aboutsummaryrefslogtreecommitdiff
path: root/tools/perf
AgeCommit message (Collapse)AuthorFilesLines
2009-10-20perf tools: Add bunch of missing headers to LIB_HArnaldo Carvalho de Melo1-0/+17
Build dependencies were not properly mapped out. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-20perf tools: Add missing tools/perf/util/include/string.hArnaldo Carvalho de Melo2-0/+2
To cure a bunch of: In file included from util/include/linux/bitmap.h:1, from util/header.h:8, from builtin-trace.c:7: util/include/../../../../include/linux/bitmap.h:8:26: error: linux/string.h: No such file or directory make: *** [builtin-trace.o] Error 1 make: *** Waiting for unfinished jobs.... Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Acked-by: Frederic Weisbecker <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-19perf stat: Count branches firstIngo Molnar1-3/+3
Count branches first, cache-misses second. The reason is that on x86 branches are not counted by all counters on all CPUs. Before: Performance counter stats for 'ls': 0.756653 task-clock-msecs # 0.802 CPUs 0 context-switches # 0.000 M/sec 0 CPU-migrations # 0.000 M/sec 250 page-faults # 0.330 M/sec 2375725 cycles # 3139.781 M/sec 1628129 instructions # 0.685 IPC 19643 cache-references # 25.960 M/sec 4608 cache-misses # 6.090 M/sec 342532 branches # 452.694 M/sec <not counted> branch-misses 0.000943356 seconds time elapsed After: Performance counter stats for 'ls': 1.056734 task-clock-msecs # 0.859 CPUs 0 context-switches # 0.000 M/sec 0 CPU-migrations # 0.000 M/sec 259 page-faults # 0.245 M/sec 3345932 cycles # 3166.295 M/sec 3074090 instructions # 0.919 IPC 616928 branches # 583.806 M/sec 39279 branch-misses # 6.367 % 21312 cache-references # 20.168 M/sec 3661 cache-misses # 3.464 M/sec 0.001230551 seconds time elapsed (also prettify the printout of branch misses, in case it's getting scaled.) Cc: Tim Blechmann <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> --- tools/perf/builtin-stat.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index c373683..95a55ea 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -59,6 +59,8 @@ static struct perf_event_attr default_attrs[] = { { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES}, { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS}, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, }; --- tools/perf/builtin-stat.c | 20 ++++++++++---------- 1 files changed, 10 insertions(+), 10 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index 95a55ea..90e0a26 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -50,17 +50,17 @@ static struct perf_event_attr default_attrs[] = { - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK }, - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES}, - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS }, - { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS }, - - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES}, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES }, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS}, - { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_TASK_CLOCK }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CONTEXT_SWITCHES }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_CPU_MIGRATIONS }, + { .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS }, + + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, };
2009-10-19perf stat: Re-align the default_attrs[] arrayIngo Molnar1-11/+11
Clean up the array definition to be vertically aligned. No functional effects. Cc: Tim Blechmann <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]> --- tools/perf/builtin-stat.c | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c index c373683..95a55ea 100644 --- a/tools/perf/builtin-stat.c +++ b/tools/perf/builtin-stat.c @@ -59,6 +59,8 @@ static struct perf_event_attr default_attrs[] = { { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS }, { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES}, { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES }, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS}, + { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES }, };
2009-10-19perf stat: Add branch performance events to default outputTim Blechmann1-0/+2
Adds performance event information about branches and branch misses to the default output of perf stat. Signed-off-by: Tim Blechmann <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-19perf tools: Display better error messages on missing packagesRandy Dunlap1-1/+5
Check for libelf headers and glibc headers separately so that the error message correctly identifies which package installation is missing/needed. Signed-off-by: Randy Dunlap <[email protected]> Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-19perf top: Fix --delay_secs 0 division by zeroTim Blechmann1-0/+2
Add delay_secs sanity check to handle_keypress, this fixes a division by zero crash. Signed-off-by: Tim Blechmann <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-19perf tools: Use DECLARE_BITMAP instead of an open-coded arrayFrederic Weisbecker4-22/+20
Use DECLARE_BITMAP instead of an open coded array for our bitmap of featured sections. This makes the array an unsigned long instead of a u64 but since we use a 256 bits bitmap, the array size shouldn't vary between different boxes. Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Steven Rostedt <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-19perf tools: Introduce bitmask'ed additional headersFrederic Weisbecker3-56/+78
This provides a new set of bitmasked headers. A new field is added in the perf headers that implements a bitmap storing optional features present in the perf.data file. The layout can be pictured like this: (Usual perf headers)(Features bitmap)[Feature 0][Feature n][Feature 255] If the bit n is set, then the feature n is used in this file. They are all set in order. This brings a backward and forward compatibility. The trace_info section has moved into such optional features, this is the first and only one for now. This is backward compatible with the .32 file version although it doesn't support the previous separate trace.info file. And finally it doesn't support the current interim development version. Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Steven Rostedt <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-19perf tools: Use kernel bitmap libraryFrederic Weisbecker14-5/+156
Use the kernel bitmap library for internal perf tools uses. Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Steven Rostedt <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-19perf stat: Add branch performance metricAnton Blanchard1-0/+11
When we count both branches and branch-misses it is useful to print out the percentage of branch-misses: # perf stat -e branches -e branch-misses /bin/true Performance counter stats for '/bin/true': 401684 branches # 0.000 M/sec 23301 branch-misses # 5.801 % Signed-off-by: Anton Blanchard <[email protected]> Cc: [email protected] Cc: [email protected] LKML-Reference: <20091018112923.GQ4808@kryten> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-17Merge commit 'v2.6.32-rc5' into perf/probesIngo Molnar15-99/+185
Conflicts: kernel/trace/trace_event_profile.c Merge reason: update to -rc5 and resolve conflict. Signed-off-by: Ingo Molnar <[email protected]>
2009-10-17perf: Add perf-probe documentMasami Hiramatsu2-0/+49
Add perf-probe subcommand document and add it to command-list. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-17perf: Add DIE_IF() macro for error checkingMasami Hiramatsu4-51/+51
Add DIE_IF() macro and replace ERR_IF() with it, and use linux/stringify.h. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-17perf: Use eprintf() for debug messages in perf-probeMasami Hiramatsu3-21/+21
Replace debug() macro with eprintf() and add -v option for showing those messages in perf-probe. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-17perf: Use die() for error cases in perf-probeMasami Hiramatsu2-68/+34
Use die() for exiting perf-probe with errors. This replaces perror_exit(), msg_exit() and fprintf()+exit() with die(), and uses die() in semantic_error(). This also renames 'die' local variables to 'dw_die' for avoiding name confliction. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-17perf: Check libdwarf APIs for perf probeMasami Hiramatsu1-2/+2
Check libdwarf APIs for perf probe in tools/perf/Makefile. Since dwarf_get_ranges() has been added from libdwarf 20081231 (and it's the newest function used in probe-finder.c), this just checks whether the function is defined. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-17perf tools: Move dereference after NULL testJulia Lawall4-8/+8
In each case, if the NULL test on thread is needed, then the dereference should be after the NULL test. A simplified version of the semantic match that detects this problem is as follows (http://coccinelle.lip6.fr/): // <smpl> @match exists@ expression x, E; identifier fld; @@ * x->fld ... when != \(x = E\|&x\) * x == NULL // </smpl> Signed-off-by: Julia Lawall <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-16perf tools: Bump version to 0.0.2Ingo Molnar1-1/+1
We released the first version of perf with 0.0.1 in v2.6.31, time to double our version number to 0.0.2 ;-) Cc: Peter Zijlstra <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf trace: Add filter SuppportLi Zefan3-3/+40
Add a new option "--filter <filter_str>" to perf record, and it should be right after "-e trace_point": #./perf record -R -f -e irq:irq_handler_entry --filter irq==18 ^C # ./perf trace perf-4303 ... irq_handler_entry: irq=18 handler=eth0 init-0 ... irq_handler_entry: irq=18 handler=eth0 init-0 ... irq_handler_entry: irq=18 handler=eth0 init-0 ... irq_handler_entry: irq=18 handler=eth0 init-0 ... irq_handler_entry: irq=18 handler=eth0 See Documentation/trace/events.txt for the syntax of filter expressions. Signed-off-by: Li Zefan <[email protected]> Acked-by: Peter Zijlstra <[email protected]> Acked-by: Frederic Weisbecker <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Tom Zanussi <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Remove all char * typecasts and use const in prototypeSteven Rostedt1-61/+61
The (char *) for all the static strings was a fix for the symptom and not the disease. The real issue was that the function prototypes needed to be declared "const char *". Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle - and + in parsing trace print formatSteven Rostedt1-0/+6
The opterators '-' and '+' are not handled in the trace print format. To do: '++' and '--'. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Add latency format to trace outputSteven Rostedt3-19/+114
Add the irqs disabled, preemption count, need resched, and other info that is shown in the latency format of ftrace. # perf trace -l perf-16457 2..s2. 53636.260344: kmem_cache_free: call_site=ffffffff811198f perf-16457 2..s2. 53636.264330: kmem_cache_free: call_site=ffffffff811198f perf-16457 2d.s4. 53636.300006: kmem_cache_free: call_site=ffffffff810d889 Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle both versions of ftrace outputSteven Rostedt1-1/+4
The ftrace output events can have either arguments or no arguments. The parser needs to be able to handle both. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Fix bprintk reading in trace outputSteven Rostedt1-8/+7
The bprintk parsing was broken in more ways than one. The file parsing was incorrect, and the words used by the arguments are always 4 bytes aligned, even on 64-bit machines. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Still continue on failed parsing of an eventSteven Rostedt2-17/+35
Even though an event may fail to parse, we should not kill the entire report. The trace should still be able to show what it can. If an event fails to parse, a warning is printed, and the output continues. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle the case with and without the "signed" trace fieldSteven Rostedt1-29/+52
The trace format files now have a "signed" field. But we should still be able to handle the kernels that do not have this field. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle newlines in trace parsing betterSteven Rostedt1-1/+7
New lines between args in the trace format can break the parsing. This should not be the case. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle * as typecast in trace parsingSteven Rostedt1-28/+22
The '*' is currently only treated as a multiplication, and it needs to be handled as a typecast pointer. This is the version used by trace-cmd. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle arrays in print fields for trace parsingSteven Rostedt1-0/+62
The array used by the ftrace stack events (caller[x]) causes issues with the parser. This adds code to handle the case, but it also assumes that the array is of type long. Note, this is a special case used (currently) only by the ftrace user and kernel stack records. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle trace parsing of < and >Steven Rostedt1-0/+2
The code to handle the '<' and '>' ops was all in place, but they were not in the switch statement to consider them as valid ops. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Fix backslash processing on trace print formatsSteven Rostedt1-2/+25
The handling of backslashes was broken. It would stop parsing when encountering one. Also, '\n', '\t', '\r' and '\\' were not converted. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15perf tools: Handle print concatenations in event format fileSteven Rostedt1-0/+16
kmem_alloc ftrace event format had a string that was broken up by two tokens. "string 1" "string 2". This patch lets the parser be able to handle the concatenation. Signed-off-by: Steven Rostedt <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-15Merge branch 'linus' into perf/coreIngo Molnar4-14/+24
Merge reason: pick up tools/perf/ changes from upstream. Signed-off-by: Ingo Molnar <[email protected]>
2009-10-13perf tools: Move threads & last_match to threads.cArnaldo Carvalho de Melo6-84/+45
This was just being copy'n'pasted all over. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Mike Galbraith <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-13perf tools: Remove expensive old debug code from perf topMike Galbraith1-14/+1
Calling gettimeofday() at high frequency is painful for handicapped boxen. The spot calling gettimeofday() is old unneeded debug code, so remove it. Reported-by: Ingo Molnar <[email protected]> Signed-off-by: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Peter Zijlstra <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-13perf tools: Do not manually count string lengthsVincent Legoll2-8/+13
Use strlen & macros instead of manually counting string lengths as this is error prone and may lend to bugs. Signed-off-by: Vincent Legoll <[email protected]> Cc: Linus Torvalds <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-12perf probe: Add perf probe command support without libdwarfMasami Hiramatsu4-12/+40
Enables 'perf probe' even if libdwarf is not installed. If libdwarf is not found, 'perf probe' just disables dwarf support. Users can use 'perf probe' to set up new events by using kprobe_events format. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Christoph Hellwig <[email protected]> Cc: Ananth N Mavinakayanahalli <[email protected]> Cc: Jim Keniston <[email protected]> Cc: Frank Ch. Eigler <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Frederic Weisbecker <[email protected]>
2009-10-12perf: Add perf probe subcommand, a kprobe-event setup helperMasami Hiramatsu6-0/+1130
Add perf probe subcommand that implements a kprobe-event setup helper to the perf command. This allows user to define kprobe events using C expressions (C line numbers, C function names, and C local variables). Usage ----- perf probe [<options>] -P 'PROBEDEF' [-P 'PROBEDEF' ...] -k, --vmlinux <file> vmlinux/module pathname -P, --probe <p|r:[GRP/]NAME FUNC[+OFFS][@SRC]|@SRC:LINE [ARG ...]> probe point definition, where p: kprobe probe r: kretprobe probe GRP: Group name (optional) NAME: Event name FUNC: Function name OFFS: Offset from function entry (in byte) SRC: Source code path LINE: Line number ARG: Probe argument (local variable name or kprobe-tracer argument format is supported.) Changes in v4: - Add _GNU_SOURCE macro for strndup(). Changes in v3: - Remove -r option because perf always be used for online kernel. - Check malloc/calloc results. Changes in v2: - Check synthesized string length. - Rename perf kprobe to perf probe. - Use spaces for separator and update usage comment. - Check error paths in parse_probepoint(). - Check optimized-out variables. Signed-off-by: Masami Hiramatsu <[email protected]> Cc: Ingo Molnar <[email protected]> Cc: Thomas Gleixner <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Steven Rostedt <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Christoph Hellwig <[email protected]> Cc: Ananth N Mavinakayanahalli <[email protected]> Cc: Jim Keniston <[email protected]> Cc: Frank Ch. Eigler <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Frederic Weisbecker <[email protected]>
2009-10-12perf tools: Remove static debugfs path from parse-eventsAshwin Chaugule1-1/+4
Timechart doesn't work if debugfs is not in /sys/kernel/debug/. Fixed by using global debugfs_path which is filled in by perf. Signed-off-by: Ashwin Chaugule <[email protected]> Cc: "Arjan van de Ven" <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-12perf tools: Fix the NO_64BIT build on pure 64-bit systemsIngo Molnar1-7/+14
Randy Dunlap reported that 'make NO_64BIT=1' fails to build a pure 32-b it binary on 64-bit/64-bit x86 systems. The reason is that we dont pass in the -m32 and GCC defaults to -m64. So pass it in - and also extend the warning message about libelf dependencies - glibc-dev[el] is needed as well beyond the libelf library. Reported-by: Randy Dunlap <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> LKML-Reference: Message-Id: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-12perf sched: Add -C option to measure on a specific CPUMike Galbraith1-1/+100
To refresh, trying to sched record only one CPU results in bogus latencies as below. I fixed^Wmade it stop doing the bad thing today, by following task migration events properly. Before: marge:/root/tmp # taskset -c 1 perf sched record -C 0 -- sleep 10 marge:/root/tmp # perf sched lat ----------------------------------------------------------------------------------------- Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | ----------------------------------------------------------------------------------------- Xorg:4943 | 1.290 ms | 1 | avg: 1670.132 ms | max: 1670.132 ms | hald-addon-stor:3569 | 0.091 ms | 3 | avg: 658.609 ms | max: 1975.797 ms | hald-addon-stor:3573 | 0.209 ms | 4 | avg: 499.138 ms | max: 1990.565 ms | audispd:4270 | 0.012 ms | 1 | avg: 0.015 ms | max: 0.015 ms | .... marge:/root/tmp # perf sched trace|grep 'Xorg:4943' swapper-0 [000] 401.184013288: sched_stat_runtime: task: Xorg:4943 runtime: 1233188 [ns], vruntime: 19105169779 [ns] rt2870TimerQHan-4947 [000] 402.854140127: sched_stat_wait: task: Xorg:4943 wait: 580073 [ns] rt2870TimerQHan-4947 [000] 402.854141770: sched_migrate_task: task Xorg:4943 [140] from: 1 to: 0 rt2870TimerQHan-4947 [000] 402.854143854: sched_stat_wait: task: Xorg:4943 wait: 0 [ns] rt2870TimerQHan-4947 [000] 402.854145397: sched_switch: task rt2870TimerQHan:4947 [140] (D) ==> Xorg:4943 [140] Xorg-4943 [000] 402.854193133: sched_stat_runtime: task: Xorg:4943 runtime: 56546 [ns], vruntime: 11766332500 [ns] Xorg-4943 [000] 402.854196842: sched_switch: task Xorg:4943 [140] (S) ==> swapper:0 [140] After: marge:/root/tmp # taskset -c 1 perf sched record -C 0 -- sleep 10 marge:/root/tmp # perf sched lat ----------------------------------------------------------------------------------------- Task | Runtime ms | Switches | Average delay ms | Maximum delay ms | ----------------------------------------------------------------------------------------- amarokapp:11150 | 271.297 ms | 878 | avg: 0.130 ms | max: 1.057 ms | konsole:5965 | 1.370 ms | 12 | avg: 0.092 ms | max: 0.855 ms | Xorg:4943 | 179.980 ms | 1109 | avg: 0.087 ms | max: 1.206 ms | hald-addon-stor:3574 | 0.212 ms | 9 | avg: 0.040 ms | max: 0.169 ms | hald-addon-stor:3570 | 0.223 ms | 9 | avg: 0.037 ms | max: 0.223 ms | klauncher:5864 | 0.550 ms | 8 | avg: 0.032 ms | max: 0.048 ms | The 'Maximum delay ms' results are now sane. Signed-off-by: Mike Galbraith <[email protected]> LKML-Reference: <new-submission> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-12perf tools: Fix counter sample frequency breakageMike Galbraith2-8/+34
Commit 42e59d7d19dc4b4 switched to a default sample frequency of 1KHz, which overrides any user supplied count, causing sched, top and timechart to miss events due to their discrete events being flagged PERF_SAMPLE_PERIOD. Override default sample frequency when the user profides a period count, and make both record and top honor that user supplied option. Signed-off-by: Mike Galbraith <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arjan van de Ven <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-12perf tools: Fix const char type propagationRandy Dunlap2-6/+6
The following perf build warnings/errors in function argument types: builtin-sched.c:1894: warning: passing argument 1 of 'sort_dimension__add' discards qualifiers from pointer target type util/trace-event-parse.c:685: warning: passing argument 2 of 'read_expected' discards qualifiers from pointer target type util/trace-event-parse.c:741: warning: passing argument 4 of 'test_type_token' discards qualifiers from pointer target type util/trace-event-parse.c:706: warning: passing argument 2 of 'read_expected_item' discards qualifiers from pointer target type ... trigger because older GCC is not able to prove that sort_dimension__add() does not change the string. Some goes for test_type_token(). Fix this by improving type consistency. Signed-off-by: Randy Dunlap <[email protected]> Acked-by: Frederic Weisbecker <[email protected]> Acked-by: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> [ Also remove ugly type cast now unnecessary. ] Signed-off-by: Ingo Molnar <[email protected]>
2009-10-08perf tools: Provide backward compatibility with previous perf.data versionFrederic Weisbecker1-1/+7
We have merged the trace.info file into perf.data by adding one section in the perf headers. This makes it incompatible with previous version: the new perf tools can't read the older perf.data. To support the previous format, we check the headers size. If they have the same size than in the previous format, then ignore the trace info section that doesn't exist. Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-08perf tools: Fix thread comm resolution in perf schedFrederic Weisbecker3-68/+13
This reverts commit 9a92b479b2f088ee2d3194243f4c8e59b1b8c9c2 ("perf tools: Improve thread comm resolution in perf sched") and fixes the real bug. The bug was elsewhere: We are failing to resolve thread names in perf sched because the table of threads we are building, on top of comm events, has a per process granularity. But perf sched, unlike the other perf tools, needs a per thread granularity as we are profiling every tasks individually. So fix it by building our threads table using the tid instead of the pid as the thread identifier. v2: Revert the previous fix - it is not really needed Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-08perf tools: Improve kernel/modules symbol lookupArnaldo Carvalho de Melo1-87/+201
This removes the ovelapping of vmlinux addresses with modules, using the ELF section name when using --vmlinux and creating a unique DSO name when using /proc/kallsyms ([kernel].N). This is done by creating multiple 'struct map' instances for address ranges backed by DSOs that have just the symbols for that range and a name that is derived from the ELF section name.o Now it is possible to ask for just the symbols in some particular kernel section: $ perf report -m --vmlinux ../build/tip-recvmmsg/vmlinux \ --dsos [kernel].vsyscall_fn | head -15 52.73% Xorg [.] vread_hpet 18.61% firefox [.] vread_hpet 14.50% npviewer.bin [.] vread_hpet 6.83% compiz [.] vread_hpet 5.73% glxgears [.] vread_hpet 0.63% java [.] vread_hpet 0.30% gnome-terminal [.] vread_hpet 0.23% perf [.] vread_hpet 0.18% xchat [.] vread_hpet $ Now we don't have to first lookup the list of modules and then, if it fails, vmlinux symbols, its just a simple lookup for the map then the symbols, just like for threads. Reports generated using /proc/kallsyms and --vmlinux should provide the same results, modulo the DSO name for sections other than ".text". But they don't right now because things like: ffffffff81011c20-ffffffff81012068 system_call ffffffff81011c30-ffffffff81011c9b system_call_after_swapgs ffffffff81011c9c-ffffffff81011cb6 system_call_fastpath ffffffff81011cb7-ffffffff81011cbb ret_from_sys_call I.e. overlapping symbols, again some ASM special case that we have to fixup. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Mike Galbraith <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-08perf tools: Up the verbose level for some really verbose stuffArnaldo Carvalho de Melo3-6/+6
Like printing every symbol created. Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Cc: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Mike Galbraith <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-08perf tools: Improve thread comm resolution in perf schedFrederic Weisbecker3-12/+67
When we get sched traces that involve a task that was already created before opening the event, we won't have the comm event for it. So if we can't find the comm event for a given thread, we look at the traces that may contain these informations. Before: ata/1:371 | 0.000 ms | 1 | avg: 3988.693 ms | max: 3988.693 ms | kondemand/1:421 | 0.096 ms | 3 | avg: 345.346 ms | max: 1035.989 ms | kondemand/0:420 | 0.025 ms | 3 | avg: 421.332 ms | max: 964.014 ms | :5124:5124 | 0.103 ms | 5 | avg: 74.082 ms | max: 277.194 ms | :6244:6244 | 0.691 ms | 9 | avg: 125.655 ms | max: 271.306 ms | firefox:5080 | 0.924 ms | 5 | avg: 53.833 ms | max: 257.828 ms | npviewer.bin:6225 | 21.871 ms | 53 | avg: 22.462 ms | max: 220.835 ms | :6245:6245 | 9.631 ms | 21 | avg: 41.864 ms | max: 213.349 ms | After: ata/1:371 | 0.000 ms | 1 | avg: 3988.693 ms | max: 3988.693 ms | kondemand/1:421 | 0.096 ms | 3 | avg: 345.346 ms | max: 1035.989 ms | kondemand/0:420 | 0.025 ms | 3 | avg: 421.332 ms | max: 964.014 ms | firefox:5124 | 0.103 ms | 5 | avg: 74.082 ms | max: 277.194 ms | npviewer.bin:6244 | 0.691 ms | 9 | avg: 125.655 ms | max: 271.306 ms | firefox:5080 | 0.924 ms | 5 | avg: 53.833 ms | max: 257.828 ms | npviewer.bin:6225 | 21.871 ms | 53 | avg: 22.462 ms | max: 220.835 ms | npviewer.bin:6245 | 9.631 ms | 21 | avg: 41.864 ms | max: 213.349 ms | Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> LKML-Reference: <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2009-10-08perf tools: Unify perf.data mapping and events handlingFrederic Weisbecker6-401/+334
This librarizes the perf.data file mapping and handling in various perf tools, roughly reducing the amount of code and fixing the places that mmap from beginning of the file whereas we want to mmap from the beginning of the data, leading to page fault because the mmap window is too small since the trace info are written in the file too. TODO: - convert perf timechart too Signed-off-by: Frederic Weisbecker <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Arnaldo Carvalho de Melo <[email protected]> Cc: Mike Galbraith <[email protected]> Cc: Paul Mackerras <[email protected]> Cc: Arjan van de Ven <[email protected]> LKML-Reference: <20091007104729.GD5043@nowhere> Signed-off-by: Ingo Molnar <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.