)]}'
{
  "log": [
    {
      "commit": "2a0c4fdb6602ea066380aaf71ff7bb2f61ffeee8",
      "tree": "b77ed1db44e1d0948ac2ab8414066e365da21ddd",
      "parents": [
        "dbfe44f02b6855efb5a596e942ec2fd96d592f60"
      ],
      "author": {
        "name": "Bernd Schmidt",
        "email": "bernds_cb1@t-online.de",
        "time": "Wed Apr 23 07:17:34 2008 +0800"
      },
      "committer": {
        "name": "Bryan Wu",
        "email": "cooloney@kernel.org",
        "time": "Wed Apr 23 07:17:34 2008 +0800"
      },
      "message": "[Blackfin] arch: fix bug - when using trace buffer with CONFIG_MPU enabled.\n\nThere were a couple of problems with the way the trace buffer state\nis saved/restored in assembly.  The DEBUG_HWTRACE_SAVE/RESTORE macros\nsave a value to the stack, which is not immediately obvious; the CPLB\nexception code needed changes to load the correct value of the stack\npointer.  The other problem is that the SAVE/RESTORE macros weren\u0027t\npushing and popping the value downwards on the stack, but rather moving\nit _upwards_, which is of course completely broken.\n\nWe also need to make sure there\u0027s a matching DEBUG_HWTRACE_RESTORE in\nthe error case of the CPLB handler.\n\nSigned-off-by: Bernd Schmidt \u003cbernds_cb1@t-online.de\u003e\nSigned-off-by: Bryan Wu \u003ccooloney@kernel.org\u003e\n\n"
    },
    {
      "commit": "80f31c8a03d2f0644d0ceaf14e7e0108a007c962",
      "tree": "71af89f61a1e9f969b9d24977e97ed5982f98237",
      "parents": [
        "a3acf52885a2312efb30a043062ef88dc3813082"
      ],
      "author": {
        "name": "Mike Frysinger",
        "email": "vapier.adi@gmail.com",
        "time": "Sat Feb 02 15:47:24 2008 +0800"
      },
      "committer": {
        "name": "Bryan Wu",
        "email": "bryan.wu@analog.com",
        "time": "Sat Feb 02 15:47:24 2008 +0800"
      },
      "message": "[Blackfin] arch: change the trace buffer control start/stop logic in the exception handlers\n\nTo save/restore the trace buffer control so that if we take an exception\nafter turning off the trace buffer at a higher level we dont inadvertently\nturn the trace buffer back on\n\nSigned-off-by: Mike Frysinger \u003cvapier.adi@gmail.com\u003e\nSigned-off-by: Bryan Wu \u003cbryan.wu@analog.com\u003e\n\n"
    },
    {
      "commit": "518039bc24cbb9ce34665814fe120eac50bedd9a",
      "tree": "6b089a05025ae224d6636b6dc9d95e7c9e4677eb",
      "parents": [
        "f16295e7e7f2a2a15876f570f10d6dc8f1f36ab8"
      ],
      "author": {
        "name": "Robin Getz",
        "email": "robin.getz@analog.com",
        "time": "Wed Jul 25 11:03:28 2007 +0800"
      },
      "committer": {
        "name": "Bryan Wu",
        "email": "bryan.wu@analog.com",
        "time": "Wed Jul 25 11:03:28 2007 +0800"
      },
      "message": "Blackfin arch: Add ability to expend the hardware trace buffer\n\nAdd ability to expend the hardware trace buffer via a configurable\nsoftware buffer - so you can have lots of history when a crash occurs.\n\nThe interesting way we do printk in the traps.c confusese the checking\nscript\n\nSigned-off-by: Robin Getz \u003crobin.getz@analog.com\u003e\nSigned-off-by: Bryan Wu \u003cbryan.wu@analog.com\u003e\n\n"
    },
    {
      "commit": "669b792c77bbc30e9f4d9c95dbc918dc348c49c2",
      "tree": "2d8403e9e3cf998eda8465a322261d45e24cd26a",
      "parents": [
        "29440a2b4cd37e32dfe0fa60ef1665775b24dab1"
      ],
      "author": {
        "name": "Robin Getz",
        "email": "robin.getz@analog.com",
        "time": "Thu Jun 21 16:34:08 2007 +0800"
      },
      "committer": {
        "name": "Bryan Wu",
        "email": "bryan.wu@analog.com",
        "time": "Thu Jun 21 16:34:08 2007 +0800"
      },
      "message": "Blackfin arch: Clean up trace buffer handling, No major functional changes.\n\nTurns on trace earlier, so crashes at kernel start should print out a\ntrace, making things easier to debug.\n\nSigned-off-by: Robin Getz \u003crobin.getz@analog.com\u003e\nSigned-off-by: Mike Frysinger \u003cmichael.frysinger@analog.com\u003e\nSigned-off-by: Bryan Wu \u003cbryan.wu@analog.com\u003e\n\n"
    }
  ]
}
