Systemd/src/core/execute.h
Lennart Poettering efdb02375b core: unified cgroup hierarchy support
This patch set adds full support the new unified cgroup hierarchy logic
of modern kernels.

A new kernel command line option "systemd.unified_cgroup_hierarchy=1" is
added. If specified the unified hierarchy is mounted to /sys/fs/cgroup
instead of a tmpfs. No further hierarchies are mounted. The kernel
command line option defaults to off. We can turn it on by default as
soon as the kernel's APIs regarding this are stabilized (but even then
downstream distros might want to turn this off, as this will break any
tools that access cgroupfs directly).

It is possibly to choose for each boot individually whether the unified
or the legacy hierarchy is used. nspawn will by default provide the
legacy hierarchy to containers if the host is using it, and the unified
otherwise. However it is possible to run containers with the unified
hierarchy on a legacy host and vice versa, by setting the
$UNIFIED_CGROUP_HIERARCHY environment variable for nspawn to 1 or 0,
respectively.

The unified hierarchy provides reliable cgroup empty notifications for
the first time, via inotify. To make use of this we maintain one
manager-wide inotify fd, and each cgroup to it.

This patch also removes cg_delete() which is unused now.

On kernel 4.2 only the "memory" controller is compatible with the
unified hierarchy, hence that's the only controller systemd exposes when
booted in unified heirarchy mode.

This introduces a new enum for enumerating supported controllers, plus a
related enum for the mask bits mapping to it. The core is changed to
make use of this everywhere.

This moves PID 1 into a new "init.scope" implicit scope unit in the root
slice. This is necessary since on the unified hierarchy cgroups may
either contain subgroups or processes but not both. PID 1 hence has to
move out of the root cgroup (strictly speaking the root cgroup is the
only one where processes and subgroups are still allowed, but in order
to support containers nicey, we move PID 1 into the new scope in all
cases.) This new unit is also used on legacy hierarchy setups. It's
actually pretty useful on all systems, as it can then be used to filter
journal messages coming from PID 1, and so on.

The root slice ("-.slice") is now implicitly created and started (and
does not require a unit file on disk anymore), since
that's where "init.scope" is located and the slice needs to be started
before the scope can.

To check whether we are in unified or legacy hierarchy mode we use
statfs() on /sys/fs/cgroup. If the .f_type field reports tmpfs we are in
legacy mode, if it reports cgroupfs we are in unified mode.

This patch set carefuly makes sure that cgls and cgtop continue to work
as desired.

When invoking nspawn as a service it will implicitly create two
subcgroups in the cgroup it is using, one to move the nspawn process
into, the other to move the actual container processes into. This is
done because of the requirement that cgroups may either contain
processes or other subgroups.
2015-09-01 23:52:27 +02:00

280 lines
7.9 KiB
C

/*-*- Mode: C; c-basic-offset: 8; indent-tabs-mode: nil -*-*/
#pragma once
/***
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 Lesser General Public License as published by
the Free Software Foundation; either version 2.1 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
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public License
along with systemd; If not, see <http://www.gnu.org/licenses/>.
***/
typedef struct ExecStatus ExecStatus;
typedef struct ExecCommand ExecCommand;
typedef struct ExecContext ExecContext;
typedef struct ExecRuntime ExecRuntime;
typedef struct ExecParameters ExecParameters;
#include <sys/capability.h>
#include <stdbool.h>
#include <stdio.h>
#include <sched.h>
#include "list.h"
#include "fdset.h"
#include "missing.h"
#include "namespace.h"
#include "bus-endpoint.h"
typedef enum ExecUtmpMode {
EXEC_UTMP_INIT,
EXEC_UTMP_LOGIN,
EXEC_UTMP_USER,
_EXEC_UTMP_MODE_MAX,
_EXEC_UTMP_MODE_INVALID = -1
} ExecUtmpMode;
typedef enum ExecInput {
EXEC_INPUT_NULL,
EXEC_INPUT_TTY,
EXEC_INPUT_TTY_FORCE,
EXEC_INPUT_TTY_FAIL,
EXEC_INPUT_SOCKET,
_EXEC_INPUT_MAX,
_EXEC_INPUT_INVALID = -1
} ExecInput;
typedef enum ExecOutput {
EXEC_OUTPUT_INHERIT,
EXEC_OUTPUT_NULL,
EXEC_OUTPUT_TTY,
EXEC_OUTPUT_SYSLOG,
EXEC_OUTPUT_SYSLOG_AND_CONSOLE,
EXEC_OUTPUT_KMSG,
EXEC_OUTPUT_KMSG_AND_CONSOLE,
EXEC_OUTPUT_JOURNAL,
EXEC_OUTPUT_JOURNAL_AND_CONSOLE,
EXEC_OUTPUT_SOCKET,
_EXEC_OUTPUT_MAX,
_EXEC_OUTPUT_INVALID = -1
} ExecOutput;
struct ExecStatus {
dual_timestamp start_timestamp;
dual_timestamp exit_timestamp;
pid_t pid;
int code; /* as in siginfo_t::si_code */
int status; /* as in sigingo_t::si_status */
};
struct ExecCommand {
char *path;
char **argv;
ExecStatus exec_status;
LIST_FIELDS(ExecCommand, command); /* useful for chaining commands */
bool ignore;
};
struct ExecRuntime {
int n_ref;
char *tmp_dir;
char *var_tmp_dir;
int netns_storage_socket[2];
};
struct ExecContext {
char **environment;
char **environment_files;
struct rlimit *rlimit[_RLIMIT_MAX];
char *working_directory, *root_directory;
bool working_directory_missing_ok;
mode_t umask;
int oom_score_adjust;
int nice;
int ioprio;
int cpu_sched_policy;
int cpu_sched_priority;
cpu_set_t *cpuset;
unsigned cpuset_ncpus;
ExecInput std_input;
ExecOutput std_output;
ExecOutput std_error;
nsec_t timer_slack_nsec;
char *tty_path;
bool tty_reset;
bool tty_vhangup;
bool tty_vt_disallocate;
bool ignore_sigpipe;
/* Since resolving these names might might involve socket
* connections and we don't want to deadlock ourselves these
* names are resolved on execution only and in the child
* process. */
char *user;
char *group;
char **supplementary_groups;
char *pam_name;
char *utmp_id;
ExecUtmpMode utmp_mode;
bool selinux_context_ignore;
char *selinux_context;
bool apparmor_profile_ignore;
char *apparmor_profile;
bool smack_process_label_ignore;
char *smack_process_label;
char **read_write_dirs, **read_only_dirs, **inaccessible_dirs;
unsigned long mount_flags;
uint64_t capability_bounding_set_drop;
cap_t capabilities;
int secure_bits;
int syslog_priority;
char *syslog_identifier;
bool syslog_level_prefix;
bool cpu_sched_reset_on_fork;
bool non_blocking;
bool private_tmp;
bool private_network;
bool private_devices;
ProtectSystem protect_system;
ProtectHome protect_home;
bool no_new_privileges;
/* This is not exposed to the user but available
* internally. We need it to make sure that whenever we spawn
* /usr/bin/mount it is run in the same process group as us so
* that the autofs logic detects that it belongs to us and we
* don't enter a trigger loop. */
bool same_pgrp;
unsigned long personality;
Set *syscall_filter;
Set *syscall_archs;
int syscall_errno;
bool syscall_whitelist:1;
Set *address_families;
bool address_families_whitelist:1;
char **runtime_directory;
mode_t runtime_directory_mode;
bool oom_score_adjust_set:1;
bool nice_set:1;
bool ioprio_set:1;
bool cpu_sched_set:1;
bool no_new_privileges_set:1;
/* custom dbus enpoint */
BusEndpoint *bus_endpoint;
};
#include "cgroup.h"
#include "cgroup-util.h"
struct ExecParameters {
char **argv;
int *fds; unsigned n_fds;
char **environment;
bool apply_permissions;
bool apply_chroot;
bool apply_tty_stdin;
bool confirm_spawn;
bool selinux_context_net;
CGroupMask cgroup_supported;
const char *cgroup_path;
bool cgroup_delegate;
const char *runtime_prefix;
usec_t watchdog_usec;
int *idle_pipe;
char *bus_endpoint_path;
int bus_endpoint_fd;
};
int exec_spawn(Unit *unit,
ExecCommand *command,
const ExecContext *context,
const ExecParameters *exec_params,
ExecRuntime *runtime,
pid_t *ret);
void exec_command_done(ExecCommand *c);
void exec_command_done_array(ExecCommand *c, unsigned n);
ExecCommand* exec_command_free_list(ExecCommand *c);
void exec_command_free_array(ExecCommand **c, unsigned n);
char *exec_command_line(char **argv);
void exec_command_dump(ExecCommand *c, FILE *f, const char *prefix);
void exec_command_dump_list(ExecCommand *c, FILE *f, const char *prefix);
void exec_command_append_list(ExecCommand **l, ExecCommand *e);
int exec_command_set(ExecCommand *c, const char *path, ...);
int exec_command_append(ExecCommand *c, const char *path, ...);
void exec_context_init(ExecContext *c);
void exec_context_done(ExecContext *c);
void exec_context_dump(ExecContext *c, FILE* f, const char *prefix);
int exec_context_destroy_runtime_directory(ExecContext *c, const char *runtime_root);
int exec_context_load_environment(Unit *unit, const ExecContext *c, char ***l);
bool exec_context_may_touch_console(ExecContext *c);
bool exec_context_maintains_privileges(ExecContext *c);
void exec_status_start(ExecStatus *s, pid_t pid);
void exec_status_exit(ExecStatus *s, ExecContext *context, pid_t pid, int code, int status);
void exec_status_dump(ExecStatus *s, FILE *f, const char *prefix);
int exec_runtime_make(ExecRuntime **rt, ExecContext *c, const char *id);
ExecRuntime *exec_runtime_ref(ExecRuntime *r);
ExecRuntime *exec_runtime_unref(ExecRuntime *r);
int exec_runtime_serialize(Unit *unit, ExecRuntime *rt, FILE *f, FDSet *fds);
int exec_runtime_deserialize_item(Unit *unit, ExecRuntime **rt, const char *key, const char *value, FDSet *fds);
void exec_runtime_destroy(ExecRuntime *rt);
const char* exec_output_to_string(ExecOutput i) _const_;
ExecOutput exec_output_from_string(const char *s) _pure_;
const char* exec_input_to_string(ExecInput i) _const_;
ExecInput exec_input_from_string(const char *s) _pure_;
const char* exec_utmp_mode_to_string(ExecUtmpMode i) _const_;
ExecUtmpMode exec_utmp_mode_from_string(const char *s) _pure_;