aboutsummaryrefslogtreecommitdiff
path: root/tools/perf
AgeCommit message (Collapse)AuthorFilesLines
2010-05-17perf tui: Add workaround for slang < 2.1.4Arnaldo Carvalho de Melo1-10/+21
Older versions of the slang library didn't used the 'const' specifier, causing problems with modern compilers of this kind: util/newt.c:252: error: passing argument 1 of ‘SLsmg_printf’ discards qualifiers from pointer target type Fix it by using some wrappers that when needed const the affected parameters back to plain (char *). Reported-by: Lin Ming <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Lin Ming <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-17perf record: Fix bug mismatch with -c option definitionStephane Eranian1-6/+5
The -c option defines the user requested sampling period. It was implemented using an unsigned int variable but the type of the option was OPT_LONG. Thus, the option parser was overwriting memory belonging to other variables, namely the mmap_pages leading to a zero page sampling buffer. The bug was exposed only when compiling at -O0, probably because the compiler was padding variables at higher optimization levels. This patch fixes this problem by declaring user_interval as u64. This also avoids wrap-around issues for large period on 32-bit systems. Commiter note: Made it use OPT_U64(user_interval) after implementing OPT_U64 in the previous patch. Cc: David S. Miller <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Stephane Eranian <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-17perf options: Introduce OPT_U64Arnaldo Carvalho de Melo2-0/+20
We have things like user_interval (-c/--count) in 'perf record' that needs this. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Stephane Eranian <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-16perf tui: Add help window to show key associationsArnaldo Carvalho de Melo1-0/+60
Suggested-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-16perf tui: Make <- exit menus tooArnaldo Carvalho de Melo1-2/+1
In fact it is now added to the hot key list when newt_form__new is used, allowing us to remove the explicit assignment in all its users. The visible change is that <- will exit the menu that pops up when -> is pressed (and Enter when callchains are not being used). Suggested-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-15perf newt: Add single key shortcuts for zoom into DSO and threadsArnaldo Carvalho de Melo1-3/+20
'D'/'d' for zooming into the DSO in the current highlighted hist entry, 'T'/'t' for zooming into the current thread. Suggested-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-15perf newt: Exit browser unconditionally when CTRL+C, q or Q is pressedArnaldo Carvalho de Melo1-3/+4
ESC still asks for confirmation. Suggested-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-15perf newt: Fix the 'A'/'a' shortcut for annotateArnaldo Carvalho de Melo1-2/+2
Reported-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-15perf newt: Make <- exit the ui_browserArnaldo Carvalho de Melo1-1/+3
Right now that means that pressing the left arrow willl make the symbol annotation window to exit back to the main symbol histogram browser. This is another improvement on the UI fastpath, i.e. just the arrows and enter are enough for most browsing. Suggested-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-14perf newt: Make <- zoom out filtersArnaldo Carvalho de Melo4-3/+120
After we use the filters to zoom into DSOs or threads, we can use <- (left arrow) to zoom out from the last filter applied. It is still possible to zoom out of order by using the popup menu. With this we now have the zoom out operation on the browsing fast path, by allowing fast navigation using just the four arrors and the enter key to expand collapse callchains. Suggested-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-14perf report: Report number of events, not samplesArnaldo Carvalho de Melo8-63/+107
Number of samples is meaningless after we switched to auto-freq, so report the number of events, i.e. not the sum of the different periods, but the number PERF_RECORD_SAMPLE emitted by the kernel. While doing this I noticed that naming "count" to the sum of all the event periods can be confusing, so rename it to .period, just like in struct sample.data, so that we become more consistent. This helps with the next step, that was to record in struct hist_entry the number of sample events for each instance, we need that because we use it to generate the number of events when applying filters to the tree of hist entries like it is being done in the TUI report browser. Suggested-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-14perf hist: Clarify events_stats fields usageArnaldo Carvalho de Melo9-34/+41
The events_stats.total field is too generic, rename it to .total_period, and also add a comment explaining that it is the sum of all the .period fields in samples, that is needed because we use auto-freq to avoid sampling artifacts. Ditto for events_stats.lost, that is the sum of all lost_event.lost fields, i.e. the number of events the kernel dropped. Looking at the users, builtin-sched.c can make use of these fields and stop doing it again. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-14perf hist: Make event__totals per histsArnaldo Carvalho de Melo8-38/+62
This is one more thing that started global but are more useful per hist or per session. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-13perf trace scripts: Fix typos in perf-trace-python.txtKirill Smelkov1-2/+2
option option -> option special special -> special Signed-off-by: Kirill Smelkov <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-13perf tools: change event inheritance logic in stat and recordStephane Eranian4-15/+15
By default, event inheritance across fork and pthread_create was on but the -i option of stat and record, which enabled inheritance, led to believe it was off by default. This patch fixes this logic by inverting the meaning of the -i option. By default inheritance is on whether you attach to a process (-p), a thread (-t) or start a process. If you pass -i, then you turn off inheritance. Turning off inheritance if you don't need it, helps limit perf resource usage as well. The patch also fixes perf stat -t xxxx and perf record -t xxxx which did not start the counters. Acked-by: Frederic Weisbecker <[email protected]> Cc: David S. Miller <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Stephane Eranian <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-13perf hist: Fix missing getline declarationFrederic Weisbecker1-0/+1
hist.c needs to include util.h so that it gets stdio.h inclusion with __GNU_SOURCE defined. Fixes: util/hist.c: In function ‘hist_entry__parse_objdump_line’: util/hist.c:931: erreur: implicit declaration of function ‘getline’ util/hist.c:931: erreur: nested extern declaration of ‘getline’ Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-13perf hist: Fix hists__browse no-newt caseFrederic Weisbecker1-1/+1
Fix mistake in a parameter type of the no-newt hists__browse() version. Fixes: builtin-report.c: In function ‘__cmd_report’: builtin-report.c:314: erreur: incompatible type for argument 1 of ‘hists__browse’ Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-13perf record: Add a fallback to the reference relocation symbolArnaldo Carvalho de Melo1-0/+3
Usually "_text" is enough, but I received reports that its not always available, so fallback to "_stext" for the symbol we use to check if we need to apply any relocation to all the symbols in the kernel symtab, for when, for instance, kexec is being used. Reported-by: Darren Hart <[email protected]> Reported-by: Steven Rostedt <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2010-05-11perf report: Librarize the annotation code and use it in the newt browserArnaldo Carvalho de Melo7-249/+544
Now we don't anymore use popen to run 'perf annotate' for the selected symbol, instead we collect per address samplings when processing samples in 'perf report' if we're using the newt browser, then we use this data directly to do annotation. Done this way we can actually traverse the objdump_line objects directly, matching the addresses to the collected samples and colouring them appropriately using lower level slang routines. The new ui_browser class will be reused for the main, callchain aware, histogram browser, when it will be made generic and don't assume that the objects are always instances of the objdump_line class maintained using list_heads. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-11perf ui: Add ui_helpline methodsArnaldo Carvalho de Melo1-19/+50
Initially this was just to be able to have a printf like method to prepare the formatted string and then pass to newtPushHelpLine, but as we already have for ui_progress, etc, its a step in identifying a restricted, highlevel set of widgets we can then have implementations for multiple widget sets (GTK, etc). Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-11perf symbols: allow forcing use of cplus_demangleKyle McMartin1-0/+3
For Fedora, I want to force perf to link against libiberty.a for cplus_demangle, rather than libbfd.a for bfd_demangle due to licensing insanity on binutils. (libiberty is LGPL2, libbfd is GPL3.) If we just rely on autodetection, we'll end up with libbfd linked against us, since they're both in binutils-static in the buildroot. Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Kyle McMartin <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-11perf probe: Check older elfutils and set NO_DWARFMasami Hiramatsu1-2/+2
Check whether elfutils is older than 0.138 (from which version checking routine has been introduced). And if so, set NO_DWARF because it is hard to check the API dependency without version checking. Signed-off-by: Masami Hiramatsu <[email protected]> Reported-by: Robert Richter <[email protected]> Cc: Robert Richter <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-11perf hist: Adopt filter by dso and by thread methods from the newt browserArnaldo Carvalho de Melo5-85/+89
Those are really not specific to the newt code, can be used by other UI frontends. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-11perf: Fix static strings treated like dynamic onesFrederic Weisbecker1-1/+1
The raw_field_ptr() helper, used to retrieve the address of a field inside a trace event, treats every strings as if they were dynamic ie: having a secondary level of indirection to retrieve their contents. FIELD_IS_STRING doesn't mean FIELD_IS_DYNAMIC, we only need to compute the secondary dereference for the latter case. This fixes perf sched segfaults, bad cmdline report and may be some other bugs. Reported-by: Jason Baron <[email protected]> Reported-by: Arnaldo Carvalho de Melo <[email protected]> Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Tom Zanussi <[email protected]>
2010-05-10perf/trace/scripting: syscall-counts script cleanupTom Zanussi1-1/+1
A small fix for the syscall counts script: - silence the match output in the shell script Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: syscall-counts-by-pid script cleanupTom Zanussi1-1/+1
A small fix for the syscall counts by pid script: - silence the match output in the shell script Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: failed-syscalls-by-pid script cleanupTom Zanussi1-1/+1
A small fixe for the failed syscalls by pid script: - silence the match output in the shell script Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: don't show script start/stop messages by defaultTom Zanussi3-7/+3
Only print the script start/stop messages in verbose mode - users normally don't care and it just clutters up the output. Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: workqueue-stats script cleanupTom Zanussi1-6/+6
Some minor fixes for the workqueue-stats script: - Fix nuisance 'use of uninitialized value' warnings Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: wakeup-latency script cleanupTom Zanussi1-4/+8
Some minor fixes for the wakeup-latency script: - Fix nuisance 'use of uninitialized value' warnings - Avoid divide-by-zero error Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: rwtop script cleanupTom Zanussi1-13/+35
A couple of fixes for the rwtop script: - printing the totals and clearing the hashes in the signal handler eventually leads to various random and serious problems when running the rwtop script continuously. Moving the print_totals() calls to the event handlers solves that problem, and the event handlers are invoked frequently enough that it doesn't affect the timeliness of the output. - Fix nuisance 'use of uninitialized value' warnings Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Message-Id: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: rw-by-pid script cleanupTom Zanussi1-23/+37
Some minor fixes for the rw-by-pid script: - Fix nuisance 'use of uninitialized value' warnings - Change the failed read/write sections to sort by error counts Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf/trace/scripting: failed-syscalls script cleanupTom Zanussi2-2/+6
A couple small fixes for the failed syscalls script: - The script description says it can be restricted to a specific comm, make it so. - silence the match output in the shell script Cc: Frédéric Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Tom Zanussi <[email protected]> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf hist: Calculate max_sym name len and nr_entriesArnaldo Carvalho de Melo5-13/+30
Better done when we are adding entries, be it initially of when we're re-sorting the histograms. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf hist: Introduce hists class and move lots of methods to itArnaldo Carvalho de Melo10-164/+146
In cbbc79a we introduced support for multiple events by introducing a new "event_stat_id" struct and then made several perf_session methods receive a point to it instead of a pointer to perf_session, and kept the event_stats and hists rb_tree in perf_session. While working on the new newt based browser, I realised that it would be better to introduce a new class, "hists" (short for "histograms"), renaming the "event_stat_id" struct and the perf_session methods that were really "hists" methods, as they manipulate only struct hists members, not touching anything in the other perf_session members. Other optimizations, such as calculating the maximum lenght of a symbol name present in an hists instance will be possible as we add them, avoiding a re-traversal just for finding that information. The rationale for the name "hists" to replace "event_stat_id" is that we may have multiple sets of hists for the same event_stat id, as, for instance, the 'perf diff' tool has, so event stat id is not what characterizes what this struct and the functions that manipulate it do. Cc: Eric B Munson <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf session: create_kernel_maps should use ->host_machineArnaldo Carvalho de Melo1-3/+2
Using machines__create_kernel_maps(..., HOST_KERNEL_ID) it would create another machine instance for the host machine, and since 1f626bc we have it out of the machines rb_tree. Fix it by using machine__create_kernel_maps(&self->host_machine) directly. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf callchains: Use zalloc to allocate objectsArnaldo Carvalho de Melo1-3/+3
Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10perf newt: Use newtAddComponent()Arnaldo Carvalho de Melo1-3/+3
Instead of newtAddComponents(just-one-entry, NULL), that is not needed if, like in this browser, we're adding just one component at a time. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-10Merge branch 'perf/test' of ↵Ingo Molnar5-105/+238
git://git.kernel.org/pub/scm/linux/kernel/git/frederic/random-tracing into perf/core
2010-05-09perf report: Allow limiting the number of entries to print in callchainsArnaldo Carvalho de Melo3-1/+15
Works by adding a third parameter to the '-g' argument, after the graph type and minimum percentage, for example: [root@doppio linux-2.6-tip]# perf report -g fractal,0.5,2 Will show only the first two symbols where at least 0.5% of the samples took place. All the other symbols that don't fall outside these constraints will be put together in the last entry, prefixed with "[...]" and the total percentage for them. Suggested-by: Arjan van de Ven <[email protected]> Acked-by: Arjan van de Ven <[email protected]> Cc: Arjan van de Ven <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-09perf session: Embed the host machine data on perf_sessionArnaldo Carvalho de Melo6-31/+29
We have just one host on a given session, and that is the most common setup right now, so embed a ->host_machine struct machine instance directly in the perf_session class, check if we're looking for it before going to the rb_tree. This also fixes a problem found when we try to process old perf.data files where we didn't have MMAP events for the kernel and modules and thus don't create the kernel maps, do it in event__preprocess_sample if it wasn't already. Reported-by: Ingo Molnar <[email protected]> Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> Cc: Zhang, Yanmin <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-09perf symbols: Check if a struct machine instance was foundArnaldo Carvalho de Melo1-2/+8
Which can happen when processing old files that had no fake kernel MMAP, events. That shouldn't result in perf_session__create_kernel_maps not being called, this will be fixed in a followup patch, for now do these checks to avoid segfaulting. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-09perf symbols: Consider unresolved DSOs in the dso__col_widt calculationArnaldo Carvalho de Melo1-0/+7
By using BITS_PER_LONG / 4, that is the number of chars that will be used in such cases as the DSO "name". Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-09perf lock: Drop "-a" option from cmd_record() default arguments setHitoshi Mitake1-1/+0
This patch drops "-a" from the default arguments passed to perf record by perf lock. If a user wants to do a system wide record of lock events, perf lock record -a <program> <argument> ... is enough for this purpose. This can reduce the size of the perf.data file. % sudo ./perf lock record whoami root [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.439 MB perf.data (~19170 samples) ] % sudo ./perf lock record -a whoami # with -a option root [ perf record: Woken up 0 times to write data ] [ perf record: Captured and wrote 48.962 MB perf.data (~2139197 samples) ] Signed-off-by: Hitoshi Mitake <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Jens Axboe <[email protected]> Cc: Jason Baron <[email protected]> Cc: Xiao Guangrong <[email protected]> LKML-Reference: Message-Id: <[email protected]> Signed-off-by: Frederic Weisbecker <[email protected]>
2010-05-09perf hist: Simplify the insertion of new hist_entry instancesArnaldo Carvalho de Melo5-44/+32
And with that fix at least one bug: The first hit for an entry, the one that calls malloc to create a new instance in __perf_session__add_hist_entry, wasn't adding the count to the per cpumode (PERF_RECORD_MISC_USER, etc) total variable. Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-09perf report: Fix leak of resolved callchains array on error pathArnaldo Carvalho de Melo1-9/+7
Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-09perf callchain: Move validate_callchain to callchain libArnaldo Carvalho de Melo3-14/+11
Cc: Frédéric Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2010-05-09perf/live-mode: Handle payload-less eventsTom Zanussi1-8/+11
Some events, such as the PERF_RECORD_FINISHED_ROUND event consist of only an event header and no data. In this case, a 0-length payload will be read, and the 0 return value will be wrongly interpreted as an 'unexpected end of event stream'. This patch allows for proper handling of data-less events by skipping 0-length reads. Signed-off-by: Tom Zanussi <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Masami Hiramatsu <[email protected]> LKML-Reference: <1273038527.6383.51.camel@tropicana> Signed-off-by: Frederic Weisbecker <[email protected]>
2010-05-09perf lock: Always check min AND max wait timeFrederic Weisbecker1-1/+1
When a lock is acquired after beeing contended, we update the wait time statistics for the given lock. But if the min wait time is updated, we don't check the max wait time. This is wrong because the first time we update the wait time, we want to update both min and max wait time. Before: Name acquired contended total wait (ns) max wait (ns) min wait (ns) key 8 1 21656 0 21656 After: Name acquired contended total wait (ns) max wait (ns) min wait (ns) key 8 1 21656 21656 21656 Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Hitoshi Mitake <[email protected]>
2010-05-09perf: Fix perf lock bad rateFrederic Weisbecker1-2/+2
Fix the cast made to get the bad rate. It is made in the result instead of the operands. We need the operands to be cast in double, otherwise the result will always be zero. Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Hitoshi Mitake <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.