From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: Clark Williams <williams@redhat.com>,
linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
Jin Yao <yao.jin@linux.intel.com>,
Alexander Shishkin <alexander.shishkin@linux.intel.com>,
Andi Kleen <ak@linux.intel.com>,
Kan Liang <kan.liang@linux.intel.com>,
Peter Zijlstra <peterz@infradead.org>,
Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 25/75] perf report: Display average IPC and IPC coverage per symbol
Date: Thu, 6 Dec 2018 18:25:12 -0300 [thread overview]
Message-ID: <20181206212602.20474-26-acme@kernel.org> (raw)
In-Reply-To: <20181206212602.20474-1-acme@kernel.org>
From: Jin Yao <yao.jin@linux.intel.com>
Support displaying the average IPC and IPC coverage per symbol in 'perf
report' --tui and --stdio modes.
For example,
$ perf record -b ...
$ perf report -s symbol
Overhead Symbol IPC [IPC Coverage]
39.60% [.] __random 2.30 [ 54.8%]
18.02% [.] main 0.43 [ 54.3%]
14.21% [.] compute_flag 2.29 [100.0%]
14.16% [.] rand 0.36 [100.0%]
7.06% [.] __random_r 2.57 [ 70.5%]
6.85% [.] rand@plt 0.00 [ 0.0%]
Jiri Olsa <jolsa@redhat.com> provided the patch to support the --stdio
mode. I merged Jiri's code in this patch.
$ perf report -s symbol --stdio
# Overhead Symbol IPC [IPC Coverage]
# ........ ........................... ....................
#
39.60% [.] __random 2.30 [ 54.8%]
18.02% [.] main 0.43 [ 54.3%]
14.21% [.] compute_flag 2.29 [100.0%]
14.16% [.] rand 0.36 [100.0%]
7.06% [.] __random_r 2.57 [ 70.5%]
6.85% [.] rand@plt 0.00 [ 0.0%]
0.02% [k] run_timer_softirq 1.60 [ 57.2%]
The columns "IPC" and "[IPC Coverage]" are automatically enabled when
the sort-key "symbol" is specified. If the perf.data file doesn't
contain timed LBR information, columns are filled with "-".
For example,
# Overhead Symbol IPC [IPC Coverage]
# ........ ........................... ....................
#
46.57% [.] main - -
17.60% [.] rand - -
15.84% [.] __random_r - -
11.90% [.] __random - -
6.50% [.] compute_flag - -
1.59% [.] rand@plt - -
0.00% [.] _dl_relocate_object - -
0.00% [k] tlb_flush_mmu - -
0.00% [k] perf_event_mmap - -
0.00% [k] native_sched_clock - -
0.00% [k] intel_pmu_handle_irq_v4 - -
0.00% [k] native_write_msr - -
v3:
---
Removed the sortkey 'ipc' from command-line. The columns "IPC"
and "[IPC Coverage]" are automatically enabled when "symbol"
is specified.
v2:
---
Merge in Jiri's patch to support stdio mode
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1543586097-27632-4-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/builtin-report.c | 26 ++++++++++++++--
tools/perf/util/hist.h | 1 +
tools/perf/util/sort.c | 61 +++++++++++++++++++++++++++++++++++++
tools/perf/util/sort.h | 2 ++
4 files changed, 87 insertions(+), 3 deletions(-)
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 257c9c18cb7e..4958095be4fc 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -85,6 +85,7 @@ struct report {
int socket_filter;
DECLARE_BITMAP(cpu_bitmap, MAX_NR_CPUS);
struct branch_type_stat brtype_stat;
+ bool symbol_ipc;
};
static int report__config(const char *var, const char *value, void *cb)
@@ -129,7 +130,7 @@ static int hist_iter__report_callback(struct hist_entry_iter *iter,
struct mem_info *mi;
struct branch_info *bi;
- if (!ui__has_annotation())
+ if (!ui__has_annotation() && !rep->symbol_ipc)
return 0;
hist__account_cycles(sample->branch_stack, al, sample,
@@ -174,7 +175,7 @@ static int hist_iter__branch_callback(struct hist_entry_iter *iter,
struct perf_evsel *evsel = iter->evsel;
int err;
- if (!ui__has_annotation())
+ if (!ui__has_annotation() && !rep->symbol_ipc)
return 0;
hist__account_cycles(sample->branch_stack, al, sample,
@@ -1133,6 +1134,7 @@ int cmd_report(int argc, const char **argv)
.mode = PERF_DATA_MODE_READ,
};
int ret = hists__init();
+ char sort_tmp[128];
if (ret < 0)
return ret;
@@ -1284,6 +1286,24 @@ int cmd_report(int argc, const char **argv)
else
use_browser = 0;
+ if (sort_order && strstr(sort_order, "ipc")) {
+ parse_options_usage(report_usage, options, "s", 1);
+ goto error;
+ }
+
+ if (sort_order && strstr(sort_order, "symbol")) {
+ if (sort__mode == SORT_MODE__BRANCH) {
+ snprintf(sort_tmp, sizeof(sort_tmp), "%s,%s",
+ sort_order, "ipc_lbr");
+ report.symbol_ipc = true;
+ } else {
+ snprintf(sort_tmp, sizeof(sort_tmp), "%s,%s",
+ sort_order, "ipc_null");
+ }
+
+ sort_order = sort_tmp;
+ }
+
if (setup_sorting(session->evlist) < 0) {
if (sort_order)
parse_options_usage(report_usage, options, "s", 1);
@@ -1311,7 +1331,7 @@ int cmd_report(int argc, const char **argv)
* so don't allocate extra space that won't be used in the stdio
* implementation.
*/
- if (ui__has_annotation()) {
+ if (ui__has_annotation() || report.symbol_ipc) {
ret = symbol__annotation_init();
if (ret < 0)
goto error;
diff --git a/tools/perf/util/hist.h b/tools/perf/util/hist.h
index 3badd7f1e1b8..664b5eda8d51 100644
--- a/tools/perf/util/hist.h
+++ b/tools/perf/util/hist.h
@@ -62,6 +62,7 @@ enum hist_column {
HISTC_TRACE,
HISTC_SYM_SIZE,
HISTC_DSO_SIZE,
+ HISTC_SYMBOL_IPC,
HISTC_NR_COLS, /* Last entry */
};
diff --git a/tools/perf/util/sort.c b/tools/perf/util/sort.c
index f96c005b3c41..047793528919 100644
--- a/tools/perf/util/sort.c
+++ b/tools/perf/util/sort.c
@@ -13,6 +13,7 @@
#include "strlist.h"
#include <traceevent/event-parse.h>
#include "mem-events.h"
+#include "annotate.h"
#include <linux/kernel.h>
regex_t parent_regex;
@@ -422,6 +423,64 @@ struct sort_entry sort_srcline_to = {
.se_width_idx = HISTC_SRCLINE_TO,
};
+static int hist_entry__sym_ipc_snprintf(struct hist_entry *he, char *bf,
+ size_t size, unsigned int width)
+{
+
+ struct symbol *sym = he->ms.sym;
+ struct map *map = he->ms.map;
+ struct perf_evsel *evsel = hists_to_evsel(he->hists);
+ struct annotation *notes;
+ double ipc = 0.0, coverage = 0.0;
+ char tmp[64];
+
+ if (!sym)
+ return repsep_snprintf(bf, size, "%-*s", width, "-");
+
+ if (!sym->annotate2 && symbol__annotate2(sym, map, evsel,
+ &annotation__default_options, NULL) < 0) {
+ return 0;
+ }
+
+ notes = symbol__annotation(sym);
+
+ if (notes->hit_cycles)
+ ipc = notes->hit_insn / ((double)notes->hit_cycles);
+
+ if (notes->total_insn) {
+ coverage = notes->cover_insn * 100.0 /
+ ((double)notes->total_insn);
+ }
+
+ snprintf(tmp, sizeof(tmp), "%-5.2f [%5.1f%%]", ipc, coverage);
+ return repsep_snprintf(bf, size, "%-*s", width, tmp);
+}
+
+struct sort_entry sort_sym_ipc = {
+ .se_header = "IPC [IPC Coverage]",
+ .se_cmp = sort__sym_cmp,
+ .se_snprintf = hist_entry__sym_ipc_snprintf,
+ .se_width_idx = HISTC_SYMBOL_IPC,
+};
+
+static int hist_entry__sym_ipc_null_snprintf(struct hist_entry *he
+ __maybe_unused,
+ char *bf, size_t size,
+ unsigned int width)
+{
+ char tmp[64];
+
+ snprintf(tmp, sizeof(tmp), "%-5s %2s", "-", "-");
+ return repsep_snprintf(bf, size, "%-*s", width, tmp);
+}
+
+struct sort_entry sort_sym_ipc_null = {
+ .se_header = "IPC [IPC Coverage]",
+ .se_cmp = sort__sym_cmp,
+ .se_snprintf = hist_entry__sym_ipc_null_snprintf,
+ .se_width_idx = HISTC_SYMBOL_IPC,
+};
+
/* --sort srcfile */
static char no_srcfile[1];
@@ -1574,6 +1633,7 @@ static struct sort_dimension common_sort_dimensions[] = {
DIM(SORT_SYM_SIZE, "symbol_size", sort_sym_size),
DIM(SORT_DSO_SIZE, "dso_size", sort_dso_size),
DIM(SORT_CGROUP_ID, "cgroup_id", sort_cgroup_id),
+ DIM(SORT_SYM_IPC_NULL, "ipc_null", sort_sym_ipc_null),
};
#undef DIM
@@ -1591,6 +1651,7 @@ static struct sort_dimension bstack_sort_dimensions[] = {
DIM(SORT_CYCLES, "cycles", sort_cycles),
DIM(SORT_SRCLINE_FROM, "srcline_from", sort_srcline_from),
DIM(SORT_SRCLINE_TO, "srcline_to", sort_srcline_to),
+ DIM(SORT_SYM_IPC, "ipc_lbr", sort_sym_ipc),
};
#undef DIM
diff --git a/tools/perf/util/sort.h b/tools/perf/util/sort.h
index a97cf8e6be86..130fe37fe2df 100644
--- a/tools/perf/util/sort.h
+++ b/tools/perf/util/sort.h
@@ -229,6 +229,7 @@ enum sort_type {
SORT_SYM_SIZE,
SORT_DSO_SIZE,
SORT_CGROUP_ID,
+ SORT_SYM_IPC_NULL,
/* branch stack specific sort keys */
__SORT_BRANCH_STACK,
@@ -242,6 +243,7 @@ enum sort_type {
SORT_CYCLES,
SORT_SRCLINE_FROM,
SORT_SRCLINE_TO,
+ SORT_SYM_IPC,
/* memory mode specific sort keys */
__SORT_MEMORY_MODE,
--
2.19.2
next prev parent reply other threads:[~2018-12-06 21:25 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-06 21:24 [GIT PULL 00/75] perf/core improvements and fixes Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 01/75] perf build: Give better hint about devel package for libssl Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 02/75] perf stat: Fix shadow stats for clock events Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 03/75] perf stat: Fix CSV mode column output for non-cgroup events Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 04/75] perf map: Remove extra indirection from map__find() Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 05/75] perf env: Also consider env->arch == NULL as local operation Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 06/75] perf machine: Record if a arch has a single user/kernel address space Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 07/75] perf thread: Add fallback functions for cases where cpumode is insufficient Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 08/75] perf tools: Use fallback for sample_addr_correlates_sym() cases Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 09/75] perf script: Use fallbacks for branch stacks Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 10/75] tools lib traceevent: Fix compile warnings in tools/lib/traceevent/event-parse.c Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 11/75] perf tests record: Allow for 'sleep' being 'coreutils' Arnaldo Carvalho de Melo
2018-12-06 21:24 ` [PATCH 12/75] perf test: Fix perf_event_attr test failure Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 13/75] tools include: Adopt ERR_CAST() from the kernel err.h header Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 14/75] perf bpf: Use ERR_CAST instead of ERR_PTR(PTR_ERR()) Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 15/75] perf top: Allow passing a kallsyms file Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 16/75] perf intel-pt: Fix error with config term "pt=0" Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 17/75] tools build feature: Check if libaio is available Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 18/75] perf mmap: Map data buffer for preserving collected data Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 19/75] perf record: Enable asynchronous trace writing Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 20/75] perf record: Extend trace writing to multi AIO Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 21/75] perf beauty mmap_flags: Check if the arch has a mmap.h file Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 22/75] tools lib traceevent: Add sanity check to is_timestamp_in_us() Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 23/75] perf annotate: Compute average IPC and IPC coverage per symbol Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 24/75] perf annotate: Create a annotate2 flag in struct symbol Arnaldo Carvalho de Melo
2018-12-06 21:25 ` Arnaldo Carvalho de Melo [this message]
2018-12-06 21:25 ` [PATCH 26/75] perf report: Documentation average IPC and IPC coverage Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 27/75] tools lib traceevent: Implement new API tep_get_ref() Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 28/75] tools lib traceevent: Added support for pkg-config Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 29/75] tools lib traceevent: Install trace-seq.h API header file Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 30/75] tools lib traceevent, perf tools: Rename 'struct tep_event_format' to 'struct tep_event' Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 31/75] tools lib traceevent: Rename tep_free_format() to tep_free_event() Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 32/75] perf tools: traceevent API cleanup, remove __tep_data2host*() Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 33/75] tools lib traceevent: traceevent API cleanup Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 34/75] perf beauty mmap_flags: Fixed syntax error Fixed missing ']' error Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 35/75] perf cs-etm: Support for ARM A32/T32 instruction sets in CoreSight trace Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 36/75] perf tests ARM: Disable breakpoint tests 32-bit Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 37/75] perf vendor events intel: Fix diverse typos Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 38/75] tools lib traceevent: Fix diverse typos in comments Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 39/75] perf tools Documentation: Fix diverse typos Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 40/75] perf bpf-loader: Fix debugging message typo Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 41/75] perf tools: Fix diverse comment typos Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 42/75] tools lib subcmd: Fix a few source code " Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 43/75] perf tools: Allow specifying proc-map-timeout in config file Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 44/75] perf trace: We need to consider "nr" if "__syscall_nr" is not there Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 45/75] perf tools: Support 'srccode' output Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 46/75] perf ordered_events: Rework show_progress for __ordered_events__flush Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 47/75] perf ordered_events: Add private data member Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 48/75] perf top: Save and display the lost count stats Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 49/75] perf top: Move lost events warning to helpline Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 50/75] perf top: Add processing thread Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 51/75] perf top: Use cond variable instead of a lock Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 52/75] perf top: Set the 'session_done' volatile variable when exiting Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 53/75] perf top: Drop samples which are behind the refresh rate Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 54/75] perf top: Save and display the drop count stats Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 55/75] perf top: Display slow reader warning when droping samples Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 56/75] perf top: Move perf_top__reset_sample_counters() to after counts display Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 57/75] perf cs-etm: Add configuration for ETMv3 trace protocol Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 58/75] perf cs-etm: Add support for ETMv3 trace decoding Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 59/75] perf cs-etm: Add support for PTMv1.1 decoding Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 60/75] perf dso: Fix unchecked usage of strncpy() Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 61/75] perf header: " Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 62/75] " Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 63/75] perf help: Remove needless use " Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 64/75] perf svghelper: Fix unchecked usage " Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 65/75] perf ui helpline: Use strlcpy() as a shorter form of strncpy() + explicit set nul Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 66/75] perf probe: Fix unchecked usage of strncpy() Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 67/75] perf parse-events: " Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 68/75] perf vendor events intel: Fix Load_Miss_Real_Latency on SKL/SKX Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 69/75] perf record: Fix memory leak on AIO objects deallocation Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 70/75] perf config: Modify size factor of snprintf Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 71/75] perf annotate: Introduce basic support for ARC Arnaldo Carvalho de Melo
2018-12-06 21:25 ` [PATCH 72/75] perf ordered_events: Add ordered_events__flush_time interface Arnaldo Carvalho de Melo
2018-12-06 21:26 ` [PATCH 73/75] perf trace: Move event delivery to a new deliver_event() function Arnaldo Carvalho de Melo
2018-12-06 21:26 ` [PATCH 74/75] perf ordered_events: Add first_time() method Arnaldo Carvalho de Melo
2018-12-06 21:26 ` [PATCH 75/75] perf trace: Add ordered processing Arnaldo Carvalho de Melo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20181206212602.20474-26-acme@kernel.org \
--to=acme@kernel.org \
--cc=acme@redhat.com \
--cc=ak@linux.intel.com \
--cc=alexander.shishkin@linux.intel.com \
--cc=kan.liang@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-perf-users@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=williams@redhat.com \
--cc=yao.jin@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).