)]}'
{
  "commit": "136221fc3219b3805c48db5da065e8e3467175d4",
  "tree": "62fa68f46cb765143b6159993eceb30a6b81f9b2",
  "parents": [
    "3d44cc3e01ee1b40317f79ed54324e25c4f848df"
  ],
  "author": {
    "name": "Wu Fengguang",
    "email": "fengguang.wu@intel.com",
    "time": "Tue Dec 23 15:21:30 2008 -0500"
  },
  "committer": {
    "name": "Trond Myklebust",
    "email": "Trond.Myklebust@netapp.com",
    "time": "Tue Dec 23 15:21:30 2008 -0500"
  },
  "message": "nfs: remove redundant tests on reading new pages\n\naops-\u003ereadpages() and its NFS helper readpage_async_filler() will only\nbe called to do readahead I/O for newly allocated pages. So it\u0027s not\nnecessary to test for the always 0 dirty/uptodate page flags.\n\nThe removal of nfs_wb_page() call also fixes a readahead bug: the NFS\nreadahead has been synchronous since 2.6.23, because that call will\nclear PG_readahead, which is the reminder for asynchronous readahead.\n\nMore background: the PG_readahead page flag is shared with PG_reclaim,\none for read path and the other for write path. clear_page_dirty_for_io()\nunconditionally clears PG_readahead to prevent possible readahead residuals,\nassuming itself to be always called in the write path. However, NFS is one\nand the only exception in that it _always_ calls clear_page_dirty_for_io()\nin the read path, i.e. for readpages()/readpage().\n\nCc: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\nSigned-off-by: Wu Fengguang \u003cwfg@linux.intel.com\u003e\nSigned-off-by: Trond Myklebust \u003cTrond.Myklebust@netapp.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "40d17987d0e847a87fa4596a9a4c9f39a464288e",
      "old_mode": 33188,
      "old_path": "fs/nfs/read.c",
      "new_id": "f856004bb7fa3c8ed4bfe99c3c61ccd6d5729861",
      "new_mode": 33188,
      "new_path": "fs/nfs/read.c"
    }
  ]
}