chiark / gitweb /
cgroup: add cgroupsification
[elogind.git] / cgroup.c
diff --git a/cgroup.c b/cgroup.c
new file mode 100644 (file)
index 0000000..04736ac
--- /dev/null
+++ b/cgroup.c
@@ -0,0 +1,523 @@
+/*-*- Mode: C; c-basic-offset: 8 -*-*/
+
+/***
+  This file is part of systemd.
+
+  Copyright 2010 Lennart Poettering
+
+  systemd is free software; you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as published by
+  the Free Software Foundation; either version 2 of the License, or
+  (at your option) any later version.
+
+  systemd is distributed in the hope that it will be useful, but
+  WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+  General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with systemd; If not, see <http://www.gnu.org/licenses/>.
+***/
+
+#include <errno.h>
+#include <assert.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <signal.h>
+#include <sys/mount.h>
+
+#include "cgroup.h"
+#include "log.h"
+
+static int translate_error(int error, int _errno) {
+
+        switch (error) {
+
+        case ECGROUPNOTCOMPILED:
+        case ECGROUPNOTMOUNTED:
+        case ECGROUPNOTEXIST:
+        case ECGROUPNOTCREATED:
+                return -ENOENT;
+
+        case ECGINVAL:
+                return -EINVAL;
+
+        case ECGROUPNOTALLOWED:
+                return -EPERM;
+
+        case ECGOTHER:
+                return -_errno;
+        }
+
+        return -EIO;
+}
+
+int cgroup_bonding_realize(CGroupBonding *b) {
+        int r;
+
+        assert(b);
+        assert(b->path);
+        assert(b->controller);
+
+        if (b->cgroup)
+                return 0;
+
+        if (!(b->cgroup = cgroup_new_cgroup(b->path)))
+                return -ENOMEM;
+
+        if (!cgroup_add_controller(b->cgroup, b->controller)) {
+                r = -ENOMEM;
+                goto fail;
+        }
+
+        if (b->inherit)
+                r = cgroup_create_cgroup_from_parent(b->cgroup, true);
+        else
+                r = cgroup_create_cgroup(b->cgroup, true);
+
+        if (r != 0) {
+                r = translate_error(r, errno);
+                goto fail;
+        }
+
+        return 0;
+
+fail:
+        cgroup_free(&b->cgroup);
+        b->cgroup = NULL;
+        return r;
+}
+
+int cgroup_bonding_realize_list(CGroupBonding *first) {
+        CGroupBonding *b;
+
+        LIST_FOREACH(by_unit, b, first) {
+                int r;
+
+                if ((r = cgroup_bonding_realize(b)) < 0)
+                        return r;
+        }
+
+        return 0;
+}
+
+void cgroup_bonding_free(CGroupBonding *b) {
+        assert(b);
+
+        if (b->unit) {
+                CGroupBonding *f;
+
+                LIST_REMOVE(CGroupBonding, by_unit, b->unit->meta.cgroup_bondings, b);
+
+                assert_se(f = hashmap_get(b->unit->meta.manager->cgroup_bondings, b->path));
+                LIST_REMOVE(CGroupBonding, by_path, f, b);
+
+                if (f)
+                        hashmap_replace(b->unit->meta.manager->cgroup_bondings, b->path, f);
+                else
+                        hashmap_remove(b->unit->meta.manager->cgroup_bondings, b->path);
+        }
+
+        free(b->controller);
+        free(b->path);
+
+        if (b->cgroup) {
+
+                if (b->only_us && b->clean_up)
+                        cgroup_delete_cgroup(b->cgroup, true);
+
+                cgroup_free(&b->cgroup);
+        }
+
+        free(b);
+}
+
+void cgroup_bonding_free_list(CGroupBonding *first) {
+        CGroupBonding *b, *n;
+
+        LIST_FOREACH_SAFE(by_unit, b, n, first)
+                cgroup_bonding_free(b);
+}
+
+int cgroup_bonding_install(CGroupBonding *b, pid_t pid) {
+        int r;
+
+        assert(b);
+        assert(pid >= 0);
+
+        if (pid == 0)
+                pid = getpid();
+
+        if (!b->cgroup)
+                return -ENOENT;
+
+        if ((r = cgroup_attach_task_pid(b->cgroup, pid)))
+                return translate_error(r, errno);
+
+        return 0;
+}
+
+int cgroup_bonding_install_list(CGroupBonding *first, pid_t pid) {
+        CGroupBonding *b;
+
+        LIST_FOREACH(by_unit, b, first) {
+                int r;
+
+                if ((r = cgroup_bonding_install(b, pid)) < 0)
+                        return r;
+        }
+
+        return 0;
+}
+
+int cgroup_bonding_kill(CGroupBonding *b, int sig) {
+        int r;
+        Set *s;
+        bool done;
+
+        assert(b);
+        assert(sig > 0);
+
+        if (!(s = set_new(trivial_hash_func, trivial_compare_func)))
+                return -ENOMEM;
+
+        do {
+                void *iterator;
+                pid_t pid;
+
+                done = true;
+
+                if ((r = cgroup_get_task_begin(b->path, b->controller, &iterator, &pid)) != 0) {
+                        if (r == ECGEOF) {
+                                r = 0;
+                                goto kill_done;
+                        } else {
+                                r = translate_error(r, errno);
+                                break;
+                        }
+                }
+
+                for (;;) {
+                        if (set_get(s, INT_TO_PTR(pid)) != INT_TO_PTR(pid)) {
+
+                                /* If we haven't killed this process
+                                 * yet, kill it */
+
+                                if (kill(pid, sig) < 0 && errno != ESRCH) {
+                                        r = -errno;
+                                        break;
+                                }
+
+                                done = false;
+
+                                if ((r = set_put(s, INT_TO_PTR(pid))) < 0)
+                                    break;
+                        }
+
+                        if ((r = cgroup_get_task_next(&iterator, &pid)) != 0) {
+
+                                if (r == ECGEOF)
+                                        r = 0;
+                                else
+                                        r = translate_error(r, errno);
+
+                                break;
+                        }
+                }
+
+        kill_done:
+                assert_se(cgroup_get_task_end(&iterator) == 0);
+
+                /* To avoid racing against processes which fork
+                 * quicker than we can kill them we repeat this until
+                 * no new pids need to be killed. */
+
+        } while (!done && r >= 0);
+
+        set_free(s);
+        return r;
+}
+
+int cgroup_bonding_kill_list(CGroupBonding *first, int sig) {
+        CGroupBonding *b;
+
+        LIST_FOREACH(by_unit, b, first) {
+                int r;
+
+                if ((r = cgroup_bonding_kill(b, sig)) < 0)
+                        return r;
+        }
+
+        return 0;
+}
+
+/* Returns 1 if the group is empty, 0 if it is not, -EAGAIN if we
+ * cannot know */
+int cgroup_bonding_is_empty(CGroupBonding *b) {
+        void *iterator;
+        pid_t pid;
+        int r;
+
+        assert(b);
+
+        r = cgroup_get_task_begin(b->path, b->controller, &iterator, &pid);
+
+        if (r == 0 || r == ECGEOF)
+                cgroup_get_task_end(&iterator);
+
+        /* Hmm, no PID in this group? Then it is definitely empty */
+        if (r == ECGEOF)
+                return 1;
+
+        /* Some error? Let's return it */
+        if (r != 0)
+                return translate_error(r, errno);
+
+        /* It's not empty, and we are the only user, then it is
+         * definitely not empty */
+        if (b->only_us)
+                return 0;
+
+        /* There are PIDs in the group but we aren't the only users,
+         * hence we cannot say */
+        return -EAGAIN;
+}
+
+int cgroup_bonding_is_empty_list(CGroupBonding *first) {
+        CGroupBonding *b;
+
+        LIST_FOREACH(by_unit, b, first) {
+                int r;
+
+                if ((r = cgroup_bonding_is_empty(b)) < 0) {
+                        /* If this returned -EAGAIN, then we don't know if the
+                         * group is empty, so let's see if another group can
+                         * tell us */
+
+                        if (r != -EAGAIN)
+                                return r;
+                } else
+                        return r;
+        }
+
+        return -EAGAIN;
+}
+
+static int install_release_agent(Manager *m, const char *mount_point) {
+        char *p, *c, *sc;
+        int r;
+
+        assert(m);
+        assert(mount_point);
+
+        if (asprintf(&p, "%s/release_agent", mount_point) < 0)
+                return -ENOMEM;
+
+        if ((r = read_one_line_file(p, &c)) < 0) {
+                free(p);
+                return r;
+        }
+
+        sc = strstrip(c);
+
+        if (sc[0] == 0) {
+                if ((r = write_one_line_file(p, CGROUP_AGENT_PATH "\n" )) < 0) {
+                        free(p);
+                        free(c);
+                        return r;
+                }
+        } else if (!streq(sc, CGROUP_AGENT_PATH)) {
+                free(p);
+                free(c);
+                return -EEXIST;
+        }
+
+        free(c);
+        free(p);
+
+        if (asprintf(&p, "%s/notify_on_release", mount_point) < 0)
+                return -ENOMEM;
+
+        if ((r = read_one_line_file(p, &c)) < 0) {
+                free(p);
+                return r;
+        }
+
+        sc = strstrip(c);
+
+        if (streq(sc, "0")) {
+                if ((r = write_one_line_file(p, "1\n")) < 0) {
+                        free(p);
+                        free(c);
+                        return r;
+                }
+        } else if (!streq(sc, "1")) {
+                free(p);
+                free(c);
+                return -EIO;
+        }
+
+        return 0;
+}
+
+static int create_hierarchy_cgroup(Manager *m) {
+        struct cgroup *cg;
+        int r;
+
+        assert(m);
+
+        if (!(cg = cgroup_new_cgroup(m->cgroup_hierarchy)))
+                return -ENOMEM;
+
+        if (!(cgroup_add_controller(cg, m->cgroup_controller))) {
+                r = -ENOMEM;
+                goto finish;
+        }
+
+        if ((r = cgroup_create_cgroup(cg, true)) != 0) {
+                log_error("Failed to create cgroup hierarchy group: %s", cgroup_strerror(r));
+                r = translate_error(r, errno);
+                goto finish;
+        }
+
+        if ((r = cgroup_attach_task(cg)) != 0) {
+                log_error("Failed to add ourselves to hierarchy group: %s", cgroup_strerror(r));
+                r = translate_error(r, errno);
+                goto finish;
+        }
+
+        r = 0;
+
+finish:
+        cgroup_free(&cg);
+        return r;
+}
+
+int manager_setup_cgroup(Manager *m) {
+        char *mp, *cp;
+        int r;
+        pid_t pid;
+
+        assert(m);
+
+        if ((r = cgroup_init()) != 0) {
+                log_error("Failed to initialize libcg: %s", cgroup_strerror(r));
+                return translate_error(r, errno);
+        }
+
+        free(m->cgroup_controller);
+        if (!(m->cgroup_controller = strdup("debug")))
+                return -ENOMEM;
+
+        if ((r = cgroup_get_subsys_mount_point(m->cgroup_controller, &mp)))
+                return translate_error(r, errno);
+
+        pid = getpid();
+
+        if ((r = cgroup_get_current_controller_path(pid, m->cgroup_controller, &cp))) {
+                free(mp);
+                return translate_error(r, errno);
+        }
+
+        free(m->cgroup_hierarchy);
+        m->cgroup_hierarchy = NULL;
+        if (asprintf(&m->cgroup_hierarchy, "%s/systemd-%llu", strcmp(cp, "/") == 0 ? "" : cp, (unsigned long long) pid) < 0) {
+                free(cp);
+                free(mp);
+                return -ENOMEM;
+        }
+
+        log_info("Using cgroup controller <%s>, hierarchy mounted at <%s>, using root group <%s>.",
+                 m->cgroup_controller,
+                 mp,
+                 m->cgroup_hierarchy);
+
+        if ((r = install_release_agent(m, mp)) < 0)
+                log_warning("Failed to install release agent, ignoring: %s", strerror(-r));
+        else
+                log_info("Installed release agent, or already installed.");
+
+        free(mp);
+        free(cp);
+
+        if ((r = create_hierarchy_cgroup(m)) < 0)
+                log_error("Failed to create root cgroup hierarchy: %s", strerror(-r));
+        else
+                log_info("Created root group.");
+
+        return r;
+}
+
+int manager_shutdown_cgroup(Manager *m) {
+        struct cgroup *cg;
+        int r;
+
+        assert(m);
+
+        if (!m->cgroup_hierarchy)
+                return 0;
+
+        if (!(cg = cgroup_new_cgroup(m->cgroup_hierarchy)))
+                return -ENOMEM;
+
+        if (!(cgroup_add_controller(cg, m->cgroup_controller))) {
+                r = -ENOMEM;
+                goto finish;
+        }
+
+        if ((r = cgroup_delete_cgroup_ext(cg, CGFLAG_DELETE_IGNORE_MIGRATION|CGFLAG_DELETE_RECURSIVE)) != 0) {
+                log_error("Failed to delete cgroup hierarchy group: %s", cgroup_strerror(r));
+                r = translate_error(r, errno);
+                goto finish;
+        }
+        r = 0;
+
+finish:
+        cgroup_free(&cg);
+        return r;
+
+}
+
+int cgroup_notify_empty(Manager *m, const char *group) {
+        CGroupBonding *l, *b;
+
+        assert(m);
+        assert(group);
+
+        if (!(l = hashmap_get(m->cgroup_bondings, group)))
+                return 0;
+
+        LIST_FOREACH(by_path, b, l) {
+                int t;
+
+                if (!b->unit)
+                        continue;
+
+                if ((t = cgroup_bonding_is_empty_list(b)) < 0) {
+
+                        /* If we don't know, we don't know */
+                        if (t != -EAGAIN)
+                                log_warning("Failed to check whether cgroup is empty: %s", strerror(errno));
+
+                        continue;
+                }
+
+                if (t > 0)
+                        if (UNIT_VTABLE(b->unit)->cgroup_notify_empty)
+                                UNIT_VTABLE(b->unit)->cgroup_notify_empty(b->unit);
+        }
+
+        return 0;
+}
+
+CGroupBonding *cgroup_bonding_find_list(CGroupBonding *first, const char *controller) {
+        CGroupBonding *b;
+
+        assert(controller);
+
+        LIST_FOREACH(by_unit, b, first)
+                if (streq(b->controller, controller))
+                        return b;
+
+        return NULL;
+}