)]}'
{
  "log": [
    {
      "commit": "8990571eb573032c1192440febb17132074c5575",
      "tree": "c5cceff1dbac91ca12917e12f5768a5ab332ec75",
      "parents": [
        "bac0abd6174e427404dd197cdbefece31e97329b"
      ],
      "author": {
        "name": "Pavel Emelyanov",
        "email": "xemul@openvz.org",
        "time": "Thu Oct 18 23:40:19 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:41 2007 -0700"
      },
      "message": "Uninline find_pid etc set of functions\n\nThe find_pid/_vpid/_pid_ns functions are used to find the struct pid by its\nid, depending on whic id - global or virtual - is used.\n\nThe find_vpid() is a macro that pushes the current-\u003ensproxy-\u003epid_ns on the\nstack to call another function - find_pid_ns().  It turned out, that this\ndereference together with the push itself cause the kernel text size to\ngrow too much.\n\nMove all these out-of-line.  Together with the previous patch this saves a\nbit less that 400 bytes from .text section.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "19b9b9b54e5f115907efd56be2c3799775a46561",
      "tree": "44340df34980166373f2a57c1283c26026682a95",
      "parents": [
        "228ebcbe634a30aec35132ea4375721bcc41bec0"
      ],
      "author": {
        "name": "Pavel Emelyanov",
        "email": "xemul@openvz.org",
        "time": "Thu Oct 18 23:40:17 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:40 2007 -0700"
      },
      "message": "pid namespaces: remove the struct pid unneeded fields\n\nSince we\u0027ve switched from using pid-\u003enr to pid-\u003eupids-\u003enr some\nfields on struct pid are no longer needed\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "3eb07c8c8adb6f0572baba844ba2d9e501654316",
      "tree": "5c3d527f6b003b316d41119320ebd5c589c8afd0",
      "parents": [
        "0fbc26a6cfab9f377e82e28225f2c0c6b4661e5c"
      ],
      "author": {
        "name": "Sukadev Bhattiprolu",
        "email": "sukadev@us.ibm.com",
        "time": "Thu Oct 18 23:40:13 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:40 2007 -0700"
      },
      "message": "pid namespaces: destroy pid namespace on init\u0027s death\n\nTerminate all processes in a namespace when the reaper of the namespace is\nexiting.  We do this by walking the pidmap of the namespace and sending\nSIGKILL to all processes.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nAcked-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "198fe21b0a17fe9c68cb519ecc566534b04f122b",
      "tree": "690825669858d0f458fc137e42adf77cdf370ea4",
      "parents": [
        "7af5729474b5b8ad385adadab78d6e723e7655a3"
      ],
      "author": {
        "name": "Pavel Emelyanov",
        "email": "xemul@openvz.org",
        "time": "Thu Oct 18 23:40:06 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:39 2007 -0700"
      },
      "message": "pid namespaces: helpers to find the task by its numerical ids\n\nWhen searching the task by numerical id on may need to find it using global\npid (as it is done now in kernel) or by its virtual id, e.g.  when sending a\nsignal to a task from one namespace the sender will specify the task\u0027s virtual\nid and we should find the task by this value.\n\n[akpm@linux-foundation.org: fix gfs2 linkage]\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "7af5729474b5b8ad385adadab78d6e723e7655a3",
      "tree": "197f8da3e7afd65554f0c0a070eb93036b720fa8",
      "parents": [
        "8ef047aaaeb811247a5639c92e2f2ae1221a28dd"
      ],
      "author": {
        "name": "Pavel Emelyanov",
        "email": "xemul@openvz.org",
        "time": "Thu Oct 18 23:40:06 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:39 2007 -0700"
      },
      "message": "pid namespaces: helpers to obtain pid numbers\n\nWhen showing pid to user or getting the pid numerical id for in-kernel use the\nvalue of this id may differ depending on the namespace.\n\nThis set of helpers is used to get the global pid nr, the virtual (i.e.  seen\nby task in its namespace) nr and the nr as it is seen from the specified\nnamespace.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "8ef047aaaeb811247a5639c92e2f2ae1221a28dd",
      "tree": "296a61f66daa8ac42b3d77a53d06a97eb71155c9",
      "parents": [
        "faacbfd3a6808bf87d8f353b42eceeaba2c78a47"
      ],
      "author": {
        "name": "Pavel Emelyanov",
        "email": "xemul@openvz.org",
        "time": "Thu Oct 18 23:40:05 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:39 2007 -0700"
      },
      "message": "pid namespaces: make alloc_pid(), free_pid() and put_pid() work with struct upid\n\nEach struct upid element of struct pid has to be initialized properly, i.e.\nits nr mst be allocated from appropriate pidmap and ns set to appropriate\nnamespace.\n\nWhen allocating a new pid, we need to know the namespace this pid will live\nin, so the additional argument is added to alloc_pid().\n\nOn the other hand, the rest of the kernel still uses the pid-\u003enr and\npid-\u003epid_chain fields, so these ones are still initialized, but this will be\nremoved soon.\n\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "4c3f2ead5a3dff9069a45560ba4d007c8ae2e2ee",
      "tree": "ab47621edf01f64485c99cf2cde56762339c86f1",
      "parents": [
        "60347f6716aa49831ac311e04d77ccdc50dc024a"
      ],
      "author": {
        "name": "Sukadev Bhattiprolu",
        "email": "sukadev@us.ibm.com",
        "time": "Thu Oct 18 23:40:03 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri Oct 19 11:53:38 2007 -0700"
      },
      "message": "pid namespaces: introduce struct upid\n\nSince task will be visible from different pid namespaces each of them have to\nbe addressed by multiple pids.  struct upid is to store the information about\nwhich id refers to which namespace.\n\nThe constuciton looks like this.  Each struct pid carried the reference\ncounter and the list of tasks attached to this pid.  At its end it has a\nvariable length array of struct upid-s.  Each struct upid has a numerical id\n(pid itself), pointer to the namespace, this ID is valid in and is hashed into\na pid_hash for searching the pids.\n\nThe nr and pid_chain fields are kept in struct pid for a while to make kernel\nstill work (no patch initialize the upids yet), but it will be removed at the\nend of this series when we switch to upids completely.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nSigned-off-by: Pavel Emelyanov \u003cxemul@openvz.org\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Paul Menage \u003cmenage@google.com\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "820e45db2380eb1545fa2bc5d34b8b2f2933faeb",
      "tree": "426e89e9205e44559d949d9cdf8bf1035bdc3710",
      "parents": [
        "e713d0dab21a68500720e222fa02567fc7dfb14b"
      ],
      "author": {
        "name": "Sukadev Bhattiprolu",
        "email": "sukadev@us.ibm.com",
        "time": "Thu May 10 22:23:00 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri May 11 08:29:35 2007 -0700"
      },
      "message": "statically initialize struct pid for swapper\n\nStatically initialize a struct pid for the swapper process (pid_t \u003d\u003d 0) and\nattach it to init_task.  This is needed so task_pid(), task_pgrp() and\ntask_session() interfaces work on the swapper process also.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Cedric Le Goater \u003cclg@fr.ibm.com\u003e\nCc: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nCc: Serge Hallyn \u003cserue@us.ibm.com\u003e\nCc: Eric Biederman \u003cebiederm@xmission.com\u003e\nCc: Herbert Poetzl \u003cherbert@13thfloor.at\u003e\nCc: \u003ccontainers@lists.osdl.org\u003e\nAcked-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "e713d0dab21a68500720e222fa02567fc7dfb14b",
      "tree": "a56f90ce94d9287b73da6db72ed0e73542a70a07",
      "parents": [
        "4ac24b3ba9016881b11646114bb5cd12cf23edd9"
      ],
      "author": {
        "name": "Sukadev Bhattiprolu",
        "email": "sukadev@us.ibm.com",
        "time": "Thu May 10 22:22:58 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Fri May 11 08:29:35 2007 -0700"
      },
      "message": "attach_pid() with struct pid parameter\n\nattach_pid() currently takes a pid_t and then uses find_pid() to find the\ncorresponding struct pid.  Sometimes we already have the struct pid.  We can\nthen skip find_pid() if attach_pid() were to take a struct pid parameter.\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nCc: Cedric Le Goater \u003cclg@fr.ibm.com\u003e\nCc: Dave Hansen \u003chaveblue@us.ibm.com\u003e\nCc: Serge Hallyn \u003cserue@us.ibm.com\u003e\nCc: \u003ccontainers@lists.osdl.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "9f57a54b6cf3f626334d97e93b5b917ad11e1efc",
      "tree": "d00c59378853d36b9508708e2d2ee38269eeb018",
      "parents": [
        "41487c65bfcce9c8e4d123da1719fcfd8df6d4d0"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Mon Feb 12 00:53:02 2007 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Mon Feb 12 09:48:32 2007 -0800"
      },
      "message": "[PATCH] pid: remove now unused do_each_task_pid and while_each_task_pid\n\nNow that I have changed all of the users remove the old version of these\nfunctions.  This should be a clear hint to any out of tree users that they\nshould use do_each_pid_task and while_each_pid_task for new code.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nCc: Alan Cox \u003calan@lxorguk.ukuu.org.uk\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "84d737866e2babdeab0c6b18ea155c6a649663b8",
      "tree": "e504da826174c2804d8c680828800aa680090686",
      "parents": [
        "6cc1b22a4acef3816eaa5f8c227d93d749b23195"
      ],
      "author": {
        "name": "Sukadev Bhattiprolu",
        "email": "sukadev@us.ibm.com",
        "time": "Fri Dec 08 02:38:01 2006 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.osdl.org",
        "time": "Fri Dec 08 08:28:52 2006 -0800"
      },
      "message": "[PATCH] add child reaper to pid_namespace\n\nAdd a per pid_namespace child-reaper.  This is needed so processes are reaped\nwithin the same pid space and do not spill over to the parent pid space.  Its\nalso needed so containers preserve existing semantic that pid \u003d\u003d 1 would reap\norphaned children.\n\nThis is based on Eric Biederman\u0027s patch: http://lkml.org/lkml/2006/2/6/285\n\nSigned-off-by: Sukadev Bhattiprolu \u003csukadev@us.ibm.com\u003e\nSigned-off-by: Cedric Le Goater \u003cclg@fr.ibm.com\u003e\nCc: Kirill Korotaev \u003cdev@openvz.org\u003e\nCc: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nCc: Herbert Poetzl \u003cherbert@13thfloor.at\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "1d32849b14bc8792e6f35ab27dd990d74b16126c",
      "tree": "caba51ce9df60e42412119063e582023a534507e",
      "parents": [
        "ad7a3bf30c27d902042bc85c3e54eb2082d633bd"
      ],
      "author": {
        "name": "Andrew Morton",
        "email": "akpm@osdl.org",
        "time": "Tue Oct 03 01:13:45 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Tue Oct 03 08:03:40 2006 -0700"
      },
      "message": "[PATCH] pid.h cleanup\n\nMake the pid.h macros look less revolting in an 80-col window.\n\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "1a657f78dcc8ea7c53eaa1f2a45ea2315738c15f",
      "tree": "a19972027792082fed505c8d540f7d877e37c0ab",
      "parents": [
        "1c0d04c9e44f4a248335c33d2be7c7f7b06ff359"
      ],
      "author": {
        "name": "Oleg Nesterov",
        "email": "oleg@tv-sign.ru",
        "time": "Mon Oct 02 02:18:59 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Oct 02 07:57:25 2006 -0700"
      },
      "message": "[PATCH] introduce get_task_pid() to fix unsafe get_pid()\n\nproc_pid_make_inode:\n\n\tei-\u003epid \u003d get_pid(task_pid(task));\n\nI think this is not safe.  get_pid() can be preempted after checking \"pid\n!\u003d NULL\".  Then the task exits, does detach_pid(), and RCU frees the pid.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "d387cae075b0aec479adbdfb71df39f7de8e9adb",
      "tree": "fe5c9a3b105f3edf4049c16dda4085c5f356fafe",
      "parents": [
        "c88be3eb2e01bbb21c9ccdc3805f0d3546c1898c"
      ],
      "author": {
        "name": "Oleg Nesterov",
        "email": "oleg@tv-sign.ru",
        "time": "Mon Oct 02 02:17:22 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Oct 02 07:57:15 2006 -0700"
      },
      "message": "[PATCH] pid: simplify pid iterators\n\nI think it is hardly possible to read the current do_each_task_pid().  The\nnew version is much simpler and makes the code smaller.\n\nOnly the do_each_task_pid change is tested, the do_each_pid_task isn\u0027t.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: \"Eric W. Biederman\" \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "5feb8f5f8403d8874a04aac443692dfe83bd63d2",
      "tree": "8dc23051350f9a4bd22c3172dc570e96e73f3e50",
      "parents": [
        "bbf73147e2d46611fbdcbc126f887c614c32350b"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Mon Oct 02 02:17:12 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Oct 02 07:57:13 2006 -0700"
      },
      "message": "[PATCH] pid: implement pid_nr\n\nAs we stop storing pid_t\u0027s and move to storing struct pid *.  We need a way to\nget the pid_t from the struct pid to report to user space what we have stored.\n\nHaving a clean well defined way to do this is especially important as we move\nto multiple pid spaces as may need to report a different value to the caller\ndepending on which pid space the caller is in.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "558cb325485aaf655130f140e8ddd25392f6c972",
      "tree": "adfea6c0b901cf6591209d29091ff5ac298843c3",
      "parents": [
        "22c935f47c03399c78e64c71b757eb36fa917ff6"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Mon Oct 02 02:17:09 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Oct 02 07:57:13 2006 -0700"
      },
      "message": "[PATCH] pid: add do_each_pid_task\n\nTo avoid pid rollover confusion the kernel needs to work with struct pid *\ninstead of pid_t.  Currently there is not an iterator that walks through all\nof the tasks of a given pid type starting with a struct pid.  This prevents us\nreplacing some pid_t instances with struct pid.  So this patch adds\ndo_each_pid_task which walks through the set of task for a given pid type\nstarting with a struct pid.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "0804ef4b0de7121261f77c565b20a11ac694e877",
      "tree": "ff12e3b999dc2ce66d97fce5d76cd7df073c0d5c",
      "parents": [
        "2bc2d61a9638dab670d8361e928d1a5a291173ef"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Mon Oct 02 02:17:04 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Oct 02 07:57:12 2006 -0700"
      },
      "message": "[PATCH] proc: readdir race fix (take 3)\n\nThe problem: An opendir, readdir, closedir sequence can fail to report\nprocess ids that are continually in use throughout the sequence of system\ncalls.  For this race to trigger the process that proc_pid_readdir stops at\nmust exit before readdir is called again.\n\nThis can cause ps to fail to report processes, and it is in violation of\nposix guarantees and normal application expectations with respect to\nreaddir.\n\nCurrently there is no way to work around this problem in user space short\nof providing a gargantuan buffer to user space so the directory read all\nhappens in on system call.\n\nThis patch implements the normal directory semantics for proc, that\nguarantee that a directory entry that is neither created nor destroyed\nwhile reading the directory entry will be returned.  For directory that are\neither created or destroyed during the readdir you may or may not see them.\n Furthermore you may seek to a directory offset you have previously seen.\n\nThese are the guarantee that ext[23] provides and that posix requires, and\nmore importantly that user space expects.  Plus it is a simple semantic to\nimplement reliable service.  It is just a matter of calling readdir a\nsecond time if you are wondering if something new has show up.\n\nThese better semantics are implemented by scanning through the pids in\nnumerical order and by making the file offset a pid plus a fixed offset.\n\nThe pid scan happens on the pid bitmap, which when you look at it is\nremarkably efficient for a brute force algorithm.  Given that a typical\ncache line is 64 bytes and thus covers space for 64*8 \u003d\u003d 200 pids.  There\nare only 40 cache lines for the entire 32K pid space.  A typical system\nwill have 100 pids or more so this is actually fewer cache lines we have to\nlook at to scan a linked list, and the worst case of having to scan the\nentire pid bitmap is pretty reasonable.\n\nIf we need something more efficient we can go to a more efficient data\nstructure for indexing the pids, but for now what we have should be\nsufficient.\n\nIn addition this takes no additional locks and is actually less code than\nwhat we are doing now.\n\nAlso another very subtle bug in this area has been fixed.  It is possible\nto catch a task in the middle of de_thread where a thread is assuming the\nthread of it\u0027s thread group leader.  This patch carefully handles that case\nso if we hit it we don\u0027t fail to return the pid, that is undergoing the\nde_thread dance.\n\nThanks to KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e for\nproviding the first fix, pointing this out and working on it.\n\n[oleg@tv-sign.ru: fix it]\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nAcked-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Jean Delvare \u003cjdelvare@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "c18258c6f0848f97e85287f6271c511a092bb784",
      "tree": "16c057a171b7623895ee208459392c1104193b84",
      "parents": [
        "35fa2048ab13d1be846be612e395c15c200bd51c"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Wed Sep 27 01:51:06 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Wed Sep 27 08:26:19 2006 -0700"
      },
      "message": "[PATCH] pid: Implement transfer_pid and use it to simplify de_thread\n\nIn de_thread we move pids from one process to another, a rather ugly case.\nThe function transfer_pid makes it clear what we are doing, and makes the\naction atomic.  This is useful we ever want to atomically traverse the\nprocess group and session lists, in a rcu safe manner.\n\nEven if the atomic properties this change should be a win as transfer_pid\nshould be less code to execute than executing both attach_pid and\ndetach_pid, and this should make de_thread slightly smaller as only a\nsingle function call needs to be emitted.  The only downside is that the\ncode might be slower to execute as the odds are against transfer_pid being\nin cache.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "92476d7fc0326a409ab1d3864a04093a6be9aca7",
      "tree": "ea50a5a31522492d9915e0763a7adc6ac87c4fbc",
      "parents": [
        "8c7904a00b06d2ee51149794b619e07369fcf9d4"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Fri Mar 31 02:31:42 2006 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Fri Mar 31 12:19:00 2006 -0800"
      },
      "message": "[PATCH] pidhash: Refactor the pid hash table\n\nSimplifies the code, reduces the need for 4 pid hash tables, and makes the\ncode more capable.\n\nIn the discussions I had with Oleg it was felt that to a large extent the\ncleanup itself justified the work.  With struct pid being dynamically\nallocated meant we could create the hash table entry when the pid was\nallocated and free the hash table entry when the pid was freed.  Instead of\nplaying with the hash lists when ever a process would attach or detach to a\nprocess.\n\nFor myself the fact that it gave what my previous task_ref patch gave for free\nwith simpler code was a big win.  The problem is that if you hold a reference\nto struct task_struct you lock in 10K of low memory.  If you do that in a user\ncontrollable way like /proc does, with an unprivileged but hostile user space\napplication with typical resource limits of 1000 fds and 100 processes I can\ntrigger the OOM killer by consuming all of low memory with task structs, on a\nmachine wight 1GB of low memory.\n\nIf I instead hold a reference to struct pid which holds a pointer to my\ntask_struct, I don\u0027t suffer from that problem because struct pid is 2 orders\nof magnitude smaller.  In fact struct pid is small enough that most other\nkernel data structures dwarf it, so simply limiting the number of referring\ndata structures is enough to prevent exhaustion of low memory.\n\nThis splits the current struct pid into two structures, struct pid and struct\npid_link, and reduces our number of hash tables from PIDTYPE_MAX to just one.\nstruct pid_link is the per process linkage into the hash tables and lives in\nstruct task_struct.  struct pid is given an indepedent lifetime, and holds\npointers to each of the pid types.\n\nThe independent life of struct pid simplifies attach_pid, and detach_pid,\nbecause we are always manipulating the list of pids and not the hash table.\nIn addition in giving struct pid an indpendent life it makes the concept much\nmore powerful.\n\nKernel data structures can now embed a struct pid * instead of a pid_t and\nnot suffer from pid wrap around problems or from keeping unnecessarily\nlarge amounts of memory allocated.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "47e65328a7b1cdfc4e3102e50d60faf94ebba7d3",
      "tree": "78e00a5321cca87767806a91e623e71b5c6637c7",
      "parents": [
        "88531f725bd52e37a7be726860e4ff3f09031d89"
      ],
      "author": {
        "name": "Oleg Nesterov",
        "email": "oleg@tv-sign.ru",
        "time": "Tue Mar 28 16:11:25 2006 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Tue Mar 28 18:36:44 2006 -0800"
      },
      "message": "[PATCH] pids: kill PIDTYPE_TGID\n\nThis patch kills PIDTYPE_TGID pid_type thus saving one hash table in\nkernel/pid.c and speeding up subthreads create/destroy a bit.  It is also a\npreparation for the further tref/pids rework.\n\nThis patch adds \u0027struct list_head thread_group\u0027 to \u0027struct task_struct\u0027\ninstead.\n\nWe don\u0027t detach group leader from PIDTYPE_PID namespace until another\nthread inherits it\u0027s -\u003epid \u003d\u003d -\u003etgid, so we are safe wrt premature\nfree_pidmap(-\u003etgid) call.\n\nCurrently there are no users of find_task_by_pid_type(PIDTYPE_TGID).\nShould the need arise, we can use find_task_by_pid()-\u003egroup_leader.\n\nSigned-off-by: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nAcked-By: Eric Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "d73d65293e3e2de7e916a89c8da30be0948afab7",
      "tree": "62b0ca76e3099c927f2960856dd060b136a8ccef",
      "parents": [
        "652486fb84a07ed750f1c11639518f55808bf555"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Tue Mar 28 16:11:03 2006 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Tue Mar 28 18:36:40 2006 -0800"
      },
      "message": "[PATCH] pidhash: kill switch_exec_pids\n\nswitch_exec_pids is only called from de_thread by way of exec, and it is\nonly called when we are exec\u0027ing from a non thread group leader.\n\nCurrently switch_exec_pids gives the leader the pid of the thread and\nunhashes and rehashes all of the process groups.  The leader is already in\nthe EXIT_DEAD state so no one cares about it\u0027s pids.  The only concern for\nthe leader is that __unhash_process called from release_task will function\ncorrectly.  If we don\u0027t touch the leader at all we know that\n__unhash_process will work fine so there is no need to touch the leader.\n\nFor the task becomming the thread group leader, we just need to give it the\npid of the old thread group leader, add it to the task list, and attach it\nto the session and the process group of the thread group.\n\nCurrently de_thread is also adding the task to the task list which is just\nsilly.\n\nCurrently the only leader of __detach_pid besides detach_pid is\nswitch_exec_pids because of the ugly extra work that was being\nperformed.\n\nSo this patch removes switch_exec_pids because it is doing too much, it is\ncreating an unnecessary special case in pid.c, duing work duplicated in\nde_thread, and generally obscuring what it is going on.\n\nThe necessary work is added to de_thread, and it seems to be a little\nclearer there what is going on.\n\nSigned-off-by: Eric W. Biederman \u003cebiederm@xmission.com\u003e\nCc: Oleg Nesterov \u003coleg@tv-sign.ru\u003e\nCc: Kirill Korotaev \u003cdev@sw.ru\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "1da177e4c3f41524e886b7f1b8a0c1fc7321cac2",
      "tree": "0bba044c4ce775e45a88a51686b5d9f90697ea9d",
      "parents": [],
      "author": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Apr 16 15:20:36 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Apr 16 15:20:36 2005 -0700"
      },
      "message": "Linux-2.6.12-rc2\n\nInitial git repository build. I\u0027m not bothering with the full history,\neven though we have it. We can create a separate \"historical\" git\narchive of that later if we want to, and in the meantime it\u0027s about\n3.2GB when imported into git - space that would just make the early\ngit days unnecessarily complicated, when we don\u0027t have a lot of good\ninfrastructure for it.\n\nLet it rip!\n"
    }
  ]
}