)]}'
{
  "commit": "a4cbbf549a9be10b7583c44249efccd64839533d",
  "tree": "4b4862e4513b629723b8853e379cd38fee08b095",
  "parents": [
    "8308756f45a12e2ff4f7749c2694fc83cdef0be9",
    "2fde4f94e0a9531251e706fa57131b51b0df042e"
  ],
  "author": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Mon Feb 09 15:43:55 2015 -0800"
  },
  "committer": {
    "name": "Linus Torvalds",
    "email": "torvalds@linux-foundation.org",
    "time": "Mon Feb 09 15:43:55 2015 -0800"
  },
  "message": "Merge branch \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip\n\nPull perf updates from Ingo Molnar:\n \"Kernel side changes:\n\n   - AMD range breakpoints support:\n\n     Extend breakpoint tools and core to support address range through\n     perf event with initial backend support for AMD extended\n     breakpoints.\n\n     The syntax is:\n\n         perf record -e mem:addr/len:type\n\n     For example set write breakpoint from 0x1000 to 0x1200 (0x1000 + 512)\n\n         perf record -e mem:0x1000/512:w\n\n   - event throttling/rotating fixes\n\n   - various event group handling fixes, cleanups and general paranoia\n     code to be more robust against bugs in the future.\n\n    - kernel stack overhead fixes\n\n  User-visible tooling side changes:\n\n   - Show precise number of samples in at the end of a \u0027record\u0027 session,\n     if processing build ids, since we will then traverse the whole\n     perf.data file and see all the PERF_RECORD_SAMPLE records,\n     otherwise stop showing the previous off-base heuristicly counted\n     number of \"samples\" (Namhyung Kim).\n\n   - Support to read compressed module from build-id cache (Namhyung\n     Kim)\n\n   - Enable sampling loads and stores simultaneously in \u0027perf mem\u0027\n     (Stephane Eranian)\n\n   - \u0027perf diff\u0027 output improvements (Namhyung Kim)\n\n   - Fix error reporting for evsel pgfault constructor (Arnaldo Carvalho\n     de Melo)\n\n  Tooling side infrastructure changes:\n\n   - Cache eh/debug frame offset for dwarf unwind (Namhyung Kim)\n\n   - Support parsing parameterized events (Cody P Schafer)\n\n   - Add support for IP address formats in libtraceevent (David Ahern)\n\n  Plus other misc fixes\"\n\n* \u0027perf-core-for-linus\u0027 of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (48 commits)\n  perf: Decouple unthrottling and rotating\n  perf: Drop module reference on event init failure\n  perf: Use POLLIN instead of POLL_IN for perf poll data in flag\n  perf: Fix put_event() ctx lock\n  perf: Fix move_group() order\n  perf: Fix event-\u003ectx locking\n  perf: Add a bit of paranoia\n  perf symbols: Convert lseek + read to pread\n  perf tools: Use perf_data_file__fd() consistently\n  perf symbols: Support to read compressed module from build-id cache\n  perf evsel: Set attr.task bit for a tracking event\n  perf header: Set header version correctly\n  perf record: Show precise number of samples\n  perf tools: Do not use __perf_session__process_events() directly\n  perf callchain: Cache eh/debug frame offset for dwarf unwind\n  perf tools: Provide stub for missing pthread_attr_setaffinity_np\n  perf evsel: Don\u0027t rely on malloc working for sz 0\n  tools lib traceevent: Add support for IP address formats\n  perf ui/tui: Show fatal error message only if exists\n  perf tests: Fix typo in sample-parsing.c\n  ...\n",
  "tree_diff": []
}