)]}' { "log": [ { "commit": "1785022e12cf0079889c519ffc7157f2143326ea", "tree": "a983e8d6527eda284dc095d87699a4d375c82895", "parents": [ "c189fab6b30c1c7240addf6f1fdf732591e758c7" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 17:16:40 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 17:16:40 2017 -0400" }, "message": "trace-cmd report: Add -O offset for function plugin\n\nWhen -O offset is added to trace-cmd report with the function plugin\nenabled, it will display the offset of the functions along with their names.\nThis helps in finding exactly where a function was called by its parent.\n\n trace-cmd report -O parent -O offset\n[..]\n rcuc/163-1330 [163] 740.653251: function: _raw_spin_lock+0x0 \u003c-- rcu_cpu_kthread+0x4d8\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "c189fab6b30c1c7240addf6f1fdf732591e758c7", "tree": "722ae4e66e2016ce2b710515fdfac64677847b89", "parents": [ "ceac02d6ec27a427b321af8865772a0d93f7a595" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:50:08 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:50:08 2017 -0400" }, "message": "trace-cmd listen: Replace tab with spaces in help output\n\nThe bash completions of trace-cmd commands uses the -h to find the options\nto finish the completions with. trace-cmd listen -h had a tab before \"-l\"\nwhich caused the completions to miss it. Not to mention, it caused -l not to\nbe in line with the other options.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "ceac02d6ec27a427b321af8865772a0d93f7a595", "tree": "600a1f6ba523ceae2385a7858e179545574c3ce6", "parents": [ "26b2695217bd820da3c1be4301b4de6fea2bbdbb" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:36:40 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:36:40 2017 -0400" }, "message": "trace-cmd report: Fix pointer check in option -r\n\nThe removal of malloc_or_die() cause a bad check for allocation failure for\nthe option of \u0027-r\u0027.\n\nFixes: b57a3336be81 (\"trace-cmd: Remove malloc_or_die() form trace-read.c\")\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "26b2695217bd820da3c1be4301b4de6fea2bbdbb", "tree": "6b31dc2cba15e444dcc0eee187e8ddcef1688a75", "parents": [ "3255cd93dd46fee8f106a6556b1114f773085a2f" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:31:44 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:31:44 2017 -0400" }, "message": "trace-cmd: Add bash completion for trace-cmd report -O\n\nShow the available options with the bash completion of trace-cmd report -O\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "3255cd93dd46fee8f106a6556b1114f773085a2f", "tree": "711dfe30c3ce7508f183d86310841cbf41cffa08", "parents": [ "91f5771c3cb04f2aa396ab0c54abb1d4b3403de9" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:06:35 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:12:22 2017 -0400" }, "message": "trace-cmd: Have bash completions look at all options by default\n\nHave all the trace-cmd commands have their options looked at for completion,\neven if there is not a specific function to handle them.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "91f5771c3cb04f2aa396ab0c54abb1d4b3403de9", "tree": "cd5cd36fff414612ae58daa48b18c18db892661b", "parents": [ "5b52ef38103ec11088863f6288dee1cb9c721b85" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 15:28:52 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:11:05 2017 -0400" }, "message": "trace-cmd: Add bash completions for instance buffers for -B option\n\nSeveral trace-cmd commands use -B to work with buffer instances. When -B is\nused in such a way, then have it search the instance directory and list the\navailable instances.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "5b52ef38103ec11088863f6288dee1cb9c721b85", "tree": "8177d37904af233afc931227074979ce0aba613c", "parents": [ "6c7c6f77ab596165f7c2c119a738a89367687292" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 15:27:43 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:01:44 2017 -0400" }, "message": "trace-cmd: Add trace-cmd show completion for bash\n\nAdd specific bash completion for trace-cmd show.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "6c7c6f77ab596165f7c2c119a738a89367687292", "tree": "171730cad72f735d1630fc4a9375736b15230403", "parents": [ "3ea6b2ddc05480ac0288f5a4b038d34d59b40c1a" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 15:30:58 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:01:34 2017 -0400" }, "message": "trace-cmd: Have trace-cmd start, profile and stream use record completion\n\nThe commands for trace-cmd start, profile and stream are the same as\ntrace-cmd record, so have them have the same completions as trace-cmd record.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "3ea6b2ddc05480ac0288f5a4b038d34d59b40c1a", "tree": "7a4efec3c3938f1000c84bcc86737b33f8028d3a", "parents": [ "ab4d1388342f5ec79afd71068b197bbae54ad4ac" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:00:45 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 16:00:45 2017 -0400" }, "message": "trace-cmd: Fix the fix for completions\n\nIn the learning process of writing completion scripts, I put in various\ndebugging. One of those was not removed before committing.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "ab4d1388342f5ec79afd71068b197bbae54ad4ac", "tree": "d36675f56cbac4aa774b03f69be03c86b4bb4e87", "parents": [ "45b17bad5a283bd572a509524f4a6ff75cdd8ddf" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 13:22:18 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 09 13:22:18 2017 -0400" }, "message": "trace-cmd: Fix completions\n\nThe completions that were installed were basically just good for trace-cmd\nrecord, which kills all other users, and makes it frustrating trying to use\ntab completion for simple file finding.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "45b17bad5a283bd572a509524f4a6ff75cdd8ddf", "tree": "610e6e5628d33264db46266dabee11e5f429d5df", "parents": [ "021710e1073fe203341b427cd1a4bac577ec899c" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 02 13:32:02 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Fri May 05 19:40:15 2017 -0400" }, "message": "trace-cmd-v2.7.dev\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "021710e1073fe203341b427cd1a4bac577ec899c", "tree": "3e1a3c46fded2ace0f146309480bec1b0826deb4", "parents": [ "326fe045ca201a1e747767cee10b4d1fd94a4d80" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 02 13:32:02 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Tue May 02 13:35:16 2017 -0400" }, "message": "trace-cmd-v2.6.1\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "326fe045ca201a1e747767cee10b4d1fd94a4d80", "tree": "d558d53176da894143562038db8d492b4c6b6e4f", "parents": [ "821d7935c7fe3fd3d412efb6343ccd4065f84ba8" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Mon May 01 15:45:23 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Mon May 01 15:45:23 2017 -0400" }, "message": "trace-cmd record: Fix filtering when using set_event_pid\n\nWhen set_event_pid exists, trace-cmd uses it for filtering tasks. A commit\nthat updates the pid filtering was not updated to handle excludes of the\ntracing, and they were added too, causing more to be filtered than what\nshould have been.\n\nFixes: a5923c5c9f81 (\"trace-cmd: Filter out specific pids\")\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "821d7935c7fe3fd3d412efb6343ccd4065f84ba8", "tree": "c2f199438967eed893ca6b920ed89de3ca5993e8", "parents": [ "adba99afe22e495ea3efd8564e67a7bf590a5ef6" ], "author": { "name": "Federico Vaga", "email": "federico.vaga@vaga.pv.it", "time": "Wed Apr 26 22:30:28 2017 +0200" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 22:36:41 2017 -0400" }, "message": "trace-cmd: Implement a main functionality lookup table\n\nSince now there is an uniform command implementation I can reduce the\n `main()` function to the minimum by using a lookup table. People can\n now directly focus on a command implementation because there is\n \"nothing\" in the `main()` function.\n\nLink: http://lkml.kernel.org/r/20170426203028.13900-3-federico.vaga@vaga.pv.it\n\nSigned-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "adba99afe22e495ea3efd8564e67a7bf590a5ef6", "tree": "0c65ba248a2abba9478080b2fec101f7001e5faa", "parents": [ "1c47f105119646c5c95c57c176239cc1d9893d5d" ], "author": { "name": "Federico Vaga", "email": "federico.vaga@vaga.pv.it", "time": "Wed Apr 26 22:30:27 2017 +0200" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 22:33:20 2017 -0400" }, "message": "trace-cmd: Uniform main command implementation\n\nMost of the commands have the implementation in a dedicated file.\nThose commands without a dedicated file are implemented in the\n`main()` function. This patch uniform the code by moving the commands\nimplemented in the `main()` function to dedicated files.\n\nLink: http://lkml.kernel.org/r/20170426203028.13900-2-federico.vaga@vaga.pv.it\n\nSigned-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "1c47f105119646c5c95c57c176239cc1d9893d5d", "tree": "a9edcb197aadc211b801bf3010687ffc99750b97", "parents": [ "49cb7d0124618bd215cdf0b421b9a816a8d2648e" ], "author": { "name": "Taeung Song", "email": "treeze.taeung@gmail.com", "time": "Wed Apr 26 23:55:40 2017 +0900" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 15:09:54 2017 -0400" }, "message": "parse-events: Fix missing break in FALSE case of pevent_filter_clear_trivial()\n\nCurrently the FILTER_TRIVIAL_FALSE case has a missing break statement, if\nthe trivial type is FALSE, it will also run into the TRUE case, and always\nbe skipped as the TRUE statement will continue the loop on the inverse\ncondition of the FALSE statement.\n\nLink: http://lkml.kernel.org/r/1493218540-12296-1-git-send-email-treeze.taeung@gmail.com\n\nReported-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nSigned-off-by: Taeung Song \u003ctreeze.taeung@gmail.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "49cb7d0124618bd215cdf0b421b9a816a8d2648e", "tree": "a80261f39d3c6152a436b2747c0cb70b463cf5ba", "parents": [ "b9a0df2f1f4923916299459e8d56942dc8c45e70" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:50:43 2017 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:51:51 2017 -0400" }, "message": "trace-view: Remove unused variable \"selection\"\n\nClean up compiler warning.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "b9a0df2f1f4923916299459e8d56942dc8c45e70", "tree": "2f6ef2690d222ea8854814904ded48f58ee02572", "parents": [ "c8bc47393693d6d6641bfcf9343b1a9adf83be2b" ], "author": { "name": "Federico Vaga", "email": "federico.vaga@vaga.pv.it", "time": "Sun Apr 23 12:22:58 2017 +0200" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: BUG fix malloc() pointer validation\n\nTo reproduce the bug\n\nmkdir /sys/kernel/debug/tracing/instances/test\n./trace-cmd show -B test -s -f\n Failed to allocate instance path snapshot\n\nLink: http://lkml.kernel.org/r/20170423102258.21609-6-federico.vaga@vaga.pv.it\n\nSigned-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "c8bc47393693d6d6641bfcf9343b1a9adf83be2b", "tree": "192daaf3d2f056d874b376068f89871467a088b6", "parents": [ "53e9b788d537ad9735c3186d93f14d1efbf62f4d" ], "author": { "name": "Federico Vaga", "email": "federico.vaga@vaga.pv.it", "time": "Sun Apr 23 12:22:56 2017 +0200" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd:read: BUG initialize input_files item to zero\n\nOn allocation the data structure was not initialized. Later on some\nattribute of this structure are used (e.g. tsoffset) assuming that the\ndefault value is zero, but it is not always true.\n\nLink: http://lkml.kernel.org/r/20170423102258.21609-4-federico.vaga@vaga.pv.it\n\nSigned-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "53e9b788d537ad9735c3186d93f14d1efbf62f4d", "tree": "861220f233c6a9acab1052e0b19c055f7ca06e31", "parents": [ "3b43d0a77d7178b797d6ba38ca21b8286b647afd" ], "author": { "name": "Federico Vaga", "email": "federico.vaga@vaga.pv.it", "time": "Sun Apr 23 12:22:55 2017 +0200" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "plugin:python: check asprintf() errors\n\nThe code was validating the string but the man page for asprintf(3)\nsays clearly:\n\n--------\nIf memory allocation wasn\u0027t possible, or some other error occurs,\nthese functions will return -1, and the contents of strp are undefined.\n--------\n\nSo, we cannot really rely on the returned string pointer.\n\nLink: http://lkml.kernel.org/r/20170423102258.21609-3-federico.vaga@vaga.pv.it\n\nSigned-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "3b43d0a77d7178b797d6ba38ca21b8286b647afd", "tree": "f599a21bdc62d30a730e1dca83d36a52287f8842", "parents": [ "83c1851afdd413e7e81b93333992f1127a558ffc" ], "author": { "name": "Federico Vaga", "email": "federico.vaga@vaga.pv.it", "time": "Sun Apr 23 12:22:54 2017 +0200" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "plugin:python: fix compiler warning\n\nThe function `load_plugin` is passed, as argument, to\n`trace_util_load_plugins()` but the prototype was not exactly the same.\n\nLink: http://lkml.kernel.org/r/20170423102258.21609-2-federico.vaga@vaga.pv.it\n\nSigned-off-by: Federico Vaga \u003cfederico.vaga@vaga.pv.it\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "83c1851afdd413e7e81b93333992f1127a558ffc", "tree": "6ca03b2e48302cd386250f361655fcf15e5c7f3f", "parents": [ "81921e3408367cbe56d5a5596fe956897643987c" ], "author": { "name": "John Kacur", "email": "jkacur@redhat.com", "time": "Tue Mar 21 17:11:53 2017 +0100" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: Fix segmentation fault in trace-snapshot\n\nTo reproduce the problem:\n\n plugin \u0027wakeup\u0027\ntrace-cmd: Device or resource busy\n Segmentation fault (core dumped)\n\nWhen the user triggers a condition such as EBUSY the program should die\ngracefully. The problem here is simply caused by an extra conversion\nspecifier in \"die\" in the write_file function\n\nLink: http://lkml.kernel.org/r/1490112713-9456-1-git-send-email-jkacur@redhat.com\n\nSigned-off-by: John Kacur \u003cjkacur@redhat.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "81921e3408367cbe56d5a5596fe956897643987c", "tree": "7011f9a2be7d840f7bc8146c7c545987a4b97130", "parents": [ "5a0e67bc510d7644f8157be27e345a652f4db632" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Mon Feb 13 12:11:44 2017 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "tools lib traceevent: Initialize lenght on OLD_RING_BUFFER_TYPE_TIME_STAMP\n\nA undefined value was being used for the OLD_RING_BUFFER_TYPE_TIME_STAMP\ncase entry, as the \u0027length\u0027 variable was not being initialized, fix it.\n\nCaught by the reporter when building tools/perf/ using clang, which emmitted\nthis warning:\n\n kbuffer-parse.c:312:7: warning: variable \u0027length\u0027 is used uninitialized whenever switch case is taken [-Wsometimes-uninitialized]\n case OLD_RINGBUF_TYPE_TIME_EXTEND:\n ^~~~~~~~~~~~~~~~~~~~~~~~~~~~\n kbuffer-parse.c:339:29: note: uninitialized use occurs here\n kbuf-\u003enext \u003d kbuf-\u003eindex + length;\n ^~~~~~\n kbuffer-parse.c:297:21: note: initialize the variable \u0027length\u0027 to silence this warning\n unsigned int length;\n ^\n \u003d 0\n\nReported-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\nCc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nCc: David Ahern \u003cdsahern@gmail.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Wang Nan \u003cwangnan0@huawei.com\u003e\nLink: http://lkml.kernel.org/r/20170213121418.47f279e8@gandalf.local.home\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "5a0e67bc510d7644f8157be27e345a652f4db632", "tree": "1cab5065abbe27270c84e09a8120f1f46efc9764", "parents": [ "d52432a9bb0583e9bfd8cabbfdc5de60aa185099" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Feb 16 20:23:01 2017 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: Remove include of string.h from event-utils.h\n\nstring.h is not needed in event-utils.h. Remove it.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "d52432a9bb0583e9bfd8cabbfdc5de60aa185099", "tree": "9f622f35de23381b8dec2770b2756f639c87a5df", "parents": [ "743bb45e49ac30884ec92a7ded21181bea74f822" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Feb 16 20:15:55 2017 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "event-parse: Change pevent_data_pc() to pevent_data_preempt_count()\n\nThis function is for external use outside of trace-cmd itself. Instead of\nusing the cryptic \"pc\" extend the name to be more meaningful, preempt_count.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "743bb45e49ac30884ec92a7ded21181bea74f822", "tree": "a3a375ea3c0462f3531aa926934497a5ce015d59", "parents": [ "39f166e86bdccb33618fd3c28672c92c6d0267ca" ], "author": { "name": "Julia Cartwright", "email": "julia@ni.com", "time": "Thu Feb 16 16:57:21 2017 -0600" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: Fixup more bogus symbol resolutions\n\nCommit ce1cbc51c (\"trace-cmd: Leave out absolute addresses to fix bogus\nsymbol resolutions\") fixed up bogus symbol resolution for \u0027A\u0027 type\nsymbols used on x86_64 for percpu offsets.\n\nHowever, this check did not cover per-cpu variables used in modules,\nwhich are generated as \u0027local\u0027 absolute symbols, and therefore appear as \u0027a\u0027 in\n/proc/kallsyms. Fix this.\n\nLink: http://lkml.kernel.org/r/20170216225721.13087-1-julia@ni.com\n\nCc: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\nSigned-off-by: Julia Cartwright \u003cjulia@ni.com\u003e\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "39f166e86bdccb33618fd3c28672c92c6d0267ca", "tree": "73e026db2f802c8bcaac21512e49f3a20f2137b7", "parents": [ "69ddfbbdd1fddcb0bf9bb9395cfd91de451a98f1" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Feb 16 17:47:26 2017 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: Add a commented single quote to the Makefile\n\nMy emacs highlighting gets confused by the single escaped quote, and makes\nthe rest of the file highlighted as if the rest of the file is part of a\nstring. By adding a commented out single quote, my emacs editor displays\neverything nicely again.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "69ddfbbdd1fddcb0bf9bb9395cfd91de451a98f1", "tree": "156be483cf0f6815061a209378845dd6197ee467", "parents": [ "e0a5f0a59d0f4fe26cbfd3a05f8f45baab7cec3a" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Feb 16 17:45:20 2017 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd build: If swig is not installed, force NO_PYTHON\u003d1\n\nWhile trying to build trace-cmd on a box without swig installed, I get an\nannoying error during build when the compile tries to build the python\nplugins. Force NO_PYTHON\u003d1 when swig is not installed, and give a warning\nabout it.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "e0a5f0a59d0f4fe26cbfd3a05f8f45baab7cec3a", "tree": "0019cf9d06e0e2894de8d9d6e56f209f3463aeb7", "parents": [ "ee713313d7e14097626e72e6c0241bebc9210e80" ], "author": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Thu Feb 16 11:50:38 2017 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: Do not initialize profiler if it isn\u0027t being used\n\nThe profiler requires scheduler tracepoints enabled, and will bug out if\nthey are not. The profiler initialization will error without these\ntracepoints defined. There\u0027s no reason to initialize the profiler on reading\na trace.dat file especially since it could cause that reading to fail if the\nprofiler required tracepoints are not there.\n\nSigned-off-by: Steven Rostedt (VMware) \u003crostedt@goodmis.org\u003e\n" }, { "commit": "ee713313d7e14097626e72e6c0241bebc9210e80", "tree": "8ced5635961b8b65de9f19530be9133de4b70d1c", "parents": [ "f2b32ffc2844858ffb21864beaa2d95e3f4b5fa9" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Nov 22 11:26:14 2016 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: Make the parse_*() functions tracecmd_parse_*()\n\nTo allow for better namespace of the tracecmd library, rename the parse_*()\nfunctions to tracecmd_parse_*() to prevent namespace collisions.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "f2b32ffc2844858ffb21864beaa2d95e3f4b5fa9", "tree": "1c2cf84c9f55781757e0f493b2020ebc96f43035", "parents": [ "7426a21e93d796c43846ae55e9761ef6f107a673" ], "author": { "name": "Josef Bacik", "email": "jbacik@fb.com", "time": "Fri Nov 20 15:57:27 2015 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: allow for custom show and handle init\n\nExternal applications that need to hook into the streaming infrastructure need a\nway to have the handle init\u0027ed in a specific way and need a way to provide their\nown read function so they can intercept events. This patch adds the neccessary\ndefinitions and hooks. Thanks,\n\nLink: http://lkml.kernel.org/r/1448053053-24188-6-git-send-email-jbacik@fb.com\n\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "7426a21e93d796c43846ae55e9761ef6f107a673", "tree": "de33221778cd410a0755d655aa290c637942aebf", "parents": [ "eaf068a84e0e238e5e06561a0bae145bf7a0b4fc" ], "author": { "name": "Josef Bacik", "email": "jbacik@fb.com", "time": "Fri Nov 20 15:57:26 2015 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: add global functions for live tracing\n\nWe need a few functions to disable/enable tracing as well as add events to be\nenabled on the first instance, this patch turns a couple of these local\nfunctions into library functions. Thanks,\n\nLink: http://lkml.kernel.org/r/1448053053-24188-5-git-send-email-jbacik@fb.com\n\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "eaf068a84e0e238e5e06561a0bae145bf7a0b4fc", "tree": "1109965cb09e3da2b2e3aaaf2b661b19326fd9e9", "parents": [ "9b01bd9aa474c58c725388e8533b0d0a8d269e57" ], "author": { "name": "Josef Bacik", "email": "jbacik@fb.com", "time": "Fri Nov 20 15:57:25 2015 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:42 2017 -0400" }, "message": "trace-cmd: add a function to create a top instance\n\nWe need a way to allow external applications to create an instance for doing\ntheir tracing, so add a helper function that takes care of creating an instance,\nsetting it as the main instance and doing the heavy lifting of creating the\ninstance in the tracing directory. Thanks,\n\nLink: http://lkml.kernel.org/r/1448053053-24188-4-git-send-email-jbacik@fb.com\n\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "9b01bd9aa474c58c725388e8533b0d0a8d269e57", "tree": "1b792e7c3533844ffd95d739fcf55497a58284d2", "parents": [ "59e4f4e9c648dada12f696efc46134418c023363" ], "author": { "name": "Josef Bacik", "email": "jbacik@fb.com", "time": "Fri Nov 20 15:57:24 2015 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:49:29 2017 -0400" }, "message": "trace-cmd: make libtracecmd a little more library friendly\n\nWhenever you link against libtracecmd you will get a bunch of debugging output\nif you don\u0027t provide your own pr_stat(). So instead default to no output if we\ndon\u0027t provide a pr_stat(). Thanks,\n\nLink: http://lkml.kernel.org/r/1448053053-24188-3-git-send-email-jbacik@fb.com\n\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\n[ Added missed updates for make gui ]\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "59e4f4e9c648dada12f696efc46134418c023363", "tree": "35a75a8668e7fee85c29c2894f5186eeab3975bd", "parents": [ "e131b50e72575b5526b592f16a3ad07748be5f11" ], "author": { "name": "Josef Bacik", "email": "jbacik@fb.com", "time": "Fri Nov 20 15:57:23 2015 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:36:26 2017 -0400" }, "message": "trace-cmd: add install_libs target to trace-cmd\n\nThis allows other commands to link against trace-cmd.\n\nLink: http://lkml.kernel.org/r/1448053053-24188-2-git-send-email-jbacik@fb.com\n\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "e131b50e72575b5526b592f16a3ad07748be5f11", "tree": "16fdf98edea55ac9e72ead02e58a6b3ad2cb2d81", "parents": [ "20cb87df8710eef56722e214a1f9958c9d7c21e2" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Nov 21 12:13:41 2016 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:36:23 2017 -0400" }, "message": "trace-cmd: Move trace-record and trace-restore out of library\n\nThe \"main\" functions should not be in the library, and trace-record and\ntrace-restore both have them. Put them back into the TRACE_CMD_OBJS and\nfigure out what breaks later.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "20cb87df8710eef56722e214a1f9958c9d7c21e2", "tree": "92378390b880e0cc74e54c5d0bc515f6a57c2813", "parents": [ "6e56f83c99f2661734f65dc1d7065b9c3c66af43" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Nov 21 11:19:57 2016 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:36:14 2017 -0400" }, "message": "trace-cmd: Add time64.h to be like the kernel\n\nAdd time64.h and use the time conversion macros there to keep parse-event.h\nsimilar to the kernel tree.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "6e56f83c99f2661734f65dc1d7065b9c3c66af43", "tree": "86fe07c69f7174ab649cb5840f9cdff489f31502", "parents": [ "ac2faf23be3d7a87d131f7e7d3fe972676be333e" ], "author": { "name": "Jon Stanley", "email": "jonstanley@gmail.com", "time": "Sat Sep 08 14:00:51 2012 -0400" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:36:09 2017 -0400" }, "message": "tools lib traceevent: Add copyright header\n\nAdding a missing copyright header to parse-utils.c. Assuminng that the\nlicense is LGPL like the rest of the trace-cmd library code.\n\nSigned-off-by: Jon Stanley \u003cjonstanley@gmail.com\u003e\nAcked-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\nLink: http://lkml.kernel.org/r/1347127251-4695-1-git-send-email-jonstanley@gmail.com\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "ac2faf23be3d7a87d131f7e7d3fe972676be333e", "tree": "0a07c2ec92205ab5906b3594e2266aead868e3d1", "parents": [ "f556063e3bee11c8bbbc02209c42669607711385" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Nov 18 18:06:43 2016 -0500" }, "committer": { "name": "Steven Rostedt (VMware)", "email": "rostedt@goodmis.org", "time": "Wed Apr 26 10:35:12 2017 -0400" }, "message": "trace-cmd: Move die() and friends out of event-utils.h and parse-utils.c\n\nAs event-utils.h and parse-utils.c are copies of the same files from the\nLinux kernel source tree, move the die() funtions out of them as they no\nlonger exist in the kernel source.\n\nMove the functions into trace-local.h. We need to update usage() of\nkernel-shark because its prototype conflicts with the usage in\ntrace-local.h.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "f556063e3bee11c8bbbc02209c42669607711385", "tree": "c67ddb1423a17698db6dc3ae83fa649a281976d1", "parents": [ "cb0b0138d04fd8efea7f5a620a29a06d6fa81dc7" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Thu Jul 14 11:23:25 2016 -0300" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Thu Nov 17 17:37:32 2016 -0500" }, "message": "tools lib traceevent: Add correct header for ipv6 definitions\n\nWe need to include netinet/in.h to get the in6_addr struct definition, needed to\nbuild it on the Android NDK:\n\n In file included from event-parse.c:36:0:\n /home/acme/android/android-ndk-r12/platforms/android-24/arch-arm/usr/include/netinet/ip6.h:82:18: error: field \u0027ip6_src\u0027 has incomplete type\n struct in6_addr ip6_src; /* source address */\n\nAnd it is the canonical way of getting IPv6 definitions, as described,\nfor instance, in Linux\u0027s \u0027man ipv6\u0027\n\nDoing that uncovers another problem: this source file uses PRIu64 but\ndoesn\u0027t include it, depending on it being included by chance via the now\nreplaced header (netinet/ip6.h), fix it.\n\nCc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nCc: Chris Phlipot \u003ccphlipot0@gmail.com\u003e\nCc: David Ahern \u003cdsahern@gmail.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Wang Nan \u003cwangnan0@huawei.com\u003e\nLink: http://lkml.kernel.org/n/tip-tilr31n3yaba1whsd47qlwa3@git.kernel.org\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "cb0b0138d04fd8efea7f5a620a29a06d6fa81dc7", "tree": "fd58a963027c2b7cdc0c6ab236a828e91205e4f8", "parents": [ "67c0816ea2bc5568e5a7f90183b1b897cade0114" ], "author": { "name": "Arnaldo Carvalho de Melo", "email": "acme@redhat.com", "time": "Fri Jul 08 15:21:37 2016 -0300" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Thu Nov 17 17:37:16 2016 -0500" }, "message": "tools lib traceevent: Use str_error_r()\n\nTo make it portable to non-glibc systems, that follow the XSI variant\ninstead of the GNU specific one that gets in place when _GNU_SOURCE is\ndefined.\n\nCc: Adrian Hunter \u003cadrian.hunter@intel.com\u003e\nCc: David Ahern \u003cdsahern@gmail.com\u003e\nCc: Jiri Olsa \u003cjolsa@kernel.org\u003e\nCc: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: Wang Nan \u003cwangnan0@huawei.com\u003e\nLink: http://lkml.kernel.org/n/tip-c1gn8x978qfop65m510wy43o@git.kernel.org\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\n[ Updated to have include/linux/string.h too ]\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "67c0816ea2bc5568e5a7f90183b1b897cade0114", "tree": "418b8c76c4c4001f6280aa7e2cbfb2af8d655aa4", "parents": [ "ee4565b8676c478a3b70c6fe351832146715fa8b" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Nov 15 16:44:01 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Nov 15 16:44:01 2016 -0500" }, "message": "trace-cmd: Update FSF address in COPYING.LIB\n\nFSF has moved, update the COPYING.LIB file.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "ee4565b8676c478a3b70c6fe351832146715fa8b", "tree": "84a30d49ca9f58b87b49ef40bf238cc241428b2d", "parents": [ "67b3610542f5b9ca4c87d46b2a02173025837c43" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Nov 15 16:34:21 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Nov 15 16:34:21 2016 -0500" }, "message": "trace-cmd: Free debug_str on error exit in tracecmd_find_tracing_dir()\n\nOn failing to mount the debugfs, free the debug_str before exiting the\nfunction.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "67b3610542f5b9ca4c87d46b2a02173025837c43", "tree": "f9c0eb93e325f9e1f8149b39d441b7c16d71dbdf", "parents": [ "8cfd82f1c16287f56f4a57306d2c709c5df1b099" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Nov 15 16:32:29 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Nov 15 16:32:29 2016 -0500" }, "message": "trace-cmd: Fix file pointer leak on error exit of tracecmd_find_tracing_dir()\n\nIf the strdup() fails to allocate, we exit tracecmd_find_tracing_dir()\nwithout closing the file descriptor just opened.\n\nReported-by: David Binderman \u003cdcb314@hotmail.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "8cfd82f1c16287f56f4a57306d2c709c5df1b099", "tree": "a9eeb4401cafb133f87eb2c31ad88e532d99817d", "parents": [ "57653fbada3060ce4aecd9361ccba0689672c49d" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Thu Sep 22 17:33:46 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Nov 01 18:02:40 2016 -0400" }, "message": "trace-cmd list: Fix the error message to show the wrong option\n\nThe index uesd optind instead of \u0027i\u0027 returning some random char.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "57653fbada3060ce4aecd9361ccba0689672c49d", "tree": "701e4072f3ae8f60f320d191930b8052788de73c", "parents": [ "17e61dba5514479fd13d45bcb6e03a7d0108720d" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Sep 16 14:41:51 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Nov 01 18:02:40 2016 -0400" }, "message": "trace-cmd msg: Add msg_write() helper for msg_do_write_check()\n\nAs the special cases in msg_do_write_check() do basically the same thing,\nadd a helper function to be called by them instead.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "17e61dba5514479fd13d45bcb6e03a7d0108720d", "tree": "80fc54f1ce80d5d16925314dd99864b7034587d0", "parents": [ "7b09ce69dafb8e4f29d2a6b54db8a4a824a7445b" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Sep 16 11:19:11 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Nov 01 18:02:40 2016 -0400" }, "message": "trace-cmd msg: Remove hook for telling trace-msg.c about debugging\n\nNow that debug is a global variable, there needs no function hook to tell\ntrace-msg.c that we are in debugging mode. Just have trace-msg.c read the\ndebug variable to know what to do.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "7b09ce69dafb8e4f29d2a6b54db8a4a824a7445b", "tree": "e1c0aa9b3c1d750544c79674a20a5ef5452787d8", "parents": [ "93d73f386a5bac3068ed7843fa96b30f8274c912" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Thu Sep 22 17:35:56 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Nov 01 18:02:40 2016 -0400" }, "message": "tracecmd list: Add --debug option\n\nAllow to debug list for agent reading.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "93d73f386a5bac3068ed7843fa96b30f8274c912", "tree": "42e0e549671136e98839e48e22f8a32b087e0865", "parents": [ "0e338e79b08567669c0f1da0e7cdd779b164f39d" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Sep 16 11:15:27 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd: Make debug a global variable\n\nIf one application enables debugging, have all files see it.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "0e338e79b08567669c0f1da0e7cdd779b164f39d", "tree": "33f1d48f645b4c6bc9ffcff436a6834a19587977", "parents": [ "7a96111611ed8e67a0a371be04779609a2483a7e" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Thu Sep 15 15:28:36 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd msg: Do not perform read when expecting no data\n\nIf the size of a read is zero, do not perform the read.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "7a96111611ed8e67a0a371be04779609a2483a7e", "tree": "162b6569a8b72760cdef14e42581139ebd6617b5", "parents": [ "7a9bb2d85096327d02836f6587de78784ce1b9ca" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Wed Sep 14 18:54:30 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd msg: Consolidate logic of allocating reads of RINIT and SENDMETA\n\nAs the logic of allocating RINIT and SENDMETA on read is the same except for\nsizes and where to store the dynamic data, create a helper funtion that\nperforms the algorithm for both of them.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "7a9bb2d85096327d02836f6587de78784ce1b9ca", "tree": "bb6e954efa751d2a9b0383da9a30e972c964e5ce", "parents": [ "5cfe15f00db02dd55f76d8f578eac20422fe2238" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Wed Sep 14 16:13:04 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd msg: Rewrote a lot of code to be more flexible in transfers\n\nCleaned up a lot of trace-msg.c to not be fixed in the sizes. Able to\nallocate only what is needed and not to be making allocation when not\nneeded. Also shortened a lot of local macro names as TRACECMD doesn\u0027t need\nto be specified when not exported for other files to share.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "5cfe15f00db02dd55f76d8f578eac20422fe2238", "tree": "23adbaf943151e38a553c07173715525754a95ea", "parents": [ "b72211fb30f605f7cc54ee066953fca8a4d7463e" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 12 15:03:58 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd listen: Simplify adding of client_pids\n\nNow that we leave holes in the client_pids array, we can remove the complex\nlogic around allocating client_pids in blocks. Just use realloc each time,\nas it only grows the array, and we never shrink it.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "b72211fb30f605f7cc54ee066953fca8a4d7463e", "tree": "746f044cdd7d356a3318334c2f86b4713ac293da", "parents": [ "9eddb072da11f13c693de2963bdb2fb9606bb521" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 12 14:15:44 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd listen: Free temp files in put_together_file()\n\nThe temp files used in put_together_file() are only used within that\nfunction. They need to be freed before leaving that function. Funny enough,\nthey are freed on failure. Since the same allocations must be freed on\nfailure as on success, have them use the same return path.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "9eddb072da11f13c693de2963bdb2fb9606bb521", "tree": "036f89995cc2af60a049ace0003d137b9be56ee3", "parents": [ "573b2fc78efaa6ec8665a6858c7656803b8626f2" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 12 14:11:54 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd listen: Free pid_list in destroy_all_readers()\n\nvalgrind showed that pid_list created by create_all_readers() was never\nfreed. The obvious place to free it is in destroy_all_readers().\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "573b2fc78efaa6ec8665a6858c7656803b8626f2", "tree": "18fed1cec90142668232cffa32f5da38f920d154", "parents": [ "3ab8dcc7dee835a0a52ad61219024b48872109b4" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 12 14:06:13 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd listen: Reuse slots in client_pid list\n\nInstead of moving around the pids in the pid list, just leave a hole. And\nhave the next allocation use it.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "3ab8dcc7dee835a0a52ad61219024b48872109b4", "tree": "9c1ae077dd66faa1fe3f7a6b2fa164ea86e3dbcd", "parents": [ "c3149d9b4cc399ab6ad4da953593c9a16b38ad72" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 12 13:26:30 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd listen: Do not clean up in signal handler\n\nDoing a memmove in the signal handler is racy with the updates of the main\nprocess. Especially if it does a realloc, and the memory moves. Do the clean\nup from the main process instead checking for EINTR in errno.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "c3149d9b4cc399ab6ad4da953593c9a16b38ad72", "tree": "6cd86eccebf00193efacb74915dcb177ddc9d20b", "parents": [ "3d463b6d949e0bef73620451ff9d31708074be28" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Sep 06 17:41:56 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:46:49 2016 -0400" }, "message": "trace-cmd listen: Make sure plog() messages get to console\n\nIn case trace-cmd seg-faults right after plog(), plog() must be flush the\nmessage to its destination before it returns, otherwise, the message could\nbe lost.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "3d463b6d949e0bef73620451ff9d31708074be28", "tree": "c1efdcfbb0dba24260a0b5ee40d968093b9b42ba", "parents": [ "511b6e4e8b087e6f13607b2aa940728ddefda539" ], "author": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Wed Oct 12 09:56:34 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Sun Oct 30 19:23:51 2016 -0400" }, "message": "trace-cmd: Add page_map for entire file\n\nAdd page mapping infrastructure between pages and the file. This will allow\nfor the entire file to be mapped, or at least much more than just a page. If\nthe mmapping fails, a smaller mapping (of a power of 2 less, or half) will\nbe tried. This will continue till the mapping is the size of a page, and if\nthat fails then the entire report will fail (as it would now).\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "511b6e4e8b087e6f13607b2aa940728ddefda539", "tree": "9eef71d38706ed5b53298fb120c1c2c88d01c52b", "parents": [ "0b4a964bcacf7db16f95eccb80d5452b1a6191ab" ], "author": { "name": "Luiz Capitulino", "email": "lcapitulino@redhat.com", "time": "Fri Oct 07 13:04:12 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Oct 28 15:50:51 2016 -0400" }, "message": "trace-cmd: Documentation: ignore all man sections\n\nToday, we only ignore section 1. But we also have a\nsection 5 man page.\n\nLink: http://lkml.kernel.org/r/20161007130412.12a445e0@redhat.com\n\nSigned-off-by: Luiz Capitulino \u003clcapitulino@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "0b4a964bcacf7db16f95eccb80d5452b1a6191ab", "tree": "20371bf85cd87493ca99c42cdb4c5094eef25ca9", "parents": [ "693ecdc486a50e1ed2a9f5542a4c70aa4d0ccaff" ], "author": { "name": "Luiz Capitulino", "email": "lcapitulino@redhat.com", "time": "Fri Oct 07 12:47:10 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Oct 28 15:30:00 2016 -0400" }, "message": "trace-cmd record: refactor set_mask()\n\nThis commit does the following:\n\n 1. Move the code that treats \"-M -1\" out of set_mask()\n 2. Drop uneeded checks from set_mask()\n 3. Make instance-\u003ecpumask point to dynamic memory\n\nThis is a preparation for supporting the \"--cpu-list\"\noption in the next commit.\n\nLink: http://lkml.kernel.org/r/1475858831-32687-3-git-send-email-lcapitulino@redhat.com\n\nSigned-off-by: Luiz Capitulino \u003clcapitulino@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "693ecdc486a50e1ed2a9f5542a4c70aa4d0ccaff", "tree": "27d9509ac60acb56a62003d7412bcf18e77c7f25", "parents": [ "e1175ca83080bb992bfb93ec0467cb040565931e" ], "author": { "name": "Luiz Capitulino", "email": "lcapitulino@redhat.com", "time": "Fri Oct 07 12:47:09 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Oct 28 15:29:07 2016 -0400" }, "message": "cpu.h: use standard types instead of glib\u0027s\n\nFuture commits will introduce standard C code that\nwants to use cpu.h. Let\u0027s convert cpu.h to standard\ntypes.\n\nI\u0027m choosing not to change the gui code that uses\ncpu.h though. I think that in all platforms where\nglib is supported, guint64 should always be uint64_t\nand guint should always be unsigned int.\n\nLink: http://lkml.kernel.org/r/1475858831-32687-2-git-send-email-lcapitulino@redhat.com\n\nSigned-off-by: Luiz Capitulino \u003clcapitulino@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "e1175ca83080bb992bfb93ec0467cb040565931e", "tree": "8336c1da77873f7c5d7bafcef8f66ddc6bd4f55e", "parents": [ "5670dbf636d1d37842a69e884dfef5f2519a2384" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Thu Oct 20 18:31:27 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Thu Oct 20 18:31:27 2016 -0400" }, "message": "trace-view: Place focus back into search box after search\n\nWhen searching the tree rows, after an item is found, it would be nice to\nrepeat the search with a keyboard instead of having to move the mouse back\nto the entry and selecting again. Instead, have the focus come back to the\nsearch entry box and allow the search to happen again with just hitting\n\"enter\".\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "5670dbf636d1d37842a69e884dfef5f2519a2384", "tree": "1f7ccd63f3740f7a08588c6b12174696b9810200", "parents": [ "5758cd040588d0fec656ee79d690920f894968ec" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Wed Oct 05 14:49:13 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Wed Oct 05 14:49:13 2016 -0400" }, "message": "kernelshark: Add infrastructure to allow kernelshark to have plugins\n\nAdd a -p plugin command line option for kernelshark to accept a plugin that\ncan modify how kernelshark works. This is just adds a loader, currently\nnothing uses it yet.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "5758cd040588d0fec656ee79d690920f894968ec", "tree": "971485769ffb7f095aecbbcb2346412b41119f3f", "parents": [ "feaf0a2cccdc2977b2924fcc9ee779e80ee8c44f" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Wed Oct 05 12:17:54 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Wed Oct 05 12:17:54 2016 -0400" }, "message": "trace-graph: Show events missed in cpu plot\n\nIf events were dropped, when hovering over the CPU plot, show that events\nwere missed.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "feaf0a2cccdc2977b2924fcc9ee779e80ee8c44f", "tree": "706ef1cef4ffb944566eabc3e3833be20ac89d8d", "parents": [ "013205a8f9e60998b39c106470957030585b5851" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Oct 03 16:11:15 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Oct 03 16:13:36 2016 -0400" }, "message": "trace-graph: Do not show \u0027interrupts\u0027 if no interrupt event was added\n\nSince showing interrupts for when events are in interrupts can cause the\ntask plots to be crazy if the interrupt events are not traced, do not do the\n\"in interrupt\" notation unless interrupts were traced.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "013205a8f9e60998b39c106470957030585b5851", "tree": "f51936bd6cc9a71beccd41f3367e5e64cd755a0d", "parents": [ "fca725695474085397339abe07de4a91a781269e" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Oct 03 10:09:04 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Oct 03 10:09:04 2016 -0400" }, "message": "kernelshark: Fix null pointer when next record does not exist\n\nWhen testing if the plot is in irq context or not, it looks at the current\nrecord as well as the next record to see the context. Unfortunately, if the\nnext record does not exist (null is returned) it would cause a crash.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "fca725695474085397339abe07de4a91a781269e", "tree": "e4ecd2c68e3d45c884815368c05f357511299c1d", "parents": [ "65ffbd349250575d0549841f0517bd64d7084a0e" ], "author": { "name": "Daniel Bristot de Oliveira", "email": "bristot@redhat.com", "time": "Mon Jul 20 14:51:00 2015 -0300" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Jul 15 15:23:37 2016 -0400" }, "message": "trace-cmd: record crashes if -f is used before -e event.\n\nThe trace-cmd record command is crashing if a -f filter is used before\na -e event.\n\nThe usage a -f before a -e is a command line error, and trace-cmd\nrecord detects this error by checking if the last_event pointer is NULL.\n\nHowever, the last_event is not initialized as NULL and this command line\nerror is not detected. Then, the last_event is used to save the filter\noptions and this causes a segmentation fault because last_event was not\ninitialized yet.\n\nThis problem is fixed by initializing last_event as NULL.\n\nHow-to reproduce:\n # trace-cmd record -f foo\n Segmentation fault\n\nExpected output:\n # trace-cmd record -f foo\n filter must come after event\n\nLink: http://lkml.kernel.org/r/894eb6c8a5665fba64811395e0d1204e201f4992.1437414265.git.bristot@redhat.com\n\nTested-by: John Kacur \u003cjkacur@redhat.com\u003e\nSigned-off-by: Daniel Bristot de Oliveira \u003cbristot@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "65ffbd349250575d0549841f0517bd64d7084a0e", "tree": "0e4c8aa80286648ccbd840b0f1d129bd011a5856", "parents": [ "73723c0bb51262715564ebb2a6cf435603779668" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Jul 12 09:28:09 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Jul 12 09:28:09 2016 -0400" }, "message": "event-parse: Add filter on task CPU id\n\nAdd a \u0027CPU\u0027 special field to allow the filter in trace-cmd report\nto filter on the task\u0027s CPU.\n\nBy adding a special field \u0027CPU\u0027 (all caps) the user can now\nfilter out tasks based on which CPU they are on. This is useful when\nfiltering out (or in) a bunch of threads.\n\n -F \u0027CPU \u003d\u003d 0\u0027\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "73723c0bb51262715564ebb2a6cf435603779668", "tree": "2963db641b69e1402b0b72c1b665fe95340bf33d", "parents": [ "64bea4bc31b27f088e38a9f941a0e93c77d5bb18" ], "author": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Wed May 11 15:09:36 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Jul 12 09:22:52 2016 -0400" }, "message": "tools lib traceevent: Do not reassign parg after collapse_tree()\n\nAt the end of process_filter(), collapse_tree() was changed to update\nthe parg parameter, but the reassignment after the call wasn\u0027t removed.\n\nWhat happens is that the \"current_op\" gets modified and freed and parg\nis assigned to the new allocated argument. But after the call to\ncollapse_tree(), parg is assigned again to the just freed \"current_op\",\nand this causes the tool to crash.\n\nThe current_op variable must also be assigned to NULL in case of error,\notherwise it will cause it to be free()ed twice.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\nAcked-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nCc: stable@vger.kernel.org # 3.14+\nFixes: 42d6194d133c (\"tools lib traceevent: Refactor process_filter()\")\nLink: http://lkml.kernel.org/r/20160511150936.678c18a1@gandalf.local.home\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "64bea4bc31b27f088e38a9f941a0e93c77d5bb18", "tree": "0e6859191a1fac89722f8747ad2911ee21405321", "parents": [ "6f79497e85dd7924365633b479b00ad1116a03d5" ], "author": { "name": "Namhyung Kim", "email": "namhyung@gmail.com", "time": "Fri Jul 08 14:56:12 2016 +0900" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Jul 12 08:32:44 2016 -0400" }, "message": "trace-cmd: Use tracecmd_peek_next_data() in fgraph_ent_handler\n\nWhen a task was migrated to other cpu in the middle of a function, the\nfgraph_exit record will be in a different cpu than the enter record.\nBut currently fgraph_ent_handler() only peeks at the same cpu so it\ncould read an incorrect record.\n\nFor example, please see following raw records:\n\n bash-10478 [007] 73.454273: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [000] 73.454650: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x111a37483c rettime\u003d0x111a3d0285 overrun\u003d0x0 depth\u003d0\n bash-10478 [000] 74.456383: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [000] 74.456655: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x1155f24337 rettime\u003d0x1155f66559 overrun\u003d0x0 depth\u003d0\n bash-10478 [000] 75.458517: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [001] 75.458849: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x1191ad9de0 rettime\u003d0x1191b2a6aa overrun\u003d0x0 depth\u003d0\n bash-10478 [001] 76.460482: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [000] 76.460679: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x11cd6662b4 rettime\u003d0x11cd695e03 overrun\u003d0x0 depth\u003d0\n bash-10478 [000] 77.462334: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [004] 77.462564: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x12091d71c4 rettime\u003d0x120920e977 overrun\u003d0x0 depth\u003d0\n bash-10478 [004] 78.464315: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [001] 78.464644: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x1244d674de rettime\u003d0x1244db7329 overrun\u003d0x0 depth\u003d0\n bash-10478 [001] 79.466018: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [004] 79.466326: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x12808b3940 rettime\u003d0x12808fe819 overrun\u003d0x0 depth\u003d0\n bash-10478 [004] 80.468005: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [002] 80.468291: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x12bc44551f rettime\u003d0x12bc48ac9a overrun\u003d0x0 depth\u003d0\n bash-10478 [002] 81.469718: funcgraph_entry: func\u003d0xffffffff8123bf90 depth\u003d0\n bash-10478 [007] 81.470088: funcgraph_exit: func\u003d0xffffffff8123bf90 calltime\u003d0x12f7f945b8 rettime\u003d0x12f7fee028 overrun\u003d0x0 depth\u003d0\n\nThe first entry was call to cma_alloc function, it was on cpu 7 but the\ntask was migrated to cpu 0 before returning from the function.\nCurrently trace-cmd shows like below:\n\n bash-10478 [007] 73.454273: funcgraph_entry: ! 367.216 us | cma_alloc();\n bash-10478 [000] 73.454650: funcgraph_exit: ! 375.369 us | }\n bash-10478 [000] 74.456383: funcgraph_entry: ! 270.882 us | cma_alloc();\n bash-10478 [000] 75.458517: funcgraph_entry: ! 195.407 us | cma_alloc();\n bash-10478 [001] 75.458849: funcgraph_exit: ! 329.930 us | }\n bash-10478 [001] 76.460482: funcgraph_entry: ! 327.243 us | cma_alloc();\n bash-10478 [000] 77.462334: funcgraph_entry: ! 293.465 us | cma_alloc();\n bash-10478 [004] 77.462564: funcgraph_exit: ! 227.251 us | }\n bash-10478 [004] 78.464315: funcgraph_entry: ! 306.905 us | cma_alloc();\n bash-10478 [001] 79.466018: funcgraph_entry: ! 303.196 us | cma_alloc();\n bash-10478 [004] 80.468005: funcgraph_entry: | cma_alloc() {\n bash-10478 [002] 80.468291: funcgraph_exit: ! 284.539 us | }\n bash-10478 [002] 81.469718: funcgraph_entry: ! 323.215 us | cma_alloc();\n\nThis is because the first funcgraph_entry on cpu 7 matched to the last\nfuncgraph_exit on cpu 7. And second funcgraph_exit on cpu 0 was shown\nalone. We need to match record from all cpu rather than the same cpu.\nIn this case, entry on cpu 7 should be paired with exit on cpu 0.\n\nWith this patch, the output look like below:\n\n bash-10478 [007] 73.454273: funcgraph_entry: ! 375.369 us | cma_alloc();\n bash-10478 [000] 74.456383: funcgraph_entry: ! 270.882 us | cma_alloc();\n bash-10478 [000] 75.458517: funcgraph_entry: ! 329.930 us | cma_alloc();\n bash-10478 [001] 76.460482: funcgraph_entry: ! 195.407 us | cma_alloc();\n bash-10478 [000] 77.462334: funcgraph_entry: ! 227.251 us | cma_alloc();\n bash-10478 [004] 78.464315: funcgraph_entry: ! 327.243 us | cma_alloc();\n bash-10478 [001] 79.466018: funcgraph_entry: ! 306.905 us | cma_alloc();\n bash-10478 [004] 80.468005: funcgraph_entry: ! 284.539 us | cma_alloc();\n bash-10478 [002] 81.469718: funcgraph_entry: ! 367.216 us | cma_alloc();\n\nMaybe we can separate enter and exit if they happened on different\ncpu. Anyway the time duration has correct value now.\n\nLink: http://lkml.kernel.org/r/20160708055612.32221-2-namhyung@gmail.com\n\nReported-by: Joonsoo Kim \u003ciamjoonsoo.kim@lge.com\u003e\nSigned-off-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "6f79497e85dd7924365633b479b00ad1116a03d5", "tree": "28960a6f74cb1e98c0f6ab1af53ef76a44bdcecf", "parents": [ "496c5d9ce71e8648af4be7bf0912f86f1ea4f19e" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Jul 11 13:00:56 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Jul 11 13:00:56 2016 -0400" }, "message": "trace-cmd: Consolidate tracecmd_{read,peak}_next_data()\n\nNow that there\u0027s a tracecmd_peak_next_data() that is a subset of the\nfunctionality of tracecmd_read_next_data(), have tracecmd_read_next_data()\ncall tracecmd_peak_next_data() and remove the duplicate code.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "496c5d9ce71e8648af4be7bf0912f86f1ea4f19e", "tree": "37829cb42ff65797f5e4a5ec0cd1ffe8b0b121ad", "parents": [ "4421d9551b903d720bb8534a02c3564187eab045" ], "author": { "name": "Namhyung Kim", "email": "namhyung@gmail.com", "time": "Fri Jul 08 14:56:11 2016 +0900" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Jul 11 12:49:16 2016 -0400" }, "message": "trace-cmd: Introduce tracecmd_peek_next_data()\n\nThe tracecmd_peek_next_data() is similar to tracecmd_read_next_data()\nbut it doesn\u0027t consume the record.\n\nLink: http://lkml.kernel.org/r/20160708055612.32221-1-namhyung@gmail.com\n\nSigned-off-by: Namhyung Kim \u003cnamhyung@kernel.org\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "4421d9551b903d720bb8534a02c3564187eab045", "tree": "49184e6d4b22a55b8cde17ae5a1823a6e41c1369", "parents": [ "dd45866263ffc10f994572daea31580910fd1bfa" ], "author": { "name": "Jan Kiszka", "email": "jan.kiszka@siemens.com", "time": "Sun May 22 18:02:42 2016 +0200" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Jul 01 11:14:05 2016 -0400" }, "message": "trace-cmd: Add more SVM exit reasons\n\nExceptions require individual decoding (only feasible intercepts\nlisted), XSETBV was missing and the AVIC brought in two new exit codes.\n\nLink: http://lkml.kernel.org/r/5741D822.3030203@web.de\n\nSigned-off-by: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "dd45866263ffc10f994572daea31580910fd1bfa", "tree": "5e3b3c1996d7224f081c3696d5f154f82054ed8e", "parents": [ "f0ed3b54ed6535055facc0fc4027e8a752fa5c6d" ], "author": { "name": "Jan Kiszka", "email": "jan.kiszka@siemens.com", "time": "Sun May 22 18:02:31 2016 +0200" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Jul 01 11:07:14 2016 -0400" }, "message": "trace-cmd: Fix reporting of unknown SVM exit reasons\n\nOn AMD, exist code -1 is also a possible value, but we use it for\nterminating the list of known exit reasons. This leads to EXIT_ERR\nbeing reported for unkown ones. Fix this by using an NULL string\npointer as terminal.\n\nLink: http://lkml.kernel.org/r/5741D817.3070902@web.de\n\nSigned-off-by: Jan Kiszka \u003cjan.kiszka@siemens.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "f0ed3b54ed6535055facc0fc4027e8a752fa5c6d", "tree": "fbc34c22d56f289171ac7fe1b9995b111865ccd3", "parents": [ "b193778aa7ee91e7677e2544353d3ce1844c4e63" ], "author": { "name": "Omar Sandoval", "email": "osandov@fb.com", "time": "Wed Jun 15 13:18:21 2016 -0700" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Jun 17 12:12:45 2016 -0400" }, "message": "trace-cmd record: add --max-graph-depth option\n\nThis is exposed in debugfs but there\u0027s no quick way to set it with\ntrace-cmd. While we\u0027re adding support for this, fix --profile so it sets\nmax_graph_depth back to whatever it was before.\n\nLink: http://lkml.kernel.org/r/52d19bb3c0daf4d03db68da438c605b318ca41a8.1466021761.git.osandov@fb.com\n\nSigned-off-by: Omar Sandoval \u003cosandov@fb.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "b193778aa7ee91e7677e2544353d3ce1844c4e63", "tree": "ee62b6c126e771a9935bba18bdd8623a7ce1d370", "parents": [ "2ef76d203d291efbf5004484bbd5c27f697801ac" ], "author": { "name": "Eric Engestrom", "email": "eric.engestrom@imgtec.com", "time": "Mon Apr 25 14:54:42 2016 +0100" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Apr 26 09:43:03 2016 -0400" }, "message": "trace-cmd record: Fix `const` placement\n\nIn `const char const * files[]`, both `const` refer to `char`.\nMoving the second one after the `*` makes the pointer const as well,\nwhich is probably what was intended.\n\nSigned-off-by: Eric Engestrom \u003ceric.engestrom@imgtec.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "2ef76d203d291efbf5004484bbd5c27f697801ac", "tree": "8709d0694b41fde8fef69c6cd360559245697631", "parents": [ "01b3337df3544cef42be5243a25ed4e005c80efa" ], "author": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Wed Mar 23 10:16:28 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Wed Mar 23 10:21:17 2016 -0400" }, "message": "tools lib traceevent: Remove redundant CPU output\n\nCommit 8533bda27f64 \"tools lib traceevent: Split pevent_print_event()\ninto specific functionality functions\" Broke apart the function\npevent_print_event() into three functions. The first function prints\nthe comm, pid and CPU, the second prints the timestamp. But that commit\nadded the printing of the CPU in the timestamp function, which now\ncauses pevent_print_event() to duplicate the CPU output.\n\nRemove the redundant printing of the record\u0027s CPU from the timestamp\nfunction.\n\nLink: http://lkml.kernel.org/r/20160323101628.459375d2@gandalf.local.home\n\nFixes: 8533bda27f64 \"tools lib traceevent: Split pevent_print_event() into specific functionality functions\"\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "01b3337df3544cef42be5243a25ed4e005c80efa", "tree": "b2f1e2c57cfc43f7b4ac85917f93b2f02bebd4dc", "parents": [ "bdf9ecb7e2b8b22484e94001530607de65b949e8" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Sep 18 15:29:18 2015 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Mar 22 10:48:53 2016 -0400" }, "message": "trace-cmd: Always use /var/run and /var/lib\n\nLooking at documentation on the Internet, it seems that even if a tool is\ninstalled in /usr/local, it still should be using /var/run and /var/lib, as\nthose are not places that will get overrided by upgrades, but all versions\nof trace-cmd should know about it.\n\nI still keep it dynamic for installing in $HOME, as that may not be done by\na administrative user.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "bdf9ecb7e2b8b22484e94001530607de65b949e8", "tree": "37e5158f41f01aa3db4de72e0573c057726075f3", "parents": [ "a5485dd1fbf166b9c29c8a8b1edb7cda0e93e15e" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Sep 18 15:13:14 2015 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Mar 22 10:41:31 2016 -0400" }, "message": "trace-cmd listen: Change install location to handle /usr/local\n\nIf prefix is \"/usr\" then use the \"/var\" directory, but for everything else\ninstall in \"$(prefix)/var\".\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "a5485dd1fbf166b9c29c8a8b1edb7cda0e93e15e", "tree": "a25068dc2641b208edae1c77ce79d2bef7973606", "parents": [ "307bdda7716074b67f074982130c1e4d4ae05d05" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri Sep 18 16:00:22 2015 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Mar 22 10:35:40 2016 -0400" }, "message": "trace-cmd listen: Add a pid file when in daemon mode\n\nWhen trace-cmd listen is started in daemon mode, create the file\n/var/run/trace-cmd-net.pid that holds the pid of the daemon.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "307bdda7716074b67f074982130c1e4d4ae05d05", "tree": "dbe9d04fe18365f1c719902cec0c8111ce8210d2", "parents": [ "42fc96f520796c7d18c7372b6ad80292d4a0bc21" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 14 17:44:12 2015 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 18:07:05 2016 -0400" }, "message": "trace-cmd: Add debug option to tell msg protocol not to timeout\n\nIn order to debug the message protocol, we do not want to time out, as we\nneed to run gdb on the client and server, and timeouts kill the connections\nwe want to debug.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "42fc96f520796c7d18c7372b6ad80292d4a0bc21", "tree": "4813b0ec37267382eb700fabc5ea54979eb314b9", "parents": [ "d4321519aec1d6aba5edd5582c21903132990b3d" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 14 14:00:16 2015 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 18:05:51 2016 -0400" }, "message": "trace-cmd: Make msgcpy() and optcpy() check for size\n\nHave msgcpy() and optcpy() test the offset + buflen to see if it would\noverwrite the size of the msg or opt. If so, return -ENIVAL. The checking of\nthis return value still needs to be done, but at least we wont be\naccidentally overwriting the memory.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "d4321519aec1d6aba5edd5582c21903132990b3d", "tree": "cf777cd1c4721ca00c80906921b3ee474f55e196", "parents": [ "8232b403ffd064fc100d5723c16069cda4ca49e4" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Sep 14 13:56:05 2015 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 18:05:48 2016 -0400" }, "message": "trace-cmd: Replace bufcpy() with msgcpy() and optcpy()\n\nThe bufcpy() wrapper to memcpy() is not very useful except that it may save\non a typecast.\n\nCreate a msgcpy() and optcpy() to use instead that is type safe and also\nallows for checking of memory size overflows in the future.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "8232b403ffd064fc100d5723c16069cda4ca49e4", "tree": "6ea50a31ff2e06ab58d2942d52846e1d8c883c6a", "parents": [ "cc042aba4a97aea452c7943a00d0147fcc5aa7ce" ], "author": { "name": "Yoshihiro YUNOMAE", "email": "yoshihiro.yunomae.ez@hitachi.com", "time": "Fri Jul 11 00:58:29 2014 +0000" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 17:18:44 2016 -0400" }, "message": "trace-cmd/msg: Use poll(2) to wait for a message\n\nUse poll(2) to wait for a message. If a client/server cannot send a message for\nany reasons, the current server/client will wait in a blocking read operation.\nSo, we use poll(2) for avoiding remaining in a blocking state.\n\nChanges in V4: Change the argument of tracecmd_msg_recv_wait()\n Fix some typos\n\nLink: http://lkml.kernel.org/r/20140711005829.25516.75538.stgit@yuno-kbuild.novalocal\n\nSigned-off-by: Yoshihiro YUNOMAE \u003cyoshihiro.yunomae.ez@hitachi.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "cc042aba4a97aea452c7943a00d0147fcc5aa7ce", "tree": "029fe861ba7eefe4efe7c97018f24b9f5c0ac601", "parents": [ "7ea7f9846352f4b607bb1cc7e7b680eca61c12e4" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Fri May 15 12:15:53 2015 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 17:18:44 2016 -0400" }, "message": "trace-cmd listen: Update the V2 protocol to handle old and new\n\nChange the protocol slightly where V1 will fail right away without creating\na blank trace.dat file by appending \u0027-1\u0027 to the V2 name. The old trace-cmd\nlisten will see that as a negative CPU count and bail.\n\nAlso change the hand shake a bit such that if we go to a V3, a V2 server\ncould just let the newer trace-cmd client know what version it can support.\n\nHere\u0027s the update:\n\n \u003cserver\u003e \u003cclient\u003e\n listen to socket fd\n connect to socket fd\n accept the client\n send \"tracecmd\"\n +------------\u003e receive \"tracecmd\"\n check \"tracecmd\"\n* send \"-1V2\\0\"\n* try_again:\n* receive \"-1V2\" \u003c------------+\n* check \"-1V2\"\n* if (!\"-1V2\") {\n* if (memcmp(\"-1V\",3))\n* error();\n* send \"V2\\0\"\n* goto try_again;\n* }\n* send \"V2\"\n* +---------------\u003e receive \"V2\"\n* check \"V2\"\n* send \"\u003cMAGIC_NUMBER\u003e\\00\" as the v2 protocol\n\n read \"\u003cMAGIC_NUMBER\u003e\\00\"\n* send \"OK\"\n* +---------------\u003e receive \"OK\"\n* check \"OK\"\n send cpus,pagesize,option(MSG_TINIT)\n receive MSG_TINIT \u003c-------+\n print \"cpus\u003dXXX\"\n print \"pagesize\u003dXXX\"\n understand option\n send port_array\n +--MSG_RINIT-\u003e receive MSG_RINIT\n understand port_array\n send meta data(MSG_SENDMETA)\n receive MSG_SENDMETA \u003c----+\n record meta data\n (snip)\n send a message to finish sending meta data\n | (MSG_FINMETA)\n receive MSG_FINMETA \u003c-----+\n read block\n --- start sending trace data on child processes ---\n\n --- When client finishes sending trace data ---\n send MSG_CLOSE\n receive MSG_CLOSE \u003c-------+\n close(socket fd) close(socket fd)\n\nThe \u0027*\u0027 represents what was updated.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "7ea7f9846352f4b607bb1cc7e7b680eca61c12e4", "tree": "041e6d0807a5bf830f1f1b65503488d037c6e38d", "parents": [ "8bbbaaaf78ecd479d149fc84ab7d50be90c3540d" ], "author": { "name": "Yoshihiro YUNOMAE", "email": "yoshihiro.yunomae.ez@hitachi.com", "time": "Fri Jul 11 00:58:26 2014 +0000" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 17:18:44 2016 -0400" }, "message": "trace-cmd/listen: Apply the trace-msg protocol for communication between a server and clients\n\nApply trace-msg protocol for communication between a server and clients.\n\nCurrently, trace-listen(server) and trace-record -N(client) operate as follows:\n\n \u003cserver\u003e \u003cclient\u003e\n listen to socket fd\n connect to socket fd\n accept the client\n send \"tracecmd\"\n +------------\u003e receive \"tracecmd\"\n check \"tracecmd\"\n send cpus\n receive cpus \u003c------------+\n print \"cpus\u003dXXX\"\n send pagesize\n |\n receive pagesize \u003c--------+\n print \"pagesize\u003dXXX\"\n send option\n |\n receive option \u003c----------+\n understand option\n send port_array\n +------------\u003e receive port_array\n understand port_array\n send meta data\n receive meta data \u003c-------+\n record meta data\n (snip)\n read block\n --- start sending trace data on child processes ---\n\n --- When client finishes sending trace data ---\n close(socket fd)\n read size \u003d 0\n close(socket fd)\n\nAll messages are unstructured character strings, so server(client) using the\nprotocol must parse the unstructured messages. Since it is hard to\nadd complex contents in the protocol, structured binary message trace-msg\nis introduced as the communication protocol.\n\nBy applying this patch, server and client operate as follows:\n\n \u003cserver\u003e \u003cclient\u003e\n listen to socket fd\n connect to socket fd\n accept the client\n send \"tracecmd\"\n +------------\u003e receive \"tracecmd\"\n check \"tracecmd\"\n send \"V2\\0\u003cMAGIC_NUMBER\u003e\\00\" as the v2 protocol\n receive \"V2\" \u003c------------+\n check \"V2\"\n read \"\u003cMAGIC_NUMBER\u003e\\00\"\n send \"V2\"\n +---------------\u003e receive \"V2\"\n check \"V2\"\n send cpus,pagesize,option(MSG_TINIT)\n receive MSG_TINIT \u003c-------+\n print \"cpus\u003dXXX\"\n print \"pagesize\u003dXXX\"\n understand option\n send port_array\n +--MSG_RINIT-\u003e receive MSG_RINIT\n understand port_array\n send meta data(MSG_SENDMETA)\n receive MSG_SENDMETA \u003c----+\n record meta data\n (snip)\n send a message to finish sending meta data\n | (MSG_FINMETA)\n receive MSG_FINMETA \u003c-----+\n read block\n --- start sending trace data on child processes ---\n\n --- When client finishes sending trace data ---\n send MSG_CLOSE\n receive MSG_CLOSE \u003c-------+\n close(socket fd) close(socket fd)\n\nBy introducing the v2 protocol, after the client checks \"tracecmd\", the client\nwill send \"V2\\0\u003cMAGIC_NUMBER\u003e\\00\\0\". This complex message is used when the\nnew client tries to connect to the old server. The new client wants to check\nwhether the reply message from the server is \"V2\" or not. However, the old\nserver does not respond to the client before receiving cpu numbers, page size,\nand options. Each message is separated with \"\\0\" in the old server, so the\nclient send \"V2\" as cpu numbers, \"\u003cMAGIC_NUMBER\u003e\" as page size, and \"0\" as\nno options. On the other hands, the old server will understand the messages\nas cpus\u003d0, pagesize\u003d\u003cMAGIC_NUMBER\u003e, and options\u003d0, and then the server will\nsend the message \"\\0\" as port numbers. Then, the message which the client\nreceives is not \"V2\" but \"\\0\", so the client will reconnect to the old server\nas the v1 protocol.\n\n\u003cHow to test\u003e\n[1] Backward compatability checks\n We need to test backward compatability of this patch for old\n trace-cmds(client/server). So, this patch was tested for [2] command checks in\n following 3 types:\n\n\t\u003cclient\u003e\t\u003cserver\u003e\n\t new\t\t old\n\t old\t\t new\n\t new\t\t new\n\n[2] Command checks\n- server (common)\n\t# trace-cmd listen -p 12345\n\n1) record\n- client\n\t# trace-cmd record -e sched -N \u003cserver IP\u003e:12345\n\t^C\n\n2) record + multiple buffers\n- client\n\t# trace-cmd record -B foo -e sched -N \u003cserver IP\u003e:12345\n\t^C\n\n3) extract\n- client\n\t# ./trace-cmd start -e sched\n\t# sleep 5\n\t# ./trace-cmd extract -N \u003cserver IP\u003e:12345\n\n4) extract + snapshot\n- client\n\t# ./trace-cmd start -e sched\n\t# sleep 5\n\t# ./trace-cmd snapshot -s\n\t# ./trace-cmd extract -N \u003cserver IP\u003e:12345 -s\n\nChanges in V4: Fix some typos, cleanups and rebase for current trace-cmd-v2.4\n Change the argument of tracecmd_msg_recv()\nChanges in V3: Change the license of trace-msg.c to LGPL v2.1\nChanges in V2: Regacy porotocol support in order to keep backward compatibility\n\nLink: http://lkml.kernel.org/r/20140711005826.25516.77711.stgit@yuno-kbuild.novalocal\n\nSigned-off-by: Yoshihiro YUNOMAE \u003cyoshihiro.yunomae.ez@hitachi.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "8bbbaaaf78ecd479d149fc84ab7d50be90c3540d", "tree": "d4098f22f8830a14dc549cb3fcd17229ab3cb251", "parents": [ "797db1e341ce7d85a41ead4cf632a28235cb8a69" ], "author": { "name": "Peter Xu", "email": "peterx@redhat.com", "time": "Thu Mar 17 09:19:27 2016 +0800" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 16:04:57 2016 -0400" }, "message": "trace-cmd: Keep default bash completion to list files\n\nThis will keep trace-cmd default completion behave like before.\n\nSigned-off-by: Peter Xu \u003cpeterx@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "797db1e341ce7d85a41ead4cf632a28235cb8a69", "tree": "d4f72eb0fc7a55e9722b11811e3b109b376bbfa0", "parents": [ "7477e7057e201466b2fa801f7a47ff5976cb06ab" ], "author": { "name": "Peter Xu", "email": "peterx@redhat.com", "time": "Mon Mar 14 14:49:32 2016 +0800" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 16:04:48 2016 -0400" }, "message": "trace-cmd: Add bash completion for \"-{e|p|l|n|g}\"\n\n\"-e\", \"-p\" and \"-{l|n|g}\" are frequently used parameters in many\nsub-commands, adding auto-completion support for them.\n\nIt may be very slow when triggering \"-{l|n|g}\" auto completion. However,\nit\u0027s better than nothing.\n\nSigned-off-by: Peter Xu \u003cpeterx@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "7477e7057e201466b2fa801f7a47ff5976cb06ab", "tree": "0fc84593b6d88a47e2c29f2ecc23f1bd83ec7976", "parents": [ "1fb72b9b91fb537c32705b2c505c3b2540f9aab2" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Wed Mar 16 11:42:49 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 21 15:29:21 2016 -0400" }, "message": "kernelshark: Handle freeing of int arrays\n\nHaving the hard irq integer arrays be set to null_int_array requires that\nthey are not freed with the normal free() function, but needs to check that\nthey have been allocated and does not equal to the null_int_array variable.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "1fb72b9b91fb537c32705b2c505c3b2540f9aab2", "tree": "c05f0532d63ed4acc5f7cb09e75c0ff80d383704", "parents": [ "77bdcb9b291e6842de79b1d421aacff74283ba02" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Mon Mar 14 15:32:43 2016 -0400" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Mar 14 15:32:43 2016 -0400" }, "message": "kernelshark: Have irq event arrays point to other than NULL\n\nIf there are no irq events, the task plots will crash as it will try to\ndereference the irq array, which would still be NULL.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "77bdcb9b291e6842de79b1d421aacff74283ba02", "tree": "206b9f616121b0f5c52aa8a814471c3e1b6e0b7d", "parents": [ "4a49c45559bdd144a775da40446a20f16fc19ba0" ], "author": { "name": "Peter Xu", "email": "peterx@redhat.com", "time": "Thu Mar 10 14:37:33 2016 +0800" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Thu Mar 10 17:17:21 2016 -0500" }, "message": "trace-cmd: Install trace-cmd.bash when installing\n\nSigned-off-by: Peter Xu \u003cpeterx@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "4a49c45559bdd144a775da40446a20f16fc19ba0", "tree": "526542e007d89e910c6d009c25f299385e73e1e7", "parents": [ "c21aae2c80718d4820a61a96aaa1533d358f1b1c" ], "author": { "name": "Peter Xu", "email": "peterx@redhat.com", "time": "Thu Mar 10 14:37:32 2016 +0800" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Thu Mar 10 17:16:23 2016 -0500" }, "message": "trace-cmd,bash: Add auto-complete for bash\n\nThis is the most simplified version. Only pop up sub-cmds and nothing\nelse.\n\nSigned-off-by: Peter Xu \u003cpeterx@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "c21aae2c80718d4820a61a96aaa1533d358f1b1c", "tree": "1d9b14ff4bd5b1a84d70b0154fc6fe70b7b010b3", "parents": [ "8533bda27f64d2b72d95c1fb585f5310f7253106" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Mar 01 17:22:32 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Mar 01 17:24:33 2016 -0500" }, "message": "trace-cmd report: Add --ts-diff to show the delta between timestamps\n\nIf --ts-diff is added to trace-cmd report, it will show the delta between\nevents. It doesn\u0027t matter which CPUs they are on. In case of filtering, the\ndelta is still only between the visible events.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "8533bda27f64d2b72d95c1fb585f5310f7253106", "tree": "381216ec2e61c97d11147b6f4fb05efba44da274", "parents": [ "0b012aab38f828e5c33f820d353c06eeff1e50c5" ], "author": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Mon Feb 29 09:01:28 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Mar 01 16:58:41 2016 -0500" }, "message": "tools lib traceevent: Split pevent_print_event() into specific functionality functions\n\nCurrently there\u0027s a single function that is used to display a record\u0027s\ndata in human readable format. That\u0027s pevent_print_event().\nUnfortunately, this gives little room for adding other output within\nthe line without updating that function call.\n\nI\u0027ve decided to split that function into 3 parts.\n\n pevent_print_event_task() which prints the task comm, pid and the CPU\n pevent_print_event_time() which outputs the record\u0027s timestamp\n pevent_print_event_data() which outputs the rest of the event data.\n\npevent_print_event() now simply calls these three functions.\n\nTo save time from doing the search for event from the record\u0027s type, I\ncreated a new helper function called pevent_find_event_by_record(),\nwhich returns the record\u0027s event, and this event has to be passed to\nthe above functions.\n\nLink: http://lkml.kernel.org/r/20160229090128.43a56704@gandalf.local.home\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "0b012aab38f828e5c33f820d353c06eeff1e50c5", "tree": "70b9d5d98fe73606bfbb24f96d2c118c9eabf4e9", "parents": [ "a6a8531330bcdff29477f9ccfefcacba34e24df9" ], "author": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Fri Feb 26 18:13:28 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Mar 01 16:58:15 2016 -0500" }, "message": "tools lib traceevent: Add \u0027~\u0027 operation within arg_num_eval()\n\nWhen evaluating values for print flags, if the value included a \u0027~\u0027\noperator, the parsing would fail. This broke kmalloc\u0027s parsing of:\n\n__print_flags(REC-\u003egfp_flags, \"|\", {(unsigned\nlong)((((((( gfp_t)(0x400000u|0x2000000u)) | (( gfp_t)0x40u) |\n(( gfp_t)0x80u) | (( gfp_t)0x20000u)) | (( gfp_t)0x02u)) |\n(( gfp_t)0x08u)) | (( gfp_t)0x4000u) | (( gfp_t)0x10000u) |\n(( gfp_t)0x1000u) | (( gfp_t)0x200u)) \u0026 ~(( gfp_t)0x2000000u))\n ^\n |\n here\n\nLink: http://lkml.kernel.org/r/20160226181328.22f47129@gandalf.local.home\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "a6a8531330bcdff29477f9ccfefcacba34e24df9", "tree": "ad5311173e5212c5158e8a5b7eb5ef90937b4b4d", "parents": [ "ee79132e06852213646775fdfd591616e666bb44" ], "author": { "name": "Daniel Bristot de Oliveira", "email": "bristot@redhat.com", "time": "Mon Feb 22 14:08:22 2016 -0300" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Mar 01 12:32:25 2016 -0500" }, "message": "tools lib traceevent: Implement \u0027%\u0027 operation\n\nThe operation \u0027%\u0027 is not implemented on event-parse.c, causing an error\nwhen parsing events with \u0027%\u0027 the operation in its printk format. For\nexample,\n\n # perf record -e sched:sched_deadline_yield ~/yield-test\n Warning: [sched:sched_deadline_yield] unknown op \u0027%\u0027\n ....\n # perf script\n Warning: [sched:sched_deadline_yield] unknown op \u0027%\u0027\n test 1641 [006] 3364.109319: sched:sched_deadline_yield: \\\n [FAILED TO PARSE] now\u003d3364109314595 \\\n deadline\u003d3364139295135 runtime\u003d19975597\n\nThis patch implements the \u0027%\u0027 operation. With this patch, we see the\ncorrect output:\n\n # perf record -e sched:sched_deadline_yield ~/yield-test\n No Warning\n\n # perf script\n yield-test 4005 [001] 4623.650978: sched:sched_deadline_yield: \\\n now\u003d4623.650974050 \\\n deadline\u003d4623.680957364 remaining_runtime\u003d19979611\n\nSigned-off-by: Daniel Bristot de Oliveira \u003cbristot@redhat.com\u003e\nReviewed-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\nCc: Juri Lelli \u003cjuri.lelli@gmail.com\u003e\nCc: Peter Zijlstra \u003cpeterz@infradead.org\u003e\nCc: Thomas Gleixner \u003ctglx@linutronix.de\u003e\nCc: linux-rt-users \u003clinux-rt-users@vger.kernel.org\u003e\nLink: http://lkml.kernel.org/r/5c96a395c56cea6d3d13d949051bdece86cc26e0.1456157869.git.bristot@redhat.com\nSigned-off-by: Arnaldo Carvalho de Melo \u003cacme@redhat.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "ee79132e06852213646775fdfd591616e666bb44", "tree": "f41bc215326d07ed676c6c63f8469a372b86793a", "parents": [ "492069a1cc8a88ef9adf374d8a1e6ba1cf2d884b" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Feb 23 17:55:35 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Feb 23 17:55:35 2016 -0500" }, "message": "trace-cmd profile: Handle NULL case in handle_fgraph_exit_event()\n\nIf there\u0027s no memory available, handle_end_event() will return a NULL task,\nwhich needs to be checked before it is dereferenced.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "492069a1cc8a88ef9adf374d8a1e6ba1cf2d884b", "tree": "b0a99c1b77f25a4bc3d6638676ccc59518cc7b90", "parents": [ "a5923c5c9f812d84472f60f8c9c3e13a6702dd97" ], "author": { "name": "Steven Rostedt (Red Hat)", "email": "rostedt@goodmis.org", "time": "Tue Feb 23 17:51:07 2016 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Feb 23 17:51:07 2016 -0500" }, "message": "trace-cmd profile: Fix error checking of find_and_update_start()\n\nCommit 6fdff2ae98a0 \"trace-cmd profiler: Removed malloc_or_die()\"\nwas a bit too aggressive in having failed mallocs exit out of called\nfunctions where it returned NULL when it should not have, which caused an\nsegfault.\n\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" }, { "commit": "a5923c5c9f812d84472f60f8c9c3e13a6702dd97", "tree": "7d6aaae55543a75e0ff5c622013d26cd3d60771a", "parents": [ "91d45fa0b2e66b8c05f3d393ee5c0ea99c377dbc" ], "author": { "name": "Josef Bacik", "email": "jbacik@fb.com", "time": "Fri Dec 04 12:03:38 2015 -0500" }, "committer": { "name": "Steven Rostedt", "email": "rostedt@goodmis.org", "time": "Tue Feb 23 17:13:44 2016 -0500" }, "message": "trace-cmd: Filter out specific pids\n\nSometimes it is a good idea to filter out events from the recording processes\nand even other pids, for example an external app that links against libtracecmd\nthat watches syscalls probably doesn\u0027t want its own syscalls showing up in the\nevents. So add the ability to filter _out_ certain pids for the events that are\nenabled. The recorder threads will add their pids to this list, but they are\nonly filtered if the library consumer filters out a pid or if the trace-cmd\nrecord user tries to filter anything out.\n\nLink: http://lkml.kernel.org/r/1449248618-30556-1-git-send-email-jbacik@fb.com\n\nSigned-off-by: Josef Bacik \u003cjbacik@fb.com\u003e\nSigned-off-by: Steven Rostedt \u003crostedt@goodmis.org\u003e\n" } ], "next": "91d45fa0b2e66b8c05f3d393ee5c0ea99c377dbc" }