perf: Stream comparison
v2 cover perf: Stream comparison
Sometimes, a small change in a hot function reducing the cycles of this function, but the overall workload doesn't get faster. It is interesting where the cycles are moved to.
What it would like is to diff before/after streams. A stream we think is a callchain which is aggregated by the branch records from samples.
By browsing the hot streams, we can understand the hot code flow. By comparing the cycles variation of same streams between old perf data and new perf data, we can understand if the cycles are moved to the unchanged code.
The before stream is the stream before source code changed (e.g. streams in perf.data.old). The after stream is the stream after source code changed (e.g. streams in perf.data).
Diffing before/after streams compares all streams (or compares top N hot streams) between two perf data files.
If all entries of one stream in perf.data.old are fully matched with all entries of another stream in perf.data, we think these two streams are matched otherwise the streams are not matched.
For example,
cycles: 1, hits: 26.80% cycles: 1, hits: 27.30%
-------------------------- --------------------------
main div.c:39 main div.c:39
main div.c:44 main div.c:44
It looks that two streams are matched and we can see for the same streams the cycles are equal and the callchain hit percents are slightly changed. That's expected in the normal range.
But that's not always true if source code is changed in perf.data (e.g. source line div.c:39 is changed). If the source line is changed, they are different streams, we can't compare them. We will think the stream in perf.data is a new stream.
The challenge is how to identify the changed source lines. The basic idea is to use linux command "diff" to compare the source file A and source file A* line by line (assume A is used in perf.data.old and A* is updated in perf.data). According to "diff" output, we can create a source line mapping table.
For example,
Execute 'diff ./before/div.c ./after/div.c'
25c25
< i = rand() % 2;
---
> i = rand() % 4;
39c39
< for (i = 0; i < 2000000000; i++) {
---
> for (i = 0; i < 20000000001; i++) { div.c (after -> before) lines mapping:
0 -> 0
1 -> 1
2 -> 2
3 -> 3
4 -> 4
5 -> 5
6 -> 6
7 -> 7
8 -> 8
9 -> 9
...
24 -> 24
25 -> -1
26 -> 26
27 -> 27
28 -> 28
29 -> 29
30 -> 30
31 -> 31
32 -> 32
33 -> 33
34 -> 34
35 -> 35
36 -> 36
37 -> 37
38 -> 38
39 -> -1
40 -> 40
...
From the table, we can easily know div.c:39 is source line changed. (mapped to -1). So these two streams are not matched.
Besides the hot streams comparison, this patch series also support the top N hottest blocks comparison.
It's also useful to figure out the top N hottest blocks from old perf data file and figure out the top N hottest blocks from new perf data file, and then compare them for the cycles diff. It can let us easily know how many cycles are moved from one block to another block.
Now let's see examples.
perf record -b ... Generate perf.data.old with branch data
perf record -b ... Generate perf.data with branch data
perf diff --stream --percent-limit 2
[ Matched hot chains between old perf data and new perf data) ]
hot chain pair 1:
cycles: 1, hits: 26.80% cycles: 1, hits: 27.30%
--------------------------- --------------------------
main div.c:39 main div.c:39
main div.c:44 main div.c:44
hot chain pair 2:
cycles: 35, hits: 21.43% cycles: 33, hits: 19.37%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
main div.c:40 main div.c:40
main div.c:39 main div.c:39
hot chain pair 3:
cycles: 18, hits: 6.10% cycles: 19, hits: 6.51%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
hot chain pair 4:
cycles: 9, hits: 5.95% cycles: 8, hits: 5.03%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380[ Hot chains in old perf data but source line changed (*) in new perf data ]
[ Hot chains in old perf data only ]
hot chain 1:
cycles: 2, hits: 4.08%
--------------------------
main div.c:42
compute_flag div.c:28[ Hot chains in new perf data only ]
hot chain 1:
cycles: 36, hits: 3.36%
--------------------------
__random_r random_r.c:357
__random random.c:293
__random random.c:293
__random random.c:291
__random random.c:291
__random random.c:291
__random random.c:288
rand rand.c:27
rand rand.c:26
rand@plt
rand@plt
compute_flag div.c:25
compute_flag div.c:22
main div.c:40
main div.c:40Ignore the rightmost columns such as '[Program Block Range]' and 'Shared Object' for saving space
# Output based on old perf data:
#
# Sampled Cycles% Avg Cycles New Stream Diff(cycles%,cycles) New Stream Sampled Cycles% New Stream Avg Cycles
# ............... .......... ............................... .......................... .....................
#
25.20% 18 -0.36%, -1 - -
15.24% 7 -0.45%, 0 - -
5.07% 2 0.09%, 0 - -
4.84% 2 0.26%, 0 - -
4.72% 2 0.30%, 0 - -
3.91% 1 0.29%, 0 - -
3.05% 1 0.11%, 0 - -
2.90% 1 0.08%, 0 - -
2.71% 1 -0.11%, 0 - -
2.44% 1 0.09%, 0 - -
2.35% 1 -0.09%, 0 - -
2.27% 1 0.15%, 0 - -
2.27% 1 0.06%, 0 - -
2.17% 1 0.09%, 0 - -If we enable the source line comparison, the output might be different.
perf diff --stream --before ./before --after ./after
[ Matched hot chains between old perf data and new perf data) ]
hot chain pair 1:
cycles: 18, hits: 6.10% cycles: 19, hits: 6.51%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
hot chain pair 2:
cycles: 9, hits: 5.95% cycles: 8, hits: 5.03%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380[ Hot chains in old perf data but source line changed (*) in new perf data ]
hot chain pair 1:
cycles: 1, hits: 26.80% cycles: 1, hits: 27.30%
--------------------------- --------------------------
main div.c:39 main div.c:39*
main div.c:44 main div.c:44
hot chain pair 2:
cycles: 35, hits: 21.43% cycles: 33, hits: 19.37%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
main div.c:40 main div.c:40
main div.c:39 main div.c:39*
[ Hot chains in old perf data only ]
hot chain 1:
cycles: 2, hits: 4.08%
--------------------------
main div.c:42
compute_flag div.c:28[ Hot chains in new perf data only ]
hot chain 1:
cycles: 36, hits: 3.36%
--------------------------
__random_r random_r.c:357
__random random.c:293
__random random.c:293
__random random.c:291
__random random.c:291
__random random.c:291
__random random.c:288
rand rand.c:27
rand rand.c:26
rand@plt
rand@plt
compute_flag div.c:25
compute_flag div.c:22
main div.c:40
main div.c:40# Output based on old perf data:
#
# Sampled Cycles% Avg Cycles New Stream Diff(cycles%,cycles) New Stream Sampled Cycles% New Stream Avg Cycles # ............... .......... ............................... .......................... .....................
#
25.20% 18 [block changed in new stream] 24.84% 17
15.24% 7 -0.45%, 0 - -
5.07% 2 0.09%, 0 - -
4.84% 2 0.26%, 0 - -
4.72% 2 0.30%, 0 - -
3.91% 1 0.29%, 0 - -
3.05% 1 0.11%, 0 - -
2.90% 1 0.08%, 0 - -
2.71% 1 -0.11%, 0 - -
2.44% 1 0.09%, 0 - -
2.35% 1 -0.09%, 0 - -
2.27% 1 0.15%, 0 - -
2.27% 1 0.06%, 0 - -
2.17% 1 0.09%, 0 - -
Sometime some changes are not reflected in the source code, e.g. changing the compiler option. So for this, we can't get the changes by diffing the source code lines.
This patch series also introduces a new perf-diff option "--changed-func". It passes the names of changed functions then perf-diff can know what functions are changed.
For example,
perf diff --stream --changed-func main --changed-func rand
NOTE:
-----
1. For the patches:
perf util: Create source line mapping table
perf util: Create streams for managing top N hottest callchains perf util: Return per-event callchain streams
perf util: Compare two streams
perf util: Calculate the sum of all streams hits
perf util: Report hot streams
perf diff: Support hot streams comparison
These patches support the hot stream comparison.
2. For the patches:
perf util: Add new block info functions for top N hot blocks comparison perf util: Add new block info fmts for showing hot blocks comparison perf util: Enable block source line comparison
perf diff: support hot blocks comparison
These patches support the hot blocks comparison.
3. For the patches
perf util: Filter out streams by name of changed functions perf util: Filter out blocks by name of changed functions perf diff: Filter out streams by changed functions
These patches support a user specified function name list which let perf-diff know these functions are changed.
v2:
---
Refine the codes for following patches:
perf util: Create source line mapping table
perf util: Create streams for managing top N hottest callchains perf util: Calculate the sum of all streams hits
perf util: Add new block info functions for top N hot blocks comparison
Jin Yao (14):
perf util: Create source line mapping table
perf util: Create streams for managing top N hottest callchains perf util: Return per-event callchain streams
perf util: Compare two streams
perf util: Calculate the sum of all streams hits
perf util: Report hot streams
perf diff: Support hot streams comparison
perf util: Add new block info functions for top N hot blocks
comparison
perf util: Add new block info fmts for showing hot blocks comparison perf util: Enable block source line comparison
perf diff: support hot blocks comparison
perf util: Filter out streams by name of changed functions perf util: Filter out blocks by name of changed functions perf diff: Filter out streams by changed functions
tools/perf/Documentation/perf-diff.txt | 19 + tools/perf/builtin-diff.c | 324 ++++++++++++--- tools/perf/util/Build | 1 + tools/perf/util/block-info.c | 433 ++++++++++++++++++- tools/perf/util/block-info.h | 38 +- tools/perf/util/callchain.c | 517 +++++++++++++++++++++++ tools/perf/util/callchain.h | 34 ++ tools/perf/util/srclist.c | 555 +++++++++++++++++++++++++ tools/perf/util/srclist.h | 74 ++++ 9 files changed, 1935 insertions(+), 60 deletions(-) create mode 100644 tools/perf/util/srclist.c create mode 100644 tools/perf/util/srclist.h
Signature · 1 line
2.17.1
PATCH v2 1/14 perf util: Create source line mapping table
Sometimes, a small change in a hot function reducing the cycles of this function, but the overall workload doesn't get faster. It is interesting where the cycles are moved to.
What it would like is to diff before/after streams. A stream is a callchain which is aggregated by the branch records from samples.
By browsing the hot streams, we can understand the hot code flow. By comparing the cycles variation of same streams between old perf data and new perf data, we can understand if the cycles are moved to the unchanged code.
The before stream is the stream before source line changed (e.g. streams in perf.data.old). The after stream is the stream after source line changed (e.g. streams in perf.data).
Diffing before/after streams compares all streams (or compares top N hot streams) between two perf data files.
If all entries of one stream in perf.data.old are fully matched with all entries of another stream in perf.data, we think these two streams are matched, otherwise the streams are not matched.
For example,
cycles: 1, hits: 26.80% cycles: 1, hits: 27.30%
-------------------------- --------------------------
main div.c:39 main div.c:39
main div.c:44 main div.c:44
It looks that two streams are matched and we can see for the same streams the cycles are equal and the stream hit percents are slightly changed. That's expected in the normal range.
But that's not always true if source lines are changed in perf.data (e.g. div.c:39 is changed). If the source line is changed, they are different streams, we can't compare them. We just think the stream in perf.data is a new one.
The challenge is how to identify the changed source lines. The basic idea is to use linux command 'diff' to compare the source file A and source file A* line by line (assume A is used in perf.data.old and A* is updated in perf.data). Create a source line mapping table.
For example,
Execute 'diff ./before/div.c ./after/div.c'
25c25
< i = rand() % 2;
---
> i = rand() % 4;
39c39
< for (i = 0; i < 2000000000; i++) {
---
> for (i = 0; i < 20000000001; i++) { div.c (after -> before) lines mapping:
0 -> 0
1 -> 1
2 -> 2
3 -> 3
4 -> 4
5 -> 5
6 -> 6
7 -> 7
8 -> 8
9 -> 9
...
24 -> 24
25 -> -1
26 -> 26
27 -> 27
28 -> 28
29 -> 29
30 -> 30
31 -> 31
32 -> 32
33 -> 33
34 -> 34
35 -> 35
36 -> 36
37 -> 37
38 -> 38
39 -> -1
40 -> 40
...
From the table, we can easily know div.c:39 is source line changed. (it's mapped to -1). So these two streams are not matched.
This patch can also handle the cases of new source lines insertion and old source lines deletion. This source line mapping table is a base for next processing for stream comparison.
This patch creates a new rblist 'srclist' to manage the source line mapping tables. Each node contains the source line mapping table of a before/after source file pair. The node is created on demand as we process the samples. So it's likely we don't need to create so many nodes.
v2:
---
Refine the code
1. calloc -> zalloc
2. Functions operating on a 'struct line_pair' have the line_pair__ prefix.
3. Check get_range() return value.
4. Use fgets to replace fgetc for getting the number of lines.
5. Use path__join to generate the full path.
6. Keep the buffer after calling getline.
--- tools/perf/util/Build | 1 + tools/perf/util/srclist.c | 555 ++++++++++++++++++++++++++++++++++++++ tools/perf/util/srclist.h | 65 +++++ 3 files changed, 621 insertions(+) create mode 100644 tools/perf/util/srclist.c create mode 100644 tools/perf/util/srclist.h
tools/perf/util/Build +1 -0
@@ -99,6 +99,7 @@ perf-y += call-path.o perf-y += rwsem.o perf-y += thread-stack.o perf-y += spark.o +perf-y += srclist.o perf-$(CONFIG_AUXTRACE) += auxtrace.o perf-$(CONFIG_AUXTRACE) += intel-pt-decoder/ perf-$(CONFIG_AUXTRACE) += intel-pt.o
tools/perf/util/srclist.c new file +555 -0
555 lines not on this page · raw message
tools/perf/util/srclist.h new file +65 -0
65 lines not on this page · raw message
Signature · 1 line
2.17.1
PATCH v2 2/14 perf util: Create streams for managing top N hottest callchains
In reply to Jin YaoWe think the stream is a callchain which is aggregated by the LBR records from samples. By browsing the stream, we can understand the code flow.
The struct callchain_node represents one callchain and we use the callchain_node->hit to measure the hot level of this callchain. Higher is hotter.
Since in perf data file, there may be many callchains so we just need to focus on the top N hottest callchains. N is a user defined parameter or just a predefined default value.
This patch saves the top N hottest callchains in 'struct stream_node' type array, which is defined in a per event 'struct callchain_streams'.
So now we can get the per-event top N hottest callchains.
v2:
---
Use zfree in free_evsel_streams().
--- tools/perf/util/callchain.c | 122 ++++++++++++++++++++++++++++++++++++ tools/perf/util/callchain.h | 16 +++++ 2 files changed, 138 insertions(+)
tools/perf/util/callchain.c +122 -0
@@ -31,6 +31,7 @@#include "callchain.h" #include "branch.h" #include "symbol.h" +#include "evlist.h" #include "../perf.h" #define CALLCHAIN_PARAM_DEFAULT \
@@ -1599,3 +1600,124 @@ void callchain_cursor_reset(struct callchain_cursor *cursor)for (node = cursor->first; node !=NULL ; node = node->next)map__zput (node->ms.map); } + +static void free_evsel_streams (struct callchain_streams *callchain_streams, +int nr_evsel) +{ +for (int i =0 ; i < nr_evsel; i++) +zfree (&callchain_streams[i].streams); + +free (callchain_streams); +} + +static struct callchain_streams *create_evsel_streams (int nr_evsel, +int nr_streams_max) +{ +struct callchain_streams *callchain_streams; + + callchain_streams =calloc (nr_evsel,sizeof (struct callchain_streams )); +if (!callchain_streams) +return NULL ; + +for (int i =0 ; i < nr_evsel; i++) { +struct callchain_streams *s = &callchain_streams[i]; + + s->streams =calloc (nr_streams_max,sizeof (struct stream_node )); +if (!s->streams) + goto err; + + s->nr_streams_max = nr_streams_max; + s->evsel_idx =-1 ; + } + +return callchain_streams; + +err: +free_evsel_streams (callchain_streams, nr_evsel); +return NULL ; +} + +/* +* The cnodes with high hit number are hot callchains. +*/ +static void set_hot_cnode (struct callchain_streams *s, +struct callchain_node *cnode) +{ +int i, idx =0 ; +u64 hit; + +if (s->nr_streams < s->nr_streams_max) { + i = s->nr_streams; + s->streams[i].cnode = cnode; + s->nr_streams++; +return ; + } + +/* +* Since only a few number of hot streams, so only use simple +* way to find the cnode with smallest hit number and replace. +*/ + hit = (s->streams[0 ].cnode)->hit; +for (i =1 ; i < s->nr_streams; i++) { +if ((s->streams[i].cnode)->hit < hit) { + hit = (s->streams[i].cnode)->hit; + idx = i; + } + } + +if (cnode->hit > hit) + s->streams[idx].cnode = cnode; +} + +static void update_hot_streams (struct hist_entry *he, +struct callchain_streams *s) +{ +struct rb_root *root = &he->sorted_chain; +struct rb_node *rb_node =rb_first (root); +struct callchain_node *node; + +while (rb_node) { + node = rb_entry(rb_node, struct callchain_node, rb_node); +set_hot_cnode (s, node); + rb_node =rb_next (rb_node); + } +} + +static void get_hot_streams (struct hists *hists, +struct callchain_streams *s) +{ +struct rb_node *next =rb_first_cached (&hists->entries); + +while (next) { +struct hist_entry *he; + + he = rb_entry(next, struct hist_entry, rb_node); +update_hot_streams (he, s); + next =rb_next (&he->rb_node); + } +} + +struct callchain_streams *callchain_evsel_streams_create (struct evlist *evlist, +int nr_streams_max, +int *nr_evsel_streams) +{ +int nr_evsel = evlist->core.nr_entries, i =0 ; +struct callchain_streams *callchain_streams; +struct evsel *pos; + + callchain_streams =create_evsel_streams (nr_evsel, nr_streams_max); +if (!callchain_streams) +return NULL ; + +evlist__for_each_entry (evlist, pos) { +struct hists *hists =evsel__hists (pos); + +hists__output_resort (hists,NULL ); +get_hot_streams (hists, &callchain_streams[i]); + callchain_streams[i].evsel_idx = pos->idx; + i++; + } + + *nr_evsel_streams = nr_evsel; +return callchain_streams; +}
tools/perf/util/callchain.h +16 -0
@@ -13,6 +13,7 @@ struct ip_callchain;struct map ;struct perf_sample ;struct thread ; +struct evlist ;#define HELP_PAD "\t\t\t\t"
@@ -159,6 +160,17 @@ struct callchain_cursor {struct callchain_cursor_node *curr; }; +struct stream_node { +struct callchain_node *cnode; +}; + +struct callchain_streams { +struct stream_node *streams; +int nr_streams_max; +int nr_streams; +int evsel_idx; +}; +extern __threadstruct callchain_cursor callchain_cursor;static inline void callchain_init (struct callchain_root *root)
@@ -289,4 +301,8 @@ int callchain_branch_counts(struct callchain_root *root, u64 *branch_count, u64 *predicted_count, u64 *abort_count, u64 *cycles_count); +struct callchain_streams *callchain_evsel_streams_create (struct evlist *evlist, +int nr_streams_max, +int *nr_evsel_streams); +#endif /* __PERF_CALLCHAIN_H */
Signature · 1 line
2.17.1
PATCH v2 3/14 perf util: Return per-event callchain streams
In reply to Jin YaoIn previous patch, we have created a 'struct callchain_streams' array and each array entry contains per-event callchain streams.
This patch returns the pointer of per-event callchain streams according to the evsel_idx.
--- tools/perf/util/callchain.c | 12 ++++++++++++ tools/perf/util/callchain.h | 4 ++++ 2 files changed, 16 insertions(+)
tools/perf/util/callchain.c +12 -0
@@ -1721,3 +1721,15 @@ struct callchain_streams *callchain_evsel_streams_create(struct evlist *evlist, *nr_evsel_streams = nr_evsel;return callchain_streams; } + +struct callchain_streams *callchain_evsel_streams_get (struct callchain_streams *cs, +int nr_streams_max, +int evsel_idx) +{ +for (int i =0 ; i < nr_streams_max; i++) { +if (cs[i].evsel_idx == evsel_idx) +return &cs[i]; + } + +return NULL ; +}
tools/perf/util/callchain.h +4 -0
@@ -305,4 +305,8 @@ struct callchain_streams *callchain_evsel_streams_create(struct evlist *evlist,int nr_streams_max, int *nr_evsel_streams); +struct callchain_streams *callchain_evsel_streams_get (struct callchain_streams *cs, +int nr_streams_max, +int evsel_idx); +#endif /* __PERF_CALLCHAIN_H */
Signature · 1 line
2.17.1
PATCH v2 4/14 perf util: Compare two streams
In reply to Jin YaoIf all callchain entries of one stream are fully matched with all entries of another stream, we think these two streams are matched.
For example,
cycles: 1, hits: 26.80% cycles: 1, hits: 27.30%
----------------------- -----------------------
main div.c:39 main div.c:39
main div.c:44 main div.c:44
Above two streams are matched if div.c is not changed.
But the streams are not matched if div.c is source line changed (e.g. div.c:39 is changed). If the source line is changed, they are different streams.
So the challenge is how to identify the changed source lines. For this purpose, we use the source line mapping table (see patch "perf util: Create source line mapping table"). The source line mapping information is saved in src_list.
So the matching logic is, if we can find the source lines, the callchain entry comparison is based on source lines. Otherwise, fallback to dso address comparison.
Once two streams are fully matched, they will be linked as a pair. From the pair, we can know which streams are matched. The pair information will be used in next patches.
--- tools/perf/util/callchain.c | 172 ++++++++++++++++++++++++++++++++++++ tools/perf/util/callchain.h | 8 ++ 2 files changed, 180 insertions(+)
tools/perf/util/callchain.c +172 -0
@@ -1733,3 +1733,175 @@ struct callchain_streams *callchain_evsel_streams_get(struct callchain_streams *return NULL ; } + +static bool chain_srclist_match (struct srclist *src_list,const char *srcline_a, +const char *srcline_b,bool *src_found, +bool *src_changed) +{ +char file_a[PATH_MAX ], file_b[PATH_MAX ]; +int line_a, line_b; +struct src_node *node; +struct line_pair *lp; + +if (sscanf (srcline_a,"%[^:]:%d" , file_a, &line_a) !=2 ) +return false; + +if (sscanf (srcline_b,"%[^:]:%d" , file_b, &line_b) !=2 ) +return false; + +if (strcmp (file_a, file_b)) +return false; + + node =srclist__find (src_list, file_a, true); +if (!node) +return false; + + *src_found = true; + + lp =srclist__line_pair (node, line_a); +if (!lp) +return false; + +if (line_a == line_b && lp->b_nr ==-1 ) { + *src_changed = true; +return true; + } + +if (lp->b_nr == line_b) +return true; + +return false; +} + +static bool chain_match (struct callchain_list *base_chain, +struct callchain_list *pair_chain, +struct srclist *src_list, +bool *src_changed) +{ +enum match_result match; +bool src_found = false; + + *src_changed = false; + +if (src_list &&chain_srclist_match (src_list, pair_chain->srcline, + base_chain->srcline, &src_found, + src_changed)) { +return true; + } + +if (!src_found) { + match =match_chain_strings (base_chain->srcline, + pair_chain->srcline); +if (match !=MATCH_ERROR ) +return match ==MATCH_EQ ; + + match =match_chain_dso_addresses (base_chain->ms.map, + base_chain->ip, + pair_chain->ms.map, + pair_chain->ip); + +return match ==MATCH_EQ ; + } + +return false; +} + +static bool callchain_node_matched (struct callchain_node *base_cnode, +struct callchain_node *pair_cnode, +struct srclist *src_list, +int *nr_changed) +{ +struct callchain_list *base_chain, *pair_chain; +bool match = false; + + pair_chain =list_first_entry (&pair_cnode->val, + struct callchain_list, + list); + +list_for_each_entry (base_chain, &base_cnode->val, list) { +bool src_changed; + +if (&pair_chain->list == &pair_cnode->val) +return false; + +if (!base_chain->srcline || !pair_chain->srcline) { + pair_chain =list_next_entry (pair_chain, list); +continue ; + } + + match =chain_match (base_chain, pair_chain, src_list, + &src_changed); + +if (src_changed) { + pair_chain->src_changed = true; + *nr_changed +=1 ; + } + +if (!match) +return false; + + pair_chain =list_next_entry (pair_chain, list); + } + +/* +* Say chain1 is ABC, chain2 is ABCD, we consider they are +* not fully matched. +*/ +if (pair_chain && (&pair_chain->list != &pair_cnode->val)) +return false; + +return match; +} + +static struct stream_node *stream_node_match (struct stream_node *base_node, +struct callchain_streams *cs_pair, +struct srclist *src_list, +bool *src_changed) +{ + *src_changed = false; + +for (int i =0 ; i < cs_pair->nr_streams; i++) { +struct stream_node *pair_node = &cs_pair->streams[i]; +int nr_changed =0 ; + +if (callchain_node_matched (base_node->cnode, pair_node->cnode, + src_list, &nr_changed)) { +if (nr_changed) + *src_changed = true; + +return pair_node; + } + } + +return NULL ; +} + +static void stream_nodes_link (struct stream_node *base_node, +struct stream_node *pair_node, +bool src_changed) +{ + base_node->pair_cnode = pair_node->cnode; + base_node->src_changed = src_changed; + pair_node->pair_cnode = base_node->cnode; +} + +void callchain_match_streams (struct callchain_streams *cs_base, +struct callchain_streams *cs_pair, +struct srclist *src_list) +{ +for (int i =0 ; i < cs_base->nr_streams; i++) { +struct stream_node *base_node = &cs_base->streams[i]; +struct stream_node *pair_node; +bool src_changed; + + pair_node =stream_node_match (base_node, cs_pair, src_list, + &src_changed); +if (pair_node) +stream_nodes_link (base_node, pair_node, src_changed); +else + base_node->src_changed = src_changed; + } + +if (src_list) +srclist__dump (src_list); +}
tools/perf/util/callchain.h +8 -0
@@ -6,6 +6,7 @@#include <linux/rbtree.h> #include "map_symbol.h" #include "branch.h" +#include "srclist.h" struct addr_location ;struct evsel ;
@@ -131,6 +132,7 @@ struct callchain_list {u64 iter_cycles;struct branch_type_stat brtype_stat;const char *srcline; +bool src_changed;struct list_head list; };
@@ -162,6 +164,8 @@ struct callchain_cursor {struct stream_node {struct callchain_node *cnode; +struct callchain_node *pair_cnode; +bool src_changed; };struct callchain_streams {
@@ -309,4 +313,8 @@ struct callchain_streams *callchain_evsel_streams_get(struct callchain_streams *int nr_streams_max, int evsel_idx); +void callchain_match_streams (struct callchain_streams *cs_base, +struct callchain_streams *cs_pair, +struct srclist *src_list); +#endif /* __PERF_CALLCHAIN_H */
Signature · 1 line
2.17.1
PATCH v2 5/14 perf util: Calculate the sum of all streams hits
In reply to Jin YaoWe have used callchain_node->hit to measure the hot level of one stream. This patch calculates the sum of hits of all streams.
Then in next patch, we can use following formula to report hot percent for one stream.
hot percent = callchain_node->hit / sum of all hits
v2:
---
Combine the variable decl line with its inital assignment in total_callchain_hits().
--- tools/perf/util/callchain.c | 34 ++++++++++++++++++++++++++++++++++ tools/perf/util/callchain.h | 1 + 2 files changed, 35 insertions(+)
tools/perf/util/callchain.c +34 -0
@@ -1683,6 +1683,38 @@ static void update_hot_streams(struct hist_entry *he, } } +static u64 count_callchain_hits (struct hist_entry *he) +{ +struct rb_root *root = &he->sorted_chain; +struct rb_node *rb_node =rb_first (root); +struct callchain_node *node; +u64 chain_hits =0 ; + +while (rb_node) { + node = rb_entry(rb_node, struct callchain_node, rb_node); + chain_hits += node->hit; + rb_node =rb_next (rb_node); + } + +return chain_hits; +} + +static u64 total_callchain_hits (struct hists *hists) +{ +struct rb_node *next =rb_first_cached (&hists->entries); +u64 chain_hits =0 ; + +while (next) { + struct hist_entry *he = rb_entry(next, struct hist_entry, + rb_node); + + chain_hits +=count_callchain_hits (he); + next =rb_next (&he->rb_node); + } + +return chain_hits; +} +static void get_hot_streams (struct hists *hists,struct callchain_streams *s) {
@@ -1695,6 +1727,8 @@ static void get_hot_streams(struct hists *hists,update_hot_streams (he, s); next =rb_next (&he->rb_node); } + + s->chain_hits =total_callchain_hits (hists); } struct callchain_streams *callchain_evsel_streams_create(struct evlist *evlist,
tools/perf/util/callchain.h +1 -0
@@ -173,6 +173,7 @@ struct callchain_streams {int nr_streams_max;int nr_streams;int evsel_idx; +u64 chain_hits; };extern __threadstruct callchain_cursor callchain_cursor;
Signature · 1 line
2.17.1
PATCH v2 6/14 perf util: Report hot streams
In reply to Jin YaoWe show the streams separately. They are divided into different sections.
1. "Matched hot chains between old perf data and new perf data"
2. "Hot chains in old perf data but source line changed in new perf data"
3. "Hot chains in old perf data only"
4. "Hot chains in new perf data only".
For each stream, we report the cycles and hot percent (hits%).
For example,
cycles: 2, hits: 4.08%
--------------------------
main div.c:42
compute_flag div.c:28--- tools/perf/util/callchain.c | 155 ++++++++++++++++++++++++++++++++++++ tools/perf/util/callchain.h | 3 + 2 files changed, 158 insertions(+)
tools/perf/util/callchain.c +155 -0
@@ -1777,6 +1777,9 @@ static bool chain_srclist_match(struct srclist *src_list, const char *srcline_a,struct src_node *node;struct line_pair *lp; +if (!srcline_a || !srcline_b) +return false; +if (sscanf (srcline_a,"%[^:]:%d" , file_a, &line_a) !=2 )return false;
@@ -1817,6 +1820,10 @@ static bool chain_match(struct callchain_list *base_chain, *src_changed = false; +/* +* Check sourceline first. If not matched, +* fallback to symbol match and address match. +*/ if (src_list &&chain_srclist_match (src_list, pair_chain->srcline, base_chain->srcline, &src_found, src_changed)) {
@@ -1939,3 +1946,151 @@ void callchain_match_streams(struct callchain_streams *cs_base,if (src_list)srclist__dump (src_list); } + +static s64 callchain_avg_cycles (struct callchain_node *cnode) +{ +struct callchain_list *chain; +s64 cycles =0 ; + +list_for_each_entry (chain, &cnode->val, list) { +if (chain->srcline && chain->branch_count) + cycles += chain->cycles_count / chain->branch_count; + } + +return cycles; +} + +static void print_callchain_pair (struct stream_node *base_node,int idx, +struct callchain_streams *cs_base, +struct callchain_streams *cs_pair) +{ +struct callchain_node *base_cnode = base_node->cnode; +struct callchain_node *pair_cnode = base_node->pair_cnode; +struct callchain_list *base_chain, *pair_chain; +char buf1[512 ], buf2[512 ], cbuf1[256 ], cbuf2[256 ]; +char *s1, *s2; +double pct; + +printf ("\nhot chain pair %d:\n" , idx); + + pct = (double )base_cnode->hit / (double )cs_base->chain_hits; +scnprintf (buf1,sizeof (buf1),"cycles: %ld, hits: %.2f%%" , +callchain_avg_cycles (base_cnode), pct *100.0 ); + + pct = (double )pair_cnode->hit / (double )cs_pair->chain_hits; +scnprintf (buf2,sizeof (buf2),"cycles: %ld, hits: %.2f%%" , +callchain_avg_cycles (pair_cnode), pct *100.0 ); + +printf ("%35s\t%35s\n" , buf1, buf2); + +printf ("%35s\t%35s\n" , +"---------------------------" , +"--------------------------" ); + + pair_chain =list_first_entry (&pair_cnode->val, + struct callchain_list, + list); + +list_for_each_entry (base_chain, &base_cnode->val, list) { +if (&pair_chain->list == &pair_cnode->val) +return ; + + s1 =callchain_list__sym_name (base_chain, cbuf1,sizeof (cbuf1), + false); + s2 =callchain_list__sym_name (pair_chain, cbuf2,sizeof (cbuf2), + false); + +if (!pair_chain->src_changed) +scnprintf (buf1,sizeof (buf1),"%35s\t%35s" , s1, s2); +else +scnprintf (buf1,sizeof (buf1),"%35s\t%35s*" , s1, s2); + +printf ("%s\n" , buf1); + pair_chain =list_next_entry (pair_chain, list); + } +} + +static void print_stream_callchain (struct stream_node *node,int idx, +struct callchain_streams *cs,bool pair) +{ +struct callchain_node *cnode = node->cnode; +struct callchain_list *chain; +char buf[512 ], cbuf[256 ], *s; +double pct; + +printf ("\nhot chain %d:\n" , idx); + + pct = (double )cnode->hit / (double )cs->chain_hits; +scnprintf (buf,sizeof (buf),"cycles: %ld, hits: %.2f%%" , +callchain_avg_cycles (cnode), pct *100.0 ); + +if (pair) { +printf ("%35s\t%35s\n" ,"" , buf); +printf ("%35s\t%35s\n" , +"" ,"--------------------------" ); + }else { +printf ("%35s\n" , buf); +printf ("%35s\n" ,"--------------------------" ); + } + +list_for_each_entry (chain, &cnode->val, list) { + s =callchain_list__sym_name (chain, cbuf,sizeof (cbuf), false); + +if (pair) +scnprintf (buf,sizeof (buf),"%35s\t%35s" ,"" , s); +else +scnprintf (buf,sizeof (buf),"%35s" , s); + +printf ("%s\n" , buf); + } +} + +void callchain_stream_report (struct callchain_streams *cs_base, +struct callchain_streams *cs_pair) +{ +struct stream_node *base_node; +int i, idx =0 ; + +printf ("[ Matched hot chains between old perf data and new perf data ]\n" ); +for (i =0 ; i < cs_base->nr_streams; i++) { + base_node = &cs_base->streams[i]; +if (base_node->pair_cnode) { +if (!base_node->src_changed) { +print_callchain_pair (base_node, ++idx, + cs_base, cs_pair); + } + } + } + + idx =0 ; +printf ("\n[ Hot chains in old perf data but source line changed (*) in new perf data ]\n" ); +for (i =0 ; i < cs_base->nr_streams; i++) { + base_node = &cs_base->streams[i]; +if (base_node->pair_cnode) { +if (base_node->src_changed) { +print_callchain_pair (base_node, ++idx, + cs_base, cs_pair); + } + } + } + + idx =0 ; +printf ("\n[ Hot chains in old perf data only ]\n" ); +for (i =0 ; i < cs_base->nr_streams; i++) { + base_node = &cs_base->streams[i]; +if (!base_node->pair_cnode) { +print_stream_callchain (base_node, ++idx, + cs_base, false); + } + } + + idx =0 ; +printf ("\n[ Hot chains in new perf data only ]\n" ); +for (i =0 ; i < cs_pair->nr_streams; i++) { + base_node = &cs_pair->streams[i]; +if (!base_node->pair_cnode) { +print_stream_callchain (base_node, ++idx, + cs_pair, true); + } + } +}
tools/perf/util/callchain.h +3 -0
@@ -318,4 +318,7 @@ void callchain_match_streams(struct callchain_streams *cs_base,struct callchain_streams *cs_pair, struct srclist *src_list); +void callchain_stream_report (struct callchain_streams *cs_base, +struct callchain_streams *cs_pair); +#endif /* __PERF_CALLCHAIN_H */
Signature · 1 line
2.17.1
PATCH v2 7/14 perf diff: Support hot streams comparison
In reply to Jin YaoThis patch enables perf-diff with "--stream", "--before" and "--after" options.
"--stream": Enable hot streams comparison
"--before": Source code directory corresponding to perf.data.old "--after" : Source code directory corresponding to perf.data
Let's see two examples
perf record -b ... Generate perf.data.old with branch data
perf record -b ... Generate perf.data with branch data
perf diff --stream
[ Matched hot chains between old perf data and new perf data ]
hot chain pair 1:
cycles: 1, hits: 26.80% cycles: 1, hits: 27.30%
--------------------------- --------------------------
main div.c:39 main div.c:39
main div.c:44 main div.c:44
hot chain pair 2:
cycles: 35, hits: 21.43% cycles: 33, hits: 19.37%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
main div.c:40 main div.c:40
main div.c:39 main div.c:39
hot chain pair 3:
cycles: 18, hits: 6.10% cycles: 19, hits: 6.51%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
hot chain pair 4:
cycles: 9, hits: 5.95% cycles: 8, hits: 5.03%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380[ Hot chains in old perf data but source line changed (*) in new perf data ]
[ Hot chains in old perf data only ]
hot chain 1:
cycles: 2, hits: 4.08%
--------------------------
main div.c:42
compute_flag div.c:28[ Hot chains in new perf data only ]
hot chain 1:
cycles: 36, hits: 3.36%
--------------------------
__random_r random_r.c:357
__random random.c:293
__random random.c:293
__random random.c:291
__random random.c:291
__random random.c:291
__random random.c:288
rand rand.c:27
rand rand.c:26
rand@plt
rand@plt
compute_flag div.c:25
compute_flag div.c:22
main div.c:40
main div.c:40If we enable the source line comparison, the output might be different.
perf diff --stream --before ./before --after ./after
[ Matched hot chains between old perf data and new perf data) ]
hot chain pair 1:
cycles: 18, hits: 6.10% cycles: 19, hits: 6.51%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
hot chain pair 2:
cycles: 9, hits: 5.95% cycles: 8, hits: 5.03%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380[ Hot chains in old perf data but source line changed (*) in new perf data ]
hot chain pair 1:
cycles: 1, hits: 26.80% cycles: 1, hits: 27.30%
--------------------------- --------------------------
main div.c:39 main div.c:39*
main div.c:44 main div.c:44
hot chain pair 2:
cycles: 35, hits: 21.43% cycles: 33, hits: 19.37%
--------------------------- --------------------------
__random_r random_r.c:360 __random_r random_r.c:360
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:388 __random_r random_r.c:388
__random_r random_r.c:380 __random_r random_r.c:380
__random_r random_r.c:357 __random_r random_r.c:357
__random random.c:293 __random random.c:293
__random random.c:293 __random random.c:293
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:291 __random random.c:291
__random random.c:288 __random random.c:288
rand rand.c:27 rand rand.c:27
rand rand.c:26 rand rand.c:26
rand@plt rand@plt
rand@plt rand@plt
compute_flag div.c:25 compute_flag div.c:25
compute_flag div.c:22 compute_flag div.c:22
main div.c:40 main div.c:40
main div.c:40 main div.c:40
main div.c:39 main div.c:39*
[ Hot chains in old perf data only ]
hot chain 1:
cycles: 2, hits: 4.08%
--------------------------
main div.c:42
compute_flag div.c:28[ Hot chains in new perf data only ]
hot chain 1:
cycles: 36, hits: 3.36%
--------------------------
__random_r random_r.c:357
__random random.c:293
__random random.c:293
__random random.c:291
__random random.c:291
__random random.c:291
__random random.c:288
rand rand.c:27
rand rand.c:26
rand@plt
rand@plt
compute_flag div.c:25
compute_flag div.c:22
main div.c:40
main div.c:40
main div.c:39--- tools/perf/Documentation/perf-diff.txt | 14 ++ tools/perf/builtin-diff.c | 170 +++++++++++++++++++++++-- 2 files changed, 171 insertions(+), 13 deletions(-)
tools/perf/Documentation/perf-diff.txt +14 -0
@@ -182,6 +182,20 @@ OPTIONS --tid=:: Only diff samples for given thread ID (comma separated list). +--stream:: + Enable hot streams comparison. Stream is a callchain which is + aggregated by the branch records from samples. The option can be + used with --before and --after if source line comparison is + required. + +--before:: + Source code directory corresponding to perf.data.old. Should be + used with --stream and --after. + +--after:: + Source code directory corresponding to perf.data. Should be + used with --stream and --before. + COMPARISON ---------- The comparison is governed by the baseline file. The baseline perf.data
tools/perf/builtin-diff.c +157 -13
275 lines not on this page · raw message
Signature · 1 line
2.17.1
PATCH v2 8/14 perf util: Add new block info functions for top N hot blocks comparison
In reply to Jin YaoIt's also useful to figure out the top N hottest blocks from old perf data file and figure out the top N hottest blocks from new perf data file, and then compare them for the cycles diff. It can let us easily know how many cycles are moved from one block to another block.
This patch adds new helper functions and data structures for the block comparison.
And it also updates the existing perf-diff to be compatible with the new interface.
v2:
---
Use name block_line__zdelete to replace free_block_line
--- tools/perf/builtin-diff.c | 45 +----- tools/perf/util/block-info.c | 288 ++++++++++++++++++++++++++++++++++- tools/perf/util/block-info.h | 32 +++- tools/perf/util/srclist.h | 9 ++ 4 files changed, 328 insertions(+), 46 deletions(-)
tools/perf/builtin-diff.c +3 -42
@@ -598,27 +598,6 @@ static void init_block_hist(struct block_hist *bh) bh->valid = true; } -static struct hist_entry *get_block_pair (struct hist_entry *he, -struct hists *hists_pair) -{ -struct rb_root_cached *root = hists_pair->entries_in; -struct rb_node *next =rb_first_cached (root); -int64_t cmp; - -while (next !=NULL ) { - struct hist_entry *he_pair = rb_entry(next, struct hist_entry, - rb_node_in); - - next =rb_next (&he_pair->rb_node_in); - - cmp =__block_info__cmp (he_pair, he); -if (!cmp) -return he_pair; - } - -return NULL ; -} -static void init_spark_values (unsigned long *svals,int num) {for (int i =0 ; i < num; i++)
@@ -665,26 +644,6 @@ static void compute_cycles_diff(struct hist_entry *he, } } -static void block_hists_match (struct hists *hists_base, -struct hists *hists_pair) -{ -struct rb_root_cached *root = hists_base->entries_in; -struct rb_node *next =rb_first_cached (root); - -while (next !=NULL ) { - struct hist_entry *he = rb_entry(next, struct hist_entry, - rb_node_in); -struct hist_entry *pair =get_block_pair (he, hists_pair); - - next =rb_next (&he->rb_node_in); - -if (pair) { -hist_entry__add_pair (pair, he); -compute_cycles_diff (he, pair); - } - } -} -static void hists__precompute (struct hists *hists) {struct rb_root_cached *root;
@@ -737,7 +696,9 @@ static void hists__precompute(struct hists *hists)if (bh->valid && pair_bh->valid) {block_hists_match (&bh->block_hists, - &pair_bh->block_hists); + &pair_bh->block_hists, +NULL , + compute_cycles_diff);hists__output_resort (&pair_bh->block_hists,NULL ); }
tools/perf/util/block-info.c +285 -3
326 lines not on this page · raw message
tools/perf/util/block-info.h +31 -1
@@ -20,6 +20,8 @@ struct block_info {int num;int num_aggr;refcount_t refcnt; +struct block_line *line; +bool srcline_matched; };struct block_fmt {
@@ -46,6 +48,7 @@ struct block_report {u64 cycles;struct block_fmt fmts[PERF_HPP_REPORT__BLOCK_MAX_INDEX ];int nr_fmts; +int evsel_idx; };struct block_hist ;
@@ -62,7 +65,8 @@ static inline void __block_info__zput(struct block_info **bi)#define block_info__zput (bi) __block_info__zput(&bi) -int64_t __block_info__cmp (struct hist_entry *left,struct hist_entry *right); +int64_t __block_info__cmp (struct hist_entry *left,struct hist_entry *right, +struct srclist *src_list__maybe_unused );int64_t block_info__cmp (struct perf_hpp_fmt *fmt__maybe_unused ,struct hist_entry *left,struct hist_entry *right);
@@ -83,4 +87,30 @@ int report__browse_block_hists(struct block_hist *bh, float min_percent,float block_info__total_cycles_percent (struct hist_entry *he); +struct block_report *block_info__get_report (struct block_report *reps, +int nr_reps,int evsel_idx); + +int block_hists_addr2line (struct hists *hists,const char *dir); + +void block_line_dump (struct block_info *bi,const char *str); + +bool block_same_srcfiles (struct block_line *bl_a,struct block_line *bl_b); + +bool block_srclist_matched (struct srclist *slist,char *rel_path, +int start_nr_a,int end_nr_a, +int start_nr_b,int end_nr_b, +bool *changed); + +void block_hists_match (struct hists *hists_base, +struct hists *hists_pair, +struct srclist *src_list, +void (*func)(struct hist_entry *, +struct hist_entry *)); + +int block_info__match_report (struct block_report *rep_base, +struct block_report *rep_pair, +struct srclist *src_list, +void (*func)(struct hist_entry *, +struct hist_entry *)); +#endif /* __PERF_BLOCK_H */
tools/perf/util/srclist.h +9 -0
@@ -33,6 +33,15 @@ struct srclist {const char *after_dir; }; +struct block_line { +char *start_file; +char *end_file; +char *start_rel; +char *end_rel; +unsigned int start_nr; +unsigned int end_nr; +}; +struct srclist *srclist__new (const char *before_dir,const char *after_dir);void srclist__delete (struct srclist *slist);
Signature · 1 line
2.17.1
PATCH v2 9/14 perf util: Add new block info fmts for showing hot blocks comparison
In reply to Jin YaoWe support three new columns in output
New Stream Diff(cycles%,cycles):
The diff of 'Sampled Cycles%' and 'Avg Cycles' between new stream and old stream for the same block.
If block is not matched, it reports "[block not in new stream]" or "[block changed in new stream]".
New Stream Sampled Cycles%:
The 'Sampled Cycles%' of the block in new stream. It's only reported when the block is changed in new stream.
New Stream Avg Cycles:
The 'Average Cycles' of the block in new stream. It's only reported when the block is changed in new stream.
--- tools/perf/util/block-info.c | 117 ++++++++++++++++++++++++++++++++++- tools/perf/util/block-info.h | 4 ++ 2 files changed, 120 insertions(+), 1 deletion(-)
tools/perf/util/block-info.c +116 -1
@@ -41,7 +41,19 @@ static struct block_header_column { [PERF_HPP_REPORT__BLOCK_DSO ] = { .name ="Shared Object" , .width =20 , - } + }, + [PERF_HPP__BLOCK_NEW_STREAM_DIFF ] = { + .name ="New Stream Diff(cycles%,cycles)" , + .width =31 , + }, + [PERF_HPP__BLOCK_NEW_STREAM_TOTAL_CYCLES_PCT ] = { + .name ="New Stream Sampled Cycles%" , + .width =26 , + }, + [PERF_HPP__BLOCK_NEW_STREAM_AVG_CYCLES ] = { + .name ="New Stream Avg Cycles" , + .width =21 , + }, };struct block_info *block_info__get (struct block_info *bi)
@@ -342,6 +354,100 @@ static int block_dso_entry(struct perf_hpp_fmt *fmt, struct perf_hpp *hpp, "[unknown]"); } +static int block_new_stream_diff_entry (struct perf_hpp_fmt *fmt, +struct perf_hpp *hpp, +struct hist_entry *he) +{ + struct block_fmt *block_fmt = container_of(fmt, struct block_fmt, fmt); +struct hist_entry *pair =hist_entry__next_pair (he); +struct block_info *bi_h, *bi_p; +double ratio_h =0.0 , ratio_p =0.0 ; +s64 cycles_diff; +char buf[32 ]; + +if (!pair) { +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, +"[block not in new stream]" ); + } + + bi_h = he->block_info; + bi_p = pair->block_info; + +if (bi_p->block_changed) { +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, +"[block changed in new stream]" ); + } + +if (bi_h->total_cycles) + ratio_h = (double )bi_h->cycles / (double )bi_h->total_cycles; + +if (bi_p->total_cycles) + ratio_p = (double )bi_p->cycles / (double )bi_p->total_cycles; + + cycles_diff = (s64)(bi_p->cycles_aggr / bi_p->num_aggr) - + (s64)(bi_h->cycles_aggr / bi_h->num_aggr); + +snprintf (buf,sizeof (buf),"%4.2f%%, %4ld" , +100.0 * (ratio_p - ratio_h), cycles_diff); + +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, buf); +} + +static int block_new_stream_total_cycles_pct_entry (struct perf_hpp_fmt *fmt, +struct perf_hpp *hpp, +struct hist_entry *he) +{ + struct block_fmt *block_fmt = container_of(fmt, struct block_fmt, fmt); +struct hist_entry *pair =hist_entry__next_pair (he); +struct block_info *bi_p; +double ratio =0.0 ; + +if (!pair) { +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, +"-" ); + } + + bi_p = pair->block_info; + +if (!bi_p->block_changed) { +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, +"-" ); + } + +if (bi_p->total_cycles) + ratio = (double )bi_p->cycles / (double )bi_p->total_cycles; + +return color_pct (hpp, block_fmt->width,100.0 * ratio); +} + +static int block_new_stream_avg_cycles_entry (struct perf_hpp_fmt *fmt, +struct perf_hpp *hpp, +struct hist_entry *he) +{ + struct block_fmt *block_fmt = container_of(fmt, struct block_fmt, fmt); +struct hist_entry *pair =hist_entry__next_pair (he); +struct block_info *bi_p; +char cycles_buf[16 ]; + +if (!pair) { +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, +"-" ); + } + + bi_p = pair->block_info; + +if (!bi_p->block_changed) { +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, +"-" ); + } + +cycles_string (bi_p->cycles_aggr / bi_p->num_aggr, cycles_buf, +sizeof (cycles_buf)); + +return scnprintf (hpp->buf, hpp->size,"%*s" , block_fmt->width, + cycles_buf); +} +static void init_block_header (struct block_fmt *block_fmt) {struct perf_hpp_fmt *fmt = &block_fmt->fmt;
@@ -385,6 +491,15 @@ static void hpp_register(struct block_fmt *block_fmt, int idx,case PERF_HPP_REPORT__BLOCK_DSO : fmt->entry = block_dso_entry;break ; +case PERF_HPP__BLOCK_NEW_STREAM_DIFF : + fmt->entry = block_new_stream_diff_entry; +break ; +case PERF_HPP__BLOCK_NEW_STREAM_TOTAL_CYCLES_PCT : + fmt->entry = block_new_stream_total_cycles_pct_entry; +break ; +case PERF_HPP__BLOCK_NEW_STREAM_AVG_CYCLES : + fmt->entry = block_new_stream_avg_cycles_entry; +break ;default :return ; }
tools/perf/util/block-info.h +4 -0
@@ -22,6 +22,7 @@ struct block_info {refcount_t refcnt;struct block_line *line;bool srcline_matched; +bool block_changed; };struct block_fmt {
@@ -40,6 +41,9 @@ enum {PERF_HPP_REPORT__BLOCK_AVG_CYCLES ,PERF_HPP_REPORT__BLOCK_RANGE ,PERF_HPP_REPORT__BLOCK_DSO , +PERF_HPP__BLOCK_NEW_STREAM_DIFF , +PERF_HPP__BLOCK_NEW_STREAM_TOTAL_CYCLES_PCT , +PERF_HPP__BLOCK_NEW_STREAM_AVG_CYCLES , PERF_HPP_REPORT__BLOCK_MAX_INDEX };
Signature · 1 line
2.17.1
PATCH v2 10/14 perf util: Enable block source line comparison
In reply to Jin YaoPreviously we only supported address comparison, but it was not good for the case that address might be changed if source code was updated.
This patch enables for the source line comparison.
1. If all of the source lines in a block are not changed (or only moved some offsets as a whole), we think the block is not changed.
2. If we can aware any source line in this block is changed, we think this block is changed.
3. If 1 and 2 are both not matched, fallback to address comparison.
--- tools/perf/util/block-info.c | 24 +++++++++++++++++++++++- 1 file changed, 23 insertions(+), 1 deletion(-)
tools/perf/util/block-info.c +23 -1
@@ -93,11 +93,12 @@ struct block_info *block_info__new(void) }int64_t __block_info__cmp (struct hist_entry *left,struct hist_entry *right, -struct srclist *src_list__maybe_unused ) +struct srclist *src_list) {struct block_info *bi_l = left->block_info;struct block_info *bi_r = right->block_info;int cmp; +bool changed;if (!bi_l->sym || !bi_r->sym) {if (!bi_l->sym && !bi_r->sym)
@@ -112,6 +113,27 @@ int64_t __block_info__cmp(struct hist_entry *left, struct hist_entry *right,if (cmp)return cmp; +if (src_list && bi_l->line && bi_r->line) { +if (block_same_srcfiles (bi_l->line, bi_r->line) && + bi_l->line->start_rel) { + +if (block_srclist_matched (src_list, + bi_l->line->start_rel, + bi_l->line->start_nr, + bi_l->line->end_nr, + bi_r->line->start_nr, + bi_r->line->end_nr, + &changed)) { + bi_l->srcline_matched = true; +return 0 ; + }else if (changed) { + bi_l->block_changed = true; +return 0 ; + }else +return -1 ; + } + } +if (bi_l->start != bi_r->start)return (int64_t )(bi_r->start - bi_l->start);
Signature · 1 line
2.17.1
PATCH v2 11/14 perf diff: support hot blocks comparison
In reply to Jin YaoThis patch reports the results of hot blocks comparison.
For example, following outputs are added to the end of hot streams reports.
# Output based on old stream (perf.data.old):
#
# Sampled Cycles% Avg Cycles New Stream Diff(cycles%,cycles) New Stream Sampled Cycles% New Stream Avg Cycles
# ............... .......... ............................... .......................... .....................
#
24.76% 17 -4.54%, -7 - -
14.72% 7 0.13%, -1 - -
5.28% 2 [block not in new stream] - -
5.09% 2 0.52%, 0 - -
5.05% 2 -0.15%, -1 - -
4.18% 1 0.12%, 0 - -
3.16% 1 0.28%, 0 - -If we enable the source line comparison, the output might be different.
# Output based on old stream (perf.data.old):
#
# Sampled Cycles% Avg Cycles New Stream Diff(cycles%,cycles) New Stream Sampled Cycles% New Stream Avg Cycles # ............... .......... ............................... .......................... .....................
#
24.76% 17 [block changed in new stream] 20.22% 10
14.72% 7 0.13%, -1 - -
5.28% 2 [block not in new stream] - -
5.09% 2 0.52%, 0 - -
5.05% 2 -0.15%, -1 - -
4.18% 1 [block changed in new stream] 4.31% 1
3.16% 1 0.28%, 0 - -
--- tools/perf/builtin-diff.c | 85 ++++++++++++++++++++++++++++++++++++++- 1 file changed, 84 insertions(+), 1 deletion(-)
tools/perf/builtin-diff.c +84 -1
@@ -49,6 +49,8 @@ struct perf_diff {bool src_cmp;bool stream;struct srclist *src_list; +u64 total_cycles; +float min_percent; };/* Diff command specific HPP columns. */
@@ -76,8 +78,10 @@ struct diff_hpp_fmt {struct data__file {struct perf_session *session; +struct block_report *block_reports;struct perf_data data;int idx; +int nr_block_reports;struct hists *hists;struct callchain_streams *evsel_streams;int nr_evsel_streams;
@@ -445,8 +449,10 @@ static int diff__process_sample_event(struct perf_tool *tool,pr_debug ("problem adding hist entry, skipping event\n" ); goto out_put; } -break ; +hist__account_cycles (sample->branch_stack, &al, sample, false, + &pdiff->total_cycles); +break ;default :if (!hists__add_entry (hists, &al,NULL ,NULL ,NULL , sample, true)) {
@@ -992,6 +998,7 @@ static int process_base_stream(struct data__file *data_base,struct evlist *evlist_pair = data_pair->session->evlist;struct evsel *evsel_base, *evsel_pair;struct callchain_streams *es_base, *es_pair; +struct block_report *rep_base, *rep_pair;evlist__for_each_entry (evlist_base, evsel_base) { evsel_pair =evsel_match (evsel_base, evlist_pair);
@@ -1014,6 +1021,37 @@ static int process_base_stream(struct data__file *data_base,callchain_stream_report (es_base, es_pair); } +evlist__for_each_entry (evlist_base, evsel_base) { + rep_base =block_info__get_report (data_base->block_reports, + data_base->nr_block_reports, + evsel_base->idx); + +if (!rep_base) +return -1 ; + +block_hists_addr2line (&rep_base->hist.block_hists, base_dir); + + evlist_pair = data_pair->session->evlist; + evsel_pair =evsel_match (evsel_base, evlist_pair); +if (!evsel_pair) +continue ; + + rep_pair =block_info__get_report (data_pair->block_reports, + data_pair->nr_block_reports, + evsel_pair->idx); + +block_hists_addr2line (&rep_pair->hist.block_hists, pair_dir); + +block_info__match_report (rep_base, rep_pair, + pdiff.src_list,NULL ); + +fprintf (stdout,"%s" , title); + + use_browser =0 ; +report__browse_block_hists (&rep_base->hist, pdiff.min_percent, + evsel_base,NULL ,NULL ); + } +return 0 ; }
@@ -1142,6 +1180,26 @@ static int check_file_brstack(void)return 0 ; } +static struct block_report *create_block_reports (struct evlist *evlist, +u64 total_cycles, +int *nr_block_reports) +{ +struct block_report *reps; +int block_hpps[7 ] = { +PERF_HPP_REPORT__BLOCK_TOTAL_CYCLES_PCT , +PERF_HPP_REPORT__BLOCK_AVG_CYCLES , +PERF_HPP__BLOCK_NEW_STREAM_DIFF , +PERF_HPP__BLOCK_NEW_STREAM_TOTAL_CYCLES_PCT , +PERF_HPP__BLOCK_NEW_STREAM_AVG_CYCLES , +PERF_HPP_REPORT__BLOCK_RANGE , +PERF_HPP_REPORT__BLOCK_DSO , + }; + + reps =block_info__create_report (evlist, total_cycles, block_hpps,7 , + nr_block_reports); +return reps; +} +static struct callchain_streams *create_evsel_streams (struct evlist *evlist,int nr_streams_max,int *nr_evsel_streams)
@@ -1197,6 +1255,8 @@ static int __cmd_diff(void) goto out_delete; } + pdiff.total_cycles =0 ; + ret =perf_session__process_events (d->session);if (ret) {pr_err ("Failed to process %s\n" , d->data.path);
@@ -1215,6 +1275,12 @@ static int __cmd_diff(void) &d->nr_evsel_streams);if (!d->evsel_streams) goto out_delete; + + d->block_reports =create_block_reports (d->session->evlist, + pdiff.total_cycles, + &d->nr_block_reports); +if (!d->block_reports) + goto out_delete; } }
@@ -1225,6 +1291,11 @@ static int __cmd_diff(void) out_delete:data__for_each_file (i, d) { +if (d->block_reports) { +block_info__free_report (d->block_reports, + d->nr_block_reports); + } +perf_session__delete (d->session);data__free (d); }
@@ -1244,6 +1315,15 @@ static int __cmd_diff(void)return ret; } +static int parse_percent_limit (const struct option *opt,const char *arg, +int unset__maybe_unused ) +{ +struct perf_diff *d = opt->value; + + d->min_percent =strtof (arg,NULL ); +return 0 ; +} +static const char *const diff_usage[] = {"perf diff [<options>] [old_file] [new_file]" ,NULL ,
@@ -1307,6 +1387,9 @@ static const struct option options[] = {OPT_STRING (0 ,"after" , &pdiff.after_dir,"dir" ,"Source code directory corresponding to perf.data. " "WARNING: use with --before and --stream" ), +OPT_CALLBACK (0 ,"percent-limit" , &pdiff,"percent" , +"Don't show entries under that percent" , + parse_percent_limit), OPT_END() };
Signature · 1 line
2.17.1
PATCH v2 12/14 perf util: Filter out streams by name of changed functions
In reply to Jin YaoSometime some changes are not reflected in the source code, e.g. changing the compiler option. So for this, we can't get the changes by diffing the source code lines.
This idea is to let user provide a list of changed functions, then perf-diff can know these functions are not matched between old perf data and new perf data.
In this patch, the names of changed function names are stored in strlist, and we will check if the symbol name is in strlist. If yes, that means this function is changed.
--- tools/perf/builtin-diff.c | 2 +- tools/perf/util/callchain.c | 30 ++++++++++++++++++++++++++---- tools/perf/util/callchain.h | 4 +++- 3 files changed, 30 insertions(+), 6 deletions(-)
tools/perf/builtin-diff.c +1 -1
@@ -1017,7 +1017,7 @@ static int process_base_stream(struct data__file *data_base,if (!es_pair)return -1 ; -callchain_match_streams (es_base, es_pair, pdiff.src_list); +callchain_match_streams (es_base, es_pair, pdiff.src_list,NULL );callchain_stream_report (es_base, es_pair); }
tools/perf/util/callchain.c +26 -4
@@ -1810,16 +1810,35 @@ static bool chain_srclist_match(struct srclist *src_list, const char *srcline_a,return false; } +static bool sym_name_changed (struct callchain_list *base_chain, +struct callchain_list *pair_chain, +struct strlist *func_list) +{ +if (!func_list || !base_chain->ms.sym || !pair_chain->ms.sym) +return false; + +if (strlist__has_entry (func_list, base_chain->ms.sym->name)) +return true; + +return false; +} +static bool chain_match (struct callchain_list *base_chain,struct callchain_list *pair_chain,struct srclist *src_list, +struct strlist *func_list,bool *src_changed) {enum match_result match;bool src_found = false; +bool func_changed; *src_changed = false; + func_changed =sym_name_changed (base_chain, pair_chain, func_list); +if (func_changed) +return false; +/* * Check sourceline first. If not matched, * fallback to symbol match and address match.
@@ -1850,6 +1869,7 @@ static bool chain_match(struct callchain_list *base_chain,static bool callchain_node_matched (struct callchain_node *base_cnode,struct callchain_node *pair_cnode,struct srclist *src_list, +struct strlist *func_list,int *nr_changed) {struct callchain_list *base_chain, *pair_chain;
@@ -1871,7 +1891,7 @@ static bool callchain_node_matched(struct callchain_node *base_cnode, } match =chain_match (base_chain, pair_chain, src_list, - &src_changed); + func_list, &src_changed);if (src_changed) { pair_chain->src_changed = true;
@@ -1897,6 +1917,7 @@ static bool callchain_node_matched(struct callchain_node *base_cnode,static struct stream_node *stream_node_match (struct stream_node *base_node,struct callchain_streams *cs_pair,struct srclist *src_list, +struct strlist *func_list,bool *src_changed) { *src_changed = false;
@@ -1906,7 +1927,7 @@ static struct stream_node *stream_node_match(struct stream_node *base_node,int nr_changed =0 ;if (callchain_node_matched (base_node->cnode, pair_node->cnode, - src_list, &nr_changed)) { + src_list, func_list, &nr_changed)) {if (nr_changed) *src_changed = true;
@@ -1928,7 +1949,8 @@ static void stream_nodes_link(struct stream_node *base_node,void callchain_match_streams (struct callchain_streams *cs_base,struct callchain_streams *cs_pair, -struct srclist *src_list) +struct srclist *src_list, +struct strlist *func_list) {for (int i =0 ; i < cs_base->nr_streams; i++) {struct stream_node *base_node = &cs_base->streams[i];
@@ -1936,7 +1958,7 @@ void callchain_match_streams(struct callchain_streams *cs_base,bool src_changed; pair_node =stream_node_match (base_node, cs_pair, src_list, - &src_changed); + func_list, &src_changed);if (pair_node)stream_nodes_link (base_node, pair_node, src_changed);else
tools/perf/util/callchain.h +3 -1
@@ -7,6 +7,7 @@#include "map_symbol.h" #include "branch.h" #include "srclist.h" +#include "strlist.h" struct addr_location ;struct evsel ;
@@ -316,7 +317,8 @@ struct callchain_streams *callchain_evsel_streams_get(struct callchain_streams *void callchain_match_streams (struct callchain_streams *cs_base,struct callchain_streams *cs_pair, -struct srclist *src_list); +struct srclist *src_list, +struct strlist *func_list);void callchain_stream_report (struct callchain_streams *cs_base,struct callchain_streams *cs_pair);
Signature · 1 line
2.17.1
PATCH v2 13/14 perf util: Filter out blocks by name of changed functions
In reply to Jin YaoSimilar as previous patch "perf util: Filter out streams by name of changed functions", user provides a list of changed functions, then perf-diff can know these functions are not matched between old perf data and new perf data.
In this patch, the names of changed function names have been stored in strlist, and we will check if the symbol name is in strlist. If yes, that means this function is changed and the related blocks are not matched.
--- tools/perf/builtin-diff.c | 4 ++-- tools/perf/util/block-info.c | 18 ++++++++++++------ tools/perf/util/block-info.h | 4 +++- 3 files changed, 17 insertions(+), 9 deletions(-)
tools/perf/builtin-diff.c +2 -2
@@ -703,7 +703,7 @@ static void hists__precompute(struct hists *hists)if (bh->valid && pair_bh->valid) {block_hists_match (&bh->block_hists, &pair_bh->block_hists, -NULL , +NULL ,NULL , compute_cycles_diff);hists__output_resort (&pair_bh->block_hists,NULL );
@@ -1043,7 +1043,7 @@ static int process_base_stream(struct data__file *data_base,block_hists_addr2line (&rep_pair->hist.block_hists, pair_dir);block_info__match_report (rep_base, rep_pair, - pdiff.src_list,NULL ); + pdiff.src_list,NULL ,NULL );fprintf (stdout,"%s" , title);
tools/perf/util/block-info.c +12 -6
@@ -93,7 +93,7 @@ struct block_info *block_info__new(void) }int64_t __block_info__cmp (struct hist_entry *left,struct hist_entry *right, -struct srclist *src_list) +struct srclist *src_list,struct strlist *func_list) {struct block_info *bi_l = left->block_info;struct block_info *bi_r = right->block_info;
@@ -113,6 +113,9 @@ int64_t __block_info__cmp(struct hist_entry *left, struct hist_entry *right,if (cmp)return cmp; +if (func_list &&strlist__has_entry (func_list, bi_l->sym->name)) +return -1 ; +if (src_list && bi_l->line && bi_r->line) {if (block_same_srcfiles (bi_l->line, bi_r->line) && bi_l->line->start_rel) {
@@ -143,7 +146,7 @@ int64_t __block_info__cmp(struct hist_entry *left, struct hist_entry *right,int64_t block_info__cmp (struct perf_hpp_fmt *fmt__maybe_unused ,struct hist_entry *left,struct hist_entry *right) { -return __block_info__cmp (left, right,NULL ); +return __block_info__cmp (left, right,NULL ,NULL ); } static void init_block_info(struct block_info *bi, struct symbol *sym,
@@ -855,7 +858,8 @@ bool block_srclist_matched(struct srclist *slist, char *rel_path,static struct hist_entry *get_block_pair (struct hist_entry *he,struct hists *hists_pair, -struct srclist *src_list) +struct srclist *src_list, +struct strlist *func_list) {struct rb_root_cached *root = hists_pair->entries_in;struct rb_node *next =rb_first_cached (root);
@@ -867,7 +871,7 @@ static struct hist_entry *get_block_pair(struct hist_entry *he, next =rb_next (&he_pair->rb_node_in); - cmp =__block_info__cmp (he_pair, he, src_list); + cmp =__block_info__cmp (he_pair, he, src_list, func_list);if (!cmp)return he_pair; }
@@ -878,6 +882,7 @@ static struct hist_entry *get_block_pair(struct hist_entry *he,void block_hists_match (struct hists *hists_base,struct hists *hists_pair,struct srclist *src_list, +struct strlist *func_list,void (*func)(struct hist_entry *,struct hist_entry *)) {
@@ -888,7 +893,7 @@ void block_hists_match(struct hists *hists_base, struct hist_entry *he = rb_entry(next, struct hist_entry, rb_node_in);struct hist_entry *pair =get_block_pair (he, hists_pair, - src_list); + src_list, func_list); next =rb_next (&he->rb_node_in);
@@ -904,6 +909,7 @@ void block_hists_match(struct hists *hists_base,int block_info__match_report (struct block_report *rep_base,struct block_report *rep_pair,struct srclist *src_list, +struct strlist *func_list,void (*func)(struct hist_entry *,struct hist_entry *)) {
@@ -911,7 +917,7 @@ int block_info__match_report(struct block_report *rep_base,struct block_hist *bh_pair = &rep_pair->hist;block_hists_match (&bh_base->block_hists, &bh_pair->block_hists, - src_list, func); + src_list, func_list, func);return 0 ; }
tools/perf/util/block-info.h +3 -1
@@ -70,7 +70,7 @@ static inline void __block_info__zput(struct block_info **bi)#define block_info__zput (bi) __block_info__zput(&bi)int64_t __block_info__cmp (struct hist_entry *left,struct hist_entry *right, -struct srclist *src_list__maybe_unused ); +struct srclist *src_list,struct strlist *func_list);int64_t block_info__cmp (struct perf_hpp_fmt *fmt__maybe_unused ,struct hist_entry *left,struct hist_entry *right);
@@ -108,12 +108,14 @@ bool block_srclist_matched(struct srclist *slist, char *rel_path,void block_hists_match (struct hists *hists_base,struct hists *hists_pair,struct srclist *src_list, +struct strlist *func_list,void (*func)(struct hist_entry *,struct hist_entry *));int block_info__match_report (struct block_report *rep_base,struct block_report *rep_pair,struct srclist *src_list, +struct strlist *func_list,void (*func)(struct hist_entry *,struct hist_entry *));
Signature · 1 line
2.17.1
PATCH v2 14/14 perf diff: Filter out streams by changed functions
In reply to Jin YaoSometime some changes are not reflected in the source code, e.g. changing the compiler option. So for this, we can't get the changes by diffing the source code lines.
This patch introduces a new perf-diff option "--changed-func". It passes the names of changed functions then perf-diff can know what functions are changed.
For example,
perf diff --stream --changed-func main --changed-func rand
It passes the function list {"main", "rand"} to perf-diff. Now perf-diff knows the functions "main" and "rand" in new perf data file are changed.
--- tools/perf/Documentation/perf-diff.txt | 5 +++++ tools/perf/builtin-diff.c | 30 ++++++++++++++++++++++++-- 2 files changed, 33 insertions(+), 2 deletions(-)
tools/perf/Documentation/perf-diff.txt +5 -0
@@ -196,6 +196,11 @@ OPTIONS Source code directory corresponding to perf.data. Should be used with --stream and --before. +--changed-func=:: + The given function is changed in new perf data file. This option + needs to be used with --stream option. Multiple functions can be given + by using this option more than once. + COMPARISON ---------- The comparison is governed by the baseline file. The baseline perf.data
tools/perf/builtin-diff.c +28 -2
@@ -27,6 +27,7 @@#include "util/block-info.h" #include "util/srclist.h" #include "util/callchain.h" +#include "util/strlist.h" #include <linux/err.h> #include <linux/zalloc.h> #include <subcmd/pager.h>
@@ -49,6 +50,7 @@ struct perf_diff {bool src_cmp;bool stream;struct srclist *src_list; +struct strlist *func_list;u64 total_cycles;float min_percent; };
@@ -1017,7 +1019,8 @@ static int process_base_stream(struct data__file *data_base,if (!es_pair)return -1 ; -callchain_match_streams (es_base, es_pair, pdiff.src_list,NULL ); +callchain_match_streams (es_base, es_pair, pdiff.src_list, + pdiff.func_list);callchain_stream_report (es_base, es_pair); }
@@ -1043,7 +1046,7 @@ static int process_base_stream(struct data__file *data_base,block_hists_addr2line (&rep_pair->hist.block_hists, pair_dir);block_info__match_report (rep_base, rep_pair, - pdiff.src_list,NULL ,NULL ); + pdiff.src_list, pdiff.func_list,NULL );fprintf (stdout,"%s" , title);
@@ -1213,6 +1216,24 @@ static struct callchain_streams *create_evsel_streams(struct evlist *evlist,return evsel_streams; } +static int parse_func (const struct option *opt__maybe_unused , +const char *str,int unset__maybe_unused ) +{ +int ret; + +if (!pdiff.func_list) { + pdiff.func_list =strlist__new (NULL ,NULL ); +if (!pdiff.func_list) +return -ENOMEM ; + } + + ret =strlist__add (pdiff.func_list, str); +if (ret <0 ) +return ret; + +return 0 ; +} +static int __cmd_diff (void ) {struct data__file *d;
@@ -1312,6 +1333,9 @@ static int __cmd_diff(void)if (pdiff.src_list)srclist__delete (pdiff.src_list); +if (pdiff.func_list) +strlist__delete (pdiff.func_list); +return ret; }
@@ -1390,6 +1414,8 @@ static const struct option options[] = {OPT_CALLBACK (0 ,"percent-limit" , &pdiff,"percent" ,"Don't show entries under that percent" , parse_percent_limit), +OPT_CALLBACK (0 ,"changed-func" ,NULL ,"func" , +"Given function is changed" , parse_func), OPT_END() };
Signature · 1 line
2.17.1
On Fri, Mar 13, 2020 at 03:11:04PM +0800, Jin Yao wrote:
SNIP
Quoted 9 lines
v2:
---
Refine the codes for following patches:
perf util: Create source line mapping table
perf util: Create streams for managing top N hottest callchains perf util: Calculate the sum of all streams hits
perf util: Add new block info functions for top N hot blocks comparison
Quoted 16 lines
Jin Yao (14):
perf util: Create source line mapping table
perf util: Create streams for managing top N hottest callchains perf util: Return per-event callchain streams
perf util: Compare two streams
perf util: Calculate the sum of all streams hits
perf util: Report hot streams
perf diff: Support hot streams comparison
perf util: Add new block info functions for top N hot blocks
comparison
perf util: Add new block info fmts for showing hot blocks comparison perf util: Enable block source line comparison
perf diff: support hot blocks comparison
perf util: Filter out streams by name of changed functions perf util: Filter out blocks by name of changed functions perf diff: Filter out streams by changed functions
I can't apply this on latest perf/core, do you have it in git tree branch?
thanks,
jirka
Quoted 17 lines
tools/perf/Documentation/perf-diff.txt | 19 + tools/perf/builtin-diff.c | 324 ++++++++++++--- tools/perf/util/Build | 1 + tools/perf/util/block-info.c | 433 ++++++++++++++++++- tools/perf/util/block-info.h | 38 +- tools/perf/util/callchain.c | 517 +++++++++++++++++++++++ tools/perf/util/callchain.h | 34 ++ tools/perf/util/srclist.c | 555 +++++++++++++++++++++++++ tools/perf/util/srclist.h | 74 ++++ 9 files changed, 1935 insertions(+), 60 deletions(-) create mode 100644 tools/perf/util/srclist.c create mode 100644 tools/perf/util/srclist.hSignature · 1 line
2.17.1
Quoted 30 lines · Jiri Olsa
On Fri, Mar 13, 2020 at 03:11:04PM +0800, Jin Yao wrote:
SNIP
Quoted 9 lines
v2:
---
Refine the codes for following patches:
perf util: Create source line mapping table
perf util: Create streams for managing top N hottest callchains perf util: Calculate the sum of all streams hits
perf util: Add new block info functions for top N hot blocks comparisonQuoted 16 lines
Jin Yao (14):
perf util: Create source line mapping table
perf util: Create streams for managing top N hottest callchains perf util: Return per-event callchain streams
perf util: Compare two streams
perf util: Calculate the sum of all streams hits
perf util: Report hot streams
perf diff: Support hot streams comparison
perf util: Add new block info functions for top N hot blocks
comparison
perf util: Add new block info fmts for showing hot blocks comparison perf util: Enable block source line comparison
perf diff: support hot blocks comparison
perf util: Filter out streams by name of changed functions perf util: Filter out blocks by name of changed functions perf diff: Filter out streams by changed functions
Quoted 5 lines
I can't apply this on latest perf/core, do you have it in git tree branch?
thanks,
jirka
Not having a git tree branch. :(
Anyway let me rebase to perf/core
Thanks
Jin Yao
Quoted 18 lines
Quoted 17 lines
tools/perf/Documentation/perf-diff.txt | 19 + tools/perf/builtin-diff.c | 324 ++++++++++++--- tools/perf/util/Build | 1 + tools/perf/util/block-info.c | 433 ++++++++++++++++++- tools/perf/util/block-info.h | 38 +- tools/perf/util/callchain.c | 517 +++++++++++++++++++++++ tools/perf/util/callchain.h | 34 ++ tools/perf/util/srclist.c | 555 +++++++++++++++++++++++++ tools/perf/util/srclist.h | 74 ++++ 9 files changed, 1935 insertions(+), 60 deletions(-)create mode 100644
tools/perf/util/srclist.c
create mode 100644tools/perf/util/srclist.hSignature · 1 line
2.17.1
Quoted 3 lines · Jin Yao
Sometimes, a small change in a hot function reducing the cycles of this function, but the overall workload doesn't get faster. It is interesting where the cycles are moved to.
I'm getting compilation fail:
BUILD: Doing 'make -j1' parallel build CC util/srclist.o util/srclist.c: In function ‘srclist__node_new’: util/srclist.c:388:35: error: ‘%s’ directive output may be truncated writing up to 4095 bytes into a region of size 4091 [-Werror=format-truncation=] 388 | snprintf(cmd, sizeof(cmd), "diff %s %s", | ^~ ...... 456 | ret = init_src_info(b_path, a_path, rel_path, &node->info); | ~~~~~~ In file included from /usr/include/stdio.h:867, from util/srclist.c:8: /usr/include/bits/stdio2.h:67:10: note: ‘__builtin___snprintf_chk’ output between 7 and 8197 bytes into a destination of size 4096 67 | return __builtin___snprintf_chk (__s, __n, __USE_FORTIFY_LEVEL - 1, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 68 | __bos (__s), __fmt, __va_arg_pack ()); | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors mv: cannot stat 'util/.srclist.o.tmp': No such file or directory make[4]: *** [/home/jolsa/kernel/linux-perf/tools/build/Makefile.build:97: util/srclist.o] Error 1 make[3]: *** [/home/jolsa/kernel/linux-perf/tools/build/Makefile.build:139: util] Error 2 make[2]: *** [Makefile.perf:617: perf-in.o] Error 2 make[1]: *** [Makefile.perf:225: sub-make] Error 2 make: *** [Makefile:70: all] Error 2
[jolsa@krava ~]$ gcc --version
gcc (GCC) 9.3.1 20200317 (Red Hat 9.3.1-1)
jirka
Hi Jiri,
Quoted 32 lines · Jiri Olsa
Quoted 3 lines · Jin Yao
Sometimes, a small change in a hot function reducing the cycles of this function, but the overall workload doesn't get faster. It is interesting where the cycles are moved to.
I'm getting compilation fail:
BUILD: Doing 'make -j1' parallel build CC util/srclist.o util/srclist.c: In function ‘srclist__node_new’: util/srclist.c:388:35: error: ‘%s’ directive output may be truncated writing up to 4095 bytes into a region of size 4091 [-Werror=format-truncation=] 388 | snprintf(cmd, sizeof(cmd), "diff %s %s", | ^~ ...... 456 | ret = init_src_info(b_path, a_path, rel_path, &node->info); | ~~~~~~ In file included from /usr/include/stdio.h:867, from util/srclist.c:8: /usr/include/bits/stdio2.h:67:10: note: ‘__builtin___snprintf_chk’ output between 7 and 8197 bytes into a destination of size 4096 67 | return __builtin___snprintf_chk (__s, __n, __USE_FORTIFY_LEVEL - 1, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 68 | __bos (__s), __fmt, __va_arg_pack ()); | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors mv: cannot stat 'util/.srclist.o.tmp': No such file or directory make[4]: *** [/home/jolsa/kernel/linux-perf/tools/build/Makefile.build:97: util/srclist.o] Error 1 make[3]: *** [/home/jolsa/kernel/linux-perf/tools/build/Makefile.build:139: util] Error 2 make[2]: *** [Makefile.perf:617: perf-in.o] Error 2 make[1]: *** [Makefile.perf:225: sub-make] Error 2 make: *** [Makefile:70: all] Error 2
Quoted 5 lines
[jolsa@krava ~]$ gcc --version
gcc (GCC) 9.3.1 20200317 (Red Hat 9.3.1-1)jirka
Can you help to add following patch on top of the patch-set? Looks we need to check the return value of snprintf for truncation checking.
jinyao@kbl:~/kbl-ws/perf-dev/lck-7589/acme/tools/perf$ git diff
tools/perf/util/srclist.c +6 -2
@@ -385,8 +385,12 @@ static int src_info__create_line_mapping(struct src_info *info, char *b_path, goto out; } -snprintf (cmd,sizeof (cmd),"diff %s %s" , - b_path, a_path); + ret =snprintf (cmd,PATH_MAX ,"diff %s %s" , + b_path, a_path); +if (ret ==PATH_MAX ) { + ret =-1 ; + goto out; + }pr_debug ("Execute '%s'\n" , cmd);
Thanks
Jin Yao
Quoted 35 lines · Yao
Hi Jiri,
Quoted 32 lines · Jiri Olsa
Quoted 3 lines · Jin Yao
Sometimes, a small change in a hot function reducing the cycles of this function, but the overall workload doesn't get faster. It is interesting where the cycles are moved to.
I'm getting compilation fail:
BUILD: Doing 'make -j1' parallel build CC util/srclist.o util/srclist.c: In function ‘srclist__node_new’: util/srclist.c:388:35: error: ‘%s’ directive output may be truncated writing up to 4095 bytes into a region of size 4091 [-Werror=format-truncation=] 388 | snprintf(cmd, sizeof(cmd), "diff %s %s", | ^~ ...... 456 | ret = init_src_info(b_path, a_path, rel_path, &node->info); | ~~~~~~ In file included from /usr/include/stdio.h:867, from util/srclist.c:8: /usr/include/bits/stdio2.h:67:10: note: ‘__builtin___snprintf_chk’ output between 7 and 8197 bytes into a destination of size 4096 67 | return __builtin___snprintf_chk (__s, __n, __USE_FORTIFY_LEVEL - 1, | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 68 | __bos (__s), __fmt, __va_arg_pack ()); | ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors mv: cannot stat 'util/.srclist.o.tmp': No such file or directory make[4]: *** [/home/jolsa/kernel/linux-perf/tools/build/Makefile.build:97: util/srclist.o] Error 1 make[3]: *** [/home/jolsa/kernel/linux-perf/tools/build/Makefile.build:139: util] Error 2 make[2]: *** [Makefile.perf:617: perf-in.o] Error 2 make[1]: *** [Makefile.perf:225: sub-make] Error 2 make: *** [Makefile:70: all] Error 2
Quoted 8 lines
Quoted 5 lines
[jolsa@krava ~]$ gcc --version
gcc (GCC) 9.3.1 20200317 (Red Hat 9.3.1-1)jirka
Can you help to add following patch on top of the patch-set? Looks we need to check the return value of snprintf for truncation checking.
yep, that helped.. I'll apply it on top for now
and try to do the review soon ;-)
thanks,
jirka
Quoted 25 lines
jinyao@kbl:~/kbl-ws/perf-dev/lck-7589/acme/tools/perf$ git difftools/perf/util/srclist.c +6
-2@@ -385,8 +385,12 @@ static int src_info__create_line_mapping(struct src_info *info, char *b_path, goto out; } -snprintf (cmd,sizeof (cmd),"diff %s %s" , - b_path, a_path); + ret =snprintf (cmd,PATH_MAX ,"diff %s %s" , + b_path, a_path); +if (ret ==PATH_MAX ) { + ret =-1 ; + goto out; + }pr_debug ("Execute '%s'\n" , cmd);Thanks
Jin Yao