linux_dsm_epyc7002/kernel
Roman Gushchin 4bfc0bb2c6 bpf: decouple the lifetime of cgroup_bpf from cgroup itself
Currently the lifetime of bpf programs attached to a cgroup is bound
to the lifetime of the cgroup itself. It means that if a user
forgets (or intentionally avoids) to detach a bpf program before
removing the cgroup, it will stay attached up to the release of the
cgroup. Since the cgroup can stay in the dying state (the state
between being rmdir()'ed and being released) for a very long time, it
leads to a waste of memory. Also, it blocks a possibility to implement
the memcg-based memory accounting for bpf objects, because a circular
reference dependency will occur. Charged memory pages are pinning the
corresponding memory cgroup, and if the memory cgroup is pinning
the attached bpf program, nothing will be ever released.

A dying cgroup can not contain any processes, so the only chance for
an attached bpf program to be executed is a live socket associated
with the cgroup. So in order to release all bpf data early, let's
count associated sockets using a new percpu refcounter. On cgroup
removal the counter is transitioned to the atomic mode, and as soon
as it reaches 0, all bpf programs are detached.

Because cgroup_bpf_release() can block, it can't be called from
the percpu ref counter callback directly, so instead an asynchronous
work is scheduled.

The reference counter is not socket specific, and can be used for any
other types of programs, which can be executed from a cgroup-bpf hook
outside of the process context, had such a need arise in the future.

Signed-off-by: Roman Gushchin <guro@fb.com>
Cc: jolsa@redhat.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
2019-05-28 09:30:02 -07:00
..
bpf bpf: decouple the lifetime of cgroup_bpf from cgroup itself 2019-05-28 09:30:02 -07:00
cgroup bpf: decouple the lifetime of cgroup_bpf from cgroup itself 2019-05-28 09:30:02 -07:00
configs
debug treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
dma treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
events mm/mmu_notifier: use correct mmu_notifier events for each invalidation 2019-05-14 09:47:49 -07:00
gcov treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
irq treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
livepatch treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 13 2019-05-21 11:28:45 +02:00
locking treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 9 2019-05-21 11:28:40 +02:00
power treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
printk treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 13 2019-05-21 11:28:45 +02:00
rcu treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
sched treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
time treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
trace bpf: check signal validity in nmi for bpf_send_signal() helper 2019-05-28 10:51:33 +02:00
.gitignore Provide in-kernel headers to make extending kernel easier 2019-04-29 16:48:03 +02:00
acct.c acct_on(): don't mess with freeze protection 2019-04-04 21:04:13 -04:00
async.c treewide: Switch printk users from %pf and %pF to %ps and %pS, respectively 2019-04-09 14:19:06 +02:00
audit_fsnotify.c audit_compare_dname_path(): switch to const struct qstr * 2019-04-28 20:33:43 -04:00
audit_tree.c fsnotify: switch send_to_group() and ->handle_event to const struct qstr * 2019-04-26 13:51:03 -04:00
audit_watch.c audit_compare_dname_path(): switch to const struct qstr * 2019-04-28 20:33:43 -04:00
audit.c audit: connect LOGIN record to its syscall record 2019-03-20 20:57:48 -04:00
audit.h audit_compare_dname_path(): switch to const struct qstr * 2019-04-28 20:33:43 -04:00
auditfilter.c Merge branch 'work.dcache' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2019-05-07 20:03:32 -07:00
auditsc.c Merge branch 'work.dcache' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2019-05-07 20:03:32 -07:00
backtracetest.c backtrace-test: Simplify stack trace handling 2019-04-29 12:37:47 +02:00
bounds.c
capability.c LSM: add SafeSetID module that gates setid calls 2019-01-25 11:22:43 -08:00
compat.c kernel/compat.c: mark expected switch fall-throughs 2019-05-15 08:16:14 -07:00
configs.c kernel/configs: use .incbin directive to embed config_data.gz 2019-03-07 18:32:02 -08:00
context_tracking.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
cpu_pm.c
cpu.c Merge branch 'timers-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2019-05-06 14:50:46 -07:00
crash_core.c kexec: export PG_offline to VMCOREINFO 2019-03-05 21:07:14 -08:00
crash_dump.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
cred.c SELinux: Remove cred security blob poisoning 2019-01-08 13:18:44 -08:00
delayacct.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 25 2019-05-21 11:52:39 +02:00
dma.c
elfcore.c
exec_domain.c
exit.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
extable.c
fail_function.c treewide: Switch printk users from %pf and %pF to %ps and %pS, respectively 2019-04-09 14:19:06 +02:00
fork.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
freezer.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
futex.c mm/gup: change GUP fast to use flags rather than a write 'bool' 2019-05-14 09:47:46 -07:00
gen_ikh_data.sh Provide in-kernel headers to make extending kernel easier 2019-04-29 16:48:03 +02:00
groups.c
hung_task.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
iomem.c mm/resource: Use resource_overlaps() to simplify region_intersects() 2019-04-19 12:59:36 +02:00
irq_work.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
jump_label.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
kallsyms.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
kcmp.c
Kconfig.freezer treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
Kconfig.hz treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
Kconfig.locks treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
Kconfig.preempt treewide: Add SPDX license identifier - Makefile/Kconfig 2019-05-21 10:50:46 +02:00
kcov.c kcov: convert kcov.refcount to refcount_t 2019-03-07 18:32:02 -08:00
kexec_core.c power/suspend: Add function to disable secondaries for suspend 2019-05-03 19:42:41 +02:00
kexec_file.c mm: memblock: make keeping memblock memory opt-in rather than opt-out 2019-05-14 09:47:50 -07:00
kexec_internal.h
kexec.c
kheaders.c Provide in-kernel headers to make extending kernel easier 2019-04-29 16:48:03 +02:00
kmod.c
kprobes.c kprobes: Fix error check when reusing optimized probes 2019-04-16 09:38:16 +02:00
ksysfs.c
kthread.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
latencytop.c kernel/latencytop.c: rename clear_all_latency_tracing to clear_tsk_latency_tracing 2019-05-14 19:52:49 -07:00
Makefile kernel/Makefile: don't assume that kernel/gen_ikh_data.sh is executable 2019-05-14 19:52:47 -07:00
memremap.c kernel/memremap.c: remove the unused device_private_entry_fault() export 2019-05-14 09:47:51 -07:00
module_signing.c
module-internal.h kallsyms: store type information in its own array 2019-03-28 15:00:37 +01:00
module.c Modules updates for v5.2 2019-05-14 10:55:54 -07:00
notifier.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
nsproxy.c
padata.c padata: Replace padata_attr_type default_attrs field with groups 2019-04-25 22:06:11 +02:00
panic.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
params.c
pid_namespace.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
pid.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
profile.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
ptrace.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
range.c
reboot.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
relay.c Merge branch 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2019-03-12 13:27:20 -07:00
resource.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
rseq.c rseq: Remove superfluous rseq_len from task_struct 2019-04-19 12:39:32 +02:00
seccomp.c audit/stable-5.2 PR 20190507 2019-05-07 19:06:04 -07:00
signal.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
smp.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
smpboot.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
smpboot.h
softirq.c softirq: Remove tasklet_hrtimer 2019-03-22 14:36:02 +01:00
stackleak.c
stacktrace.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
stop_machine.c treewide: Switch printk users from %pf and %pF to %ps and %pS, respectively 2019-04-09 14:19:06 +02:00
sys_ni.c signal: support CLONE_PIDFD with pidfd_send_signal 2019-05-07 14:31:03 +02:00
sys.c kernel/sys.c: prctl: fix false positive in validate_prctl_map() 2019-05-14 09:47:44 -07:00
sysctl_binary.c kernel/sysctl: add panic_print into sysctl 2019-01-04 13:13:47 -08:00
sysctl.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
task_work.c
taskstats.c genetlink: optionally validate strictly/dumps 2019-04-27 17:07:22 -04:00
test_kprobes.c treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 25 2019-05-21 11:52:39 +02:00
torture.c torture: Don't try to offline the last CPU 2019-03-26 14:42:53 -07:00
tracepoint.c
tsacct.c
ucount.c
uid16.c
uid16.h
umh.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
up.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
user_namespace.c
user-return-notifier.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
user.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00
utsname_sysctl.c
utsname.c
watchdog_hld.c kernel/watchdog_hld.c: hard lockup message should end with a newline 2019-04-19 09:46:05 -07:00
watchdog.c watchdog: Fix typo in comment 2019-04-18 14:05:51 +02:00
workqueue_internal.h sched/core, workqueues: Distangle worker accounting from rq lock 2019-04-16 16:55:15 +02:00
workqueue.c treewide: Add SPDX license identifier for missed files 2019-05-21 10:50:45 +02:00