From: Jean Pihet <jean.pihet@linaro.org>
To: Borislav Petkov <bp@alien8.de>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@kernel.org>,
Arnaldo Carvalho de Melo <acme@infradead.org>,
Jiri Olsa <jolsa@redhat.com>,
linux-kernel@vger.kernel.org, Robert Richter <rric@kernel.org>
Cc: Robert Richter <robert.richter@linaro.org>,
Namhyung Kim <namhyung@kernel.org>,
Jean Pihet <jean.pihet@linaro.org>
Subject: [PATCH 08/16] perf, persistent: Exposing persistent events using sysfs
Date: Mon, 7 Apr 2014 17:04:30 +0200 [thread overview]
Message-ID: <1396883078-25320-9-git-send-email-jean.pihet@linaro.org> (raw)
In-Reply-To: <1396883078-25320-1-git-send-email-jean.pihet@linaro.org>
From: Robert Richter <robert.richter@linaro.org>
Expose persistent events in the system to userland using sysfs. Perf
tools are able to read existing pmu events from sysfs. Now we use a
persistent pmu as an event container containing all registered
persistent events of the system. This patch adds dynamically
registration of persistent events to sysfs. E.g. something like this:
/sys/bus/event_source/devices/persistent/events/mce_record:persistent,config=106
/sys/bus/event_source/devices/persistent/format/persistent:attr5:23
Perf tools need to support the attr<num> syntax that is added in a
separate patch set. With it we are able to run perf tool commands to
read persistent events, e.g.:
# perf record -e persistent/mce_record/ sleep 10
# perf top -e persistent/mce_record/
[ Jiri: Document attr<index> syntax in sysfs ABI ]
[ Namhyung: Fix sysfs registration with lockdep enabled ]
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Robert Richter <robert.richter@linaro.org>
Signed-off-by: Robert Richter <rric@kernel.org>
Signed-off-by: Jean Pihet <jean.pihet@linaro.org>
---
.../testing/sysfs-bus-event_source-devices-format | 43 ++++++++++++----
kernel/events/persistent.c | 60 ++++++++++++++++++++++
2 files changed, 92 insertions(+), 11 deletions(-)
diff --git a/Documentation/ABI/testing/sysfs-bus-event_source-devices-format b/Documentation/ABI/testing/sysfs-bus-event_source-devices-format
index 77f47ff..744d53a 100644
--- a/Documentation/ABI/testing/sysfs-bus-event_source-devices-format
+++ b/Documentation/ABI/testing/sysfs-bus-event_source-devices-format
@@ -1,13 +1,14 @@
-Where: /sys/bus/event_source/devices/<dev>/format
+Where: /sys/bus/event_source/devices/<pmu>/format/<name>
Date: January 2012
-Kernel Version: 3.3
+Kernel Version: 3.3
+ 3.xx (added attr<index>:<bits>)
Contact: Jiri Olsa <jolsa@redhat.com>
-Description:
- Attribute group to describe the magic bits that go into
- perf_event_attr::config[012] for a particular pmu.
- Each attribute of this group defines the 'hardware' bitmask
- we want to export, so that userspace can deal with sane
- name/value pairs.
+
+Description: Define formats for bit ranges in perf_event_attr
+
+ Attribute group to describe the magic bits that go
+ into struct perf_event_attr for a particular pmu. Bit
+ range may be any bit mask of an u64 (bits 0 to 63).
Userspace must be prepared for the possibility that attributes
define overlapping bit ranges. For example:
@@ -15,6 +16,26 @@ Description:
attr2 = 'config:0-7'
attr3 = 'config:12-35'
- Example: 'config1:1,6-10,44'
- Defines contents of attribute that occupies bits 1,6-10,44 of
- perf_event_attr::config1.
+ Syntax Description
+
+ config[012]*:<bits> Each attribute of this group
+ defines the 'hardware' bitmask
+ we want to export, so that
+ userspace can deal with sane
+ name/value pairs.
+
+ attr<index>:<bits> Set any field of the event
+ attribute. The index is a
+ decimal number that specifies
+ the u64 value to be set within
+ struct perf_event_attr.
+
+ Examples:
+
+ 'config1:1,6-10,44' Defines contents of attribute
+ that occupies bits 1,6-10,44
+ of perf_event_attr::config1.
+
+ 'attr5:24' Define the persistent event
+ flag (bit 24 of the attribute
+ flags)
diff --git a/kernel/events/persistent.c b/kernel/events/persistent.c
index 6b039f5..6578acf 100644
--- a/kernel/events/persistent.c
+++ b/kernel/events/persistent.c
@@ -8,6 +8,7 @@
#define CPU_BUFFER_NR_PAGES ((512 * 1024) / PAGE_SIZE)
struct pevent {
+ struct perf_pmu_events_attr sysfs;
char *name;
int id;
};
@@ -119,6 +120,8 @@ static void persistent_event_close(int cpu, struct pevent *pevent)
persistent_event_release(event);
}
+static int pevent_sysfs_register(struct pevent *event);
+
static int __maybe_unused
persistent_open(char *name, struct perf_event_attr *attr, int nr_pages)
{
@@ -144,12 +147,18 @@ persistent_open(char *name, struct perf_event_attr *attr, int nr_pages)
goto fail;
}
+ pevent->sysfs.id = pevent->id;
+
for_each_possible_cpu(cpu) {
ret = persistent_event_open(cpu, pevent, attr, nr_pages);
if (ret)
goto fail;
}
+ ret = pevent_sysfs_register(pevent);
+ if (ret)
+ goto fail;
+
return 0;
fail:
for_each_possible_cpu(cpu)
@@ -223,10 +232,61 @@ static struct attribute_group persistent_format_group = {
.attrs = persistent_format_attrs,
};
+#define MAX_EVENTS 16
+
+static struct attribute *pevents_attr[MAX_EVENTS + 1] = { };
+
+static struct attribute_group pevents_group = {
+ .name = "events",
+ .attrs = pevents_attr,
+};
+
static const struct attribute_group *persistent_attr_groups[] = {
&persistent_format_group,
+ NULL, /* placeholder: &pevents_group */
NULL,
};
+#define EVENTS_GROUP_PTR (&persistent_attr_groups[1])
+
+static ssize_t pevent_sysfs_show(struct device *dev,
+ struct device_attribute *__attr, char *page)
+{
+ struct perf_pmu_events_attr *attr =
+ container_of(__attr, struct perf_pmu_events_attr, attr);
+ return sprintf(page, "persistent,config=%lld",
+ (unsigned long long)attr->id);
+}
+
+static int pevent_sysfs_register(struct pevent *pevent)
+{
+ struct perf_pmu_events_attr *sysfs = &pevent->sysfs;
+ struct attribute *attr = &sysfs->attr.attr;
+ struct device *dev = persistent_pmu.dev;
+ const struct attribute_group **group = EVENTS_GROUP_PTR;
+ int idx;
+
+ sysfs->id = pevent->id;
+ sysfs->attr = (struct device_attribute)
+ __ATTR(, 0444, pevent_sysfs_show, NULL);
+ attr->name = pevent->name;
+ sysfs_attr_init(attr);
+
+ /* add sysfs attr to events: */
+ for (idx = 0; idx < MAX_EVENTS; idx++) {
+ if (!cmpxchg(pevents_attr + idx, NULL, attr))
+ break;
+ }
+
+ if (idx >= MAX_EVENTS)
+ return -ENOSPC;
+ if (!idx)
+ *group = &pevents_group;
+ if (!dev)
+ return 0; /* sysfs not yet initialized */
+ if (idx)
+ return sysfs_add_file_to_group(&dev->kobj, attr, (*group)->name);
+ return sysfs_create_group(&persistent_pmu.dev->kobj, *group);
+}
static int persistent_pmu_init(struct perf_event *event)
{
--
1.7.11.7
next prev parent reply other threads:[~2014-04-07 15:08 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-07 15:04 [PATCH v4 00/16] perf, persistent: Add persistent events Jean Pihet
2014-04-07 15:04 ` [PATCH 01/16] perf, mmap: Factor out ring_buffer_detach_all() Jean Pihet
2014-04-07 15:04 ` [PATCH 02/16] perf, mmap: Factor out try_get_event()/put_event() Jean Pihet
2014-04-07 15:04 ` [PATCH 03/16] perf, mmap: Factor out perf_alloc/free_rb() Jean Pihet
2014-04-22 14:25 ` Peter Zijlstra
2014-04-07 15:04 ` [PATCH 04/16] perf, mmap: Factor out perf_get_fd() Jean Pihet
2014-04-22 14:27 ` Peter Zijlstra
2014-04-25 13:54 ` Robert Richter
2014-04-25 14:43 ` Peter Zijlstra
2014-04-25 14:52 ` Peter Zijlstra
2014-04-29 15:34 ` Robert Richter
2014-04-07 15:04 ` [PATCH 05/16] perf: Add persistent events Jean Pihet
2014-04-07 15:04 ` [PATCH 06/16] mce, x86: Enable " Jean Pihet
2014-04-07 15:04 ` [PATCH 07/16] perf, persistent: Implementing a persistent pmu Jean Pihet
2014-04-07 15:04 ` Jean Pihet [this message]
2014-04-07 15:04 ` [PATCH 09/16] perf, persistent: Use unique event ids Jean Pihet
2014-04-07 15:04 ` [PATCH 10/16] perf, persistent: Implement reference counter for events Jean Pihet
2014-04-07 15:04 ` [PATCH 11/16] perf, persistent: Dynamically resize list of sysfs entries Jean Pihet
2014-04-07 15:04 ` [PATCH 12/16] perf, persistent: ioctl functions to control persistency Jean Pihet
2014-04-07 15:04 ` [PATCH 13/16] perf tools: Rename flex conditions to avoid name conflicts Jean Pihet
2014-04-07 15:04 ` [PATCH 14/16] perf tools: Modify event parser to update event attribute by index Jean Pihet
2014-04-07 15:04 ` [PATCH 15/16] perf tools: Add attr<num> syntax to event parser Jean Pihet
2014-04-07 15:04 ` [PATCH 16/16] perf tools: Retry mapping buffers readonly on EACCES Jean Pihet
2014-04-17 12:44 ` [PATCH v4 00/16] perf, persistent: Add persistent events Jean Pihet
2014-04-17 12:50 ` Borislav Petkov
2014-04-17 13:17 ` Jean Pihet
2014-04-17 13:21 ` Borislav Petkov
2014-04-22 8:20 ` Jean Pihet
2014-04-22 10:07 ` Jean Pihet
2014-04-22 10:20 ` Borislav Petkov
2014-05-06 12:39 ` Robert Richter
2014-05-06 18:50 ` Borislav Petkov
2014-05-06 18:53 ` Borislav Petkov
2014-05-07 16:44 ` Robert Richter
2014-05-08 18:23 ` Borislav Petkov
2014-05-09 9:17 ` Robert Richter
2014-05-06 18:58 ` Borislav Petkov
2014-05-07 17:01 ` Robert Richter
2014-05-08 18:36 ` Borislav Petkov
2014-05-09 8:52 ` Robert Richter
2014-05-09 10:17 ` Borislav Petkov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1396883078-25320-9-git-send-email-jean.pihet@linaro.org \
--to=jean.pihet@linaro.org \
--cc=acme@infradead.org \
--cc=bp@alien8.de \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=peterz@infradead.org \
--cc=robert.richter@linaro.org \
--cc=rric@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®