- 17 Aug, 2015 11 commits
-
-
Adrian Hunter authored
To record an AUX area, the weak function auxtrace_record__init() must be implemented. Equally to decode an AUX area, the AUX area tracing type must be added to the perf_event__process_auxtrace_info() function. This patch makes those two changes plus hooks up default config for the intel_pt PMU. Also some brief documentation is provided for using the tools with intel_pt. Commiter note: E.g: [root@perf4 ~]# dmesg 451 [0.405807] Performance Events: PEBS fmt2+, 16-deep LBR, Broadwell events, full-width counters, Intel PMU driver. [root@perf4 ~]# perf --version perf version 4.1.g53874a [root@perf4 ~]# perf record -e intel_pt//u -a sleep 10 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.383 MB perf.data ] [root@perf4 ~]# perf evlist intel_pt//u sched:sched_switch dummy:u [root@perf4 ~]# perf report --stdio # To display the perf.data header info, please use --header/--header-only options. # # # Total Lost Samples: 0 # # Samples: 0 of event 'intel_pt//u' # Event count (approx.): 0 # # Overhead Command Shared Object Symbol # ........ ....... ............. ...... # # Samples: 393 of event 'sched:sched_switch' # Event count (approx.): 393 # # Overhead Command Shared Object Symbol # ........ .............. ................ .............. 49.62% swapper [kernel.vmlinux] [k] __schedule 10.69% rcu_sched [kernel.vmlinux] [k] __schedule 6.62% rcuos/0 [kernel.vmlinux] [k] __schedule 5.60% kworker/0:1 [kernel.vmlinux] [k] __schedule 3.56% rcuos/3 [kernel.vmlinux] [k] __schedule 3.05% kworker/u384:2 [kernel.vmlinux] [k] __schedule 2.54% kworker/2:0 [kernel.vmlinux] [k] __schedule 2.54% tuned [kernel.vmlinux] [k] __schedule <SNIP> # Samples: 0 of event 'dummy:u' # Event count (approx.): 0 # # Overhead Command Shared Object Symbol # ........ ....... ............. ...... # Samples: 28 of event 'instructions:u' # Event count (approx.): 5030172 # # Overhead Command Shared Object Symbol # ........ .......... ................... ................................ # 21.43% tuned libpython2.7.so.1.0 [.] PyEval_EvalFrameEx | ---PyEval_EvalFrameEx | |--83.33%-- PyEval_EvalCodeEx | PyEval_EvalFrameEx | | | |--60.00%-- PyEval_EvalCodeEx | | PyEval_EvalFrameEx | | PyEval_EvalFrameEx | | | --40.00%-- PyEval_EvalFrameEx | --16.67%-- PyEval_EvalFrameEx PyEval_EvalCodeEx PyEval_EvalFrameEx PyEval_EvalCodeEx PyEval_EvalFrameEx PyEval_EvalFrameEx 14.29% tuned libpython2.7.so.1.0 [.] _PyType_Lookup | ---_PyType_Lookup _PyObject_GenericGetAttrWithDict PyEval_EvalFrameEx PyEval_EvalCodeEx PyEval_EvalFrameEx PyEval_EvalCodeEx PyEval_EvalFrameEx | |--75.00%-- PyEval_EvalFrameEx | --25.00%-- PyEval_EvalCodeEx PyEval_EvalFrameEx PyEval_EvalFrameEx 3.57% irqbalance irqbalance [.] 0x0000000000004038 | ---0x4038 0x4761 0x4761 0x4761 0x49f1 0x2295 3.57% irqbalance libc-2.17.so [.] __GI_____strtoull_l_internal | ---__GI_____strtoull_l_internal 0x6f49 0x229a 3.57% irqbalance libc-2.17.so [.] __strchrnul | ---__strchrnul vfprintf __vsprintf_chk __sprintf_chk 0x2724 0x4038 0x2331 3.57% irqbalance libc-2.17.so [.] __strstr_sse42 | ---__strstr_sse42 0x71e0 0x229f # And now to some userspace ftrace on uninstrumented binaries 8-) : # Hand edited to make it a bit more compact, replacing /home/acme/bin/perf # with /bin/perf: [root@perf4 ~]# perf script perf 8921 [3] 7.310889: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310889: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310889: 1 branches:u: 481694 perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310889: 1 branches:u: 481630 perf_evlist__enable (/bin/perf) => 4816d8 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310889: 1 branches:u: 4816de perf_evlist__enable (/bin/perf) => 48164f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310889: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310889: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.310889: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310889: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.310890: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310890: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310890: 1 branches:u: 481694 perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310890: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310890: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.310890: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310890: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.310893: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310893: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310893: 1 branches:u: 4816a8 perf_evlist__enable (/bin/perf) => 4815f8 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310893: 1 branches:u: 4815fe perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310893: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310893: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.310893: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310893: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.310956: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310956: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310956: 1 branches:u: 481694 perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310956: 1 branches:u: 481630 perf_evlist__enable (/bin/perf) => 4816d8 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310956: 1 branches:u: 4816de perf_evlist__enable (/bin/perf) => 48164f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310956: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310956: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.310956: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310956: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.310961: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310961: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310961: 1 branches:u: 481694 perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310961: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310961: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.310961: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310961: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.310968: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310968: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310968: 1 branches:u: 4816a8 perf_evlist__enable (/bin/perf) => 4815f8 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310968: 1 branches:u: 4815fe perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310968: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.310968: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.310968: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.310968: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.311040: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.311040: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311040: 1 branches:u: 481694 perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311040: 1 branches:u: 481630 perf_evlist__enable (/bin/perf) => 4816d8 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311040: 1 branches:u: 4816de perf_evlist__enable (/bin/perf) => 48164f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311040: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311040: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.311040: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.311040: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.311046: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.311046: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311046: 1 branches:u: 481694 perf_evlist__enable (/bin/perf) => 481614 perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311046: 1 branches:u: 481652 perf_evlist__enable (/bin/perf) => 48165f perf_evlist__enable (/bin/perf) perf 8921 [3] 7.311046: 1 branches:u: 481684 perf_evlist__enable (/bin/perf) => 41d250 ioctl@plt (/bin/perf) perf 8921 [3] 7.311046: 1 branches:u: 41d250 ioctl@plt (/bin/perf) => 7fcecadbf250 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.311046: 1 branches:u: 7fcecadbf255 __GI___ioctl (/usr/lib64/libc-2.17.so) => 0 [unknown] ([unknown]) perf 8921 [3] 7.311050: 1 branches:u: 0 [unknown] ([unknown]) => 7fcecadbf257 __GI___ioctl (/usr/lib64/libc-2.17.so) perf 8921 [3] 7.311050: 1 branches:u: 7fcecadbf25f __GI___ioctl (/usr/lib64/libc-2.17.so) => 481689 perf_evlist__enable (/bin/perf) : Signed-off-by:Adrian Hunter <adrian.hunter@intel.com> Tested-by:
Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1437150840-31811-8-git-send-email-adrian.hunter@intel.comSigned-off-by:
-
Adrian Hunter authored
Add support for Intel Processor Trace. Intel PT support fits within the new auxtrace infrastructure. Recording is supporting by identifying the Intel PT PMU, parsing options and setting up events. Decoding is supported by queuing up trace data by cpu or thread and then decoding synchronously delivering synthesized event samples into the session processing for tools to consume. Signed-off-by:
-
Adrian Hunter authored
Add support for decoding an Intel Processor Trace. Intel PT trace data must be 'decoded' which involves walking the object code and matching the trace data packets. The decoder requests a buffer of binary data via a get_trace() call-back, which it decodes using instruction information which it gets via another call-back walk_insn(). Signed-off-by:
-
Adrian Hunter authored
Add a facility to log Intel Processor Trace decoding. The log is intended for debugging purposes only. The log file name is "intel_pt.log" and is opened in the current directory. The log contains a record of all packets and instructions decoded and can get very large (10 MB would be a small one). Signed-off-by:
-
Adrian Hunter authored
Add support for decoding instructions for Intel Processor Trace. The kernel x86 instruction decoder is copied for this. This essentially provides intel_pt_get_insn() which takes a binary buffer, uses the kernel's x86 instruction decoder to get details of the instruction and then categorizes it for consumption by an Intel PT decoder. Signed-off-by:
-
Adrian Hunter authored
Add support for decoding Intel Processor Trace packets. This essentially provides intel_pt_get_packet() which takes a buffer of binary data and returns the decoded packet. Signed-off-by:
-
Adrian Hunter authored
Add the Intel Processor Trace type constant PERF_AUXTRACE_INTEL_PT. Signed-off-by:
Jiri Olsa <jolsa@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Link: http://lkml.kernel.org/r/1437150840-31811-2-git-send-email-adrian.hunter@intel.comSigned-off-by:
-
Adrian Hunter authored
Add a helper function to probe whether cpu-wide tracing is possible. Signed-off-by:
-
Adrian Hunter authored
Older kernels attempt to prelink vdso to its virtual address. To permit annotation using objdump, the map__rip_2objdump() calculation must result in that same address which we can infer from the start and offset of the text section. Signed-off-by:
-
Adrian Hunter authored
Fix the following 32-bit compilation errors: util/annotate.c: In function ‘addr_map_symbol__account_cycles’: util/annotate.c:643:3: error: format ‘%lx’ expects argument of type ‘long unsigned int’, but argument 4 has type ‘u64’ [-Werror=format=] pr_debug2("BB with bad start: addr %lx start %lx sym %lx saddr %lx\n", ^ util/annotate.c:643:3: error: format ‘%lx’ expects argument of type ‘long unsigned int’, but argument 5 has type ‘u64’ [-Werror=format=] util/annotate.c:643:3: error: format ‘%lx’ expects argument of type ‘long unsigned int’, but argument 6 has type ‘u64’ [-Werror=format=] These were introduced by the patch: "perf report: Add infrastructure for a cycles histogram" Also change the 'saddr' variable from 'unsigned long' to 'u64' noting that theoretically we could be processing data captured on a 64-bit machine but processing it on a 32-bit machine. Signed-off-by: -
Jiri Olsa authored
Milian Wolff reported non functional DWARF unwind under perf script. The reason is that perf script does not properly configure callchain_param.record_mode, which is needed by unwind code. Stealing the code from report and leaving the place for more initialization code in a hope we could merge it with report__setup_sample_type one day. Reported-by:
Milian Wolff <mail@milianw.de> Signed-off-by:
Milian Wolff <milian.wolff@kdab.com> Cc: David Ahern <dsahern@gmail.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/20150813071724.GA21322@krava.brq.redhat.comSigned-off-by:
-
- 14 Aug, 2015 1 commit
-
-
Arnaldo Carvalho de Melo authored
We were storing the vfs_getname payload (i.e. ptr->string) into the trace wide storage area (struct trace), so that we could use the last payload when setting up the fd->pathname per thread tables, oops, not a good idea for multi cpu tracing sessions... Fix it by moving it to the per thread area (struct thread_trace). Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-3j05ttqyaem7kh7oubvr1keo@git.kernel.orgSigned-off-by:
-
- 13 Aug, 2015 3 commits
-
-
Masami Hiramatsu authored
The commit 75186a9b (perf probe: Fix to show lines of sys_ functions correctly) introduced a bug by a missed brace around if block. This fixes to add it. Signed-off-by:
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Fixes: 75186a9b ("perf probe: Fix to show lines of sys_ functions correctly") Link: http://lkml.kernel.org/r/20150812215541.9088.62425.stgit@localhost.localdomainSigned-off-by:
-
Andi Kleen authored
The Fedora 22 version of libdw requires a couple of extra libraries to link. With a dynamic link the dependencies are pulled in automatically, but this doesn't work for static linking. Add the needed libraries explicitely to the feature probe and the Makefile. v2: Explicitly check for static linking and only add the dependencies when -static is set. This is to avoid regressions on Arnaldo's system. Signed-off-by:Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Link: http://lkml.kernel.org/r/1439419717-20601-1-git-send-email-andi@firstfloor.orgSigned-off-by:
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo: User visible changes: - Allow selecting the type of callchains per event, including disabling callchains in all but one entry in an event list, to save space, and also to ask for the callchains collected in one event to be used in other events. (Kan Liang) - Beautify more syscall arguments in 'perf trace': (Arnaldo Carvalho de Melo) - A bunch more translate file/pathnames from pointers to strings. - Convert numbers to strings for the 'keyctl' syscall 'option' arg. - Add missing 'clockid' entries. - Fix 'perf probe -L sys_*' as it was not showing all the source code for syscall functions in the kernel. (Masami Hiramatsu) - Make ESC unzoom as well in the hists browser, i.e. in 'report' and 'top', as we're considering repurposing the right and left arrow keys to use in horizontal scrolling, i.e. leave just ESC to be used for what <- works now, and ENTER for what -> does (they are already aliases for ages). (Arnaldo Carvalho de Melo) Infrastructure fixes: - Check for SRCLINE_UNKNOWN case in "srcfile" processing (Andi Kleen) - Wrap the slsmg_{printf,write_nstring} slang functions behind ui_browser, so that we can make the ui_browser based browsers (annotate, menus, hists, etc) UI library agnostic and usable with multiple backends (slang now, GTK+ and others in the future, maybe) (Arnaldo Carvalho de Melo) Signed-off-by:
Ingo Molnar <mingo@kernel.org>
-
- 12 Aug, 2015 22 commits
-
-
Kan Liang authored
Add tests in tests/parse-events.c to check call-graph and time option. Signed-off-by:
Kan Liang <kan.liang@intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: http://lkml.kernel.org/r/1439289050-40510-4-git-send-email-kan.liang@intel.comSigned-off-by:
-
Kan Liang authored
Introduce --show-ref-call-graph for perf report to print reference callgraph for no callgraph event. Here is an example. perf report --show-ref-call-graph --stdio # To display the perf.data header info, please use --header/--header-only options. # # # Total Lost Samples: 0 # # Samples: 5 of event 'cpu/cpu-cycles,call-graph=fp/' # Event count (approx.): 144985 # # Children Self Command Shared Object Symbol # ........ ........ ....... ................ ........................................ # 72.30% 0.00% sleep [kernel.vmlinux] [k] entry_SYSCALL_64_fastpath | ---entry_SYSCALL_64_fastpath | |--22.62%-- __GI___libc_nanosleep --77.38%-- [...] ...... # Samples: 6 of event 'cpu/instructions,call-graph=no/', show reference callgraph # Event count (approx.): 172780 # # Children Self Command Shared Object Symbol # ........ ........ ....... ................ ........................................ # 73.16% 0.00% sleep [kernel.vmlinux] [k] entry_SYSCALL_64_fastpath | ---entry_SYSCALL_64_fastpath | |--31.44%-- __GI___libc_nanosleep --68.56%-- [...] Signed-off-by: -
Kan Liang authored
This patch introduce "call-graph=no" to disable per-event callgraph. Here is an example. perf record -e 'cpu/cpu-cycles,call-graph=fp/,cpu/instructions,call-graph=no/' sleep 1 perf report --stdio # To display the perf.data header info, please use --header/--header-only options. # # # Total Lost Samples: 0 # # Samples: 6 of event 'cpu/cpu-cycles,call-graph=fp/' # Event count (approx.): 774218 # # Children Self Command Shared Object Symbol # ........ ........ ....... ................ ........................................ # 61.94% 0.00% sleep [kernel.vmlinux] [k] entry_SYSCALL_64_fastpath | ---entry_SYSCALL_64_fastpath | |--97.30%-- __brk | --2.70%-- mmap64 _dl_check_map_versions _dl_check_all_versions 61.94% 0.00% sleep [kernel.vmlinux] [k] perf_event_mmap | ---perf_event_mmap | |--97.30%-- do_brk | sys_brk | entry_SYSCALL_64_fastpath | __brk | --2.70%-- mmap_region do_mmap_pgoff vm_mmap_pgoff sys_mmap_pgoff sys_mmap entry_SYSCALL_64_fastpath mmap64 _dl_check_map_versions _dl_check_all_versions ...... # Samples: 6 of event 'cpu/instructions,call-graph=no/' # Event count (approx.): 359692 # # Children Self Command Shared Object Symbol # ........ ........ ....... ................ ................................. # 89.03% 0.00% sleep [unknown] [.] 0xffff6598ffff6598 89.03% 0.00% sleep ld-2.17.so [.] _dl_resolve_conflicts 89.03% 0.00% sleep [kernel.vmlinux] [k] page_fault Signed-off-by: -
Kan Liang authored
This patchkit adds the ability to set callgraph mode (fp, dwarf, lbr) per event. This in term can reduce sampling overhead and the size of the perf.data. Here is an example. perf record -e 'cpu/cpu-cycles,period=1000,call-graph=fp,time=1/,cpu/instructions,call-graph=lbr/' sleep 1 perf evlist -v cpu/cpu-cycles,period=1000,call-graph=fp,time=1/: type: 4, size: 112, config: 0x3c, { sample_period, sample_freq }: 1000, sample_type: IP|TID|TIME|CALLCHAIN|PERIOD|IDENTIFIER, read_format: ID, disabled: 1, inherit: 1, mmap: 1, comm: 1, enable_on_exec: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1 cpu/instructions,call-graph=lbr/: type: 4, size: 112, config: 0xc0, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CALLCHAIN|PERIOD|BRANCH_STACK|IDENTIFIER, read_format: ID, disabled: 1, inherit: 1, freq: 1, enable_on_exec: 1, sample_id_all: 1, exclude_guest: 1 Signed-off-by: -
Masami Hiramatsu authored
"perf probe --lines sys_poll" shows only the first line of sys_poll, because the SYSCALL_DEFINE macro: ---- SYSCALL_DEFINE*(foo,...) { body; } ---- is expanded as below (on debuginfo) ---- static inline int SYSC_foo(...) { body; } int SyS_foo(...) <- is an alias of sys_foo. { return SYSC_foo(...); } ---- So, "perf probe --lines sys_foo" decodes SyS_foo function and it also skips inlined functions(SYSC_foo) inside the target function because those functions are usually defined somewhere else. To fix this issue, this fix checks whether the inlined function is defined at the same point of the target function, and if so, it doesn't skip the inline function. Reported-by:Arnaldo Carvalho de Melo <acme@kernel.org> Signed-off-by:
-
Arnaldo Carvalho de Melo authored
In addition to <-, that may be repurposed for horizontal scrolling. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-w3rctelxr4yxrjufx7z3fclb@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
To remove direct access to libslang functions, with the immediate goal of implementing horizontal scrolling at the ui_browser level, but also because we may at some point want to implement ui_browser with other UIs in addition to the current libslang implementation. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-w0niblabqrkecs4o0eogfy6c@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
To remove direct access to libslang functions, with the immediate goal of implementing horizontal scrolling at the ui_browser level, but also because we may at some point want to implement ui_browser with other UIs in addition to the current libslang implementation. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-437ineavoejzou727mr9bxpi@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
8.697 (0.103 ms): pool/2343 keyctl(option: GET_PERSISTENT, arg2: 1000, arg3: 4294967294, arg4: 140703061514067, arg5: 140703692383680) = 1023192809 8.763 (0.049 ms): pool/2343 keyctl(option: SEARCH, arg2: 1023192809, arg3: 140703745767772, arg4: 140703745767832, arg5: 4294967294) = 140224497 8.789 (0.016 ms): pool/2343 keyctl(option: SEARCH, arg2: 140224497, arg3: 140703745767814, arg4: 140703745767900) = 512300257 8.807 (0.011 ms): pool/2343 keyctl(option: READ, arg2: 512300257 ) = 13 8.822 (0.008 ms): pool/2343 keyctl(option: READ, arg2: 512300257, arg3: 140703061514000, arg4: 13 ) = 13 8.837 (0.007 ms): pool/2343 keyctl(option: READ, arg2: 140224497 ) = 4 8.852 (0.009 ms): pool/2343 keyctl(option: READ, arg2: 140224497, arg3: 140703061514000, arg4: 4 ) = 4 8.869 (0.010 ms): pool/2343 keyctl(option: SEARCH, arg2: 140224497, arg3: 140703745767772, arg4: 140703061514032) = -1 ENOKEY Required key not available 8.892 (0.017 ms): pool/2343 keyctl(option: DESCRIBE, arg2: 512300257 ) = 43 8.910 (0.012 ms): pool/2343 keyctl(option: DESCRIBE, arg2: 512300257, arg3: 140703061544384, arg4: 43) = 43 Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-013ab219irsxngyumrf5gp8s@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
But we really should have something like 'strace -yy' here... Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-eyrt1ypfq68u4ljagyk2nj1i@git.kernel.orgSigned-off-by:
-
Andi Kleen authored
Handle the SRCLINE_UNKNOWN case correctly when processing "srcfile". Commiter note: We can't just free it, as it was't allocated via malloc, its a guard variable. Reported-by:
Namhyung Kim <namhyung@kernel.org> Signed-off-by:
-
Arnaldo Carvalho de Melo authored
We were missing: CLOCK_BOOTTIME, CLOCK_REALTIME_ALARM, CLOCK_BOOTTIME_ALARM, CLOCK_SGI_CYCLE and CLOCK_TAI. Add them. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Milian Wolff <milian.wolff@kdab.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-d67rwqtwm9jyenwes98kr0cr@git.kernel.orgSigned-off-by:
-
Arnaldo Carvalho de Melo authored
This time using 'trinity' to test these: fchmodat, futimesat, llistxattr, lremovexattr, lstat, mknodat, mq_unlink, stat and vmsplice. Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Borislav Petkov <bp@suse.de> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Milian Wolff <milian.wolff@kdab.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-a1uqu249nwwh0ixrhm80k4a4@git.kernel.orgSigned-off-by:
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo: User visible changes: - Introduce 'srcfile' sort key: (Andi Kleen) # perf record -F 10000 usleep 1 # perf report --stdio --dsos '[kernel.vmlinux]' -s srcfile <SNIP> # Overhead Source File 26.49% copy_page_64.S 5.49% signal.c 0.51% msr.h # It can be combined with other fields, for instance, experiment with '-s srcfile,symbol'. There are some oddities in some distros and with some specific DSOs, being investigated, so your mileage may vary. - Update the column width for the "srcline" sort key (Arnaldo Carvalho de Melo) - Support per-event 'freq' term: (Namhyung Kim) $ perf record -e 'cpu/instructions,freq=1234/',cycles -c 1000 sleep 1 $ perf evlist -F cpu/instructions,freq=1234/: sample_freq=1234 cycles: sample_period=1000 $ Infrastructure changes: - Move perf_counts struct and functions into separate object (Jiri Olsa) - Unset perf_event_attr::freq when period term is set (Jiri Olsa) - Move callchain option parsing code to util.c (Kan Liang) Signed-off-by:
-
Takao Indoh authored
This patch just cleans up some files of Intel Processor Trace, does not change its behavior. This patch removes unused definitions and replaces a constant value with a macro. Signed-off-by:
Takao Indoh <indou.takao@jp.fujitsu.com> Signed-off-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Cc: Alexander Shishkin<alexander.shishkin@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: H.Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1438681015-5124-1-git-send-email-indou.takao@jp.fujitsu.comSigned-off-by:
-
Alexander Shishkin authored
A question [1] was raised about the use of page::private in AUX buffer allocations, so let's add a clarification about its intended use. The private field and flag are used by perf's rb_alloc_aux() path to tell the pmu driver the size of each high-order allocation, so that the driver can program those appropriately into its hardware. This only matters for PMUs that don't support hardware scatter tables. Otherwise, every page in the buffer is just a page. This patch adds a comment about the private field to the AUX buffer allocation path. [1] http://marc.info/?l=linux-kernel&m=143803696607968Reported-by:
Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by:
Alexander Shishkin <alexander.shishkin@linux.intel.com> Signed-off-by:
-
Peter Zijlstra authored
Currently we only update the sysfs event files per available MSR, we didn't actually disallow creating unlisted events. Rework things such that the dectection, sysfs listing and event creation are better coordinated. Sadly it appears it's impossible to probe R/O MSRs under virt. This means we have to do the full model table to avoid listing all MSRs all the time. Tested-by:
Andy Lutomirski <luto@amacapital.net> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Ingo Molnar authored
Signed-off-by: -
Matt Fleming authored
Tony reports that booting his 144-cpu machine with maxcpus=10 triggers the following WARN_ON(): [ 21.045727] WARNING: CPU: 8 PID: 647 at arch/x86/kernel/cpu/perf_event_intel_cqm.c:1267 intel_cqm_cpu_prepare+0x75/0x90() [ 21.045744] CPU: 8 PID: 647 Comm: systemd-udevd Not tainted 4.2.0-rc4 #1 [ 21.045745] Hardware name: Intel Corporation BRICKLAND/BRICKLAND, BIOS BRHSXSD1.86B.0066.R00.1506021730 06/02/2015 [ 21.045747] 0000000000000000 0000000082771b09 ffff880856333ba8 ffffffff81669b67 [ 21.045748] 0000000000000000 0000000000000000 ffff880856333be8 ffffffff8107b02a [ 21.045750] ffff88085b789800 ffff88085f68a020 ffffffff819e2470 000000000000000a [ 21.045750] Call Trace: [ 21.045757] [<ffffffff81669b67>] dump_stack+0x45/0x57 [ 21.045759] [<ffffffff8107b02a>] warn_slowpath_common+0x8a/0xc0 [ 21.045761] [<ffffffff8107b15a>] warn_slowpath_null+0x1a/0x20 [ 21.045762] [<ffffffff81036725>] intel_cqm_cpu_prepare+0x75/0x90 [ 21.045764] [<ffffffff81036872>] intel_cqm_cpu_notifier+0x42/0x160 [ 21.045767] [<ffffffff8109a33d>] notifier_call_chain+0x4d/0x80 [ 21.045769] [<ffffffff8109a44e>] __raw_notifier_call_chain+0xe/0x10 [ 21.045770] [<ffffffff8107b538>] _cpu_up+0xe8/0x190 [ 21.045771] [<ffffffff8107b65a>] cpu_up+0x7a/0xa0 [ 21.045774] [<ffffffff8165e920>] cpu_subsys_online+0x40/0x90 [ 21.045777] [<ffffffff81433b37>] device_online+0x67/0x90 [ 21.045778] [<ffffffff81433bea>] online_store+0x8a/0xa0 [ 21.045782] [<ffffffff81430e78>] dev_attr_store+0x18/0x30 [ 21.045785] [<ffffffff8126b6ba>] sysfs_kf_write+0x3a/0x50 [ 21.045786] [<ffffffff8126ad40>] kernfs_fop_write+0x120/0x170 [ 21.045789] [<ffffffff811f0b77>] __vfs_write+0x37/0x100 [ 21.045791] [<ffffffff811f38b8>] ? __sb_start_write+0x58/0x110 [ 21.045795] [<ffffffff81296d2d>] ? security_file_permission+0x3d/0xc0 [ 21.045796] [<ffffffff811f1279>] vfs_write+0xa9/0x190 [ 21.045797] [<ffffffff811f2075>] SyS_write+0x55/0xc0 [ 21.045800] [<ffffffff81067300>] ? do_page_fault+0x30/0x80 [ 21.045804] [<ffffffff816709ae>] entry_SYSCALL_64_fastpath+0x12/0x71 [ 21.045805] ---[ end trace fe228b836d8af405 ]--- The root cause is that CPU_UP_PREPARE is completely the wrong notifier action from which to access cpu_data(), because smp_store_cpu_info() won't have been executed by the target CPU at that point, which in turn means that ->x86_cache_max_rmid and ->x86_cache_occ_scale haven't been filled out. Instead let's invoke our handler from CPU_STARTING and rename it appropriately. Reported-by:
Tony Luck <tony.luck@intel.com> Signed-off-by:
Matt Fleming <matt.fleming@intel.com> Signed-off-by:
-
Peter Zijlstra authored
We fail to free the shared_regs allocation if the constraint_list allocation fails. Cure this and be more consistent in NULL-ing the pointers after free. Signed-off-by:
-
Peter Zijlstra authored
I ran the perf fuzzer, which triggered some WARN()s which are due to trying to stop/restart an event on the wrong CPU. Use the normal IPI pattern to ensure we run the code on the correct CPU. Signed-off-by:
-
Ben Hutchings authored
If rb->aux_refcount is decremented to zero before rb->refcount, __rb_free_aux() may be called twice resulting in a double free of rb->aux_pages. Fix this by adding a check to __rb_free_aux(). Signed-off-by:
Ben Hutchings <ben@decadent.org.uk> Signed-off-by:
-
- 10 Aug, 2015 3 commits
-
-
Namhyung Kim authored
Currently perf evlist -F shows the number as if it's always sampling frequency. But we now support per-event freq/period settings. So it'd better to show more detailed info whether it's freq or period. $ perf record -e 'cpu/config=1/,cpu/config=2,period=300000/' sleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.017 MB perf.data ] $ perf evlist -F cpu/config=1/: sample_freq=4000 cpu/config=2,period=300000/: sample_period=300000 Signed-off-by:
-
Namhyung Kim authored
Now perf can set per-event value of time and (sampling) period. But I guess most users like me just want to set frequency rather than period. So add the 'freq' term in the event parser. Signed-off-by:
-
Andi Kleen authored
In some cases it's useful to characterize samples by file. This is useful to get a higher level categorization, for example to map cost to subsystems. Add a srcfile sort key to perf report. It builds on top of the existing srcline support. Commiter notes: E.g.: # perf record -F 10000 usleep 1 [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.016 MB perf.data (13 samples) ] [root@zoo ~]# perf report -s srcfile --stdio # Total Lost Samples: 0 # # Samples: 13 of event 'cycles' # Event count (approx.): 869878 # # Overhead Source File # ........ ........... 60.99% . 20.62% paravirt.h 14.23% rmap.c 4.04% signal.c 0.11% msr.h # The first line is collecting all the files for which srcfiles couldn't somehow get resolved to: # perf report -s srcfile,dso --stdio # Total Lost Samples: 0 # # Samples: 13 of event 'cycles' # Event count (approx.): 869878 # # Overhead Source File Shared Object # ........ ........... ................ 40.97% . ld-2.20.so 20.62% paravirt.h [kernel.vmlinux] 20.02% . libc-2.20.so 14.23% rmap.c [kernel.vmlinux] 4.04% signal.c [kernel.vmlinux] 0.11% msr.h [kernel.vmlinux] # XXX: Investigate why that is not resolving on Fedora 21, Andi says he hasn't seen this on Fedora 22. Signed-off-by:
-
