)]}'
{
  "log": [
    {
      "commit": "b0c9778b1d07ed3aa7e411db201275553527b1b1",
      "tree": "8649c1b27edebc6addef5087eb1830fc8bccdb74",
      "parents": [
        "9f6455325618821dcf6775d7972881fde32e77c5"
      ],
      "author": {
        "name": "Tejun Heo",
        "email": "tj@kernel.org",
        "time": "Fri Apr 09 18:57:01 2010 +0900"
      },
      "committer": {
        "name": "Tejun Heo",
        "email": "tj@kernel.org",
        "time": "Sat May 01 08:30:50 2010 +0200"
      },
      "message": "percpu: implement kernel memory based chunk allocation\n\nImplement an alternate percpu chunk management based on kernel memeory\nfor nommu SMP architectures.  Instead of mapping into vmalloc area,\nchunks are allocated as a contiguous kernel memory using\nalloc_pages().  As such, percpu allocator on nommu will have the\nfollowing restrictions.\n\n* It can\u0027t fill chunks on-demand page-by-page.  It has to allocate\n  each chunk fully upfront.\n\n* It can\u0027t support sparse chunk for NUMA configurations.  SMP w/o mmu\n  is crazy enough.  Let\u0027s hope no one does NUMA w/o mmu.  :-P\n\n* If chunk size isn\u0027t power-of-two multiple of PAGE_SIZE, the\n  unaligned amount will be wasted on each chunk.  So, archs which use\n  this better align chunk size.\n\nFor instructions on how to use this, read the comment on top of\nmm/percpu-km.c.\n\nSigned-off-by: Tejun Heo \u003ctj@kernel.org\u003e\nReviewed-by: David Howells \u003cdhowells@redhat.com\u003e\nCc: Graff Yang \u003cgraff.yang@gmail.com\u003e\nCc: Sonic Zhang \u003csonic.adi@gmail.com\u003e\n"
    }
  ]
}