leandrof/linux
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
f5bdb34bf0
linux/kernel
History
David Vernet f5bdb34bf0 livepatch: Avoid CPU hogging with cond_resched 
When initializing a 'struct klp_object' in klp_init_object_loaded(), and
performing relocations in klp_resolve_symbols(), klp_find_object_symbol()
is invoked to look up the address of a symbol in an already-loaded module
(or vmlinux). This, in turn, calls kallsyms_on_each_symbol() or
module_kallsyms_on_each_symbol() to find the address of the symbol that is
being patched.

It turns out that symbol lookups often take up the most CPU time when
enabling and disabling a patch, and may hog the CPU and cause other tasks
on that CPU's runqueue to starve -- even in paths where interrupts are
enabled.  For example, under certain workloads, enabling a KLP patch with
many objects or functions may cause ksoftirqd to be starved, and thus for
interrupts to be backlogged and delayed. This may end up causing TCP
retransmits on the host where the KLP patch is being applied, and in
general, may cause any interrupts serviced by softirqd to be delayed while
the patch is being applied.

So as to ensure that kallsyms_on_each_symbol() does not end up hogging the
CPU, this patch adds a call to cond_resched() in kallsyms_on_each_symbol()
and module_kallsyms_on_each_symbol(), which are invoked when doing a symbol
lookup in vmlinux and a module respectively.  Without this patch, if a
live-patch is applied on a 36-core Intel host with heavy TCP traffic, a
~10x spike is observed in TCP retransmits while the patch is being applied.
Additionally, collecting sched events with perf indicates that ksoftirqd is
awakened ~1.3 seconds before it's eventually scheduled.  With the patch, no
increase in TCP retransmit events is observed, and ksoftirqd is scheduled
shortly after it's awakened.

Signed-off-by: David Vernet <void@manifault.com>
Acked-by: Miroslav Benes <mbenes@suse.cz>
Acked-by: Song Liu <song@kernel.org>
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20211229215646.830451-1-void@manifault.com
2022-01-07 12:00:53 +01:00
..
bpf Merge https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next 2021-08-30 16:42:47 -07:00
cgroup Merge branch 'akpm' (patches from Andrew) 2021-09-03 10:08:28 -07:00
configs drivers/char: remove /dev/kmem for good 2021-05-07 00:26:34 -07:00
debug kernel: debug: Fix unreachable code in gdb_serial_stub() 2021-07-12 11:03:35 -05:00
dma Merge branch 'stable/for-linus-5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/swiotlb 2021-09-03 10:34:44 -07:00
entry tick/nohz: Only check for RCU deferred wakeup on user/guest entry when needed 2021-05-31 10:14:49 +02:00
events Core: 2021-08-31 16:43:06 -07:00
gcov Kconfig: Introduce ARCH_WANTS_NO_INSTR and CC_HAS_NO_PROFILE_FN_ATTR 2021-06-22 11:07:18 -07:00
irq Updates to the interrupt core and driver subsystems: 2021-08-30 14:38:37 -07:00
kcsan LKMM updates: 2021-09-02 13:00:15 -07:00
livepatch livepatch: Replace deprecated CPU-hotplug functions. 2021-08-19 12:00:24 +02:00
locking Updates for locking and atomics: 2021-08-30 14:26:36 -07:00
power Merge branches 'pm-pci', 'pm-sleep', 'pm-domains' and 'powercap' 2021-08-30 19:25:42 +02:00
printk Merge branch 'rework/printk_safe-removal' into for-linus 2021-08-30 16:36:10 +02:00
rcu Updates for locking and atomics: 2021-08-30 14:26:36 -07:00
sched Power management updates for 5.15-rc1 2021-08-31 13:21:58 -07:00
time Merge branch 'akpm' (patches from Andrew) 2021-09-03 10:08:28 -07:00
trace printk changes for 5.15 2021-09-01 18:41:13 -07:00
.gitignore .gitignore: prefix local generated files with a slash 2021-05-02 00:43:35 +09:00
acct.c kernel/acct.c: use #elif instead of #end and #elif 2020-12-15 22:46:15 -08:00
async.c kernel/async.c: remove async_unregister_domain() 2021-05-07 00:26:33 -07:00
audit_fsnotify.c audit_alloc_mark(): don't open-code ERR_CAST() 2021-02-23 10:25:27 -05:00
audit_tree.c audit: move put_tree() to avoid trim_trees refcount underflow and UAF 2021-08-24 18:52:36 -04:00
audit_watch.c fsnotify: generalize handle_inode_event() 2020-12-03 14:58:35 +01:00
audit.c lsm: separate security_task_getsecid() into subjective and objective variants 2021-03-22 15:23:32 -04:00
audit.h audit: add header protection to kernel/audit.h 2021-07-19 22:38:24 -04:00
auditfilter.c lsm: separate security_task_getsecid() into subjective and objective variants 2021-03-22 15:23:32 -04:00
auditsc.c audit: remove trailing spaces and tabs 2021-06-10 20:59:05 -04:00
backtracetest.c
bounds.c
capability.c capability: handle idmapped mounts 2021-01-24 14:27:16 +01:00
cfi.c cfi: Use rcu_read_{un}lock_sched_notrace 2021-08-11 13:11:12 -07:00
compat.c
configs.c
context_tracking.c
cpu_pm.c PM: cpu: Make notifier chain use a raw_spinlock_t 2021-08-16 18:55:32 +02:00
cpu.c cpu/hotplug: Add debug printks for hotplug callback failures 2021-08-10 18:31:32 +02:00
crash_core.c kdump: use vmlinux_build_id to simplify 2021-07-08 11:48:22 -07:00
crash_dump.c
cred.c ucounts: Increase ucounts reference counter before the security hook 2021-08-23 16:13:04 -05:00
delayacct.c delayacct: Add sysctl to enable at runtime 2021-05-12 11:43:25 +02:00
dma.c
exec_domain.c
exit.c io_uring: remove files pointer in cancellation functions 2021-08-23 13:10:37 -06:00
extable.c
fail_function.c fault-injection: handle EI_ETYPE_TRUE 2020-12-15 22:46:19 -08:00
fork.c Core: 2021-08-31 16:43:06 -07:00
freezer.c sched: Add get_current_state() 2021-06-18 11:43:08 +02:00
futex.c futex: Prevent requeue_pi() lock nesting issue on RT 2021-08-17 19:05:59 +02:00
gen_kheaders.sh kbuild: clean up ${quiet} checks in shell scripts 2021-05-27 04:01:50 +09:00
groups.c groups: simplify struct group_info allocation 2021-02-26 09:41:03 -08:00
hung_task.c Merge branch 'akpm' (patches from Andrew) 2021-07-02 12:08:10 -07:00
iomem.c
irq_work.c irq_work: Make irq_work_queue() NMI-safe again 2021-06-10 10:00:08 +02:00
jump_label.c jump_label: Fix jump_label_text_reserved() vs __init 2021-07-05 10:46:20 +02:00
kallsyms.c livepatch: Avoid CPU hogging with cond_resched 2022-01-07 12:00:53 +01:00
kcmp.c Merge branch 'exec-update-lock-for-v5.11' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace 2020-12-15 19:36:48 -08:00
Kconfig.freezer
Kconfig.hz
Kconfig.locks locking/rwlock: Provide RT variant 2021-08-17 17:50:51 +02:00
Kconfig.preempt sched/core: Disable CONFIG_SCHED_CORE by default 2021-06-28 22:43:05 +02:00
kcov.c kernel: make kcov_common_handle consider the current context 2020-11-02 18:00:20 -08:00
kexec_core.c Merge branch 'rework/printk_safe-removal' into for-linus 2021-08-30 16:36:10 +02:00
kexec_elf.c
kexec_file.c kernel: kexec_file: fix error return code of kexec_calculate_store_digests() 2021-05-07 00:26:32 -07:00
kexec_internal.h kexec: move machine_kexec_post_load() to public interface 2021-02-22 12:33:26 +00:00
kexec.c
kheaders.c
kmod.c modules: add CONFIG_MODPROBE_PATH 2021-05-07 00:26:33 -07:00
kprobes.c Locking fixes: 2021-07-11 11:06:09 -07:00
ksysfs.c
kthread.c Merge branch 'akpm' (patches from Andrew) 2021-06-29 17:29:11 -07:00
latencytop.c
Makefile kbuild: update config_data.gz only when the content of .config is changed 2021-05-02 00:43:35 +09:00
module_signature.c module: harden ELF info handling 2021-01-19 10:24:45 +01:00
module_signing.c module: harden ELF info handling 2021-01-19 10:24:45 +01:00
module-internal.h
module.c livepatch: Avoid CPU hogging with cond_resched 2022-01-07 12:00:53 +01:00
notifier.c notifier: Remove atomic_notifier_call_chain_robust() 2021-08-16 18:55:32 +02:00
nsproxy.c memcg: enable accounting for new namesapces and struct nsproxy 2021-09-03 09:58:12 -07:00
padata.c padata: Remove repeated verbose license text 2021-08-27 16:30:18 +08:00
panic.c Merge branch 'rework/printk_safe-removal' into for-linus 2021-08-30 16:36:10 +02:00
params.c params: lift param_set_uint_minmax to common code 2021-08-16 14:42:22 +02:00
pid_namespace.c memcg: enable accounting for new namesapces and struct nsproxy 2021-09-03 09:58:12 -07:00
pid.c kernel/pid.c: implement additional checks upon pidfd_create() parameters 2021-08-10 12:53:07 +02:00
profile.c kernel: Initialize cpumask before parsing 2021-04-10 13:35:54 +02:00
ptrace.c sched: Change task_struct::state 2021-06-18 11:43:09 +02:00
range.c
reboot.c reboot: Add hardware protection power-off 2021-06-21 13:08:36 +01:00
regset.c
relay.c relay: allow the use of const callback structs 2020-12-15 22:46:18 -08:00
resource_kunit.c resource: provide meaningful MODULE_LICENSE() in test suite 2020-11-25 18:52:35 +01:00
resource.c kernel/resource: fix return code check in __request_free_mem_region 2021-05-14 19:41:32 -07:00
rseq.c rseq: Optimise rseq_get_rseq_cs() and clear_rseq_cs() 2021-04-14 18:04:09 +02:00
scftorture.c scftorture: Avoid NULL pointer exception on early exit 2021-07-27 11:39:30 -07:00
scs.c scs: switch to vmapped shadow stacks 2020-12-01 10:30:28 +00:00
seccomp.c Merge branch 'exit-cleanups-for-v5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace 2021-09-01 14:52:05 -07:00
signal.c Merge branch 'akpm' (patches from Andrew) 2021-09-03 10:08:28 -07:00
smp.c smp: Fix all kernel-doc warnings 2021-08-11 14:47:16 +02:00
smpboot.c smpboot: Replace deprecated CPU-hotplug functions. 2021-08-10 14:57:42 +02:00
smpboot.h
softirq.c genirq: Change force_irqthreads to a static key 2021-08-10 22:50:07 +02:00
stackleak.c
stacktrace.c
static_call.c static_call: Fix static_call_text_reserved() vs __init 2021-07-05 10:46:33 +02:00
stop_machine.c stop_machine: Add caller debug info to queue_stop_cpus_work 2021-03-23 16:01:58 +01:00
sys_ni.c Merge branch 'akpm' (patches from Andrew) 2021-09-03 10:08:28 -07:00
sys.c set_user: add capability check when rlimit(RLIMIT_NPROC) exceeds 2021-08-12 14:54:25 +02:00
sysctl-test.c kernel/sysctl-test: Remove some casts which are no-longer required 2021-06-23 16:41:24 -06:00
sysctl.c Merge branch 'akpm' (patches from Andrew) 2021-09-03 10:08:28 -07:00
task_work.c kasan: record task_work_add() call stack 2021-04-30 11:20:42 -07:00
taskstats.c treewide: rename nla_strlcpy to nla_strscpy. 2020-11-16 08:08:54 -08:00
test_kprobes.c
torture.c torture: Replace deprecated CPU-hotplug functions. 2021-08-10 10:48:07 -07:00
tracepoint.c tracepoint: Use rcu get state and cond sync for static call updates 2021-08-06 10:54:41 -04:00
tsacct.c
ucount.c ucounts: add missing data type changes 2021-08-09 15:45:02 -05:00
uid16.c
uid16.h
umh.c kernel/umh.c: fix some spelling mistakes 2021-05-07 00:26:34 -07:00
up.c A set of locking related fixes and updates: 2021-05-09 13:07:03 -07:00
user_namespace.c memcg: enable accounting for new namesapces and struct nsproxy 2021-09-03 09:58:12 -07:00
user-return-notifier.c
user.c Reimplement RLIMIT_MEMLOCK on top of ucounts 2021-04-30 14:14:02 -05:00
usermode_driver.c Merge branch 'work.namei' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs 2021-07-03 11:41:14 -07:00
utsname_sysctl.c
utsname.c
watch_queue.c watch_queue: rectify kernel-doc for init_watch() 2021-01-26 11:16:34 +00:00
watchdog_hld.c
watchdog.c kernel: watchdog: modify the explanation related to watchdog thread 2021-06-29 10:53:46 -07:00
workqueue_internal.h workqueue: Assign a color to barrier work items 2021-08-17 07:49:10 -10:00
workqueue.c workqueue: Assign a color to barrier work items 2021-08-17 07:49:10 -10:00