From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tejun Heo Subject: [PATCH 6/6] cgroup, perf_event: make perf_event controller work on cgroup2 hierarchy Date: Thu, 7 Jan 2016 17:29:50 -0500 Message-ID: <1452205790-21331-7-git-send-email-tj@kernel.org> References: <1452205790-21331-1-git-send-email-tj@kernel.org> Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=sender:from:to:cc:subject:date:message-id:in-reply-to:references; bh=NflOlPrruRx+OZ0HtisXP6ejMw5QtjEY/YTcv6ZyNh0=; b=jW8Se6bj2vctC+v091yt4vyc1B3XXOkehkwC8fidUIRpSgSb6LbxLKjCbRR8allwxr aytL889F+CslB5Ywt87xI0CmDpKtjHtemTkZ1h6nSajxmQEX5vCJn0jBvEw7+CXfBvIm iBHv3SLaHIiVTZpK2Q449yWIcpDYWpyQAzTgXi9ijiGAzJqOqS4y+pAs7UabNcu9M7ER OSbmffSousR/ziFzDNkW7VqS5UClqBHOb05RimjIuBOG3L5a6Tu+T6BhsMBpElg4tBPv mhi2NQv5TNkeA9UMTHAm0Q1Um3isT/fgPSiEMZNw6hHY+jghhJEjZrKIJ6fsd0ozDatf Vr4Q== In-Reply-To: <1452205790-21331-1-git-send-email-tj@kernel.org> Sender: linux-kernel-owner@vger.kernel.org List-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit 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 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 Cc: Peter Zijlstra Cc: Ingo Molnar Cc: Arnaldo Carvalho de Melo --- 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