android/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
19f3e328b4
android_kernel_xiaomi_sm8450/kernel
History
Jens Axboe eb42e7b304 task_work: Use TIF_NOTIFY_SIGNAL if available 
[ Upstream commit 114518eb6430b832d2f9f5a008043b913ccf0e24 ]

If the arch supports TIF_NOTIFY_SIGNAL, then use that for TWA_SIGNAL as
it's more efficient than using the signal delivery method. This is
especially true on threaded applications, where ->sighand is shared across
threads, but it's also lighter weight on non-shared cases.

io_uring is a heavy consumer of TWA_SIGNAL based task_work. A test with
threads shows a nice improvement running an io_uring based echo server.

stock kernel:
0.01% <= 0.1 milliseconds
95.86% <= 0.2 milliseconds
98.27% <= 0.3 milliseconds
99.71% <= 0.4 milliseconds
100.00% <= 0.5 milliseconds
100.00% <= 0.6 milliseconds
100.00% <= 0.7 milliseconds
100.00% <= 0.8 milliseconds
100.00% <= 0.9 milliseconds
100.00% <= 1.0 milliseconds
100.00% <= 1.1 milliseconds
100.00% <= 2 milliseconds
100.00% <= 3 milliseconds
100.00% <= 3 milliseconds
1378930.00 requests per second
~1600% CPU

1.38M requests/second, and all 16 CPUs are maxed out.

patched kernel:
0.01% <= 0.1 milliseconds
98.24% <= 0.2 milliseconds
99.47% <= 0.3 milliseconds
99.99% <= 0.4 milliseconds
100.00% <= 0.5 milliseconds
100.00% <= 0.6 milliseconds
100.00% <= 0.7 milliseconds
100.00% <= 0.8 milliseconds
100.00% <= 0.9 milliseconds
100.00% <= 1.2 milliseconds
1666111.38 requests per second
~1450% CPU

1.67M requests/second, and we're no longer just hammering on the sighand
lock. The original reporter states:

"For 5.7.15 my benchmark achieves 1.6M qps and system cpu is at ~80%.
 for 5.7.16 or later it achieves only 1M qps and the system cpu is is
 at ~100%"

with the only difference there being that TWA_SIGNAL is used
unconditionally in 5.7.16, since it's required to be able to handle the
inability to run task_work if the application is waiting in the kernel
already on an event that needs task_work run to be satisfied. Also see
commit 0ba9c9edcd.

Reported-by: Roman Gershman <romger@amazon.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Link: https://lore.kernel.org/r/20201026203230.386348-5-axboe@kernel.dk
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2023-01-04 11:39:18 +01:00
..
bpf bpf: Do not copy spin lock field from user in bpf_selem_alloc 2022-12-08 11:23:55 +01:00
cgroup memcg: fix possible use-after-free in memcg_write_event_control() 2022-12-14 11:31:57 +01:00
configs
debug lockdown: also lock down previous kgdb use 2022-05-30 09:33:22 +02:00
dma swiotlb: max mapping size takes min align mask into account 2022-10-05 10:38:40 +02:00
entry entry: Add support for TIF_NOTIFY_SIGNAL 2023-01-04 11:39:18 +01:00
events signal: Add task_sigpending() helper 2023-01-04 11:39:17 +01:00
gcov gcov: clang: fix the buffer overflow issue 2022-12-02 17:40:04 +01:00
irq genirq: Take the proposed affinity at face value if force==true 2022-12-02 17:40:06 +01:00
kcsan kcsan: Fix debugfs initcall return type 2021-05-26 12:06:54 +02:00
livepatch livepatch: fix race between fork and KLP transition 2022-10-26 13:25:14 +02:00
locking locking/lockdep: Fix lockdep_init_map_*() confusion 2022-08-21 15:15:33 +02:00
power PM: hibernate: Allow hybrid sleep to work with s2idle 2022-11-03 23:57:52 +09:00
printk printk: fix return value of printk.devkmsg __setup handler 2022-04-08 14:40:08 +02:00
rcu rcu-tasks: Convert RCU_LOCKDEP_WARN() to WARN_ONCE() 2022-10-26 13:25:44 +02:00
sched sched/fair: Fix fault in reweight_entity 2022-08-21 15:16:26 +02:00
time timekeeping: contribute wall clock to rng on time change 2022-08-21 15:16:20 +02:00
trace tracing: Free buffers when a used dynamic event is removed 2022-12-08 11:23:58 +01:00
.gitignore kbuild: update config_data.gz only when the content of .config is changed 2021-05-11 14:47:37 +02:00
acct.c kernel: acct.c: fix some kernel-doc nits 2020-10-16 11:11:19 -07:00
async.c Revert "module, async: async_synchronize_full() on module init iff async is used" 2022-02-23 12:01:00 +01:00
audit_fsnotify.c audit: fix potential double free on error path from fsnotify_add_inode_mark 2022-08-31 17:15:13 +02:00
audit_tree.c audit: move put_tree() to avoid trim_trees refcount underflow and UAF 2021-09-03 10:09:31 +02:00
audit_watch.c fsnotify: generalize handle_inode_event() 2020-12-30 11:54:18 +01:00
audit.c audit: improve audit queue handling when "audit=1" on cmdline 2022-02-08 18:30:34 +01:00
audit.h audit: log AUDIT_TIME_* records only from rules 2022-04-08 14:40:00 +02:00
auditfilter.c treewide: Use fallthrough pseudo-keyword 2020-08-23 17:36:59 -05:00
auditsc.c audit: log AUDIT_TIME_* records only from rules 2022-04-08 14:40:00 +02:00
backtracetest.c treewide: Replace DECLARE_TASKLET() with DECLARE_TASKLET_OLD() 2020-07-30 11:15:58 -07:00
bounds.c
capability.c LSM: Signal to SafeSetID when setting group IDs 2020-10-13 09:17:34 -07:00
compat.c treewide: Use fallthrough pseudo-keyword 2020-08-23 17:36:59 -05:00
configs.c
context_tracking.c context_tracking: Ensure that the critical path cannot be instrumented 2020-06-11 15:14:36 +02:00
cpu_pm.c PM: cpu: Make notifier chain use a raw_spinlock_t 2021-09-15 09:50:40 +02:00
cpu.c random: clear fast pool, crng, and batches in cpuhp bring up 2022-05-30 09:33:36 +02:00
crash_core.c crash_core, vmcoreinfo: append 'SECTION_SIZE_BITS' to vmcoreinfo 2021-06-23 14:42:52 +02:00
crash_dump.c
cred.c Revert "Add a reference to ucounts for each cred" 2021-09-08 08:49:00 +02:00
delayacct.c
dma.c
exec_domain.c
exit.c fix race between exit_itimers() and /proc/pid/timers 2022-07-21 21:19:59 +02:00
extable.c
fail_function.c fail_function: Remove a redundant mutex unlock 2020-11-19 11:58:16 -08:00
fork.c kernel: provide create_io_thread() helper 2023-01-04 11:39:16 +01:00
freezer.c Revert "kernel: freezer should treat PF_IO_WORKER like PF_KTHREAD for freezing" 2021-04-07 15:00:14 +02:00
futex.c mm, futex: fix shared futex pgoff on shmem huge page 2021-06-30 08:47:29 -04:00
gen_kheaders.sh kbuild: add variables for compression tools 2020-06-06 23:42:01 +09:00
groups.c LSM: Signal to SafeSetID when setting group IDs 2020-10-13 09:17:34 -07:00
hung_task.c kernel/hung_task.c: make type annotations consistent 2020-11-02 12:14:19 -08:00
iomem.c
irq_work.c irq_work, smp: Allow irq_work on call_single_queue 2020-05-28 10:54:15 +02:00
jump_label.c jump_label: Fix jump_label_text_reserved() vs __init 2021-07-20 16:05:58 +02:00
kallsyms.c treewide: Convert macro and uses of __section(foo) to __section("foo") 2020-10-25 14:51:49 -07:00
kcmp.c exec: Transform exec_update_mutex into a rw_semaphore 2021-01-09 13:46:24 +01:00
Kconfig.freezer
Kconfig.hz
Kconfig.locks
Kconfig.preempt
kcov.c kcov: make some symbols static 2020-08-12 10:58:02 -07:00
kexec_core.c kernel: kexec: remove the lock operation of system_transition_mutex 2021-02-03 23:28:37 +01:00
kexec_elf.c
kexec_file.c ima: force signature verification when CONFIG_KEXEC_SIG is configured 2022-07-21 21:20:11 +02:00
kexec_internal.h
kexec.c LSM: Introduce kernel_post_load_data() hook 2020-10-05 13:37:03 +02:00
kheaders.c
kmod.c kmod: remove redundant "be an" in the comment 2020-08-12 10:58:01 -07:00
kprobes.c kprobes: Skip clearing aggrprobe's post_handler in kprobe-on-ftrace case 2022-11-25 17:45:55 +01:00
ksysfs.c
kthread.c kthread: Fix PF_KTHREAD vs to_kthread() race 2021-09-03 10:09:31 +02:00
latencytop.c sysctl: pass kernel pointers to ->proc_handler 2020-04-27 02:07:40 -04:00
Makefile kbuild: update config_data.gz only when the content of .config is changed 2021-05-11 14:47:37 +02:00
module_signature.c module: harden ELF info handling 2021-03-25 09:04:11 +01:00
module_signing.c module: harden ELF info handling 2021-03-25 09:04:11 +01:00
module-internal.h
module.c module: check for exit sections in layout_sections() instead of module_init_section() 2022-05-25 09:18:02 +02:00
notifier.c notifier: Fix broken error handling pattern 2020-09-01 09:58:03 +02:00
nsproxy.c nsproxy: support CLONE_NEWTIME with setns() 2020-07-08 11:14:22 +02:00
padata.c padata: fix possible padata_works_lock deadlock 2020-09-04 17:51:55 +10:00
panic.c panic: don't dump stack twice on warn 2020-11-14 11:26:04 -08:00
params.c params: Replace zero-length array with flexible-array member 2020-10-29 17:22:59 -05:00
pid_namespace.c memcg: enable accounting for pids in nested pid namespaces 2021-09-18 13:40:36 +02:00
pid.c exec: Transform exec_update_mutex into a rw_semaphore 2021-01-09 13:46:24 +01:00
profile.c profiling: fix shift too large makes kernel panic 2022-08-21 15:16:05 +02:00
ptrace.c ptrace: Reimplement PTRACE_KILL by always sending SIGKILL 2022-06-09 10:20:49 +02:00
range.c kernel.h: split out min()/max() et al. helpers 2020-10-16 11:11:19 -07:00
reboot.c reboot: fix overflow parsing reboot cpu number 2020-11-14 11:26:03 -08:00
regset.c regset: kill ->get() 2020-07-27 14:31:12 -04:00
relay.c kernel/relay.c: drop unneeded initialization 2020-10-16 11:11:22 -07:00
resource.c kernel/resource: make walk_mem_res() find all busy IORESOURCE_MEM resources 2021-05-19 10:13:09 +02:00
rseq.c rseq: Remove broken uapi field layout on 32-bit little endian 2022-04-08 14:40:03 +02:00
scftorture.c scftorture: Fix distribution of short handler delays 2022-06-09 10:21:01 +02:00
scs.c mm: memcontrol: account kernel stack per node 2020-08-07 11:33:25 -07:00
seccomp.c seccomp: Fix setting loaded filter count during TSYNC 2021-08-18 08:59:06 +02:00
signal.c entry: Add support for TIF_NOTIFY_SIGNAL 2023-01-04 11:39:18 +01:00
smp.c smp: Fix offline cpu check in flush_smp_call_function_queue() 2022-04-20 09:23:29 +02:00
smpboot.c sched/core: Initialize the idle task with preemption disabled 2021-07-14 16:55:50 +02:00
smpboot.h
softirq.c softirq: Add debug check to __raise_softirq_irqoff() 2020-09-16 15:18:56 +02:00
stackleak.c gcc-plugins/stackleak: Use noinstr in favor of notrace 2022-02-23 12:01:00 +01:00
stacktrace.c stacktrace: Remove reliable argument from arch_stack_walk() callback 2020-09-18 14:24:16 +01:00
static_call.c static_call: Fix unused variable warn w/o MODULE 2021-09-08 08:49:00 +02:00
stop_machine.c stop_machine, rcu: Mark functions as notrace 2020-10-26 12:12:27 +01:00
sys_ni.c kernel/sys_ni: add compat entry for fadvise64_64 2022-08-31 17:15:13 +02:00
sys.c prctl: allow to setup brk for et_dyn executables 2021-09-26 14:08:57 +02:00
sysctl-test.c
sysctl.c proc: proc_skip_spaces() shouldn't think it is working on C strings 2022-12-08 11:24:00 +01:00
task_work.c task_work: Use TIF_NOTIFY_SIGNAL if available 2023-01-04 11:39:18 +01:00
taskstats.c taskstats: move specifying netlink policy back to ops 2020-10-02 19:11:12 -07:00
test_kprobes.c
torture.c torture: Dump ftrace at shutdown only if requested 2020-06-29 12:01:45 -07:00
tracepoint.c tracepoint: Use rcu get state and cond sync for static call updates 2021-09-03 10:09:30 +02:00
tsacct.c taskstats: Cleanup the use of task->exit_code 2022-01-27 10:54:33 +01:00
ucount.c Revert "Add a reference to ucounts for each cred" 2021-09-08 08:49:00 +02:00
uid16.c
uid16.h
umh.c usermodehelper: reset umask to default before executing user process 2020-10-06 10:31:52 -07:00
up.c smp: Fix smp_call_function_single_async prototype 2021-05-14 09:50:46 +02:00
user_namespace.c Revert "Add a reference to ucounts for each cred" 2021-09-08 08:49:00 +02:00
user-return-notifier.c
user.c user.c: make uidhash_table static 2020-06-04 19:06:24 -07:00
usermode_driver.c bpf: Fix umd memory leak in copy_process() 2021-03-30 14:32:03 +02:00
utsname_sysctl.c sysctl: pass kernel pointers to ->proc_handler 2020-04-27 02:07:40 -04:00
utsname.c nsproxy: add struct nsset 2020-05-09 13:57:12 +02:00
watch_queue.c watch_queue: Fix missing locking in add_watch_to_object() 2022-08-03 12:00:44 +02:00
watchdog_hld.c
watchdog.c watchdog: export lockup_detector_reconfigure 2022-08-25 11:38:20 +02:00
workqueue_internal.h
workqueue.c workqueue: don't skip lockdep work dependency in cancel_work_sync() 2022-09-28 11:10:40 +02:00