)]}'
{
  "log": [
    {
      "commit": "801c135ce73d5df1caf3eca35b66a10824ae0707",
      "tree": "eaf6e7859650557192533b70746479de686c56e1",
      "parents": [
        "de46c33745f5e2ad594c72f2cf5f490861b16ce1"
      ],
      "author": {
        "name": "Artem B. Bityutskiy",
        "email": "dedekind@linutronix.de",
        "time": "Tue Jun 27 12:22:22 2006 +0400"
      },
      "committer": {
        "name": "Frank Haverkamp",
        "email": "haver@vnet.ibm.com",
        "time": "Fri Apr 27 14:23:33 2007 +0300"
      },
      "message": "UBI: Unsorted Block Images\n\nUBI (Latin: \"where?\") manages multiple logical volumes on a single\nflash device, specifically supporting NAND flash devices. UBI provides\na flexible partitioning concept which still allows for wear-levelling\nacross the whole flash device.\n\nIn a sense, UBI may be compared to the Logical Volume Manager\n(LVM). Whereas LVM maps logical sector numbers to physical HDD sector\nnumbers, UBI maps logical eraseblocks to physical eraseblocks.\n\nMore information may be found at\nhttp://www.linux-mtd.infradead.org/doc/ubi.html\n\nPartitioning/Re-partitioning\n\n  An UBI volume occupies a certain number of erase blocks. This is\n  limited by a configured maximum volume size, which could also be\n  viewed as the partition size. Each individual UBI volume\u0027s size can\n  be changed independently of the other UBI volumes, provided that the\n  sum of all volume sizes doesn\u0027t exceed a certain limit.\n\n  UBI supports dynamic volumes and static volumes. Static volumes are\n  read-only and their contents are protected by CRC check sums.\n\nBad eraseblocks handling\n\n  UBI transparently handles bad eraseblocks. When a physical\n  eraseblock becomes bad, it is substituted by a good physical\n  eraseblock, and the user does not even notice this.\n\nScrubbing\n\n  On a NAND flash bit flips can occur on any write operation,\n  sometimes also on read. If bit flips persist on the device, at first\n  they can still be corrected by ECC, but once they accumulate,\n  correction will become impossible. Thus it is best to actively scrub\n  the affected eraseblock, by first copying it to a free eraseblock\n  and then erasing the original. The UBI layer performs this type of\n  scrubbing under the covers, transparently to the UBI volume users.\n\nErase Counts\n\n  UBI maintains an erase count header per eraseblock. This frees\n  higher-level layers (like file systems) from doing this and allows\n  for centralized erase count management instead. The erase counts are\n  used by the wear-levelling algorithm in the UBI layer. The algorithm\n  itself is exchangeable.\n\nBooting from NAND\n\n  For booting directly from NAND flash the hardware must at least be\n  capable of fetching and executing a small portion of the NAND\n  flash. Some NAND flash controllers have this kind of support. They\n  usually limit the window to a few kilobytes in erase block 0. This\n  \"initial program loader\" (IPL) must then contain sufficient logic to\n  load and execute the next boot phase.\n\n  Due to bad eraseblocks, which may be randomly scattered over the\n  flash device, it is problematic to store the \"secondary program\n  loader\" (SPL) statically. Also, due to bit-flips it may become\n  corrupted over time. UBI allows to solve this problem gracefully by\n  storing the SPL in a small static UBI volume.\n\nUBI volumes vs. static partitions\n\n  UBI volumes are still very similar to static MTD partitions:\n\n    * both consist of eraseblocks (logical eraseblocks in case of UBI\n      volumes, and physical eraseblocks in case of static partitions;\n    * both support three basic operations - read, write, erase.\n\n  But UBI volumes have the following advantages over traditional\n  static MTD partitions:\n\n    * there are no eraseblock wear-leveling constraints in case of UBI\n      volumes, so the user should not care about this;\n    * there are no bit-flips and bad eraseblocks in case of UBI volumes.\n\n  So, UBI volumes may be considered as flash devices with relaxed\n  restrictions.\n\nWhere can it be found?\n\n  Documentation, kernel code and applications can be found in the MTD\n  gits.\n\nWhat are the applications for?\n\n  The applications help to create binary flash images for two purposes: pfi\n  files (partial flash images) for in-system update of UBI volumes, and plain\n  binary images, with or without OOB data in case of NAND, for a manufacturing\n  step. Furthermore some tools are/and will be created that allow flash content\n  analysis after a system has crashed..\n\nWho did UBI?\n\n  The original ideas, where UBI is based on, were developed by Andreas\n  Arnez, Frank Haverkamp and Thomas Gleixner. Josh W. Boyer and some others\n  were involved too. The implementation of the kernel layer was done by Artem\n  B. Bityutskiy. The user-space applications and tools were written by Oliver\n  Lohmann with contributions from Frank Haverkamp, Andreas Arnez, and Artem.\n  Joern Engel contributed a patch which modifies JFFS2 so that it can be run on\n  a UBI volume. Thomas Gleixner did modifications to the NAND layer. Alexander\n  Schmidt made some testing work as well as core functionality improvements.\n\nSigned-off-by: Artem B. Bityutskiy \u003cdedekind@linutronix.de\u003e\nSigned-off-by: Frank Haverkamp \u003chaver@vnet.ibm.com\u003e\n"
    }
  ]
}