)]}'
{
  "log": [
    {
      "commit": "d521b63b27e3a397e0ef7ca86b6e813861083c83",
      "tree": "d27bc29f9e0c7beb4b1e36b252ce71f6d008909f",
      "parents": [
        "0514f8a9c0cbd26afa70dc56406cc0ee1e134dcf"
      ],
      "author": {
        "name": "Andy Grover",
        "email": "andy.grover@oracle.com",
        "time": "Fri Oct 30 08:51:57 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Fri Oct 30 15:06:39 2009 -0700"
      },
      "message": "RDS/IB+IW: Move recv processing to a tasklet\n\nMove receive processing from event handler to a tasklet.\nThis should help prevent hangcheck timer from going off\nwhen RDS is under heavy load.\n\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "a870d62726721785c34fa73d852bd35e5d1b295b",
      "tree": "bab3e33c8030e296477d4719817dfbe75303f04f",
      "parents": [
        "edacaeae52ade6cbb3a0704db32a9fb4a219dee3"
      ],
      "author": {
        "name": "Andy Grover",
        "email": "andy.grover@oracle.com",
        "time": "Fri Jul 17 13:13:33 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Jul 20 08:03:13 2009 -0700"
      },
      "message": "RDS/IB: Always use PAGE_SIZE for FMR page size\n\nWhile FMRs allow significant flexibility in what size of pages they can use,\nwe really just want FMR pages to match CPU page size. Roland says we can\ncount on this always being supported, so this simplifies things.\n\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "02a6a2592e41d27644d647f3bce23598649961bc",
      "tree": "7455566e6f3ff13279bb5949f06e398dbb8cff1a",
      "parents": [
        "9ddbcfa098bae757d3760dd1dbf2847a0bd5a525"
      ],
      "author": {
        "name": "Andy Grover",
        "email": "andy.grover@oracle.com",
        "time": "Fri Jul 17 13:13:24 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Jul 20 08:03:03 2009 -0700"
      },
      "message": "RDS/IB: Handle connections using RDS 3.0 wire protocol\n\nThe big differences between RDS 3.0 and 3.1 are protocol-level\nflow control, and with 3.1 the header is in front of the data. The header\nalways ends up in the header buffer, and the data goes in the data page.\n\nIn 3.0 our \"header\" is a trailer, and will end up either in the data\npage, the header buffer, or split across the two. Since 3.1 is backwards-\ncompatible with 3.0, we need to continue to support these cases. This\npatch does that -- if using RDS 3.0 wire protocol, it will copy the header\nfrom wherever it ended up into the header buffer.\n\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "3ba23ade464cca7c4a7ba5628c613339d3f2e161",
      "tree": "b0c5e487a9b2461fbbfdcedde870d2b545b5deb3",
      "parents": [
        "4edf547b4d0f886acf5aa5a0c8f8edbaff280830"
      ],
      "author": {
        "name": "Andy Grover",
        "email": "andy.grover@oracle.com",
        "time": "Fri Jul 17 13:13:22 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Mon Jul 20 08:03:01 2009 -0700"
      },
      "message": "RDS: Set retry_count to 2 and make modifiable via modparam\n\nThis will be default cause IB connections to failover faster,\nbut allow a longer retry count to be used if desired.\n\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "7b70d0336da777c00395cc7a503497c2cdabd1a8",
      "tree": "60da1fba7bdfb7d905ab0fe9b23b5289f068265f",
      "parents": [
        "d39e0602bb987133321d358d9b837d67c27b223d"
      ],
      "author": {
        "name": "Steve Wise",
        "email": "swise@opengridcomputing.com",
        "time": "Thu Apr 09 14:09:39 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Apr 09 17:21:17 2009 -0700"
      },
      "message": "RDS/IW+IB: Allow max credit advertise window.\n\nFix hack that restricts the credit advertisement to 127.\n\nSigned-off-by: Steve Wise \u003cswise@opengridcomputing.com\u003e\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "8cbd9606a6367c221a7bbcc47f3ab1a8c31b6437",
      "tree": "9ff2adf88be32383fd59dc133f1cd6670b9ee815",
      "parents": [
        "745cbccac3fe8cead529a1b3358e1e86a1505bfa"
      ],
      "author": {
        "name": "Andy Grover",
        "email": "andy.grover@oracle.com",
        "time": "Wed Apr 01 08:20:20 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Apr 02 00:52:22 2009 -0700"
      },
      "message": "RDS: Use spinlock to protect 64b value update on 32b archs\n\nWe have a 64bit value that needs to be set atomically.\nThis is easy and quick on all 64bit archs, and can also be done\non x86/32 with set_64bit() (uses cmpxchg8b). However other\n32b archs don\u0027t have this.\n\nI actually changed this to the current state in preparation for\nmainline because the old way (using a spinlock on 32b) resulted in\nunsightly #ifdefs in the code. But obviously, being correct takes\nprecedence.\n\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "745cbccac3fe8cead529a1b3358e1e86a1505bfa",
      "tree": "6c865b19881dbdd64556cc1827f6966d8a646a38",
      "parents": [
        "f1cffcbfcc53b825da7d1d26244aabd8dccb24aa"
      ],
      "author": {
        "name": "Andy Grover",
        "email": "andy.grover@oracle.com",
        "time": "Wed Apr 01 08:20:19 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Apr 02 00:52:22 2009 -0700"
      },
      "message": "RDS: Rewrite connection cleanup, fixing oops on rmmod\n\nThis fixes a bug where a connection was unexpectedly\nnot on *any* list while being destroyed. It also\ncleans up some code duplication and regularizes some\nfunction names.\n\n* Grab appropriate lock in conn_free() and explain in comment\n* Ensure via locking that a conn is never not on either\n  a dev\u0027s list or the nodev list\n* Add rds_xx_remove_conn() to match rds_xx_add_conn()\n* Make rds_xx_add_conn() return void\n* Rename remove_{,nodev_}conns() to\n  destroy_{,nodev_}conns() and unify their implementation\n  in a helper function\n* Document lock ordering as nodev conn_lock before\n  dev_conn_lock\n\nReported-by: Yosef Etigin \u003cyosefe@voltaire.com\u003e\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    },
    {
      "commit": "ec16227e14141e4fd7ae76354c09dadfe2449d9e",
      "tree": "9a20de7d9a0688d721517c2b38d6be83956c7ca4",
      "parents": [
        "eff5f53bef75c0864a5da06bb688939092b848dc"
      ],
      "author": {
        "name": "Andy Grover",
        "email": "andy.grover@oracle.com",
        "time": "Tue Feb 24 15:30:30 2009 +0000"
      },
      "committer": {
        "name": "David S. Miller",
        "email": "davem@davemloft.net",
        "time": "Thu Feb 26 23:39:30 2009 -0800"
      },
      "message": "RDS/IB: Infiniband transport\n\nRegisters as an RDS transport and an IB client, and uses IB CM\nAPI to allocate ids, queue pairs, and the rest of that fun stuff.\n\nSigned-off-by: Andy Grover \u003candy.grover@oracle.com\u003e\nSigned-off-by: David S. Miller \u003cdavem@davemloft.net\u003e\n"
    }
  ]
}