)]}'
{
  "log": [
    {
      "commit": "6cb54819d7b1867053e2dfd8c0ca3a8dc65a7eff",
      "tree": "1a1422dc2e103fe92dd86bfa26b8b39b3f2413d5",
      "parents": [
        "5d546f54324e04747e82ccbb4ea85f54bdcacd6d"
      ],
      "author": {
        "name": "Ingo Molnar",
        "email": "mingo@elte.hu",
        "time": "Mon Aug 01 13:39:13 2005 +0200"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Mon Aug 01 10:03:56 2005 -0700"
      },
      "message": "[PATCH] remove sys_set_zone_reclaim()\n\nThis removes sys_set_zone_reclaim() for now.  While i\u0027m sure Martin is\ntrying to solve a real problem, we must not hard-code an incomplete and\ninsufficient approach into a syscall, because syscalls are pretty much\nfor eternity.  I am quite strongly convinced that this syscall must not\nhit v2.6.13 in its current form.\n\nFirstly, the syscall lacks basic syscall design: e.g. it allows the\nglobal setting of VM policy for unprivileged users. (!) [ Imagine an\nOracle installation and a SAP installation on the same NUMA box fighting\nover the \u0027optimal\u0027 setting for this flag. What will they do? Will they\ntry to set the flag to their own preferred value every second or so? ]\n\nSecondly, it was added based on a single datapoint from Martin:\n\n http://marc.theaimsgroup.com/?l\u003dlinux-mm\u0026m\u003d111763597218177\u0026w\u003d2\n\nwhere Martin characterizes the numbers the following way:\n\n \u0027 Run-to-run variability for \"make -j\" is huge, so these numbers aren\u0027t\n   terribly useful except to see that with reclaim the benchmark still\n   finishes in a reasonable amount of time. \u0027\n\nin other words: the fundamental problem has likely not been solved, only\na tendential move into the right direction has been observed, and a\nhandful of numbers were picked out of a set of hugely variable results,\nwithout showing the variability data. How much variance is there\nrun-to-run?\n\nI\u0027d really suggest to first walk the walk and see what\u0027s needed to get\nstable \u0026 predictable kernel compilation numbers on that NUMA box, before\nadding random syscalls to tune a particular aspect of the VM ... which\napproach might not even matter once the whole picture has been analyzed\nand understood!\n\nThe third, most important point is that the syscall exposes VM tuning\ninternals in a completely unstructured way. What sense does it make to\nhave a _GLOBAL_ per-node setting for \u0027should we go to another node for\nreclaim\u0027? If then it might make sense to do this per-app, via numalib or\nso.\n\nThe change is minimalistic in that it doesnt remove the syscall and the\nunderlying infrastructure changes, only the user-visible changes.  We\ncould perhaps add a CAP_SYS_ADMIN-only sysctl for this hack, a\u0027ka\n/proc/sys/vm/swappiness, but even that looks quite counterproductive\nwhen the generic approach is that we are trying to reduce the number of\nexternal factors in the VM balance picture.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "0eeca28300df110bd6ed54b31193c83b87921443",
      "tree": "7db42d8a18d80eca538f5b7d25e0532b8fa38b85",
      "parents": [
        "bd4c625c061c2a38568d0add3478f59172455159"
      ],
      "author": {
        "name": "Robert Love",
        "email": "rml@novell.com",
        "time": "Tue Jul 12 17:06:03 2005 -0400"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Tue Jul 12 20:38:38 2005 -0700"
      },
      "message": "[PATCH] inotify\n\ninotify is intended to correct the deficiencies of dnotify, particularly\nits inability to scale and its terrible user interface:\n\n        * dnotify requires the opening of one fd per each directory\n          that you intend to watch. This quickly results in too many\n          open files and pins removable media, preventing unmount.\n        * dnotify is directory-based. You only learn about changes to\n          directories. Sure, a change to a file in a directory affects\n          the directory, but you are then forced to keep a cache of\n          stat structures.\n        * dnotify\u0027s interface to user-space is awful.  Signals?\n\ninotify provides a more usable, simple, powerful solution to file change\nnotification:\n\n        * inotify\u0027s interface is a system call that returns a fd, not SIGIO.\n\t  You get a single fd, which is select()-able.\n        * inotify has an event that says \"the filesystem that the item\n          you were watching is on was unmounted.\"\n        * inotify can watch directories or files.\n\nInotify is currently used by Beagle (a desktop search infrastructure),\nGamin (a FAM replacement), and other projects.\n\nSee Documentation/filesystems/inotify.txt.\n\nSigned-off-by: Robert Love \u003crml@novell.com\u003e\nCc: John McCutchan \u003cttb@tentacle.dhs.org\u003e\nCc: Christoph Hellwig \u003chch@lst.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "dc009d92435f99498cbc579ce76bf28e837e2c14",
      "tree": "2ba8732b28225593d996b8faa079dc6ab4bbc9bc",
      "parents": [
        "d0537508a9921efced238b20967e50e519ac34af"
      ],
      "author": {
        "name": "Eric W. Biederman",
        "email": "ebiederm@xmission.com",
        "time": "Sat Jun 25 14:57:52 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Jun 25 16:24:48 2005 -0700"
      },
      "message": "[PATCH] kexec: add kexec syscalls\n\nThis patch introduces the architecture independent implementation the\nsys_kexec_load, the compat_sys_kexec_load system calls.\n\nKexec on panic support has been integrated into the core patch and is\nrelatively clean.\n\nIn addition the hopefully architecture independent option\ncrashkernel\u003dsize@location has been docuemented.  It\u0027s purpose is to reserve\nspace for the panic kernel to live, and where no DMA transfer will ever be\nsetup to access.\n\nSigned-off-by: Eric Biederman \u003cebiederm@xmission.com\u003e\nSigned-off-by: Alexander Nyberg \u003calexn@telia.com\u003e\nSigned-off-by: Adrian Bunk \u003cbunk@stusta.de\u003e\nSigned-off-by: Vivek Goyal \u003cvgoyal@in.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "753ee728964e5afb80c17659cc6c3a6fd0a42fe0",
      "tree": "41c9a7700d0858c1f77c5bdaba97e5b636f69b06",
      "parents": [
        "bfbb38fb808ac23ef44472d05d9bb36edfb49ed0"
      ],
      "author": {
        "name": "Martin Hicks",
        "email": "mort@sgi.com",
        "time": "Tue Jun 21 17:14:41 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Tue Jun 21 18:46:14 2005 -0700"
      },
      "message": "[PATCH] VM: early zone reclaim\n\nThis is the core of the (much simplified) early reclaim.  The goal of this\npatch is to reclaim some easily-freed pages from a zone before falling back\nonto another zone.\n\nOne of the major uses of this is NUMA machines.  With the default allocator\nbehavior the allocator would look for memory in another zone, which might be\noff-node, before trying to reclaim from the current zone.\n\nThis adds a zone tuneable to enable early zone reclaim.  It is selected on a\nper-zone basis and is turned on/off via syscall.\n\nAdding some extra throttling on the reclaim was also required (patch\n4/4).  Without the machine would grind to a crawl when doing a \"make -j\"\nkernel build.  Even with this patch the System Time is higher on\naverage, but it seems tolerable.  Here are some numbers for kernbench\nruns on a 2-node, 4cpu, 8Gig RAM Altix in the \"make -j\" run:\n\n\t\t\twall  user   sys   %cpu  ctx sw.  sleeps\n\t\t\t----  ----   ---   ----   ------  ------\nNo patch\t\t1009  1384   847   258   298170   504402\nw/patch, no reclaim     880   1376   667   288   254064   396745\nw/patch \u0026 reclaim       1079  1385   926   252   291625   548873\n\nThese numbers are the average of 2 runs of 3 \"make -j\" runs done right\nafter system boot.  Run-to-run variability for \"make -j\" is huge, so\nthese numbers aren\u0027t terribly useful except to seee that with reclaim\nthe benchmark still finishes in a reasonable amount of time.\n\nI also looked at the NUMA hit/miss stats for the \"make -j\" runs and the\nreclaim doesn\u0027t make any difference when the machine is thrashing away.\n\nDoing a \"make -j8\" on a single node that is filled with page cache pages\ntakes 700 seconds with reclaim turned on and 735 seconds without reclaim\n(due to remote memory accesses).\n\nThe simple zone_reclaim syscall program is at\nhttp://www.bork.org/~mort/sgi/zone_reclaim.c\n\nSigned-off-by: Martin Hicks \u003cmort@sgi.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "7d87e14c236d6c4cab66d87cf0bc1e0f0375d308",
      "tree": "0c0826cdc102286b541e3e56b59c81752d34c90d",
      "parents": [
        "434498d5323445b59167fd7aa5633b74ebbce901"
      ],
      "author": {
        "name": "Stephen Rothwell",
        "email": "sfr@canb.auug.org.au",
        "time": "Sun May 01 08:59:12 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sun May 01 08:59:12 2005 -0700"
      },
      "message": "[PATCH] consolidate sys_shmat\n\nSigned-off-by: Stephen Rothwell \u003csfr@canb.auug.org.au\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"
    }
  ]
}