)]}'
{
  "log": [
    {
      "commit": "13e4b57f6a4e23ceb99794a650d777e74831f4a6",
      "tree": "a585fdc3a2374dbedcf0114fa3c38d2fb7920774",
      "parents": [
        "75bcc8c5e1de78616b04ef9f317a293a7c1c163c"
      ],
      "author": {
        "name": "Nishanth Aravamudan",
        "email": "nacc@us.ibm.com",
        "time": "Sat Sep 10 00:27:25 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Sat Sep 10 10:06:37 2005 -0700"
      },
      "message": "[PATCH] mm: fix-up schedule_timeout() usage\n\nUse schedule_timeout_{,un}interruptible() instead of\nset_current_state()/schedule_timeout() to reduce kernel size.\n\nSigned-off-by: Nishanth Aravamudan \u003cnacc@us.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "5d337b9194b1ce3b6fd5f3cb2799455ed2f9a3d1",
      "tree": "91ed9ef6f4cb5f6a1832f2baaaabd53fcd83513e",
      "parents": [
        "048c27fd72816b44e096997d1c6901c3abbfd45b"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:41 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:42 2005 -0700"
      },
      "message": "[PATCH] swap: swap_lock replace list+device\n\nThe idea of a swap_device_lock per device, and a swap_list_lock over them all,\nis appealing; but in practice almost every holder of swap_device_lock must\nalready hold swap_list_lock, which defeats the purpose of the split.\n\nThe only exceptions have been swap_duplicate, valid_swaphandles and an\nuntrodden path in try_to_unuse (plus a few places added in this series).\nvalid_swaphandles doesn\u0027t show up high in profiles, but swap_duplicate does\ndemand attention.  However, with the hold time in get_swap_pages so much\nreduced, I\u0027ve not yet found a load and set of swap device priorities to show\neven swap_duplicate benefitting from the split.  Certainly the split is mere\noverhead in the common case of a single swap device.\n\nSo, replace swap_list_lock and swap_device_lock by spinlock_t swap_lock\n(generally we seem to prefer an _ in the name, and not hide in a macro).\n\nIf someone can show a regression in swap_duplicate, then probably we should\nadd a hashlock for the swap_map entries alone (shorts being anatomic), so as\nto help the case of the single swap device too.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "048c27fd72816b44e096997d1c6901c3abbfd45b",
      "tree": "159a00b71ce9db161a48e4fc07b212db455a2cf1",
      "parents": [
        "52b7efdbe5f5696fc80338560a3fc51e0b0a993c"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:40 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:41 2005 -0700"
      },
      "message": "[PATCH] swap: scan_swap_map latency breaks\n\nThe get_swap_page/scan_swap_map latency can be so bad that even those without\npreemption configured deserve relief: periodically cond_resched.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "52b7efdbe5f5696fc80338560a3fc51e0b0a993c",
      "tree": "30162de9fc8fe3dddb6462f8ff82f1594067cadd",
      "parents": [
        "7dfad4183bf9cd92f977caa3c12cc74f0eefc0e6"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:39 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:41 2005 -0700"
      },
      "message": "[PATCH] swap: scan_swap_map drop swap_device_lock\n\nget_swap_page has often shown up on latency traces, doing lengthy scans while\nholding two spinlocks.  swap_list_lock is already dropped, now scan_swap_map\ndrop swap_device_lock before scanning the swap_map.\n\nWhile scanning for an empty cluster, don\u0027t worry that racing tasks may\nallocate what was free and free what was allocated; but when allocating an\nentry, check it\u0027s still free after retaking the lock.  Avoid dropping the lock\nin the expected common path.  No barriers beyond the locks, just let the\ncookie crumble; highest_bit limit is volatile, but benign.\n\nGuard against swapoff: must check SWP_WRITEOK before allocating, must raise\nSWP_SCANNING reference count while in scan_swap_map, swapoff wait for that to\nfall - just use schedule_timeout, we don\u0027t want to burden scan_swap_map\nitself, and it\u0027s very unlikely that anyone can really still be in\nscan_swap_map once swapoff gets this far.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "7dfad4183bf9cd92f977caa3c12cc74f0eefc0e6",
      "tree": "e11306e1e82acba1832f07fdfb3296c34cf6e934",
      "parents": [
        "fb4f88dcabdc716c7c350e09cf4a38a419b007e1"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:38 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:41 2005 -0700"
      },
      "message": "[PATCH] swap: scan_swap_map restyled\n\nRewrite scan_swap_map to allocate in just the same way as before (taking the\nnext free entry SWAPFILE_CLUSTER-1 times, then restarting at the lowest wholly\nempty cluster, falling back to lowest entry if none), but with a view towards\ndropping the lock in the next patch.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "fb4f88dcabdc716c7c350e09cf4a38a419b007e1",
      "tree": "1a7806e3ac8bd45eded544763324c13c9f95c0e6",
      "parents": [
        "89d09a2c80ea6baafb559b86d545fada05e14ab5"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:37 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:41 2005 -0700"
      },
      "message": "[PATCH] swap: get_swap_page drop swap_list_lock\n\nRewrite get_swap_page to allocate in just the same sequence as before, but\nwithout holding swap_list_lock across its scan_swap_map.  Decrement\nnr_swap_pages and update swap_list.next in advance, while still holding\nswap_list_lock.  Skip full devices by testing highest_bit.  Swapoff hold\nswap_device_lock as well as swap_list_lock to clear SWP_WRITEOK.  Reduces lock\ncontention when there are parallel swap devices of the same priority.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "89d09a2c80ea6baafb559b86d545fada05e14ab5",
      "tree": "90ff3c2a62c144b93c89b33d66583046f1a7efae",
      "parents": [
        "6eb396dc4a9781c5e7951143ab56ce5710687ab3"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:36 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:41 2005 -0700"
      },
      "message": "[PATCH] swap: freeing update swap_list.next\n\nThis makes negligible difference in practice: but swap_list.next should not be\nupdated to a higher prio in the general helper swap_info_get, but rather in\nswap_entry_free; and then only in the case when entry is actually freed.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "6eb396dc4a9781c5e7951143ab56ce5710687ab3",
      "tree": "48b165c8934a0f3852344725fee060740dcdfa48",
      "parents": [
        "53092a7402f227151a681b0c92ec8598c5618b1a"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:35 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:41 2005 -0700"
      },
      "message": "[PATCH] swap: swap unsigned int consistency\n\nThe swap header\u0027s unsigned int last_page determines the range of swap pages,\nbut swap_info has been using int or unsigned long in some cases: use unsigned\nint throughout (except, in several places a local unsigned long is useful to\navoid overflows when adding).\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Jens Axboe \u003caxboe@suse.de\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "53092a7402f227151a681b0c92ec8598c5618b1a",
      "tree": "ea482a18567c8b3b1778331ee3c4864e2f4dae6f",
      "parents": [
        "11d31886dbcb61039ed3789e583d21c6e70960fd"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:34 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:40 2005 -0700"
      },
      "message": "[PATCH] swap: show span of swap extents\n\nThe \"Adding %dk swap\" message shows the number of swap extents, as a guide to\nhow fragmented the swapfile may be.  But a useful further guide is what total\nextent they span across (sometimes scarily large).\n\nAnd there\u0027s no need to keep nr_extents in swap_info: it\u0027s unused after the\ninitial message, so save a little space by keeping it on stack.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "11d31886dbcb61039ed3789e583d21c6e70960fd",
      "tree": "ddaebfd35080a530a30c56587707c2c5ef452591",
      "parents": [
        "4cd3bb10ff0b21b77b5a4cd13b4bd36694e054c4"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:34 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:40 2005 -0700"
      },
      "message": "[PATCH] swap: swap extent list is ordered\n\nThere are several comments that swap\u0027s extent_list.prev points to the lowest\nextent: that\u0027s not so, it\u0027s extent_list.next which points to it, as you\u0027d\nexpect.  And a couple of loops in add_swap_extent which go all the way through\nthe list, when they should just add to the other end.\n\nFix those up, and let map_swap_page search the list forwards: profiles shows\nit to be twice as quick that way - because prefetch works better on how the\nstructs are typically kmalloc\u0027ed?  or because usually more is written to than\nread from swap, and swap is allocated ascendingly?\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "4cd3bb10ff0b21b77b5a4cd13b4bd36694e054c4",
      "tree": "e2a76a229b2c030a7d4aec1155e505fd4c087067",
      "parents": [
        "e2244ec2efa4ee1edf391d0001d314933e2b2974"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:33 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:40 2005 -0700"
      },
      "message": "[PATCH] swap: move destroy_swap_extents calls\n\nsys_swapon\u0027s call to destroy_swap_extents on failure is made after the final\nswap_list_unlock, which is faintly unsafe: another sys_swapon might already be\nsetting up that swap_info_struct.  Calling it earlier, before taking\nswap_list_lock, is safe.  sys_swapoff\u0027s call to destroy_swap_extents was safe,\nbut likewise move it earlier, before taking the locks (once try_to_unuse has\ncompleted, nothing can be needing the swap extents).\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "e2244ec2efa4ee1edf391d0001d314933e2b2974",
      "tree": "c9e43f9b6e11ce41d976025c6f9d94c273521132",
      "parents": [
        "b0d9bcd4bb79a7834f8492f2ae5c2655a551f23d"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:32 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:40 2005 -0700"
      },
      "message": "[PATCH] swap: correct swapfile nr_good_pages\n\nIf a regular swapfile lies on a filesystem whose blocksize is less than\nPAGE_SIZE, then setup_swap_extents may have to cut the number of usable swap\npages; but sys_swapon\u0027s nr_good_pages was not expecting that.  Also,\nsetup_swap_extents takes no account of badpages listed in the swap header: not\nworth doing so, but ensure nr_badpages is 0 for a regular swapfile.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "b0d9bcd4bb79a7834f8492f2ae5c2655a551f23d",
      "tree": "8978577cf4b132e08e5abc7101e934ffd3106d3c",
      "parents": [
        "fd4fd5aac1282825195c6816ed40a2a6d42db5bf"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Sat Sep 03 15:54:31 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@evo.osdl.org",
        "time": "Mon Sep 05 00:05:40 2005 -0700"
      },
      "message": "[PATCH] swap: update swapfile i_sem comment\n\nUpdate swap extents comment: nowadays we guard with S_SWAPFILE not i_sem.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "c475a8ab625d567eacf5e30ec35d6d8704558062",
      "tree": "0971bef7b876f1b3eb160621fc2b61cb5313827b",
      "parents": [
        "d296e9cd02c92e576ecce5344026a4df4353cdb2"
      ],
      "author": {
        "name": "Hugh Dickins",
        "email": "hugh@veritas.com",
        "time": "Tue Jun 21 17:15:12 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Tue Jun 21 18:46:21 2005 -0700"
      },
      "message": "[PATCH] can_share_swap_page: use page_mapcount\n\nRemember that ironic get_user_pages race?  when the raised page_count on a\npage swapped out led do_wp_page to decide that it had to copy on write, so\nsubstituted a different page into userspace.  2.6.7 onwards have Andrea\u0027s\nsolution, where try_to_unmap_one backs out if it finds page_count raised.\n\nWhich works, but is unsatisfying (rmap.c has no other page_count heuristics),\nand was found a few months ago to hang an intensive page migration test.  A\nyear ago I was hesitant to engage page_mapcount, now it seems the right fix.\n\nSo remove the page_count hack from try_to_unmap_one; and use activate_page in\nunuse_mm when dropping lock, to replace its secondary effect of helping\nswapoff to make progress in that case.\n\nSimplify can_share_swap_page (now called only on anonymous pages) to check\npage_mapcount + page_swapcount \u003d\u003d 1: still needs the page lock to stabilize\ntheir (pessimistic) sum, but does not need swapper_space.tree_lock for that.\n\nIn do_swap_page, move swap_free and unlock_page below page_add_anon_rmap, to\nkeep sum on the high side, and correct when can_share_swap_page called.\n\nSigned-off-by: Hugh Dickins \u003chugh@veritas.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "ba32311eb73f624a85a5fc2e043cda8e076f86ef",
      "tree": "1baa53ef8f4b708ec1807ca3a267c713d00da266",
      "parents": [
        "59afda786a58f2e8f74a48f2af90534d76d5338d"
      ],
      "author": {
        "name": "McMullan, Jason",
        "email": "jason.mcmullan@timesys.com",
        "time": "Mon May 16 21:53:40 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Tue May 17 07:59:18 2005 -0700"
      },
      "message": "[PATCH] swapout oops fix\n\nFix OOPS when swapping on a device that doesn\u0027t have an unplug_io_fn defined\n(eg, ATA Over Ethernet)\n\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"
    }
  ]
}