2009-06-26 22:28:00 +08:00
|
|
|
/*
|
2011-01-14 11:51:58 +08:00
|
|
|
* Copyright (C) 2009-2011, Frederic Weisbecker <fweisbec@gmail.com>
|
2009-06-26 22:28:00 +08:00
|
|
|
*
|
|
|
|
* Handle the callchains from the stream in an ad-hoc radix tree and then
|
|
|
|
* sort them in an rbtree.
|
|
|
|
*
|
2009-07-01 11:35:15 +08:00
|
|
|
* Using a radix for code path provides a fast retrieval and factorizes
|
|
|
|
* memory use. Also that lets us use the paths in a hierarchical graph view.
|
|
|
|
*
|
2009-06-26 22:28:00 +08:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <stdlib.h>
|
|
|
|
#include <stdio.h>
|
|
|
|
#include <stdbool.h>
|
|
|
|
#include <errno.h>
|
2009-08-09 10:19:15 +08:00
|
|
|
#include <math.h>
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2014-01-14 23:37:15 +08:00
|
|
|
#include "asm/bug.h"
|
|
|
|
|
2013-07-19 06:33:57 +08:00
|
|
|
#include "hist.h"
|
2010-05-20 23:15:33 +08:00
|
|
|
#include "util.h"
|
2014-01-14 13:25:35 +08:00
|
|
|
#include "sort.h"
|
|
|
|
#include "machine.h"
|
2009-06-26 22:28:00 +08:00
|
|
|
#include "callchain.h"
|
|
|
|
|
2012-05-31 13:43:26 +08:00
|
|
|
__thread struct callchain_cursor callchain_cursor;
|
|
|
|
|
2015-08-04 16:30:20 +08:00
|
|
|
int parse_callchain_record_opt(const char *arg, struct callchain_param *param)
|
2014-09-23 09:01:42 +08:00
|
|
|
{
|
2015-08-07 03:44:52 +08:00
|
|
|
return parse_callchain_record(arg, param);
|
2014-09-23 09:01:42 +08:00
|
|
|
}
|
|
|
|
|
2014-09-23 09:01:43 +08:00
|
|
|
static int parse_callchain_mode(const char *value)
|
|
|
|
{
|
|
|
|
if (!strncmp(value, "graph", strlen(value))) {
|
|
|
|
callchain_param.mode = CHAIN_GRAPH_ABS;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (!strncmp(value, "flat", strlen(value))) {
|
|
|
|
callchain_param.mode = CHAIN_FLAT;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (!strncmp(value, "fractal", strlen(value))) {
|
|
|
|
callchain_param.mode = CHAIN_GRAPH_REL;
|
|
|
|
return 0;
|
|
|
|
}
|
2015-11-09 13:45:37 +08:00
|
|
|
if (!strncmp(value, "folded", strlen(value))) {
|
|
|
|
callchain_param.mode = CHAIN_FOLDED;
|
|
|
|
return 0;
|
|
|
|
}
|
2017-01-25 00:44:10 +08:00
|
|
|
|
|
|
|
pr_err("Invalid callchain mode: %s\n", value);
|
2014-09-23 09:01:43 +08:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int parse_callchain_order(const char *value)
|
|
|
|
{
|
|
|
|
if (!strncmp(value, "caller", strlen(value))) {
|
|
|
|
callchain_param.order = ORDER_CALLER;
|
2015-10-22 15:45:46 +08:00
|
|
|
callchain_param.order_set = true;
|
2014-09-23 09:01:43 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (!strncmp(value, "callee", strlen(value))) {
|
|
|
|
callchain_param.order = ORDER_CALLEE;
|
2015-10-22 15:45:46 +08:00
|
|
|
callchain_param.order_set = true;
|
2014-09-23 09:01:43 +08:00
|
|
|
return 0;
|
|
|
|
}
|
2017-01-25 00:44:10 +08:00
|
|
|
|
|
|
|
pr_err("Invalid callchain order: %s\n", value);
|
2014-09-23 09:01:43 +08:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static int parse_callchain_sort_key(const char *value)
|
|
|
|
{
|
|
|
|
if (!strncmp(value, "function", strlen(value))) {
|
|
|
|
callchain_param.key = CCKEY_FUNCTION;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (!strncmp(value, "address", strlen(value))) {
|
|
|
|
callchain_param.key = CCKEY_ADDRESS;
|
|
|
|
return 0;
|
|
|
|
}
|
perf callchain: Support handling complete branch stacks as histograms
Currently branch stacks can be only shown as edge histograms for
individual branches. I never found this display particularly useful.
This implements an alternative mode that creates histograms over
complete branch traces, instead of individual branches, similar to how
normal callgraphs are handled. This is done by putting it in front of
the normal callgraph and then using the normal callgraph histogram
infrastructure to unify them.
This way in complex functions we can understand the control flow that
lead to a particular sample, and may even see some control flow in the
caller for short functions.
Example (simplified, of course for such simple code this is usually not
needed), please run this after the whole patchkit is in, as at this
point in the patch order there is no --branch-history, that will be
added in a patch after this one:
tcall.c:
volatile a = 10000, b = 100000, c;
__attribute__((noinline)) f2()
{
c = a / b;
}
__attribute__((noinline)) f1()
{
f2();
f2();
}
main()
{
int i;
for (i = 0; i < 1000000; i++)
f1();
}
% perf record -b -g ./tsrc/tcall
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.044 MB perf.data (~1923 samples) ]
% perf report --no-children --branch-history
...
54.91% tcall.c:6 [.] f2 tcall
|
|--65.53%-- f2 tcall.c:5
| |
| |--70.83%-- f1 tcall.c:11
| | f1 tcall.c:10
| | main tcall.c:18
| | main tcall.c:18
| | main tcall.c:17
| | main tcall.c:17
| | f1 tcall.c:13
| | f1 tcall.c:13
| | f2 tcall.c:7
| | f2 tcall.c:5
| | f1 tcall.c:12
| | f1 tcall.c:12
| | f2 tcall.c:7
| | f2 tcall.c:5
| | f1 tcall.c:11
| |
| --29.17%-- f1 tcall.c:12
| f1 tcall.c:12
| f2 tcall.c:7
| f2 tcall.c:5
| f1 tcall.c:11
| f1 tcall.c:10
| main tcall.c:18
| main tcall.c:18
| main tcall.c:17
| main tcall.c:17
| f1 tcall.c:13
| f1 tcall.c:13
| f2 tcall.c:7
| f2 tcall.c:5
| f1 tcall.c:12
The default output is unchanged.
This is only implemented in perf report, no change to record or anywhere
else.
This adds the basic code to report:
- add a new "branch" option to the -g option parser to enable this mode
- when the flag is set include the LBR into the callstack in machine.c.
The rest of the history code is unchanged and doesn't know the
difference between LBR entry and normal call entry.
- detect overlaps with the callchain
- remove small loop duplicates in the LBR
Current limitations:
- The LBR flags (mispredict etc.) are not shown in the history
and LBR entries have no special marker.
- It would be nice if annotate marked the LBR entries somehow
(e.g. with arrows)
v2: Various fixes.
v3: Merge further patches into this one. Fix white space.
v4: Improve manpage. Address review feedback.
v5: Rename functions. Better error message without -g. Fix crash without
-b.
v6: Rebase
v7: Rebase. Use NO_ENTRY in memset.
v8: Port to latest tip. Move add_callchain_ip to separate
patch. Skip initial entries in callchain. Minor cleanups.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1415844328-4884-3-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2014-11-13 10:05:20 +08:00
|
|
|
if (!strncmp(value, "branch", strlen(value))) {
|
|
|
|
callchain_param.branch_callstack = 1;
|
|
|
|
return 0;
|
|
|
|
}
|
2017-01-25 00:44:10 +08:00
|
|
|
|
|
|
|
pr_err("Invalid callchain sort key: %s\n", value);
|
2014-09-23 09:01:43 +08:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
2015-11-09 13:45:41 +08:00
|
|
|
static int parse_callchain_value(const char *value)
|
|
|
|
{
|
|
|
|
if (!strncmp(value, "percent", strlen(value))) {
|
|
|
|
callchain_param.value = CCVAL_PERCENT;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (!strncmp(value, "period", strlen(value))) {
|
|
|
|
callchain_param.value = CCVAL_PERIOD;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
if (!strncmp(value, "count", strlen(value))) {
|
|
|
|
callchain_param.value = CCVAL_COUNT;
|
|
|
|
return 0;
|
|
|
|
}
|
2017-01-25 00:44:10 +08:00
|
|
|
|
|
|
|
pr_err("Invalid callchain config key: %s\n", value);
|
2015-11-09 13:45:41 +08:00
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
|
perf top: Support call-graph display options also
Currently 'perf top --call-graph' option is same as 'perf record'. But
'perf top' also need to receive display options in 'perf report'. To do
that, change parse_callchain_report_opt() to allow record options too.
Now perf top can receive display options like below:
$ perf top --call-graph
Error: option `call-graph' requires a value
Usage: perf top [<options>]
--call-graph
<mode[,dump_size],output_type,min_percent[,print_limit],call_order[,branch]>
setup and enables call-graph (stack chain/backtrace)
recording: fp dwarf lbr, output_type (graph, flat,
fractal, or none), min percent threshold, optional
print limit, callchain order, key (function or
address), add branches
$ perf top --call-graph callee,graph,fp
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Chandler Carruth <chandlerc@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1445495330-25416-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-10-22 14:28:49 +08:00
|
|
|
static int
|
|
|
|
__parse_callchain_report_opt(const char *arg, bool allow_record_opt)
|
2014-04-08 02:55:24 +08:00
|
|
|
{
|
2014-08-14 14:01:38 +08:00
|
|
|
char *tok;
|
2014-04-08 02:55:24 +08:00
|
|
|
char *endptr;
|
2014-08-14 14:01:38 +08:00
|
|
|
bool minpcnt_set = false;
|
perf top: Support call-graph display options also
Currently 'perf top --call-graph' option is same as 'perf record'. But
'perf top' also need to receive display options in 'perf report'. To do
that, change parse_callchain_report_opt() to allow record options too.
Now perf top can receive display options like below:
$ perf top --call-graph
Error: option `call-graph' requires a value
Usage: perf top [<options>]
--call-graph
<mode[,dump_size],output_type,min_percent[,print_limit],call_order[,branch]>
setup and enables call-graph (stack chain/backtrace)
recording: fp dwarf lbr, output_type (graph, flat,
fractal, or none), min percent threshold, optional
print limit, callchain order, key (function or
address), add branches
$ perf top --call-graph callee,graph,fp
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Chandler Carruth <chandlerc@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1445495330-25416-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-10-22 14:28:49 +08:00
|
|
|
bool record_opt_set = false;
|
|
|
|
bool try_stack_size = false;
|
2014-04-08 02:55:24 +08:00
|
|
|
|
2016-04-18 22:53:07 +08:00
|
|
|
callchain_param.enabled = true;
|
2014-04-08 02:55:24 +08:00
|
|
|
symbol_conf.use_callchain = true;
|
|
|
|
|
|
|
|
if (!arg)
|
|
|
|
return 0;
|
|
|
|
|
2014-08-14 14:01:38 +08:00
|
|
|
while ((tok = strtok((char *)arg, ",")) != NULL) {
|
|
|
|
if (!strncmp(tok, "none", strlen(tok))) {
|
|
|
|
callchain_param.mode = CHAIN_NONE;
|
2016-04-18 22:53:07 +08:00
|
|
|
callchain_param.enabled = false;
|
2014-08-14 14:01:38 +08:00
|
|
|
symbol_conf.use_callchain = false;
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2014-09-23 09:01:43 +08:00
|
|
|
if (!parse_callchain_mode(tok) ||
|
|
|
|
!parse_callchain_order(tok) ||
|
2015-11-09 13:45:41 +08:00
|
|
|
!parse_callchain_sort_key(tok) ||
|
|
|
|
!parse_callchain_value(tok)) {
|
2014-09-23 09:01:43 +08:00
|
|
|
/* parsing ok - move on to the next */
|
perf top: Support call-graph display options also
Currently 'perf top --call-graph' option is same as 'perf record'. But
'perf top' also need to receive display options in 'perf report'. To do
that, change parse_callchain_report_opt() to allow record options too.
Now perf top can receive display options like below:
$ perf top --call-graph
Error: option `call-graph' requires a value
Usage: perf top [<options>]
--call-graph
<mode[,dump_size],output_type,min_percent[,print_limit],call_order[,branch]>
setup and enables call-graph (stack chain/backtrace)
recording: fp dwarf lbr, output_type (graph, flat,
fractal, or none), min percent threshold, optional
print limit, callchain order, key (function or
address), add branches
$ perf top --call-graph callee,graph,fp
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Chandler Carruth <chandlerc@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1445495330-25416-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-10-22 14:28:49 +08:00
|
|
|
try_stack_size = false;
|
|
|
|
goto next;
|
|
|
|
} else if (allow_record_opt && !record_opt_set) {
|
|
|
|
if (parse_callchain_record(tok, &callchain_param))
|
|
|
|
goto try_numbers;
|
|
|
|
|
|
|
|
/* assume that number followed by 'dwarf' is stack size */
|
|
|
|
if (callchain_param.record_mode == CALLCHAIN_DWARF)
|
|
|
|
try_stack_size = true;
|
|
|
|
|
|
|
|
record_opt_set = true;
|
|
|
|
goto next;
|
|
|
|
}
|
|
|
|
|
|
|
|
try_numbers:
|
|
|
|
if (try_stack_size) {
|
|
|
|
unsigned long size = 0;
|
|
|
|
|
|
|
|
if (get_stack_size(tok, &size) < 0)
|
|
|
|
return -1;
|
|
|
|
callchain_param.dump_size = size;
|
|
|
|
try_stack_size = false;
|
2014-09-23 09:01:43 +08:00
|
|
|
} else if (!minpcnt_set) {
|
|
|
|
/* try to get the min percent */
|
2014-08-14 14:01:38 +08:00
|
|
|
callchain_param.min_percent = strtod(tok, &endptr);
|
|
|
|
if (tok == endptr)
|
|
|
|
return -1;
|
|
|
|
minpcnt_set = true;
|
|
|
|
} else {
|
|
|
|
/* try print limit at last */
|
|
|
|
callchain_param.print_limit = strtoul(tok, &endptr, 0);
|
|
|
|
if (tok == endptr)
|
|
|
|
return -1;
|
|
|
|
}
|
perf top: Support call-graph display options also
Currently 'perf top --call-graph' option is same as 'perf record'. But
'perf top' also need to receive display options in 'perf report'. To do
that, change parse_callchain_report_opt() to allow record options too.
Now perf top can receive display options like below:
$ perf top --call-graph
Error: option `call-graph' requires a value
Usage: perf top [<options>]
--call-graph
<mode[,dump_size],output_type,min_percent[,print_limit],call_order[,branch]>
setup and enables call-graph (stack chain/backtrace)
recording: fp dwarf lbr, output_type (graph, flat,
fractal, or none), min percent threshold, optional
print limit, callchain order, key (function or
address), add branches
$ perf top --call-graph callee,graph,fp
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Chandler Carruth <chandlerc@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1445495330-25416-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-10-22 14:28:49 +08:00
|
|
|
next:
|
2014-08-14 14:01:38 +08:00
|
|
|
arg = NULL;
|
2014-04-08 02:55:24 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
if (callchain_register_param(&callchain_param) < 0) {
|
|
|
|
pr_err("Can't register callchain params\n");
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
perf top: Support call-graph display options also
Currently 'perf top --call-graph' option is same as 'perf record'. But
'perf top' also need to receive display options in 'perf report'. To do
that, change parse_callchain_report_opt() to allow record options too.
Now perf top can receive display options like below:
$ perf top --call-graph
Error: option `call-graph' requires a value
Usage: perf top [<options>]
--call-graph
<mode[,dump_size],output_type,min_percent[,print_limit],call_order[,branch]>
setup and enables call-graph (stack chain/backtrace)
recording: fp dwarf lbr, output_type (graph, flat,
fractal, or none), min percent threshold, optional
print limit, callchain order, key (function or
address), add branches
$ perf top --call-graph callee,graph,fp
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: Brendan Gregg <brendan.d.gregg@gmail.com>
Cc: Chandler Carruth <chandlerc@gmail.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1445495330-25416-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2015-10-22 14:28:49 +08:00
|
|
|
int parse_callchain_report_opt(const char *arg)
|
|
|
|
{
|
|
|
|
return __parse_callchain_report_opt(arg, false);
|
|
|
|
}
|
|
|
|
|
|
|
|
int parse_callchain_top_opt(const char *arg)
|
|
|
|
{
|
|
|
|
return __parse_callchain_report_opt(arg, true);
|
|
|
|
}
|
|
|
|
|
2014-09-23 09:01:43 +08:00
|
|
|
int perf_callchain_config(const char *var, const char *value)
|
|
|
|
{
|
|
|
|
char *endptr;
|
|
|
|
|
|
|
|
if (prefixcmp(var, "call-graph."))
|
|
|
|
return 0;
|
|
|
|
var += sizeof("call-graph.") - 1;
|
|
|
|
|
|
|
|
if (!strcmp(var, "record-mode"))
|
2015-08-04 16:30:20 +08:00
|
|
|
return parse_callchain_record_opt(value, &callchain_param);
|
2014-09-23 09:01:43 +08:00
|
|
|
if (!strcmp(var, "dump-size")) {
|
|
|
|
unsigned long size = 0;
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
ret = get_stack_size(value, &size);
|
|
|
|
callchain_param.dump_size = size;
|
|
|
|
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
if (!strcmp(var, "print-type"))
|
|
|
|
return parse_callchain_mode(value);
|
|
|
|
if (!strcmp(var, "order"))
|
|
|
|
return parse_callchain_order(value);
|
|
|
|
if (!strcmp(var, "sort-key"))
|
|
|
|
return parse_callchain_sort_key(value);
|
|
|
|
if (!strcmp(var, "threshold")) {
|
|
|
|
callchain_param.min_percent = strtod(value, &endptr);
|
2017-01-25 00:44:10 +08:00
|
|
|
if (value == endptr) {
|
|
|
|
pr_err("Invalid callchain threshold: %s\n", value);
|
2014-09-23 09:01:43 +08:00
|
|
|
return -1;
|
2017-01-25 00:44:10 +08:00
|
|
|
}
|
2014-09-23 09:01:43 +08:00
|
|
|
}
|
|
|
|
if (!strcmp(var, "print-limit")) {
|
|
|
|
callchain_param.print_limit = strtod(value, &endptr);
|
2017-01-25 00:44:10 +08:00
|
|
|
if (value == endptr) {
|
|
|
|
pr_err("Invalid callchain print limit: %s\n", value);
|
2014-09-23 09:01:43 +08:00
|
|
|
return -1;
|
2017-01-25 00:44:10 +08:00
|
|
|
}
|
2014-09-23 09:01:43 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
static void
|
2009-07-02 23:58:21 +08:00
|
|
|
rb_insert_callchain(struct rb_root *root, struct callchain_node *chain,
|
|
|
|
enum chain_mode mode)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
|
|
|
struct rb_node **p = &root->rb_node;
|
|
|
|
struct rb_node *parent = NULL;
|
|
|
|
struct callchain_node *rnode;
|
2011-01-14 11:51:59 +08:00
|
|
|
u64 chain_cumul = callchain_cumul_hits(chain);
|
2009-06-26 22:28:00 +08:00
|
|
|
|
|
|
|
while (*p) {
|
2009-08-07 13:11:05 +08:00
|
|
|
u64 rnode_cumul;
|
|
|
|
|
2009-06-26 22:28:00 +08:00
|
|
|
parent = *p;
|
|
|
|
rnode = rb_entry(parent, struct callchain_node, rb_node);
|
2011-01-14 11:51:59 +08:00
|
|
|
rnode_cumul = callchain_cumul_hits(rnode);
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2009-07-02 23:58:21 +08:00
|
|
|
switch (mode) {
|
2009-07-05 13:39:21 +08:00
|
|
|
case CHAIN_FLAT:
|
2015-11-09 13:45:37 +08:00
|
|
|
case CHAIN_FOLDED:
|
2009-07-02 23:58:21 +08:00
|
|
|
if (rnode->hit < chain->hit)
|
|
|
|
p = &(*p)->rb_left;
|
|
|
|
else
|
|
|
|
p = &(*p)->rb_right;
|
|
|
|
break;
|
2009-07-05 13:39:21 +08:00
|
|
|
case CHAIN_GRAPH_ABS: /* Falldown */
|
|
|
|
case CHAIN_GRAPH_REL:
|
2009-08-07 13:11:05 +08:00
|
|
|
if (rnode_cumul < chain_cumul)
|
2009-07-02 23:58:21 +08:00
|
|
|
p = &(*p)->rb_left;
|
|
|
|
else
|
|
|
|
p = &(*p)->rb_right;
|
|
|
|
break;
|
2009-08-15 18:26:57 +08:00
|
|
|
case CHAIN_NONE:
|
2009-07-02 23:58:21 +08:00
|
|
|
default:
|
|
|
|
break;
|
|
|
|
}
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
rb_link_node(&chain->rb_node, parent, p);
|
|
|
|
rb_insert_color(&chain->rb_node, root);
|
|
|
|
}
|
|
|
|
|
2009-07-05 13:39:21 +08:00
|
|
|
static void
|
|
|
|
__sort_chain_flat(struct rb_root *rb_root, struct callchain_node *node,
|
|
|
|
u64 min_hit)
|
|
|
|
{
|
2013-10-11 13:15:36 +08:00
|
|
|
struct rb_node *n;
|
2009-07-05 13:39:21 +08:00
|
|
|
struct callchain_node *child;
|
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
n = rb_first(&node->rb_root_in);
|
|
|
|
while (n) {
|
|
|
|
child = rb_entry(n, struct callchain_node, rb_node_in);
|
|
|
|
n = rb_next(n);
|
|
|
|
|
2009-07-05 13:39:21 +08:00
|
|
|
__sort_chain_flat(rb_root, child, min_hit);
|
2013-10-11 13:15:36 +08:00
|
|
|
}
|
2009-07-05 13:39:21 +08:00
|
|
|
|
|
|
|
if (node->hit && node->hit >= min_hit)
|
|
|
|
rb_insert_callchain(rb_root, node, CHAIN_FLAT);
|
|
|
|
}
|
|
|
|
|
2009-06-26 22:28:00 +08:00
|
|
|
/*
|
|
|
|
* Once we get every callchains from the stream, we can now
|
|
|
|
* sort them by hit
|
|
|
|
*/
|
2009-07-05 13:39:21 +08:00
|
|
|
static void
|
2010-08-23 02:05:22 +08:00
|
|
|
sort_chain_flat(struct rb_root *rb_root, struct callchain_root *root,
|
2012-09-11 06:15:03 +08:00
|
|
|
u64 min_hit, struct callchain_param *param __maybe_unused)
|
2009-07-05 13:39:21 +08:00
|
|
|
{
|
2015-11-26 15:08:18 +08:00
|
|
|
*rb_root = RB_ROOT;
|
2010-08-23 02:05:22 +08:00
|
|
|
__sort_chain_flat(rb_root, &root->node, min_hit);
|
2009-07-05 13:39:21 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
static void __sort_chain_graph_abs(struct callchain_node *node,
|
|
|
|
u64 min_hit)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
2013-10-11 13:15:36 +08:00
|
|
|
struct rb_node *n;
|
2009-06-26 22:28:00 +08:00
|
|
|
struct callchain_node *child;
|
|
|
|
|
2009-07-05 13:39:21 +08:00
|
|
|
node->rb_root = RB_ROOT;
|
2013-10-11 13:15:36 +08:00
|
|
|
n = rb_first(&node->rb_root_in);
|
|
|
|
|
|
|
|
while (n) {
|
|
|
|
child = rb_entry(n, struct callchain_node, rb_node_in);
|
|
|
|
n = rb_next(n);
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2009-07-05 13:39:21 +08:00
|
|
|
__sort_chain_graph_abs(child, min_hit);
|
2011-01-14 11:51:59 +08:00
|
|
|
if (callchain_cumul_hits(child) >= min_hit)
|
2009-07-05 13:39:21 +08:00
|
|
|
rb_insert_callchain(&node->rb_root, child,
|
|
|
|
CHAIN_GRAPH_ABS);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void
|
2010-08-23 02:05:22 +08:00
|
|
|
sort_chain_graph_abs(struct rb_root *rb_root, struct callchain_root *chain_root,
|
2012-09-11 06:15:03 +08:00
|
|
|
u64 min_hit, struct callchain_param *param __maybe_unused)
|
2009-07-05 13:39:21 +08:00
|
|
|
{
|
2010-08-23 02:05:22 +08:00
|
|
|
__sort_chain_graph_abs(&chain_root->node, min_hit);
|
|
|
|
rb_root->rb_node = chain_root->node.rb_root.rb_node;
|
2009-07-02 23:58:21 +08:00
|
|
|
}
|
|
|
|
|
2009-07-05 13:39:21 +08:00
|
|
|
static void __sort_chain_graph_rel(struct callchain_node *node,
|
|
|
|
double min_percent)
|
2009-07-02 23:58:21 +08:00
|
|
|
{
|
2013-10-11 13:15:36 +08:00
|
|
|
struct rb_node *n;
|
2009-07-02 23:58:21 +08:00
|
|
|
struct callchain_node *child;
|
2009-07-05 13:39:21 +08:00
|
|
|
u64 min_hit;
|
2009-07-02 23:58:21 +08:00
|
|
|
|
|
|
|
node->rb_root = RB_ROOT;
|
2009-08-09 10:19:15 +08:00
|
|
|
min_hit = ceil(node->children_hit * min_percent);
|
2009-07-02 23:58:21 +08:00
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
n = rb_first(&node->rb_root_in);
|
|
|
|
while (n) {
|
|
|
|
child = rb_entry(n, struct callchain_node, rb_node_in);
|
|
|
|
n = rb_next(n);
|
|
|
|
|
2009-07-05 13:39:21 +08:00
|
|
|
__sort_chain_graph_rel(child, min_percent);
|
2011-01-14 11:51:59 +08:00
|
|
|
if (callchain_cumul_hits(child) >= min_hit)
|
2009-07-05 13:39:21 +08:00
|
|
|
rb_insert_callchain(&node->rb_root, child,
|
|
|
|
CHAIN_GRAPH_REL);
|
2009-07-02 23:58:21 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-07-05 13:39:21 +08:00
|
|
|
static void
|
2010-08-23 02:05:22 +08:00
|
|
|
sort_chain_graph_rel(struct rb_root *rb_root, struct callchain_root *chain_root,
|
2012-09-11 06:15:03 +08:00
|
|
|
u64 min_hit __maybe_unused, struct callchain_param *param)
|
2009-07-02 23:58:21 +08:00
|
|
|
{
|
2010-08-23 02:05:22 +08:00
|
|
|
__sort_chain_graph_rel(&chain_root->node, param->min_percent / 100.0);
|
|
|
|
rb_root->rb_node = chain_root->node.rb_root.rb_node;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2011-01-14 11:52:00 +08:00
|
|
|
int callchain_register_param(struct callchain_param *param)
|
2009-07-05 13:39:21 +08:00
|
|
|
{
|
|
|
|
switch (param->mode) {
|
|
|
|
case CHAIN_GRAPH_ABS:
|
|
|
|
param->sort = sort_chain_graph_abs;
|
|
|
|
break;
|
|
|
|
case CHAIN_GRAPH_REL:
|
|
|
|
param->sort = sort_chain_graph_rel;
|
|
|
|
break;
|
|
|
|
case CHAIN_FLAT:
|
2015-11-09 13:45:37 +08:00
|
|
|
case CHAIN_FOLDED:
|
2009-07-05 13:39:21 +08:00
|
|
|
param->sort = sort_chain_flat;
|
|
|
|
break;
|
2009-08-15 18:26:57 +08:00
|
|
|
case CHAIN_NONE:
|
2009-07-05 13:39:21 +08:00
|
|
|
default:
|
|
|
|
return -1;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
/*
|
|
|
|
* Create a child for a parent. If inherit_children, then the new child
|
|
|
|
* will become the new parent of it's parent children
|
|
|
|
*/
|
|
|
|
static struct callchain_node *
|
|
|
|
create_child(struct callchain_node *parent, bool inherit_children)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
|
|
|
struct callchain_node *new;
|
|
|
|
|
2010-05-10 21:56:50 +08:00
|
|
|
new = zalloc(sizeof(*new));
|
2009-06-26 22:28:00 +08:00
|
|
|
if (!new) {
|
|
|
|
perror("not enough memory to create child for code path tree");
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
new->parent = parent;
|
|
|
|
INIT_LIST_HEAD(&new->val);
|
2015-11-09 13:45:43 +08:00
|
|
|
INIT_LIST_HEAD(&new->parent_val);
|
2009-07-01 11:35:15 +08:00
|
|
|
|
|
|
|
if (inherit_children) {
|
2013-10-11 13:15:36 +08:00
|
|
|
struct rb_node *n;
|
|
|
|
struct callchain_node *child;
|
|
|
|
|
|
|
|
new->rb_root_in = parent->rb_root_in;
|
|
|
|
parent->rb_root_in = RB_ROOT;
|
2009-07-01 11:35:15 +08:00
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
n = rb_first(&new->rb_root_in);
|
|
|
|
while (n) {
|
|
|
|
child = rb_entry(n, struct callchain_node, rb_node_in);
|
|
|
|
child->parent = new;
|
|
|
|
n = rb_next(n);
|
|
|
|
}
|
2009-07-01 11:35:15 +08:00
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
/* make it the first child */
|
|
|
|
rb_link_node(&new->rb_node_in, NULL, &parent->rb_root_in.rb_node);
|
|
|
|
rb_insert_color(&new->rb_node_in, &parent->rb_root_in);
|
2009-07-01 11:35:15 +08:00
|
|
|
}
|
2009-06-26 22:28:00 +08:00
|
|
|
|
|
|
|
return new;
|
|
|
|
}
|
|
|
|
|
2010-03-23 00:09:33 +08:00
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
/*
|
|
|
|
* Fill the node with callchain values
|
|
|
|
*/
|
2016-02-16 22:08:21 +08:00
|
|
|
static int
|
2011-01-14 11:51:58 +08:00
|
|
|
fill_node(struct callchain_node *node, struct callchain_cursor *cursor)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
2011-01-14 11:51:58 +08:00
|
|
|
struct callchain_cursor_node *cursor_node;
|
|
|
|
|
|
|
|
node->val_nr = cursor->nr - cursor->pos;
|
|
|
|
if (!node->val_nr)
|
|
|
|
pr_warning("Warning: empty node in callchain tree\n");
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
cursor_node = callchain_cursor_current(cursor);
|
|
|
|
|
|
|
|
while (cursor_node) {
|
2009-06-26 22:28:00 +08:00
|
|
|
struct callchain_list *call;
|
|
|
|
|
2010-05-10 21:56:50 +08:00
|
|
|
call = zalloc(sizeof(*call));
|
2009-06-26 22:28:00 +08:00
|
|
|
if (!call) {
|
|
|
|
perror("not enough memory for the code path tree");
|
2016-02-16 22:08:21 +08:00
|
|
|
return -1;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
2011-01-14 11:51:58 +08:00
|
|
|
call->ip = cursor_node->ip;
|
|
|
|
call->ms.sym = cursor_node->sym;
|
2017-01-06 14:23:31 +08:00
|
|
|
call->ms.map = map__get(cursor_node->map);
|
2016-10-31 09:19:51 +08:00
|
|
|
|
|
|
|
if (cursor_node->branch) {
|
|
|
|
call->branch_count = 1;
|
|
|
|
|
|
|
|
if (cursor_node->branch_flags.predicted)
|
|
|
|
call->predicted_count = 1;
|
|
|
|
|
|
|
|
if (cursor_node->branch_flags.abort)
|
|
|
|
call->abort_count = 1;
|
|
|
|
|
|
|
|
call->cycles_count = cursor_node->branch_flags.cycles;
|
|
|
|
call->iter_count = cursor_node->nr_loop_iter;
|
|
|
|
call->samples_count = cursor_node->samples;
|
|
|
|
}
|
|
|
|
|
2009-06-26 22:28:00 +08:00
|
|
|
list_add_tail(&call->list, &node->val);
|
2011-01-14 11:51:58 +08:00
|
|
|
|
|
|
|
callchain_cursor_advance(cursor);
|
|
|
|
cursor_node = callchain_cursor_current(cursor);
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
2016-02-16 22:08:21 +08:00
|
|
|
return 0;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
static struct callchain_node *
|
2011-01-14 11:51:58 +08:00
|
|
|
add_child(struct callchain_node *parent,
|
|
|
|
struct callchain_cursor *cursor,
|
|
|
|
u64 period)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
|
|
|
struct callchain_node *new;
|
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
new = create_child(parent, false);
|
2016-02-16 22:08:20 +08:00
|
|
|
if (new == NULL)
|
|
|
|
return NULL;
|
|
|
|
|
2016-02-16 22:08:21 +08:00
|
|
|
if (fill_node(new, cursor) < 0) {
|
|
|
|
struct callchain_list *call, *tmp;
|
|
|
|
|
|
|
|
list_for_each_entry_safe(call, tmp, &new->val, list) {
|
|
|
|
list_del(&call->list);
|
2017-01-06 14:23:31 +08:00
|
|
|
map__zput(call->ms.map);
|
2016-02-16 22:08:21 +08:00
|
|
|
free(call);
|
|
|
|
}
|
|
|
|
free(new);
|
|
|
|
return NULL;
|
|
|
|
}
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2009-08-07 13:11:05 +08:00
|
|
|
new->children_hit = 0;
|
2010-07-08 09:41:46 +08:00
|
|
|
new->hit = period;
|
2015-11-09 13:45:40 +08:00
|
|
|
new->children_count = 0;
|
|
|
|
new->count = 1;
|
2013-10-11 13:15:36 +08:00
|
|
|
return new;
|
|
|
|
}
|
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
enum match_result {
|
|
|
|
MATCH_ERROR = -1,
|
|
|
|
MATCH_EQ,
|
|
|
|
MATCH_LT,
|
|
|
|
MATCH_GT,
|
|
|
|
};
|
|
|
|
|
|
|
|
static enum match_result match_chain(struct callchain_cursor_node *node,
|
|
|
|
struct callchain_list *cnode)
|
2013-10-11 13:15:36 +08:00
|
|
|
{
|
|
|
|
struct symbol *sym = node->sym;
|
2016-02-16 22:08:22 +08:00
|
|
|
u64 left, right;
|
2013-10-11 13:15:36 +08:00
|
|
|
|
|
|
|
if (cnode->ms.sym && sym &&
|
2016-02-16 22:08:22 +08:00
|
|
|
callchain_param.key == CCKEY_FUNCTION) {
|
|
|
|
left = cnode->ms.sym->start;
|
|
|
|
right = sym->start;
|
|
|
|
} else {
|
|
|
|
left = cnode->ip;
|
|
|
|
right = node->ip;
|
|
|
|
}
|
|
|
|
|
2016-10-31 09:19:51 +08:00
|
|
|
if (left == right) {
|
|
|
|
if (node->branch) {
|
|
|
|
cnode->branch_count++;
|
|
|
|
|
|
|
|
if (node->branch_flags.predicted)
|
|
|
|
cnode->predicted_count++;
|
|
|
|
|
|
|
|
if (node->branch_flags.abort)
|
|
|
|
cnode->abort_count++;
|
|
|
|
|
|
|
|
cnode->cycles_count += node->branch_flags.cycles;
|
|
|
|
cnode->iter_count += node->nr_loop_iter;
|
|
|
|
cnode->samples_count += node->samples;
|
|
|
|
}
|
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
return MATCH_EQ;
|
2016-10-31 09:19:51 +08:00
|
|
|
}
|
2016-02-16 22:08:22 +08:00
|
|
|
|
|
|
|
return left > right ? MATCH_GT : MATCH_LT;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
/*
|
|
|
|
* Split the parent in two parts (a new child is created) and
|
|
|
|
* give a part of its callchain to the created child.
|
|
|
|
* Then create another child to host the given callchain of new branch
|
|
|
|
*/
|
2016-02-16 22:08:23 +08:00
|
|
|
static int
|
2011-01-14 11:51:58 +08:00
|
|
|
split_add_child(struct callchain_node *parent,
|
|
|
|
struct callchain_cursor *cursor,
|
|
|
|
struct callchain_list *to_split,
|
|
|
|
u64 idx_parents, u64 idx_local, u64 period)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
|
|
|
struct callchain_node *new;
|
2009-07-01 11:35:15 +08:00
|
|
|
struct list_head *old_tail;
|
2009-07-01 18:37:06 +08:00
|
|
|
unsigned int idx_total = idx_parents + idx_local;
|
2009-06-26 22:28:00 +08:00
|
|
|
|
|
|
|
/* split */
|
2009-07-01 11:35:15 +08:00
|
|
|
new = create_child(parent, true);
|
2016-02-16 22:08:23 +08:00
|
|
|
if (new == NULL)
|
|
|
|
return -1;
|
2009-07-01 11:35:15 +08:00
|
|
|
|
|
|
|
/* split the callchain and move a part to the new child */
|
|
|
|
old_tail = parent->val.prev;
|
|
|
|
list_del_range(&to_split->list, old_tail);
|
|
|
|
new->val.next = &to_split->list;
|
|
|
|
new->val.prev = old_tail;
|
|
|
|
to_split->list.prev = &new->val;
|
|
|
|
old_tail->next = &new->val;
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
/* split the hits */
|
|
|
|
new->hit = parent->hit;
|
2009-08-07 13:11:05 +08:00
|
|
|
new->children_hit = parent->children_hit;
|
2011-01-14 11:51:59 +08:00
|
|
|
parent->children_hit = callchain_cumul_hits(new);
|
2009-07-01 11:35:15 +08:00
|
|
|
new->val_nr = parent->val_nr - idx_local;
|
|
|
|
parent->val_nr = idx_local;
|
2015-11-09 13:45:40 +08:00
|
|
|
new->count = parent->count;
|
|
|
|
new->children_count = parent->children_count;
|
|
|
|
parent->children_count = callchain_cumul_counts(new);
|
2009-07-01 11:35:15 +08:00
|
|
|
|
|
|
|
/* create a new child for the new branch if any */
|
2011-01-14 11:51:58 +08:00
|
|
|
if (idx_total < cursor->nr) {
|
2013-10-11 13:15:36 +08:00
|
|
|
struct callchain_node *first;
|
|
|
|
struct callchain_list *cnode;
|
|
|
|
struct callchain_cursor_node *node;
|
|
|
|
struct rb_node *p, **pp;
|
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
parent->hit = 0;
|
2010-07-08 09:41:46 +08:00
|
|
|
parent->children_hit += period;
|
2015-11-09 13:45:40 +08:00
|
|
|
parent->count = 0;
|
|
|
|
parent->children_count += 1;
|
2013-10-11 13:15:36 +08:00
|
|
|
|
|
|
|
node = callchain_cursor_current(cursor);
|
|
|
|
new = add_child(parent, cursor, period);
|
2016-02-16 22:08:20 +08:00
|
|
|
if (new == NULL)
|
2016-02-16 22:08:23 +08:00
|
|
|
return -1;
|
2013-10-11 13:15:36 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* This is second child since we moved parent's children
|
|
|
|
* to new (first) child above.
|
|
|
|
*/
|
|
|
|
p = parent->rb_root_in.rb_node;
|
|
|
|
first = rb_entry(p, struct callchain_node, rb_node_in);
|
|
|
|
cnode = list_first_entry(&first->val, struct callchain_list,
|
|
|
|
list);
|
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
if (match_chain(node, cnode) == MATCH_LT)
|
2013-10-11 13:15:36 +08:00
|
|
|
pp = &p->rb_left;
|
|
|
|
else
|
|
|
|
pp = &p->rb_right;
|
|
|
|
|
|
|
|
rb_link_node(&new->rb_node_in, p, pp);
|
|
|
|
rb_insert_color(&new->rb_node_in, &parent->rb_root_in);
|
2009-07-01 11:35:15 +08:00
|
|
|
} else {
|
2010-07-08 09:41:46 +08:00
|
|
|
parent->hit = period;
|
2015-11-09 13:45:40 +08:00
|
|
|
parent->count = 1;
|
2009-07-01 11:35:15 +08:00
|
|
|
}
|
2016-02-16 22:08:23 +08:00
|
|
|
return 0;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
static enum match_result
|
2011-01-14 11:51:58 +08:00
|
|
|
append_chain(struct callchain_node *root,
|
|
|
|
struct callchain_cursor *cursor,
|
|
|
|
u64 period);
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2016-02-16 22:08:24 +08:00
|
|
|
static int
|
2011-01-14 11:51:58 +08:00
|
|
|
append_chain_children(struct callchain_node *root,
|
|
|
|
struct callchain_cursor *cursor,
|
|
|
|
u64 period)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
|
|
|
struct callchain_node *rnode;
|
2013-10-11 13:15:36 +08:00
|
|
|
struct callchain_cursor_node *node;
|
|
|
|
struct rb_node **p = &root->rb_root_in.rb_node;
|
|
|
|
struct rb_node *parent = NULL;
|
|
|
|
|
|
|
|
node = callchain_cursor_current(cursor);
|
|
|
|
if (!node)
|
2016-02-16 22:08:24 +08:00
|
|
|
return -1;
|
2009-06-26 22:28:00 +08:00
|
|
|
|
|
|
|
/* lookup in childrens */
|
2013-10-11 13:15:36 +08:00
|
|
|
while (*p) {
|
2016-02-16 22:08:22 +08:00
|
|
|
enum match_result ret;
|
2009-07-01 18:37:06 +08:00
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
parent = *p;
|
|
|
|
rnode = rb_entry(parent, struct callchain_node, rb_node_in);
|
|
|
|
|
2014-01-14 23:37:15 +08:00
|
|
|
/* If at least first entry matches, rely to children */
|
|
|
|
ret = append_chain(rnode, cursor, period);
|
2016-02-16 22:08:22 +08:00
|
|
|
if (ret == MATCH_EQ)
|
2009-08-07 13:11:05 +08:00
|
|
|
goto inc_children_hit;
|
2016-02-16 22:08:24 +08:00
|
|
|
if (ret == MATCH_ERROR)
|
|
|
|
return -1;
|
2013-10-11 13:15:36 +08:00
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
if (ret == MATCH_LT)
|
2013-10-11 13:15:36 +08:00
|
|
|
p = &parent->rb_left;
|
|
|
|
else
|
|
|
|
p = &parent->rb_right;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
2009-07-01 11:35:15 +08:00
|
|
|
/* nothing in children, add to the current node */
|
2013-10-11 13:15:36 +08:00
|
|
|
rnode = add_child(root, cursor, period);
|
2016-02-16 22:08:20 +08:00
|
|
|
if (rnode == NULL)
|
2016-02-16 22:08:24 +08:00
|
|
|
return -1;
|
2016-02-16 22:08:20 +08:00
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
rb_link_node(&rnode->rb_node_in, parent, p);
|
|
|
|
rb_insert_color(&rnode->rb_node_in, &root->rb_root_in);
|
2009-07-05 13:39:20 +08:00
|
|
|
|
2009-08-07 13:11:05 +08:00
|
|
|
inc_children_hit:
|
2010-07-08 09:41:46 +08:00
|
|
|
root->children_hit += period;
|
2015-11-09 13:45:40 +08:00
|
|
|
root->children_count++;
|
2016-02-16 22:08:24 +08:00
|
|
|
return 0;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
static enum match_result
|
2011-01-14 11:51:58 +08:00
|
|
|
append_chain(struct callchain_node *root,
|
|
|
|
struct callchain_cursor *cursor,
|
|
|
|
u64 period)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
|
|
|
struct callchain_list *cnode;
|
2011-01-14 11:51:58 +08:00
|
|
|
u64 start = cursor->pos;
|
2009-06-26 22:28:00 +08:00
|
|
|
bool found = false;
|
2011-01-14 11:51:58 +08:00
|
|
|
u64 matches;
|
2016-02-16 22:08:22 +08:00
|
|
|
enum match_result cmp = MATCH_ERROR;
|
2009-06-26 22:28:00 +08:00
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
/*
|
|
|
|
* Lookup in the current node
|
|
|
|
* If we have a symbol, then compare the start to match
|
2013-07-19 06:33:57 +08:00
|
|
|
* anywhere inside a function, unless function
|
|
|
|
* mode is disabled.
|
2009-07-01 11:35:15 +08:00
|
|
|
*/
|
2009-06-26 22:28:00 +08:00
|
|
|
list_for_each_entry(cnode, &root->val, list) {
|
2011-01-14 11:51:58 +08:00
|
|
|
struct callchain_cursor_node *node;
|
2010-03-23 00:09:33 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
node = callchain_cursor_current(cursor);
|
|
|
|
if (!node)
|
2009-07-01 11:35:15 +08:00
|
|
|
break;
|
2010-03-23 00:09:33 +08:00
|
|
|
|
2014-01-14 23:37:15 +08:00
|
|
|
cmp = match_chain(node, cnode);
|
2016-02-16 22:08:22 +08:00
|
|
|
if (cmp != MATCH_EQ)
|
2009-06-26 22:28:00 +08:00
|
|
|
break;
|
2010-03-23 00:09:33 +08:00
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
found = true;
|
2011-01-14 11:51:58 +08:00
|
|
|
|
|
|
|
callchain_cursor_advance(cursor);
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
/* matches not, relay no the parent */
|
2011-01-14 11:51:58 +08:00
|
|
|
if (!found) {
|
2016-02-16 22:08:22 +08:00
|
|
|
WARN_ONCE(cmp == MATCH_ERROR, "Chain comparison error\n");
|
2014-01-14 23:37:15 +08:00
|
|
|
return cmp;
|
2011-01-14 11:51:58 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
matches = cursor->pos - start;
|
2009-06-26 22:28:00 +08:00
|
|
|
|
|
|
|
/* we match only a part of the node. Split it and add the new chain */
|
2011-01-14 11:51:58 +08:00
|
|
|
if (matches < root->val_nr) {
|
2016-02-16 22:08:23 +08:00
|
|
|
if (split_add_child(root, cursor, cnode, start, matches,
|
|
|
|
period) < 0)
|
|
|
|
return MATCH_ERROR;
|
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
return MATCH_EQ;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/* we match 100% of the path, increment the hit */
|
2011-01-14 11:51:58 +08:00
|
|
|
if (matches == root->val_nr && cursor->pos == cursor->nr) {
|
2010-07-08 09:41:46 +08:00
|
|
|
root->hit += period;
|
2015-11-09 13:45:40 +08:00
|
|
|
root->count++;
|
2016-02-16 22:08:22 +08:00
|
|
|
return MATCH_EQ;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2009-07-01 11:35:15 +08:00
|
|
|
/* We match the node and still have a part remaining */
|
2016-02-16 22:08:24 +08:00
|
|
|
if (append_chain_children(root, cursor, period) < 0)
|
|
|
|
return MATCH_ERROR;
|
2009-07-01 11:35:15 +08:00
|
|
|
|
2016-02-16 22:08:22 +08:00
|
|
|
return MATCH_EQ;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
int callchain_append(struct callchain_root *root,
|
|
|
|
struct callchain_cursor *cursor,
|
|
|
|
u64 period)
|
2009-06-26 22:28:00 +08:00
|
|
|
{
|
2011-01-14 11:51:58 +08:00
|
|
|
if (!cursor->nr)
|
2010-03-23 00:09:33 +08:00
|
|
|
return 0;
|
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
callchain_cursor_commit(cursor);
|
2010-03-23 00:09:33 +08:00
|
|
|
|
2016-02-16 22:08:24 +08:00
|
|
|
if (append_chain_children(&root->node, cursor, period) < 0)
|
|
|
|
return -1;
|
2010-08-23 02:05:22 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
if (cursor->nr > root->max_depth)
|
|
|
|
root->max_depth = cursor->nr;
|
2010-03-23 00:09:33 +08:00
|
|
|
|
|
|
|
return 0;
|
2009-06-26 22:28:00 +08:00
|
|
|
}
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
|
|
|
static int
|
2011-01-14 11:51:58 +08:00
|
|
|
merge_chain_branch(struct callchain_cursor *cursor,
|
|
|
|
struct callchain_node *dst, struct callchain_node *src)
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
{
|
2011-01-14 11:51:58 +08:00
|
|
|
struct callchain_cursor_node **old_last = cursor->last;
|
2013-10-11 13:15:36 +08:00
|
|
|
struct callchain_node *child;
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
struct callchain_list *list, *next_list;
|
2013-10-11 13:15:36 +08:00
|
|
|
struct rb_node *n;
|
2011-01-14 11:51:58 +08:00
|
|
|
int old_pos = cursor->nr;
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
int err = 0;
|
|
|
|
|
|
|
|
list_for_each_entry_safe(list, next_list, &src->val, list) {
|
2011-01-14 11:51:58 +08:00
|
|
|
callchain_cursor_append(cursor, list->ip,
|
perf report: Add branch flag to callchain cursor node
Since the branch ip has been added to call stack for easier browsing,
this patch adds more branch information. For example, add a flag to
indicate if this ip is a branch, and also add with the branch flag.
Then we can know if the cursor node represents a branch and know what
the branch flag it has.
The branch history code has a loop detection pass that removes loops. It
would be nice for knowing how many loops were removed then in next
steps, we can compute out the average number of iterations.
For example:
Before remove_loops(),
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x300, to = 0x250
entry3: from = 0x300, to = 0x250
entry4: from = 0x700, to = 0x800
After remove_loops()
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x700, to = 0x800
The original entry2 and entry3 are removed. So the number of iterations
(from = 0x300, to = 0x250) is equal to removed number + 1 (2 + 1).
iterations = removed number + 1;
average iteractions = Sum(iteractions) / number of samples
This formula ignores other cases, for example, iterations cross multiple
buffers and one buffer contains 2+ loops. Because in practice, it's good
enough.
Signed-off-by: Yao Jin <yao.jin@linux.intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Linux-kernel@vger.kernel.org
Cc: Yao Jin <yao.jin@linux.intel.com>
Link: http://lkml.kernel.org/n/1477876794-30749-2-git-send-email-yao.jin@linux.intel.com
[ Renamed 'iter' to 'nr_loop_iter' for clarity ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-10-31 09:19:49 +08:00
|
|
|
list->ms.map, list->ms.sym,
|
|
|
|
false, NULL, 0, 0);
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
list_del(&list->list);
|
2017-01-06 14:23:31 +08:00
|
|
|
map__zput(list->ms.map);
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
free(list);
|
|
|
|
}
|
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
if (src->hit) {
|
|
|
|
callchain_cursor_commit(cursor);
|
2016-02-16 22:08:24 +08:00
|
|
|
if (append_chain_children(dst, cursor, src->hit) < 0)
|
|
|
|
return -1;
|
2011-01-14 11:51:58 +08:00
|
|
|
}
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
2013-10-11 13:15:36 +08:00
|
|
|
n = rb_first(&src->rb_root_in);
|
|
|
|
while (n) {
|
|
|
|
child = container_of(n, struct callchain_node, rb_node_in);
|
|
|
|
n = rb_next(n);
|
|
|
|
rb_erase(&child->rb_node_in, &src->rb_root_in);
|
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
err = merge_chain_branch(cursor, dst, child);
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
if (err)
|
|
|
|
break;
|
|
|
|
|
|
|
|
free(child);
|
|
|
|
}
|
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
cursor->nr = old_pos;
|
|
|
|
cursor->last = old_last;
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
int callchain_merge(struct callchain_cursor *cursor,
|
|
|
|
struct callchain_root *dst, struct callchain_root *src)
|
|
|
|
{
|
|
|
|
return merge_chain_branch(cursor, &dst->node, &src->node);
|
|
|
|
}
|
|
|
|
|
|
|
|
int callchain_cursor_append(struct callchain_cursor *cursor,
|
perf report: Add branch flag to callchain cursor node
Since the branch ip has been added to call stack for easier browsing,
this patch adds more branch information. For example, add a flag to
indicate if this ip is a branch, and also add with the branch flag.
Then we can know if the cursor node represents a branch and know what
the branch flag it has.
The branch history code has a loop detection pass that removes loops. It
would be nice for knowing how many loops were removed then in next
steps, we can compute out the average number of iterations.
For example:
Before remove_loops(),
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x300, to = 0x250
entry3: from = 0x300, to = 0x250
entry4: from = 0x700, to = 0x800
After remove_loops()
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x700, to = 0x800
The original entry2 and entry3 are removed. So the number of iterations
(from = 0x300, to = 0x250) is equal to removed number + 1 (2 + 1).
iterations = removed number + 1;
average iteractions = Sum(iteractions) / number of samples
This formula ignores other cases, for example, iterations cross multiple
buffers and one buffer contains 2+ loops. Because in practice, it's good
enough.
Signed-off-by: Yao Jin <yao.jin@linux.intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Linux-kernel@vger.kernel.org
Cc: Yao Jin <yao.jin@linux.intel.com>
Link: http://lkml.kernel.org/n/1477876794-30749-2-git-send-email-yao.jin@linux.intel.com
[ Renamed 'iter' to 'nr_loop_iter' for clarity ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-10-31 09:19:49 +08:00
|
|
|
u64 ip, struct map *map, struct symbol *sym,
|
|
|
|
bool branch, struct branch_flags *flags,
|
|
|
|
int nr_loop_iter, int samples)
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
{
|
2011-01-14 11:51:58 +08:00
|
|
|
struct callchain_cursor_node *node = *cursor->last;
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
if (!node) {
|
2013-01-31 09:05:49 +08:00
|
|
|
node = calloc(1, sizeof(*node));
|
2011-01-14 11:51:58 +08:00
|
|
|
if (!node)
|
|
|
|
return -ENOMEM;
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
*cursor->last = node;
|
|
|
|
}
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
node->ip = ip;
|
2017-01-06 14:23:31 +08:00
|
|
|
map__zput(node->map);
|
|
|
|
node->map = map__get(map);
|
2011-01-14 11:51:58 +08:00
|
|
|
node->sym = sym;
|
perf report: Add branch flag to callchain cursor node
Since the branch ip has been added to call stack for easier browsing,
this patch adds more branch information. For example, add a flag to
indicate if this ip is a branch, and also add with the branch flag.
Then we can know if the cursor node represents a branch and know what
the branch flag it has.
The branch history code has a loop detection pass that removes loops. It
would be nice for knowing how many loops were removed then in next
steps, we can compute out the average number of iterations.
For example:
Before remove_loops(),
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x300, to = 0x250
entry3: from = 0x300, to = 0x250
entry4: from = 0x700, to = 0x800
After remove_loops()
entry0: from = 0x100, to = 0x200
entry1: from = 0x300, to = 0x250
entry2: from = 0x700, to = 0x800
The original entry2 and entry3 are removed. So the number of iterations
(from = 0x300, to = 0x250) is equal to removed number + 1 (2 + 1).
iterations = removed number + 1;
average iteractions = Sum(iteractions) / number of samples
This formula ignores other cases, for example, iterations cross multiple
buffers and one buffer contains 2+ loops. Because in practice, it's good
enough.
Signed-off-by: Yao Jin <yao.jin@linux.intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Linux-kernel@vger.kernel.org
Cc: Yao Jin <yao.jin@linux.intel.com>
Link: http://lkml.kernel.org/n/1477876794-30749-2-git-send-email-yao.jin@linux.intel.com
[ Renamed 'iter' to 'nr_loop_iter' for clarity ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2016-10-31 09:19:49 +08:00
|
|
|
node->branch = branch;
|
|
|
|
node->nr_loop_iter = nr_loop_iter;
|
|
|
|
node->samples = samples;
|
|
|
|
|
|
|
|
if (flags)
|
|
|
|
memcpy(&node->branch_flags, flags,
|
|
|
|
sizeof(struct branch_flags));
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
cursor->nr++;
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
|
2011-01-14 11:51:58 +08:00
|
|
|
cursor->last = &node->next;
|
|
|
|
|
|
|
|
return 0;
|
perf: Support for callchains merge
If we sort the histograms by comm, which is the default,
we need to merge some of them, typically different thread
histograms of a same process, or just same comm. But during
this merge, we forgot to merge callchains.
So imagine we have three threads (tids: 1000, 1001, 1002) that
belong to comm "foo".
tid 1000 got 100 events
tid 1001 got 10 events
tid 1002 got 3 events
Once we merge these histograms to get a per comm result, we'll
finally get:
"foo" got 113 events
The problem is if we merge 1000 and 1001 histograms into 1002, then
the end merge result, wrt callchains, will be only callchains that
belong to 1002.
This is because we haven't handled callchains in the merge. Only those
from one of the threads inside a common comm survive.
It means during this merge, we can lose a lot of callchains.
Fix this by implementing callchains merge and apply it on histograms
that collapse.
Reported-by: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
2010-08-23 03:10:35 +08:00
|
|
|
}
|
2014-01-14 13:25:35 +08:00
|
|
|
|
2016-04-15 01:48:07 +08:00
|
|
|
int sample__resolve_callchain(struct perf_sample *sample,
|
|
|
|
struct callchain_cursor *cursor, struct symbol **parent,
|
2014-01-14 13:25:35 +08:00
|
|
|
struct perf_evsel *evsel, struct addr_location *al,
|
|
|
|
int max_stack)
|
|
|
|
{
|
|
|
|
if (sample->callchain == NULL)
|
|
|
|
return 0;
|
|
|
|
|
2012-09-11 13:13:04 +08:00
|
|
|
if (symbol_conf.use_callchain || symbol_conf.cumulate_callchain ||
|
2016-05-03 19:54:43 +08:00
|
|
|
perf_hpp_list.parent) {
|
2016-04-15 01:48:07 +08:00
|
|
|
return thread__resolve_callchain(al->thread, cursor, evsel, sample,
|
2014-10-24 02:26:17 +08:00
|
|
|
parent, al, max_stack);
|
2014-01-14 13:25:35 +08:00
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int hist_entry__append_callchain(struct hist_entry *he, struct perf_sample *sample)
|
|
|
|
{
|
2014-07-14 18:02:35 +08:00
|
|
|
if (!symbol_conf.use_callchain || sample->callchain == NULL)
|
2014-01-14 13:25:35 +08:00
|
|
|
return 0;
|
|
|
|
return callchain_append(he->callchain, &callchain_cursor, sample->period);
|
|
|
|
}
|
2013-10-31 12:58:30 +08:00
|
|
|
|
|
|
|
int fill_callchain_info(struct addr_location *al, struct callchain_cursor_node *node,
|
|
|
|
bool hide_unresolved)
|
|
|
|
{
|
|
|
|
al->map = node->map;
|
|
|
|
al->sym = node->sym;
|
|
|
|
if (node->map)
|
|
|
|
al->addr = node->map->map_ip(node->map, node->ip);
|
|
|
|
else
|
|
|
|
al->addr = node->ip;
|
|
|
|
|
|
|
|
if (al->sym == NULL) {
|
|
|
|
if (hide_unresolved)
|
|
|
|
return 0;
|
|
|
|
if (al->map == NULL)
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (al->map->groups == &al->machine->kmaps) {
|
|
|
|
if (machine__is_host(al->machine)) {
|
|
|
|
al->cpumode = PERF_RECORD_MISC_KERNEL;
|
|
|
|
al->level = 'k';
|
|
|
|
} else {
|
|
|
|
al->cpumode = PERF_RECORD_MISC_GUEST_KERNEL;
|
|
|
|
al->level = 'g';
|
|
|
|
}
|
|
|
|
} else {
|
|
|
|
if (machine__is_host(al->machine)) {
|
|
|
|
al->cpumode = PERF_RECORD_MISC_USER;
|
|
|
|
al->level = '.';
|
|
|
|
} else if (perf_guest) {
|
|
|
|
al->cpumode = PERF_RECORD_MISC_GUEST_USER;
|
|
|
|
al->level = 'u';
|
|
|
|
} else {
|
|
|
|
al->cpumode = PERF_RECORD_MISC_HYPERVISOR;
|
|
|
|
al->level = 'H';
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
out:
|
|
|
|
return 1;
|
|
|
|
}
|
2014-11-13 10:05:23 +08:00
|
|
|
|
|
|
|
char *callchain_list__sym_name(struct callchain_list *cl,
|
|
|
|
char *bf, size_t bfsize, bool show_dso)
|
|
|
|
{
|
|
|
|
int printed;
|
|
|
|
|
|
|
|
if (cl->ms.sym) {
|
2014-11-13 10:05:24 +08:00
|
|
|
if (callchain_param.key == CCKEY_ADDRESS &&
|
|
|
|
cl->ms.map && !cl->srcline)
|
|
|
|
cl->srcline = get_srcline(cl->ms.map->dso,
|
|
|
|
map__rip_2objdump(cl->ms.map,
|
2014-11-13 10:05:27 +08:00
|
|
|
cl->ip),
|
|
|
|
cl->ms.sym, false);
|
2014-11-13 10:05:24 +08:00
|
|
|
if (cl->srcline)
|
|
|
|
printed = scnprintf(bf, bfsize, "%s %s",
|
|
|
|
cl->ms.sym->name, cl->srcline);
|
|
|
|
else
|
|
|
|
printed = scnprintf(bf, bfsize, "%s", cl->ms.sym->name);
|
2014-11-13 10:05:23 +08:00
|
|
|
} else
|
|
|
|
printed = scnprintf(bf, bfsize, "%#" PRIx64, cl->ip);
|
|
|
|
|
|
|
|
if (show_dso)
|
|
|
|
scnprintf(bf + printed, bfsize - printed, " %s",
|
|
|
|
cl->ms.map ?
|
|
|
|
cl->ms.map->dso->short_name :
|
|
|
|
"unknown");
|
|
|
|
|
|
|
|
return bf;
|
|
|
|
}
|
2014-12-30 13:38:13 +08:00
|
|
|
|
2015-11-09 13:45:39 +08:00
|
|
|
char *callchain_node__scnprintf_value(struct callchain_node *node,
|
|
|
|
char *bf, size_t bfsize, u64 total)
|
|
|
|
{
|
|
|
|
double percent = 0.0;
|
|
|
|
u64 period = callchain_cumul_hits(node);
|
2015-11-09 13:45:41 +08:00
|
|
|
unsigned count = callchain_cumul_counts(node);
|
2015-11-09 13:45:39 +08:00
|
|
|
|
2015-11-09 13:45:41 +08:00
|
|
|
if (callchain_param.mode == CHAIN_FOLDED) {
|
2015-11-09 13:45:39 +08:00
|
|
|
period = node->hit;
|
2015-11-09 13:45:41 +08:00
|
|
|
count = node->count;
|
|
|
|
}
|
2015-11-09 13:45:39 +08:00
|
|
|
|
2015-11-09 13:45:41 +08:00
|
|
|
switch (callchain_param.value) {
|
|
|
|
case CCVAL_PERIOD:
|
|
|
|
scnprintf(bf, bfsize, "%"PRIu64, period);
|
|
|
|
break;
|
|
|
|
case CCVAL_COUNT:
|
|
|
|
scnprintf(bf, bfsize, "%u", count);
|
|
|
|
break;
|
|
|
|
case CCVAL_PERCENT:
|
|
|
|
default:
|
|
|
|
if (total)
|
|
|
|
percent = period * 100.0 / total;
|
|
|
|
scnprintf(bf, bfsize, "%.2f%%", percent);
|
|
|
|
break;
|
|
|
|
}
|
2015-11-09 13:45:39 +08:00
|
|
|
return bf;
|
|
|
|
}
|
|
|
|
|
|
|
|
int callchain_node__fprintf_value(struct callchain_node *node,
|
|
|
|
FILE *fp, u64 total)
|
|
|
|
{
|
|
|
|
double percent = 0.0;
|
|
|
|
u64 period = callchain_cumul_hits(node);
|
2015-11-09 13:45:41 +08:00
|
|
|
unsigned count = callchain_cumul_counts(node);
|
2015-11-09 13:45:39 +08:00
|
|
|
|
2015-11-09 13:45:41 +08:00
|
|
|
if (callchain_param.mode == CHAIN_FOLDED) {
|
2015-11-09 13:45:39 +08:00
|
|
|
period = node->hit;
|
2015-11-09 13:45:41 +08:00
|
|
|
count = node->count;
|
|
|
|
}
|
2015-11-09 13:45:39 +08:00
|
|
|
|
2015-11-09 13:45:41 +08:00
|
|
|
switch (callchain_param.value) {
|
|
|
|
case CCVAL_PERIOD:
|
|
|
|
return fprintf(fp, "%"PRIu64, period);
|
|
|
|
case CCVAL_COUNT:
|
|
|
|
return fprintf(fp, "%u", count);
|
|
|
|
case CCVAL_PERCENT:
|
|
|
|
default:
|
|
|
|
if (total)
|
|
|
|
percent = period * 100.0 / total;
|
|
|
|
return percent_color_fprintf(fp, "%.2f%%", percent);
|
|
|
|
}
|
|
|
|
return 0;
|
2015-11-09 13:45:39 +08:00
|
|
|
}
|
|
|
|
|
2016-10-31 09:19:51 +08:00
|
|
|
static void callchain_counts_value(struct callchain_node *node,
|
|
|
|
u64 *branch_count, u64 *predicted_count,
|
|
|
|
u64 *abort_count, u64 *cycles_count)
|
|
|
|
{
|
|
|
|
struct callchain_list *clist;
|
|
|
|
|
|
|
|
list_for_each_entry(clist, &node->val, list) {
|
|
|
|
if (branch_count)
|
|
|
|
*branch_count += clist->branch_count;
|
|
|
|
|
|
|
|
if (predicted_count)
|
|
|
|
*predicted_count += clist->predicted_count;
|
|
|
|
|
|
|
|
if (abort_count)
|
|
|
|
*abort_count += clist->abort_count;
|
|
|
|
|
|
|
|
if (cycles_count)
|
|
|
|
*cycles_count += clist->cycles_count;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static int callchain_node_branch_counts_cumul(struct callchain_node *node,
|
|
|
|
u64 *branch_count,
|
|
|
|
u64 *predicted_count,
|
|
|
|
u64 *abort_count,
|
|
|
|
u64 *cycles_count)
|
|
|
|
{
|
|
|
|
struct callchain_node *child;
|
|
|
|
struct rb_node *n;
|
|
|
|
|
|
|
|
n = rb_first(&node->rb_root_in);
|
|
|
|
while (n) {
|
|
|
|
child = rb_entry(n, struct callchain_node, rb_node_in);
|
|
|
|
n = rb_next(n);
|
|
|
|
|
|
|
|
callchain_node_branch_counts_cumul(child, branch_count,
|
|
|
|
predicted_count,
|
|
|
|
abort_count,
|
|
|
|
cycles_count);
|
|
|
|
|
|
|
|
callchain_counts_value(child, branch_count,
|
|
|
|
predicted_count, abort_count,
|
|
|
|
cycles_count);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
int callchain_branch_counts(struct callchain_root *root,
|
|
|
|
u64 *branch_count, u64 *predicted_count,
|
|
|
|
u64 *abort_count, u64 *cycles_count)
|
|
|
|
{
|
|
|
|
if (branch_count)
|
|
|
|
*branch_count = 0;
|
|
|
|
|
|
|
|
if (predicted_count)
|
|
|
|
*predicted_count = 0;
|
|
|
|
|
|
|
|
if (abort_count)
|
|
|
|
*abort_count = 0;
|
|
|
|
|
|
|
|
if (cycles_count)
|
|
|
|
*cycles_count = 0;
|
|
|
|
|
|
|
|
return callchain_node_branch_counts_cumul(&root->node,
|
|
|
|
branch_count,
|
|
|
|
predicted_count,
|
|
|
|
abort_count,
|
|
|
|
cycles_count);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int callchain_counts_printf(FILE *fp, char *bf, int bfsize,
|
|
|
|
u64 branch_count, u64 predicted_count,
|
|
|
|
u64 abort_count, u64 cycles_count,
|
|
|
|
u64 iter_count, u64 samples_count)
|
|
|
|
{
|
|
|
|
double predicted_percent = 0.0;
|
|
|
|
const char *null_str = "";
|
|
|
|
char iter_str[32];
|
|
|
|
char *str;
|
|
|
|
u64 cycles = 0;
|
|
|
|
|
|
|
|
if (branch_count == 0) {
|
|
|
|
if (fp)
|
|
|
|
return fprintf(fp, " (calltrace)");
|
|
|
|
|
|
|
|
return scnprintf(bf, bfsize, " (calltrace)");
|
|
|
|
}
|
|
|
|
|
|
|
|
if (iter_count && samples_count) {
|
|
|
|
scnprintf(iter_str, sizeof(iter_str),
|
|
|
|
", iterations:%" PRId64 "",
|
|
|
|
iter_count / samples_count);
|
|
|
|
str = iter_str;
|
|
|
|
} else
|
|
|
|
str = (char *)null_str;
|
|
|
|
|
|
|
|
predicted_percent = predicted_count * 100.0 / branch_count;
|
|
|
|
cycles = cycles_count / branch_count;
|
|
|
|
|
|
|
|
if ((predicted_percent >= 100.0) && (abort_count == 0)) {
|
|
|
|
if (fp)
|
|
|
|
return fprintf(fp, " (cycles:%" PRId64 "%s)",
|
|
|
|
cycles, str);
|
|
|
|
|
|
|
|
return scnprintf(bf, bfsize, " (cycles:%" PRId64 "%s)",
|
|
|
|
cycles, str);
|
|
|
|
}
|
|
|
|
|
|
|
|
if ((predicted_percent < 100.0) && (abort_count == 0)) {
|
|
|
|
if (fp)
|
|
|
|
return fprintf(fp,
|
|
|
|
" (predicted:%.1f%%, cycles:%" PRId64 "%s)",
|
|
|
|
predicted_percent, cycles, str);
|
|
|
|
|
|
|
|
return scnprintf(bf, bfsize,
|
|
|
|
" (predicted:%.1f%%, cycles:%" PRId64 "%s)",
|
|
|
|
predicted_percent, cycles, str);
|
|
|
|
}
|
|
|
|
|
|
|
|
if (fp)
|
|
|
|
return fprintf(fp,
|
|
|
|
" (predicted:%.1f%%, abort:%" PRId64 ", cycles:%" PRId64 "%s)",
|
|
|
|
predicted_percent, abort_count, cycles, str);
|
|
|
|
|
|
|
|
return scnprintf(bf, bfsize,
|
|
|
|
" (predicted:%.1f%%, abort:%" PRId64 ", cycles:%" PRId64 "%s)",
|
|
|
|
predicted_percent, abort_count, cycles, str);
|
|
|
|
}
|
|
|
|
|
|
|
|
int callchain_list_counts__printf_value(struct callchain_node *node,
|
|
|
|
struct callchain_list *clist,
|
|
|
|
FILE *fp, char *bf, int bfsize)
|
|
|
|
{
|
|
|
|
u64 branch_count, predicted_count;
|
|
|
|
u64 abort_count, cycles_count;
|
|
|
|
u64 iter_count = 0, samples_count = 0;
|
|
|
|
|
|
|
|
branch_count = clist->branch_count;
|
|
|
|
predicted_count = clist->predicted_count;
|
|
|
|
abort_count = clist->abort_count;
|
|
|
|
cycles_count = clist->cycles_count;
|
|
|
|
|
|
|
|
if (node) {
|
|
|
|
struct callchain_list *call;
|
|
|
|
|
|
|
|
list_for_each_entry(call, &node->val, list) {
|
|
|
|
iter_count += call->iter_count;
|
|
|
|
samples_count += call->samples_count;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return callchain_counts_printf(fp, bf, bfsize, branch_count,
|
|
|
|
predicted_count, abort_count,
|
|
|
|
cycles_count, iter_count, samples_count);
|
|
|
|
}
|
|
|
|
|
2014-12-30 13:38:13 +08:00
|
|
|
static void free_callchain_node(struct callchain_node *node)
|
|
|
|
{
|
|
|
|
struct callchain_list *list, *tmp;
|
|
|
|
struct callchain_node *child;
|
|
|
|
struct rb_node *n;
|
|
|
|
|
2015-11-09 13:45:43 +08:00
|
|
|
list_for_each_entry_safe(list, tmp, &node->parent_val, list) {
|
|
|
|
list_del(&list->list);
|
2017-01-06 14:23:31 +08:00
|
|
|
map__zput(list->ms.map);
|
2015-11-09 13:45:43 +08:00
|
|
|
free(list);
|
|
|
|
}
|
|
|
|
|
2014-12-30 13:38:13 +08:00
|
|
|
list_for_each_entry_safe(list, tmp, &node->val, list) {
|
|
|
|
list_del(&list->list);
|
2017-01-06 14:23:31 +08:00
|
|
|
map__zput(list->ms.map);
|
2014-12-30 13:38:13 +08:00
|
|
|
free(list);
|
|
|
|
}
|
|
|
|
|
|
|
|
n = rb_first(&node->rb_root_in);
|
|
|
|
while (n) {
|
|
|
|
child = container_of(n, struct callchain_node, rb_node_in);
|
|
|
|
n = rb_next(n);
|
|
|
|
rb_erase(&child->rb_node_in, &node->rb_root_in);
|
|
|
|
|
|
|
|
free_callchain_node(child);
|
|
|
|
free(child);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void free_callchain(struct callchain_root *root)
|
|
|
|
{
|
|
|
|
if (!symbol_conf.use_callchain)
|
|
|
|
return;
|
|
|
|
|
|
|
|
free_callchain_node(&root->node);
|
|
|
|
}
|
2015-11-09 13:45:43 +08:00
|
|
|
|
2016-01-05 11:06:00 +08:00
|
|
|
static u64 decay_callchain_node(struct callchain_node *node)
|
|
|
|
{
|
|
|
|
struct callchain_node *child;
|
|
|
|
struct rb_node *n;
|
|
|
|
u64 child_hits = 0;
|
|
|
|
|
|
|
|
n = rb_first(&node->rb_root_in);
|
|
|
|
while (n) {
|
|
|
|
child = container_of(n, struct callchain_node, rb_node_in);
|
|
|
|
|
|
|
|
child_hits += decay_callchain_node(child);
|
|
|
|
n = rb_next(n);
|
|
|
|
}
|
|
|
|
|
|
|
|
node->hit = (node->hit * 7) / 8;
|
|
|
|
node->children_hit = child_hits;
|
|
|
|
|
|
|
|
return node->hit;
|
|
|
|
}
|
|
|
|
|
|
|
|
void decay_callchain(struct callchain_root *root)
|
|
|
|
{
|
|
|
|
if (!symbol_conf.use_callchain)
|
|
|
|
return;
|
|
|
|
|
|
|
|
decay_callchain_node(&root->node);
|
|
|
|
}
|
|
|
|
|
2015-11-09 13:45:43 +08:00
|
|
|
int callchain_node__make_parent_list(struct callchain_node *node)
|
|
|
|
{
|
|
|
|
struct callchain_node *parent = node->parent;
|
|
|
|
struct callchain_list *chain, *new;
|
|
|
|
LIST_HEAD(head);
|
|
|
|
|
|
|
|
while (parent) {
|
|
|
|
list_for_each_entry_reverse(chain, &parent->val, list) {
|
|
|
|
new = malloc(sizeof(*new));
|
|
|
|
if (new == NULL)
|
|
|
|
goto out;
|
|
|
|
*new = *chain;
|
|
|
|
new->has_children = false;
|
2017-01-06 14:23:31 +08:00
|
|
|
map__get(new->ms.map);
|
2015-11-09 13:45:43 +08:00
|
|
|
list_add_tail(&new->list, &head);
|
|
|
|
}
|
|
|
|
parent = parent->parent;
|
|
|
|
}
|
|
|
|
|
|
|
|
list_for_each_entry_safe_reverse(chain, new, &head, list)
|
|
|
|
list_move_tail(&chain->list, &node->parent_val);
|
|
|
|
|
|
|
|
if (!list_empty(&node->parent_val)) {
|
|
|
|
chain = list_first_entry(&node->parent_val, struct callchain_list, list);
|
|
|
|
chain->has_children = rb_prev(&node->rb_node) || rb_next(&node->rb_node);
|
|
|
|
|
|
|
|
chain = list_first_entry(&node->val, struct callchain_list, list);
|
|
|
|
chain->has_children = false;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
out:
|
|
|
|
list_for_each_entry_safe(chain, new, &head, list) {
|
|
|
|
list_del(&chain->list);
|
2017-01-06 14:23:31 +08:00
|
|
|
map__zput(chain->ms.map);
|
2015-11-09 13:45:43 +08:00
|
|
|
free(chain);
|
|
|
|
}
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
2016-12-06 11:40:02 +08:00
|
|
|
|
|
|
|
int callchain_cursor__copy(struct callchain_cursor *dst,
|
|
|
|
struct callchain_cursor *src)
|
|
|
|
{
|
|
|
|
int rc = 0;
|
|
|
|
|
|
|
|
callchain_cursor_reset(dst);
|
|
|
|
callchain_cursor_commit(src);
|
|
|
|
|
|
|
|
while (true) {
|
|
|
|
struct callchain_cursor_node *node;
|
|
|
|
|
|
|
|
node = callchain_cursor_current(src);
|
|
|
|
if (node == NULL)
|
|
|
|
break;
|
|
|
|
|
|
|
|
rc = callchain_cursor_append(dst, node->ip, node->map, node->sym,
|
|
|
|
node->branch, &node->branch_flags,
|
|
|
|
node->nr_loop_iter, node->samples);
|
|
|
|
if (rc)
|
|
|
|
break;
|
|
|
|
|
|
|
|
callchain_cursor_advance(src);
|
|
|
|
}
|
|
|
|
|
|
|
|
return rc;
|
|
|
|
}
|