)]}'
{
  "log": [
    {
      "commit": "7ea931c9fc80c4d0a4306c30ec92eb0f1d922a0b",
      "tree": "4f2c85d658c75e8679e01943e3e95866b3729b54",
      "parents": [
        "f5b087b52f1710eb0bf15a2d2b030c51a6a1ca9e"
      ],
      "author": {
        "name": "Paul Jackson",
        "email": "pj@sgi.com",
        "time": "Mon Apr 28 02:12:29 2008 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@linux-foundation.org",
        "time": "Mon Apr 28 08:58:19 2008 -0700"
      },
      "message": "mempolicy: add bitmap_onto() and bitmap_fold() operations\n\nThe following adds two more bitmap operators, bitmap_onto() and bitmap_fold(),\nwith the usual cpumask and nodemask wrappers.\n\nThe bitmap_onto() operator computes one bitmap relative to another.  If the\nn-th bit in the origin mask is set, then the m-th bit of the destination mask\nwill be set, where m is the position of the n-th set bit in the relative mask.\n\nThe bitmap_fold() operator folds a bitmap into a second that has bit m set iff\nthe input bitmap has some bit n set, where m \u003d\u003d n mod sz, for the specified sz\nvalue.\n\nThere are two substantive changes between this patch and its\npredecessor bitmap_relative:\n 1) Renamed bitmap_relative() to be bitmap_onto().\n 2) Added bitmap_fold().\n\nThe essential motivation for bitmap_onto() is to provide a mechanism for\nconverting a cpuset-relative CPU or Node mask to an absolute mask.  Cpuset\nrelative masks are written as if the current task were in a cpuset whose CPUs\nor Nodes were just the consecutive ones numbered 0..N-1, for some N.  The\nbitmap_onto() operator is provided in anticipation of adding support for the\nfirst such cpuset relative mask, by the mbind() and set_mempolicy() system\ncalls, using a planned flag of MPOL_F_RELATIVE_NODES.  These bitmap operators\n(and their nodemask wrappers, in particular) will be used in code that\nconverts the user specified cpuset relative memory policy to a specific system\nnode numbered policy, given the current mems_allowed of the tasks cpuset.\n\nSuch cpuset relative mempolicies will address two deficiencies\nof the existing interface between cpusets and mempolicies:\n 1) A task cannot at present reliably establish a cpuset\n    relative mempolicy because there is an essential race\n    condition, in that the tasks cpuset may be changed in\n    between the time the task can query its cpuset placement,\n    and the time the task can issue the applicable mbind or\n    set_memplicy system call.\n 2) A task cannot at present establish what cpuset relative\n    mempolicy it would like to have, if it is in a smaller\n    cpuset than it might have mempolicy preferences for,\n    because the existing interface only allows specifying\n    mempolicies for nodes currently allowed by the cpuset.\n\nCpuset relative mempolicies are useful for tasks that don\u0027t distinguish\nparticularly between one CPU or Node and another, but only between how many of\neach are allowed, and the proper placement of threads and memory pages on the\nvarious CPUs and Nodes available.\n\nThe motivation for the added bitmap_fold() can be seen in the following\nexample.\n\nLet\u0027s say an application has specified some mempolicies that presume 16 memory\nnodes, including say a mempolicy that specified MPOL_F_RELATIVE_NODES (cpuset\nrelative) nodes 12-15.  Then lets say that application is crammed into a\ncpuset that only has 8 memory nodes, 0-7.  If one just uses bitmap_onto(),\nthis mempolicy, mapped to that cpuset, would ignore the requested relative\nnodes above 7, leaving it empty of nodes.  That\u0027s not good; better to fold the\nhigher nodes down, so that some nodes are included in the resulting mapped\nmempolicy.  In this case, the mempolicy nodes 12-15 are taken modulo 8 (the\nweight of the mems_allowed of the confining cpuset), resulting in a mempolicy\nspecifying nodes 4-7.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: David Rientjes \u003crientjes@google.com\u003e\nCc: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Andi Kleen \u003cak@suse.de\u003e\nCc: Mel Gorman \u003cmel@csn.ul.ie\u003e\nCc: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: \u003ckosaki.motohiro@jp.fujitsu.com\u003e\nCc: \u003cray-lk@madrabbit.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "37c0708dbee5825df3bd9ce6ef2199c6c1713970",
      "tree": "747551aa58484e7f872da118b864c8f3ca6e892d",
      "parents": [
        "56bbd65df0e92a4a8eb70c5f2b416ae2b6c5fb31"
      ],
      "author": {
        "name": "Christoph Lameter",
        "email": "clameter@sgi.com",
        "time": "Tue Oct 16 01:25:36 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Tue Oct 16 09:42:58 2007 -0700"
      },
      "message": "Memoryless nodes: Add N_CPU node state\n\nWe need the check for a node with cpu in zone reclaim.  Zone reclaim will not\nallow remote zone reclaim if a node has a cpu.\n\n[Lee.Schermerhorn@hp.com: Move setup of N_CPU node state mask]\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nTested-by:  Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nAcked-by: Bob Picco \u003cbob.picco@hp.com\u003e\nCc: Nishanth Aravamudan \u003cnacc@us.ibm.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@skynet.ie\u003e\nSigned-off-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "7ea1530ab3fdfa85441061909cc8040e84776fd4",
      "tree": "f3af0f8ed40a6df90bdbb9396d6163d59798a821",
      "parents": [
        "13808910713a98cc1159291e62cdfec92cc94d05"
      ],
      "author": {
        "name": "Christoph Lameter",
        "email": "clameter@sgi.com",
        "time": "Tue Oct 16 01:25:29 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Tue Oct 16 09:42:58 2007 -0700"
      },
      "message": "Memoryless nodes: introduce mask of nodes with memory\n\nIt is necessary to know if nodes have memory since we have recently begun to\nadd support for memoryless nodes.  For that purpose we introduce a two new\nnode states: N_HIGH_MEMORY and N_NORMAL_MEMORY.\n\nA node has its bit in N_HIGH_MEMORY set if it has any memory regardless of the\ntype of mmemory.  If a node has memory then it has at least one zone defined\nin its pgdat structure that is located in the pgdat itself.\n\nA node has its bit in N_NORMAL_MEMORY set if it has a lower zone than\nZONE_HIGHMEM.  This means it is possible to allocate memory that is not\nsubject to kmap.\n\nN_HIGH_MEMORY and N_NORMAL_MEMORY can then be used in various places to insure\nthat we do the right thing when we encounter a memoryless node.\n\n[akpm@linux-foundation.org: build fix]\n[Lee.Schermerhorn@hp.com: update N_HIGH_MEMORY node state for memory hotadd]\n[y-goto@jp.fujitsu.com: Fix memory hotplug + sparsemem build]\nSigned-off-by: Lee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e\nSigned-off-by: Nishanth Aravamudan \u003cnacc@us.ibm.com\u003e\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nAcked-by: Bob Picco \u003cbob.picco@hp.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@skynet.ie\u003e\nSigned-off-by: Yasunori Goto \u003cy-goto@jp.fujitsu.com\u003e\nSigned-off-by: Paul Mundt \u003clethal@linux-sh.org\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "13808910713a98cc1159291e62cdfec92cc94d05",
      "tree": "0fd7189dc2a76e1ae165ca5d6e8c6b4e6f1761af",
      "parents": [
        "55144768e100b68447f44c5e5c9deb155ad661bd"
      ],
      "author": {
        "name": "Christoph Lameter",
        "email": "clameter@sgi.com",
        "time": "Tue Oct 16 01:25:27 2007 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Tue Oct 16 09:42:58 2007 -0700"
      },
      "message": "Memoryless nodes: Generic management of nodemasks for various purposes\n\nWhy do we need to support memoryless nodes?\n\nKAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e wrote:\n\n\u003e For fujitsu, problem is called \"empty\" node.\n\u003e\n\u003e When ACPI\u0027s SRAT table includes \"possible nodes\", ia64 bootstrap(acpi_numa_init)\n\u003e creates nodes, which includes no memory, no cpu.\n\u003e\n\u003e I tried to remove empty-node in past, but that was denied.\n\u003e It was because we can hot-add cpu to the empty node.\n\u003e (node-hotplug triggered by cpu is not implemented now. and it will be ugly.)\n\u003e\n\u003e\n\u003e For HP, (Lee can comment on this later), they have memory-less-node.\n\u003e As far as I hear, HP\u0027s machine can have following configration.\n\u003e\n\u003e (example)\n\u003e Node0: CPU0   memory AAA MB\n\u003e Node1: CPU1   memory AAA MB\n\u003e Node2: CPU2   memory AAA MB\n\u003e Node3: CPU3   memory AAA MB\n\u003e Node4: Memory XXX GB\n\u003e\n\u003e AAA is very small value (below 16MB)  and will be omitted by ia64 bootstrap.\n\u003e After boot, only Node 4 has valid memory (but have no cpu.)\n\u003e\n\u003e Maybe this is memory-interleave by firmware config.\n\nChristoph Lameter \u003cclameter@sgi.com\u003e wrote:\n\n\u003e Future SGI platforms (actually also current one can have but nothing like\n\u003e that is deployed to my knowledge) have nodes with only cpus. Current SGI\n\u003e platforms have nodes with just I/O that we so far cannot manage in the\n\u003e core. So the arch code maps them to the nearest memory node.\n\nLee Schermerhorn \u003cLee.Schermerhorn@hp.com\u003e wrote:\n\n\u003e For the HP platforms, we can configure each cell with from 0% to 100%\n\u003e \"cell local memory\".  When we configure with \u003c100% CLM, the \"missing\n\u003e percentages\" are interleaved by hardware on a cache-line granularity to\n\u003e improve bandwidth at the expense of latency for numa-challenged\n\u003e applications [and OSes, but not our problem ;-)].  When we boot Linux on\n\u003e such a config, all of the real nodes have no memory--it all resides in a\n\u003e single interleaved pseudo-node.\n\u003e\n\u003e When we boot Linux on a 100% CLM configuration [\u003d\u003d NUMA], we still have\n\u003e the interleaved pseudo-node.  It contains a few hundred MB stolen from\n\u003e the real nodes to contain the DMA zone.  [Interleaved memory resides at\n\u003e phys addr 0].  The memoryless-nodes patches, along with the zoneorder\n\u003e patches, support this config as well.\n\u003e\n\u003e Also, when we boot a NUMA config with the \"mem\u003d\" command line,\n\u003e specifying less memory than actually exists, Linux takes the excluded\n\u003e memory \"off the top\" rather than distributing it across the nodes.  This\n\u003e can result in memoryless nodes, as well.\n\u003e\n\nThis patch:\n\nPreparation for memoryless node patches.\n\nProvide a generic way to keep nodemasks describing various characteristics of\nNUMA nodes.\n\nRemove the node_online_map and the node_possible map and realize the same\nfunctionality using two nodes stats: N_POSSIBLE and N_ONLINE.\n\n[Lee.Schermerhorn@hp.com: Initialize N_*_MEMORY and N_CPU masks for non-NUMA config]\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nTested-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nAcked-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nAcked-by: Bob Picco \u003cbob.picco@hp.com\u003e\nCc: Nishanth Aravamudan \u003cnacc@us.ibm.com\u003e\nCc: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nCc: Mel Gorman \u003cmel@skynet.ie\u003e\nSigned-off-by: Lee Schermerhorn \u003clee.schermerhorn@hp.com\u003e\nCc: \"Serge E. Hallyn\" \u003cserge@hallyn.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "74c7aa8b8581e0ba8d6d17c623b9279aaabbb0cf",
      "tree": "e8bfdd1d4bd5a7d4ee0e0bbf83c45c9f2b5deb59",
      "parents": [
        "5ec553a90448b3edbd26c1acc72464f877614bfa"
      ],
      "author": {
        "name": "Christoph Lameter",
        "email": "clameter@sgi.com",
        "time": "Tue Feb 20 13:57:51 2007 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@woody.linux-foundation.org",
        "time": "Tue Feb 20 17:10:13 2007 -0800"
      },
      "message": "[PATCH] Replace highest_possible_node_id() with nr_node_ids\n\nhighest_possible_node_id() is currently used to calculate the last possible\nnode idso that the network subsystem can figure out how to size per node\narrays.\n\nI think having the ability to determine the maximum amount of nodes in a\nsystem at runtime is useful but then we should name this entry\ncorrespondingly, it should return the number of node_ids, and the the value\nneeds to be setup only once on bootup.  The node_possible_map does not\nchange after bootup.\n\nThis patch introduces nr_node_ids and replaces the use of\nhighest_possible_node_id().  nr_node_ids is calculated on bootup when the\npage allocators pagesets are initialized.\n\n[deweerdt@free.fr: fix oops]\nSigned-off-by: Christoph Lameter \u003cclameter@sgi.com\u003e\nCc: Neil Brown \u003cneilb@suse.de\u003e\nCc: Trond Myklebust \u003ctrond.myklebust@fys.uio.no\u003e\nSigned-off-by: Frederik Deweerdt \u003cfrederik.deweerdt@gmail.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@linux-foundation.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@linux-foundation.org\u003e\n"
    },
    {
      "commit": "01a3ee2b203e511e20f98b85a9172fd32c53e87c",
      "tree": "0dd90d81dc86f231828af23bdb97522405b06cab",
      "parents": [
        "39484e53bb00f55b6303a908070db133608ef2a5"
      ],
      "author": {
        "name": "Reinette Chatre",
        "email": "reinette.chatre@linux.intel.com",
        "time": "Wed Oct 11 01:21:55 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Wed Oct 11 11:14:22 2006 -0700"
      },
      "message": "[PATCH] bitmap: parse input from kernel and user buffers\n\nlib/bitmap.c:bitmap_parse() is a library function that received as input a\nuser buffer.  This seemed to have originated from the way the write_proc\nfunction of the /proc filesystem operates.\n\nThis has been reworked to not use kmalloc and eliminates a lot of\nget_user() overhead by performing one access_ok before using __get_user().\n\nWe need to test if we are in kernel or user space (is_user) and access the\nbuffer differently.  We cannot use __get_user() to access kernel addresses\nin all cases, for example in architectures with separate address space for\nkernel and user.\n\nThis function will be useful for other uses as well; for example, taking\ninput for /sysfs instead of /proc, so it was changed to accept kernel\nbuffers.  We have this use for the Linux UWB project, as part as the\nupcoming bandwidth allocator code.\n\nOnly a few routines used this function and they were changed too.\n\nSigned-off-by: Reinette Chatre \u003creinette.chatre@linux.intel.com\u003e\nSigned-off-by: Inaky Perez-Gonzalez \u003cinaky@linux.intel.com\u003e\nCc: Paul Jackson \u003cpj@sgi.com\u003e\nCc: Joe Korty \u003cjoe.korty@ccur.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "0f532f3861d2c4e5aa7dcd33fb18e9975eb28457",
      "tree": "4b4cf436bbc94a1659fdb25314ff7b17624e80a1",
      "parents": [
        "ca8af486765852302931bb69075871d5564e1e5b"
      ],
      "author": {
        "name": "Greg Banks",
        "email": "gnb@melbourne.sgi.com",
        "time": "Mon Oct 02 02:17:39 2006 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Oct 02 07:57:17 2006 -0700"
      },
      "message": "[PATCH] cpumask: add highest_possible_node_id\n\ncpumask: add highest_possible_node_id(), analogous to\nhighest_possible_processor_id().\n\n[pj@sgi.com: fix typo]\nSigned-off-by: Greg Banks \u003cgnb@melbourne.sgi.com\u003e\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "8357f8695d58b50fbf2bd507b4b0fc2cd1e43bd6",
      "tree": "8f7326f570ee80d129add7356c7b5c44fb995447",
      "parents": [
        "a0140c1d85637ee5f4ea7c78f066e3611a6a79dc"
      ],
      "author": {
        "name": "KAMEZAWA Hiroyuki",
        "email": "kamezawa.hiroyu@jp.fujitsu.com",
        "time": "Mon Mar 27 01:15:57 2006 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Mar 27 08:44:47 2006 -0800"
      },
      "message": "[PATCH] define for_each_online_pgdat\n\nThis patch defines for_each_online_pgdat() as a replacement of\nfor_each_pgdat()\n\nNow, online nodes are managed by node_online_map.  But for_each_pgdat()\nuses pgdat_link to iterate over all nodes(pgdat).  This means management\nstructure for online pgdat is duplicated.\n\nI think using node_online_map for for_each_pgdat() is simple and sane\nrather ather than pgdat_link.  New macro is named as\nfor_each_online_pgdat().  Following patch will fix callers of\nfor_each_pgdat().\n\nThe bootmem allocater uses for_each_pgdat() before pgdat initialization.  I\ndon\u0027t think it\u0027s sane.  Following patch will fix it.\n\nSigned-off-by: Yasunori Goto     \u003cy-goto@jp.fujitsu.com\u003e\nSigned-off-by: KAMEZAWA Hiroyuki \u003ckamezawa.hiroyu@jp.fujitsu.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "1b8623545b42c03eb92e51b28c84acf4b8ba00a3",
      "tree": "071045ad9c60d2697292c523c77322a70a248fb9",
      "parents": [
        "92118c739df879497b8cc5a2eb3a9dc255f01b20"
      ],
      "author": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Thu Dec 15 01:07:03 2005 -0500"
      },
      "committer": {
        "name": "Al Viro",
        "email": "viro@zeniv.linux.org.uk",
        "time": "Tue Feb 07 20:56:35 2006 -0500"
      },
      "message": "[PATCH] remove bogus asm/bug.h includes.\n\nA bunch of asm/bug.h includes are both not needed (since it will get\npulled anyway) and bogus (since they are done too early).  Removed.\n\nSigned-off-by: Al Viro \u003cviro@zeniv.linux.org.uk\u003e\n"
    },
    {
      "commit": "fb5eeeee44edb248b4837416966f19731f497f79",
      "tree": "f947a4dcf103f55d526bb5c71f69b657d8f22e61",
      "parents": [
        "28a42b9ea7e42e1efb02cc2dcacba0b6af234e1b"
      ],
      "author": {
        "name": "Paul Jackson",
        "email": "pj@sgi.com",
        "time": "Sun Oct 30 15:02:33 2005 -0800"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Sun Oct 30 17:37:21 2005 -0800"
      },
      "message": "[PATCH] cpusets: bitmap and mask remap operators\n\nIn the forthcoming task migration support, a key calculation will be\nmapping cpu and node numbers from the old set to the new set while\npreserving cpuset-relative offset.\n\nFor example, if a task and its pages on nodes 8-11 are being migrated to\nnodes 24-27, then pages on node 9 (the 2nd node in the old set) should be\nmoved to node 25 (the 2nd node in the new set.)\n\nAs with other bitmap operations, the proper way to code this is to provide\nthe underlying calculation in lib/bitmap.c, and then to provide the usual\ncpumask and nodemask wrappers.\n\nThis patch provides that.  These operations are termed \u0027remap\u0027 operations.\nBoth remapping a single bit and a set of bits is supported.\n\nSigned-off-by: Paul Jackson \u003cpj@sgi.com\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"
    }
  ]
}