- 17 Jul, 2013 6 commits
-
-
Jiri Olsa authored
Removing attr_offset from perf_header as it's possible to use it as a local variable. Signed-off-by:
Jiri Olsa <jolsa@redhat.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Ingo Molnar <mingo@elte.hu> 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/r/1374083403-14591-4-git-send-email-jolsa@redhat.comSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Jiri Olsa authored
Removing data_offset seek as it's not needed, because data are not read by syscall but mmaped instead. Signed-off-by:
-
Jiri Olsa authored
Using session->fd instead of passing fd as argument because it's always session->fd that's passed as fd argument. Signed-off-by:
-
Jiri Olsa authored
The prev_state is defined as long which is 4 bytes long on 32-bit x86. Changing the check against sizeof(long). Signed-off-by:
-
Namhyung Kim authored
If an user gives both of --symfs and --vmlinux option, the vmlinux will be searched under the symfs directory. This is somewhat confusing since vmlinux often lives in kernel build directory or somewhere other than user space binaries. So it'd be better not adding symfs prefix for a vmlinux if it has an absolute pathname. Reported-by:
Kwanghyun Yoo <ykh815.yoo@lge.com> Signed-off-by:
Namhyung Kim <namhyung@kernel.org> Acked-by:
David Ahern <dsahern@gmail.com> Cc: David Ahern <dsahern@gmail.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1374048495-3643-1-git-send-email-namhyung@kernel.orgSigned-off-by:
-
Robert Richter authored
Boris just raised another variant of building perf tools which is broken: $ make tools/perf ... LINK /home/robert/cx/linux/tools/perf/perf gcc: error: ../linux/tools/lib/lk/liblk.a: No such file or directory The variant wasn't considered by: 107de372 perf tools: Fix build errors with O and DESTDIR make vars set There are other variant of building perf too: $ make -C tools perf $ make -C tools/perf Plus variants with O= and DESTDIR set. This patch fixes the above and was tested with the following: $ make O=... DESTDIR=... tools/perf $ make O=... DESTDIR=... -C tools/ perf $ make O=... DESTDIR=... -C tools/perf $ make tools/perf $ make -C tools/ perf $ make -C tools/perf Reported-by:Borislav Petkov <bp@alien8.de> Signed-off-by:
Robert Richter <robert.richter@linaro.org> Signed-off-by:
Robert Richter <rric@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-tip-commits@vger.kernel.org Link: http://lkml.kernel.org/r/20130716145036.GH8731@rric.localhostSigned-off-by:
-
- 15 Jul, 2013 5 commits
-
-
Jiri Olsa authored
Removing event types framework completely. The only remainder (apart from few comments) is following enum: enum perf_user_event_type { ... PERF_RECORD_HEADER_EVENT_TYPE = 65, /* deprecated */ ... } It's kept as deprecated, resulting in error when processed in perf_session__process_user_event function. Signed-off-by: -
Jiri Olsa authored
Removing event types data pushing from record command. It's no longer needed, because this data is ignored. Signed-off-by:
-
Jiri Olsa authored
Removing event types data storing/reading to/from perf data file as it's no longer needed. The only user of this data 'perf timechart' was switched to use tracepoints handler callbacks. The event_types offset and size stay in the perf data file header but are ignored from now on. Signed-off-by:
-
Jiri Olsa authored
The only user of the event types data is 'perf timechart' command and uses this info to identify proper tracepoints based on its name. Switching this code to use tracepoint callbacks handlers same as another commands like builtin-{kmem,lock,sched}.c using the perf_session__set_tracepoints_handlers function. This way we get rid of the only event types user and can remove them completely in next patches. Signed-off-by: -
Jiri Olsa authored
Adding traceevent lib event-parse.h include to timechart command and removing duplicated local 'enum trace_flag_type' definition. Signed-off-by:
-
- 12 Jul, 2013 29 commits
-
-
Jiri Olsa authored
Moving hist_entry__period_snprintf function into stdio code and making it static, as it's no longer used anywhere else. Signed-off-by:
-
Jiri Olsa authored
Adding option 'o' to allow sorting based on the input file number. By default (without -o option) the output is sorted on baseline. Also removing '+' sorting support from -c option, because it's not needed anymore. Signed-off-by:
-
Jiri Olsa authored
All compute functions are now local to the diff command, making them static. Signed-off-by:
-
Jiri Olsa authored
Updating perf diff documentation to include multiple perf data files comparison. Signed-off-by:
-
Jiri Olsa authored
Adding diff command the flexibility to specify multiple data files on input. If not input file is given the standard behaviour stands and diff inspects 'perf.data' and 'perf.data.old' files. Signed-off-by:
-
Jiri Olsa authored
Another step towards multiple data files support. Having columns definition within struct data__file force each data file having its own columns. Signed-off-by:
-
Jiri Olsa authored
Moving diff related columns into diff command, because they are not used by any other command. Also moving the column entry functions under generic one with baseline as an exception. Signed-off-by:
-
Jiri Olsa authored
Data files are referenced through the index of the file on the command line. Adding list of data files for each index to ease up navigation for user. It's displayed only if in verbose mode. Signed-off-by:
-
Jiri Olsa authored
It does not make sense to make some computation (ratio, wdiff), when the hist_entry is 'dummy' - added via hists__link. Adding dummy field to struct hist_entry which indicates that it was added by hists__link and avoiding some of the processing for such entries. Signed-off-by:
-
Jiri Olsa authored
Making the baseline hists to act as a pairs head. So far we don't care which hists act as a pairs head, because we have only 2 files to deal with and any of them is suitable to do the job. But if we want to process more files, we need to pick up one hists to act as pairs head, and the baseline hists is the most suitable. Signed-off-by:
-
Jiri Olsa authored
Introducing struct diff_data to hold data file specifics. It will be handy when dealing with more than 2 data files. Signed-off-by:
-
Jiri Olsa authored
Now when diff command is separated from other standard outputs, we can use perf_hpp__init to initialize all standard columns. Moving PERF_HPP__OVERHEAD column init back to perf_hpp__init, and removing extra enable calls. Signed-off-by:
-
Jiri Olsa authored
Adding 'struct perf_hpp_fmt' into hpp callbacks, so commands can access their private data. It'll be handy for diff command in future to be able to access file related data for each column. Signed-off-by:
-
Greg Price authored
For example, in an application with an expensive function implemented with deeply nested recursive calls, the default call-graph presentation is dominated by the different callchains within that function. By ignoring these callees, we can collect the callchains leading into the function and compactly identify what to blame for expensive calls. For example, in this report the callers of garbage_collect() are scattered across the tree: $ perf report -d ruby 2>- | grep -m10 ^[^#]*[a-z] 22.03% ruby [.] gc_mark --- gc_mark |--59.40%-- mark_keyvalue | st_foreach | gc_mark_children | |--99.75%-- rb_gc_mark | | rb_vm_mark | | gc_mark_children | | gc_marks | | |--99.00%-- garbage_collect If we ignore the callees of garbage_collect(), its callers are coalesced: $ perf report --ignore-callees garbage_collect -d ruby 2>- | grep -m10 ^[^#]*[a-z] 72.92% ruby [.] garbage_collect --- garbage_collect vm_xmalloc |--47.08%-- ruby_xmalloc | st_insert2 | rb_hash_aset | |--98.45%-- features_index_add | | rb_provide_feature | | rb_require_safe | | vm_call_method Signed-off-by:Greg Price <price@mit.edu> Tested-by:
-
Andi Kleen authored
List the kernel supplied pmu event aliases in perf list It's better when the users can actually see them. Signed-off-by:
Andi Kleen <ak@linux.intel.com> Link: http://lkml.kernel.org/r/1366480949-32292-2-git-send-email-andi@firstfloor.orgSigned-off-by:
-
Andi Kleen authored
When an event fails to parse and it's not in a new style format, try to parse it again as a cpu event. This allows to use sysfs exported events directly without //, so you can use perf record -e mem-loads ... instead of perf record -e cpu/mem-loads/ Signed-off-by:
-
Adrian Hunter authored
As evident from 'machine__process_fork_event()' and 'machine__process_exit_event()' the 'pid' member of struct thread is actually the tid. Rename 'pid' to 'tid' in struct thread accordingly. Signed-off-by:
Adrian Hunter <adrian.hunter@intel.com> Acked-by:
-
Adrian Hunter authored
The 'size' variable includes the header so must be at least 'sizeof(struct perf_event_header)'. Error out immediately if that is not the case. Also don't byte-swap the header until it is actually "fetched" from the mmap region. Signed-off-by:
-
Adrian Hunter authored
The same lines of code are used in three places. Make it a new function '__perf_evlist__munmap()'. Signed-off-by:
-
Adrian Hunter authored
The name parameter is constant, declare it so. Signed-off-by:
-
Adrian Hunter authored
By default, perf inject should "repipe" all events including 'finished_round'. Signed-off-by:
-
Adrian Hunter authored
The 'inject' command expects to get a reference to 'struct perf_inject' from its 'tool' member. For that to work, 'tool' needs to be a parameter of all tool callbacks. Make it so. Signed-off-by:
-
Adrian Hunter authored
The 'machine' parameter is unused in 'perf_event__repipe_synth()' and some callers pass NULL anyway. So remove it. Signed-off-by:
-
Namhyung Kim authored
The 'rules' means that every second line of the tree view has a shaded background, which makes it easier to see which cell belongs to which row in the tree view. It can be useful for a tree view that has a lot of rows. Reviewed-by:
Pekka Enberg <penberg@kernel.org> Signed-off-by:
-
Namhyung Kim authored
If callchain is displayed, add "row-activated" signal handler for handling double-click or pressing ENTER key action. Reviewed-by:
-
Namhyung Kim authored
Sometimes it's annoying to see when some symbols have very wierd long names. So it might be a good idea to make column size changable. Reviewed-by:
-
Namhyung Kim authored
Display callchain percent value in the overhead column. Signed-off-by:
-
Namhyung Kim authored
Display callchain information in the symbol column. It's only enabled when recorded with -g and has symbol sort key. Reviewed-by:
-
Namhyung Kim authored
The GtkTreeStore can save items in a tree-like way. This is a preparation for supporting callgraphs in the hist browser. Reviewed-by:
-
