriteHdr(Wal *pWal){ - volatile WalIndexHdr *aHdr = walIndexHdr(pWal); - const int nCksum = offsetof(WalIndexHdr, aCksum); - - assert( pWal->writeLock ); - pWal->hdr.isInit = 1; - pWal->hdr.iVersion = WALINDEX_MAX_VERSION; - walChecksumBytes(1, (u8*)&pWal->hdr, nCksum, 0, pWal->hdr.aCksum); - memcpy((void *)&aHdr[1], (void *)&pWal->hdr, sizeof(WalIndexHdr)); - walShmBarrier(pWal); - memcpy((void *)&aHdr[0], (void *)&pWal->hdr, sizeof(WalIndexHdr)); -} - -/* -** This function encodes a single frame header and writes it to a buffer -** supplied by the caller. A frame-header is made up of a series of -** 4-byte big-endian integers, as follows: -** -** 0: Page number. -** 4: For commit records, the size of the database image in pages -** after the commit. For all other records, zero. -** 8: Salt-1 (copied from the wal-header) -** 12: Salt-2 (copied from the wal-header) -** 16: Checksum-1. -** 20: Checksum-2. -*/ -static void walEncodeFrame( - Wal *pWal, /* The write-ahead log */ - u32 iPage, /* Database page number for frame */ - u32 nTruncate, /* New db size (or 0 for non-commit frames) */ - u8 *aData, /* Pointer to page data */ - u8 *aFrame /* OUT: Write encoded frame here */ -){ - int nativeCksum; /* True for native byte-order checksums */ - u32 *aCksum = pWal->hdr.aFrameCksum; - assert( WAL_FRAME_HDRSIZE==24 ); - sqlite3Put4byte(&aFrame[0], iPage); - sqlite3Put4byte(&aFrame[4], nTruncate); - memcpy(&aFrame[8], pWal->hdr.aSalt, 8); - - nativeCksum = (pWal->hdr.bigEndCksum==SQLITE_BIGENDIAN); - walChecksumBytes(nativeCksum, aFrame, 8, aCksum, aCksum); - walChecksumBytes(nativeCksum, aData, pWal->szPage, aCksum, aCksum); - - sqlite3Put4byte(&aFrame[16], aCksum[0]); - sqlite3Put4byte(&aFrame[20], aCksum[1]); -} - -/* -** Check to see if the frame with header in aFrame[] and content -** in aData[] is valid. If it is a valid frame, fill *piPage and -** *pnTruncate and return true. Return if the frame is not valid. -*/ -static int walDecodeFrame( - Wal *pWal, /* The write-ahead log */ - u32 *piPage, /* OUT: Database page number for frame */ - u32 *pnTruncate, /* OUT: New db size (or 0 if not commit) */ - u8 *aData, /* Pointer to page data (for checksum) */ - u8 *aFrame /* Frame data */ -){ - int nativeCksum; /* True for native byte-order checksums */ - u32 *aCksum = pWal->hdr.aFrameCksum; - u32 pgno; /* Page number of the frame */ - assert( WAL_FRAME_HDRSIZE==24 ); - - /* A frame is only valid if the salt values in the frame-header - ** match the salt values in the wal-header. - */ - if( memcmp(&pWal->hdr.aSalt, &aFrame[8], 8)!=0 ){ - return 0; - } - - /* A frame is only valid if the page number is creater than zero. - */ - pgno = sqlite3Get4byte(&aFrame[0]); - if( pgno==0 ){ - return 0; - } - - /* A frame is only valid if a checksum of the WAL header, - ** all prior frams, the first 16 bytes of this frame-header, - ** and the frame-data matches the checksum in the last 8 - ** bytes of this frame-header. - */ - nativeCksum = (pWal->hdr.bigEndCksum==SQLITE_BIGENDIAN); - walChecksumBytes(nativeCksum, aFrame, 8, aCksum, aCksum); - walChecksumBytes(nativeCksum, aData, pWal->szPage, aCksum, aCksum); - if( aCksum[0]!=sqlite3Get4byte(&aFrame[16]) - || aCksum[1]!=sqlite3Get4byte(&aFrame[20]) - ){ - /* Checksum failed. */ - return 0; - } - - /* If we reach this point, the frame is valid. Return the page number - ** and the new database size. - */ - *piPage = pgno; - *pnTruncate = sqlite3Get4byte(&aFrame[4]); - return 1; -} - - -#if defined(SQLITE_TEST) && defined(SQLITE_DEBUG) -/* -** Names of locks. This routine is used to provide debugging output and is not -** a part of an ordinary build. -*/ -static const char *walLockName(int lockIdx){ - if( lockIdx==WAL_WRITE_LOCK ){ - return "WRITE-LOCK"; - }else if( lockIdx==WAL_CKPT_LOCK ){ - return "CKPT-LOCK"; - }else if( lockIdx==WAL_RECOVER_LOCK ){ - return "RECOVER-LOCK"; - }else{ - static char zName[15]; - sqlite3_snprintf(sizeof(zName), zName, "READ-LOCK[%d]", - lockIdx-WAL_READ_LOCK(0)); - return zName; - } -} -#endif /*defined(SQLITE_TEST) || defined(SQLITE_DEBUG) */ - - -/* -** Set or release locks on the WAL. Locks are either shared or exclusive. -** A lock cannot be moved directly between shared and exclusive - it must go -** through the unlocked state first. -** -** In locking_mode=EXCLUSIVE, all of these routines become no-ops. -*/ -static int walLockShared(Wal *pWal, int lockIdx){ - int rc; - if( pWal->exclusiveMode ) return SQLITE_OK; - rc = sqlite3OsShmLock(pWal->pDbFd, lockIdx, 1, - SQLITE_SHM_LOCK | SQLITE_SHM_SHARED); - WALTRACE(("WAL%p: acquire SHARED-%s %s\n", pWal, - walLockName(lockIdx), rc ? "failed" : "ok")); - VVA_ONLY( pWal->lockError = (u8)(rc!=SQLITE_OK && rc!=SQLITE_BUSY); ) - return rc; -} -static void walUnlockShared(Wal *pWal, int lockIdx){ - if( pWal->exclusiveMode ) return; - (void)sqlite3OsShmLock(pWal->pDbFd, lockIdx, 1, - SQLITE_SHM_UNLOCK | SQLITE_SHM_SHARED); - WALTRACE(("WAL%p: release SHARED-%s\n", pWal, walLockName(lockIdx))); -} -static int walLockExclusive(Wal *pWal, int lockIdx, int n){ - int rc; - if( pWal->exclusiveMode ) return SQLITE_OK; - rc = sqlite3OsShmLock(pWal->pDbFd, lockIdx, n, - SQLITE_SHM_LOCK | SQLITE_SHM_EXCLUSIVE); - WALTRACE(("WAL%p: acquire EXCLUSIVE-%s cnt=%d %s\n", pWal, - walLockName(lockIdx), n, rc ? "failed" : "ok")); - VVA_ONLY( pWal->lockError = (u8)(rc!=SQLITE_OK && rc!=SQLITE_BUSY); ) - return rc; -} -static void walUnlockExclusive(Wal *pWal, int lockIdx, int n){ - if( pWal->exclusiveMode ) return; - (void)sqlite3OsShmLock(pWal->pDbFd, lockIdx, n, - SQLITE_SHM_UNLOCK | SQLITE_SHM_EXCLUSIVE); - WALTRACE(("WAL%p: release EXCLUSIVE-%s cnt=%d\n", pWal, - walLockName(lockIdx), n)); -} - -/* -** Compute a hash on a page number. The resulting hash value must land -** between 0 and (HASHTABLE_NSLOT-1). The walHashNext() function advances -** the hash to the next value in the event of a collision. -*/ -static int walHash(u32 iPage){ - assert( iPage>0 ); - assert( (HASHTABLE_NSLOT & (HASHTABLE_NSLOT-1))==0 ); - return (iPage*HASHTABLE_HASH_1) & (HASHTABLE_NSLOT-1); -} -static int walNextHash(int iPriorHash){ - return (iPriorHash+1)&(HASHTABLE_NSLOT-1); -} - -/* -** Return pointers to the hash table and page number array stored on -** page iHash of the wal-index. The wal-index is broken into 32KB pages -** numbered starting from 0. -** -** Set output variable *paHash to point to the start of the hash table -** in the wal-index file. Set *piZero to one less than the frame -** number of the first frame indexed by this hash table. If a -** slot in the hash table is set to N, it refers to frame number -** (*piZero+N) in the log. -** -** Finally, set *paPgno so that *paPgno[1] is the page number of the -** first frame indexed by the hash table, frame (*piZero+1). -*/ -static int walHashGet( - Wal *pWal, /* WAL handle */ - int iHash, /* Find the iHash'th table */ - volatile ht_slot **paHash, /* OUT: Pointer to hash index */ - volatile u32 **paPgno, /* OUT: Pointer to page number array */ - u32 *piZero /* OUT: Frame associated with *paPgno[0] */ -){ - int rc; /* Return code */ - volatile u32 *aPgno; - - rc = walIndexPage(pWal, iHash, &aPgno); - assert( rc==SQLITE_OK || iHash>0 ); - - if( rc==SQLITE_OK ){ - u32 iZero; - volatile ht_slot *aHash; - - aHash = (volatile ht_slot *)&aPgno[HASHTABLE_NPAGE]; - if( iHash==0 ){ - aPgno = &aPgno[WALINDEX_HDR_SIZE/sizeof(u32)]; - iZero = 0; - }else{ - iZero = HASHTABLE_NPAGE_ONE + (iHash-1)*HASHTABLE_NPAGE; - } - - *paPgno = &aPgno[-1]; - *paHash = aHash; - *piZero = iZero; - } - return rc; -} - -/* -** Return the number of the wal-index page that contains the hash-table -** and page-number array that contain entries corresponding to WAL frame -** iFrame. The wal-index is broken up into 32KB pages. Wal-index pages -** are numbered starting from 0. -*/ -static int walFramePage(u32 iFrame){ - int iHash = (iFrame+HASHTABLE_NPAGE-HASHTABLE_NPAGE_ONE-1) / HASHTABLE_NPAGE; - assert( (iHash==0 || iFrame>HASHTABLE_NPAGE_ONE) - && (iHash>=1 || iFrame<=HASHTABLE_NPAGE_ONE) - && (iHash<=1 || iFrame>(HASHTABLE_NPAGE_ONE+HASHTABLE_NPAGE)) - && (iHash>=2 || iFrame<=HASHTABLE_NPAGE_ONE+HASHTABLE_NPAGE) - && (iHash<=2 || iFrame>(HASHTABLE_NPAGE_ONE+2*HASHTABLE_NPAGE)) - ); - return iHash; -} - -/* -** Return the page number associated with frame iFrame in this WAL. -*/ -static u32 walFramePgno(Wal *pWal, u32 iFrame){ - int iHash = walFramePage(iFrame); - if( iHash==0 ){ - return pWal->apWiData[0][WALINDEX_HDR_SIZE/sizeof(u32) + iFrame - 1]; - } - return pWal->apWiData[iHash][(iFrame-1-HASHTABLE_NPAGE_ONE)%HASHTABLE_NPAGE]; -} - -/* -** Remove entries from the hash table that point to WAL slots greater -** than pWal->hdr.mxFrame. -** -** This function is called whenever pWal->hdr.mxFrame is decreased due -** to a rollback or savepoint. -** -** At most only the hash table containing pWal->hdr.mxFrame needs to be -** updated. Any later hash tables will be automatically cleared when -** pWal->hdr.mxFrame advances to the point where those hash tables are -** actually needed. -*/ -static void walCleanupHash(Wal *pWal){ - volatile ht_slot *aHash = 0; /* Pointer to hash table to clear */ - volatile u32 *aPgno = 0; /* Page number array for hash table */ - u32 iZero = 0; /* frame == (aHash[x]+iZero) */ - int iLimit = 0; /* Zero values greater than this */ - int nByte; /* Number of bytes to zero in aPgno[] */ - int i; /* Used to iterate through aHash[] */ - - assert( pWal->writeLock ); - testcase( pWal->hdr.mxFrame==HASHTABLE_NPAGE_ONE-1 ); - testcase( pWal->hdr.mxFrame==HASHTABLE_NPAGE_ONE ); - testcase( pWal->hdr.mxFrame==HASHTABLE_NPAGE_ONE+1 ); - - if( pWal->hdr.mxFrame==0 ) return; - - /* Obtain pointers to the hash-table and page-number array containing - ** the entry that corresponds to frame pWal->hdr.mxFrame. It is guaranteed - ** that the page said hash-table and array reside on is already mapped. - */ - assert( pWal->nWiData>walFramePage(pWal->hdr.mxFrame) ); - assert( pWal->apWiData[walFramePage(pWal->hdr.mxFrame)] ); - walHashGet(pWal, walFramePage(pWal->hdr.mxFrame), &aHash, &aPgno, &iZero); - - /* Zero all hash-table entries that correspond to frame numbers greater - ** than pWal->hdr.mxFrame. - */ - iLimit = pWal->hdr.mxFrame - iZero; - assert( iLimit>0 ); - for(i=0; iiLimit ){ - aHash[i] = 0; - } - } - - /* Zero the entries in the aPgno array that correspond to frames with - ** frame numbers greater than pWal->hdr.mxFrame. - */ - nByte = (int)((char *)aHash - (char *)&aPgno[iLimit+1]); - memset((void *)&aPgno[iLimit+1], 0, nByte); - -#ifdef SQLITE_ENABLE_EXPENSIVE_ASSERT - /* Verify that the every entry in the mapping region is still reachable - ** via the hash table even after the cleanup. - */ - if( iLimit ){ - int i; /* Loop counter */ - int iKey; /* Hash key */ - for(i=1; i<=iLimit; i++){ - for(iKey=walHash(aPgno[i]); aHash[iKey]; iKey=walNextHash(iKey)){ - if( aHash[iKey]==i ) break; - } - assert( aHash[iKey]==i ); - } - } -#endif /* SQLITE_ENABLE_EXPENSIVE_ASSERT */ -} - - -/* -** Set an entry in the wal-index that will map database page number -** pPage into WAL frame iFrame. -*/ -static int walIndexAppend(Wal *pWal, u32 iFrame, u32 iPage){ - int rc; /* Return code */ - u32 iZero = 0; /* One less than frame number of aPgno[1] */ - volatile u32 *aPgno = 0; /* Page number array */ - volatile ht_slot *aHash = 0; /* Hash table */ - - rc = walHashGet(pWal, walFramePage(iFrame), &aHash, &aPgno, &iZero); - - /* Assuming the wal-index file was successfully mapped, populate the - ** page number array and hash table entry. - */ - if( rc==SQLITE_OK ){ - int iKey; /* Hash table key */ - int idx; /* Value to write to hash-table slot */ - int nCollide; /* Number of hash collisions */ - - idx = iFrame - iZero; - assert( idx <= HASHTABLE_NSLOT/2 + 1 ); - - /* If this is the first entry to be added to this hash-table, zero the - ** entire hash table and aPgno[] array before proceding. - */ - if( idx==1 ){ - int nByte = (int)((u8 *)&aHash[HASHTABLE_NSLOT] - (u8 *)&aPgno[1]); - memset((void*)&aPgno[1], 0, nByte); - } - - /* If the entry in aPgno[] is already set, then the previous writer - ** must have exited unexpectedly in the middle of a transaction (after - ** writing one or more dirty pages to the WAL to free up memory). - ** Remove the remnants of that writers uncommitted transaction from - ** the hash-table before writing any new entries. - */ - if( aPgno[idx] ){ - walCleanupHash(pWal); - assert( !aPgno[idx] ); - } - - /* Write the aPgno[] array entry and the hash-table slot. */ - nCollide = idx; - for(iKey=walHash(iPage); aHash[iKey]; iKey=walNextHash(iKey)){ - if( (nCollide--)==0 ) return SQLITE_CORRUPT_BKPT; - } - aPgno[idx] = iPage; - aHash[iKey] = (ht_slot)idx; - -#ifdef SQLITE_ENABLE_EXPENSIVE_ASSERT - /* Verify that the number of entries in the hash table exactly equals - ** the number of entries in the mapping region. - */ - { - int i; /* Loop counter */ - int nEntry = 0; /* Number of entries in the hash table */ - for(i=0; iapWiData, nByte); + if( !apNew ){ + *ppPage = 0; + return SQLITE_NOMEM; } + memset((void*)&apNew[pWal->nWiData], 0, + sizeof(u32*)*(iPage+1-pWal->nWiData)); + pWal->apWiData = apNew; + pWal->nWiData = iPage+1; + } - /* Verify that the every entry in the mapping region is reachable - ** via the hash table. This turns out to be a really, really expensive - ** thing to check, so only do this occasionally - not on every - ** iteration. - */ - if( (idx&0x3ff)==0 ){ - int i; /* Loop counter */ - for(i=1; i<=idx; i++){ - for(iKey=walHash(aPgno[i]); aHash[iKey]; iKey=walNextHash(iKey)){ - if( aHash[iKey]==i ) break; - } - assert( aHash[iKey]==i ); + /* Request a pointer to the required page from the VFS */ + if( pWal->apWiData[iPage]==0 ){ + if( pWal->exclusiveMode==WAL_HEAPMEMORY_MODE ){ + pWal->apWiData[iPage] = (u32 volatile *)sqlite3MallocZero(WALINDEX_PGSZ); + if( !pWal->apWiData[iPage] ) rc = SQLITE_NOMEM; + }else{ + rc = sqlite3OsShmMap(pWal->pDbFd, iPage, WALINDEX_PGSZ, + pWal->writeLock, (void volatile **)&pWal->apWiData[iPage] + ); + if( rc==SQLITE_READONLY ){ + pWal->readOnly |= WAL_SHM_RDONLY; + rc = SQLITE_OK; } } -#endif /* SQLITE_ENABLE_EXPENSIVE_ASSERT */ } - + *ppPage = pWal->apWiData[iPage]; + assert( iPage==0 || *ppPage || rc!=SQLITE_OK ); return rc; } +/* +** Return a pointer to the WalCkptInfo structure in the wal-index. +*/ +static volatile WalCkptInfo *walCkptInfo(Wal *pWal){ + assert( pWal->nWiData>0 && pWal->apWiData[0] ); + return (volatile WalCkptInfo*)&(pWal->apWiData[0][sizeof(WalIndexHdr)/2]); +} /* -** Recover the wal-index by reading the write-ahead log file. -** -** This routine first tries to establish an exclusive lock on the -** wal-index to prevent other threads/processes from doing anything -** with the WAL or wal-index while recovery is running. The -** WAL_RECOVER_LOCK is also held so that other threads will know -** that this thread is running recovery. If unable to establish -** the necessary locks, this routine returns SQLITE_BUSY. +** Return a pointer to the WalIndexHdr structure in the wal-index. */ -static int walIndexRecover(Wal *pWal){ - int rc; /* Return Code */ - i64 nSize; /* Size of log file */ - u32 aFrameCksum[2] = {0, 0}; - int iLock; /* Lock offset to lock for checkpoint */ - int nLock; /* Number of locks to hold */ +static volatile WalIndexHdr *walIndexHdr(Wal *pWal){ + assert( pWal->nWiData>0 && pWal->apWiData[0] ); + return (volatile WalIndexHdr*)pWal->apWiData[0]; +} - /* Obtain an exclusive lock on all byte in the locking range not already - ** locked by the caller. The caller is guaranteed to have locked the - ** WAL_WRITE_LOCK byte, and may have also locked the WAL_CKPT_LOCK byte. - ** If successful, the same bytes that are locked here are unlocked before - ** this function returns. - */ - assert( pWal->ckptLock==1 || pWal->ckptLock==0 ); - assert( WAL_ALL_BUT_WRITE==WAL_WRITE_LOCK+1 ); - assert( WAL_CKPT_LOCK==WAL_ALL_BUT_WRITE ); - assert( pWal->writeLock ); - iLock = WAL_ALL_BUT_WRITE + pWal->ckptLock; - nLock = SQLITE_SHM_NLOCK - iLock; - rc = walLockExclusive(pWal, iLock, nLock); - if( rc ){ - return rc; - } - WALTRACE(("WAL%p: recovery begin...\n", pWal)); +/* +** The argument to this macro must be of type u32. On a little-endian +** architecture, it returns the u32 value that results from interpreting +** the 4 bytes as a big-endian value. On a big-endian architecture, it +** returns the value that would be produced by intepreting the 4 bytes +** of the input value as a little-endian integer. +*/ +#define BYTESWAP32(x) ( \ + (((x)&0x000000FF)<<24) + (((x)&0x0000FF00)<<8) \ + + (((x)&0x00FF0000)>>8) + (((x)&0xFF000000)>>24) \ +) - memset(&pWal->hdr, 0, sizeof(WalIndexHdr)); +/* +** Generate or extend an 8 byte checksum based on the data in +** array aByte[] and the initial values of aIn[0] and aIn[1] (or +** initial values of 0 and 0 if aIn==NULL). +** +** The checksum is written back into aOut[] before returning. +** +** nByte must be a positive multiple of 8. +*/ +static void walChecksumBytes( + int nativeCksum, /* True for native byte-order, false for non-native */ + u8 *a, /* Content to be checksummed */ + int nByte, /* Bytes of content in a[]. Must be a multiple of 8. */ + const u32 *aIn, /* Initial checksum value input */ + u32 *aOut /* OUT: Final checksum value output */ +){ + u32 s1, s2; + u32 *aData = (u32 *)a; + u32 *aEnd = (u32 *)&a[nByte]; - rc = sqlite3OsFileSize(pWal->pWalFd, &nSize); - if( rc!=SQLITE_OK ){ - goto recovery_error; + if( aIn ){ + s1 = aIn[0]; + s2 = aIn[1]; + }else{ + s1 = s2 = 0; } - if( nSize>WAL_HDRSIZE ){ - u8 aBuf[WAL_HDRSIZE]; /* Buffer to load WAL header into */ - u8 *aFrame = 0; /* Malloc'd buffer to load entire frame */ - int szFrame; /* Number of bytes in buffer aFrame[] */ - u8 *aData; /* Pointer to data part of aFrame buffer */ - int iFrame; /* Index of last frame read */ - i64 iOffset; /* Next offset to read from log file */ - int szPage; /* Page size according to the log */ - u32 magic; /* Magic value read from WAL header */ - u32 version; /* Magic value read from WAL header */ - int isValid; /* True if this frame is valid */ - - /* Read in the WAL header. */ - rc = sqlite3OsRead(pWal->pWalFd, aBuf, WAL_HDRSIZE, 0); - if( rc!=SQLITE_OK ){ - goto recovery_error; - } - - /* If the database page size is not a power of two, or is greater than - ** SQLITE_MAX_PAGE_SIZE, conclude that the WAL file contains no valid - ** data. Similarly, if the 'magic' value is invalid, ignore the whole - ** WAL file. - */ - magic = sqlite3Get4byte(&aBuf[0]); - szPage = sqlite3Get4byte(&aBuf[8]); - if( (magic&0xFFFFFFFE)!=WAL_MAGIC - || szPage&(szPage-1) - || szPage>SQLITE_MAX_PAGE_SIZE - || szPage<512 - ){ - goto finished; - } - pWal->hdr.bigEndCksum = (u8)(magic&0x00000001); - pWal->szPage = szPage; - pWal->nCkpt = sqlite3Get4byte(&aBuf[12]); - memcpy(&pWal->hdr.aSalt, &aBuf[16], 8); - - /* Verify that the WAL header checksum is correct */ - walChecksumBytes(pWal->hdr.bigEndCksum==SQLITE_BIGENDIAN, - aBuf, WAL_HDRSIZE-2*4, 0, pWal->hdr.aFrameCksum - ); - if( pWal->hdr.aFrameCksum[0]!=sqlite3Get4byte(&aBuf[24]) - || pWal->hdr.aFrameCksum[1]!=sqlite3Get4byte(&aBuf[28]) - ){ - goto finished; - } - - /* Verify that the version number on the WAL format is one that - ** are able to understand */ - version = sqlite3Get4byte(&aBuf[4]); - if( version!=WAL_MAX_VERSION ){ - rc = SQLITE_CANTOPEN_BKPT; - goto finished; - } - - /* Malloc a buffer to read frames into. */ - szFrame = szPage + WAL_FRAME_HDRSIZE; - aFrame = (u8 *)sqlite3_malloc(szFrame); - if( !aFrame ){ - rc = SQLITE_NOMEM; - goto recovery_error; - } - aData = &aFrame[WAL_FRAME_HDRSIZE]; - - /* Read all frames from the log file. */ - iFrame = 0; - for(iOffset=WAL_HDRSIZE; (iOffset+szFrame)<=nSize; iOffset+=szFrame){ - u32 pgno; /* Database page number for frame */ - u32 nTruncate; /* dbsize field from frame header */ - - /* Read and decode the next log frame. */ - iFrame++; - rc = sqlite3OsRead(pWal->pWalFd, aFrame, szFrame, iOffset); - if( rc!=SQLITE_OK ) break; - isValid = walDecodeFrame(pWal, &pgno, &nTruncate, aData, aFrame); - if( !isValid ) break; - rc = walIndexAppend(pWal, iFrame, pgno); - if( rc!=SQLITE_OK ) break; - - /* If nTruncate is non-zero, this is a commit record. */ - if( nTruncate ){ - pWal->hdr.mxFrame = iFrame; - pWal->hdr.nPage = nTruncate; - pWal->hdr.szPage = (u16)((szPage&0xff00) | (szPage>>16)); - testcase( szPage<=32768 ); - testcase( szPage>=65536 ); - aFrameCksum[0] = pWal->hdr.aFrameCksum[0]; - aFrameCksum[1] = pWal->hdr.aFrameCksum[1]; - } - } + assert( nByte>=8 ); + assert( (nByte&0x00000007)==0 ); - sqlite3_free(aFrame); + if( nativeCksum ){ + do { + s1 += *aData++ + s2; + s2 += *aData++ + s1; + }while( aDatahdr.aFrameCksum[0] = aFrameCksum[0]; - pWal->hdr.aFrameCksum[1] = aFrameCksum[1]; - walIndexWriteHdr(pWal); - - /* Reset the checkpoint-header. This is safe because this thread is - ** currently holding locks that exclude all other readers, writers and - ** checkpointers. - */ - pInfo = walCkptInfo(pWal); - pInfo->nBackfill = 0; - pInfo->aReadMark[0] = 0; - for(i=1; iaReadMark[i] = READMARK_NOT_USED; - if( pWal->hdr.mxFrame ) pInfo->aReadMark[1] = pWal->hdr.mxFrame; + aOut[0] = s1; + aOut[1] = s2; +} - /* If more than one frame was recovered from the log file, report an - ** event via sqlite3_log(). This is to help with identifying performance - ** problems caused by applications routinely shutting down without - ** checkpointing the log file. - */ - if( pWal->hdr.nPage ){ - sqlite3_log(SQLITE_NOTICE_RECOVER_WAL, - "recovered %d frames from WAL file %s", - pWal->hdr.mxFrame, pWal->zWalName - ); - } +static void walShmBarrier(Wal *pWal){ + if( pWal->exclusiveMode!=WAL_HEAPMEMORY_MODE ){ + sqlite3OsShmBarrier(pWal->pDbFd); } - -recovery_error: - WALTRACE(("WAL%p: recovery %s\n", pWal, rc ? "failed" : "ok")); - walUnlockExclusive(pWal, iLock, nLock); - return rc; } /* -** Close an open wal-index. +** Write the header information in pWal->hdr into the wal-index. +** +** The checksum on pWal->hdr is updated before it is written. */ -static void walIndexClose(Wal *pWal, int isDelete){ - if( pWal->exclusiveMode==WAL_HEAPMEMORY_MODE ){ - int i; - for(i=0; inWiData; i++){ - sqlite3_free((void *)pWal->apWiData[i]); - pWal->apWiData[i] = 0; - } - }else{ - sqlite3OsShmUnmap(pWal->pDbFd, isDelete); - } +static void walIndexWriteHdr(Wal *pWal){ + volatile WalIndexHdr *aHdr = walIndexHdr(pWal); + const int nCksum = offsetof(WalIndexHdr, aCksum); + + assert( pWal->writeLock ); + pWal->hdr.isInit = 1; + pWal->hdr.iVersion = WALINDEX_MAX_VERSION; + walChecksumBytes(1, (u8*)&pWal->hdr, nCksum, 0, pWal->hdr.aCksum); + memcpy((void *)&aHdr[1], (void *)&pWal->hdr, sizeof(WalIndexHdr)); + walShmBarrier(pWal); + memcpy((void *)&aHdr[0], (void *)&pWal->hdr, sizeof(WalIndexHdr)); } -/* -** Open a connection to the WAL file zWalName. The database file must -** already be opened on connection pDbFd. The buffer that zWalName points -** to must remain valid for the lifetime of the returned Wal* handle. -** -** A SHARED lock should be held on the database file when this function -** is called. The purpose of this SHARED lock is to prevent any other -** client from unlinking the WAL or wal-index file. If another process -** were to do this just after this client opened one of these files, the -** system would be badly broken. +/* +** This function encodes a single frame header and writes it to a buffer +** supplied by the caller. A frame-header is made up of a series of +** 4-byte big-endian integers, as follows: ** -** If the log file is successfully opened, SQLITE_OK is returned and -** *ppWal is set to point to a new WAL handle. If an error occurs, -** an SQLite error code is returned and *ppWal is left unmodified. +** 0: Page number. +** 4: For commit records, the size of the database image in pages +** after the commit. For all other records, zero. +** 8: Salt-1 (copied from the wal-header) +** 12: Salt-2 (copied from the wal-header) +** 16: Checksum-1. +** 20: Checksum-2. */ -SQLITE_PRIVATE int sqlite3WalOpen( - sqlite3_vfs *pVfs, /* vfs module to open wal and wal-index */ - sqlite3_file *pDbFd, /* The open database file */ - const char *zWalName, /* Name of the WAL file */ - int bNoShm, /* True to run in heap-memory mode */ - i64 mxWalSize, /* Truncate WAL to this size on reset */ - Wal **ppWal /* OUT: Allocated Wal handle */ +static void walEncodeFrame( + Wal *pWal, /* The write-ahead log */ + u32 iPage, /* Database page number for frame */ + u32 nTruncate, /* New db size (or 0 for non-commit frames) */ + u8 *aData, /* Pointer to page data */ + u8 *aFrame /* OUT: Write encoded frame here */ ){ - int rc; /* Return Code */ - Wal *pRet; /* Object to allocate and return */ - int flags; /* Flags passed to OsOpen() */ + int nativeCksum; /* True for native byte-order checksums */ + u32 *aCksum = pWal->hdr.aFrameCksum; + assert( WAL_FRAME_HDRSIZE==24 ); + sqlite3Put4byte(&aFrame[0], iPage); + sqlite3Put4byte(&aFrame[4], nTruncate); + memcpy(&aFrame[8], pWal->hdr.aSalt, 8); - assert( zWalName && zWalName[0] ); - assert( pDbFd ); + nativeCksum = (pWal->hdr.bigEndCksum==SQLITE_BIGENDIAN); + walChecksumBytes(nativeCksum, aFrame, 8, aCksum, aCksum); + walChecksumBytes(nativeCksum, aData, pWal->szPage, aCksum, aCksum); - /* In the amalgamation, the os_unix.c and os_win.c source files come before - ** this source file. Verify that the #defines of the locking byte offsets - ** in os_unix.c and os_win.c agree with the WALINDEX_LOCK_OFFSET value. - */ -#ifdef WIN_SHM_BASE - assert( WIN_SHM_BASE==WALINDEX_LOCK_OFFSET ); -#endif -#ifdef UNIX_SHM_BASE - assert( UNIX_SHM_BASE==WALINDEX_LOCK_OFFSET ); -#endif + sqlite3Put4byte(&aFrame[16], aCksum[0]); + sqlite3Put4byte(&aFrame[20], aCksum[1]); +} +/* +** Check to see if the frame with header in aFrame[] and content +** in aData[] is valid. If it is a valid frame, fill *piPage and +** *pnTruncate and return true. Return if the frame is not valid. +*/ +static int walDecodeFrame( + Wal *pWal, /* The write-ahead log */ + u32 *piPage, /* OUT: Database page number for frame */ + u32 *pnTruncate, /* OUT: New db size (or 0 if not commit) */ + u8 *aData, /* Pointer to page data (for checksum) */ + u8 *aFrame /* Frame data */ +){ + int nativeCksum; /* True for native byte-order checksums */ + u32 *aCksum = pWal->hdr.aFrameCksum; + u32 pgno; /* Page number of the frame */ + assert( WAL_FRAME_HDRSIZE==24 ); - /* Allocate an instance of struct Wal to return. */ - *ppWal = 0; - pRet = (Wal*)sqlite3MallocZero(sizeof(Wal) + pVfs->szOsFile); - if( !pRet ){ - return SQLITE_NOMEM; + /* A frame is only valid if the salt values in the frame-header + ** match the salt values in the wal-header. + */ + if( memcmp(&pWal->hdr.aSalt, &aFrame[8], 8)!=0 ){ + return 0; } - pRet->pVfs = pVfs; - pRet->pWalFd = (sqlite3_file *)&pRet[1]; - pRet->pDbFd = pDbFd; - pRet->readLock = -1; - pRet->mxWalSize = mxWalSize; - pRet->zWalName = zWalName; - pRet->syncHeader = 1; - pRet->padToSectorBoundary = 1; - pRet->exclusiveMode = (bNoShm ? WAL_HEAPMEMORY_MODE: WAL_NORMAL_MODE); - - /* Open file handle on the write-ahead log file. */ - flags = (SQLITE_OPEN_READWRITE|SQLITE_OPEN_CREATE|SQLITE_OPEN_WAL); - rc = sqlite3OsOpen(pVfs, zWalName, pRet->pWalFd, flags, &flags); - if( rc==SQLITE_OK && flags&SQLITE_OPEN_READONLY ){ - pRet->readOnly = WAL_RDONLY; + /* A frame is only valid if the page number is creater than zero. + */ + pgno = sqlite3Get4byte(&aFrame[0]); + if( pgno==0 ){ + return 0; } - if( rc!=SQLITE_OK ){ - walIndexClose(pRet, 0); - sqlite3OsClose(pRet->pWalFd); - sqlite3_free(pRet); - }else{ - int iDC = sqlite3OsDeviceCharacteristics(pRet->pWalFd); - if( iDC & SQLITE_IOCAP_SEQUENTIAL ){ pRet->syncHeader = 0; } - if( iDC & SQLITE_IOCAP_POWERSAFE_OVERWRITE ){ - pRet->padToSectorBoundary = 0; - } - *ppWal = pRet; - WALTRACE(("WAL%d: opened\n", pRet)); + /* A frame is only valid if a checksum of the WAL header, + ** all prior frams, the first 16 bytes of this frame-header, + ** and the frame-data matches the checksum in the last 8 + ** bytes of this frame-header. + */ + nativeCksum = (pWal->hdr.bigEndCksum==SQLITE_BIGENDIAN); + walChecksumBytes(nativeCksum, aFrame, 8, aCksum, aCksum); + walChecksumBytes(nativeCksum, aData, pWal->szPage, aCksum, aCksum); + if( aCksum[0]!=sqlite3Get4byte(&aFrame[16]) + || aCksum[1]!=sqlite3Get4byte(&aFrame[20]) + ){ + /* Checksum failed. */ + return 0; } - return rc; + + /* If we reach this point, the frame is valid. Return the page number + ** and the new database size. + */ + *piPage = pgno; + *pnTruncate = sqlite3Get4byte(&aFrame[4]); + return 1; } + +#if defined(SQLITE_TEST) && defined(SQLITE_DEBUG) /* -** Change the size to which the WAL file is trucated on each reset. +** Names of locks. This routine is used to provide debugging output and is not +** a part of an ordinary build. */ -SQLITE_PRIVATE void sqlite3WalLimit(Wal *pWal, i64 iLimit){ - if( pWal ) pWal->mxWalSize = iLimit; +static const char *walLockName(int lockIdx){ + if( lockIdx==WAL_WRITE_LOCK ){ + return "WRITE-LOCK"; + }else if( lockIdx==WAL_CKPT_LOCK ){ + return "CKPT-LOCK"; + }else if( lockIdx==WAL_RECOVER_LOCK ){ + return "RECOVER-LOCK"; + }else{ + static char zName[15]; + sqlite3_snprintf(sizeof(zName), zName, "READ-LOCK[%d]", + lockIdx-WAL_READ_LOCK(0)); + return zName; + } } +#endif /*defined(SQLITE_TEST) || defined(SQLITE_DEBUG) */ + /* -** Find the smallest page number out of all pages held in the WAL that -** has not been returned by any prior invocation of this method on the -** same WalIterator object. Write into *piFrame the frame index where -** that page was last written into the WAL. Write into *piPage the page -** number. +** Set or release locks on the WAL. Locks are either shared or exclusive. +** A lock cannot be moved directly between shared and exclusive - it must go +** through the unlocked state first. ** -** Return 0 on success. If there are no pages in the WAL with a page -** number larger than *piPage, then return 1. +** In locking_mode=EXCLUSIVE, all of these routines become no-ops. */ -static int walIteratorNext( - WalIterator *p, /* Iterator */ - u32 *piPage, /* OUT: The page number of the next page */ - u32 *piFrame /* OUT: Wal frame index of next page */ -){ - u32 iMin; /* Result pgno must be greater than iMin */ - u32 iRet = 0xFFFFFFFF; /* 0xffffffff is never a valid page number */ - int i; /* For looping through segments */ - - iMin = p->iPrior; - assert( iMin<0xffffffff ); - for(i=p->nSegment-1; i>=0; i--){ - struct WalSegment *pSegment = &p->aSegment[i]; - while( pSegment->iNextnEntry ){ - u32 iPg = pSegment->aPgno[pSegment->aIndex[pSegment->iNext]]; - if( iPg>iMin ){ - if( iPgiZero + pSegment->aIndex[pSegment->iNext]; - } - break; - } - pSegment->iNext++; - } - } - - *piPage = p->iPrior = iRet; - return (iRet==0xFFFFFFFF); +static int walLockShared(Wal *pWal, int lockIdx){ + int rc; + if( pWal->exclusiveMode ) return SQLITE_OK; + rc = sqlite3OsShmLock(pWal->pDbFd, lockIdx, 1, + SQLITE_SHM_LOCK | SQLITE_SHM_SHARED); + WALTRACE(("WAL%p: acquire SHARED-%s %s\n", pWal, + walLockName(lockIdx), rc ? "failed" : "ok")); + VVA_ONLY( pWal->lockError = (u8)(rc!=SQLITE_OK && rc!=SQLITE_BUSY); ) + return rc; +} +static void walUnlockShared(Wal *pWal, int lockIdx){ + if( pWal->exclusiveMode ) return; + (void)sqlite3OsShmLock(pWal->pDbFd, lockIdx, 1, + SQLITE_SHM_UNLOCK | SQLITE_SHM_SHARED); + WALTRACE(("WAL%p: release SHARED-%s\n", pWal, walLockName(lockIdx))); +} +static int walLockExclusive(Wal *pWal, int lockIdx, int n){ + int rc; + if( pWal->exclusiveMode ) return SQLITE_OK; + rc = sqlite3OsShmLock(pWal->pDbFd, lockIdx, n, + SQLITE_SHM_LOCK | SQLITE_SHM_EXCLUSIVE); + WALTRACE(("WAL%p: acquire EXCLUSIVE-%s cnt=%d %s\n", pWal, + walLockName(lockIdx), n, rc ? "failed" : "ok")); + VVA_ONLY( pWal->lockError = (u8)(rc!=SQLITE_OK && rc!=SQLITE_BUSY); ) + return rc; +} +static void walUnlockExclusive(Wal *pWal, int lockIdx, int n){ + if( pWal->exclusiveMode ) return; + (void)sqlite3OsShmLock(pWal->pDbFd, lockIdx, n, + SQLITE_SHM_UNLOCK | SQLITE_SHM_EXCLUSIVE); + WALTRACE(("WAL%p: release EXCLUSIVE-%s cnt=%d\n", pWal, + walLockName(lockIdx), n)); } /* -** This function merges two sorted lists into a single sorted list. -** -** aLeft[] and aRight[] are arrays of indices. The sort key is -** aContent[aLeft[]] and aContent[aRight[]]. Upon entry, the following -** is guaranteed for all J0 ); + assert( (HASHTABLE_NSLOT & (HASHTABLE_NSLOT-1))==0 ); + return (iPage*HASHTABLE_HASH_1) & (HASHTABLE_NSLOT-1); +} +static int walNextHash(int iPriorHash){ + return (iPriorHash+1)&(HASHTABLE_NSLOT-1); +} + +/* +** Return pointers to the hash table and page number array stored on +** page iHash of the wal-index. The wal-index is broken into 32KB pages +** numbered starting from 0. ** -** aLeft[X]!=aRight[Y] && aContent[aLeft[X]] == aContent[aRight[Y]] +** Set output variable *paHash to point to the start of the hash table +** in the wal-index file. Set *piZero to one less than the frame +** number of the first frame indexed by this hash table. If a +** slot in the hash table is set to N, it refers to frame number +** (*piZero+N) in the log. ** -** When that happens, omit the aLeft[X] and use the aRight[Y] index. +** Finally, set *paPgno so that *paPgno[1] is the page number of the +** first frame indexed by the hash table, frame (*piZero+1). */ -static void walMerge( - const u32 *aContent, /* Pages in wal - keys for the sort */ - ht_slot *aLeft, /* IN: Left hand input list */ - int nLeft, /* IN: Elements in array *paLeft */ - ht_slot **paRight, /* IN/OUT: Right hand input list */ - int *pnRight, /* IN/OUT: Elements in *paRight */ - ht_slot *aTmp /* Temporary buffer */ +static int walHashGet( + Wal *pWal, /* WAL handle */ + int iHash, /* Find the iHash'th table */ + volatile ht_slot **paHash, /* OUT: Pointer to hash index */ + volatile u32 **paPgno, /* OUT: Pointer to page number array */ + u32 *piZero /* OUT: Frame associated with *paPgno[0] */ ){ - int iLeft = 0; /* Current index in aLeft */ - int iRight = 0; /* Current index in aRight */ - int iOut = 0; /* Current index in output buffer */ - int nRight = *pnRight; - ht_slot *aRight = *paRight; + int rc; /* Return code */ + volatile u32 *aPgno; - assert( nLeft>0 && nRight>0 ); - while( iRight0 ); - if( (iLeft=nRight || aContent[aLeft[iLeft]]HASHTABLE_NPAGE_ONE) + && (iHash>=1 || iFrame<=HASHTABLE_NPAGE_ONE) + && (iHash<=1 || iFrame>(HASHTABLE_NPAGE_ONE+HASHTABLE_NPAGE)) + && (iHash>=2 || iFrame<=HASHTABLE_NPAGE_ONE+HASHTABLE_NPAGE) + && (iHash<=2 || iFrame>(HASHTABLE_NPAGE_ONE+2*HASHTABLE_NPAGE)) + ); + return iHash; +} - assert( iLeft>=nLeft || aContent[aLeft[iLeft]]>dbpage ); - assert( iRight>=nRight || aContent[aRight[iRight]]>dbpage ); +/* +** Return the page number associated with frame iFrame in this WAL. +*/ +static u32 walFramePgno(Wal *pWal, u32 iFrame){ + int iHash = walFramePage(iFrame); + if( iHash==0 ){ + return pWal->apWiData[0][WALINDEX_HDR_SIZE/sizeof(u32) + iFrame - 1]; } - - *paRight = aLeft; - *pnRight = iOut; - memcpy(aLeft, aTmp, sizeof(aTmp[0])*iOut); + return pWal->apWiData[iHash][(iFrame-1-HASHTABLE_NPAGE_ONE)%HASHTABLE_NPAGE]; } /* -** Sort the elements in list aList using aContent[] as the sort key. -** Remove elements with duplicate keys, preferring to keep the -** larger aList[] values. -** -** The aList[] entries are indices into aContent[]. The values in -** aList[] are to be sorted so that for all Jhdr.mxFrame. ** -** aContent[aList[X]] == aContent[aList[Y]] +** This function is called whenever pWal->hdr.mxFrame is decreased due +** to a rollback or savepoint. ** -** Keep the larger of the two values aList[X] and aList[Y] and discard -** the smaller. +** At most only the hash table containing pWal->hdr.mxFrame needs to be +** updated. Any later hash tables will be automatically cleared when +** pWal->hdr.mxFrame advances to the point where those hash tables are +** actually needed. */ -static void walMergesort( - const u32 *aContent, /* Pages in wal */ - ht_slot *aBuffer, /* Buffer of at least *pnList items to use */ - ht_slot *aList, /* IN/OUT: List to sort */ - int *pnList /* IN/OUT: Number of elements in aList[] */ -){ - struct Sublist { - int nList; /* Number of elements in aList */ - ht_slot *aList; /* Pointer to sub-list content */ - }; +static void walCleanupHash(Wal *pWal){ + volatile ht_slot *aHash = 0; /* Pointer to hash table to clear */ + volatile u32 *aPgno = 0; /* Page number array for hash table */ + u32 iZero = 0; /* frame == (aHash[x]+iZero) */ + int iLimit = 0; /* Zero values greater than this */ + int nByte; /* Number of bytes to zero in aPgno[] */ + int i; /* Used to iterate through aHash[] */ - const int nList = *pnList; /* Size of input list */ - int nMerge = 0; /* Number of elements in list aMerge */ - ht_slot *aMerge = 0; /* List to be merged */ - int iList; /* Index into input list */ - int iSub = 0; /* Index into aSub array */ - struct Sublist aSub[13]; /* Array of sub-lists */ + assert( pWal->writeLock ); + testcase( pWal->hdr.mxFrame==HASHTABLE_NPAGE_ONE-1 ); + testcase( pWal->hdr.mxFrame==HASHTABLE_NPAGE_ONE ); + testcase( pWal->hdr.mxFrame==HASHTABLE_NPAGE_ONE+1 ); - memset(aSub, 0, sizeof(aSub)); - assert( nList<=HASHTABLE_NPAGE && nList>0 ); - assert( HASHTABLE_NPAGE==(1<<(ArraySize(aSub)-1)) ); + if( pWal->hdr.mxFrame==0 ) return; - for(iList=0; iListaList && p->nList<=(1<aList==&aList[iList&~((2<aList, p->nList, &aMerge, &nMerge, aBuffer); - } - aSub[iSub].aList = aMerge; - aSub[iSub].nList = nMerge; - } + /* Obtain pointers to the hash-table and page-number array containing + ** the entry that corresponds to frame pWal->hdr.mxFrame. It is guaranteed + ** that the page said hash-table and array reside on is already mapped. + */ + assert( pWal->nWiData>walFramePage(pWal->hdr.mxFrame) ); + assert( pWal->apWiData[walFramePage(pWal->hdr.mxFrame)] ); + walHashGet(pWal, walFramePage(pWal->hdr.mxFrame), &aHash, &aPgno, &iZero); - for(iSub++; iSubnList<=(1<aList==&aList[nList&~((2<aList, p->nList, &aMerge, &nMerge, aBuffer); + /* Zero all hash-table entries that correspond to frame numbers greater + ** than pWal->hdr.mxFrame. + */ + iLimit = pWal->hdr.mxFrame - iZero; + assert( iLimit>0 ); + for(i=0; iiLimit ){ + aHash[i] = 0; } } - assert( aMerge==aList ); - *pnList = nMerge; + + /* Zero the entries in the aPgno array that correspond to frames with + ** frame numbers greater than pWal->hdr.mxFrame. + */ + nByte = (int)((char *)aHash - (char *)&aPgno[iLimit+1]); + memset((void *)&aPgno[iLimit+1], 0, nByte); -#ifdef SQLITE_DEBUG - { - int i; - for(i=1; i<*pnList; i++){ - assert( aContent[aList[i]] > aContent[aList[i-1]] ); +#ifdef SQLITE_ENABLE_EXPENSIVE_ASSERT + /* Verify that the every entry in the mapping region is still reachable + ** via the hash table even after the cleanup. + */ + if( iLimit ){ + int i; /* Loop counter */ + int iKey; /* Hash key */ + for(i=1; i<=iLimit; i++){ + for(iKey=walHash(aPgno[i]); aHash[iKey]; iKey=walNextHash(iKey)){ + if( aHash[iKey]==i ) break; + } + assert( aHash[iKey]==i ); } } -#endif +#endif /* SQLITE_ENABLE_EXPENSIVE_ASSERT */ } -/* -** Free an iterator allocated by walIteratorInit(). -*/ -static void walIteratorFree(WalIterator *p){ - sqlite3ScratchFree(p); -} /* -** Construct a WalInterator object that can be used to loop over all -** pages in the WAL in ascending order. The caller must hold the checkpoint -** lock. -** -** On success, make *pp point to the newly allocated WalInterator object -** return SQLITE_OK. Otherwise, return an error code. If this routine -** returns an error, the value of *pp is undefined. -** -** The calling routine should invoke walIteratorFree() to destroy the -** WalIterator object when it has finished with it. +** Set an entry in the wal-index that will map database page number +** pPage into WAL frame iFrame. */ -static int walIteratorInit(Wal *pWal, WalIterator **pp){ - WalIterator *p; /* Return value */ - int nSegment; /* Number of segments to merge */ - u32 iLast; /* Last frame in log */ - int nByte; /* Number of bytes to allocate */ - int i; /* Iterator variable */ - ht_slot *aTmp; /* Temp space used by merge-sort */ - int rc = SQLITE_OK; /* Return Code */ +static int walIndexAppend(Wal *pWal, u32 iFrame, u32 iPage){ + int rc; /* Return code */ + u32 iZero = 0; /* One less than frame number of aPgno[1] */ + volatile u32 *aPgno = 0; /* Page number array */ + volatile ht_slot *aHash = 0; /* Hash table */ - /* This routine only runs while holding the checkpoint lock. And - ** it only runs if there is actually content in the log (mxFrame>0). + rc = walHashGet(pWal, walFramePage(iFrame), &aHash, &aPgno, &iZero); + + /* Assuming the wal-index file was successfully mapped, populate the + ** page number array and hash table entry. */ - assert( pWal->ckptLock && pWal->hdr.mxFrame>0 ); - iLast = pWal->hdr.mxFrame; + if( rc==SQLITE_OK ){ + int iKey; /* Hash table key */ + int idx; /* Value to write to hash-table slot */ + int nCollide; /* Number of hash collisions */ - /* Allocate space for the WalIterator object. */ - nSegment = walFramePage(iLast) + 1; - nByte = sizeof(WalIterator) - + (nSegment-1)*sizeof(struct WalSegment) - + iLast*sizeof(ht_slot); - p = (WalIterator *)sqlite3ScratchMalloc(nByte); - if( !p ){ - return SQLITE_NOMEM; - } - memset(p, 0, nByte); - p->nSegment = nSegment; + idx = iFrame - iZero; + assert( idx <= HASHTABLE_NSLOT/2 + 1 ); + + /* If this is the first entry to be added to this hash-table, zero the + ** entire hash table and aPgno[] array before proceding. + */ + if( idx==1 ){ + int nByte = (int)((u8 *)&aHash[HASHTABLE_NSLOT] - (u8 *)&aPgno[1]); + memset((void*)&aPgno[1], 0, nByte); + } - /* Allocate temporary space used by the merge-sort routine. This block - ** of memory will be freed before this function returns. - */ - aTmp = (ht_slot *)sqlite3ScratchMalloc( - sizeof(ht_slot) * (iLast>HASHTABLE_NPAGE?HASHTABLE_NPAGE:iLast) - ); - if( !aTmp ){ - rc = SQLITE_NOMEM; - } + /* If the entry in aPgno[] is already set, then the previous writer + ** must have exited unexpectedly in the middle of a transaction (after + ** writing one or more dirty pages to the WAL to free up memory). + ** Remove the remnants of that writers uncommitted transaction from + ** the hash-table before writing any new entries. + */ + if( aPgno[idx] ){ + walCleanupHash(pWal); + assert( !aPgno[idx] ); + } - for(i=0; rc==SQLITE_OK && iaSegment[p->nSegment])[iZero]; - iZero++; - - for(j=0; jaSegment[i].iZero = iZero; - p->aSegment[i].nEntry = nEntry; - p->aSegment[i].aIndex = aIndex; - p->aSegment[i].aPgno = (u32 *)aPgno; } +#endif /* SQLITE_ENABLE_EXPENSIVE_ASSERT */ } - sqlite3ScratchFree(aTmp); - if( rc!=SQLITE_OK ){ - walIteratorFree(p); - } - *pp = p; - return rc; -} -/* -** Attempt to obtain the exclusive WAL lock defined by parameters lockIdx and -** n. If the attempt fails and parameter xBusy is not NULL, then it is a -** busy-handler function. Invoke it and retry the lock until either the -** lock is successfully obtained or the busy-handler returns 0. -*/ -static int walBusyLock( - Wal *pWal, /* WAL connection */ - int (*xBusy)(void*), /* Function to call when busy */ - void *pBusyArg, /* Context argument for xBusyHandler */ - int lockIdx, /* Offset of first byte to lock */ - int n /* Number of bytes to lock */ -){ - int rc; - do { - rc = walLockExclusive(pWal, lockIdx, n); - }while( xBusy && rc==SQLITE_BUSY && xBusy(pBusyArg) ); return rc; } -/* -** The cache of the wal-index header must be valid to call this function. -** Return the page-size in bytes used by the database. -*/ -static int walPagesize(Wal *pWal){ - return (pWal->hdr.szPage&0xfe00) + ((pWal->hdr.szPage&0x0001)<<16); -} /* -** Copy as much content as we can from the WAL back into the database file -** in response to an sqlite3_wal_checkpoint() request or the equivalent. -** -** The amount of information copies from WAL to database might be limited -** by active readers. This routine will never overwrite a database page -** that a concurrent reader might be using. -** -** All I/O barrier operations (a.k.a fsyncs) occur in this routine when -** SQLite is in WAL-mode in synchronous=NORMAL. That means that if -** checkpoints are always run by a background thread or background -** process, foreground threads will never block on a lengthy fsync call. -** -** Fsync is called on the WAL before writing content out of the WAL and -** into the database. This ensures that if the new content is persistent -** in the WAL and can be recovered following a power-loss or hard reset. -** -** Fsync is also called on the database file if (and only if) the entire -** WAL content is copied into the database file. This second fsync makes -** it safe to delete the WAL since the new content will persist in the -** database file. -** -** This routine uses and updates the nBackfill field of the wal-index header. -** This is the only routine tha will increase the value of nBackfill. -** (A WAL reset or recovery will revert nBackfill to zero, but not increase -** its value.) +** Recover the wal-index by reading the write-ahead log file. ** -** The caller must be holding sufficient locks to ensure that no other -** checkpoint is running (in any other thread or process) at the same -** time. +** This routine first tries to establish an exclusive lock on the +** wal-index to prevent other threads/processes from doing anything +** with the WAL or wal-index while recovery is running. The +** WAL_RECOVER_LOCK is also held so that other threads will know +** that this thread is running recovery. If unable to establish +** the necessary locks, this routine returns SQLITE_BUSY. */ -static int walCheckpoint( - Wal *pWal, /* Wal connection */ - int eMode, /* One of PASSIVE, FULL or RESTART */ - int (*xBusyCall)(void*), /* Function to call when busy */ - void *pBusyArg, /* Context argument for xBusyHandler */ - int sync_flags, /* Flags for OsSync() (or 0) */ - u8 *zBuf /* Temporary buffer to use */ -){ - int rc; /* Return code */ - int szPage; /* Database page-size */ - WalIterator *pIter = 0; /* Wal iterator context */ - u32 iDbpage = 0; /* Next database page to write */ - u32 iFrame = 0; /* Wal frame containing data for iDbpage */ - u32 mxSafeFrame; /* Max frame that can be backfilled */ - u32 mxPage; /* Max database page to write */ - int i; /* Loop counter */ - volatile WalCkptInfo *pInfo; /* The checkpoint status information */ - int (*xBusy)(void*) = 0; /* Function to call when waiting for locks */ +static int walIndexRecover(Wal *pWal){ + int rc; /* Return Code */ + i64 nSize; /* Size of log file */ + u32 aFrameCksum[2] = {0, 0}; + int iLock; /* Lock offset to lock for checkpoint */ + int nLock; /* Number of locks to hold */ - szPage = walPagesize(pWal); - testcase( szPage<=32768 ); - testcase( szPage>=65536 ); - pInfo = walCkptInfo(pWal); - if( pInfo->nBackfill>=pWal->hdr.mxFrame ) return SQLITE_OK; + /* Obtain an exclusive lock on all byte in the locking range not already + ** locked by the caller. The caller is guaranteed to have locked the + ** WAL_WRITE_LOCK byte, and may have also locked the WAL_CKPT_LOCK byte. + ** If successful, the same bytes that are locked here are unlocked before + ** this function returns. + */ + assert( pWal->ckptLock==1 || pWal->ckptLock==0 ); + assert( WAL_ALL_BUT_WRITE==WAL_WRITE_LOCK+1 ); + assert( WAL_CKPT_LOCK==WAL_ALL_BUT_WRITE ); + assert( pWal->writeLock ); + iLock = WAL_ALL_BUT_WRITE + pWal->ckptLock; + nLock = SQLITE_SHM_NLOCK - iLock; + rc = walLockExclusive(pWal, iLock, nLock); + if( rc ){ + return rc; + } + WALTRACE(("WAL%p: recovery begin...\n", pWal)); - /* Allocate the iterator */ - rc = walIteratorInit(pWal, &pIter); + memset(&pWal->hdr, 0, sizeof(WalIndexHdr)); + + rc = sqlite3OsFileSize(pWal->pWalFd, &nSize); if( rc!=SQLITE_OK ){ - return rc; + goto recovery_error; } - assert( pIter ); - if( eMode!=SQLITE_CHECKPOINT_PASSIVE ) xBusy = xBusyCall; + if( nSize>WAL_HDRSIZE ){ + u8 aBuf[WAL_HDRSIZE]; /* Buffer to load WAL header into */ + u8 *aFrame = 0; /* Malloc'd buffer to load entire frame */ + int szFrame; /* Number of bytes in buffer aFrame[] */ + u8 *aData; /* Pointer to data part of aFrame buffer */ + int iFrame; /* Index of last frame read */ + i64 iOffset; /* Next offset to read from log file */ + int szPage; /* Page size according to the log */ + u32 magic; /* Magic value read from WAL header */ + u32 version; /* Magic value read from WAL header */ + int isValid; /* True if this frame is valid */ + + /* Read in the WAL header. */ + rc = sqlite3OsRead(pWal->pWalFd, aBuf, WAL_HDRSIZE, 0); + if( rc!=SQLITE_OK ){ + goto recovery_error; + } - /* Compute in mxSafeFrame the index of the last frame of the WAL that is - ** safe to write into the database. Frames beyond mxSafeFrame might - ** overwrite database pages that are in use by active readers and thus - ** cannot be backfilled from the WAL. - */ - mxSafeFrame = pWal->hdr.mxFrame; - mxPage = pWal->hdr.nPage; - for(i=1; iaReadMark[i]; - if( mxSafeFrame>y ){ - assert( y<=pWal->hdr.mxFrame ); - rc = walBusyLock(pWal, xBusy, pBusyArg, WAL_READ_LOCK(i), 1); - if( rc==SQLITE_OK ){ - pInfo->aReadMark[i] = (i==1 ? mxSafeFrame : READMARK_NOT_USED); - walUnlockExclusive(pWal, WAL_READ_LOCK(i), 1); - }else if( rc==SQLITE_BUSY ){ - mxSafeFrame = y; - xBusy = 0; - }else{ - goto walcheckpoint_out; - } + /* If the database page size is not a power of two, or is greater than + ** SQLITE_MAX_PAGE_SIZE, conclude that the WAL file contains no valid + ** data. Similarly, if the 'magic' value is invalid, ignore the whole + ** WAL file. + */ + magic = sqlite3Get4byte(&aBuf[0]); + szPage = sqlite3Get4byte(&aBuf[8]); + if( (magic&0xFFFFFFFE)!=WAL_MAGIC + || szPage&(szPage-1) + || szPage>SQLITE_MAX_PAGE_SIZE + || szPage<512 + ){ + goto finished; } - } + pWal->hdr.bigEndCksum = (u8)(magic&0x00000001); + pWal->szPage = szPage; + pWal->nCkpt = sqlite3Get4byte(&aBuf[12]); + memcpy(&pWal->hdr.aSalt, &aBuf[16], 8); - if( pInfo->nBackfillnBackfill; + /* Verify that the WAL header checksum is correct */ + walChecksumBytes(pWal->hdr.bigEndCksum==SQLITE_BIGENDIAN, + aBuf, WAL_HDRSIZE-2*4, 0, pWal->hdr.aFrameCksum + ); + if( pWal->hdr.aFrameCksum[0]!=sqlite3Get4byte(&aBuf[24]) + || pWal->hdr.aFrameCksum[1]!=sqlite3Get4byte(&aBuf[28]) + ){ + goto finished; + } - /* Sync the WAL to disk */ - if( sync_flags ){ - rc = sqlite3OsSync(pWal->pWalFd, sync_flags); + /* Verify that the version number on the WAL format is one that + ** are able to understand */ + version = sqlite3Get4byte(&aBuf[4]); + if( version!=WAL_MAX_VERSION ){ + rc = SQLITE_CANTOPEN_BKPT; + goto finished; } - /* If the database may grow as a result of this checkpoint, hint - ** about the eventual size of the db file to the VFS layer. - */ - if( rc==SQLITE_OK ){ - i64 nReq = ((i64)mxPage * szPage); - rc = sqlite3OsFileSize(pWal->pDbFd, &nSize); - if( rc==SQLITE_OK && nSizepDbFd, SQLITE_FCNTL_SIZE_HINT, &nReq); - } + /* Malloc a buffer to read frames into. */ + szFrame = szPage + WAL_FRAME_HDRSIZE; + aFrame = (u8 *)sqlite3_malloc(szFrame); + if( !aFrame ){ + rc = SQLITE_NOMEM; + goto recovery_error; } + aData = &aFrame[WAL_FRAME_HDRSIZE]; + /* Read all frames from the log file. */ + iFrame = 0; + for(iOffset=WAL_HDRSIZE; (iOffset+szFrame)<=nSize; iOffset+=szFrame){ + u32 pgno; /* Database page number for frame */ + u32 nTruncate; /* dbsize field from frame header */ - /* Iterate through the contents of the WAL, copying data to the db file. */ - while( rc==SQLITE_OK && 0==walIteratorNext(pIter, &iDbpage, &iFrame) ){ - i64 iOffset; - assert( walFramePgno(pWal, iFrame)==iDbpage ); - if( iFrame<=nBackfill || iFrame>mxSafeFrame || iDbpage>mxPage ) continue; - iOffset = walFrameOffset(iFrame, szPage) + WAL_FRAME_HDRSIZE; - /* testcase( IS_BIG_INT(iOffset) ); // requires a 4GiB WAL file */ - rc = sqlite3OsRead(pWal->pWalFd, zBuf, szPage, iOffset); + /* Read and decode the next log frame. */ + iFrame++; + rc = sqlite3OsRead(pWal->pWalFd, aFrame, szFrame, iOffset); if( rc!=SQLITE_OK ) break; - iOffset = (iDbpage-1)*(i64)szPage; - testcase( IS_BIG_INT(iOffset) ); - rc = sqlite3OsWrite(pWal->pDbFd, zBuf, szPage, iOffset); + isValid = walDecodeFrame(pWal, &pgno, &nTruncate, aData, aFrame); + if( !isValid ) break; + rc = walIndexAppend(pWal, iFrame, pgno); if( rc!=SQLITE_OK ) break; - } - /* If work was actually accomplished... */ - if( rc==SQLITE_OK ){ - if( mxSafeFrame==walIndexHdr(pWal)->mxFrame ){ - i64 szDb = pWal->hdr.nPage*(i64)szPage; - testcase( IS_BIG_INT(szDb) ); - rc = sqlite3OsTruncate(pWal->pDbFd, szDb); - if( rc==SQLITE_OK && sync_flags ){ - rc = sqlite3OsSync(pWal->pDbFd, sync_flags); - } - } - if( rc==SQLITE_OK ){ - pInfo->nBackfill = mxSafeFrame; + /* If nTruncate is non-zero, this is a commit record. */ + if( nTruncate ){ + pWal->hdr.mxFrame = iFrame; + pWal->hdr.nPage = nTruncate; + pWal->hdr.szPage = (u16)((szPage&0xff00) | (szPage>>16)); + testcase( szPage<=32768 ); + testcase( szPage>=65536 ); + aFrameCksum[0] = pWal->hdr.aFrameCksum[0]; + aFrameCksum[1] = pWal->hdr.aFrameCksum[1]; } } - /* Release the reader lock held while backfilling */ - walUnlockExclusive(pWal, WAL_READ_LOCK(0), 1); + sqlite3_free(aFrame); } - if( rc==SQLITE_BUSY ){ - /* Reset the return code so as not to report a checkpoint failure - ** just because there are active readers. */ - rc = SQLITE_OK; - } +finished: + if( rc==SQLITE_OK ){ + volatile WalCkptInfo *pInfo; + int i; + pWal->hdr.aFrameCksum[0] = aFrameCksum[0]; + pWal->hdr.aFrameCksum[1] = aFrameCksum[1]; + walIndexWriteHdr(pWal); - /* If this is an SQLITE_CHECKPOINT_RESTART operation, and the entire wal - ** file has been copied into the database file, then block until all - ** readers have finished using the wal file. This ensures that the next - ** process to write to the database restarts the wal file. - */ - if( rc==SQLITE_OK && eMode!=SQLITE_CHECKPOINT_PASSIVE ){ - assert( pWal->writeLock ); - if( pInfo->nBackfillhdr.mxFrame ){ - rc = SQLITE_BUSY; - }else if( eMode==SQLITE_CHECKPOINT_RESTART ){ - assert( mxSafeFrame==pWal->hdr.mxFrame ); - rc = walBusyLock(pWal, xBusy, pBusyArg, WAL_READ_LOCK(1), WAL_NREADER-1); - if( rc==SQLITE_OK ){ - walUnlockExclusive(pWal, WAL_READ_LOCK(1), WAL_NREADER-1); - } + /* Reset the checkpoint-header. This is safe because this thread is + ** currently holding locks that exclude all other readers, writers and + ** checkpointers. + */ + pInfo = walCkptInfo(pWal); + pInfo->nBackfill = 0; + pInfo->aReadMark[0] = 0; + for(i=1; iaReadMark[i] = READMARK_NOT_USED; + if( pWal->hdr.mxFrame ) pInfo->aReadMark[1] = pWal->hdr.mxFrame; + + /* If more than one frame was recovered from the log file, report an + ** event via sqlite3_log(). This is to help with identifying performance + ** problems caused by applications routinely shutting down without + ** checkpointing the log file. + */ + if( pWal->hdr.nPage ){ + sqlite3_log(SQLITE_NOTICE_RECOVER_WAL, + "recovered %d frames from WAL file %s", + pWal->hdr.mxFrame, pWal->zWalName + ); } } - walcheckpoint_out: - walIteratorFree(pIter); +recovery_error: + WALTRACE(("WAL%p: recovery %s\n", pWal, rc ? "failed" : "ok")); + walUnlockExclusive(pWal, iLock, nLock); return rc; } /* -** If the WAL file is currently larger than nMax bytes in size, truncate -** it to exactly nMax bytes. If an error occurs while doing so, ignore it. -*/ -static void walLimitSize(Wal *pWal, i64 nMax){ - i64 sz; - int rx; - sqlite3BeginBenignMalloc(); - rx = sqlite3OsFileSize(pWal->pWalFd, &sz); - if( rx==SQLITE_OK && (sz > nMax ) ){ - rx = sqlite3OsTruncate(pWal->pWalFd, nMax); - } - sqlite3EndBenignMalloc(); - if( rx ){ - sqlite3_log(rx, "cannot limit WAL size: %s", pWal->zWalName); - } -} - -/* -** Close a connection to a log file. +** Close an open wal-index. */ -SQLITE_PRIVATE int sqlite3WalClose( - Wal *pWal, /* Wal to close */ - int sync_flags, /* Flags to pass to OsSync() (or 0) */ - int nBuf, - u8 *zBuf /* Buffer of at least nBuf bytes */ -){ - int rc = SQLITE_OK; - if( pWal ){ - int isDelete = 0; /* True to unlink wal and wal-index files */ - - /* If an EXCLUSIVE lock can be obtained on the database file (using the - ** ordinary, rollback-mode locking methods, this guarantees that the - ** connection associated with this log file is the only connection to - ** the database. In this case checkpoint the database and unlink both - ** the wal and wal-index files. - ** - ** The EXCLUSIVE lock is not released before returning. - */ - rc = sqlite3OsLock(pWal->pDbFd, SQLITE_LOCK_EXCLUSIVE); - if( rc==SQLITE_OK ){ - if( pWal->exclusiveMode==WAL_NORMAL_MODE ){ - pWal->exclusiveMode = WAL_EXCLUSIVE_MODE; - } - rc = sqlite3WalCheckpoint( - pWal, SQLITE_CHECKPOINT_PASSIVE, 0, 0, sync_flags, nBuf, zBuf, 0, 0 - ); - if( rc==SQLITE_OK ){ - int bPersist = -1; - sqlite3OsFileControlHint( - pWal->pDbFd, SQLITE_FCNTL_PERSIST_WAL, &bPersist - ); - if( bPersist!=1 ){ - /* Try to delete the WAL file if the checkpoint completed and - ** fsyned (rc==SQLITE_OK) and if we are not in persistent-wal - ** mode (!bPersist) */ - isDelete = 1; - }else if( pWal->mxWalSize>=0 ){ - /* Try to truncate the WAL file to zero bytes if the checkpoint - ** completed and fsynced (rc==SQLITE_OK) and we are in persistent - ** WAL mode (bPersist) and if the PRAGMA journal_size_limit is a - ** non-negative value (pWal->mxWalSize>=0). Note that we truncate - ** to zero bytes as truncating to the journal_size_limit might - ** leave a corrupt WAL file on disk. */ - walLimitSize(pWal, 0); - } - } - } - - walIndexClose(pWal, isDelete); - sqlite3OsClose(pWal->pWalFd); - if( isDelete ){ - sqlite3BeginBenignMalloc(); - sqlite3OsDelete(pWal->pVfs, pWal->zWalName, 0); - sqlite3EndBenignMalloc(); +static void walIndexClose(Wal *pWal, int isDelete){ + if( pWal->exclusiveMode==WAL_HEAPMEMORY_MODE ){ + int i; + for(i=0; inWiData; i++){ + sqlite3_free((void *)pWal->apWiData[i]); + pWal->apWiData[i] = 0; } - WALTRACE(("WAL%p: closed\n", pWal)); - sqlite3_free((void *)pWal->apWiData); - sqlite3_free(pWal); + }else{ + sqlite3OsShmUnmap(pWal->pDbFd, isDelete); } - return rc; } -/* -** Try to read the wal-index header. Return 0 on success and 1 if -** there is a problem. -** -** The wal-index is in shared memory. Another thread or process might -** be writing the header at the same time this procedure is trying to -** read it, which might result in inconsistency. A dirty read is detected -** by verifying that both copies of the header are the same and also by -** a checksum on the header. +/* +** Open a connection to the WAL file zWalName. The database file must +** already be opened on connection pDbFd. The buffer that zWalName points +** to must remain valid for the lifetime of the returned Wal* handle. ** -** If and only if the read is consistent and the header is different from -** pWal->hdr, then pWal->hdr is updated to the content of the new header -** and *pChanged is set to 1. +** A SHARED lock should be held on the database file when this function +** is called. The purpose of this SHARED lock is to prevent any other +** client from unlinking the WAL or wal-index file. If another process +** were to do this just after this client opened one of these files, the +** system would be badly broken. ** -** If the checksum cannot be verified return non-zero. If the header -** is read successfully and the checksum verified, return zero. +** If the log file is successfully opened, SQLITE_OK is returned and +** *ppWal is set to point to a new WAL handle. If an error occurs, +** an SQLite error code is returned and *ppWal is left unmodified. */ -static int walIndexTryHdr(Wal *pWal, int *pChanged){ - u32 aCksum[2]; /* Checksum on the header content */ - WalIndexHdr h1, h2; /* Two copies of the header content */ - WalIndexHdr volatile *aHdr; /* Header in shared memory */ +SQLITE_PRIVATE int sqlite3WalOpen( + sqlite3_vfs *pVfs, /* vfs module to open wal and wal-index */ + sqlite3_file *pDbFd, /* The open database file */ + const char *zWalName, /* Name of the WAL file */ + int bNoShm, /* True to run in heap-memory mode */ + i64 mxWalSize, /* Truncate WAL to this size on reset */ + Wal **ppWal /* OUT: Allocated Wal handle */ +){ + int rc; /* Return Code */ + Wal *pRet; /* Object to allocate and return */ + int flags; /* Flags passed to OsOpen() */ - /* The first page of the wal-index must be mapped at this point. */ - assert( pWal->nWiData>0 && pWal->apWiData[0] ); + assert( zWalName && zWalName[0] ); + assert( pDbFd ); - /* Read the header. This might happen concurrently with a write to the - ** same area of shared memory on a different CPU in a SMP, - ** meaning it is possible that an inconsistent snapshot is read - ** from the file. If this happens, return non-zero. - ** - ** There are two copies of the header at the beginning of the wal-index. - ** When reading, read [0] first then [1]. Writes are in the reverse order. - ** Memory barriers are used to prevent the compiler or the hardware from - ** reordering the reads and writes. + /* In the amalgamation, the os_unix.c and os_win.c source files come before + ** this source file. Verify that the #defines of the locking byte offsets + ** in os_unix.c and os_win.c agree with the WALINDEX_LOCK_OFFSET value. */ - aHdr = walIndexHdr(pWal); - memcpy(&h1, (void *)&aHdr[0], sizeof(h1)); - walShmBarrier(pWal); - memcpy(&h2, (void *)&aHdr[1], sizeof(h2)); +#ifdef WIN_SHM_BASE + assert( WIN_SHM_BASE==WALINDEX_LOCK_OFFSET ); +#endif +#ifdef UNIX_SHM_BASE + assert( UNIX_SHM_BASE==WALINDEX_LOCK_OFFSET ); +#endif - if( memcmp(&h1, &h2, sizeof(h1))!=0 ){ - return 1; /* Dirty read */ - } - if( h1.isInit==0 ){ - return 1; /* Malformed header - probably all zeros */ + + /* Allocate an instance of struct Wal to return. */ + *ppWal = 0; + pRet = (Wal*)sqlite3MallocZero(sizeof(Wal) + pVfs->szOsFile); + if( !pRet ){ + return SQLITE_NOMEM; } - walChecksumBytes(1, (u8*)&h1, sizeof(h1)-sizeof(h1.aCksum), 0, aCksum); - if( aCksum[0]!=h1.aCksum[0] || aCksum[1]!=h1.aCksum[1] ){ - return 1; /* Checksum does not match */ + + pRet->pVfs = pVfs; + pRet->pWalFd = (sqlite3_file *)&pRet[1]; + pRet->pDbFd = pDbFd; + pRet->readLock = -1; + pRet->mxWalSize = mxWalSize; + pRet->zWalName = zWalName; + pRet->syncHeader = 1; + pRet->padToSectorBoundary = 1; + pRet->exclusiveMode = (bNoShm ? WAL_HEAPMEMORY_MODE: WAL_NORMAL_MODE); + + /* Open file handle on the write-ahead log file. */ + flags = (SQLITE_OPEN_READWRITE|SQLITE_OPEN_CREATE|SQLITE_OPEN_WAL); + rc = sqlite3OsOpen(pVfs, zWalName, pRet->pWalFd, flags, &flags); + if( rc==SQLITE_OK && flags&SQLITE_OPEN_READONLY ){ + pRet->readOnly = WAL_RDONLY; } - if( memcmp(&pWal->hdr, &h1, sizeof(WalIndexHdr)) ){ - *pChanged = 1; - memcpy(&pWal->hdr, &h1, sizeof(WalIndexHdr)); - pWal->szPage = (pWal->hdr.szPage&0xfe00) + ((pWal->hdr.szPage&0x0001)<<16); - testcase( pWal->szPage<=32768 ); - testcase( pWal->szPage>=65536 ); + if( rc!=SQLITE_OK ){ + walIndexClose(pRet, 0); + sqlite3OsClose(pRet->pWalFd); + sqlite3_free(pRet); + }else{ + int iDC = sqlite3OsDeviceCharacteristics(pRet->pWalFd); + if( iDC & SQLITE_IOCAP_SEQUENTIAL ){ pRet->syncHeader = 0; } + if( iDC & SQLITE_IOCAP_POWERSAFE_OVERWRITE ){ + pRet->padToSectorBoundary = 0; + } + *ppWal = pRet; + WALTRACE(("WAL%d: opened\n", pRet)); } + return rc; +} - /* The header was successfully read. Return zero. */ - return 0; +/* +** Change the size to which the WAL file is trucated on each reset. +*/ +SQLITE_PRIVATE void sqlite3WalLimit(Wal *pWal, i64 iLimit){ + if( pWal ) pWal->mxWalSize = iLimit; } /* -** Read the wal-index header from the wal-index and into pWal->hdr. -** If the wal-header appears to be corrupt, try to reconstruct the -** wal-index from the WAL before returning. -** -** Set *pChanged to 1 if the wal-index header value in pWal->hdr is -** changed by this opertion. If pWal->hdr is unchanged, set *pChanged -** to 0. +** Find the smallest page number out of all pages held in the WAL that +** has not been returned by any prior invocation of this method on the +** same WalIterator object. Write into *piFrame the frame index where +** that page was last written into the WAL. Write into *piPage the page +** number. ** -** If the wal-index header is successfully read, return SQLITE_OK. -** Otherwise an SQLite error code. +** Return 0 on success. If there are no pages in the WAL with a page +** number larger than *piPage, then return 1. */ -static int walIndexReadHdr(Wal *pWal, int *pChanged){ - int rc; /* Return code */ - int badHdr; /* True if a header read failed */ - volatile u32 *page0; /* Chunk of wal-index containing header */ - - /* Ensure that page 0 of the wal-index (the page that contains the - ** wal-index header) is mapped. Return early if an error occurs here. - */ - assert( pChanged ); - rc = walIndexPage(pWal, 0, &page0); - if( rc!=SQLITE_OK ){ - return rc; - }; - assert( page0 || pWal->writeLock==0 ); - - /* If the first page of the wal-index has been mapped, try to read the - ** wal-index header immediately, without holding any lock. This usually - ** works, but may fail if the wal-index header is corrupt or currently - ** being modified by another thread or process. - */ - badHdr = (page0 ? walIndexTryHdr(pWal, pChanged) : 1); +static int walIteratorNext( + WalIterator *p, /* Iterator */ + u32 *piPage, /* OUT: The page number of the next page */ + u32 *piFrame /* OUT: Wal frame index of next page */ +){ + u32 iMin; /* Result pgno must be greater than iMin */ + u32 iRet = 0xFFFFFFFF; /* 0xffffffff is never a valid page number */ + int i; /* For looping through segments */ - /* If the first attempt failed, it might have been due to a race - ** with a writer. So get a WRITE lock and try again. - */ - assert( badHdr==0 || pWal->writeLock==0 ); - if( badHdr ){ - if( pWal->readOnly & WAL_SHM_RDONLY ){ - if( SQLITE_OK==(rc = walLockShared(pWal, WAL_WRITE_LOCK)) ){ - walUnlockShared(pWal, WAL_WRITE_LOCK); - rc = SQLITE_READONLY_RECOVERY; - } - }else if( SQLITE_OK==(rc = walLockExclusive(pWal, WAL_WRITE_LOCK, 1)) ){ - pWal->writeLock = 1; - if( SQLITE_OK==(rc = walIndexPage(pWal, 0, &page0)) ){ - badHdr = walIndexTryHdr(pWal, pChanged); - if( badHdr ){ - /* If the wal-index header is still malformed even while holding - ** a WRITE lock, it can only mean that the header is corrupted and - ** needs to be reconstructed. So run recovery to do exactly that. - */ - rc = walIndexRecover(pWal); - *pChanged = 1; + iMin = p->iPrior; + assert( iMin<0xffffffff ); + for(i=p->nSegment-1; i>=0; i--){ + struct WalSegment *pSegment = &p->aSegment[i]; + while( pSegment->iNextnEntry ){ + u32 iPg = pSegment->aPgno[pSegment->aIndex[pSegment->iNext]]; + if( iPg>iMin ){ + if( iPgiZero + pSegment->aIndex[pSegment->iNext]; } + break; } - pWal->writeLock = 0; - walUnlockExclusive(pWal, WAL_WRITE_LOCK, 1); + pSegment->iNext++; + } + } + + *piPage = p->iPrior = iRet; + return (iRet==0xFFFFFFFF); +} + +/* +** This function merges two sorted lists into a single sorted list. +** +** aLeft[] and aRight[] are arrays of indices. The sort key is +** aContent[aLeft[]] and aContent[aRight[]]. Upon entry, the following +** is guaranteed for all J0 && nRight>0 ); + while( iRight=nRight || aContent[aLeft[iLeft]]hdr.iVersion!=WALINDEX_MAX_VERSION ){ - rc = SQLITE_CANTOPEN_BKPT; + aTmp[iOut++] = logpage; + if( iLeft=nLeft || aContent[aLeft[iLeft]]>dbpage ); + assert( iRight>=nRight || aContent[aRight[iRight]]>dbpage ); } - return rc; + *paRight = aLeft; + *pnRight = iOut; + memcpy(aLeft, aTmp, sizeof(aTmp[0])*iOut); } /* -** This is the value that walTryBeginRead returns when it needs to -** be retried. -*/ -#define WAL_RETRY (-1) - -/* -** Attempt to start a read transaction. This might fail due to a race or -** other transient condition. When that happens, it returns WAL_RETRY to -** indicate to the caller that it is safe to retry immediately. +** Sort the elements in list aList using aContent[] as the sort key. +** Remove elements with duplicate keys, preferring to keep the +** larger aList[] values. ** -** On success return SQLITE_OK. On a permanent failure (such an -** I/O error or an SQLITE_BUSY because another process is running -** recovery) return a positive error code. +** The aList[] entries are indices into aContent[]. The values in +** aList[] are to be sorted so that for all Jhdr. If the -** wal-index header has changed, *pChanged is set to 1 (as an indication -** to the caller that the local paget cache is obsolete and needs to be -** flushed.) When useWal==1, the wal-index header is assumed to already -** be loaded and the pChanged parameter is unused. +** aContent[aList[J]] < aContent[aList[K]] ** -** The caller must set the cnt parameter to the number of prior calls to -** this routine during the current read attempt that returned WAL_RETRY. -** This routine will start taking more aggressive measures to clear the -** race conditions after multiple WAL_RETRY returns, and after an excessive -** number of errors will ultimately return SQLITE_PROTOCOL. The -** SQLITE_PROTOCOL return indicates that some other process has gone rogue -** and is not honoring the locking protocol. There is a vanishingly small -** chance that SQLITE_PROTOCOL could be returned because of a run of really -** bad luck when there is lots of contention for the wal-index, but that -** possibility is so small that it can be safely neglected, we believe. +** For any X and Y such that ** -** On success, this routine obtains a read lock on -** WAL_READ_LOCK(pWal->readLock). The pWal->readLock integer is -** in the range 0 <= pWal->readLock < WAL_NREADER. If pWal->readLock==(-1) -** that means the Wal does not hold any read lock. The reader must not -** access any database page that is modified by a WAL frame up to and -** including frame number aReadMark[pWal->readLock]. The reader will -** use WAL frames up to and including pWal->hdr.mxFrame if pWal->readLock>0 -** Or if pWal->readLock==0, then the reader will ignore the WAL -** completely and get all content directly from the database file. -** If the useWal parameter is 1 then the WAL will never be ignored and -** this routine will always set pWal->readLock>0 on success. -** When the read transaction is completed, the caller must release the -** lock on WAL_READ_LOCK(pWal->readLock) and set pWal->readLock to -1. +** aContent[aList[X]] == aContent[aList[Y]] ** -** This routine uses the nBackfill and aReadMark[] fields of the header -** to select a particular WAL_READ_LOCK() that strives to let the -** checkpoint process do as much work as possible. This routine might -** update values of the aReadMark[] array in the header, but if it does -** so it takes care to hold an exclusive lock on the corresponding -** WAL_READ_LOCK() while changing values. +** Keep the larger of the two values aList[X] and aList[Y] and discard +** the smaller. */ -static int walTryBeginRead(Wal *pWal, int *pChanged, int useWal, int cnt){ - volatile WalCkptInfo *pInfo; /* Checkpoint information in wal-index */ - u32 mxReadMark; /* Largest aReadMark[] value */ - int mxI; /* Index of largest aReadMark[] value */ - int i; /* Loop counter */ - int rc = SQLITE_OK; /* Return code */ +static void walMergesort( + const u32 *aContent, /* Pages in wal */ + ht_slot *aBuffer, /* Buffer of at least *pnList items to use */ + ht_slot *aList, /* IN/OUT: List to sort */ + int *pnList /* IN/OUT: Number of elements in aList[] */ +){ + struct Sublist { + int nList; /* Number of elements in aList */ + ht_slot *aList; /* Pointer to sub-list content */ + }; - assert( pWal->readLock<0 ); /* Not currently locked */ + const int nList = *pnList; /* Size of input list */ + int nMerge = 0; /* Number of elements in list aMerge */ + ht_slot *aMerge = 0; /* List to be merged */ + int iList; /* Index into input list */ + int iSub = 0; /* Index into aSub array */ + struct Sublist aSub[13]; /* Array of sub-lists */ - /* Take steps to avoid spinning forever if there is a protocol error. - ** - ** Circumstances that cause a RETRY should only last for the briefest - ** instances of time. No I/O or other system calls are done while the - ** locks are held, so the locks should not be held for very long. But - ** if we are unlucky, another process that is holding a lock might get - ** paged out or take a page-fault that is time-consuming to resolve, - ** during the few nanoseconds that it is holding the lock. In that case, - ** it might take longer than normal for the lock to free. - ** - ** After 5 RETRYs, we begin calling sqlite3OsSleep(). The first few - ** calls to sqlite3OsSleep() have a delay of 1 microsecond. Really this - ** is more of a scheduler yield than an actual delay. But on the 10th - ** an subsequent retries, the delays start becoming longer and longer, - ** so that on the 100th (and last) RETRY we delay for 21 milliseconds. - ** The total delay time before giving up is less than 1 second. - */ - if( cnt>5 ){ - int nDelay = 1; /* Pause time in microseconds */ - if( cnt>100 ){ - VVA_ONLY( pWal->lockError = 1; ) - return SQLITE_PROTOCOL; - } - if( cnt>=10 ) nDelay = (cnt-9)*238; /* Max delay 21ms. Total delay 996ms */ - sqlite3OsSleep(pWal->pVfs, nDelay); - } + memset(aSub, 0, sizeof(aSub)); + assert( nList<=HASHTABLE_NPAGE && nList>0 ); + assert( HASHTABLE_NPAGE==(1<<(ArraySize(aSub)-1)) ); - if( !useWal ){ - rc = walIndexReadHdr(pWal, pChanged); - if( rc==SQLITE_BUSY ){ - /* If there is not a recovery running in another thread or process - ** then convert BUSY errors to WAL_RETRY. If recovery is known to - ** be running, convert BUSY to BUSY_RECOVERY. There is a race here - ** which might cause WAL_RETRY to be returned even if BUSY_RECOVERY - ** would be technically correct. But the race is benign since with - ** WAL_RETRY this routine will be called again and will probably be - ** right on the second iteration. - */ - if( pWal->apWiData[0]==0 ){ - /* This branch is taken when the xShmMap() method returns SQLITE_BUSY. - ** We assume this is a transient condition, so return WAL_RETRY. The - ** xShmMap() implementation used by the default unix and win32 VFS - ** modules may return SQLITE_BUSY due to a race condition in the - ** code that determines whether or not the shared-memory region - ** must be zeroed before the requested page is returned. - */ - rc = WAL_RETRY; - }else if( SQLITE_OK==(rc = walLockShared(pWal, WAL_RECOVER_LOCK)) ){ - walUnlockShared(pWal, WAL_RECOVER_LOCK); - rc = WAL_RETRY; - }else if( rc==SQLITE_BUSY ){ - rc = SQLITE_BUSY_RECOVERY; - } - } - if( rc!=SQLITE_OK ){ - return rc; + for(iList=0; iListaList && p->nList<=(1<aList==&aList[iList&~((2<aList, p->nList, &aMerge, &nMerge, aBuffer); } + aSub[iSub].aList = aMerge; + aSub[iSub].nList = nMerge; } - pInfo = walCkptInfo(pWal); - if( !useWal && pInfo->nBackfill==pWal->hdr.mxFrame ){ - /* The WAL has been completely backfilled (or it is empty). - ** and can be safely ignored. - */ - rc = walLockShared(pWal, WAL_READ_LOCK(0)); - walShmBarrier(pWal); - if( rc==SQLITE_OK ){ - if( memcmp((void *)walIndexHdr(pWal), &pWal->hdr, sizeof(WalIndexHdr)) ){ - /* It is not safe to allow the reader to continue here if frames - ** may have been appended to the log before READ_LOCK(0) was obtained. - ** When holding READ_LOCK(0), the reader ignores the entire log file, - ** which implies that the database file contains a trustworthy - ** snapshoT. Since holding READ_LOCK(0) prevents a checkpoint from - ** happening, this is usually correct. - ** - ** However, if frames have been appended to the log (or if the log - ** is wrapped and written for that matter) before the READ_LOCK(0) - ** is obtained, that is not necessarily true. A checkpointer may - ** have started to backfill the appended frames but crashed before - ** it finished. Leaving a corrupt image in the database file. - */ - walUnlockShared(pWal, WAL_READ_LOCK(0)); - return WAL_RETRY; - } - pWal->readLock = 0; - return SQLITE_OK; - }else if( rc!=SQLITE_BUSY ){ - return rc; + for(iSub++; iSubnList<=(1<aList==&aList[nList&~((2<aList, p->nList, &aMerge, &nMerge, aBuffer); } } + assert( aMerge==aList ); + *pnList = nMerge; - /* If we get this far, it means that the reader will want to use - ** the WAL to get at content from recent commits. The job now is - ** to select one of the aReadMark[] entries that is closest to - ** but not exceeding pWal->hdr.mxFrame and lock that entry. - */ - mxReadMark = 0; - mxI = 0; - for(i=1; iaReadMark[i]; - if( mxReadMark<=thisMark && thisMark<=pWal->hdr.mxFrame ){ - assert( thisMark!=READMARK_NOT_USED ); - mxReadMark = thisMark; - mxI = i; - } - } - /* There was once an "if" here. The extra "{" is to preserve indentation. */ +#ifdef SQLITE_DEBUG { - if( (pWal->readOnly & WAL_SHM_RDONLY)==0 - && (mxReadMarkhdr.mxFrame || mxI==0) - ){ - for(i=1; iaReadMark[i] = pWal->hdr.mxFrame; - mxI = i; - walUnlockExclusive(pWal, WAL_READ_LOCK(i), 1); - break; - }else if( rc!=SQLITE_BUSY ){ - return rc; - } - } - } - if( mxI==0 ){ - assert( rc==SQLITE_BUSY || (pWal->readOnly & WAL_SHM_RDONLY)!=0 ); - return rc==SQLITE_BUSY ? WAL_RETRY : SQLITE_READONLY_CANTLOCK; - } - - rc = walLockShared(pWal, WAL_READ_LOCK(mxI)); - if( rc ){ - return rc==SQLITE_BUSY ? WAL_RETRY : rc; - } - /* Now that the read-lock has been obtained, check that neither the - ** value in the aReadMark[] array or the contents of the wal-index - ** header have changed. - ** - ** It is necessary to check that the wal-index header did not change - ** between the time it was read and when the shared-lock was obtained - ** on WAL_READ_LOCK(mxI) was obtained to account for the possibility - ** that the log file may have been wrapped by a writer, or that frames - ** that occur later in the log than pWal->hdr.mxFrame may have been - ** copied into the database by a checkpointer. If either of these things - ** happened, then reading the database with the current value of - ** pWal->hdr.mxFrame risks reading a corrupted snapshot. So, retry - ** instead. - ** - ** This does not guarantee that the copy of the wal-index header is up to - ** date before proceeding. That would not be possible without somehow - ** blocking writers. It only guarantees that a dangerous checkpoint or - ** log-wrap (either of which would require an exclusive lock on - ** WAL_READ_LOCK(mxI)) has not occurred since the snapshot was valid. - */ - walShmBarrier(pWal); - if( pInfo->aReadMark[mxI]!=mxReadMark - || memcmp((void *)walIndexHdr(pWal), &pWal->hdr, sizeof(WalIndexHdr)) - ){ - walUnlockShared(pWal, WAL_READ_LOCK(mxI)); - return WAL_RETRY; - }else{ - assert( mxReadMark<=pWal->hdr.mxFrame ); - pWal->readLock = (i16)mxI; + int i; + for(i=1; i<*pnList; i++){ + assert( aContent[aList[i]] > aContent[aList[i-1]] ); } } - return rc; -} - -/* -** Begin a read transaction on the database. -** -** This routine used to be called sqlite3OpenSnapshot() and with good reason: -** it takes a snapshot of the state of the WAL and wal-index for the current -** instant in time. The current thread will continue to use this snapshot. -** Other threads might append new content to the WAL and wal-index but -** that extra content is ignored by the current thread. -** -** If the database contents have changes since the previous read -** transaction, then *pChanged is set to 1 before returning. The -** Pager layer will use this to know that is cache is stale and -** needs to be flushed. -*/ -SQLITE_PRIVATE int sqlite3WalBeginReadTransaction(Wal *pWal, int *pChanged){ - int rc; /* Return code */ - int cnt = 0; /* Number of TryBeginRead attempts */ - - do{ - rc = walTryBeginRead(pWal, pChanged, 0, ++cnt); - }while( rc==WAL_RETRY ); - testcase( (rc&0xff)==SQLITE_BUSY ); - testcase( (rc&0xff)==SQLITE_IOERR ); - testcase( rc==SQLITE_PROTOCOL ); - testcase( rc==SQLITE_OK ); - return rc; +#endif } -/* -** Finish with a read transaction. All this does is release the -** read-lock. +/* +** Free an iterator allocated by walIteratorInit(). */ -SQLITE_PRIVATE void sqlite3WalEndReadTransaction(Wal *pWal){ - sqlite3WalEndWriteTransaction(pWal); - if( pWal->readLock>=0 ){ - walUnlockShared(pWal, WAL_READ_LOCK(pWal->readLock)); - pWal->readLock = -1; - } +static void walIteratorFree(WalIterator *p){ + sqlite3ScratchFree(p); } /* -** Search the wal file for page pgno. If found, set *piRead to the frame that -** contains the page. Otherwise, if pgno is not in the wal file, set *piRead -** to zero. +** Construct a WalInterator object that can be used to loop over all +** pages in the WAL in ascending order. The caller must hold the checkpoint +** lock. ** -** Return SQLITE_OK if successful, or an error code if an error occurs. If an -** error does occur, the final value of *piRead is undefined. +** On success, make *pp point to the newly allocated WalInterator object +** return SQLITE_OK. Otherwise, return an error code. If this routine +** returns an error, the value of *pp is undefined. +** +** The calling routine should invoke walIteratorFree() to destroy the +** WalIterator object when it has finished with it. */ -SQLITE_PRIVATE int sqlite3WalFindFrame( - Wal *pWal, /* WAL handle */ - Pgno pgno, /* Database page number to read data for */ - u32 *piRead /* OUT: Frame number (or zero) */ -){ - u32 iRead = 0; /* If !=0, WAL frame to return data from */ - u32 iLast = pWal->hdr.mxFrame; /* Last page in WAL for this reader */ - int iHash; /* Used to loop through N hash tables */ - - /* This routine is only be called from within a read transaction. */ - assert( pWal->readLock>=0 || pWal->lockError ); +static int walIteratorInit(Wal *pWal, WalIterator **pp){ + WalIterator *p; /* Return value */ + int nSegment; /* Number of segments to merge */ + u32 iLast; /* Last frame in log */ + int nByte; /* Number of bytes to allocate */ + int i; /* Iterator variable */ + ht_slot *aTmp; /* Temp space used by merge-sort */ + int rc = SQLITE_OK; /* Return Code */ - /* If the "last page" field of the wal-index header snapshot is 0, then - ** no data will be read from the wal under any circumstances. Return early - ** in this case as an optimization. Likewise, if pWal->readLock==0, - ** then the WAL is ignored by the reader so return early, as if the - ** WAL were empty. + /* This routine only runs while holding the checkpoint lock. And + ** it only runs if there is actually content in the log (mxFrame>0). */ - if( iLast==0 || pWal->readLock==0 ){ - *piRead = 0; - return SQLITE_OK; + assert( pWal->ckptLock && pWal->hdr.mxFrame>0 ); + iLast = pWal->hdr.mxFrame; + + /* Allocate space for the WalIterator object. */ + nSegment = walFramePage(iLast) + 1; + nByte = sizeof(WalIterator) + + (nSegment-1)*sizeof(struct WalSegment) + + iLast*sizeof(ht_slot); + p = (WalIterator *)sqlite3ScratchMalloc(nByte); + if( !p ){ + return SQLITE_NOMEM; } + memset(p, 0, nByte); + p->nSegment = nSegment; - /* Search the hash table or tables for an entry matching page number - ** pgno. Each iteration of the following for() loop searches one - ** hash table (each hash table indexes up to HASHTABLE_NPAGE frames). - ** - ** This code might run concurrently to the code in walIndexAppend() - ** that adds entries to the wal-index (and possibly to this hash - ** table). This means the value just read from the hash - ** slot (aHash[iKey]) may have been added before or after the - ** current read transaction was opened. Values added after the - ** read transaction was opened may have been written incorrectly - - ** i.e. these slots may contain garbage data. However, we assume - ** that any slots written before the current read transaction was - ** opened remain unmodified. - ** - ** For the reasons above, the if(...) condition featured in the inner - ** loop of the following block is more stringent that would be required - ** if we had exclusive access to the hash-table: - ** - ** (aPgno[iFrame]==pgno): - ** This condition filters out normal hash-table collisions. - ** - ** (iFrame<=iLast): - ** This condition filters out entries that were added to the hash - ** table after the current read-transaction had started. + /* Allocate temporary space used by the merge-sort routine. This block + ** of memory will be freed before this function returns. */ - for(iHash=walFramePage(iLast); iHash>=0 && iRead==0; iHash--){ - volatile ht_slot *aHash; /* Pointer to hash table */ - volatile u32 *aPgno; /* Pointer to array of page numbers */ - u32 iZero; /* Frame number corresponding to aPgno[0] */ - int iKey; /* Hash slot index */ - int nCollide; /* Number of hash collisions remaining */ - int rc; /* Error code */ + aTmp = (ht_slot *)sqlite3ScratchMalloc( + sizeof(ht_slot) * (iLast>HASHTABLE_NPAGE?HASHTABLE_NPAGE:iLast) + ); + if( !aTmp ){ + rc = SQLITE_NOMEM; + } - rc = walHashGet(pWal, iHash, &aHash, &aPgno, &iZero); - if( rc!=SQLITE_OK ){ - return rc; - } - nCollide = HASHTABLE_NSLOT; - for(iKey=walHash(pgno); aHash[iKey]; iKey=walNextHash(iKey)){ - u32 iFrame = aHash[iKey] + iZero; - if( iFrame<=iLast && aPgno[aHash[iKey]]==pgno ){ - /* assert( iFrame>iRead ); -- not true if there is corruption */ - iRead = iFrame; + for(i=0; rc==SQLITE_OK && iaSegment[p->nSegment])[iZero]; + iZero++; + + for(j=0; jaSegment[i].iZero = iZero; + p->aSegment[i].nEntry = nEntry; + p->aSegment[i].aIndex = aIndex; + p->aSegment[i].aPgno = (u32 *)aPgno; } } + sqlite3ScratchFree(aTmp); -#ifdef SQLITE_ENABLE_EXPENSIVE_ASSERT - /* If expensive assert() statements are available, do a linear search - ** of the wal-index file content. Make sure the results agree with the - ** result obtained using the hash indexes above. */ - { - u32 iRead2 = 0; - u32 iTest; - for(iTest=iLast; iTest>0; iTest--){ - if( walFramePgno(pWal, iTest)==pgno ){ - iRead2 = iTest; - break; - } - } - assert( iRead==iRead2 ); + if( rc!=SQLITE_OK ){ + walIteratorFree(p); } -#endif - - *piRead = iRead; - return SQLITE_OK; + *pp = p; + return rc; } /* -** Read the contents of frame iRead from the wal file into buffer pOut -** (which is nOut bytes in size). Return SQLITE_OK if successful, or an -** error code otherwise. +** Attempt to obtain the exclusive WAL lock defined by parameters lockIdx and +** n. If the attempt fails and parameter xBusy is not NULL, then it is a +** busy-handler function. Invoke it and retry the lock until either the +** lock is successfully obtained or the busy-handler returns 0. */ -SQLITE_PRIVATE int sqlite3WalReadFrame( - Wal *pWal, /* WAL handle */ - u32 iRead, /* Frame to read */ - int nOut, /* Size of buffer pOut in bytes */ - u8 *pOut /* Buffer to write page data to */ +static int walBusyLock( + Wal *pWal, /* WAL connection */ + int (*xBusy)(void*), /* Function to call when busy */ + void *pBusyArg, /* Context argument for xBusyHandler */ + int lockIdx, /* Offset of first byte to lock */ + int n /* Number of bytes to lock */ ){ - int sz; - i64 iOffset; - sz = pWal->hdr.szPage; - sz = (sz&0xfe00) + ((sz&0x0001)<<16); - testcase( sz<=32768 ); - testcase( sz>=65536 ); - iOffset = walFrameOffset(iRead, sz) + WAL_FRAME_HDRSIZE; - /* testcase( IS_BIG_INT(iOffset) ); // requires a 4GiB WAL */ - return sqlite3OsRead(pWal->pWalFd, pOut, (nOut>sz ? sz : nOut), iOffset); + int rc; + do { + rc = walLockExclusive(pWal, lockIdx, n); + }while( xBusy && rc==SQLITE_BUSY && xBusy(pBusyArg) ); + return rc; } -/* -** Return the size of the database in pages (or zero, if unknown). +/* +** The cache of the wal-index header must be valid to call this function. +** Return the page-size in bytes used by the database. */ -SQLITE_PRIVATE Pgno sqlite3WalDbsize(Wal *pWal){ - if( pWal && ALWAYS(pWal->readLock>=0) ){ - return pWal->hdr.nPage; - } - return 0; +static int walPagesize(Wal *pWal){ + return (pWal->hdr.szPage&0xfe00) + ((pWal->hdr.szPage&0x0001)<<16); } - -/* -** This function starts a write transaction on the WAL. +/* +** Copy as much content as we can from the WAL back into the database file +** in response to an sqlite3_wal_checkpoint() request or the equivalent. ** -** A read transaction must have already been started by a prior call -** to sqlite3WalBeginReadTransaction(). +** The amount of information copies from WAL to database might be limited +** by active readers. This routine will never overwrite a database page +** that a concurrent reader might be using. ** -** If another thread or process has written into the database since -** the read transaction was started, then it is not possible for this -** thread to write as doing so would cause a fork. So this routine -** returns SQLITE_BUSY in that case and no write transaction is started. +** All I/O barrier operations (a.k.a fsyncs) occur in this routine when +** SQLite is in WAL-mode in synchronous=NORMAL. That means that if +** checkpoints are always run by a background thread or background +** process, foreground threads will never block on a lengthy fsync call. ** -** There can only be a single writer active at a time. +** Fsync is called on the WAL before writing content out of the WAL and +** into the database. This ensures that if the new content is persistent +** in the WAL and can be recovered following a power-loss or hard reset. +** +** Fsync is also called on the database file if (and only if) the entire +** WAL content is copied into the database file. This second fsync makes +** it safe to delete the WAL since the new content will persist in the +** database file. +** +** This routine uses and updates the nBackfill field of the wal-index header. +** This is the only routine tha will increase the value of nBackfill. +** (A WAL reset or recovery will revert nBackfill to zero, but not increase +** its value.) +** +** The caller must be holding sufficient locks to ensure that no other +** checkpoint is running (in any other thread or process) at the same +** time. */ -SQLITE_PRIVATE int sqlite3WalBeginWriteTransaction(Wal *pWal){ - int rc; - - /* Cannot start a write transaction without first holding a read - ** transaction. */ - assert( pWal->readLock>=0 ); +static int walCheckpoint( + Wal *pWal, /* Wal connection */ + int eMode, /* One of PASSIVE, FULL or RESTART */ + int (*xBusyCall)(void*), /* Function to call when busy */ + void *pBusyArg, /* Context argument for xBusyHandler */ + int sync_flags, /* Flags for OsSync() (or 0) */ + u8 *zBuf /* Temporary buffer to use */ +){ + int rc; /* Return code */ + int szPage; /* Database page-size */ + WalIterator *pIter = 0; /* Wal iterator context */ + u32 iDbpage = 0; /* Next database page to write */ + u32 iFrame = 0; /* Wal frame containing data for iDbpage */ + u32 mxSafeFrame; /* Max frame that can be backfilled */ + u32 mxPage; /* Max database page to write */ + int i; /* Loop counter */ + volatile WalCkptInfo *pInfo; /* The checkpoint status information */ + int (*xBusy)(void*) = 0; /* Function to call when waiting for locks */ - if( pWal->readOnly ){ - return SQLITE_READONLY; - } + szPage = walPagesize(pWal); + testcase( szPage<=32768 ); + testcase( szPage>=65536 ); + pInfo = walCkptInfo(pWal); + if( pInfo->nBackfill>=pWal->hdr.mxFrame ) return SQLITE_OK; - /* Only one writer allowed at a time. Get the write lock. Return - ** SQLITE_BUSY if unable. - */ - rc = walLockExclusive(pWal, WAL_WRITE_LOCK, 1); - if( rc ){ + /* Allocate the iterator */ + rc = walIteratorInit(pWal, &pIter); + if( rc!=SQLITE_OK ){ return rc; } - pWal->writeLock = 1; + assert( pIter ); - /* If another connection has written to the database file since the - ** time the read transaction on this connection was started, then - ** the write is disallowed. + if( eMode!=SQLITE_CHECKPOINT_PASSIVE ) xBusy = xBusyCall; + + /* Compute in mxSafeFrame the index of the last frame of the WAL that is + ** safe to write into the database. Frames beyond mxSafeFrame might + ** overwrite database pages that are in use by active readers and thus + ** cannot be backfilled from the WAL. */ - if( memcmp(&pWal->hdr, (void *)walIndexHdr(pWal), sizeof(WalIndexHdr))!=0 ){ - walUnlockExclusive(pWal, WAL_WRITE_LOCK, 1); - pWal->writeLock = 0; - rc = SQLITE_BUSY_SNAPSHOT; + mxSafeFrame = pWal->hdr.mxFrame; + mxPage = pWal->hdr.nPage; + for(i=1; iaReadMark[i]; + if( mxSafeFrame>y ){ + assert( y<=pWal->hdr.mxFrame ); + rc = walBusyLock(pWal, xBusy, pBusyArg, WAL_READ_LOCK(i), 1); + if( rc==SQLITE_OK ){ + pInfo->aReadMark[i] = (i==1 ? mxSafeFrame : READMARK_NOT_USED); + walUnlockExclusive(pWal, WAL_READ_LOCK(i), 1); + }else if( rc==SQLITE_BUSY ){ + mxSafeFrame = y; + xBusy = 0; + }else{ + goto walcheckpoint_out; + } + } } - return rc; -} + if( pInfo->nBackfillnBackfill; -/* -** End a write transaction. The commit has already been done. This -** routine merely releases the lock. -*/ -SQLITE_PRIVATE int sqlite3WalEndWriteTransaction(Wal *pWal){ - if( pWal->writeLock ){ - walUnlockExclusive(pWal, WAL_WRITE_LOCK, 1); - pWal->writeLock = 0; - pWal->truncateOnCommit = 0; - } - return SQLITE_OK; -} + /* Sync the WAL to disk */ + if( sync_flags ){ + rc = sqlite3OsSync(pWal->pWalFd, sync_flags); + } -/* -** If any data has been written (but not committed) to the log file, this -** function moves the write-pointer back to the start of the transaction. -** -** Additionally, the callback function is invoked for each frame written -** to the WAL since the start of the transaction. If the callback returns -** other than SQLITE_OK, it is not invoked again and the error code is -** returned to the caller. -** -** Otherwise, if the callback function does not return an error, this -** function returns SQLITE_OK. -*/ -SQLITE_PRIVATE int sqlite3WalUndo(Wal *pWal, int (*xUndo)(void *, Pgno), void *pUndoCtx){ - int rc = SQLITE_OK; - if( ALWAYS(pWal->writeLock) ){ - Pgno iMax = pWal->hdr.mxFrame; - Pgno iFrame; - - /* Restore the clients cache of the wal-index header to the state it - ** was in before the client began writing to the database. + /* If the database may grow as a result of this checkpoint, hint + ** about the eventual size of the db file to the VFS layer. */ - memcpy(&pWal->hdr, (void *)walIndexHdr(pWal), sizeof(WalIndexHdr)); - - for(iFrame=pWal->hdr.mxFrame+1; - ALWAYS(rc==SQLITE_OK) && iFrame<=iMax; - iFrame++ - ){ - /* This call cannot fail. Unless the page for which the page number - ** is passed as the second argument is (a) in the cache and - ** (b) has an outstanding reference, then xUndo is either a no-op - ** (if (a) is false) or simply expels the page from the cache (if (b) - ** is false). - ** - ** If the upper layer is doing a rollback, it is guaranteed that there - ** are no outstanding references to any page other than page 1. And - ** page 1 is never written to the log until the transaction is - ** committed. As a result, the call to xUndo may not fail. - */ - assert( walFramePgno(pWal, iFrame)!=1 ); - rc = xUndo(pUndoCtx, walFramePgno(pWal, iFrame)); + if( rc==SQLITE_OK ){ + i64 nReq = ((i64)mxPage * szPage); + rc = sqlite3OsFileSize(pWal->pDbFd, &nSize); + if( rc==SQLITE_OK && nSizepDbFd, SQLITE_FCNTL_SIZE_HINT, &nReq); + } } - if( iMax!=pWal->hdr.mxFrame ) walCleanupHash(pWal); - } - assert( rc==SQLITE_OK ); - return rc; -} -/* -** Argument aWalData must point to an array of WAL_SAVEPOINT_NDATA u32 -** values. This function populates the array with values required to -** "rollback" the write position of the WAL handle back to the current -** point in the event of a savepoint rollback (via WalSavepointUndo()). -*/ -SQLITE_PRIVATE void sqlite3WalSavepoint(Wal *pWal, u32 *aWalData){ - assert( pWal->writeLock ); - aWalData[0] = pWal->hdr.mxFrame; - aWalData[1] = pWal->hdr.aFrameCksum[0]; - aWalData[2] = pWal->hdr.aFrameCksum[1]; - aWalData[3] = pWal->nCkpt; -} -/* -** Move the write position of the WAL back to the point identified by -** the values in the aWalData[] array. aWalData must point to an array -** of WAL_SAVEPOINT_NDATA u32 values that has been previously populated -** by a call to WalSavepoint(). -*/ -SQLITE_PRIVATE int sqlite3WalSavepointUndo(Wal *pWal, u32 *aWalData){ - int rc = SQLITE_OK; + /* Iterate through the contents of the WAL, copying data to the db file. */ + while( rc==SQLITE_OK && 0==walIteratorNext(pIter, &iDbpage, &iFrame) ){ + i64 iOffset; + assert( walFramePgno(pWal, iFrame)==iDbpage ); + if( iFrame<=nBackfill || iFrame>mxSafeFrame || iDbpage>mxPage ) continue; + iOffset = walFrameOffset(iFrame, szPage) + WAL_FRAME_HDRSIZE; + /* testcase( IS_BIG_INT(iOffset) ); // requires a 4GiB WAL file */ + rc = sqlite3OsRead(pWal->pWalFd, zBuf, szPage, iOffset); + if( rc!=SQLITE_OK ) break; + iOffset = (iDbpage-1)*(i64)szPage; + testcase( IS_BIG_INT(iOffset) ); + rc = sqlite3OsWrite(pWal->pDbFd, zBuf, szPage, iOffset); + if( rc!=SQLITE_OK ) break; + } - assert( pWal->writeLock ); - assert( aWalData[3]!=pWal->nCkpt || aWalData[0]<=pWal->hdr.mxFrame ); + /* If work was actually accomplished... */ + if( rc==SQLITE_OK ){ + if( mxSafeFrame==walIndexHdr(pWal)->mxFrame ){ + i64 szDb = pWal->hdr.nPage*(i64)szPage; + testcase( IS_BIG_INT(szDb) ); + rc = sqlite3OsTruncate(pWal->pDbFd, szDb); + if( rc==SQLITE_OK && sync_flags ){ + rc = sqlite3OsSync(pWal->pDbFd, sync_flags); + } + } + if( rc==SQLITE_OK ){ + pInfo->nBackfill = mxSafeFrame; + } + } - if( aWalData[3]!=pWal->nCkpt ){ - /* This savepoint was opened immediately after the write-transaction - ** was started. Right after that, the writer decided to wrap around - ** to the start of the log. Update the savepoint values to match. - */ - aWalData[0] = 0; - aWalData[3] = pWal->nCkpt; + /* Release the reader lock held while backfilling */ + walUnlockExclusive(pWal, WAL_READ_LOCK(0), 1); } - if( aWalData[0]hdr.mxFrame ){ - pWal->hdr.mxFrame = aWalData[0]; - pWal->hdr.aFrameCksum[0] = aWalData[1]; - pWal->hdr.aFrameCksum[1] = aWalData[2]; - walCleanupHash(pWal); + if( rc==SQLITE_BUSY ){ + /* Reset the return code so as not to report a checkpoint failure + ** just because there are active readers. */ + rc = SQLITE_OK; } - return rc; -} - - -/* -** This function is called just before writing a set of frames to the log -** file (see sqlite3WalFrames()). It checks to see if, instead of appending -** to the current log file, it is possible to overwrite the start of the -** existing log file with the new frames (i.e. "reset" the log). If so, -** it sets pWal->hdr.mxFrame to 0. Otherwise, pWal->hdr.mxFrame is left -** unchanged. -** -** SQLITE_OK is returned if no error is encountered (regardless of whether -** or not pWal->hdr.mxFrame is modified). An SQLite error code is returned -** if an error occurs. -*/ -static int walRestartLog(Wal *pWal){ - int rc = SQLITE_OK; - int cnt; - - if( pWal->readLock==0 ){ - volatile WalCkptInfo *pInfo = walCkptInfo(pWal); - assert( pInfo->nBackfill==pWal->hdr.mxFrame ); - if( pInfo->nBackfill>0 ){ - u32 salt1; - sqlite3_randomness(4, &salt1); - rc = walLockExclusive(pWal, WAL_READ_LOCK(1), WAL_NREADER-1); + /* If this is an SQLITE_CHECKPOINT_RESTART operation, and the entire wal + ** file has been copied into the database file, then block until all + ** readers have finished using the wal file. This ensures that the next + ** process to write to the database restarts the wal file. + */ + if( rc==SQLITE_OK && eMode!=SQLITE_CHECKPOINT_PASSIVE ){ + assert( pWal->writeLock ); + if( pInfo->nBackfillhdr.mxFrame ){ + rc = SQLITE_BUSY; + }else if( eMode==SQLITE_CHECKPOINT_RESTART ){ + assert( mxSafeFrame==pWal->hdr.mxFrame ); + rc = walBusyLock(pWal, xBusy, pBusyArg, WAL_READ_LOCK(1), WAL_NREADER-1); if( rc==SQLITE_OK ){ - /* If all readers are using WAL_READ_LOCK(0) (in other words if no - ** readers are currently using the WAL), then the transactions - ** frames will overwrite the start of the existing log. Update the - ** wal-index header to reflect this. - ** - ** In theory it would be Ok to update the cache of the header only - ** at this point. But updating the actual wal-index header is also - ** safe and means there is no special case for sqlite3WalUndo() - ** to handle if this transaction is rolled back. - */ - int i; /* Loop counter */ - u32 *aSalt = pWal->hdr.aSalt; /* Big-endian salt values */ - - pWal->nCkpt++; - pWal->hdr.mxFrame = 0; - sqlite3Put4byte((u8*)&aSalt[0], 1 + sqlite3Get4byte((u8*)&aSalt[0])); - aSalt[1] = salt1; - walIndexWriteHdr(pWal); - pInfo->nBackfill = 0; - pInfo->aReadMark[1] = 0; - for(i=2; iaReadMark[i] = READMARK_NOT_USED; - assert( pInfo->aReadMark[0]==0 ); walUnlockExclusive(pWal, WAL_READ_LOCK(1), WAL_NREADER-1); - }else if( rc!=SQLITE_BUSY ){ - return rc; } } - walUnlockShared(pWal, WAL_READ_LOCK(0)); - pWal->readLock = -1; - cnt = 0; - do{ - int notUsed; - rc = walTryBeginRead(pWal, ¬Used, 1, ++cnt); - }while( rc==WAL_RETRY ); - assert( (rc&0xff)!=SQLITE_BUSY ); /* BUSY not possible when useWal==1 */ - testcase( (rc&0xff)==SQLITE_IOERR ); - testcase( rc==SQLITE_PROTOCOL ); - testcase( rc==SQLITE_OK ); } + + walcheckpoint_out: + walIteratorFree(pIter); return rc; } /* -** Information about the current state of the WAL file and where -** the next fsync should occur - passed from sqlite3WalFrames() into -** walWriteToLog(). -*/ -typedef struct WalWriter { - Wal *pWal; /* The complete WAL information */ - sqlite3_file *pFd; /* The WAL file to which we write */ - sqlite3_int64 iSyncPoint; /* Fsync at this offset */ - int syncFlags; /* Flags for the fsync */ - int szPage; /* Size of one page */ -} WalWriter; - -/* -** Write iAmt bytes of content into the WAL file beginning at iOffset. -** Do a sync when crossing the p->iSyncPoint boundary. -** -** In other words, if iSyncPoint is in between iOffset and iOffset+iAmt, -** first write the part before iSyncPoint, then sync, then write the -** rest. +** If the WAL file is currently larger than nMax bytes in size, truncate +** it to exactly nMax bytes. If an error occurs while doing so, ignore it. */ -static int walWriteToLog( - WalWriter *p, /* WAL to write to */ - void *pContent, /* Content to be written */ - int iAmt, /* Number of bytes to write */ - sqlite3_int64 iOffset /* Start writing at this offset */ -){ - int rc; - if( iOffsetiSyncPoint && iOffset+iAmt>=p->iSyncPoint ){ - int iFirstAmt = (int)(p->iSyncPoint - iOffset); - rc = sqlite3OsWrite(p->pFd, pContent, iFirstAmt, iOffset); - if( rc ) return rc; - iOffset += iFirstAmt; - iAmt -= iFirstAmt; - pContent = (void*)(iFirstAmt + (char*)pContent); - assert( p->syncFlags & (SQLITE_SYNC_NORMAL|SQLITE_SYNC_FULL) ); - rc = sqlite3OsSync(p->pFd, p->syncFlags); - if( iAmt==0 || rc ) return rc; +static void walLimitSize(Wal *pWal, i64 nMax){ + i64 sz; + int rx; + sqlite3BeginBenignMalloc(); + rx = sqlite3OsFileSize(pWal->pWalFd, &sz); + if( rx==SQLITE_OK && (sz > nMax ) ){ + rx = sqlite3OsTruncate(pWal->pWalFd, nMax); + } + sqlite3EndBenignMalloc(); + if( rx ){ + sqlite3_log(rx, "cannot limit WAL size: %s", pWal->zWalName); } - rc = sqlite3OsWrite(p->pFd, pContent, iAmt, iOffset); - return rc; } /* -** Write out a single frame of the WAL -*/ -static int walWriteOneFrame( - WalWriter *p, /* Where to write the frame */ - PgHdr *pPage, /* The page of the frame to be written */ - int nTruncate, /* The commit flag. Usually 0. >0 for commit */ - sqlite3_int64 iOffset /* Byte offset at which to write */ -){ - int rc; /* Result code from subfunctions */ - void *pData; /* Data actually written */ - u8 aFrame[WAL_FRAME_HDRSIZE]; /* Buffer to assemble frame-header in */ -#if defined(SQLITE_HAS_CODEC) - if( (pData = sqlite3PagerCodec(pPage))==0 ) return SQLITE_NOMEM; -#else - pData = pPage->pData; -#endif - walEncodeFrame(p->pWal, pPage->pgno, nTruncate, pData, aFrame); - rc = walWriteToLog(p, aFrame, sizeof(aFrame), iOffset); - if( rc ) return rc; - /* Write the page data */ - rc = walWriteToLog(p, pData, p->szPage, iOffset+sizeof(aFrame)); - return rc; -} - -/* -** Write a set of frames to the log. The caller must hold the write-lock -** on the log file (obtained using sqlite3WalBeginWriteTransaction()). +** Close a connection to a log file. */ -SQLITE_PRIVATE int sqlite3WalFrames( - Wal *pWal, /* Wal handle to write to */ - int szPage, /* Database page-size in bytes */ - PgHdr *pList, /* List of dirty pages to write */ - Pgno nTruncate, /* Database size after this commit */ - int isCommit, /* True if this is a commit */ - int sync_flags /* Flags to pass to OsSync() (or 0) */ +SQLITE_PRIVATE int sqlite3WalClose( + Wal *pWal, /* Wal to close */ + int sync_flags, /* Flags to pass to OsSync() (or 0) */ + int nBuf, + u8 *zBuf /* Buffer of at least nBuf bytes */ ){ - int rc; /* Used to catch return codes */ - u32 iFrame; /* Next frame address */ - PgHdr *p; /* Iterator to run through pList with. */ - PgHdr *pLast = 0; /* Last frame in list */ - int nExtra = 0; /* Number of extra copies of last page */ - int szFrame; /* The size of a single frame */ - i64 iOffset; /* Next byte to write in WAL file */ - WalWriter w; /* The writer */ - - assert( pList ); - assert( pWal->writeLock ); + int rc = SQLITE_OK; + if( pWal ){ + int isDelete = 0; /* True to unlink wal and wal-index files */ - /* If this frame set completes a transaction, then nTruncate>0. If - ** nTruncate==0 then this frame set does not complete the transaction. */ - assert( (isCommit!=0)==(nTruncate!=0) ); + /* If an EXCLUSIVE lock can be obtained on the database file (using the + ** ordinary, rollback-mode locking methods, this guarantees that the + ** connection associated with this log file is the only connection to + ** the database. In this case checkpoint the database and unlink both + ** the wal and wal-index files. + ** + ** The EXCLUSIVE lock is not released before returning. + */ + rc = sqlite3OsLock(pWal->pDbFd, SQLITE_LOCK_EXCLUSIVE); + if( rc==SQLITE_OK ){ + if( pWal->exclusiveMode==WAL_NORMAL_MODE ){ + pWal->exclusiveMode = WAL_EXCLUSIVE_MODE; + } + rc = sqlite3WalCheckpoint( + pWal, SQLITE_CHECKPOINT_PASSIVE, 0, 0, sync_flags, nBuf, zBuf, 0, 0 + ); + if( rc==SQLITE_OK ){ + int bPersist = -1; + sqlite3OsFileControlHint( + pWal->pDbFd, SQLITE_FCNTL_PERSIST_WAL, &bPersist + ); + if( bPersist!=1 ){ + /* Try to delete the WAL file if the checkpoint completed and + ** fsyned (rc==SQLITE_OK) and if we are not in persistent-wal + ** mode (!bPersist) */ + isDelete = 1; + }else if( pWal->mxWalSize>=0 ){ + /* Try to truncate the WAL file to zero bytes if the checkpoint + ** completed and fsynced (rc==SQLITE_OK) and we are in persistent + ** WAL mode (bPersist) and if the PRAGMA journal_size_limit is a + ** non-negative value (pWal->mxWalSize>=0). Note that we truncate + ** to zero bytes as truncating to the journal_size_limit might + ** leave a corrupt WAL file on disk. */ + walLimitSize(pWal, 0); + } + } + } -#if defined(SQLITE_TEST) && defined(SQLITE_DEBUG) - { int cnt; for(cnt=0, p=pList; p; p=p->pDirty, cnt++){} - WALTRACE(("WAL%p: frame write begin. %d frames. mxFrame=%d. %s\n", - pWal, cnt, pWal->hdr.mxFrame, isCommit ? "Commit" : "Spill")); + walIndexClose(pWal, isDelete); + sqlite3OsClose(pWal->pWalFd); + if( isDelete ){ + sqlite3BeginBenignMalloc(); + sqlite3OsDelete(pWal->pVfs, pWal->zWalName, 0); + sqlite3EndBenignMalloc(); + } + WALTRACE(("WAL%p: closed\n", pWal)); + sqlite3_free((void *)pWal->apWiData); + sqlite3_free(pWal); } -#endif + return rc; +} - /* See if it is possible to write these frames into the start of the - ** log file, instead of appending to it at pWal->hdr.mxFrame. - */ - if( SQLITE_OK!=(rc = walRestartLog(pWal)) ){ - return rc; - } +/* +** Try to read the wal-index header. Return 0 on success and 1 if +** there is a problem. +** +** The wal-index is in shared memory. Another thread or process might +** be writing the header at the same time this procedure is trying to +** read it, which might result in inconsistency. A dirty read is detected +** by verifying that both copies of the header are the same and also by +** a checksum on the header. +** +** If and only if the read is consistent and the header is different from +** pWal->hdr, then pWal->hdr is updated to the content of the new header +** and *pChanged is set to 1. +** +** If the checksum cannot be verified return non-zero. If the header +** is read successfully and the checksum verified, return zero. +*/ +static int walIndexTryHdr(Wal *pWal, int *pChanged){ + u32 aCksum[2]; /* Checksum on the header content */ + WalIndexHdr h1, h2; /* Two copies of the header content */ + WalIndexHdr volatile *aHdr; /* Header in shared memory */ - /* If this is the first frame written into the log, write the WAL - ** header to the start of the WAL file. See comments at the top of - ** this source file for a description of the WAL header format. - */ - iFrame = pWal->hdr.mxFrame; - if( iFrame==0 ){ - u8 aWalHdr[WAL_HDRSIZE]; /* Buffer to assemble wal-header in */ - u32 aCksum[2]; /* Checksum for wal-header */ + /* The first page of the wal-index must be mapped at this point. */ + assert( pWal->nWiData>0 && pWal->apWiData[0] ); - sqlite3Put4byte(&aWalHdr[0], (WAL_MAGIC | SQLITE_BIGENDIAN)); - sqlite3Put4byte(&aWalHdr[4], WAL_MAX_VERSION); - sqlite3Put4byte(&aWalHdr[8], szPage); - sqlite3Put4byte(&aWalHdr[12], pWal->nCkpt); - if( pWal->nCkpt==0 ) sqlite3_randomness(8, pWal->hdr.aSalt); - memcpy(&aWalHdr[16], pWal->hdr.aSalt, 8); - walChecksumBytes(1, aWalHdr, WAL_HDRSIZE-2*4, 0, aCksum); - sqlite3Put4byte(&aWalHdr[24], aCksum[0]); - sqlite3Put4byte(&aWalHdr[28], aCksum[1]); - - pWal->szPage = szPage; - pWal->hdr.bigEndCksum = SQLITE_BIGENDIAN; - pWal->hdr.aFrameCksum[0] = aCksum[0]; - pWal->hdr.aFrameCksum[1] = aCksum[1]; - pWal->truncateOnCommit = 1; + /* Read the header. This might happen concurrently with a write to the + ** same area of shared memory on a different CPU in a SMP, + ** meaning it is possible that an inconsistent snapshot is read + ** from the file. If this happens, return non-zero. + ** + ** There are two copies of the header at the beginning of the wal-index. + ** When reading, read [0] first then [1]. Writes are in the reverse order. + ** Memory barriers are used to prevent the compiler or the hardware from + ** reordering the reads and writes. + */ + aHdr = walIndexHdr(pWal); + memcpy(&h1, (void *)&aHdr[0], sizeof(h1)); + walShmBarrier(pWal); + memcpy(&h2, (void *)&aHdr[1], sizeof(h2)); - rc = sqlite3OsWrite(pWal->pWalFd, aWalHdr, sizeof(aWalHdr), 0); - WALTRACE(("WAL%p: wal-header write %s\n", pWal, rc ? "failed" : "ok")); - if( rc!=SQLITE_OK ){ - return rc; - } + if( memcmp(&h1, &h2, sizeof(h1))!=0 ){ + return 1; /* Dirty read */ + } + if( h1.isInit==0 ){ + return 1; /* Malformed header - probably all zeros */ + } + walChecksumBytes(1, (u8*)&h1, sizeof(h1)-sizeof(h1.aCksum), 0, aCksum); + if( aCksum[0]!=h1.aCksum[0] || aCksum[1]!=h1.aCksum[1] ){ + return 1; /* Checksum does not match */ + } - /* Sync the header (unless SQLITE_IOCAP_SEQUENTIAL is true or unless - ** all syncing is turned off by PRAGMA synchronous=OFF). Otherwise - ** an out-of-order write following a WAL restart could result in - ** database corruption. See the ticket: - ** - ** http://localhost:591/sqlite/info/ff5be73dee - */ - if( pWal->syncHeader && sync_flags ){ - rc = sqlite3OsSync(pWal->pWalFd, sync_flags & SQLITE_SYNC_MASK); - if( rc ) return rc; - } + if( memcmp(&pWal->hdr, &h1, sizeof(WalIndexHdr)) ){ + *pChanged = 1; + memcpy(&pWal->hdr, &h1, sizeof(WalIndexHdr)); + pWal->szPage = (pWal->hdr.szPage&0xfe00) + ((pWal->hdr.szPage&0x0001)<<16); + testcase( pWal->szPage<=32768 ); + testcase( pWal->szPage>=65536 ); } - assert( (int)pWal->szPage==szPage ); - /* Setup information needed to write frames into the WAL */ - w.pWal = pWal; - w.pFd = pWal->pWalFd; - w.iSyncPoint = 0; - w.syncFlags = sync_flags; - w.szPage = szPage; - iOffset = walFrameOffset(iFrame+1, szPage); - szFrame = szPage + WAL_FRAME_HDRSIZE; + /* The header was successfully read. Return zero. */ + return 0; +} - /* Write all frames into the log file exactly once */ - for(p=pList; p; p=p->pDirty){ - int nDbSize; /* 0 normally. Positive == commit flag */ - iFrame++; - assert( iOffset==walFrameOffset(iFrame, szPage) ); - nDbSize = (isCommit && p->pDirty==0) ? nTruncate : 0; - rc = walWriteOneFrame(&w, p, nDbSize, iOffset); - if( rc ) return rc; - pLast = p; - iOffset += szFrame; - } +/* +** Read the wal-index header from the wal-index and into pWal->hdr. +** If the wal-header appears to be corrupt, try to reconstruct the +** wal-index from the WAL before returning. +** +** Set *pChanged to 1 if the wal-index header value in pWal->hdr is +** changed by this opertion. If pWal->hdr is unchanged, set *pChanged +** to 0. +** +** If the wal-index header is successfully read, return SQLITE_OK. +** Otherwise an SQLite error code. +*/ +static int walIndexReadHdr(Wal *pWal, int *pChanged){ + int rc; /* Return code */ + int badHdr; /* True if a header read failed */ + volatile u32 *page0; /* Chunk of wal-index containing header */ - /* If this is the end of a transaction, then we might need to pad - ** the transaction and/or sync the WAL file. - ** - ** Padding and syncing only occur if this set of frames complete a - ** transaction and if PRAGMA synchronous=FULL. If synchronous==NORMAL - ** or synchonous==OFF, then no padding or syncing are needed. - ** - ** If SQLITE_IOCAP_POWERSAFE_OVERWRITE is defined, then padding is not - ** needed and only the sync is done. If padding is needed, then the - ** final frame is repeated (with its commit mark) until the next sector - ** boundary is crossed. Only the part of the WAL prior to the last - ** sector boundary is synced; the part of the last frame that extends - ** past the sector boundary is written after the sync. + /* Ensure that page 0 of the wal-index (the page that contains the + ** wal-index header) is mapped. Return early if an error occurs here. */ - if( isCommit && (sync_flags & WAL_SYNC_TRANSACTIONS)!=0 ){ - if( pWal->padToSectorBoundary ){ - int sectorSize = sqlite3SectorSize(pWal->pWalFd); - w.iSyncPoint = ((iOffset+sectorSize-1)/sectorSize)*sectorSize; - while( iOffsetwriteLock==0 ); - /* If this frame set completes the first transaction in the WAL and - ** if PRAGMA journal_size_limit is set, then truncate the WAL to the - ** journal size limit, if possible. + /* If the first page of the wal-index has been mapped, try to read the + ** wal-index header immediately, without holding any lock. This usually + ** works, but may fail if the wal-index header is corrupt or currently + ** being modified by another thread or process. */ - if( isCommit && pWal->truncateOnCommit && pWal->mxWalSize>=0 ){ - i64 sz = pWal->mxWalSize; - if( walFrameOffset(iFrame+nExtra+1, szPage)>pWal->mxWalSize ){ - sz = walFrameOffset(iFrame+nExtra+1, szPage); - } - walLimitSize(pWal, sz); - pWal->truncateOnCommit = 0; - } + badHdr = (page0 ? walIndexTryHdr(pWal, pChanged) : 1); - /* Append data to the wal-index. It is not necessary to lock the - ** wal-index to do this as the SQLITE_SHM_WRITE lock held on the wal-index - ** guarantees that there are no other writers, and no data that may - ** be in use by existing readers is being overwritten. + /* If the first attempt failed, it might have been due to a race + ** with a writer. So get a WRITE lock and try again. */ - iFrame = pWal->hdr.mxFrame; - for(p=pList; p && rc==SQLITE_OK; p=p->pDirty){ - iFrame++; - rc = walIndexAppend(pWal, iFrame, p->pgno); - } - while( rc==SQLITE_OK && nExtra>0 ){ - iFrame++; - nExtra--; - rc = walIndexAppend(pWal, iFrame, pLast->pgno); + assert( badHdr==0 || pWal->writeLock==0 ); + if( badHdr ){ + if( pWal->readOnly & WAL_SHM_RDONLY ){ + if( SQLITE_OK==(rc = walLockShared(pWal, WAL_WRITE_LOCK)) ){ + walUnlockShared(pWal, WAL_WRITE_LOCK); + rc = SQLITE_READONLY_RECOVERY; + } + }else if( SQLITE_OK==(rc = walLockExclusive(pWal, WAL_WRITE_LOCK, 1)) ){ + pWal->writeLock = 1; + if( SQLITE_OK==(rc = walIndexPage(pWal, 0, &page0)) ){ + badHdr = walIndexTryHdr(pWal, pChanged); + if( badHdr ){ + /* If the wal-index header is still malformed even while holding + ** a WRITE lock, it can only mean that the header is corrupted and + ** needs to be reconstructed. So run recovery to do exactly that. + */ + rc = walIndexRecover(pWal); + *pChanged = 1; + } + } + pWal->writeLock = 0; + walUnlockExclusive(pWal, WAL_WRITE_LOCK, 1); + } } - if( rc==SQLITE_OK ){ - /* Update the private copy of the header. */ - pWal->hdr.szPage = (u16)((szPage&0xff00) | (szPage>>16)); - testcase( szPage<=32768 ); - testcase( szPage>=65536 ); - pWal->hdr.mxFrame = iFrame; - if( isCommit ){ - pWal->hdr.iChange++; - pWal->hdr.nPage = nTruncate; - } - /* If this is a commit, update the wal-index header too. */ - if( isCommit ){ - walIndexWriteHdr(pWal); - pWal->iCallback = iFrame; - } + /* If the header is read successfully, check the version number to make + ** sure the wal-index was not constructed with some future format that + ** this version of SQLite cannot understand. + */ + if( badHdr==0 && pWal->hdr.iVersion!=WALINDEX_MAX_VERSION ){ + rc = SQLITE_CANTOPEN_BKPT; } - WALTRACE(("WAL%p: frame write %s\n", pWal, rc ? "failed" : "ok")); return rc; } -/* -** This routine is called to implement sqlite3_wal_checkpoint() and -** related interfaces. +/* +** This is the value that walTryBeginRead returns when it needs to +** be retried. +*/ +#define WAL_RETRY (-1) + +/* +** Attempt to start a read transaction. This might fail due to a race or +** other transient condition. When that happens, it returns WAL_RETRY to +** indicate to the caller that it is safe to retry immediately. ** -** Obtain a CHECKPOINT lock and then backfill as much information as -** we can from WAL into the database. +** On success return SQLITE_OK. On a permanent failure (such an +** I/O error or an SQLITE_BUSY because another process is running +** recovery) return a positive error code. ** -** If parameter xBusy is not NULL, it is a pointer to a busy-handler -** callback. In this case this function runs a blocking checkpoint. +** The useWal parameter is true to force the use of the WAL and disable +** the case where the WAL is bypassed because it has been completely +** checkpointed. If useWal==0 then this routine calls walIndexReadHdr() +** to make a copy of the wal-index header into pWal->hdr. If the +** wal-index header has changed, *pChanged is set to 1 (as an indication +** to the caller that the local paget cache is obsolete and needs to be +** flushed.) When useWal==1, the wal-index header is assumed to already +** be loaded and the pChanged parameter is unused. +** +** The caller must set the cnt parameter to the number of prior calls to +** this routine during the current read attempt that returned WAL_RETRY. +** This routine will start taking more aggressive measures to clear the +** race conditions after multiple WAL_RETRY returns, and after an excessive +** number of errors will ultimately return SQLITE_PROTOCOL. The +** SQLITE_PROTOCOL return indicates that some other process has gone rogue +** and is not honoring the locking protocol. There is a vanishingly small +** chance that SQLITE_PROTOCOL could be returned because of a run of really +** bad luck when there is lots of contention for the wal-index, but that +** possibility is so small that it can be safely neglected, we believe. +** +** On success, this routine obtains a read lock on +** WAL_READ_LOCK(pWal->readLock). The pWal->readLock integer is +** in the range 0 <= pWal->readLock < WAL_NREADER. If pWal->readLock==(-1) +** that means the Wal does not hold any read lock. The reader must not +** access any database page that is modified by a WAL frame up to and +** including frame number aReadMark[pWal->readLock]. The reader will +** use WAL frames up to and including pWal->hdr.mxFrame if pWal->readLock>0 +** Or if pWal->readLock==0, then the reader will ignore the WAL +** completely and get all content directly from the database file. +** If the useWal parameter is 1 then the WAL will never be ignored and +** this routine will always set pWal->readLock>0 on success. +** When the read transaction is completed, the caller must release the +** lock on WAL_READ_LOCK(pWal->readLock) and set pWal->readLock to -1. +** +** This routine uses the nBackfill and aReadMark[] fields of the header +** to select a particular WAL_READ_LOCK() that strives to let the +** checkpoint process do as much work as possible. This routine might +** update values of the aReadMark[] array in the header, but if it does +** so it takes care to hold an exclusive lock on the corresponding +** WAL_READ_LOCK() while changing values. */ -SQLITE_PRIVATE int sqlite3WalCheckpoint( - Wal *pWal, /* Wal connection */ - int eMode, /* PASSIVE, FULL or RESTART */ - int (*xBusy)(void*), /* Function to call when busy */ - void *pBusyArg, /* Context argument for xBusyHandler */ - int sync_flags, /* Flags to sync db file with (or 0) */ - int nBuf, /* Size of temporary buffer */ - u8 *zBuf, /* Temporary buffer to use */ - int *pnLog, /* OUT: Number of frames in WAL */ - int *pnCkpt /* OUT: Number of backfilled frames in WAL */ -){ - int rc; /* Return code */ - int isChanged = 0; /* True if a new wal-index header is loaded */ - int eMode2 = eMode; /* Mode to pass to walCheckpoint() */ - - assert( pWal->ckptLock==0 ); - assert( pWal->writeLock==0 ); +static int walTryBeginRead(Wal *pWal, int *pChanged, int useWal, int cnt){ + volatile WalCkptInfo *pInfo; /* Checkpoint information in wal-index */ + u32 mxReadMark; /* Largest aReadMark[] value */ + int mxI; /* Index of largest aReadMark[] value */ + int i; /* Loop counter */ + int rc = SQLITE_OK; /* Return code */ - if( pWal->readOnly ) return SQLITE_READONLY; - WALTRACE(("WAL%p: checkpoint begins\n", pWal)); - rc = walLockExclusive(pWal, WAL_CKPT_LOCK, 1); - if( rc ){ - /* Usually this is SQLITE_BUSY meaning that another thread or process - ** is already running a checkpoint, or maybe a recovery. But it might - ** also be SQLITE_IOERR. */ - return rc; - } - pWal->ckptLock = 1; + assert( pWal->readLock<0 ); /* Not currently locked */ - /* If this is a blocking-checkpoint, then obtain the write-lock as well - ** to prevent any writers from running while the checkpoint is underway. - ** This has to be done before the call to walIndexReadHdr() below. + /* Take steps to avoid spinning forever if there is a protocol error. ** - ** If the writer lock cannot be obtained, then a passive checkpoint is - ** run instead. Since the checkpointer is not holding the writer lock, - ** there is no point in blocking waiting for any readers. Assuming no - ** other error occurs, this function will return SQLITE_BUSY to the caller. + ** Circumstances that cause a RETRY should only last for the briefest + ** instances of time. No I/O or other system calls are done while the + ** locks are held, so the locks should not be held for very long. But + ** if we are unlucky, another process that is holding a lock might get + ** paged out or take a page-fault that is time-consuming to resolve, + ** during the few nanoseconds that it is holding the lock. In that case, + ** it might take longer than normal for the lock to free. + ** + ** After 5 RETRYs, we begin calling sqlite3OsSleep(). The first few + ** calls to sqlite3OsSleep() have a delay of 1 microsecond. Really this + ** is more of a scheduler yield than an actual delay. But on the 10th + ** an subsequent retries, the delays start becoming longer and longer, + ** so that on the 100th (and last) RETRY we delay for 21 milliseconds. + ** The total delay time before giving up is less than 1 second. */ - if( eMode!=SQLITE_CHECKPOINT_PASSIVE ){ - rc = walBusyLock(pWal, xBusy, pBusyArg, WAL_WRITE_LOCK, 1); + if( cnt>5 ){ + int nDelay = 1; /* Pause time in microseconds */ + if( cnt>100 ){ + VVA_ONLY( pWal->lockError = 1; ) + return SQLITE_PROTOCOL; + } + if( cnt>=10 ) nDelay = (cnt-9)*238; /* Max delay 21ms. Total delay 996ms */ + sqlite3OsSleep(pWal->pVfs, nDelay); + } + + if( !useWal ){ + rc = walIndexReadHdr(pWal, pChanged); + if( rc==SQLITE_BUSY ){ + /* If there is not a recovery running in another thread or process + ** then convert BUSY errors to WAL_RETRY. If recovery is known to + ** be running, convert BUSY to BUSY_RECOVERY. There is a race here + ** which might cause WAL_RETRY to be returned even if BUSY_RECOVERY + ** would be technically correct. But the race is benign since with + ** WAL_RETRY this routine will be called again and will probably be + ** right on the second iteration. + */ + if( pWal->apWiData[0]==0 ){ + /* This branch is taken when the xShmMap() method returns SQLITE_BUSY. + ** We assume this is a transient condition, so return WAL_RETRY. The + ** xShmMap() implementation used by the default unix and win32 VFS + ** modules may return SQLITE_BUSY due to a race condition in the + ** code that determines whether or not the shared-memory region + ** must be zeroed before the requested page is returned. + */ + rc = WAL_RETRY; + }else if( SQLITE_OK==(rc = walLockShared(pWal, WAL_RECOVER_LOCK)) ){ + walUnlockShared(pWal, WAL_RECOVER_LOCK); + rc = WAL_RETRY; + }else if( rc==SQLITE_BUSY ){ + rc = SQLITE_BUSY_RECOVERY; + } + } + if( rc!=SQLITE_OK ){ + return rc; + } + } + + pInfo = walCkptInfo(pWal); + if( !useWal && pInfo->nBackfill==pWal->hdr.mxFrame ){ + /* The WAL has been completely backfilled (or it is empty). + ** and can be safely ignored. + */ + rc = walLockShared(pWal, WAL_READ_LOCK(0)); + walShmBarrier(pWal); if( rc==SQLITE_OK ){ - pWal->writeLock = 1; - }else if( rc==SQLITE_BUSY ){ - eMode2 = SQLITE_CHECKPOINT_PASSIVE; - rc = SQLITE_OK; + if( memcmp((void *)walIndexHdr(pWal), &pWal->hdr, sizeof(WalIndexHdr)) ){ + /* It is not safe to allow the reader to continue here if frames + ** may have been appended to the log before READ_LOCK(0) was obtained. + ** When holding READ_LOCK(0), the reader ignores the entire log file, + ** which implies that the database file contains a trustworthy + ** snapshoT. Since holding READ_LOCK(0) prevents a checkpoint from + ** happening, this is usually correct. + ** + ** However, if frames have been appended to the log (or if the log + ** is wrapped and written for that matter) before the READ_LOCK(0) + ** is obtained, that is not necessarily true. A checkpointer may + ** have started to backfill the appended frames but crashed before + ** it finished. Leaving a corrupt image in the database file. + */ + walUnlockShared(pWal, WAL_READ_LOCK(0)); + return WAL_RETRY; + } + pWal->readLock = 0; + return SQLITE_OK; + }else if( rc!=SQLITE_BUSY ){ + return rc; } } - /* Read the wal-index header. */ - if( rc==SQLITE_OK ){ - rc = walIndexReadHdr(pWal, &isChanged); - if( isChanged && pWal->pDbFd->pMethods->iVersion>=3 ){ - sqlite3OsUnfetch(pWal->pDbFd, 0, 0); + /* If we get this far, it means that the reader will want to use + ** the WAL to get at content from recent commits. The job now is + ** to select one of the aReadMark[] entries that is closest to + ** but not exceeding pWal->hdr.mxFrame and lock that entry. + */ + mxReadMark = 0; + mxI = 0; + for(i=1; iaReadMark[i]; + if( mxReadMark<=thisMark && thisMark<=pWal->hdr.mxFrame ){ + assert( thisMark!=READMARK_NOT_USED ); + mxReadMark = thisMark; + mxI = i; } } + /* There was once an "if" here. The extra "{" is to preserve indentation. */ + { + if( (pWal->readOnly & WAL_SHM_RDONLY)==0 + && (mxReadMarkhdr.mxFrame || mxI==0) + ){ + for(i=1; iaReadMark[i] = pWal->hdr.mxFrame; + mxI = i; + walUnlockExclusive(pWal, WAL_READ_LOCK(i), 1); + break; + }else if( rc!=SQLITE_BUSY ){ + return rc; + } + } + } + if( mxI==0 ){ + assert( rc==SQLITE_BUSY || (pWal->readOnly & WAL_SHM_RDONLY)!=0 ); + return rc==SQLITE_BUSY ? WAL_RETRY : SQLITE_READONLY_CANTLOCK; + } - /* Copy data from the log to the database file. */ - if( rc==SQLITE_OK ){ - if( pWal->hdr.mxFrame && walPagesize(pWal)!=nBuf ){ - rc = SQLITE_CORRUPT_BKPT; + rc = walLockShared(pWal, WAL_READ_LOCK(mxI)); + if( rc ){ + return rc==SQLITE_BUSY ? WAL_RETRY : rc; + } + /* Now that the read-lock has been obtained, check that neither the + ** value in the aReadMark[] array or the contents of the wal-index + ** header have changed. + ** + ** It is necessary to check that the wal-index header did not change + ** between the time it was read and when the shared-lock was obtained + ** on WAL_READ_LOCK(mxI) was obtained to account for the possibility + ** that the log file may have been wrapped by a writer, or that frames + ** that occur later in the log than pWal->hdr.mxFrame may have been + ** copied into the database by a checkpointer. If either of these things + ** happened, then reading the database with the current value of + ** pWal->hdr.mxFrame risks reading a corrupted snapshot. So, retry + ** instead. + ** + ** This does not guarantee that the copy of the wal-index header is up to + ** date before proceeding. That would not be possible without somehow + ** blocking writers. It only guarantees that a dangerous checkpoint or + ** log-wrap (either of which would require an exclusive lock on + ** WAL_READ_LOCK(mxI)) has not occurred since the snapshot was valid. + */ + walShmBarrier(pWal); + if( pInfo->aReadMark[mxI]!=mxReadMark + || memcmp((void *)walIndexHdr(pWal), &pWal->hdr, sizeof(WalIndexHdr)) + ){ + walUnlockShared(pWal, WAL_READ_LOCK(mxI)); + return WAL_RETRY; }else{ - rc = walCheckpoint(pWal, eMode2, xBusy, pBusyArg, sync_flags, zBuf); - } - - /* If no error occurred, set the output variables. */ - if( rc==SQLITE_OK || rc==SQLITE_BUSY ){ - if( pnLog ) *pnLog = (int)pWal->hdr.mxFrame; - if( pnCkpt ) *pnCkpt = (int)(walCkptInfo(pWal)->nBackfill); + assert( mxReadMark<=pWal->hdr.mxFrame ); + pWal->readLock = (i16)mxI; } } + return rc; +} - if( isChanged ){ - /* If a new wal-index header was loaded before the checkpoint was - ** performed, then the pager-cache associated with pWal is now - ** out of date. So zero the cached wal-index header to ensure that - ** next time the pager opens a snapshot on this database it knows that - ** the cache needs to be reset. - */ - memset(&pWal->hdr, 0, sizeof(WalIndexHdr)); - } +/* +** Begin a read transaction on the database. +** +** This routine used to be called sqlite3OpenSnapshot() and with good reason: +** it takes a snapshot of the state of the WAL and wal-index for the current +** instant in time. The current thread will continue to use this snapshot. +** Other threads might append new content to the WAL and wal-index but +** that extra content is ignored by the current thread. +** +** If the database contents have changes since the previous read +** transaction, then *pChanged is set to 1 before returning. The +** Pager layer will use this to know that is cache is stale and +** needs to be flushed. +*/ +SQLITE_PRIVATE int sqlite3WalBeginReadTransaction(Wal *pWal, int *pChanged){ + int rc; /* Return code */ + int cnt = 0; /* Number of TryBeginRead attempts */ - /* Release the locks. */ - sqlite3WalEndWriteTransaction(pWal); - walUnlockExclusive(pWal, WAL_CKPT_LOCK, 1); - pWal->ckptLock = 0; - WALTRACE(("WAL%p: checkpoint %s\n", pWal, rc ? "failed" : "ok")); - return (rc==SQLITE_OK && eMode!=eMode2 ? SQLITE_BUSY : rc); + do{ + rc = walTryBeginRead(pWal, pChanged, 0, ++cnt); + }while( rc==WAL_RETRY ); + testcase( (rc&0xff)==SQLITE_BUSY ); + testcase( (rc&0xff)==SQLITE_IOERR ); + testcase( rc==SQLITE_PROTOCOL ); + testcase( rc==SQLITE_OK ); + return rc; } -/* Return the value to pass to a sqlite3_wal_hook callback, the -** number of frames in the WAL at the point of the last commit since -** sqlite3WalCallback() was called. If no commits have occurred since -** the last call, then return 0. +/* +** Finish with a read transaction. All this does is release the +** read-lock. */ -SQLITE_PRIVATE int sqlite3WalCallback(Wal *pWal){ - u32 ret = 0; - if( pWal ){ - ret = pWal->iCallback; - pWal->iCallback = 0; +SQLITE_PRIVATE void sqlite3WalEndReadTransaction(Wal *pWal){ + sqlite3WalEndWriteTransaction(pWal); + if( pWal->readLock>=0 ){ + walUnlockShared(pWal, WAL_READ_LOCK(pWal->readLock)); + pWal->readLock = -1; } - return (int)ret; } /* -** This function is called to change the WAL subsystem into or out -** of locking_mode=EXCLUSIVE. -** -** If op is zero, then attempt to change from locking_mode=EXCLUSIVE -** into locking_mode=NORMAL. This means that we must acquire a lock -** on the pWal->readLock byte. If the WAL is already in locking_mode=NORMAL -** or if the acquisition of the lock fails, then return 0. If the -** transition out of exclusive-mode is successful, return 1. This -** operation must occur while the pager is still holding the exclusive -** lock on the main database file. -** -** If op is one, then change from locking_mode=NORMAL into -** locking_mode=EXCLUSIVE. This means that the pWal->readLock must -** be released. Return 1 if the transition is made and 0 if the -** WAL is already in exclusive-locking mode - meaning that this -** routine is a no-op. The pager must already hold the exclusive lock -** on the main database file before invoking this operation. +** Search the wal file for page pgno. If found, set *piRead to the frame that +** contains the page. Otherwise, if pgno is not in the wal file, set *piRead +** to zero. ** -** If op is negative, then do a dry-run of the op==1 case but do -** not actually change anything. The pager uses this to see if it -** should acquire the database exclusive lock prior to invoking -** the op==1 case. +** Return SQLITE_OK if successful, or an error code if an error occurs. If an +** error does occur, the final value of *piRead is undefined. */ -SQLITE_PRIVATE int sqlite3WalExclusiveMode(Wal *pWal, int op){ - int rc; - assert( pWal->writeLock==0 ); - assert( pWal->exclusiveMode!=WAL_HEAPMEMORY_MODE || op==-1 ); +SQLITE_PRIVATE int sqlite3WalFindFrame( + Wal *pWal, /* WAL handle */ + Pgno pgno, /* Database page number to read data for */ + u32 *piRead /* OUT: Frame number (or zero) */ +){ + u32 iRead = 0; /* If !=0, WAL frame to return data from */ + u32 iLast = pWal->hdr.mxFrame; /* Last page in WAL for this reader */ + int iHash; /* Used to loop through N hash tables */ - /* pWal->readLock is usually set, but might be -1 if there was a - ** prior error while attempting to acquire are read-lock. This cannot - ** happen if the connection is actually in exclusive mode (as no xShmLock - ** locks are taken in this case). Nor should the pager attempt to - ** upgrade to exclusive-mode following such an error. - */ + /* This routine is only be called from within a read transaction. */ assert( pWal->readLock>=0 || pWal->lockError ); - assert( pWal->readLock>=0 || (op<=0 && pWal->exclusiveMode==0) ); - if( op==0 ){ - if( pWal->exclusiveMode ){ - pWal->exclusiveMode = 0; - if( walLockShared(pWal, WAL_READ_LOCK(pWal->readLock))!=SQLITE_OK ){ - pWal->exclusiveMode = 1; + /* If the "last page" field of the wal-index header snapshot is 0, then + ** no data will be read from the wal under any circumstances. Return early + ** in this case as an optimization. Likewise, if pWal->readLock==0, + ** then the WAL is ignored by the reader so return early, as if the + ** WAL were empty. + */ + if( iLast==0 || pWal->readLock==0 ){ + *piRead = 0; + return SQLITE_OK; + } + + /* Search the hash table or tables for an entry matching page number + ** pgno. Each iteration of the following for() loop searches one + ** hash table (each hash table indexes up to HASHTABLE_NPAGE frames). + ** + ** This code might run concurrently to the code in walIndexAppend() + ** that adds entries to the wal-index (and possibly to this hash + ** table). This means the value just read from the hash + ** slot (aHash[iKey]) may have been added before or after the + ** current read transaction was opened. Values added after the + ** read transaction was opened may have been written incorrectly - + ** i.e. these slots may contain garbage data. However, we assume + ** that any slots written before the current read transaction was + ** opened remain unmodified. + ** + ** For the reasons above, the if(...) condition featured in the inner + ** loop of the following block is more stringent that would be required + ** if we had exclusive access to the hash-table: + ** + ** (aPgno[iFrame]==pgno): + ** This condition filters out normal hash-table collisions. + ** + ** (iFrame<=iLast): + ** This condition filters out entries that were added to the hash + ** table after the current read-transaction had started. + */ + for(iHash=walFramePage(iLast); iHash>=0 && iRead==0; iHash--){ + volatile ht_slot *aHash; /* Pointer to hash table */ + volatile u32 *aPgno; /* Pointer to array of page numbers */ + u32 iZero; /* Frame number corresponding to aPgno[0] */ + int iKey; /* Hash slot index */ + int nCollide; /* Number of hash collisions remaining */ + int rc; /* Error code */ + + rc = walHashGet(pWal, iHash, &aHash, &aPgno, &iZero); + if( rc!=SQLITE_OK ){ + return rc; + } + nCollide = HASHTABLE_NSLOT; + for(iKey=walHash(pgno); aHash[iKey]; iKey=walNextHash(iKey)){ + u32 iFrame = aHash[iKey] + iZero; + if( iFrame<=iLast && aPgno[aHash[iKey]]==pgno ){ + /* assert( iFrame>iRead ); -- not true if there is corruption */ + iRead = iFrame; + } + if( (nCollide--)==0 ){ + return SQLITE_CORRUPT_BKPT; } - rc = pWal->exclusiveMode==0; - }else{ - /* Already in locking_mode=NORMAL */ - rc = 0; } - }else if( op>0 ){ - assert( pWal->exclusiveMode==0 ); - assert( pWal->readLock>=0 ); - walUnlockShared(pWal, WAL_READ_LOCK(pWal->readLock)); - pWal->exclusiveMode = 1; - rc = 1; - }else{ - rc = pWal->exclusiveMode==0; } - return rc; + +#ifdef SQLITE_ENABLE_EXPENSIVE_ASSERT + /* If expensive assert() statements are available, do a linear search + ** of the wal-index file content. Make sure the results agree with the + ** result obtained using the hash indexes above. */ + { + u32 iRead2 = 0; + u32 iTest; + for(iTest=iLast; iTest>0; iTest--){ + if( walFramePgno(pWal, iTest)==pgno ){ + iRead2 = iTest; + break; + } + } + assert( iRead==iRead2 ); + } +#endif + + *piRead = iRead; + return SQLITE_OK; } -/* -** Return true if the argument is non-NULL and the WAL module is using -** heap-memory for the wal-index. Otherwise, if the argument is NULL or the -** WAL module is using shared-memory, return false. +/* +** Read the contents of frame iRead from the wal file into buffer pOut +** (which is nOut bytes in size). Return SQLITE_OK if successful, or an +** error code otherwise. */ -SQLITE_PRIVATE int sqlite3WalHeapMemory(Wal *pWal){ - return (pWal && pWal->exclusiveMode==WAL_HEAPMEMORY_MODE ); +SQLITE_PRIVATE int sqlite3WalReadFrame( + Wal *pWal, /* WAL handle */ + u32 iRead, /* Frame to read */ + int nOut, /* Size of buffer pOut in bytes */ + u8 *pOut /* Buffer to write page data to */ +){ + int sz; + i64 iOffset; + sz = pWal->hdr.szPage; + sz = (sz&0xfe00) + ((sz&0x0001)<<16); + testcase( sz<=32768 ); + testcase( sz>=65536 ); + iOffset = walFrameOffset(iRead, sz) + WAL_FRAME_HDRSIZE; + /* testcase( IS_BIG_INT(iOffset) ); // requires a 4GiB WAL */ + return sqlite3OsRead(pWal->pWalFd, pOut, (nOut>sz ? sz : nOut), iOffset); } -#ifdef SQLITE_ENABLE_ZIPVFS -/* -** If the argument is not NULL, it points to a Wal object that holds a -** read-lock. This function returns the database page-size if it is known, -** or zero if it is not (or if pWal is NULL). +/* +** Return the size of the database in pages (or zero, if unknown). */ -SQLITE_PRIVATE int sqlite3WalFramesize(Wal *pWal){ - assert( pWal==0 || pWal->readLock>=0 ); - return (pWal ? pWal->szPage : 0); +SQLITE_PRIVATE Pgno sqlite3WalDbsize(Wal *pWal){ + if( pWal && ALWAYS(pWal->readLock>=0) ){ + return pWal->hdr.nPage; + } + return 0; } -#endif -#endif /* #ifndef SQLITE_OMIT_WAL */ -/************** End of wal.c *************************************************/ -/************** Begin file btmutex.c *****************************************/ -/* -** 2007 August 27 -** -** The author disclaims copyright to this source code. In place of -** a legal notice, here is a blessing: +/* +** This function starts a write transaction on the WAL. ** -** May you do good and not evil. -** May you find forgiveness for yourself and forgive others. -** May you share freely, never taking more than you give. +** A read transaction must have already been started by a prior call +** to sqlite3WalBeginReadTransaction(). ** -************************************************************************* +** If another thread or process has written into the database since +** the read transaction was started, then it is not possible for this +** thread to write as doing so would cause a fork. So this routine +** returns SQLITE_BUSY in that case and no write transaction is started. ** -** This file contains code used to implement mutexes on Btree objects. -** This code really belongs in btree.c. But btree.c is getting too -** big and we want to break it down some. This packaged seemed like -** a good breakout. +** There can only be a single writer active at a time. */ -/************** Include btreeInt.h in the middle of btmutex.c ****************/ -/************** Begin file btreeInt.h ****************************************/ -/* -** 2004 April 6 -** -** The author disclaims copyright to this source code. In place of -** a legal notice, here is a blessing: -** -** May you do good and not evil. -** May you find forgiveness for yourself and forgive others. -** May you share freely, never taking more than you give. -** -************************************************************************* -** This file implements a external (disk-based) database using BTrees. -** For a detailed discussion of BTrees, refer to -** -** Donald E. Knuth, THE ART OF COMPUTER PROGRAMMING, Volume 3: -** "Sorting And Searching", pages 473-480. Addison-Wesley -** Publishing Company, Reading, Massachusetts. -** -** The basic idea is that each page of the file contains N database -** entries and N+1 pointers to subpages. -** -** ---------------------------------------------------------------- -** | Ptr(0) | Key(0) | Ptr(1) | Key(1) | ... | Key(N-1) | Ptr(N) | -** ---------------------------------------------------------------- -** -** All of the keys on the page that Ptr(0) points to have values less -** than Key(0). All of the keys on page Ptr(1) and its subpages have -** values greater than Key(0) and less than Key(1). All of the keys -** on Ptr(N) and its subpages have values greater than Key(N-1). And -** so forth. -** -** Finding a particular key requires reading O(log(M)) pages from the -** disk where M is the number of entries in the tree. -** -** In this implementation, a single file can hold one or more separate -** BTrees. Each BTree is identified by the index of its root page. The -** key and data for any entry are combined to form the "payload". A -** fixed amount of payload can be carried directly on the database -** page. If the payload is larger than the preset amount then surplus -** bytes are stored on overflow pages. The payload for an entry -** and the preceding pointer are combined to form a "Cell". Each -** page has a small header which contains the Ptr(N) pointer and other -** information such as the size of key and data. -** -** FORMAT DETAILS -** -** The file is divided into pages. The first page is called page 1, -** the second is page 2, and so forth. A page number of zero indicates -** "no such page". The page size can be any power of 2 between 512 and 65536. -** Each page can be either a btree page, a freelist page, an overflow -** page, or a pointer-map page. -** -** The first page is always a btree page. The first 100 bytes of the first -** page contain a special header (the "file header") that describes the file. -** The format of the file header is as follows: -** -** OFFSET SIZE DESCRIPTION -** 0 16 Header string: "SQLite format 3\000" -** 16 2 Page size in bytes. -** 18 1 File format write version -** 19 1 File format read version -** 20 1 Bytes of unused space at the end of each page -** 21 1 Max embedded payload fraction -** 22 1 Min embedded payload fraction -** 23 1 Min leaf payload fraction -** 24 4 File change counter -** 28 4 Reserved for future use -** 32 4 First freelist page -** 36 4 Number of freelist pages in the file -** 40 60 15 4-byte meta values passed to higher layers -** -** 40 4 Schema cookie -** 44 4 File format of schema layer -** 48 4 Size of page cache -** 52 4 Largest root-page (auto/incr_vacuum) -** 56 4 1=UTF-8 2=UTF16le 3=UTF16be -** 60 4 User version -** 64 4 Incremental vacuum mode -** 68 4 unused -** 72 4 unused -** 76 4 unused -** -** All of the integer values are big-endian (most significant byte first). -** -** The file change counter is incremented when the database is changed -** This counter allows other processes to know when the file has changed -** and thus when they need to flush their cache. -** -** The max embedded payload fraction is the amount of the total usable -** space in a page that can be consumed by a single cell for standard -** B-tree (non-LEAFDATA) tables. A value of 255 means 100%. The default -** is to limit the maximum cell size so that at least 4 cells will fit -** on one page. Thus the default max embedded payload fraction is 64. -** -** If the payload for a cell is larger than the max payload, then extra -** payload is spilled to overflow pages. Once an overflow page is allocated, -** as many bytes as possible are moved into the overflow pages without letting -** the cell size drop below the min embedded payload fraction. -** -** The min leaf payload fraction is like the min embedded payload fraction -** except that it applies to leaf nodes in a LEAFDATA tree. The maximum -** payload fraction for a LEAFDATA tree is always 100% (or 255) and it -** not specified in the header. -** -** Each btree pages is divided into three sections: The header, the -** cell pointer array, and the cell content area. Page 1 also has a 100-byte -** file header that occurs before the page header. -** -** |----------------| -** | file header | 100 bytes. Page 1 only. -** |----------------| -** | page header | 8 bytes for leaves. 12 bytes for interior nodes -** |----------------| -** | cell pointer | | 2 bytes per cell. Sorted order. -** | array | | Grows downward -** | | v -** |----------------| -** | unallocated | -** | space | -** |----------------| ^ Grows upwards -** | cell content | | Arbitrary order interspersed with freeblocks. -** | area | | and free space fragments. -** |----------------| -** -** The page headers looks like this: -** -** OFFSET SIZE DESCRIPTION -** 0 1 Flags. 1: intkey, 2: zerodata, 4: leafdata, 8: leaf -** 1 2 byte offset to the first freeblock -** 3 2 number of cells on this page -** 5 2 first byte of the cell content area -** 7 1 number of fragmented free bytes -** 8 4 Right child (the Ptr(N) value). Omitted on leaves. -** -** The flags define the format of this btree page. The leaf flag means that -** this page has no children. The zerodata flag means that this page carries -** only keys and no data. The intkey flag means that the key is a integer -** which is stored in the key size entry of the cell header rather than in -** the payload area. -** -** The cell pointer array begins on the first byte after the page header. -** The cell pointer array contains zero or more 2-byte numbers which are -** offsets from the beginning of the page to the cell content in the cell -** content area. The cell pointers occur in sorted order. The system strives -** to keep free space after the last cell pointer so that new cells can -** be easily added without having to defragment the page. -** -** Cell content is stored at the very end of the page and grows toward the -** beginning of the page. -** -** Unused space within the cell content area is collected into a linked list of -** freeblocks. Each freeblock is at least 4 bytes in size. The byte offset -** to the first freeblock is given in the header. Freeblocks occur in -** increasing order. Because a freeblock must be at least 4 bytes in size, -** any group of 3 or fewer unused bytes in the cell content area cannot -** exist on the freeblock chain. A group of 3 or fewer free bytes is called -** a fragment. The total number of bytes in all fragments is recorded. -** in the page header at offset 7. -** -** SIZE DESCRIPTION -** 2 Byte offset of the next freeblock -** 2 Bytes in this freeblock -** -** Cells are of variable length. Cells are stored in the cell content area at -** the end of the page. Pointers to the cells are in the cell pointer array -** that immediately follows the page header. Cells is not necessarily -** contiguous or in order, but cell pointers are contiguous and in order. -** -** Cell content makes use of variable length integers. A variable -** length integer is 1 to 9 bytes where the lower 7 bits of each -** byte are used. The integer consists of all bytes that have bit 8 set and -** the first byte with bit 8 clear. The most significant byte of the integer -** appears first. A variable-length integer may not be more than 9 bytes long. -** As a special case, all 8 bytes of the 9th byte are used as data. This -** allows a 64-bit integer to be encoded in 9 bytes. -** -** 0x00 becomes 0x00000000 -** 0x7f becomes 0x0000007f -** 0x81 0x00 becomes 0x00000080 -** 0x82 0x00 becomes 0x00000100 -** 0x80 0x7f becomes 0x0000007f -** 0x8a 0x91 0xd1 0xac 0x78 becomes 0x12345678 -** 0x81 0x81 0x81 0x81 0x01 becomes 0x10204081 -** -** Variable length integers are used for rowids and to hold the number of -** bytes of key and data in a btree cell. -** -** The content of a cell looks like this: -** -** SIZE DESCRIPTION -** 4 Page number of the left child. Omitted if leaf flag is set. -** var Number of bytes of data. Omitted if the zerodata flag is set. -** var Number of bytes of key. Or the key itself if intkey flag is set. -** * Payload -** 4 First page of the overflow chain. Omitted if no overflow -** -** Overflow pages form a linked list. Each page except the last is completely -** filled with data (pagesize - 4 bytes). The last page can have as little -** as 1 byte of data. -** -** SIZE DESCRIPTION -** 4 Page number of next overflow page -** * Data +SQLITE_PRIVATE int sqlite3WalBeginWriteTransaction(Wal *pWal){ + int rc; + + /* Cannot start a write transaction without first holding a read + ** transaction. */ + assert( pWal->readLock>=0 ); + + if( pWal->readOnly ){ + return SQLITE_READONLY; + } + + /* Only one writer allowed at a time. Get the write lock. Return + ** SQLITE_BUSY if unable. + */ + rc = walLockExclusive(pWal, WAL_WRITE_LOCK, 1); + if( rc ){ + return rc; + } + pWal->writeLock = 1; + + /* If another connection has written to the database file since the + ** time the read transaction on this connection was started, then + ** the write is disallowed. + */ + if( memcmp(&pWal->hdr, (void *)walIndexHdr(pWal), sizeof(WalIndexHdr))!=0 ){ + walUnlockExclusive(pWal, WAL_WRITE_LOCK, 1); + pWal->writeLock = 0; + rc = SQLITE_BUSY_SNAPSHOT; + } + + return rc; +} + +/* +** End a write transaction. The commit has already been done. This +** routine merely releases the lock. +*/ +SQLITE_PRIVATE int sqlite3WalEndWriteTransaction(Wal *pWal){ + if( pWal->writeLock ){ + walUnlockExclusive(pWal, WAL_WRITE_LOCK, 1); + pWal->writeLock = 0; + pWal->truncateOnCommit = 0; + } + return SQLITE_OK; +} + +/* +** If any data has been written (but not committed) to the log file, this +** function moves the write-pointer back to the start of the transaction. ** -** Freelist pages come in two subtypes: trunk pages and leaf pages. The -** file header points to the first in a linked list of trunk page. Each trunk -** page points to multiple leaf pages. The content of a leaf page is -** unspecified. A trunk page looks like this: +** Additionally, the callback function is invoked for each frame written +** to the WAL since the start of the transaction. If the callback returns +** other than SQLITE_OK, it is not invoked again and the error code is +** returned to the caller. ** -** SIZE DESCRIPTION -** 4 Page number of next trunk page -** 4 Number of leaf pointers on this page -** * zero or more pages numbers of leaves +** Otherwise, if the callback function does not return an error, this +** function returns SQLITE_OK. */ +SQLITE_PRIVATE int sqlite3WalUndo(Wal *pWal, int (*xUndo)(void *, Pgno), void *pUndoCtx){ + int rc = SQLITE_OK; + if( ALWAYS(pWal->writeLock) ){ + Pgno iMax = pWal->hdr.mxFrame; + Pgno iFrame; + + /* Restore the clients cache of the wal-index header to the state it + ** was in before the client began writing to the database. + */ + memcpy(&pWal->hdr, (void *)walIndexHdr(pWal), sizeof(WalIndexHdr)); + for(iFrame=pWal->hdr.mxFrame+1; + ALWAYS(rc==SQLITE_OK) && iFrame<=iMax; + iFrame++ + ){ + /* This call cannot fail. Unless the page for which the page number + ** is passed as the second argument is (a) in the cache and + ** (b) has an outstanding reference, then xUndo is either a no-op + ** (if (a) is false) or simply expels the page from the cache (if (b) + ** is false). + ** + ** If the upper layer is doing a rollback, it is guaranteed that there + ** are no outstanding references to any page other than page 1. And + ** page 1 is never written to the log until the transaction is + ** committed. As a result, the call to xUndo may not fail. + */ + assert( walFramePgno(pWal, iFrame)!=1 ); + rc = xUndo(pUndoCtx, walFramePgno(pWal, iFrame)); + } + if( iMax!=pWal->hdr.mxFrame ) walCleanupHash(pWal); + } + assert( rc==SQLITE_OK ); + return rc; +} -/* The following value is the maximum cell size assuming a maximum page -** size give above. +/* +** Argument aWalData must point to an array of WAL_SAVEPOINT_NDATA u32 +** values. This function populates the array with values required to +** "rollback" the write position of the WAL handle back to the current +** point in the event of a savepoint rollback (via WalSavepointUndo()). */ -#define MX_CELL_SIZE(pBt) ((int)(pBt->pageSize-8)) +SQLITE_PRIVATE void sqlite3WalSavepoint(Wal *pWal, u32 *aWalData){ + assert( pWal->writeLock ); + aWalData[0] = pWal->hdr.mxFrame; + aWalData[1] = pWal->hdr.aFrameCksum[0]; + aWalData[2] = pWal->hdr.aFrameCksum[1]; + aWalData[3] = pWal->nCkpt; +} -/* The maximum number of cells on a single page of the database. This -** assumes a minimum cell size of 6 bytes (4 bytes for the cell itself -** plus 2 bytes for the index to the cell in the page header). Such -** small cells will be rare, but they are possible. +/* +** Move the write position of the WAL back to the point identified by +** the values in the aWalData[] array. aWalData must point to an array +** of WAL_SAVEPOINT_NDATA u32 values that has been previously populated +** by a call to WalSavepoint(). */ -#define MX_CELL(pBt) ((pBt->pageSize-8)/6) +SQLITE_PRIVATE int sqlite3WalSavepointUndo(Wal *pWal, u32 *aWalData){ + int rc = SQLITE_OK; + + assert( pWal->writeLock ); + assert( aWalData[3]!=pWal->nCkpt || aWalData[0]<=pWal->hdr.mxFrame ); + + if( aWalData[3]!=pWal->nCkpt ){ + /* This savepoint was opened immediately after the write-transaction + ** was started. Right after that, the writer decided to wrap around + ** to the start of the log. Update the savepoint values to match. + */ + aWalData[0] = 0; + aWalData[3] = pWal->nCkpt; + } + + if( aWalData[0]hdr.mxFrame ){ + pWal->hdr.mxFrame = aWalData[0]; + pWal->hdr.aFrameCksum[0] = aWalData[1]; + pWal->hdr.aFrameCksum[1] = aWalData[2]; + walCleanupHash(pWal); + } + + return rc; +} -/* Forward declarations */ -typedef struct MemPage MemPage; -typedef struct BtLock BtLock; /* -** This is a magic string that appears at the beginning of every -** SQLite database in order to identify the file as a real database. +** This function is called just before writing a set of frames to the log +** file (see sqlite3WalFrames()). It checks to see if, instead of appending +** to the current log file, it is possible to overwrite the start of the +** existing log file with the new frames (i.e. "reset" the log). If so, +** it sets pWal->hdr.mxFrame to 0. Otherwise, pWal->hdr.mxFrame is left +** unchanged. ** -** You can change this value at compile-time by specifying a -** -DSQLITE_FILE_HEADER="..." on the compiler command-line. The -** header must be exactly 16 bytes including the zero-terminator so -** the string itself should be 15 characters long. If you change -** the header, then your custom library will not be able to read -** databases generated by the standard tools and the standard tools -** will not be able to read databases created by your custom library. +** SQLITE_OK is returned if no error is encountered (regardless of whether +** or not pWal->hdr.mxFrame is modified). An SQLite error code is returned +** if an error occurs. */ -#ifndef SQLITE_FILE_HEADER /* 123456789 123456 */ -# define SQLITE_FILE_HEADER "SQLite format 3" -#endif +static int walRestartLog(Wal *pWal){ + int rc = SQLITE_OK; + int cnt; + + if( pWal->readLock==0 ){ + volatile WalCkptInfo *pInfo = walCkptInfo(pWal); + assert( pInfo->nBackfill==pWal->hdr.mxFrame ); + if( pInfo->nBackfill>0 ){ + u32 salt1; + sqlite3_randomness(4, &salt1); + rc = walLockExclusive(pWal, WAL_READ_LOCK(1), WAL_NREADER-1); + if( rc==SQLITE_OK ){ + /* If all readers are using WAL_READ_LOCK(0) (in other words if no + ** readers are currently using the WAL), then the transactions + ** frames will overwrite the start of the existing log. Update the + ** wal-index header to reflect this. + ** + ** In theory it would be Ok to update the cache of the header only + ** at this point. But updating the actual wal-index header is also + ** safe and means there is no special case for sqlite3WalUndo() + ** to handle if this transaction is rolled back. + */ + int i; /* Loop counter */ + u32 *aSalt = pWal->hdr.aSalt; /* Big-endian salt values */ + + pWal->nCkpt++; + pWal->hdr.mxFrame = 0; + sqlite3Put4byte((u8*)&aSalt[0], 1 + sqlite3Get4byte((u8*)&aSalt[0])); + aSalt[1] = salt1; + walIndexWriteHdr(pWal); + pInfo->nBackfill = 0; + pInfo->aReadMark[1] = 0; + for(i=2; iaReadMark[i] = READMARK_NOT_USED; + assert( pInfo->aReadMark[0]==0 ); + walUnlockExclusive(pWal, WAL_READ_LOCK(1), WAL_NREADER-1); + }else if( rc!=SQLITE_BUSY ){ + return rc; + } + } + walUnlockShared(pWal, WAL_READ_LOCK(0)); + pWal->readLock = -1; + cnt = 0; + do{ + int notUsed; + rc = walTryBeginRead(pWal, ¬Used, 1, ++cnt); + }while( rc==WAL_RETRY ); + assert( (rc&0xff)!=SQLITE_BUSY ); /* BUSY not possible when useWal==1 */ + testcase( (rc&0xff)==SQLITE_IOERR ); + testcase( rc==SQLITE_PROTOCOL ); + testcase( rc==SQLITE_OK ); + } + return rc; +} /* -** Page type flags. An ORed combination of these flags appear as the -** first byte of on-disk image of every BTree page. +** Information about the current state of the WAL file and where +** the next fsync should occur - passed from sqlite3WalFrames() into +** walWriteToLog(). */ -#define PTF_INTKEY 0x01 -#define PTF_ZERODATA 0x02 -#define PTF_LEAFDATA 0x04 -#define PTF_LEAF 0x08 +typedef struct WalWriter { + Wal *pWal; /* The complete WAL information */ + sqlite3_file *pFd; /* The WAL file to which we write */ + sqlite3_int64 iSyncPoint; /* Fsync at this offset */ + int syncFlags; /* Flags for the fsync */ + int szPage; /* Size of one page */ +} WalWriter; /* -** As each page of the file is loaded into memory, an instance of the following -** structure is appended and initialized to zero. This structure stores -** information about the page that is decoded from the raw file page. -** -** The pParent field points back to the parent page. This allows us to -** walk up the BTree from any leaf to the root. Care must be taken to -** unref() the parent page pointer when this page is no longer referenced. -** The pageDestructor() routine handles that chore. +** Write iAmt bytes of content into the WAL file beginning at iOffset. +** Do a sync when crossing the p->iSyncPoint boundary. ** -** Access to all fields of this structure is controlled by the mutex -** stored in MemPage.pBt->mutex. +** In other words, if iSyncPoint is in between iOffset and iOffset+iAmt, +** first write the part before iSyncPoint, then sync, then write the +** rest. */ -struct MemPage { - u8 isInit; /* True if previously initialized. MUST BE FIRST! */ - u8 nOverflow; /* Number of overflow cell bodies in aCell[] */ - u8 intKey; /* True if intkey flag is set */ - u8 leaf; /* True if leaf flag is set */ - u8 hasData; /* True if this page stores data */ - u8 hdrOffset; /* 100 for page 1. 0 otherwise */ - u8 childPtrSize; /* 0 if leaf==1. 4 if leaf==0 */ - u8 max1bytePayload; /* min(maxLocal,127) */ - u16 maxLocal; /* Copy of BtShared.maxLocal or BtShared.maxLeaf */ - u16 minLocal; /* Copy of BtShared.minLocal or BtShared.minLeaf */ - u16 cellOffset; /* Index in aData of first cell pointer */ - u16 nFree; /* Number of free bytes on the page */ - u16 nCell; /* Number of cells on this page, local and ovfl */ - u16 maskPage; /* Mask for page offset */ - u16 aiOvfl[5]; /* Insert the i-th overflow cell before the aiOvfl-th - ** non-overflow cell */ - u8 *apOvfl[5]; /* Pointers to the body of overflow cells */ - BtShared *pBt; /* Pointer to BtShared that this page is part of */ - u8 *aData; /* Pointer to disk image of the page data */ - u8 *aDataEnd; /* One byte past the end of usable data */ - u8 *aCellIdx; /* The cell index area */ - DbPage *pDbPage; /* Pager page handle */ - Pgno pgno; /* Page number for this page */ -}; +static int walWriteToLog( + WalWriter *p, /* WAL to write to */ + void *pContent, /* Content to be written */ + int iAmt, /* Number of bytes to write */ + sqlite3_int64 iOffset /* Start writing at this offset */ +){ + int rc; + if( iOffsetiSyncPoint && iOffset+iAmt>=p->iSyncPoint ){ + int iFirstAmt = (int)(p->iSyncPoint - iOffset); + rc = sqlite3OsWrite(p->pFd, pContent, iFirstAmt, iOffset); + if( rc ) return rc; + iOffset += iFirstAmt; + iAmt -= iFirstAmt; + pContent = (void*)(iFirstAmt + (char*)pContent); + assert( p->syncFlags & (SQLITE_SYNC_NORMAL|SQLITE_SYNC_FULL) ); + rc = sqlite3OsSync(p->pFd, p->syncFlags); + if( iAmt==0 || rc ) return rc; + } + rc = sqlite3OsWrite(p->pFd, pContent, iAmt, iOffset); + return rc; +} /* -** The in-memory image of a disk page has the auxiliary information appended -** to the end. EXTRA_SIZE is the number of bytes of space needed to hold -** that extra information. +** Write out a single frame of the WAL */ -#define EXTRA_SIZE sizeof(MemPage) +static int walWriteOneFrame( + WalWriter *p, /* Where to write the frame */ + PgHdr *pPage, /* The page of the frame to be written */ + int nTruncate, /* The commit flag. Usually 0. >0 for commit */ + sqlite3_int64 iOffset /* Byte offset at which to write */ +){ + int rc; /* Result code from subfunctions */ + void *pData; /* Data actually written */ + u8 aFrame[WAL_FRAME_HDRSIZE]; /* Buffer to assemble frame-header in */ +#if defined(SQLITE_HAS_CODEC) + if( (pData = sqlite3PagerCodec(pPage))==0 ) return SQLITE_NOMEM; +#else + pData = pPage->pData; +#endif + walEncodeFrame(p->pWal, pPage->pgno, nTruncate, pData, aFrame); + rc = walWriteToLog(p, aFrame, sizeof(aFrame), iOffset); + if( rc ) return rc; + /* Write the page data */ + rc = walWriteToLog(p, pData, p->szPage, iOffset+sizeof(aFrame)); + return rc; +} -/* -** A linked list of the following structures is stored at BtShared.pLock. -** Locks are added (or upgraded from READ_LOCK to WRITE_LOCK) when a cursor -** is opened on the table with root page BtShared.iTable. Locks are removed -** from this list when a transaction is committed or rolled back, or when -** a btree handle is closed. +/* +** Write a set of frames to the log. The caller must hold the write-lock +** on the log file (obtained using sqlite3WalBeginWriteTransaction()). */ -struct BtLock { - Btree *pBtree; /* Btree handle holding this lock */ - Pgno iTable; /* Root page of table */ - u8 eLock; /* READ_LOCK or WRITE_LOCK */ - BtLock *pNext; /* Next in BtShared.pLock list */ -}; +SQLITE_PRIVATE int sqlite3WalFrames( + Wal *pWal, /* Wal handle to write to */ + int szPage, /* Database page-size in bytes */ + PgHdr *pList, /* List of dirty pages to write */ + Pgno nTruncate, /* Database size after this commit */ + int isCommit, /* True if this is a commit */ + int sync_flags /* Flags to pass to OsSync() (or 0) */ +){ + int rc; /* Used to catch return codes */ + u32 iFrame; /* Next frame address */ + PgHdr *p; /* Iterator to run through pList with. */ + PgHdr *pLast = 0; /* Last frame in list */ + int nExtra = 0; /* Number of extra copies of last page */ + int szFrame; /* The size of a single frame */ + i64 iOffset; /* Next byte to write in WAL file */ + WalWriter w; /* The writer */ -/* Candidate values for BtLock.eLock */ -#define READ_LOCK 1 -#define WRITE_LOCK 2 + assert( pList ); + assert( pWal->writeLock ); -/* A Btree handle -** -** A database connection contains a pointer to an instance of -** this object for every database file that it has open. This structure -** is opaque to the database connection. The database connection cannot -** see the internals of this structure and only deals with pointers to -** this structure. -** -** For some database files, the same underlying database cache might be -** shared between multiple connections. In that case, each connection -** has it own instance of this object. But each instance of this object -** points to the same BtShared object. The database cache and the -** schema associated with the database file are all contained within -** the BtShared object. -** -** All fields in this structure are accessed under sqlite3.mutex. -** The pBt pointer itself may not be changed while there exists cursors -** in the referenced BtShared that point back to this Btree since those -** cursors have to go through this Btree to find their BtShared and -** they often do so without holding sqlite3.mutex. -*/ -struct Btree { - sqlite3 *db; /* The database connection holding this btree */ - BtShared *pBt; /* Sharable content of this btree */ - u8 inTrans; /* TRANS_NONE, TRANS_READ or TRANS_WRITE */ - u8 sharable; /* True if we can share pBt with another db */ - u8 locked; /* True if db currently has pBt locked */ - int wantToLock; /* Number of nested calls to sqlite3BtreeEnter() */ - int nBackup; /* Number of backup operations reading this btree */ - Btree *pNext; /* List of other sharable Btrees from the same db */ - Btree *pPrev; /* Back pointer of the same list */ -#ifndef SQLITE_OMIT_SHARED_CACHE - BtLock lock; /* Object used to lock page 1 */ + /* If this frame set completes a transaction, then nTruncate>0. If + ** nTruncate==0 then this frame set does not complete the transaction. */ + assert( (isCommit!=0)==(nTruncate!=0) ); + +#if defined(SQLITE_TEST) && defined(SQLITE_DEBUG) + { int cnt; for(cnt=0, p=pList; p; p=p->pDirty, cnt++){} + WALTRACE(("WAL%p: frame write begin. %d frames. mxFrame=%d. %s\n", + pWal, cnt, pWal->hdr.mxFrame, isCommit ? "Commit" : "Spill")); + } #endif -}; -/* -** Btree.inTrans may take one of the following values. -** -** If the shared-data extension is enabled, there may be multiple users -** of the Btree structure. At most one of these may open a write transaction, -** but any number may have active read transactions. -*/ -#define TRANS_NONE 0 -#define TRANS_READ 1 -#define TRANS_WRITE 2 + /* See if it is possible to write these frames into the start of the + ** log file, instead of appending to it at pWal->hdr.mxFrame. + */ + if( SQLITE_OK!=(rc = walRestartLog(pWal)) ){ + return rc; + } + + /* If this is the first frame written into the log, write the WAL + ** header to the start of the WAL file. See comments at the top of + ** this source file for a description of the WAL header format. + */ + iFrame = pWal->hdr.mxFrame; + if( iFrame==0 ){ + u8 aWalHdr[WAL_HDRSIZE]; /* Buffer to assemble wal-header in */ + u32 aCksum[2]; /* Checksum for wal-header */ + + sqlite3Put4byte(&aWalHdr[0], (WAL_MAGIC | SQLITE_BIGENDIAN)); + sqlite3Put4byte(&aWalHdr[4], WAL_MAX_VERSION); + sqlite3Put4byte(&aWalHdr[8], szPage); + sqlite3Put4byte(&aWalHdr[12], pWal->nCkpt); + if( pWal->nCkpt==0 ) sqlite3_randomness(8, pWal->hdr.aSalt); + memcpy(&aWalHdr[16], pWal->hdr.aSalt, 8); + walChecksumBytes(1, aWalHdr, WAL_HDRSIZE-2*4, 0, aCksum); + sqlite3Put4byte(&aWalHdr[24], aCksum[0]); + sqlite3Put4byte(&aWalHdr[28], aCksum[1]); + + pWal->szPage = szPage; + pWal->hdr.bigEndCksum = SQLITE_BIGENDIAN; + pWal->hdr.aFrameCksum[0] = aCksum[0]; + pWal->hdr.aFrameCksum[1] = aCksum[1]; + pWal->truncateOnCommit = 1; + + rc = sqlite3OsWrite(pWal->pWalFd, aWalHdr, sizeof(aWalHdr), 0); + WALTRACE(("WAL%p: wal-header write %s\n", pWal, rc ? "failed" : "ok")); + if( rc!=SQLITE_OK ){ + return rc; + } + + /* Sync the header (unless SQLITE_IOCAP_SEQUENTIAL is true or unless + ** all syncing is turned off by PRAGMA synchronous=OFF). Otherwise + ** an out-of-order write following a WAL restart could result in + ** database corruption. See the ticket: + ** + ** http://localhost:591/sqlite/info/ff5be73dee + */ + if( pWal->syncHeader && sync_flags ){ + rc = sqlite3OsSync(pWal->pWalFd, sync_flags & SQLITE_SYNC_MASK); + if( rc ) return rc; + } + } + assert( (int)pWal->szPage==szPage ); + + /* Setup information needed to write frames into the WAL */ + w.pWal = pWal; + w.pFd = pWal->pWalFd; + w.iSyncPoint = 0; + w.syncFlags = sync_flags; + w.szPage = szPage; + iOffset = walFrameOffset(iFrame+1, szPage); + szFrame = szPage + WAL_FRAME_HDRSIZE; + + /* Write all frames into the log file exactly once */ + for(p=pList; p; p=p->pDirty){ + int nDbSize; /* 0 normally. Positive == commit flag */ + iFrame++; + assert( iOffset==walFrameOffset(iFrame, szPage) ); + nDbSize = (isCommit && p->pDirty==0) ? nTruncate : 0; + rc = walWriteOneFrame(&w, p, nDbSize, iOffset); + if( rc ) return rc; + pLast = p; + iOffset += szFrame; + } + + /* If this is the end of a transaction, then we might need to pad + ** the transaction and/or sync the WAL file. + ** + ** Padding and syncing only occur if this set of frames complete a + ** transaction and if PRAGMA synchronous=FULL. If synchronous==NORMAL + ** or synchonous==OFF, then no padding or syncing are needed. + ** + ** If SQLITE_IOCAP_POWERSAFE_OVERWRITE is defined, then padding is not + ** needed and only the sync is done. If padding is needed, then the + ** final frame is repeated (with its commit mark) until the next sector + ** boundary is crossed. Only the part of the WAL prior to the last + ** sector boundary is synced; the part of the last frame that extends + ** past the sector boundary is written after the sync. + */ + if( isCommit && (sync_flags & WAL_SYNC_TRANSACTIONS)!=0 ){ + if( pWal->padToSectorBoundary ){ + int sectorSize = sqlite3SectorSize(pWal->pWalFd); + w.iSyncPoint = ((iOffset+sectorSize-1)/sectorSize)*sectorSize; + while( iOffsettruncateOnCommit && pWal->mxWalSize>=0 ){ + i64 sz = pWal->mxWalSize; + if( walFrameOffset(iFrame+nExtra+1, szPage)>pWal->mxWalSize ){ + sz = walFrameOffset(iFrame+nExtra+1, szPage); + } + walLimitSize(pWal, sz); + pWal->truncateOnCommit = 0; + } + + /* Append data to the wal-index. It is not necessary to lock the + ** wal-index to do this as the SQLITE_SHM_WRITE lock held on the wal-index + ** guarantees that there are no other writers, and no data that may + ** be in use by existing readers is being overwritten. + */ + iFrame = pWal->hdr.mxFrame; + for(p=pList; p && rc==SQLITE_OK; p=p->pDirty){ + iFrame++; + rc = walIndexAppend(pWal, iFrame, p->pgno); + } + while( rc==SQLITE_OK && nExtra>0 ){ + iFrame++; + nExtra--; + rc = walIndexAppend(pWal, iFrame, pLast->pgno); + } + + if( rc==SQLITE_OK ){ + /* Update the private copy of the header. */ + pWal->hdr.szPage = (u16)((szPage&0xff00) | (szPage>>16)); + testcase( szPage<=32768 ); + testcase( szPage>=65536 ); + pWal->hdr.mxFrame = iFrame; + if( isCommit ){ + pWal->hdr.iChange++; + pWal->hdr.nPage = nTruncate; + } + /* If this is a commit, update the wal-index header too. */ + if( isCommit ){ + walIndexWriteHdr(pWal); + pWal->iCallback = iFrame; + } + } -/* -** An instance of this object represents a single database file. -** -** A single database file can be in use at the same time by two -** or more database connections. When two or more connections are -** sharing the same database file, each connection has it own -** private Btree object for the file and each of those Btrees points -** to this one BtShared object. BtShared.nRef is the number of -** connections currently sharing this database file. -** -** Fields in this structure are accessed under the BtShared.mutex -** mutex, except for nRef and pNext which are accessed under the -** global SQLITE_MUTEX_STATIC_MASTER mutex. The pPager field -** may not be modified once it is initially set as long as nRef>0. -** The pSchema field may be set once under BtShared.mutex and -** thereafter is unchanged as long as nRef>0. -** -** isPending: -** -** If a BtShared client fails to obtain a write-lock on a database -** table (because there exists one or more read-locks on the table), -** the shared-cache enters 'pending-lock' state and isPending is -** set to true. -** -** The shared-cache leaves the 'pending lock' state when either of -** the following occur: -** -** 1) The current writer (BtShared.pWriter) concludes its transaction, OR -** 2) The number of locks held by other connections drops to zero. + WALTRACE(("WAL%p: frame write %s\n", pWal, rc ? "failed" : "ok")); + return rc; +} + +/* +** This routine is called to implement sqlite3_wal_checkpoint() and +** related interfaces. ** -** while in the 'pending-lock' state, no connection may start a new -** transaction. +** Obtain a CHECKPOINT lock and then backfill as much information as +** we can from WAL into the database. ** -** This feature is included to help prevent writer-starvation. +** If parameter xBusy is not NULL, it is a pointer to a busy-handler +** callback. In this case this function runs a blocking checkpoint. */ -struct BtShared { - Pager *pPager; /* The page cache */ - sqlite3 *db; /* Database connection currently using this Btree */ - BtCursor *pCursor; /* A list of all open cursors */ - MemPage *pPage1; /* First page of the database */ - u8 openFlags; /* Flags to sqlite3BtreeOpen() */ -#ifndef SQLITE_OMIT_AUTOVACUUM - u8 autoVacuum; /* True if auto-vacuum is enabled */ - u8 incrVacuum; /* True if incr-vacuum is enabled */ - u8 bDoTruncate; /* True to truncate db on commit */ -#endif - u8 inTransaction; /* Transaction state */ - u8 max1bytePayload; /* Maximum first byte of cell for a 1-byte payload */ - u16 btsFlags; /* Boolean parameters. See BTS_* macros below */ - u16 maxLocal; /* Maximum local payload in non-LEAFDATA tables */ - u16 minLocal; /* Minimum local payload in non-LEAFDATA tables */ - u16 maxLeaf; /* Maximum local payload in a LEAFDATA table */ - u16 minLeaf; /* Minimum local payload in a LEAFDATA table */ - u32 pageSize; /* Total number of bytes on a page */ - u32 usableSize; /* Number of usable bytes on each page */ - int nTransaction; /* Number of open transactions (read + write) */ - u32 nPage; /* Number of pages in the database */ - void *pSchema; /* Pointer to space allocated by sqlite3BtreeSchema() */ - void (*xFreeSchema)(void*); /* Destructor for BtShared.pSchema */ - sqlite3_mutex *mutex; /* Non-recursive mutex required to access this object */ - Bitvec *pHasContent; /* Set of pages moved to free-list this transaction */ -#ifndef SQLITE_OMIT_SHARED_CACHE - int nRef; /* Number of references to this structure */ - BtShared *pNext; /* Next on a list of sharable BtShared structs */ - BtLock *pLock; /* List of locks held on this shared-btree struct */ - Btree *pWriter; /* Btree with currently open write transaction */ -#endif - u8 *pTmpSpace; /* BtShared.pageSize bytes of space for tmp use */ -}; +SQLITE_PRIVATE int sqlite3WalCheckpoint( + Wal *pWal, /* Wal connection */ + int eMode, /* PASSIVE, FULL or RESTART */ + int (*xBusy)(void*), /* Function to call when busy */ + void *pBusyArg, /* Context argument for xBusyHandler */ + int sync_flags, /* Flags to sync db file with (or 0) */ + int nBuf, /* Size of temporary buffer */ + u8 *zBuf, /* Temporary buffer to use */ + int *pnLog, /* OUT: Number of frames in WAL */ + int *pnCkpt /* OUT: Number of backfilled frames in WAL */ +){ + int rc; /* Return code */ + int isChanged = 0; /* True if a new wal-index header is loaded */ + int eMode2 = eMode; /* Mode to pass to walCheckpoint() */ -/* -** Allowed values for BtShared.btsFlags -*/ -#define BTS_READ_ONLY 0x0001 /* Underlying file is readonly */ -#define BTS_PAGESIZE_FIXED 0x0002 /* Page size can no longer be changed */ -#define BTS_SECURE_DELETE 0x0004 /* PRAGMA secure_delete is enabled */ -#define BTS_INITIALLY_EMPTY 0x0008 /* Database was empty at trans start */ -#define BTS_NO_WAL 0x0010 /* Do not open write-ahead-log files */ -#define BTS_EXCLUSIVE 0x0020 /* pWriter has an exclusive lock */ -#define BTS_PENDING 0x0040 /* Waiting for read-locks to clear */ + assert( pWal->ckptLock==0 ); + assert( pWal->writeLock==0 ); -/* -** An instance of the following structure is used to hold information -** about a cell. The parseCellPtr() function fills in this structure -** based on information extract from the raw disk page. -*/ -typedef struct CellInfo CellInfo; -struct CellInfo { - i64 nKey; /* The key for INTKEY tables, or number of bytes in key */ - u8 *pCell; /* Pointer to the start of cell content */ - u32 nData; /* Number of bytes of data */ - u32 nPayload; /* Total amount of payload */ - u16 nHeader; /* Size of the cell content header in bytes */ - u16 nLocal; /* Amount of payload held locally */ - u16 iOverflow; /* Offset to overflow page number. Zero if no overflow */ - u16 nSize; /* Size of the cell content on the main b-tree page */ -}; + if( pWal->readOnly ) return SQLITE_READONLY; + WALTRACE(("WAL%p: checkpoint begins\n", pWal)); + rc = walLockExclusive(pWal, WAL_CKPT_LOCK, 1); + if( rc ){ + /* Usually this is SQLITE_BUSY meaning that another thread or process + ** is already running a checkpoint, or maybe a recovery. But it might + ** also be SQLITE_IOERR. */ + return rc; + } + pWal->ckptLock = 1; -/* -** Maximum depth of an SQLite B-Tree structure. Any B-Tree deeper than -** this will be declared corrupt. This value is calculated based on a -** maximum database size of 2^31 pages a minimum fanout of 2 for a -** root-node and 3 for all other internal nodes. -** -** If a tree that appears to be taller than this is encountered, it is -** assumed that the database is corrupt. -*/ -#define BTCURSOR_MAX_DEPTH 20 + /* If this is a blocking-checkpoint, then obtain the write-lock as well + ** to prevent any writers from running while the checkpoint is underway. + ** This has to be done before the call to walIndexReadHdr() below. + ** + ** If the writer lock cannot be obtained, then a passive checkpoint is + ** run instead. Since the checkpointer is not holding the writer lock, + ** there is no point in blocking waiting for any readers. Assuming no + ** other error occurs, this function will return SQLITE_BUSY to the caller. + */ + if( eMode!=SQLITE_CHECKPOINT_PASSIVE ){ + rc = walBusyLock(pWal, xBusy, pBusyArg, WAL_WRITE_LOCK, 1); + if( rc==SQLITE_OK ){ + pWal->writeLock = 1; + }else if( rc==SQLITE_BUSY ){ + eMode2 = SQLITE_CHECKPOINT_PASSIVE; + rc = SQLITE_OK; + } + } -/* -** A cursor is a pointer to a particular entry within a particular -** b-tree within a database file. -** -** The entry is identified by its MemPage and the index in -** MemPage.aCell[] of the entry. -** -** A single database file can be shared by two more database connections, -** but cursors cannot be shared. Each cursor is associated with a -** particular database connection identified BtCursor.pBtree.db. -** -** Fields in this structure are accessed under the BtShared.mutex -** found at self->pBt->mutex. -*/ -struct BtCursor { - Btree *pBtree; /* The Btree to which this cursor belongs */ - BtShared *pBt; /* The BtShared this cursor points to */ - BtCursor *pNext, *pPrev; /* Forms a linked list of all cursors */ - struct KeyInfo *pKeyInfo; /* Argument passed to comparison function */ -#ifndef SQLITE_OMIT_INCRBLOB - Pgno *aOverflow; /* Cache of overflow page locations */ -#endif - Pgno pgnoRoot; /* The root page of this tree */ - sqlite3_int64 cachedRowid; /* Next rowid cache. 0 means not valid */ - CellInfo info; /* A parse of the cell we are pointing at */ - i64 nKey; /* Size of pKey, or last integer key */ - void *pKey; /* Saved key that was cursor's last known position */ - int skipNext; /* Prev() is noop if negative. Next() is noop if positive */ - u8 wrFlag; /* True if writable */ - u8 atLast; /* Cursor pointing to the last entry */ - u8 validNKey; /* True if info.nKey is valid */ - u8 eState; /* One of the CURSOR_XXX constants (see below) */ -#ifndef SQLITE_OMIT_INCRBLOB - u8 isIncrblobHandle; /* True if this cursor is an incr. io handle */ -#endif - u8 hints; /* As configured by CursorSetHints() */ - i16 iPage; /* Index of current page in apPage */ - u16 aiIdx[BTCURSOR_MAX_DEPTH]; /* Current index in apPage[i] */ - MemPage *apPage[BTCURSOR_MAX_DEPTH]; /* Pages from root to current page */ -}; + /* Read the wal-index header. */ + if( rc==SQLITE_OK ){ + rc = walIndexReadHdr(pWal, &isChanged); + if( isChanged && pWal->pDbFd->pMethods->iVersion>=3 ){ + sqlite3OsUnfetch(pWal->pDbFd, 0, 0); + } + } -/* -** Potential values for BtCursor.eState. -** -** CURSOR_INVALID: -** Cursor does not point to a valid entry. This can happen (for example) -** because the table is empty or because BtreeCursorFirst() has not been -** called. -** -** CURSOR_VALID: -** Cursor points to a valid entry. getPayload() etc. may be called. -** -** CURSOR_SKIPNEXT: -** Cursor is valid except that the Cursor.skipNext field is non-zero -** indicating that the next sqlite3BtreeNext() or sqlite3BtreePrevious() -** operation should be a no-op. -** -** CURSOR_REQUIRESEEK: -** The table that this cursor was opened on still exists, but has been -** modified since the cursor was last used. The cursor position is saved -** in variables BtCursor.pKey and BtCursor.nKey. When a cursor is in -** this state, restoreCursorPosition() can be called to attempt to -** seek the cursor to the saved position. -** -** CURSOR_FAULT: -** A unrecoverable error (an I/O error or a malloc failure) has occurred -** on a different connection that shares the BtShared cache with this -** cursor. The error has left the cache in an inconsistent state. -** Do nothing else with this cursor. Any attempt to use the cursor -** should return the error code stored in BtCursor.skip -*/ -#define CURSOR_INVALID 0 -#define CURSOR_VALID 1 -#define CURSOR_SKIPNEXT 2 -#define CURSOR_REQUIRESEEK 3 -#define CURSOR_FAULT 4 + /* Copy data from the log to the database file. */ + if( rc==SQLITE_OK ){ + if( pWal->hdr.mxFrame && walPagesize(pWal)!=nBuf ){ + rc = SQLITE_CORRUPT_BKPT; + }else{ + rc = walCheckpoint(pWal, eMode2, xBusy, pBusyArg, sync_flags, zBuf); + } -/* -** The database page the PENDING_BYTE occupies. This page is never used. -*/ -# define PENDING_BYTE_PAGE(pBt) PAGER_MJ_PGNO(pBt) + /* If no error occurred, set the output variables. */ + if( rc==SQLITE_OK || rc==SQLITE_BUSY ){ + if( pnLog ) *pnLog = (int)pWal->hdr.mxFrame; + if( pnCkpt ) *pnCkpt = (int)(walCkptInfo(pWal)->nBackfill); + } + } -/* -** These macros define the location of the pointer-map entry for a -** database page. The first argument to each is the number of usable -** bytes on each page of the database (often 1024). The second is the -** page number to look up in the pointer map. -** -** PTRMAP_PAGENO returns the database page number of the pointer-map -** page that stores the required pointer. PTRMAP_PTROFFSET returns -** the offset of the requested map entry. -** -** If the pgno argument passed to PTRMAP_PAGENO is a pointer-map page, -** then pgno is returned. So (pgno==PTRMAP_PAGENO(pgsz, pgno)) can be -** used to test if pgno is a pointer-map page. PTRMAP_ISPAGE implements -** this test. + if( isChanged ){ + /* If a new wal-index header was loaded before the checkpoint was + ** performed, then the pager-cache associated with pWal is now + ** out of date. So zero the cached wal-index header to ensure that + ** next time the pager opens a snapshot on this database it knows that + ** the cache needs to be reset. + */ + memset(&pWal->hdr, 0, sizeof(WalIndexHdr)); + } + + /* Release the locks. */ + sqlite3WalEndWriteTransaction(pWal); + walUnlockExclusive(pWal, WAL_CKPT_LOCK, 1); + pWal->ckptLock = 0; + WALTRACE(("WAL%p: checkpoint %s\n", pWal, rc ? "failed" : "ok")); + return (rc==SQLITE_OK && eMode!=eMode2 ? SQLITE_BUSY : rc); +} + +/* Return the value to pass to a sqlite3_wal_hook callback, the +** number of frames in the WAL at the point of the last commit since +** sqlite3WalCallback() was called. If no commits have occurred since +** the last call, then return 0. */ -#define PTRMAP_PAGENO(pBt, pgno) ptrmapPageno(pBt, pgno) -#define PTRMAP_PTROFFSET(pgptrmap, pgno) (5*(pgno-pgptrmap-1)) -#define PTRMAP_ISPAGE(pBt, pgno) (PTRMAP_PAGENO((pBt),(pgno))==(pgno)) +SQLITE_PRIVATE int sqlite3WalCallback(Wal *pWal){ + u32 ret = 0; + if( pWal ){ + ret = pWal->iCallback; + pWal->iCallback = 0; + } + return (int)ret; +} /* -** The pointer map is a lookup table that identifies the parent page for -** each child page in the database file. The parent page is the page that -** contains a pointer to the child. Every page in the database contains -** 0 or 1 parent pages. (In this context 'database page' refers -** to any page that is not part of the pointer map itself.) Each pointer map -** entry consists of a single byte 'type' and a 4 byte parent page number. -** The PTRMAP_XXX identifiers below are the valid types. -** -** The purpose of the pointer map is to facility moving pages from one -** position in the file to another as part of autovacuum. When a page -** is moved, the pointer in its parent must be updated to point to the -** new location. The pointer map is used to locate the parent page quickly. -** -** PTRMAP_ROOTPAGE: The database page is a root-page. The page-number is not -** used in this case. -** -** PTRMAP_FREEPAGE: The database page is an unused (free) page. The page-number -** is not used in this case. +** This function is called to change the WAL subsystem into or out +** of locking_mode=EXCLUSIVE. ** -** PTRMAP_OVERFLOW1: The database page is the first page in a list of -** overflow pages. The page number identifies the page that -** contains the cell with a pointer to this overflow page. +** If op is zero, then attempt to change from locking_mode=EXCLUSIVE +** into locking_mode=NORMAL. This means that we must acquire a lock +** on the pWal->readLock byte. If the WAL is already in locking_mode=NORMAL +** or if the acquisition of the lock fails, then return 0. If the +** transition out of exclusive-mode is successful, return 1. This +** operation must occur while the pager is still holding the exclusive +** lock on the main database file. ** -** PTRMAP_OVERFLOW2: The database page is the second or later page in a list of -** overflow pages. The page-number identifies the previous -** page in the overflow page list. +** If op is one, then change from locking_mode=NORMAL into +** locking_mode=EXCLUSIVE. This means that the pWal->readLock must +** be released. Return 1 if the transition is made and 0 if the +** WAL is already in exclusive-locking mode - meaning that this +** routine is a no-op. The pager must already hold the exclusive lock +** on the main database file before invoking this operation. ** -** PTRMAP_BTREE: The database page is a non-root btree page. The page number -** identifies the parent page in the btree. +** If op is negative, then do a dry-run of the op==1 case but do +** not actually change anything. The pager uses this to see if it +** should acquire the database exclusive lock prior to invoking +** the op==1 case. */ -#define PTRMAP_ROOTPAGE 1 -#define PTRMAP_FREEPAGE 2 -#define PTRMAP_OVERFLOW1 3 -#define PTRMAP_OVERFLOW2 4 -#define PTRMAP_BTREE 5 +SQLITE_PRIVATE int sqlite3WalExclusiveMode(Wal *pWal, int op){ + int rc; + assert( pWal->writeLock==0 ); + assert( pWal->exclusiveMode!=WAL_HEAPMEMORY_MODE || op==-1 ); -/* A bunch of assert() statements to check the transaction state variables -** of handle p (type Btree*) are internally consistent. -*/ -#define btreeIntegrity(p) \ - assert( p->pBt->inTransaction!=TRANS_NONE || p->pBt->nTransaction==0 ); \ - assert( p->pBt->inTransaction>=p->inTrans ); + /* pWal->readLock is usually set, but might be -1 if there was a + ** prior error while attempting to acquire are read-lock. This cannot + ** happen if the connection is actually in exclusive mode (as no xShmLock + ** locks are taken in this case). Nor should the pager attempt to + ** upgrade to exclusive-mode following such an error. + */ + assert( pWal->readLock>=0 || pWal->lockError ); + assert( pWal->readLock>=0 || (op<=0 && pWal->exclusiveMode==0) ); + + if( op==0 ){ + if( pWal->exclusiveMode ){ + pWal->exclusiveMode = 0; + if( walLockShared(pWal, WAL_READ_LOCK(pWal->readLock))!=SQLITE_OK ){ + pWal->exclusiveMode = 1; + } + rc = pWal->exclusiveMode==0; + }else{ + /* Already in locking_mode=NORMAL */ + rc = 0; + } + }else if( op>0 ){ + assert( pWal->exclusiveMode==0 ); + assert( pWal->readLock>=0 ); + walUnlockShared(pWal, WAL_READ_LOCK(pWal->readLock)); + pWal->exclusiveMode = 1; + rc = 1; + }else{ + rc = pWal->exclusiveMode==0; + } + return rc; +} +/* +** Return true if the argument is non-NULL and the WAL module is using +** heap-memory for the wal-index. Otherwise, if the argument is NULL or the +** WAL module is using shared-memory, return false. +*/ +SQLITE_PRIVATE int sqlite3WalHeapMemory(Wal *pWal){ + return (pWal && pWal->exclusiveMode==WAL_HEAPMEMORY_MODE ); +} +#ifdef SQLITE_ENABLE_ZIPVFS /* -** The ISAUTOVACUUM macro is used within balance_nonroot() to determine -** if the database supports auto-vacuum or not. Because it is used -** within an expression that is an argument to another macro -** (sqliteMallocRaw), it is not possible to use conditional compilation. -** So, this macro is defined instead. +** If the argument is not NULL, it points to a Wal object that holds a +** read-lock. This function returns the database page-size if it is known, +** or zero if it is not (or if pWal is NULL). */ -#ifndef SQLITE_OMIT_AUTOVACUUM -#define ISAUTOVACUUM (pBt->autoVacuum) -#else -#define ISAUTOVACUUM 0 +SQLITE_PRIVATE int sqlite3WalFramesize(Wal *pWal){ + assert( pWal==0 || pWal->readLock>=0 ); + return (pWal ? pWal->szPage : 0); +} #endif +#endif /* #ifndef SQLITE_OMIT_WAL */ +/************** End of wal.c *************************************************/ +/************** Begin file btmutex.c *****************************************/ /* -** This structure is passed around through all the sanity checking routines -** in order to keep track of some global state information. +** 2007 August 27 ** -** The aRef[] array is allocated so that there is 1 bit for each page in -** the database. As the integrity-check proceeds, for each page used in -** the database the corresponding bit is set. This allows integrity-check to -** detect pages that are used twice and orphaned pages (both of which -** indicate corruption). -*/ -typedef struct IntegrityCk IntegrityCk; -struct IntegrityCk { - BtShared *pBt; /* The tree being checked out */ - Pager *pPager; /* The associated pager. Also accessible by pBt->pPager */ - u8 *aPgRef; /* 1 bit per page in the db (see above) */ - Pgno nPage; /* Number of pages in the database */ - int mxErr; /* Stop accumulating errors when this reaches zero */ - int nErr; /* Number of messages written to zErrMsg so far */ - int mallocFailed; /* A memory allocation error has occurred */ - StrAccum errMsg; /* Accumulate the error message text here */ -}; - -/* -** Routines to read or write a two- and four-byte big-endian integer values. +** The author disclaims copyright to this source code. In place of +** a legal notice, here is a blessing: +** +** May you do good and not evil. +** May you find forgiveness for yourself and forgive others. +** May you share freely, never taking more than you give. +** +************************************************************************* +** +** This file contains code used to implement mutexes on Btree objects. +** This code really belongs in btree.c. But btree.c is getting too +** big and we want to break it down some. This packaged seemed like +** a good breakout. */ -#define get2byte(x) ((x)[0]<<8 | (x)[1]) -#define put2byte(p,v) ((p)[0] = (u8)((v)>>8), (p)[1] = (u8)(v)) -#define get4byte sqlite3Get4byte -#define put4byte sqlite3Put4byte - -/************** End of btreeInt.h ********************************************/ -/************** Continuing where we left off in btmutex.c ********************/ #ifndef SQLITE_OMIT_SHARED_CACHE #if SQLITE_THREADSAFE @@ -88995,10 +91735,24 @@ */ SQLITE_PRIVATE void sqlite3RegisterBuiltinFunctions(sqlite3 *db){ int rc = sqlite3_overload_function(db, "MATCH", 2); +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +#ifndef OMIT_EXPORT + extern void sqlcipher_exportFunc(sqlite3_context *, int, sqlite3_value **); +#endif +#endif +/* END SQLCIPHER */ assert( rc==SQLITE_NOMEM || rc==SQLITE_OK ); if( rc==SQLITE_NOMEM ){ db->mallocFailed = 1; } +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC +#ifndef OMIT_EXPORT + sqlite3CreateFunc(db, "sqlcipher_export", 1, SQLITE_TEXT, 0, sqlcipher_exportFunc, 0, 0, 0); +#endif +#endif +/* END SQLCIPHER */ } /* @@ -94082,6 +96836,12 @@ sqlite3 *db = pParse->db; /* The database connection */ Db *pDb; /* The specific database being pragmaed */ Vdbe *v = sqlite3GetVdbe(pParse); /* Prepared statement */ +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC + extern int sqlcipher_codec_pragma(sqlite3*, int, Parse *, const char *, const char *); +#endif +/* END SQLCIPHER */ + if( v==0 ) return; sqlite3VdbeRunOnlyOnce(v); @@ -94142,6 +96902,13 @@ pParse->rc = rc; }else +/* BEGIN SQLCIPHER */ +#ifdef SQLITE_HAS_CODEC + if(sqlcipher_codec_pragma(db, iDb, pParse, zLeft, zRight)) { + /* sqlcipher_codec_pragma executes internal */ + }else + #endif +/* END SQLCIPHER */ #if !defined(SQLITE_OMIT_PAGER_PRAGMAS) && !defined(SQLITE_OMIT_DEPRECATED) /* @@ -94739,60 +97506,6 @@ #ifndef SQLITE_OMIT_SCHEMA_PRAGMAS /* - ** PRAGMA proc_list - ** - ** Return a single row for each procedure, the returned data set are: - ** - ** name: Procedure name - ** is_aggregate: True is procedure is an aggregate - ** nargs: Number of arguments of the procedure, or -1 if unlimited - ** spe_name: Specific name (unique procedure name) - */ - if( sqlite3StrICmp(zLeft, "proc_list")==0 ){ - if( sqlite3ReadSchema(pParse) ) goto pragma_out; - - sqlite3VdbeSetNumCols(v, 4); - pParse->nMem = 4; - sqlite3VdbeSetColName(v, 0, COLNAME_NAME, "name", SQLITE_STATIC); - sqlite3VdbeSetColName(v, 1, COLNAME_NAME, "is_aggregate", SQLITE_STATIC); - sqlite3VdbeSetColName(v, 2, COLNAME_NAME, "nargs", SQLITE_STATIC); - sqlite3VdbeSetColName(v, 3, COLNAME_NAME, "spe_name", SQLITE_STATIC); - int j; - for(j=0; jaFunc.a); j++){ - FuncDef *func; - for (func =db->aFunc.a[j]; func; func = func->pNext) { - char *sname; - int size; - size = strlen (func->zName) + 25; - sname = sqlite3_malloc (sizeof (char) * size); - snprintf (sname, size-1, "%s_%d_%d", func->zName, func->nArg, func->iPrefEnc); - sqlite3VdbeAddOp4(v, OP_String8, 0, 1, 0, func->zName, 0); - sqlite3VdbeAddOp2(v, OP_Integer, func->xFinalize ? 1 : 0, 2); - sqlite3VdbeAddOp2(v, OP_Integer, func->nArg, 3); - sqlite3VdbeAddOp4(v, OP_String8, 0, 4, 0, sname, 0); - sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 4); - sqlite3_free (sname); - } - } - for(j=0; jpNext) { - char *sname; - int size; - size = strlen (func->zName) + 25; - sname = sqlite3_malloc (sizeof (char) * size); - snprintf (sname, size-1, "%s_%d_%d", func->zName, func->nArg, func->iPrefEnc); - sqlite3VdbeAddOp4(v, OP_String8, 0, 1, 0, func->zName, 0); - sqlite3VdbeAddOp2(v, OP_Integer, func->xFinalize ? 1 : 0, 2); - sqlite3VdbeAddOp2(v, OP_Integer, func->nArg, 3); - sqlite3VdbeAddOp4(v, OP_String8, 0, 4, 0, sname, 0); - sqlite3VdbeAddOp2(v, OP_ResultRow, 1, 4); - sqlite3_free (sname); - } - } - }else - - /* ** PRAGMA table_info() ** ** Return a single row for each column of the named table. The columns of