)]}'
{
  "log": [
    {
      "commit": "4440095c8268c1a5e11577097d2be429cec036ca",
      "tree": "7e1ca48bcd8fd0b947a7cc0edf31c0af9bbc4ec3",
      "parents": [
        "f42ecb2808db5386f983d593a7c08d3ea3b94a27"
      ],
      "author": {
        "name": "Andi Kleen",
        "email": "ak@linux.intel.com",
        "time": "Wed Dec 23 21:00:20 2009 +0100"
      },
      "committer": {
        "name": "Andi Kleen",
        "email": "ak@linux.intel.com",
        "time": "Wed Dec 23 21:00:20 2009 +0100"
      },
      "message": "SYSCTL: Print binary sysctl warnings (nearly) only once\n\nWhen printing legacy sysctls print the warning message\nfor each of them only once.  This way there is a guarantee\nthe syslog won\u0027t be flooded for any sane program.\n\nThe original attempt at this made the tables non const and stored\nthe flag inline.\n\nLinus suggested using a separate hash table for this, this is based on a\ncode snippet from him.\n\nThe hash implies this is not exact and can sometimes not print a\nnew sysctl due to a hash collision, but in practice this should not\nbe a problem\n\nI used a FNV32 hash over the binary string with a 32byte bitmap. This\ngives relatively little collisions when all the predefined binary sysctls\nare hashed:\n\nsize 256\nbucket\nlength      number\n0:          [25]\n1:          [67]\n2:          [88]\n3:          [47]\n4:          [22]\n5:          [6]\n6:          [1]\n\nThe worst case is a single collision of 6 hash values.\n\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\n"
    },
    {
      "commit": "61cf693159d6a968a7014e24905143f71ed8ddcf",
      "tree": "595a4bddf8b005f7345cc123b19f4da1f5d07aa7",
      "parents": [
        "288f02bbb6e9609cbaf1eb7a9cb97ae45ce090b2"
      ],
      "author": {
        "name": "Andi Kleen",
        "email": "andi@firstfloor.org",
        "time": "Wed Dec 16 12:28:44 2009 +0100"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Wed Dec 16 12:36:18 2009 -0800"
      },
      "message": "[sysctl] Fix breakage on systems with older glibc\n\nAs predicted during code review, the sysctl(2) changes made systems with\nold glibc nearly unusable.  About every command gives a:\n\n  warning: process `ls\u0027 used the deprecated sysctl system call with 1.4\n\nwarning in the log.\n\nI see this on a SUSE 10.0 system with glibc 2.3.5.\n\nDon\u0027t warn for this common case.\n\nSigned-off-by: Andi Kleen \u003cak@linux.intel.com\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "757010f026ab3044c594003e216d00a33ed95c56",
      "tree": "4e9d44c52c34c09b5a347688af20840be7a1c61a",
      "parents": [
        "63395b65972c07edce595c9cc8a983016738cdac"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Thu Nov 12 01:39:06 2009 -0800"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Thu Nov 12 01:42:31 2009 -0800"
      },
      "message": "sysctl binary: Reorder the tests to process wild card entries first.\n\nA malicious user could have passed in a ctl_name of 0 and triggered\nthe well know ctl_name to procname mapping code, instead of the wild\ncard matching code.  This is a slight problem as wild card entries don\u0027t\nhave procnames, and because in some alternate universe a network device\nmight have ifindex 0.  So test for and handle wild card entries first.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    },
    {
      "commit": "63395b65972c07edce595c9cc8a983016738cdac",
      "tree": "4ba406874098df734119d282ba634337c7541ba2",
      "parents": [
        "2fb10732c3b3c9671b1a391996ce7e551876c25e"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Thu Nov 12 00:35:55 2009 -0800"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Thu Nov 12 01:25:43 2009 -0800"
      },
      "message": "sysctl: sysctl_binary.c Fix compilation when !CONFIG_NET\n\ndev_get_by_index does not exist when the network stack is not\ncompiled in, so only include the code to follow wild card paths\nwhen the network stack is present.\n\nI have shuffled the code around a little to make it clear\nthat dev_put is called after dev_get_by_index showing that\nthere is no leak.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    },
    {
      "commit": "2fb10732c3b3c9671b1a391996ce7e551876c25e",
      "tree": "84944564aaefa5c702c79dc20ccc9ecec34dc0a8",
      "parents": [
        "2315ffa0a9f789c588c7139effa7404a387d8685"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Wed Nov 11 19:32:48 2009 -0800"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Wed Nov 11 19:35:52 2009 -0800"
      },
      "message": "sysctl:  Warn about all uses of sys_sysctl.\n\nNow that the glibc pthread implemenation no longers uses sysctl() users\nof sysctl are as rare as hen\u0027s teeth.  So remove the glibc exception\nfrom the warning, and use the standard printk_ratelimit instead of\nrolling our own.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    },
    {
      "commit": "26a7034b40ba80f82f64fa251a2cbf49f9971c6a",
      "tree": "b2f72e19e51d04c6f18398ca08f8b2e8519ee65e",
      "parents": [
        "642c6d946b5cdc27d0146c41dc20b7c4d4c3ccd8"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Thu Nov 05 05:26:41 2009 -0800"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Wed Nov 11 00:42:05 2009 -0800"
      },
      "message": "sysctl: Reduce sys_sysctl to a compatibility wrapper around /proc/sys\n\nTo simply maintenance and to be able to remove all of the binary\nsysctl support from various subsystems I have rewritten the binary\nsysctl code as a compatibility wrapper around proc/sys.\n\nThe code is built around a hard coded table based on the table\nin sysctl_check.c that lists all of our current binary sysctls\nand provides enough information to convert from the sysctl\nbinary input into into ascii and back again.  New in this\npatch is the realization that the only dynamic entries\nthat need to be handled have ifname as the asscii string\nand ifindex as their ctl_name.\n\nWhen a sys_sysctl is called the code now looks in the\ntranslation table converting the binary name to the\npath under /proc where the value is to be found.  Opens\nthat file, and calls into a format conversion wrapper\nthat calls fop-\u003eread and then fop-\u003ewrite as appropriate.\n\nSince in practice the practically no one uses or tests\nsys_sysctl rewritting the code to be beautiful is a little\nsilly.  The redeeming merit of this work is it allows us to\nrip out all of the binary sysctl syscall support from\neverywhere else in the tree.  Allowing us to remove\na lot of dead (after this patch) and barely maintained code.\n\nIn addition it becomes much easier to optimize the sysctl\nimplementation for being the backing store of /proc/sys,\nwithout having to worry about sys_sysctl.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    },
    {
      "commit": "642c6d946b5cdc27d0146c41dc20b7c4d4c3ccd8",
      "tree": "90badbad210a63e66fd5a07bc593511efad7a3d2",
      "parents": [
        "942405f36038b8f930ab67e24aa1ad72bee96a25"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Apr 03 01:08:48 2009 -0700"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Nov 06 03:53:59 2009 -0800"
      },
      "message": "sysctl: Make do_sysctl static\n\nNow that all of the architectures use compat_sys_sysctl do_sysctl\ncan become static.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    },
    {
      "commit": "da3f6f9b3e0d1e73975ca81ae124406bf1587d40",
      "tree": "ab2e5bb6366b0bdc00ec1bc5c02e610c1c77195d",
      "parents": [
        "2830b68361a9f58354ad043c6d85043ea917f907"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Apr 03 00:36:27 2009 -0700"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Nov 06 03:52:55 2009 -0800"
      },
      "message": "sysctl: Introduce a generic compat sysctl sysctl\n\nThis uses compat_alloc_userspace to remove the various\nhacks to allow do_sysctl to write to throuh oldlenp.\n\nThe rest of our mature compat syscall helper facitilies\nare used as well to ensure we have a nice clean maintainable\ncompat syscall that can be used on all architectures.\n\nThe motiviation for a generic compat sysctl (besides the\nobvious hack removal) is to reduce the number of compat\nsysctl defintions out there so I can refactor the\nbinary sysctl implementation.\n\nppc already used the name compat_sys_sysctl so I remove the\nppcs version here.\n\nCc: Benjamin Herrenschmidt \u003cbenh@kernel.crashing.org\u003e\nCc: Paul Mackerras \u003cpaulus@samba.org\u003e\nAcked-by: Arnd Bergmann \u003carnd@arndb.de\u003e\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    },
    {
      "commit": "2830b68361a9f58354ad043c6d85043ea917f907",
      "tree": "d4ab2ecbeb4340365274a8fc81c5dd0e0e092cb7",
      "parents": [
        "afa588b2651a03da4bc601a17a244b1cd97264f2"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Apr 03 00:09:33 2009 -0700"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Nov 06 03:23:35 2009 -0800"
      },
      "message": "sysctl: Refactor the binary sysctl handling to remove duplicate code\n\nRead in the binary sysctl path once, instead of reread it\nfrom user space each time the code needs to access a path\nelement.\n\nThe deprecated sysctl warning is moved to do_sysctl so\nthat the compat_sysctl entries syscalls will also warn.\n\nThe return of -ENOSYS when !CONFIG_SYSCTL_SYSCALL is moved\nto binary_sysctl.  Always leaving a do_sysctl available\nthat handles !CONFIG_SYSCTL_SYSCALL and printing the\ndeprecated sysctl warning allows for a single defitition\nof the sysctl syscall.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    },
    {
      "commit": "afa588b2651a03da4bc601a17a244b1cd97264f2",
      "tree": "cb7a03753f282b4e459305fec307caf0837e756b",
      "parents": [
        "b419148e567728f6af0c3b01965c1cc141e3e13a"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Thu Apr 02 23:44:59 2009 -0700"
      },
      "committer": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Nov 06 03:20:07 2009 -0800"
      },
      "message": "sysctl: Separate the binary sysctl logic into it\u0027s own file.\n\nIn preparation for more invasive cleanups separate the core\nbinary sysctl logic into it\u0027s own file.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\n"
    }
  ]
}