Skip to contentSkip to conversation

perf top: demo of how to use the sysfs interface

RFC PATCH In review Lin Ming 3 participants 10 messages May 18, 2010 – May 21, 2010 perf events

lore
Share & tools
lore

Lin Ming

author
#0 lore raw

RFC PATCH v2 11/11 perf top: demo of how to use the sysfs interface

Just a temporary patch to show how to use the pmu sysfs interface...

---
 tools/perf/builtin-top.c |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)
tools/perf/builtin-top.c +13 -0
index adc179d..eaa9405 100644
@@ -1206,6 +1206,7 @@ static void start_counter(int i, int counter)
 	struct perf_event_attr *attr;
 	int cpu;
 	int thread_index;
+	int sys_fd;
 
 	cpu = profile_cpu;
 	if (target_tid == -1 && profile_cpu == -1)
@@ -1226,9 +1227,21 @@ static void start_counter(int i, int counter)
 
 	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
+		/*
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the <event-name> and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */
+		sys_fd = open("/sys/devices/system/cpu/event_source/id", O_RDONLY);
+		if (sys_fd != -1)
+			attr->type |= (sys_fd << PERF_EVENT_ATTR_TYPE_BITS) | PERF_EVENT_FD;
+
 		fd[i][counter][thread_index] = sys_perf_event_open(attr,
 				all_tids[thread_index], cpu, group_fd, 0);
 
+		if (sys_fd != -1)
+			close(sys_fd);
+
 		if (fd[i][counter][thread_index] < 0) {
 			int err = errno;
 
#1 lore raw
Quoted 20 lines · Lin Ming

Just a temporary patch to show how to use the pmu sysfs interface...

---
 tools/perf/builtin-top.c |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)
tools/perf/builtin-top.c +1 -0
index adc179d..eaa9405 100644
@@ -1206,6 +1206,7 @@ static void start_counter(int i, int counter)
 	struct perf_event_attr *attr;
 	int cpu;
 	int thread_index;
+	int sys_fd;
 
 	cpu = profile_cpu;
 	if (target_tid == -1 && profile_cpu == -1)

tools/perf/builtin-top.c static void start_counter(int i, int counter) · line 1234

Quoted 8 lines
tools/perf/builtin-top.c +5 -0
@@ -1226,9 +1227,21 @@ static void start_counter(int i, int counter)
 
 	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
+		/*
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the <event-name> and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */

In this above case, does this sys_fd also specify the event I am going to open, in addition to its event source? I'd assume not since event_source is a symlink to /sys/devices/system/cpu/event_source (right?)

How do I specify the exact event id via the sysfs interface?

Thanks,

- Corey

author
#2 lore raw
Quoted 23 lines · Corey Ashford
Quoted 21 lines · Lin Ming

Just a temporary patch to show how to use the pmu sysfs interface...

---
 tools/perf/builtin-top.c |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)
tools/perf/builtin-top.c +1 -0
index adc179d..eaa9405 100644
@@ -1206,6 +1206,7 @@ static void start_counter(int i, int counter)
 	struct perf_event_attr *attr;
 	int cpu;
 	int thread_index;
+	int sys_fd;
 
 	cpu = profile_cpu;
 	if (target_tid == -1 && profile_cpu == -1)
@@ -1226,9 +1227,21 @@ static void start_counter(int i, int counter)
 
Quoted 9 lines

tools/perf/builtin-top.c static void start_counter(int i, int counter) · line 1234

Quoted 7 lines
tools/perf/builtin-top.c +5 -0
 	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
+		/*
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the <event-name> and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */

In this above case, does this sys_fd also specify the event I am going to open, in addition to its event source? I'd assume not since event_source is a symlink to /sys/devices/system/cpu/event_source (right?)

Right, this sys_fd only specifies the event source.

Quoted 1 line

How do I specify the exact event id via the sysfs interface?

/sys/devices/system/cpu/events/<event-name>/id

But in this patch series, the event id sysfs interface is not used yet.

Thanks,
Lin Ming

Quoted 5 lines

Thanks,

- Corey

#3 lore raw
Quoted 34 lines · Lin Ming
Quoted 23 lines · Corey Ashford
Quoted 21 lines · Lin Ming

Just a temporary patch to show how to use the pmu sysfs interface...

---
 tools/perf/builtin-top.c |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)
tools/perf/builtin-top.c +1 -0
index adc179d..eaa9405 100644
@@ -1206,6 +1206,7 @@ static void start_counter(int i, int counter)
 	struct perf_event_attr *attr;
 	int cpu;
 	int thread_index;
+	int sys_fd;
 
 	cpu = profile_cpu;
 	if (target_tid == -1 && profile_cpu == -1)
@@ -1226,9 +1227,21 @@ static void start_counter(int i, int counter)
 
Quoted 9 lines

tools/perf/builtin-top.c static void start_counter(int i, int counter) · line 1234

Quoted 7 lines
tools/perf/builtin-top.c +5 -0
 	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
+		/*
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the <event-name> and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */

In this above case, does this sys_fd also specify the event I am going to open, in addition to its event source? I'd assume not since event_source is a symlink to /sys/devices/system/cpu/event_source (right?)

Quoted 7 lines

Right, this sys_fd only specifies the event source.

Quoted 1 line

How do I specify the exact event id via the sysfs interface?

/sys/devices/system/cpu/events/<event-name>/id

But in this patch series, the event id sysfs interface is not used yet.

So, I would open that id and then read the id code and place it in attr->config or maybe place the fd into attr (somewhere) ?

We also need to take into account event "attributes" - other data that is needed to configure a specific event. For example, think about a memory controller which has a PMU can count events in a particular memory range; we need to be able to supply the memory range somehow, and I don't think that can be accomplished by passing in the fd of a sysfs file that we've opened.

- Corey

author
#4 lore raw
Quoted 37 lines · Corey Ashford
Quoted 36 lines · Lin Ming
Quoted 23 lines · Corey Ashford
Quoted 21 lines · Lin Ming

Just a temporary patch to show how to use the pmu sysfs interface...

---
 tools/perf/builtin-top.c |   13 +++++++++++++
 1 files changed, 13 insertions(+), 0 deletions(-)
tools/perf/builtin-top.c +1 -0
index adc179d..eaa9405 100644
@@ -1206,6 +1206,7 @@ static void start_counter(int i, int counter)
 	struct perf_event_attr *attr;
 	int cpu;
 	int thread_index;
+	int sys_fd;
 
 	cpu = profile_cpu;
 	if (target_tid == -1 && profile_cpu == -1)
@@ -1226,9 +1227,21 @@ static void start_counter(int i, int counter)
 
Quoted 9 lines

tools/perf/builtin-top.c static void start_counter(int i, int counter) · line 1234

Quoted 7 lines
tools/perf/builtin-top.c +5 -0
 	for (thread_index = 0; thread_index < thread_num; thread_index++) {
 try_again:
+		/*
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the <event-name> and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */

In this above case, does this sys_fd also specify the event I am going to open, in addition to its event source? I'd assume not since event_source is a symlink to /sys/devices/system/cpu/event_source (right?)

Right, this sys_fd only specifies the event source.

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

How do I specify the exact event id via the sysfs interface?

/sys/devices/system/cpu/events/<event-name>/id

But in this patch series, the event id sysfs interface is not used yet.

So, I would open that id and then read the id code and place it in attr->config or maybe place the fd into attr (somewhere) ?

Place the id code in attr->config.

Quoted 3 lines

We also need to take into account event "attributes" - other data that is needed to configure a specific event. For example, think about a memory controller which has a PMU can count events in a particular memory range; we need to be able to supply the memory range somehow, and I don't think that can be accomplished by passing in the fd of a sysfs file that we've opened.

Each event is a directory in the sysfs, so we can put all the event "attributes" under it.

For your example, /sys/devices/system/node/events/<event-name>/id /sys/devices/system/node/events/<event-name>/memory_range
....

Then we can read these attributes and pass the value into the syscall.

Quoted 2 lines

- Corey

#5 lore raw
Quoted 56 lines · Lin Ming
Quoted 37 lines · Corey Ashford
Quoted 36 lines · Lin Ming
Quoted 23 lines · Corey Ashford
Quoted 21 lines · Lin Ming

Just a temporary patch to show how to use the pmu sysfs interface...

---
  tools/perf/builtin-top.c |   13 +++++++++++++
  1 files changed, 13 insertions(+), 0 deletions(-)
tools/perf/builtin-top.c +1 -0
index adc179d..eaa9405 100644
@@ -1206,6 +1206,7 @@ static void start_counter(int i, int counter)
  	struct perf_event_attr *attr;
  	int cpu;
  	int thread_index;
+	int sys_fd;
 
  	cpu = profile_cpu;
  	if (target_tid == -1&&  profile_cpu == -1)
@@ -1226,9 +1227,21 @@ static void start_counter(int i, int counter)
 
Quoted 9 lines

tools/perf/builtin-top.c static void start_counter(int i, int counter) · line 1234

Quoted 7 lines
tools/perf/builtin-top.c +5 -0
  	for (thread_index = 0; thread_index<  thread_num; thread_index++) {
  try_again:
+		/*
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the<event-name>  and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */

In this above case, does this sys_fd also specify the event I am going to open, in addition to its event source? I'd assume not since event_source is a symlink to /sys/devices/system/cpu/event_source (right?)

Right, this sys_fd only specifies the event source.

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

How do I specify the exact event id via the sysfs interface?

/sys/devices/system/cpu/events/<event-name>/id

But in this patch series, the event id sysfs interface is not used yet.

So, I would open that id and then read the id code and place it in attr->config or maybe place the fd into attr (somewhere) ?

Place the id code in attr->config.

Quoted 3 lines

We also need to take into account event "attributes" - other data that is needed to configure a specific event. For example, think about a memory controller which has a PMU can count events in a particular memory range; we need to be able to supply the memory range somehow, and I don't think that can be accomplished by passing in the fd of a sysfs file that we've opened.

Each event is a directory in the sysfs, so we can put all the event "attributes" under it.

Quoted 7 lines

For your example, /sys/devices/system/node/events/<event-name>/id /sys/devices/system/node/events/<event-name>/memory_range
....

Then we can read these attributes and pass the value into the syscall.

I'm not sure I made the example clear.

Let's say I have a memory controller event called memory_write and it has two attributes: low_addr and high_addr... writes to addresses between the low_addr and high_addr will increment the counter.

As a user, I want to be able to specify a particular memory range, let's say 0x1000000..0x2000000

A sysfs structure like this might be constructed:
/sys/devices/system/node/events/memory_write /sys/devices/system/node/events/memory_write/attr /sys/devices/system/node/events/memory_write/attr/low_addr /sys/devices/system/node/events/memory_write/attr/low_addr/min /sys/devices/system/node/events/memory_write/attr/low_addr/max

In another posting I had also added a bit shift value, but there are so few bits left in the attr->config, that I'm not sure this is a very extensible mechanism, but just for the sake of illustration of the basic idea, I'll add it here:
/sys/devices/system/node/events/memory_write/attr/low_addr/shift

Then the same thing is repeated for the other attribute:
/sys/devices/system/node/events/memory_write/attr/high_addr /sys/devices/system/node/events/memory_write/attr/high_addr/min /sys/devices/system/node/events/memory_write/attr/high_addr/max /sys/devices/system/node/events/memory_write/attr/high_addr/shift

In this scenario, a user tool (like perf) would be able to see that there are attributes associated with the memory_write event, and it knows the names and range of allowed values for these attributes. The shift value tells the tool how much to shift the attribute value before OR'ing it into the attr->config value.

If we find that more than 64 bits are needed for the event code plus the attribute values, perhaps shift values greater than 64 would denote to place the attribute bits into a new attr field, like attr->config_extra[shift / 64]

On the perf command line, then, a user could specify something like:

perf stat -e node::memory_write:low_addr=0x1000000:high_addr=0x2000000

What do you think?

- Corey

author
#6 lore raw
Quoted 93 lines · Corey Ashford
Quoted 56 lines · Lin Ming
Quoted 37 lines · Corey Ashford
Quoted 36 lines · Lin Ming
Quoted 23 lines · Corey Ashford
Quoted 21 lines · Lin Ming

Just a temporary patch to show how to use the pmu sysfs interface...

---
  tools/perf/builtin-top.c |   13 +++++++++++++
  1 files changed, 13 insertions(+), 0 deletions(-)
tools/perf/builtin-top.c +1 -0
index adc179d..eaa9405 100644
@@ -1206,6 +1206,7 @@ static void start_counter(int i, int counter)
  	struct perf_event_attr *attr;
  	int cpu;
  	int thread_index;
+	int sys_fd;
 
  	cpu = profile_cpu;
  	if (target_tid == -1&&  profile_cpu == -1)
@@ -1226,9 +1227,21 @@ static void start_counter(int i, int counter)
 
Quoted 9 lines

tools/perf/builtin-top.c static void start_counter(int i, int counter) · line 1234

Quoted 7 lines
tools/perf/builtin-top.c +5 -0
  	for (thread_index = 0; thread_index<  thread_num; thread_index++) {
  try_again:
+		/*
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the<event-name>  and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */

In this above case, does this sys_fd also specify the event I am going to open, in addition to its event source? I'd assume not since event_source is a symlink to /sys/devices/system/cpu/event_source (right?)

Right, this sys_fd only specifies the event source.

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

How do I specify the exact event id via the sysfs interface?

/sys/devices/system/cpu/events/<event-name>/id

But in this patch series, the event id sysfs interface is not used yet.

So, I would open that id and then read the id code and place it in attr->config or maybe place the fd into attr (somewhere) ?

Place the id code in attr->config.

Quoted 3 lines

We also need to take into account event "attributes" - other data that is needed to configure a specific event. For example, think about a memory controller which has a PMU can count events in a particular memory range; we need to be able to supply the memory range somehow, and I don't think that can be accomplished by passing in the fd of a sysfs file that we've opened.

Each event is a directory in the sysfs, so we can put all the event "attributes" under it.

Quoted 7 lines

For your example, /sys/devices/system/node/events/<event-name>/id /sys/devices/system/node/events/<event-name>/memory_range
....

Then we can read these attributes and pass the value into the syscall.

I'm not sure I made the example clear.

Let's say I have a memory controller event called memory_write and it has two attributes: low_addr and high_addr... writes to addresses between the low_addr and high_addr will increment the counter.

As a user, I want to be able to specify a particular memory range, let's say 0x1000000..0x2000000

A sysfs structure like this might be constructed:
/sys/devices/system/node/events/memory_write /sys/devices/system/node/events/memory_write/attr /sys/devices/system/node/events/memory_write/attr/low_addr /sys/devices/system/node/events/memory_write/attr/low_addr/min /sys/devices/system/node/events/memory_write/attr/low_addr/max

In another posting I had also added a bit shift value, but there are so few bits left in the attr->config, that I'm not sure this is a very extensible mechanism, but just for the sake of illustration of the basic idea, I'll add it here:
/sys/devices/system/node/events/memory_write/attr/low_addr/shift

Then the same thing is repeated for the other attribute:
/sys/devices/system/node/events/memory_write/attr/high_addr /sys/devices/system/node/events/memory_write/attr/high_addr/min /sys/devices/system/node/events/memory_write/attr/high_addr/max /sys/devices/system/node/events/memory_write/attr/high_addr/shift

Quoted 5 lines

In this scenario, a user tool (like perf) would be able to see that there are attributes associated with the memory_write event, and it knows the names and range of allowed values for these attributes. The shift value tells the tool how much to shift the attribute value before OR'ing it into the attr->config value.

So different events may have different shift value for its attributes. If an event has 4 attributes, each attribute will have a "shift" in sysfs, right?

Quoted 11 lines

If we find that more than 64 bits are needed for the event code plus the attribute values, perhaps shift values greater than 64 would denote to place the attribute bits into a new attr field, like attr->config_extra[shift / 64]

On the perf command line, then, a user could specify something like:

perf stat -e node::memory_write:low_addr=0x1000000:high_addr=0x2000000

What do you think?

I like your idea.

Thanks,
Lin Ming

Quoted 2 lines

- Corey

#7 lore raw
Quoted 10 lines · Lin Ming
Quoted 9 lines · Corey Ashford
Quoted 8 lines · Lin Ming
Quoted 7 lines · Corey Ashford
Quoted 6 lines · Lin Ming
Quoted 5 lines · Corey Ashford
Quoted 3 lines · Lin Ming

Just a temporary patch to show how to use the pmu sysfs interface...

-- snip --

Quoted 63 lines
Quoted 63 lines
Quoted 27 lines
Quoted 9 lines
Quoted 9 lines
Quoted 6 lines

Review

Quoted 4 lines
(quoted diff) +4 -0
+		 * This is just an ugly demo of how to use the sysfs interface.
+		 * You can also parse the<event-name>  and open sys file as,
+		 * sys_fd = open("/sys/devices/system/cpu/events/<event-name>/event_source/id", O_RDONLY);
+		 */

In this above case, does this sys_fd also specify the event I am going to open, in addition to its event source? I'd assume not since event_source is a symlink to /sys/devices/system/cpu/event_source (right?)

Right, this sys_fd only specifies the event source.

Quoted 8 lines
Quoted 5 lines
Quoted 1 line

How do I specify the exact event id via the sysfs interface?

/sys/devices/system/cpu/events/<event-name>/id

But in this patch series, the event id sysfs interface is not used yet.

So, I would open that id and then read the id code and place it in attr->config or maybe place the fd into attr (somewhere) ?

Place the id code in attr->config.

Quoted 3 lines

We also need to take into account event "attributes" - other data that is needed to configure a specific event. For example, think about a memory controller which has a PMU can count events in a particular memory range; we need to be able to supply the memory range somehow, and I don't think that can be accomplished by passing in the fd of a sysfs file that we've opened.

Each event is a directory in the sysfs, so we can put all the event "attributes" under it.

Quoted 7 lines

For your example, /sys/devices/system/node/events/<event-name>/id /sys/devices/system/node/events/<event-name>/memory_range
....

Then we can read these attributes and pass the value into the syscall.

I'm not sure I made the example clear.

Let's say I have a memory controller event called memory_write and it has two attributes: low_addr and high_addr... writes to addresses between the low_addr and high_addr will increment the counter.

As a user, I want to be able to specify a particular memory range, let's say 0x1000000..0x2000000

A sysfs structure like this might be constructed:
/sys/devices/system/node/events/memory_write /sys/devices/system/node/events/memory_write/attr /sys/devices/system/node/events/memory_write/attr/low_addr /sys/devices/system/node/events/memory_write/attr/low_addr/min /sys/devices/system/node/events/memory_write/attr/low_addr/max

In another posting I had also added a bit shift value, but there are so few bits left in the attr->config, that I'm not sure this is a very extensible mechanism, but just for the sake of illustration of the basic idea, I'll add it here:
/sys/devices/system/node/events/memory_write/attr/low_addr/shift

Then the same thing is repeated for the other attribute:
/sys/devices/system/node/events/memory_write/attr/high_addr /sys/devices/system/node/events/memory_write/attr/high_addr/min /sys/devices/system/node/events/memory_write/attr/high_addr/max /sys/devices/system/node/events/memory_write/attr/high_addr/shift

Quoted 7 lines
Quoted 5 lines

In this scenario, a user tool (like perf) would be able to see that there are attributes associated with the memory_write event, and it knows the names and range of allowed values for these attributes. The shift value tells the tool how much to shift the attribute value before OR'ing it into the attr->config value.

So different events may have different shift value for its attributes.

Yes, depending on the hardware, different attributes will be needed with different
[sets of] events. When more than one event shares a set of attributes (for example memory_read and memory_write events would share low_addr and high_addr), the attributes can be placed at the node level and then symlinked from the event.

So for example:
/sys/devices/system/node/attrs/memory/low_addr /sys/devices/system/node/attrs/memory/low_addr/min /sys/devices/system/node/attrs/memory/low_addr/max /sys/devices/system/node/attrs/memory/low_addr/shift

Then in the specific event:
/sys/devices/system/node/events/memory_write/attr/low_addr --> /sys/devices/system/node/attrs/memory/low_addr /sys/devices/system/node/events/memory_read/attr/low_addr --> /sys/devices/system/node/attrs/memory/low_addr

So we could save some space this way.

Quoted 2 lines

If an event has 4 attributes, each attribute will have a "shift" in sysfs, right?

Exactly. It allows the user tool to construct a attr->config (+ config_extra[]) value with many multi-bit fields.

I'm not convinced that this is the best way to go, so any alternative ideas would be welcome.

We are doing something similar with the IBM Wirespeed processor implementation of perf_events, where we are packing the attribute values into the config value, but we are using Stephane Eranian's libpfm4 to handle all the details of doing the packing, instead of exporting the packing info via sysfs.

- Corey

maintainer
#8 lore raw
In reply to Corey Ashford
Quoted 5 lines · Corey Ashford

We also need to take into account event "attributes" - other data that is needed to configure a specific event. For example, think about a memory controller which has a PMU can count events in a particular memory range; we need to be able to supply the memory range somehow,

That's that perf_event_attr is for.

#9 lore raw
Quoted 9 lines · Peter Zijlstra
Quoted 5 lines · Corey Ashford

We also need to take into account event "attributes" - other data that is needed to configure a specific event. For example, think about a memory controller which has a PMU can count events in a particular memory range; we need to be able to supply the memory range somehow,

That's that perf_event_attr is for.

I think the issue is that what is stuffed into perf_event_attr's fields should be done in a way that userspace tools can figure out via attribute info which is in exported by the kernel via sysfs (or eventfs, etc.)

- Corey