Skip to contentSkip to conversation

perf mem: improves DSO long names search speed with RB tree

PATCH Superseded Waiman Long 3 participants 6 messages Sep 16, 2014 – Sep 17, 2014 perf events

v2 of 6 Superseded Latest is v6 →

lore
Share & tools
lore
#0 lore raw

v2 cover perf mem: improves DSO long names search speed with RB tree

v1->v2:
- Rename DSO longname RBtree find function to segregate its two different uses of searching and linking DSO into RB tree.

This patch addes a RBtree to sort the DSO structures of the perf tool by its long name so that searching for a match will be much quicker especially if a large number of DSOs are beining profiled.

Waiman Long (1):
perf mem: improves DSO long names search speed with RB tree

 tools/perf/util/dso.c |   87 ++++++++++++++++++++++++++++++++++++++++++++++--
 tools/perf/util/dso.h |    1 +
 2 files changed, 84 insertions(+), 4 deletions(-)
#1 lore raw

PATCH v2 1/1 perf mem: improves DSO long names search speed with RB tree

With workload that spawns and destroys many threads and processes, it was found that perf-mem could took a long time to post-process the perf data after the target workload had completed its operation. The performance bottleneck was found to be searching and insertion of the new DSO structures (thousands of them in this case).

In a dual-socket Ivy-Bridge E7-4890 v2 machine (30-core, 60-thread), the perf profile below shows what perf was doing after the profiled AIM7 shared workload completed:

- 83.94% perf libc-2.11.3.so [.] __strcmp_sse42
- __strcmp_sse42
- 99.82% map__new
machine__process_mmap_event
perf_session_deliver_event
perf_session__process_event
__perf_session__process_events
cmd_record
cmd_mem
run_builtin
main
__libc_start_main
- 13.17% perf perf [.] __dsos__findnew

     __dsos__findnew
     map__new
     machine__process_mmap_event
     perf_session_deliver_event
     perf_session__process_event
     __perf_session__process_events
     cmd_record
     cmd_mem
     run_builtin
     main
     __libc_start_main

So about 97% of CPU times were spent in the map__new() function trying to insert new DSO entry into the DSO linked list. The whole post-processing step took about 9 minutes.

The DSO structures are currently searched linearly. So the total processing time will be proportional to n^2.

To overcome this performance problem, the DSO code is modified to also put the DSO structures in a RB tree sorted by its long name in additional to being in a simple linked list. With this change, the processing time will become proportional to n*log(n) which will be much quicker for large n. However, the short name will still be searched using the old linear searching method which is slow. With that patch in place, the same perf-mem post-processing step took less than 30 seconds to complete.

---
 tools/perf/util/dso.c |   87 ++++++++++++++++++++++++++++++++++++++++++++++--
 tools/perf/util/dso.h |    1 +
 2 files changed, 84 insertions(+), 4 deletions(-)
tools/perf/util/dso.c +83 -4
index 819f104..fccb2f0 100644
@@ -611,17 +611,93 @@ struct dso *dso__kernel_findnew(struct machine *machine, const char *name,
 	return dso;
 }
 
+/*
+ * RB root of DSOs sorted by the long name
+ */
+static struct rb_root dso__longname_root = { NULL };
+
+/*
+ * Find a matching entry and/or link current entry to RB tree.
+ * Either one of the dso or name parameter must be non-NULL or the
+ * function will not work.
+ */
+static struct dso *
+dso__findlink_by_longname(struct dso *dso, const char *name)
+{
+	struct rb_node **p = &dso__longname_root.rb_node;
+	struct rb_node  *parent = NULL;
+	int warned = false;
+
+	if (!name)
+		name = dso->long_name;
+	/*
+	 * Find node with the matching name
+	 */
+	while (*p) {
+		struct dso *this = rb_entry(*p, struct dso, longname_rb_node);
+		long rc = (long)strcmp(name, this->long_name);
+
+		parent = *p;
+		if (rc == 0) {
+			/*
+			 * In case the new DSO is a duplicate of an existing
+			 * one, print an one-time warning & sort the entry
+			 * by its DSO address.
+			 */
+			if (!dso || (dso == this))
+				return this;	/* Find matching dso */
+			if (!warned) {
+				pr_warning("Duplicated dso long name: %s\n",
+					   name);
+				warned = true;
+			}
+			rc = (long)dso - (long)this;
+		}
+		if (rc < 0)
+			p = &parent->rb_left;
+		else
+			p = &parent->rb_right;
+	}
+	if (dso) {
+		/* Add new node and rebalance tree */
+		rb_link_node(&dso->longname_rb_node, parent, p);
+		rb_insert_color(&dso->longname_rb_node, &dso__longname_root);
+	}
+	return NULL;
+}
+
+static inline struct dso *
+dso__find_by_longname(const char *name)
+{
+	return dso__findlink_by_longname(NULL, name);
+}
+
+/*
+ * Unlink the longname-sorted RB tree node
+ */
+static inline void dso__unlink_longname_node(struct dso *dso)
+{
+	rb_erase(&dso->longname_rb_node, &dso__longname_root);
+}
+
 void dso__set_long_name(struct dso *dso, const char *name, bool name_allocated)
 {
 	if (name == NULL)
 		return;
 
+	if (dso->long_name) {
+		if (!strcmp(dso->long_name, name))
+			return;
+		dso__unlink_longname_node(dso);
+	}
+
 	if (dso->long_name_allocated)
 		free((char *)dso->long_name);
 
 	dso->long_name		 = name;
 	dso->long_name_len	 = strlen(name);
 	dso->long_name_allocated = name_allocated;
+	(void)dso__findlink_by_longname(dso, name);
 }
 
 void dso__set_short_name(struct dso *dso, const char *name, bool name_allocated)
@@ -695,6 +771,8 @@ struct dso *dso__new(const char *name)
 	if (dso != NULL) {
 		int i;
 		strcpy(dso->name, name);
+		RB_CLEAR_NODE(&dso->longname_rb_node);
+		dso->long_name = NULL;
 		dso__set_long_name(dso, dso->name, false);
 		dso__set_short_name(dso, dso->name, false);
 		for (i = 0; i < MAP__NR_TYPES; ++i)
@@ -733,6 +811,10 @@ void dso__delete(struct dso *dso)
 		zfree((char **)&dso->long_name);
 		dso->long_name_allocated = false;
 	}
+	if (dso->long_name) {
+		dso__unlink_longname_node(dso);
+		dso->long_name = NULL;
+	}
 
 	dso__data_close(dso);
 	dso_cache__free(&dso->data.cache);
@@ -822,10 +904,7 @@ struct dso *dsos__find(const struct list_head *head, const char *name, bool cmp_
 				return pos;
 		return NULL;
 	}
-	list_for_each_entry(pos, head, node)
-		if (strcmp(pos->long_name, name) == 0)
-			return pos;
-	return NULL;
+	return dso__find_by_longname(name);
 }
 
 struct dso *__dsos__findnew(struct list_head *head, const char *name)
tools/perf/util/dso.h +1 -0
index ad553ba..81a75ee 100644
@@ -76,6 +76,7 @@ struct dso {
 	struct list_head node;
 	struct rb_root	 symbols[MAP__NR_TYPES];
 	struct rb_root	 symbol_names[MAP__NR_TYPES];
+	struct rb_node   longname_rb_node;
 	void		 *a2l;
 	char		 *symsrc_filename;
 	unsigned int	 a2l_fails;
Signature · 1 line
1.7.1
reviewer
#2 lore raw
Quoted 62 lines · Waiman Long

With workload that spawns and destroys many threads and processes, it was found that perf-mem could took a long time to post-process the perf data after the target workload had completed its operation. The performance bottleneck was found to be searching and insertion of the new DSO structures (thousands of them in this case).

In a dual-socket Ivy-Bridge E7-4890 v2 machine (30-core, 60-thread), the perf profile below shows what perf was doing after the profiled AIM7 shared workload completed:

(quoted diff) +0 -2
-     83.94%  perf  libc-2.11.3.so     [.] __strcmp_sse42
   - __strcmp_sse42
      - 99.82% map__new
           machine__process_mmap_event
           perf_session_deliver_event
           perf_session__process_event
           __perf_session__process_events
           cmd_record
           cmd_mem
           run_builtin
           main
           __libc_start_main
-     13.17%  perf  perf               [.] __dsos__findnew
     __dsos__findnew
     map__new
     machine__process_mmap_event
     perf_session_deliver_event
     perf_session__process_event
     __perf_session__process_events
     cmd_record
     cmd_mem
     run_builtin
     main
     __libc_start_main

So about 97% of CPU times were spent in the map__new() function trying to insert new DSO entry into the DSO linked list. The whole post-processing step took about 9 minutes.

The DSO structures are currently searched linearly. So the total processing time will be proportional to n^2.

To overcome this performance problem, the DSO code is modified to also put the DSO structures in a RB tree sorted by its long name in additional to being in a simple linked list. With this change, the processing time will become proportional to n*log(n) which will be much quicker for large n. However, the short name will still be searched using the old linear searching method which is slow. With that patch in place, the same perf-mem post-processing step took less than 30 seconds to complete.

---
 tools/perf/util/dso.c |   87 ++++++++++++++++++++++++++++++++++++++++++++++--
 tools/perf/util/dso.h |    1 +
 2 files changed, 84 insertions(+), 4 deletions(-)
tools/perf/util/dso.c +0 -0
index 819f104..fccb2f0 100644

tools/perf/util/dso.c struct dso *dso__kernel_findnew(struct machine *machine, const char *name, · line 617

Quoted 7 lines
tools/perf/util/dso.c +4 -0
@@ -611,17 +611,93 @@ struct dso *dso__kernel_findnew(struct machine *machine, const char *name,
 	return dso;
 }
 
+/*
+ * RB root of DSOs sorted by the long name
+ */
+static struct rb_root dso__longname_root = { NULL };

Global variable!

Why not just change the lists to rbtrees i.e.

tools/perf/util/machine.h +2 -2
index 6a6bcc1..fa30780 100644                                                                                                                               
@@ -32,8 +32,8 @@ struct machine {                                                                                                                          
        struct list_head  dead_threads;                                                                                                                     
        struct thread     *last_match;                                                                                                                      
        struct vdso_info  *vdso_info;                                                                                                                       
-       struct list_head  user_dsos;                                                                                                                        
-       struct list_head  kernel_dsos;                                                                                                                      
+       struct rb_root    user_dsos;                                                                                                                        
+       struct rb_root    kernel_dsos;                                                                                                                      
        struct map_groups kmaps;                                                                                                                            
        struct map        *vmlinux_maps[MAP__NR_TYPES];                                                                                                     
        u64               kernel_start;

And make all the resulting adjustments.

#3 lore raw
Quoted 66 lines · Adrian Hunter
Quoted 65 lines · Waiman Long

With workload that spawns and destroys many threads and processes, it was found that perf-mem could took a long time to post-process the perf data after the target workload had completed its operation. The performance bottleneck was found to be searching and insertion of the new DSO structures (thousands of them in this case).

In a dual-socket Ivy-Bridge E7-4890 v2 machine (30-core, 60-thread), the perf profile below shows what perf was doing after the profiled AIM7 shared workload completed:

(quoted diff) +0 -2
-     83.94%  perf  libc-2.11.3.so     [.] __strcmp_sse42
    - __strcmp_sse42
       - 99.82% map__new
            machine__process_mmap_event
            perf_session_deliver_event
            perf_session__process_event
            __perf_session__process_events
            cmd_record
            cmd_mem
            run_builtin
            main
            __libc_start_main
-     13.17%  perf  perf               [.] __dsos__findnew
      __dsos__findnew
      map__new
      machine__process_mmap_event
      perf_session_deliver_event
      perf_session__process_event
      __perf_session__process_events
      cmd_record
      cmd_mem
      run_builtin
      main
      __libc_start_main

So about 97% of CPU times were spent in the map__new() function trying to insert new DSO entry into the DSO linked list. The whole post-processing step took about 9 minutes.

The DSO structures are currently searched linearly. So the total processing time will be proportional to n^2.

To overcome this performance problem, the DSO code is modified to also put the DSO structures in a RB tree sorted by its long name in additional to being in a simple linked list. With this change, the processing time will become proportional to n*log(n) which will be much quicker for large n. However, the short name will still be searched using the old linear searching method which is slow. With that patch in place, the same perf-mem post-processing step took less than 30 seconds to complete.

---
  tools/perf/util/dso.c |   87 ++++++++++++++++++++++++++++++++++++++++++++++--
  tools/perf/util/dso.h |    1 +
  2 files changed, 84 insertions(+), 4 deletions(-)
tools/perf/util/dso.c +0 -0
index 819f104..fccb2f0 100644
@@ -611,17 +611,93 @@ struct dso *dso__kernel_findnew(struct machine *machine, const char *name,
  	return dso;
  }
 
Quoted 7 lines

tools/perf/util/dso.c struct dso *dso__kernel_findnew(struct machine *machine, const char *name, · line 617

Quoted 4 lines
tools/perf/util/dso.c +4 -0
+/*
+ * RB root of DSOs sorted by the long name
+ */
+static struct rb_root dso__longname_root = { NULL };

Global variable!

Why not just change the lists to rbtrees i.e.

The DSOs can be searched either by its short name or its long name. The patch enables long name search with RB tree. The short name search is still using the linked list as the sort order will be different from the long name.

I think I can remove the linked list and do the short name search linearly by scanning from the leftmost to the rightmost in the RBtree. In that way, we can remove the redundant list head field in the DSO structure.

Cheers,
Longman

#4 lore raw
In reply to Waiman Long

Em Tue, Sep 16, 2014 at 03:08:19PM -0400, Waiman Long escreveu:

tools/perf/util/dso.c struct dso *dso__kernel_findnew(struct machine *machine, const char *name, · line 617

Quoted 9 lines

+++ b/tools/perf/util/dso.c

tools/perf/util/dso.c +4 -0
@@ -611,17 +611,93 @@ struct dso *dso__kernel_findnew(struct machine *machine, const char *name,
 	return dso;
 }
 
+/*
+ * RB root of DSOs sorted by the long name
+ */
+static struct rb_root dso__longname_root = { NULL };

Use RB_ROOT, like in:

[acme@zoo linux]$ grep -w rb_root mm/vmalloc.c
static struct rb_root vmap_area_root = RB_ROOT;
[acme@zoo linux]$

But then can't this be made non-static, i.e. at the 'struct machine' level? I.e. it is more likely that DSOs with the same long name are really the same thing on a single 'struct machine', not accross multiple ones or even multiple sessions (i.e. across multiple 'struct machines').

IIRC Adrian also pointed this out.

- Arnaldo

Quoted 131 lines
tools/perf/util/dso.c +78 -4
+/*
+ * Find a matching entry and/or link current entry to RB tree.
+ * Either one of the dso or name parameter must be non-NULL or the
+ * function will not work.
+ */
+static struct dso *
+dso__findlink_by_longname(struct dso *dso, const char *name)
+{
+	struct rb_node **p = &dso__longname_root.rb_node;
+	struct rb_node  *parent = NULL;
+	int warned = false;
+
+	if (!name)
+		name = dso->long_name;
+	/*
+	 * Find node with the matching name
+	 */
+	while (*p) {
+		struct dso *this = rb_entry(*p, struct dso, longname_rb_node);
+		long rc = (long)strcmp(name, this->long_name);
+
+		parent = *p;
+		if (rc == 0) {
+			/*
+			 * In case the new DSO is a duplicate of an existing
+			 * one, print an one-time warning & sort the entry
+			 * by its DSO address.
+			 */
+			if (!dso || (dso == this))
+				return this;	/* Find matching dso */
+			if (!warned) {
+				pr_warning("Duplicated dso long name: %s\n",
+					   name);
+				warned = true;
+			}
+			rc = (long)dso - (long)this;
+		}
+		if (rc < 0)
+			p = &parent->rb_left;
+		else
+			p = &parent->rb_right;
+	}
+	if (dso) {
+		/* Add new node and rebalance tree */
+		rb_link_node(&dso->longname_rb_node, parent, p);
+		rb_insert_color(&dso->longname_rb_node, &dso__longname_root);
+	}
+	return NULL;
+}
+
+static inline struct dso *
+dso__find_by_longname(const char *name)
+{
+	return dso__findlink_by_longname(NULL, name);
+}
+
+/*
+ * Unlink the longname-sorted RB tree node
+ */
+static inline void dso__unlink_longname_node(struct dso *dso)
+{
+	rb_erase(&dso->longname_rb_node, &dso__longname_root);
+}
+
 void dso__set_long_name(struct dso *dso, const char *name, bool name_allocated)
 {
 	if (name == NULL)
 		return;
 
+	if (dso->long_name) {
+		if (!strcmp(dso->long_name, name))
+			return;
+		dso__unlink_longname_node(dso);
+	}
+
 	if (dso->long_name_allocated)
 		free((char *)dso->long_name);
 
 	dso->long_name		 = name;
 	dso->long_name_len	 = strlen(name);
 	dso->long_name_allocated = name_allocated;
+	(void)dso__findlink_by_longname(dso, name);
 }
 
 void dso__set_short_name(struct dso *dso, const char *name, bool name_allocated)
@@ -695,6 +771,8 @@ struct dso *dso__new(const char *name)
 	if (dso != NULL) {
 		int i;
 		strcpy(dso->name, name);
+		RB_CLEAR_NODE(&dso->longname_rb_node);
+		dso->long_name = NULL;
 		dso__set_long_name(dso, dso->name, false);
 		dso__set_short_name(dso, dso->name, false);
 		for (i = 0; i < MAP__NR_TYPES; ++i)
@@ -733,6 +811,10 @@ void dso__delete(struct dso *dso)
 		zfree((char **)&dso->long_name);
 		dso->long_name_allocated = false;
 	}
+	if (dso->long_name) {
+		dso__unlink_longname_node(dso);
+		dso->long_name = NULL;
+	}
 
 	dso__data_close(dso);
 	dso_cache__free(&dso->data.cache);
@@ -822,10 +904,7 @@ struct dso *dsos__find(const struct list_head *head, const char *name, bool cmp_
 				return pos;
 		return NULL;
 	}
-	list_for_each_entry(pos, head, node)
-		if (strcmp(pos->long_name, name) == 0)
-			return pos;
-	return NULL;
+	return dso__find_by_longname(name);
 }
 
 struct dso *__dsos__findnew(struct list_head *head, const char *name)
tools/perf/util/dso.h +1 -0
index ad553ba..81a75ee 100644
@@ -76,6 +76,7 @@ struct dso {
 	struct list_head node;
 	struct rb_root	 symbols[MAP__NR_TYPES];
 	struct rb_root	 symbol_names[MAP__NR_TYPES];
+	struct rb_node   longname_rb_node;
 	void		 *a2l;
 	char		 *symsrc_filename;
 	unsigned int	 a2l_fails;
Signature · 1 line
1.7.1
#5 lore raw
Quoted 12 lines · Arnaldo Carvalho de Melo

Em Tue, Sep 16, 2014 at 03:08:19PM -0400, Waiman Long escreveu:

tools/perf/util/dso.c struct dso *dso__kernel_findnew(struct machine *machine, const char *name, · line 617

Quoted 9 lines

+++ b/tools/perf/util/dso.c

tools/perf/util/dso.c +4 -0
@@ -611,17 +611,93 @@ struct dso *dso__kernel_findnew(struct machine *machine, const char *name,
  	return dso;
  }
 
+/*
+ * RB root of DSOs sorted by the long name
+ */
+static struct rb_root dso__longname_root = { NULL };

Use RB_ROOT, like in:

Quoted 3 lines

[acme@zoo linux]$ grep -w rb_root mm/vmalloc.c
static struct rb_root vmap_area_root = RB_ROOT;
[acme@zoo linux]$

I don't use RB_ROOT here because it gave compilation error.

Quoted 10 lines

But then can't this be made non-static, i.e. at the 'struct machine' level? I.e. it is more likely that DSOs with the same long name are really the same thing on a single 'struct machine', not accross multiple ones or even multiple sessions (i.e. across multiple 'struct machines').

IIRC Adrian also pointed this out.

- Arnaldo

Yes, I am going to put the rb_root in the machine structure level as suggested by Adrian. So the static rb_root will be gone. I also found out that the dso__load_sym() function in util/symbol-elf.c may create DSOs of the same long name which can be either "[kernel.allsyms]" or "/lib/modules/.../vmlinux".

Cheers,
Longman