- 23 Jul, 2014 8 commits
-
-
Adrian Hunter authored
Add an array to struct machine to store the current tid running on each cpu. Add machine functions to get / set the tid for a cpu. This will be used to determine the tid when decoding a per-cpu Instruction Trace. Signed-off-by:
Adrian Hunter <adrian.hunter@intel.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1406035081-14301-17-git-send-email-adrian.hunter@intel.comSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Adrian Hunter authored
Tools may wish to track on which cpu a thread is running. Add 'cpu' to struct thread for that purpose. This will be used to determine the cpu when decoding a per-thread Instruction Trace. E.g: Intel PT decoding uses sched_switch events to determine which task is running on which cpu. The Intel PT data comes straight from the hardware which doesn't know about linux threads. Signed-off-by:
-
Adrian Hunter authored
Add ability to mark all dsos as hit. This is needed in the case of Instruction Tracing. It takes so long to decode an Instruction Trace that it is not worth doing just to determine which dsos are hit. A later patch takes this into use. Signed-off-by:
-
Adrian Hunter authored
Add a function to track whether a caller has seen the data status of a dso. This is needed to enable callers to report the error exactly once only per dso. Signed-off-by:
-
Adrian Hunter authored
Add 'data.status' to record whether a dso has data (i.e. an object file). This is used to avoid repeatedly creating the file name and attempting to open a file that is not present. Signed-off-by:
-
Adrian Hunter authored
When using: perf record -e branches:u -c1 usleep 1 perf script -f ip lines are displayed like: ffffffff813b23d5 => Change so that the dangling '=>' does not appear. Signed-off-by:
-
Adrian Hunter authored
Change the order of the output to put the srcline last. It puts the branch 'from address' and 'to address' on the same line, which is how it would be without the source line reference. So it makes it consistent and much easier to read. E.g. old format: 4028fc main+0x2c (/bin/ls) /build/buildd/coreutils-8.20/src/ls.c:1269 => 40d8a0 set_program_name+0x0 (/bin/ls) new format: 4028fc main+0x2c (/bin/ls) => 40d8a0 set_program_name+0x0 (/bin/ls) /build/buildd/coreutils-8.20/src/ls.c:1269 Signed-off-by: -
Adrian Hunter authored
Intel PT decoding walks the object code to reconstruct the trace. A jump label change during tracing causes decoding errors. The "Enable close-on-exec flag on perf file descriptor" patch caused there to be always a jump label change. It was found that using a per-cpu context instead of a per-thread context for the probe of the close-on-exec feature, made the problem go away. Signed-off-by:
-
- 22 Jul, 2014 1 commit
-
-
Adrian Hunter authored
Zero is a valid fd. Error comparison should check for negative fd. Signed-off-by:
Jiri Olsa <jolsa@kernel.org> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1405586590-13657-2-git-send-email-adrian.hunter@intel.comSigned-off-by:
-
- 18 Jul, 2014 4 commits
-
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf into perf/core Pull perf/core improvements and fixes from Jiri Olsa: . Update attr test with PERF_FLAG_FD_CLOEXEC flag (Jiri Olsa) . Enable close-on-exec flag on perf file descriptor (Yann Droneaud) Signed-off-by:
Ingo Molnar <mingo@kernel.org>
-
Jiri Olsa authored
We not support PERF_FLAG_FD_CLOEXEC flag, updating attr tests accordingly. Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/n/tip-2mgzn5p9u74drmdmxo5unar0@git.kernel.orgSigned-off-by:
-
Yann Droneaud authored
In commit a21b0b35 ('perf: Introduce a flag to enable close-on-exec in perf_event_open()'), flag PERF_FLAG_FD_CLOEXEC was added to perf_event_open(2) syscall to allows userspace to atomically enable close-on-exec behavor when creating the file descriptor. This patch makes perf tools use the new flag if supported by the kernel, so that the event file descriptors got automatically closed if perf tool exec a sub-command. Signed-off-by:
Yann Droneaud <ydroneaud@opteya.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Arnaldo Carvalho de Melo <acme@ghostprotocols.net> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: http://lkml.kernel.org/n/1404160127-7475-1-git-send-email-ydroneaud@opteya.comSigned-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: o Support S/390 in 'perf kvm stat' (Alexander Yarygin) Developer Stuff: o Various fixes and prep work related to supporting Intel PT (Adrian Hunter) o Introduce multiple debug variables control (Jiri Olsa) o Add callchain and additional sample information for python scripts (Joseph Schuchart) Signed-off-by:
-
- 17 Jul, 2014 9 commits
-
-
Adrian Hunter authored
It is possible to record a perf.data file on one architecture and process it on another. Consequently, TSC conversion functions need to be moved out of the arch directory. Signed-off-by:
-
Jiri Olsa authored
We don't use getopt.h interfaces. Signed-off-by:
-
Jiri Olsa authored
Adding --debug option as a way to setup debug variables. Starting with support for verbose, more will come. It's possible to use it now with report command: $ perf --debug verbose ... $ perf --debug verbose=2 ... I'll need this support to add separated debug variable for ordered events change in order to separate debug output out of standard verbose stream. Signed-off-by:
-
Jiri Olsa authored
This way we can easily reuse current debug functions for another debug variables other than verbose. Signed-off-by:
-
Jiri Olsa authored
Moving pr_* debug macros to have it with in same object as debug variables, becase we will change them to use verbose variable in next patch. Signed-off-by:
-
Jiri Olsa authored
And use verbose as an global object in following functions: __map_groups__fprintf_maps __map_groups__fprintf_removed_maps map_groups__fprintf_maps map_groups__fprintf Also making map_groups__fprintf_maps static. Signed-off-by:
-
Adrian Hunter authored
__machine__findnew_thread() creates a 'struct thread' but does not free it on the error path. Fix it. Signed-off-by:
-
Adrian Hunter authored
It needs to be possible to call thread__delete() on a thread with no map groups. This is needed for a subsequent patch which deletes a thread on the error path before map groups have been attached. Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/r/1405495184-20441-2-git-send-email-adrian.hunter@intel.comSigned-off-by:
-
Adrian Hunter authored
Events like sched_switch do not provide a pid (tgid) which can result in threads with an unknown pid. If the pid is later discovered, join the map groups. Note the thread's map groups should be empty because they are populated by MMAP events which do provide the pid and tid. Signed-off-by:
-
- 16 Jul, 2014 18 commits
-
-
Adrian Hunter authored
Add an option to cause a selected event to be enabled immediately when configured by perf_evsel__config(). This is needed when using the sched_switch tracepoint to follow object code execution. By having sched_switch enabled immediately the first sched_switch event precedes the start of other tracing. Signed-off-by:
-
Adrian Hunter authored
Add an option to prevent additional samples being added to a selected event by perf_evsel__config(). This is needed when using the sched_switch tracepoint to follow object code execution. Since sched_switch will be used only for switch information, additional sampling is wasteful. Signed-off-by:
-
Adrian Hunter authored
Add an option macro that is the same as OPT_CALLBACK except that the argument is optional and it is possible to associate additional data with it. Signed-off-by:
-
Adrian Hunter authored
Add a feature test for __sync_val_compare_and_swap() and __sync_bool_compare_and_swap() Signed-off-by:
-
Adrian Hunter authored
In preparation for adding more mmap parameters, pass existing parameters in a struct. Signed-off-by:
-
Adrian Hunter authored
Flag if the event stream is a file that has been mmapped in one go. This is useful, for example, if a tool needs to keep an event for later reference. If the new flag is set, a pointer to the event can be retained, otherwise the event must be copied. Signed-off-by:
-
Adrian Hunter authored
Expose dso__first_symbol() and dso__next_symbol() to make it possible to iterate over a dso's symbols. Signed-off-by:
-
Adrian Hunter authored
Record kallsyms binary type so that tools will not attempt to read binary data from it. Signed-off-by:
-
Adrian Hunter authored
Add a flag to 'struct dso' to record if the dso is 64-bit or not. Update the flag when reading the ELF. This is needed for instruction decoding. For example, x86 instruction decoding depends on whether or not the 64-bit instruction set is used. Signed-off-by:
-
Adrian Hunter authored
Currently a copy of kcore is not made if there is one already with the same modules at the same addresses. Change this to make a copy anyway if the force (-f) option is also used. Signed-off-by:
-
Adrian Hunter authored
hist_entry__append_callchain() must check if the sample has a callcahin or it will append the callchain from a previous sample. Signed-off-by:
-
Adrian Hunter authored
Build Ids won't be injected unless the build id feature flag is set. Signed-off-by:
-
Adrian Hunter authored
Symbols of type STT_GNU_IFUNC are functions so accept them as such. Signed-off-by:
-
Adrian Hunter authored
thread__find_addr_map() falls back to trying the kernel maps if the address is negative and is not found in userspace maps. As commented in the code, the kernel maps must be "loaded" before use. This patch ensures that happens under the fallback condition also. Signed-off-by:
-
Adrian Hunter authored
The comm_exec flag on the attribute can later be found in the perf.data file allowing a tool to know in advance if the captured data has the flag. Signed-off-by:
-
Adrian Hunter authored
Signed-off-by:
-
Adrian Hunter authored
The value used for unknown pids cannot be zero because that is used by the "idle" task. Use -1 instead. Also handle the unknown pid case when creating map groups. Note that, threads with an unknown pid should not occur because fork (or synthesized) events precede the thread's existence. Signed-off-by:
-
Joseph Schuchart authored
To python scripts, including pid, tid, and cpu for which the event was recorded. At the moment, the pointer to the sample struct is passed to scripts, which seems to be of little use. The patch puts this information in dictionaries for easy access by Python scripts. Signed-off-by:
Joseph Schuchart <joseph.schuchart@tu-dresden.de> Acked-by:
Thomas Ilsche <thomas.ilsche@tu-dresden.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Namhyung Kim <namhyung@gmail.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: Thomas Ilsche <thomas.ilsche@tu-dresden.de> Link: http://lkml.kernel.org/r/53BE7E20.8080500@tu-dresden.deSigned-off-by:
-
