* Re: [PATCH v3] perf annotate/report: Remove hist__account_cycles from callback
2019-03-15 21:16 [PATCH v3] perf annotate/report: Remove hist__account_cycles from callback Jin Yao
@ 2019-03-20 0:35 ` Jin, Yao
2019-03-20 1:47 ` Andi Kleen
2019-04-10 13:00 ` Arnaldo Carvalho de Melo
2019-05-18 8:46 ` [tip:perf/core] perf annotate: Remove hist__account_cycles() " tip-bot for Jin Yao
2 siblings, 1 reply; 5+ messages in thread
From: Jin, Yao @ 2019-03-20 0:35 UTC (permalink / raw)
To: acme, jolsa, peterz, mingo, alexander.shishkin
Cc: Linux-kernel, ak, kan.liang, yao.jin
Nobody like this patch? :)
In my test, for a 72MB perf.data (with LBR data), the perf annotate
processing speed:
Before: 10.2s
After: 2.2s
Thanks
Jin Yao
On 3/16/2019 5:16 AM, Jin Yao wrote:
> The hist__account_cycles is executed when the hist_iter__branch_callback
> is called. But it looks it's not necessary. In hist__account_cycles, it
> already walks on all branch entries.
>
> This patch moves the hist__account_cycles out of callback, now the data
> processing is much faster than before.
>
> Previous code has an issue that the ch[offset].num++
> (in __symbol__account_cycles) is executed repeatedly since
> hist__account_cycles is called in each hist_iter__branch_callback,
> so the counting of ch[offset].num is not correct (too big).
> With this patch, the issue is fixed. And we don't need the code of
> "ch->reset >= ch->num / 2" to check if there are too many overlaps
> (in annotation__count_and_fill), otherwise some data would be
> hidden.
>
> Now, we can try, for example:
>
> perf record -b ...
> perf annotate or perf report -s symbol
>
> The before/after output should be no change.
>
> v3:
> ---
> Fix the crash in stdio mode.
> Like previous code, it needs the checking of ui__has_annotation()
> before hist__account_cycles()
>
> v2:
> ---
> 1. Cover the similar perf report
> 2. Remove the checking code "ch->reset >= ch->num / 2"
>
> Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
> ---
> tools/perf/builtin-annotate.c | 4 ++--
> tools/perf/builtin-report.c | 11 +++++------
> tools/perf/util/annotate.c | 2 +-
> 3 files changed, 8 insertions(+), 9 deletions(-)
>
> diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
> index 67f9d9f..77deb3a 100644
> --- a/tools/perf/builtin-annotate.c
> +++ b/tools/perf/builtin-annotate.c
> @@ -159,8 +159,6 @@ static int hist_iter__branch_callback(struct hist_entry_iter *iter,
> struct perf_evsel *evsel = iter->evsel;
> int err;
>
> - hist__account_cycles(sample->branch_stack, al, sample, false);
> -
> bi = he->branch_info;
> err = addr_map_symbol__inc_samples(&bi->from, sample, evsel);
>
> @@ -199,6 +197,8 @@ static int process_branch_callback(struct perf_evsel *evsel,
> if (a.map != NULL)
> a.map->dso->hit = 1;
>
> + hist__account_cycles(sample->branch_stack, al, sample, false);
> +
> ret = hist_entry_iter__add(&iter, &a, PERF_MAX_STACK_DEPTH, ann);
> return ret;
> }
> diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
> index 1921aaa..493e115 100644
> --- a/tools/perf/builtin-report.c
> +++ b/tools/perf/builtin-report.c
> @@ -136,9 +136,6 @@ static int hist_iter__report_callback(struct hist_entry_iter *iter,
> if (!ui__has_annotation() && !rep->symbol_ipc)
> return 0;
>
> - hist__account_cycles(sample->branch_stack, al, sample,
> - rep->nonany_branch_mode);
> -
> if (sort__mode == SORT_MODE__BRANCH) {
> bi = he->branch_info;
> err = addr_map_symbol__inc_samples(&bi->from, sample, evsel);
> @@ -181,9 +178,6 @@ static int hist_iter__branch_callback(struct hist_entry_iter *iter,
> if (!ui__has_annotation() && !rep->symbol_ipc)
> return 0;
>
> - hist__account_cycles(sample->branch_stack, al, sample,
> - rep->nonany_branch_mode);
> -
> bi = he->branch_info;
> err = addr_map_symbol__inc_samples(&bi->from, sample, evsel);
> if (err)
> @@ -282,6 +276,11 @@ static int process_sample_event(struct perf_tool *tool,
> if (al.map != NULL)
> al.map->dso->hit = 1;
>
> + if (ui__has_annotation() || rep->symbol_ipc) {
> + hist__account_cycles(sample->branch_stack, &al, sample,
> + rep->nonany_branch_mode);
> + }
> +
> ret = hist_entry_iter__add(&iter, &al, rep->max_stack, rep);
> if (ret < 0)
> pr_debug("problem adding hist entry, skipping event\n");
> diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
> index 5f6dbbf..e8080c0 100644
> --- a/tools/perf/util/annotate.c
> +++ b/tools/perf/util/annotate.c
> @@ -1015,7 +1015,7 @@ static void annotation__count_and_fill(struct annotation *notes, u64 start, u64
> float ipc = n_insn / ((double)ch->cycles / (double)ch->num);
>
> /* Hide data when there are too many overlaps. */
> - if (ch->reset >= 0x7fff || ch->reset >= ch->num / 2)
> + if (ch->reset >= 0x7fff)
> return;
>
> for (offset = start; offset <= end; offset++) {
>
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [PATCH v3] perf annotate/report: Remove hist__account_cycles from callback
2019-03-20 0:35 ` Jin, Yao
@ 2019-03-20 1:47 ` Andi Kleen
0 siblings, 0 replies; 5+ messages in thread
From: Andi Kleen @ 2019-03-20 1:47 UTC (permalink / raw)
To: Jin, Yao
Cc: acme, jolsa, peterz, mingo, alexander.shishkin, Linux-kernel,
kan.liang, yao.jin
On Wed, Mar 20, 2019 at 08:35:17AM +0800, Jin, Yao wrote:
> Nobody like this patch? :)
>
> In my test, for a 72MB perf.data (with LBR data), the perf annotate
> processing speed:
>
> Before: 10.2s
> After: 2.2s
That's great!
-Andi
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH v3] perf annotate/report: Remove hist__account_cycles from callback
2019-03-15 21:16 [PATCH v3] perf annotate/report: Remove hist__account_cycles from callback Jin Yao
2019-03-20 0:35 ` Jin, Yao
@ 2019-04-10 13:00 ` Arnaldo Carvalho de Melo
2019-05-18 8:46 ` [tip:perf/core] perf annotate: Remove hist__account_cycles() " tip-bot for Jin Yao
2 siblings, 0 replies; 5+ messages in thread
From: Arnaldo Carvalho de Melo @ 2019-04-10 13:00 UTC (permalink / raw)
To: Jin Yao
Cc: jolsa, peterz, mingo, alexander.shishkin, Linux-kernel, ak,
kan.liang, yao.jin
Em Sat, Mar 16, 2019 at 05:16:17AM +0800, Jin Yao escreveu:
> The hist__account_cycles is executed when the hist_iter__branch_callback
> is called. But it looks it's not necessary. In hist__account_cycles, it
> already walks on all branch entries.
>
> This patch moves the hist__account_cycles out of callback, now the data
> processing is much faster than before.
>
> Previous code has an issue that the ch[offset].num++
> (in __symbol__account_cycles) is executed repeatedly since
> hist__account_cycles is called in each hist_iter__branch_callback,
> so the counting of ch[offset].num is not correct (too big).
> With this patch, the issue is fixed. And we don't need the code of
> "ch->reset >= ch->num / 2" to check if there are too many overlaps
> (in annotation__count_and_fill), otherwise some data would be
> hidden.
>
> Now, we can try, for example:
Tested with:
perf annotate --stdio
perf annotate --stdio2
perf annotate --tui
perf report --tui -> press annotation hotkey 'a'
Applied,
- Arnaldo
^ permalink raw reply [flat|nested] 5+ messages in thread
* [tip:perf/core] perf annotate: Remove hist__account_cycles() from callback
2019-03-15 21:16 [PATCH v3] perf annotate/report: Remove hist__account_cycles from callback Jin Yao
2019-03-20 0:35 ` Jin, Yao
2019-04-10 13:00 ` Arnaldo Carvalho de Melo
@ 2019-05-18 8:46 ` tip-bot for Jin Yao
2 siblings, 0 replies; 5+ messages in thread
From: tip-bot for Jin Yao @ 2019-05-18 8:46 UTC (permalink / raw)
To: linux-tip-commits
Cc: acme, ak, peterz, kan.liang, yao.jin, linux-kernel, hpa, jolsa,
yao.jin, alexander.shishkin, mingo, tglx
Commit-ID: bdd1666b3d03d675bdb7f8d92b29f2797acbc5e8
Gitweb: https://git.kernel.org/tip/bdd1666b3d03d675bdb7f8d92b29f2797acbc5e8
Author: Jin Yao <yao.jin@linux.intel.com>
AuthorDate: Sat, 16 Mar 2019 05:16:17 +0800
Committer: Arnaldo Carvalho de Melo <acme@redhat.com>
CommitDate: Wed, 15 May 2019 16:36:46 -0300
perf annotate: Remove hist__account_cycles() from callback
The hist__account_cycles() function is executed when the
hist_iter__branch_callback() is called.
But it looks it's not necessary. In hist__account_cycles, it already
walks on all branch entries.
This patch moves the hist__account_cycles out of callback, now the data
processing is much faster than before.
Previous code has an issue that the ch[offset].num++ (in
__symbol__account_cycles) is executed repeatedly since
hist__account_cycles is called in each hist_iter__branch_callback, so
the counting of ch[offset].num is not correct (too big).
With this patch, the issue is fixed. And we don't need the code of
"ch->reset >= ch->num / 2" to check if there are too many overlaps (in
annotation__count_and_fill), otherwise some data would be hidden.
Now, we can try, for example:
perf record -b ...
perf annotate or perf report -s symbol
The before/after output should be no change.
v3:
---
Fix the crash in stdio mode.
Like previous code, it needs the checking of ui__has_annotation()
before hist__account_cycles()
v2:
---
1. Cover the similar perf report
2. Remove the checking code "ch->reset >= ch->num / 2"
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1552684577-29041-1-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/builtin-annotate.c | 4 ++--
tools/perf/builtin-report.c | 11 +++++------
tools/perf/util/annotate.c | 2 +-
3 files changed, 8 insertions(+), 9 deletions(-)
diff --git a/tools/perf/builtin-annotate.c b/tools/perf/builtin-annotate.c
index 67f9d9ffacfb..77deb3a40596 100644
--- a/tools/perf/builtin-annotate.c
+++ b/tools/perf/builtin-annotate.c
@@ -159,8 +159,6 @@ static int hist_iter__branch_callback(struct hist_entry_iter *iter,
struct perf_evsel *evsel = iter->evsel;
int err;
- hist__account_cycles(sample->branch_stack, al, sample, false);
-
bi = he->branch_info;
err = addr_map_symbol__inc_samples(&bi->from, sample, evsel);
@@ -199,6 +197,8 @@ static int process_branch_callback(struct perf_evsel *evsel,
if (a.map != NULL)
a.map->dso->hit = 1;
+ hist__account_cycles(sample->branch_stack, al, sample, false);
+
ret = hist_entry_iter__add(&iter, &a, PERF_MAX_STACK_DEPTH, ann);
return ret;
}
diff --git a/tools/perf/builtin-report.c b/tools/perf/builtin-report.c
index 4054eb1f98ac..91e27ac297c2 100644
--- a/tools/perf/builtin-report.c
+++ b/tools/perf/builtin-report.c
@@ -136,9 +136,6 @@ static int hist_iter__report_callback(struct hist_entry_iter *iter,
if (!ui__has_annotation() && !rep->symbol_ipc)
return 0;
- hist__account_cycles(sample->branch_stack, al, sample,
- rep->nonany_branch_mode);
-
if (sort__mode == SORT_MODE__BRANCH) {
bi = he->branch_info;
err = addr_map_symbol__inc_samples(&bi->from, sample, evsel);
@@ -181,9 +178,6 @@ static int hist_iter__branch_callback(struct hist_entry_iter *iter,
if (!ui__has_annotation() && !rep->symbol_ipc)
return 0;
- hist__account_cycles(sample->branch_stack, al, sample,
- rep->nonany_branch_mode);
-
bi = he->branch_info;
err = addr_map_symbol__inc_samples(&bi->from, sample, evsel);
if (err)
@@ -282,6 +276,11 @@ static int process_sample_event(struct perf_tool *tool,
if (al.map != NULL)
al.map->dso->hit = 1;
+ if (ui__has_annotation() || rep->symbol_ipc) {
+ hist__account_cycles(sample->branch_stack, &al, sample,
+ rep->nonany_branch_mode);
+ }
+
ret = hist_entry_iter__add(&iter, &al, rep->max_stack, rep);
if (ret < 0)
pr_debug("problem adding hist entry, skipping event\n");
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 09762985c713..0b8573fd9b05 100644
--- a/tools/perf/util/annotate.c
+++ b/tools/perf/util/annotate.c
@@ -1021,7 +1021,7 @@ static void annotation__count_and_fill(struct annotation *notes, u64 start, u64
float ipc = n_insn / ((double)ch->cycles / (double)ch->num);
/* Hide data when there are too many overlaps. */
- if (ch->reset >= 0x7fff || ch->reset >= ch->num / 2)
+ if (ch->reset >= 0x7fff)
return;
for (offset = start; offset <= end; offset++) {
^ permalink raw reply related [flat|nested] 5+ messages in thread