)]}'
{
  "log": [
    {
      "commit": "8b6490e5faafb3a16ea45654fb55f9ff086f1495",
      "tree": "7af6f19fb36afe14a3405a4a656c29ad7ce251eb",
      "parents": [
        "0f97a931b337e4662e736ca67f1fab0a187f5852"
      ],
      "author": {
        "name": "Dipankar Sarma",
        "email": "dipankar@in.ibm.com",
        "time": "Fri Sep 09 13:04:07 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@g5.osdl.org",
        "time": "Fri Sep 09 13:57:54 2005 -0700"
      },
      "message": "[PATCH] files: fix rcu initializers\n\nFirst of a number of files_lock scaability patches.\n\n Here are the x86 numbers -\n\n tiobench on a 4(8)-way (HT) P4 system on ramdisk :\n\n                                         (lockfree)\n Test            2.6.10-vanilla  Stdev   2.6.10-fd       Stdev\n -------------------------------------------------------------\n Seqread         1400.8          11.52   1465.4          34.27\n Randread        1594            8.86    2397.2          29.21\n Seqwrite        242.72          3.47    238.46          6.53\n Randwrite       445.74          9.15    446.4           9.75\n\n The performance improvement is very significant.\n We are getting killed by the cacheline bouncing of the files_struct\n lock here. Writes on ramdisk (ext2) seems to vary just too\n much to get any meaningful number.\n\n Also, With Tridge\u0027s thread_perf test on a 4(8)-way (HT) P4 xeon system :\n\n 2.6.12-rc5-vanilla :\n\n Running test \u0027readwrite\u0027 with 8 tasks\n Threads     0.34 +/- 0.01 seconds\n Processes   0.16 +/- 0.00 seconds\n\n 2.6.12-rc5-fd :\n\n Running test \u0027readwrite\u0027 with 8 tasks\n Threads     0.17 +/- 0.02 seconds\n Processes   0.17 +/- 0.02 seconds\n\n I repeated the measurements on ramfs (as opposed to ext2 on ramdisk in\n the earlier measurement) and I got more consistent results from tiobench :\n\n 4(8) way xeon P4\n -----------------\n                                         (lock-free)\n Test            2.6.12-rc5      Stdev   2.6.12-rc5-fd   Stdev\n -------------------------------------------------------------\n Seqread         1282            18.59   1343.6          26.37\n Randread        1517            7       2415            34.27\n Seqwrite        702.2           5.27    709.46           5.9\n Randwrite       846.86          15.15   919.68          21.4\n\n 4-way ppc64\n ------------\n                                         (lock-free)\n Test            2.6.12-rc5      Stdev   2.6.12-rc5-fd   Stdev\n -------------------------------------------------------------\n Seqread         1549            91.16   1569.6          47.2\n Randread        1473.6          25.11   1585.4          69.99\n Seqwrite        1096.8          20.03   1136            29.61\n Randwrite       1189.6           4.04   1275.2          32.96\n\n Also running Tridge\u0027s thread_perf test on ppc64 :\n\n 2.6.12-rc5-vanilla\n --------------------\n Running test \u0027readwrite\u0027 with 4 tasks\n Threads     0.20 +/- 0.02 seconds\n Processes   0.16 +/- 0.01 seconds\n\n 2.6.12-rc5-fd\n --------------------\n Running test \u0027readwrite\u0027 with 4 tasks\n Threads     0.18 +/- 0.04 seconds\n Processes   0.16 +/- 0.01 seconds\n\n The benefits are huge (upto ~60%) in some cases on x86 primarily\n due to the atomic operations during acquisition of -\u003efile_lock\n and cache line bouncing in fast path. ppc64 benefits are modest\n due to LL/SC based locking, but still statistically significant.\n\nThis patch:\n\nRCU head initilizer no longer needs the head varible name since we don\u0027t use\nlist.h lists anymore.\n\nSigned-off-by: Dipankar Sarma \u003cdipankar@in.ibm.com\u003e\nSigned-off-by: Andrew Morton \u003cakpm@osdl.org\u003e\nSigned-off-by: Linus Torvalds \u003ctorvalds@osdl.org\u003e\n"
    },
    {
      "commit": "9b06e818985d139fd9e82c28297f7744e1b484e1",
      "tree": "3cd2ba33b22fc2812f714f86ea52c68704f8e2c3",
      "parents": [
        "512345be2549308b8ae8e85a3ff7f6d56a38e5f6"
      ],
      "author": {
        "name": "Paul E. McKenney",
        "email": "paulmck@us.ibm.com",
        "time": "Sun May 01 08:59:04 2005 -0700"
      },
      "committer": {
        "name": "Linus Torvalds",
        "email": "torvalds@ppc970.osdl.org",
        "time": "Sun May 01 08:59:04 2005 -0700"
      },
      "message": "[PATCH] Deprecate synchronize_kernel, GPL replacement\n\nThe synchronize_kernel() primitive is used for quite a few different purposes:\nwaiting for RCU readers, waiting for NMIs, waiting for interrupts, and so on.\nThis makes RCU code harder to read, since synchronize_kernel() might or might\nnot have matching rcu_read_lock()s.  This patch creates a new\nsynchronize_rcu() that is to be used for RCU readers and a new\nsynchronize_sched() that is used for the rest.  These two new primitives\ncurrently have the same implementation, but this is might well change with\nadditional real-time support.  Both new primitives are GPL-only, the old\nprimitive is deprecated.\n\nSigned-off-by: Paul E. McKenney \u003cpaulmck@us.ibm.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"
    }
  ]
}