)]}'
{
  "log": [
    {
      "commit": "f8cbd99bd3a023db8d6356d19a5f6f539d367327",
      "tree": "f7472cc26a2a1dad631c35a4eb6d0c10cf8cd66e",
      "parents": [
        "f704f56af95bec3c1ca719d64d0becef74d40899"
      ],
      "author": {
        "name": "Ingo Molnar",
        "email": "mingo@elte.hu",
        "time": "Sat Jun 25 14:57:39 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Jun 25 16:24:45 2005 -0700"
      },
      "message": "[PATCH] sched: voluntary kernel preemption\n\nThis patch adds a new preemption model: \u0027Voluntary Kernel Preemption\u0027.  The\n3 models can be selected from a new menu:\n\n            (X) No Forced Preemption (Server)\n            ( ) Voluntary Kernel Preemption (Desktop)\n            ( ) Preemptible Kernel (Low-Latency Desktop)\n\nwe still default to the stock (Server) preemption model.\n\nVoluntary preemption works by adding a cond_resched()\n(reschedule-if-needed) call to every might_sleep() check.  It is lighter\nthan CONFIG_PREEMPT - at the cost of not having as tight latencies.  It\nrepresents a different latency/complexity/overhead tradeoff.\n\nIt has no runtime impact at all if disabled.  Here are size stats that show\nhow the various preemption models impact the kernel\u0027s size:\n\n    text    data     bss     dec     hex filename\n 3618774  547184  179896 4345854  424ffe vmlinux.stock\n 3626406  547184  179896 4353486  426dce vmlinux.voluntary   +0.2%\n 3748414  548640  179896 4476950  445016 vmlinux.preempt     +3.5%\n\nvoluntary-preempt is +0.2% of .text, preempt is +3.5%.\n\nThis feature has been tested for many months by lots of people (and it\u0027s\nalso included in the RHEL4 distribution and earlier variants were in Fedora\nas well), and it\u0027s intended for users and distributions who dont want to\nuse full-blown CONFIG_PREEMPT for one reason or another.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "f704f56af95bec3c1ca719d64d0becef74d40899",
      "tree": "f024287878246703cf28f738d2c553c476c53c34",
      "parents": [
        "cc19ca86a023fcd552c78e77a7be6ce271f92a28"
      ],
      "author": {
        "name": "Ingo Molnar",
        "email": "mingo@elte.hu",
        "time": "Sat Jun 25 14:57:38 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Jun 25 16:24:45 2005 -0700"
      },
      "message": "[PATCH] enable PREEMPT_BKL on !PREEMPT+SMP too\n\nThe only sane way to clean up the current 3 lock_kernel() variants seems to\nbe to remove the spinlock-based BKL implementations altogether, and to keep\nthe semaphore-based one only.  If we dont want to do that for whatever\nreason then i\u0027m afraid we have to live with the current complexity.  (but\ni\u0027m open for other cleanup suggestions as well.)\n\nTo explore this possibility we\u0027ll (at a minimum) have to know whether the\nsemaphore-based BKL works fine on plain SMP too.  The patch below enables\nthis.\n\nThe patch may make sense in isolation as well, as it might bring\nperformance benefits: code that would formerly spin on the BKL spinlock\nwill now schedule away and give up the CPU.  It might introduce performance\nregressions as well, if any performance-critical code uses the BKL heavily\nand gets overscheduled due to the semaphore.  I very much hope there is no\nsuch performance-critical codepath left though.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "cc19ca86a023fcd552c78e77a7be6ce271f92a28",
      "tree": "b4e9c9b02b73105d363070c49a765208b99cee1a",
      "parents": [
        "7f1867a5b3dc3034cbea403b229d65eed4a7f62e"
      ],
      "author": {
        "name": "Ingo Molnar",
        "email": "mingo@elte.hu",
        "time": "Sat Jun 25 14:57:36 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sat Jun 25 16:24:45 2005 -0700"
      },
      "message": "[PATCH] consolidate PREEMPT options into kernel/Kconfig.preempt\n\nThis patch consolidates the CONFIG_PREEMPT and CONFIG_PREEMPT_BKL\npreemption options into kernel/Kconfig.preempt.  This, besides reducing\nsource-code, also enables more centralized tweaking of preemption related\noptions.\n\nSigned-off-by: Ingo Molnar \u003cmingo@elte.hu\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    }
  ]
}