2017-11-18 17:09:20 +01:00
|
|
|
/* SPDX-License-Identifier: LGPL-2.1+ */
|
2013-11-11 10:03:31 +01:00
|
|
|
|
|
|
|
|
#include <stdio.h>
|
|
|
|
|
|
|
|
|
|
#include "macro.h"
|
2015-11-16 22:09:36 +01:00
|
|
|
#include "manager.h"
|
2016-04-08 18:54:05 +02:00
|
|
|
#include "rm-rf.h"
|
2017-11-08 19:01:18 +01:00
|
|
|
#include "string-util.h"
|
2013-11-11 10:03:31 +01:00
|
|
|
#include "test-helper.h"
|
2016-04-08 18:54:05 +02:00
|
|
|
#include "tests.h"
|
2015-11-16 22:09:36 +01:00
|
|
|
#include "unit.h"
|
2013-11-11 10:03:31 +01:00
|
|
|
|
|
|
|
|
static int test_cgroup_mask(void) {
|
2017-02-14 22:33:52 +01:00
|
|
|
_cleanup_(rm_rf_physical_and_freep) char *runtime_dir = NULL;
|
2018-03-09 21:55:55 +01:00
|
|
|
_cleanup_(manager_freep) Manager *m = NULL;
|
2014-02-14 19:11:07 +01:00
|
|
|
Unit *son, *daughter, *parent, *root, *grandchild, *parent_deep;
|
2013-11-11 10:03:31 +01:00
|
|
|
int r;
|
|
|
|
|
|
2017-10-10 20:55:20 +02:00
|
|
|
r = enter_cgroup_subroot();
|
2018-09-13 13:34:12 +02:00
|
|
|
if (r == -ENOMEDIUM)
|
|
|
|
|
return log_tests_skipped("cgroupfs not available");
|
tests: when running a manager object in a test, migrate to private cgroup subroot first (#6576)
Without this "meson test" will end up running all tests in the same
cgroup root, and they all will try to manage it. Which usually isn't too
bad, except when they end up clearing up each other's cgroups. This race
is hard to trigger but has caused various CI runs to fail spuriously.
With this change we simply move every test that runs a manager object
into their own private cgroup. Note that we don't clean up the cgroup at
the end, we leave that to the cgroup manager around it.
This fixes races that become visible by test runs throwing out errors
like this:
```
exec-systemcallfilter-failing.service: Passing 0 fds to service
exec-systemcallfilter-failing.service: About to execute: /bin/echo 'This should not be seen'
exec-systemcallfilter-failing.service: Forked /bin/echo as 5693
exec-systemcallfilter-failing.service: Changed dead -> start
exec-systemcallfilter-failing.service: Failed to attach to cgroup /exec-systemcallfilter-failing.service: No such file or directory
Received SIGCHLD from PID 5693 ((echo)).
Child 5693 ((echo)) died (code=exited, status=219/CGROUP)
exec-systemcallfilter-failing.service: Child 5693 belongs to exec-systemcallfilter-failing.service
exec-systemcallfilter-failing.service: Main process exited, code=exited, status=219/CGROUP
exec-systemcallfilter-failing.service: Changed start -> failed
exec-systemcallfilter-failing.service: Unit entered failed state.
exec-systemcallfilter-failing.service: Failed with result 'exit-code'.
exec-systemcallfilter-failing.service: cgroup is empty
Assertion 'service->main_exec_status.status == status_expected' failed at ../src/src/test/test-execute.c:71, function check(). Aborting.
```
BTW, I tracked this race down by using perf:
```
# perf record -e cgroup:cgroup_mkdir,cgroup_rmdir
…
# perf script
```
Thanks a lot @iaguis, @alban for helping me how to use perf for this.
Fixes #5895.
2017-08-09 15:42:49 +02:00
|
|
|
|
2013-11-11 10:03:31 +01:00
|
|
|
/* Prepare the manager. */
|
2018-09-12 08:15:09 +02:00
|
|
|
assert_se(set_unit_path(get_testdata_dir()) >= 0);
|
2017-02-14 22:33:52 +01:00
|
|
|
assert_se(runtime_dir = setup_fake_runtime_dir());
|
2018-03-10 11:02:18 +01:00
|
|
|
r = manager_new(UNIT_FILE_USER, MANAGER_TEST_RUN_BASIC, &m);
|
2017-10-04 16:01:32 +02:00
|
|
|
if (IN_SET(r, -EPERM, -EACCES)) {
|
2018-09-13 13:34:12 +02:00
|
|
|
log_error_errno(r, "manager_new: %m");
|
|
|
|
|
return log_tests_skipped("cannot create manager");
|
2013-11-11 10:03:31 +01:00
|
|
|
}
|
2018-09-13 13:34:12 +02:00
|
|
|
|
2016-01-20 02:20:02 +01:00
|
|
|
assert_se(r >= 0);
|
2015-11-13 19:28:32 +01:00
|
|
|
|
|
|
|
|
/* Turn off all kinds of default accouning, so that we can
|
|
|
|
|
* verify the masks resulting of our configuration and nothing
|
|
|
|
|
* else. */
|
|
|
|
|
m->default_cpu_accounting =
|
|
|
|
|
m->default_memory_accounting =
|
|
|
|
|
m->default_blockio_accounting =
|
2016-05-19 02:35:12 +02:00
|
|
|
m->default_io_accounting =
|
2015-11-13 19:28:32 +01:00
|
|
|
m->default_tasks_accounting = false;
|
|
|
|
|
m->default_tasks_max = (uint64_t) -1;
|
|
|
|
|
|
2014-10-04 23:51:45 +02:00
|
|
|
assert_se(r >= 0);
|
2018-06-20 22:19:26 +02:00
|
|
|
assert_se(manager_startup(m, NULL, NULL) >= 0);
|
2013-11-11 10:03:31 +01:00
|
|
|
|
|
|
|
|
/* Load units and verify hierarchy. */
|
2018-04-12 15:51:39 +02:00
|
|
|
assert_se(manager_load_startable_unit_or_warn(m, "parent.slice", NULL, &parent) >= 0);
|
|
|
|
|
assert_se(manager_load_startable_unit_or_warn(m, "son.service", NULL, &son) >= 0);
|
|
|
|
|
assert_se(manager_load_startable_unit_or_warn(m, "daughter.service", NULL, &daughter) >= 0);
|
|
|
|
|
assert_se(manager_load_startable_unit_or_warn(m, "grandchild.service", NULL, &grandchild) >= 0);
|
|
|
|
|
assert_se(manager_load_startable_unit_or_warn(m, "parent-deep.slice", NULL, &parent_deep) >= 0);
|
2014-10-04 23:51:45 +02:00
|
|
|
assert_se(UNIT_DEREF(son->slice) == parent);
|
|
|
|
|
assert_se(UNIT_DEREF(daughter->slice) == parent);
|
|
|
|
|
assert_se(UNIT_DEREF(parent_deep->slice) == parent);
|
|
|
|
|
assert_se(UNIT_DEREF(grandchild->slice) == parent_deep);
|
2013-11-11 10:03:31 +01:00
|
|
|
root = UNIT_DEREF(parent->slice);
|
|
|
|
|
|
|
|
|
|
/* Verify per-unit cgroups settings. */
|
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 19:22:36 +02:00
|
|
|
assert_se(unit_get_own_mask(son) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT));
|
|
|
|
|
assert_se(unit_get_own_mask(daughter) == 0);
|
|
|
|
|
assert_se(unit_get_own_mask(grandchild) == 0);
|
|
|
|
|
assert_se(unit_get_own_mask(parent_deep) == CGROUP_MASK_MEMORY);
|
2016-05-19 02:35:12 +02:00
|
|
|
assert_se(unit_get_own_mask(parent) == (CGROUP_MASK_IO | CGROUP_MASK_BLKIO));
|
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 19:22:36 +02:00
|
|
|
assert_se(unit_get_own_mask(root) == 0);
|
2014-02-14 19:11:07 +01:00
|
|
|
|
|
|
|
|
/* Verify aggregation of member masks */
|
2014-10-04 23:51:45 +02:00
|
|
|
assert_se(unit_get_members_mask(son) == 0);
|
|
|
|
|
assert_se(unit_get_members_mask(daughter) == 0);
|
|
|
|
|
assert_se(unit_get_members_mask(grandchild) == 0);
|
|
|
|
|
assert_se(unit_get_members_mask(parent_deep) == 0);
|
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 19:22:36 +02:00
|
|
|
assert_se(unit_get_members_mask(parent) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_MEMORY));
|
2016-05-19 02:35:12 +02:00
|
|
|
assert_se(unit_get_members_mask(root) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_IO | CGROUP_MASK_BLKIO | CGROUP_MASK_MEMORY));
|
2014-02-14 19:11:07 +01:00
|
|
|
|
|
|
|
|
/* Verify aggregation of sibling masks. */
|
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 19:22:36 +02:00
|
|
|
assert_se(unit_get_siblings_mask(son) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_MEMORY));
|
|
|
|
|
assert_se(unit_get_siblings_mask(daughter) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_MEMORY));
|
2014-10-04 23:51:45 +02:00
|
|
|
assert_se(unit_get_siblings_mask(grandchild) == 0);
|
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 19:22:36 +02:00
|
|
|
assert_se(unit_get_siblings_mask(parent_deep) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_MEMORY));
|
2016-05-19 02:35:12 +02:00
|
|
|
assert_se(unit_get_siblings_mask(parent) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_IO | CGROUP_MASK_BLKIO | CGROUP_MASK_MEMORY));
|
|
|
|
|
assert_se(unit_get_siblings_mask(root) == (CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_IO | CGROUP_MASK_BLKIO | CGROUP_MASK_MEMORY));
|
2014-02-14 19:11:07 +01:00
|
|
|
|
|
|
|
|
/* Verify aggregation of target masks. */
|
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 19:22:36 +02:00
|
|
|
assert_se(unit_get_target_mask(son) == ((CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_MEMORY) & m->cgroup_supported));
|
|
|
|
|
assert_se(unit_get_target_mask(daughter) == ((CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_MEMORY) & m->cgroup_supported));
|
2014-10-04 23:51:45 +02:00
|
|
|
assert_se(unit_get_target_mask(grandchild) == 0);
|
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 19:22:36 +02:00
|
|
|
assert_se(unit_get_target_mask(parent_deep) == ((CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_MEMORY) & m->cgroup_supported));
|
2016-05-19 02:35:12 +02:00
|
|
|
assert_se(unit_get_target_mask(parent) == ((CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_IO | CGROUP_MASK_BLKIO | CGROUP_MASK_MEMORY) & m->cgroup_supported));
|
|
|
|
|
assert_se(unit_get_target_mask(root) == ((CGROUP_MASK_CPU | CGROUP_MASK_CPUACCT | CGROUP_MASK_IO | CGROUP_MASK_BLKIO | CGROUP_MASK_MEMORY) & m->cgroup_supported));
|
2013-11-11 10:03:31 +01:00
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
2017-11-08 19:01:18 +01:00
|
|
|
static void test_cg_mask_to_string_one(CGroupMask mask, const char *t) {
|
|
|
|
|
_cleanup_free_ char *b = NULL;
|
|
|
|
|
|
|
|
|
|
assert_se(cg_mask_to_string(mask, &b) >= 0);
|
|
|
|
|
assert_se(streq_ptr(b, t));
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void test_cg_mask_to_string(void) {
|
|
|
|
|
test_cg_mask_to_string_one(0, NULL);
|
|
|
|
|
test_cg_mask_to_string_one(_CGROUP_MASK_ALL, "cpu cpuacct io blkio memory devices pids");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_CPU, "cpu");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_CPUACCT, "cpuacct");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_IO, "io");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_BLKIO, "blkio");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_MEMORY, "memory");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_DEVICES, "devices");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_PIDS, "pids");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_CPU|CGROUP_MASK_CPUACCT, "cpu cpuacct");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_CPU|CGROUP_MASK_PIDS, "cpu pids");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_CPUACCT|CGROUP_MASK_PIDS, "cpuacct pids");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_DEVICES|CGROUP_MASK_PIDS, "devices pids");
|
|
|
|
|
test_cg_mask_to_string_one(CGROUP_MASK_IO|CGROUP_MASK_BLKIO, "io blkio");
|
|
|
|
|
}
|
|
|
|
|
|
2013-11-11 10:03:31 +01:00
|
|
|
int main(int argc, char* argv[]) {
|
2018-09-13 13:34:12 +02:00
|
|
|
int rc = EXIT_SUCCESS;
|
2016-04-08 18:54:05 +02:00
|
|
|
|
2018-09-13 14:31:13 +02:00
|
|
|
test_setup_logging(LOG_DEBUG);
|
2017-11-08 19:01:18 +01:00
|
|
|
|
|
|
|
|
test_cg_mask_to_string();
|
2018-09-13 13:34:12 +02:00
|
|
|
TEST_REQ_RUNNING_SYSTEMD(rc = test_cgroup_mask());
|
2016-04-08 18:54:05 +02:00
|
|
|
|
2013-11-11 10:03:31 +01:00
|
|
|
return rc;
|
|
|
|
|
}
|