From: Tejun Heo <tj@kernel.org>
To: lizefan@huawei.com, hannes@cmpxchg.org, a.p.zijlstra@chello.nl,
mingo@redhat.com, acme@kernel.org
Cc: linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
kernel-team@fb.com, Tejun Heo <tj@kernel.org>
Subject: [PATCH 6/6] cgroup, perf_event: make perf_event controller work on cgroup2 hierarchy
Date: Thu, 7 Jan 2016 17:29:50 -0500 [thread overview]
Message-ID: <1452205790-21331-7-git-send-email-tj@kernel.org> (raw)
In-Reply-To: <1452205790-21331-1-git-send-email-tj@kernel.org>
perf_event is a utility controller whose primary role is identifying
cgroup membership to filter perf events; however, because it also
tracks some per-css state, it can't be replaced by pure cgroup
membership test. Mark the controller as implicitly enabled on the
default hierarchy so that perf events can always be filtered based on
cgroup v2 path as long as the controller is not mounted on a legacy
hierarchy.
"perf record" is updated accordingly so that it searches for both v1
and v2 hierarchies. A v1 hierarchy is used if perf_event is mounted
on it; otherwise, it uses the v2 hierarchy.
Signed-off-by: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
---
Documentation/cgroup.txt | 13 +++++++++++++
kernel/events/core.c | 6 ++++++
tools/perf/util/cgroup.c | 26 +++++++++++++++++++-------
3 files changed, 38 insertions(+), 7 deletions(-)
diff --git a/Documentation/cgroup.txt b/Documentation/cgroup.txt
index 31d1f7b..721e43e 100644
--- a/Documentation/cgroup.txt
+++ b/Documentation/cgroup.txt
@@ -47,6 +47,8 @@ CONTENTS
5-3. IO
5-3-1. IO Interface Files
5-3-2. Writeback
+ 5-4. Misc
+ 5-4-1. perf_event
P. Information on Kernel Programming
P-1. Filesystem Support for Writeback
D. Deprecated v1 Core Features
@@ -1013,6 +1015,17 @@ writeback as follows.
vm.dirty[_background]_ratio.
+5-4. Misc
+
+5-4-1. perf_event
+
+perf_event controller, if not mounted on a legacy hierarchy on the
+first mount of cgroup v2, is automatically enabled on the v2 hierarchy
+so that perf events can always be filtered by cgroup v2 path. The
+controller can still be moved to a legacy hierarchy after v2 hierarchy
+is populated.
+
+
P. Information on Kernel Programming
This section contains kernel programming information in the areas
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 026305d..02ad5b3 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -9469,5 +9469,11 @@ struct cgroup_subsys perf_event_cgrp_subsys = {
.css_alloc = perf_cgroup_css_alloc,
.css_free = perf_cgroup_css_free,
.attach = perf_cgroup_attach,
+ /*
+ * Implicitly enable on dfl hierarchy so that perf events can
+ * always be filtered by cgroup2 path as long as perf_event
+ * controller is not mounted on a legacy hierarchy.
+ */
+ .implicit_on_dfl = true,
};
#endif /* CONFIG_CGROUP_PERF */
diff --git a/tools/perf/util/cgroup.c b/tools/perf/util/cgroup.c
index 32e12ec..54b5bf4 100644
--- a/tools/perf/util/cgroup.c
+++ b/tools/perf/util/cgroup.c
@@ -12,8 +12,8 @@ cgroupfs_find_mountpoint(char *buf, size_t maxlen)
{
FILE *fp;
char mountpoint[PATH_MAX + 1], tokens[PATH_MAX + 1], type[PATH_MAX + 1];
+ char path_v1[PATH_MAX + 1], path_v2[PATH_MAX + 2], *path;
char *token, *saved_ptr = NULL;
- int found = 0;
fp = fopen("/proc/mounts", "r");
if (!fp)
@@ -24,31 +24,43 @@ cgroupfs_find_mountpoint(char *buf, size_t maxlen)
* and inspect every cgroupfs mount point to find one that has
* perf_event subsystem
*/
+ path_v1[0] = '\0';
+ path_v2[0] = '\0';
+
while (fscanf(fp, "%*s %"STR(PATH_MAX)"s %"STR(PATH_MAX)"s %"
STR(PATH_MAX)"s %*d %*d\n",
mountpoint, type, tokens) == 3) {
- if (!strcmp(type, "cgroup")) {
+ if (!path_v1[0] && !strcmp(type, "cgroup")) {
token = strtok_r(tokens, ",", &saved_ptr);
while (token != NULL) {
if (!strcmp(token, "perf_event")) {
- found = 1;
+ strcpy(path_v1, mountpoint);
break;
}
token = strtok_r(NULL, ",", &saved_ptr);
}
}
- if (found)
+
+ if (!path_v2[0] && !strcmp(type, "cgroup2"))
+ strcpy(path_v2, mountpoint);
+
+ if (path_v1[0] && path_v2[0])
break;
}
fclose(fp);
- if (!found)
+
+ if (path_v1[0])
+ path = path_v1;
+ else if (path_v2[0])
+ path = path_v2;
+ else
return -1;
- if (strlen(mountpoint) < maxlen) {
- strcpy(buf, mountpoint);
+ if (strlen(path) < maxlen) {
+ strcpy(buf, path);
return 0;
}
return -1;
--
2.5.0
prev parent reply other threads:[~2016-01-07 22:30 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-07 22:29 [PATCHSET] cgroup, perf_event: make perf_event work on v2 hierarchy Tejun Heo
2016-01-07 22:29 ` [PATCH 1/6] cgroup: s/cgrp_dfl_root_/cgrp_dfl_/ Tejun Heo
2016-01-07 22:29 ` [PATCH 2/6] cgroup: convert cgroup_subsys flag fields to bool bitfields Tejun Heo
2016-01-07 22:29 ` [PATCH 3/6] cgroup: make css_tryget_online_from_dir() also recognize cgroup2 fs Tejun Heo
2016-01-07 22:29 ` [PATCH 4/6] cgroup: use ->subtree_control when testing no internal process rule Tejun Heo
2016-02-23 14:59 ` Tejun Heo
2016-01-07 22:29 ` [PATCH 5/6] cgroup: implement cgroup_subsys->implicit_on_dfl Tejun Heo
2016-01-07 22:29 ` Tejun Heo [this message]
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=1452205790-21331-7-git-send-email-tj@kernel.org \
--to=tj@kernel.org \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@kernel.org \
--cc=cgroups@vger.kernel.org \
--cc=hannes@cmpxchg.org \
--cc=kernel-team@fb.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lizefan@huawei.com \
--cc=mingo@redhat.com \
/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
Powered by JetHome