Difference between revisions of "セーブデータ"

From 3dbrew
Jump to navigation Jump to search
Line 621: Line 621:

Latest revision as of 08:25, 21 July 2012


This page describes the format, encryption, etc. of savegames stored in 3DS game cartridges/gamecards and SD/NAND. You can find savegames from various 3DS games on the Games page.

Gamecard Savegame Encryption[edit]

On the 3DS savegames are stored much like on the DS, that is on a FLASH chip in the gamecart. On the DS these savegames were stored in plain-text but on the 3DS a layer of encryption was added. This is highly likely a streamcipher, as the contents of several savegames exhibit the odd behavior that xor-ing certain parts of the savegame together will result in the plain-text appearing.

The reason this works is because the stream cipher used has a period of 512 bytes. That is to say, it will repeat the same keystream after 512 bytes. The way you encrypt with a stream cipher is you XOR your data with the keystream as it is produced. Unfortunately, if your streamcipher repeats and you are encrypting a known plain-text (in our case, zeros) you are basically giving away your valuable keystream.

So how do you use this to decrypt a savegame on a 3DS? First off, you chunk up the savegame into 512 byte chunks. Then, you bin these chunks by their contents, discarding any that contain only FF. Now look for the most common chunk. This is your keystream. Now XOR the keystream with your original savegame and you should have a fully decrypted savegame. XOR with the keystream again to produce an encrypted savegame.

All savegames, including non-gamecard savegames, are encrypted with AES-CTR. The base CTR never changes after savegame creation. The CTR used for gamecard savegames eventually repeats, while non-gamecard savegames don't use a repeating CTR. For the old gamecard CTR method, it repeated every 0x200-bytes. With the new method it repeats at least every 0x1000-bytes, but the exact period isn't known for certain.


With system version 2.0.0-4 the system now uses a different CTR method, which fixed the above flaw. The xorpad seems to repeat in the file but not every 0x200 bytes. So for now it is unknown how to decrypt the newer save files.

Games to use the new encryption:

  • Super Mario 3D Land
  • Mario Kart 7
  • Need for Speed - The Run

Some information:

  • Old games saves still use the old 0x200-bytes AES-CTR xorpad.
  • New games saves can be backed-up and restored.
  • The wearleveling stayed the same.
  • Xoring two files together can produce some clear text
  • It's been spotted that the AES-CTR xorpad repeated after 0x1000 bytes (so it might be the maximum length but still it's not proved).

Wear leveling[edit]

The 3DS employs a wear leveling scheme on the savegame FLASH chips. This is done through the usage of blockmaps and a journal. The blockmap is located at offset 0 of the flash chip, and is immediately followed by the journal. The initial state is dictated by the blockmap, and the journal is then applied to that.

First, there are 8 bytes whose purposes are currently unknown. Then comes the actual blockmap. The blockmap structure is simple:

struct header_entry {
        uint8_t phys_sec; // when bit7 is set, block has checksums, otherwise checksums are all zero
        uint8_t alloc_cnt;
        uint8_t chksums[8];
} __attribute__((__packed__));

There's one entry per sector, counting from physical sector 1 (sector 0 contains the blockmap/journal).

The 2 bytes that follow the blockmap are the CRC16 (with starting value 0xFFFF (like modbus)) of the first 8 bytes and the blockmap.

Then comes the journal. The journal structure is as follows:

struct sector_entry {
        uint8_t virt_sec;       // Mapped to sector
        uint8_t prev_virt_sec;  // Physical sector previously mapped to
        uint8_t phys_sec;       // Mapped from sector
        uint8_t prev_phys_sec;  // Virtual sector previously mapped to
        uint8_t phys_realloc_cnt;       // Amount of times physical sector has been remapped
        uint8_t virt_realloc_cnt;       // Amount of times virtual sector has been remapped
        uint8_t chksums[8];
} __attribute__((__packed__));

struct long_sector_entry{
        struct sector_entry sector;
        struct sector_entry dupe;
        uint32_t magic;

With magic being a constant 0x080d6ce0.

The checksums in the blockmap/journal entries work as follows:

  • each byte is the checksum of an encrypted 0x200 bytes large block
  • to calculate the checksum, a CRC16 of the block (with starting value 0xFFFF) is calculated, and the two bytes of the CRC16 are XORed together to produce the 8bit checksum


There can be multiple partitions on the chip. The partitions are represented by tables of DIFI blobs inside a DISA structure. The order of the DIFI blobs is the order of the partitions in the chip.


  • If the uint32 @ 0x168 into the image in the DISA(the low 8-bits) is non-zero, then first table is is hashed, otherwise the second DIFI table is hashed.
  • If the table has more then 1 DIFI then the uint32 @ 0x168 is the offset from the DATA partition to the file base (masked with 0xFFFFFFFE).
  • At offset 0x0 in the image is a 0x10-byte MAC over the 0x100-byte DISA/DIFF, it might be AES-CCM MAC but it's unknown for certain. The following 0xf0-bytes after the MAC normally must be zero, it's unknown whether this can ever be non-zero.
Start Length Description
0x00 4 Magic ("DISA")
0x04 4 Magic Number (0x40000)
0x08 8 Partition table size
0x10 8 Offset to primary partition table in DISA
0x18 8 Offset to secondary partition table in DISA
0x20 8 Partition table's length
0x28 8 SAVE Partition entry offset in the partition table
0x30 8 SAVE Partition entry length in the partition table
0x38 8 DATA Partition entry offset in the partition table
0x40 8 DATA Partition entry length in the partition table
0x48 8 SAVE Partition offset
0x50 8 SAVE Partition length
0x58 8 DATA Partition offset
0x60 8 DATA Partition length
0x68 4 Active table (and the offset to the filebase)
0x6C 0x20 Hash from active table
0x8C 4*29 Unknown
  • The hash in the DISA hashes the Active Table (starting from tables's offset to tables's offset + table length) with SHA256.
  • The partitions offsets points to a 0x1000 long block which isn't understood yet. The actual information starts after that block.

The DIFIs table @ 0x200 (into the image) is written twice, (Meaning, if there's 4 DIFI blobs then the table is 2 DIFIs long).

The second table is for backup. The active table is mentioned at 0x13C into the image (1=First table, other=Second Table)


  • This is the extdata equivalent of DISA, for extdata which use FS. DIFF is *only* used with extdata, not regular savegames.
  • When the active-table field low 8-bits is non-zero, the primary partition is used. Otherwise, the secondary partition is used.
Start Length Description
0x00 4 Magic ("DIFF")
0x04 4 Magic Number (0x30000)
0x08 8 Primary partition table offset
0x10 8 Secondary partition table offset
0x18 8 Partition table length
0x20 4 Active table (and the offset to the filebase)
0x24 0x20 Unknown
0x34 0x20 Hash of the active partition table
0x54 0x1ac Unknown


These 0x130 large blobs describe the partitions. Every DIFI blob describes a partition. Partitions are catted together, so after the end of one partition is the beginning of the next.

Actually DIFI blobs are 0x12C large because the last 4 are not used and appear 0xFFFFFFFF at the encrypted image.

For most games there's only 1 partition (The SAVE partition) and some (like Asphalt 3D, Steel Diver & Lego Star Wars III) has 2 partitions.

  • 2 Partitions means that the files inside the SAVE partition is on the other partition (we would call it DATA partition).
  • No more than 2 partitions have been seen yet (and can't be because of the DISA known structure).
Start Length Description
0x00 4 Magic ("DIFI")
0x04 4 Magic Number (0x10000)
0x08 8 Offset to "IVFC" blob in DIFI (usually 0x44)
0x10 8 Size of "IVFC" blob
0x18 8 Offset to "DPFS" blob in DIFI (usually 0xBC)
0x20 8 Size of "DPFS" blob
0x28 8 Offset to the hash in DIFI (usually 0x010C)
0x30 8 Size of this hash
0x38 4 Flags (when this byte is non-zero, this is a DATA partition)
0x3C 8 File base offset (for DATA partitions)


Start Length Description
0x00 4 Magic ("IVFC")
0x04 4 Magic Number (0x20000)
0x08 8 Unknown (0x20?)
0x10 8 First Hash Offset
0x18 8 First Hash Length
0x20 8 First Hash Block Size (1<<value)
0x28 8 Second Hash Offset
0x30 8 Second Hash Length
0x38 8 Second Hash Block Size (1<<value)
0x40 8 HashTable Offset
0x48 8 HashTable Length
0x50 8 HashTable Block Size (1<<value)
0x58 8 FileSystem Offset
0x60 8 FileSystem Length
0x68 8 FileSystem Block Size (1<<value)
0x70 8 Unknown (usually 0x78=120)
  • First & Second hash are not understood yet.


Start Length Description
0x00 4 Magic ("DPFS")
0x04 4 Magic Number (0x10000)
0x08 8 Offset To First table
0x10 8 First table length
0x18 8 First table block size (1<<value)
0x20 8 Offset To Second table
0x28 8 Second table length
0x30 8 Second table block size (1<<value)
0x38 8 Offset to Data
0x40 8 Data Length
0x48 8 Data block size (1<<value)
  • Every block this table point to is written twice (concatenated). You can see that the offset to the next block is twice the length (except the data which always begin after 0x1000).

The first partition's data starts at 0x2000. First comes the hashtable (usually start @ 0x40 into the partition) and then the filesystem.

The hashtable entries' size is 2^x where x is the 'Hashed block size' from the IVFC block.


After the DIFI,IVFC & DPFS comes a 0x20 long hash, it is unknown what it's hashing.

Summary Drawing

Sfimg drawing.png

The SAVE partition[edit]

  • The SAVE filesystem works with a backup. There are two SAVE blocks inside the partition concatenated. Which SAVE block is the updated one is unknown yet.. (I'm guessing from experience that (image[0x100B] & 0x20) == 0x20 --> 1st SAVE --Elisherer 01:30, 18 October 2011 (CEST))

Finding the folders table:

  • If DATA partition exists: At folder table exact offset from the SAVE struct (from the beginning of the struct).
  • Otherwise: The 'folder table offset' * 'folder table media' (=0x200) from the 'filestore offset'. (usually 0 from filebase)

Finding the files table:

  • If DATA partition exists: At file table exact offset from the SAVE struct (from the beginning of the struct).
  • Otherwise: The 'file table offset' * 'file table media' (=0x200) from the 'filestore offset'.

Detemining the filestore base:

  • If DATA partition exists: At file base from the DATA's DIFI struct into the DATA partition.
  • Otherwise: At the 'filestore offset' from the beginning of the SAVE struct.

Folder's entry structure:

 struct folder_entry {
     u32 parent_folder_index;
     u8  filename[0x10];
     u32 folder_index;
     u32 unk1; 
     u32 last_file_index;
     u32 unk3; 
     u32 unk4;

File's entry structure:

 struct file_entry {
     u32 parent_folder_index;
     u8  filename[0x10];
     u32 index;
     u32 unk1; // magic?
     u32 block_offset;
     u64 file_size;
     u32 unk2; // flags?
     u32 unk3;

The first entry in both tables is the count of the table, the parent directory index will be the amount of table rows. The root includes itself, so there are the amount - 1 (minus one) folders in the root directory (or files). The entries that follow after the root are the actual folders/files.

Reading the files out is as simple as taking the file base offset and adding (block_offset * 0x200) to it.

Here's a follow-up example from the Legend of Zelda: Ocarina of Time 3D:

//FST entry = SAVE base + File base + (FST offset * 0x200) + (FST entry # * 0x30)
//0x2600    = 0x2000    + 0x400     + (0x1        * 0x200) + (0x0         * 0x30)

00002600: 03000000 09000000 00000000 00000000  ................
00002610: 00000000 00000000 00000000 00000000  ................
00002620: 00000000 00000000 00000000 00000000  ................
00002630: 01000000 73797374 656D2E64 61740000  ....system.dat..
00002640: 00000000 00000000 D57B1100 02000000  ........Õ{......
00002650: 22000000 00000000 E8121500 00000000  ".......è.......
00002660: 01000000 73617665 30302E62 696E0000  ....save00.bin..
00002670: 00000000 01000000 69921100 03000000  ........i’......
00002680: DC140000 00000000 04000000 00000000  Ü...............
Start Length Description
0x00 4 Magic ("SAVE")
0x04 4 Magic Number (0x40000)
0x08 8 Offset to data in this SAVE header(normally 0x20)
0x10 8 Partition Size [medias]
0x18 4 Partition Media Size
0x1C 8 Unknown
0x24 4 Unknown
0x28 8 FolderMap Offset
0x30 4 FolderMap Size
0x34 4 FolderMap Media Size
0x38 8 FileMap Offset
0x40 4 FileMap Size
0x44 4 FileMap Media Size
0x48 8 BlockMap Offset
0x50 4 BlockMap Size
0x54 4 BlockMap Media Size
0x58 8 File store offset (from SAVE)
0x60 4 File store length [medias]
0x64 4 File store media size
0x68 4/8 Folders Table offset (8 bytes in DATA)
0x6C 4 Folders Table Length (medias) (Only in no DATA)
0x70 4 Folders Table unknown
0x74 4 Folders Table Media size
0x78 4/8 Files Table offset (8 bytes in DATA)
0x7C 4 Files Table Length (medias) (Only in no DATA)
0x80 4 Files Table unknown
0x84 4 Files Table Media size
  • The FolderMap and FileMap still unknown. They are tables of uint32.
  • The BlockMap is a map of the blocks in the filestore. An entry in the BlockMap is 2 uint32: {uint32 start_block; uint32 end_block; }. This is still being researched. (You can use 3DSExplorer to see those maps.

Summary Drawing

Sfsave drawing.png


When a save FLASH contains all xFFFF blocks it's assumed uninitialized by the game cartridges and it initializes default data in place, without prompting the user. The 0xFFFFFFFF blocks are uninitialized data. When creating a non-gamecard savegame and other images/files, it's initially all 0xFFFFFFFF until it's formatted where some of the blocks are overwritten with encrypted data.

I got a new game SplinterCell3D-Pal and I downloaded the save and it was 128KB of 0xFF, except the first 0x10 bytes which were the letter 'Z' (uppercase) --Elisherer 22:41, 15 October 2011 (CEST)

Fun Facts[edit]

If you have facts that you found out by looking at the binary files please share them here:

  • From one save to another the game backups the last files that were in the partition and the entire image header in "random" locations.. --Elisherer 22:41, 15 October 2011 (CEST)