[ Upstream commit 3002153a91a9732a6d1d0bb95138593c7da15743 ]
The RCU priority boosting can fail in two situations:
1) If (nr_cpus= > maxcpus=), which means if the total number of CPUs
is higher than those brought online at boot, then torture_onoff() may
later bring up CPUs that weren't online on boot. Now since rcutorture
initialization only boosts the ksoftirqds of the CPUs that have been
set online on boot, the CPUs later set online by torture_onoff won't
benefit from the boost, making RCU priority boosting fail.
2) The ksoftirqd kthreads are boosted after the creation of
rcu_torture_boost() kthreads, which opens a window large enough for these
rcu_torture_boost() kthreads to wait (despite running at FIFO priority)
for ksoftirqds that are still running at SCHED_NORMAL priority.
The issues can trigger for example with:
./kvm.sh --configs TREE01 --kconfig "CONFIG_RCU_BOOST=y"
[ 34.968561] rcu-torture: !!!
[ 34.968627] ------------[ cut here ]------------
[ 35.014054] WARNING: CPU: 4 PID: 114 at kernel/rcu/rcutorture.c:1979 rcu_torture_stats_print+0x5ad/0x610
[ 35.052043] Modules linked in:
[ 35.069138] CPU: 4 PID: 114 Comm: rcu_torture_sta Not tainted 5.18.0-rc1 #1
[ 35.096424] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.14.0-0-g155821a-rebuilt.opensuse.org 04/01/2014
[ 35.154570] RIP: 0010:rcu_torture_stats_print+0x5ad/0x610
[ 35.198527] Code: 63 1b 02 00 74 02 0f 0b 48 83 3d 35 63 1b 02 00 74 02 0f 0b 48 83 3d 21 63 1b 02 00 74 02 0f 0b 48 83 3d 0d 63 1b 02 00 74 02 <0f> 0b 83 eb 01 0f 8e ba fc ff ff 0f 0b e9 b3 fc ff f82
[ 37.251049] RSP: 0000:ffffa92a0050bdf8 EFLAGS: 00010202
[ 37.277320] rcu: De-offloading 8
[ 37.290367] RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000001
[ 37.290387] RDX: 0000000000000000 RSI: 00000000ffffbfff RDI: 00000000ffffffff
[ 37.290398] RBP: 000000000000007b R08: 0000000000000000 R09: c0000000ffffbfff
[ 37.290407] R10: 000000000000002a R11: ffffa92a0050bc18 R12: ffffa92a0050be20
[ 37.290417] R13: ffffa92a0050be78 R14: 0000000000000000 R15: 000000000001bea0
[ 37.290427] FS: 0000000000000000(0000) GS:ffff96045eb00000(0000) knlGS:0000000000000000
[ 37.290448] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 37.290460] CR2: 0000000000000000 CR3: 000000001dc0c000 CR4: 00000000000006e0
[ 37.290470] Call Trace:
[ 37.295049] <TASK>
[ 37.295065] ? preempt_count_add+0x63/0x90
[ 37.295095] ? _raw_spin_lock_irqsave+0x12/0x40
[ 37.295125] ? rcu_torture_stats_print+0x610/0x610
[ 37.295143] rcu_torture_stats+0x29/0x70
[ 37.295160] kthread+0xe3/0x110
[ 37.295176] ? kthread_complete_and_exit+0x20/0x20
[ 37.295193] ret_from_fork+0x22/0x30
[ 37.295218] </TASK>
Fix this with boosting the ksoftirqds kthreads from the boosting
hotplug callback itself and before the boosting kthreads are created.
Fixes: ea6d962e80b6 ("rcutorture: Judge RCU priority boosting on grace periods, not callbacks")
Signed-off-by: Frederic Weisbecker <frederic@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
Pull RCU fix from Paul McKenney:
"This contains a pair of commits that fix 282d8998e997 ("srcu: Prevent
expedited GPs and blocking readers from consuming CPU"), which was
itself a fix to an SRCU expedited grace-period problem that could
prevent kernel live patching (KLP) from completing.
That SRCU fix for KLP introduced large (as in minutes) boot-time
delays to embedded Linux kernels running on qemu/KVM. These delays
were due to the emulation of certain MMIO operations controlling
memory layout, which were emulated with one expedited grace period per
access. Common configurations required thousands of boot-time MMIO
accesses, and thus thousands of boot-time expedited SRCU grace
periods.
In these configurations, the occasional sleeps that allowed KLP to
proceed caused excessive boot delays. These commits preserve enough
sleeps to permit KLP to proceed, but few enough that the virtual
embedded kernels still boot reasonably quickly.
This represents a regression introduced in the v5.19 merge window, and
the bug is causing significant inconvenience"
* tag 'rcu-urgent.2022.07.21a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
srcu: Make expedited RCU grace periods block even less frequently
srcu: Block less aggressively for expedited grace periods
The purpose of commit 282d8998e997 ("srcu: Prevent expedited GPs
and blocking readers from consuming CPU") was to prevent a long
series of never-blocking expedited SRCU grace periods from blocking
kernel-live-patching (KLP) progress. Although it was successful, it also
resulted in excessive boot times on certain embedded workloads running
under qemu with the "-bios QEMU_EFI.fd" command line. Here "excessive"
means increasing the boot time up into the three-to-four minute range.
This increase in boot time was due to the more than 6000 back-to-back
invocations of synchronize_rcu_expedited() within the KVM host OS, which
in turn resulted from qemu's emulation of a long series of MMIO accesses.
Commit 640a7d37c3f4 ("srcu: Block less aggressively for expedited grace
periods") did not significantly help this particular use case.
Zhangfei Gao and Shameerali Kolothum Thodi did experiments varying the
value of SRCU_MAX_NODELAY_PHASE with HZ=250 and with various values
of non-sleeping per phase counts on a system with preemption enabled,
and observed the following boot times:
+──────────────────────────+────────────────+
| SRCU_MAX_NODELAY_PHASE | Boot time (s) |
+──────────────────────────+────────────────+
| 100 | 30.053 |
| 150 | 25.151 |
| 200 | 20.704 |
| 250 | 15.748 |
| 500 | 11.401 |
| 1000 | 11.443 |
| 10000 | 11.258 |
| 1000000 | 11.154 |
+──────────────────────────+────────────────+
Analysis on the experiment results show additional improvements with
CPU-bound delays approaching one jiffy in duration. This improvement was
also seen when number of per-phase iterations were scaled to one jiffy.
This commit therefore scales per-grace-period phase number of non-sleeping
polls so that non-sleeping polls extend for about one jiffy. In addition,
the delay-calculation call to srcu_get_delay() in srcu_gp_end() is
replaced with a simple check for an expedited grace period. This change
schedules callback invocation immediately after expedited grace periods
complete, which results in greatly improved boot times. Testing done
by Marc and Zhangfei confirms that this change recovers most of the
performance degradation in boottime; for CONFIG_HZ_250 configuration,
specifically, boot times improve from 3m50s to 41s on Marc's setup;
and from 2m40s to ~9.7s on Zhangfei's setup.
In addition to the changes to default per phase delays, this
change adds 3 new kernel parameters - srcutree.srcu_max_nodelay,
srcutree.srcu_max_nodelay_phase, and srcutree.srcu_retry_check_delay.
This allows users to configure the srcu grace period scanning delays in
order to more quickly react to additional use cases.
Fixes: 640a7d37c3f4 ("srcu: Block less aggressively for expedited grace periods")
Fixes: 282d8998e997 ("srcu: Prevent expedited GPs and blocking readers from consuming CPU")
Reported-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Reported-by: yueluck <yueluck@163.com>
Signed-off-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Tested-by: Marc Zyngier <maz@kernel.org>
Tested-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Link: https://lore.kernel.org/all/20615615-0013-5adc-584f-2b1d5c03ebfc@linaro.org/
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Commit 282d8998e997 ("srcu: Prevent expedited GPs and blocking readers
from consuming CPU") fixed a problem where a long-running expedited SRCU
grace period could block kernel live patching. It did so by giving up
on expediting once a given SRCU expedited grace period grew too old.
Unfortunately, this added excessive delays to boots of virtual embedded
systems specifying "-bios QEMU_EFI.fd" to qemu. This commit therefore
makes the transition away from expediting less aggressive, increasing
the per-grace-period phase number of non-sleeping polls of readers from
one to three and increasing the required grace-period age from one jiffy
(actually from zero to one jiffies) to two jiffies (actually from one
to two jiffies).
Fixes: 282d8998e997 ("srcu: Prevent expedited GPs and blocking readers from consuming CPU")
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Reported-by: Zhangfei Gao <zhangfei.gao@linaro.org>
Reported-by: chenxiang (M)" <chenxiang66@hisilicon.com>
Cc: Shameerali Kolothum Thodi <shameerali.kolothum.thodi@huawei.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Link: https://lore.kernel.org/all/20615615-0013-5adc-584f-2b1d5c03ebfc@linaro.org/
This reverts commit 2bb2b7b57f81255c13f4395ea911d6bdc70c9fe2. The testing of 5.19 release candidates revealed missing synchronization between early and regular console functionality. It would be possible to start the console kthreads later as a workaround. But it is clear that console lock serialized console drivers between each other. It opens a big area of possible problems that were not considered by people involved in the development and review. printk() is crucial for debugging kernel issues and console output is very important part of it. The number of consoles is huge and a proper review would take some time. As a result it need to be reverted for 5.19. Link: https://lore.kernel.org/r/YrBdjVwBOVgLfHyb@alley Signed-off-by: Petr Mladek <pmladek@suse.com> Link: https://lore.kernel.org/r/20220623145157.21938-7-pmladek@suse.com
Pull sysctl updates from Luis Chamberlain: "For two kernel releases now kernel/sysctl.c has been being cleaned up slowly, since the tables were grossly long, sprinkled with tons of #ifdefs and all this caused merge conflicts with one susbystem or another. This tree was put together to help try to avoid conflicts with these cleanups going on different trees at time. So nothing exciting on this pull request, just cleanups. Thanks a lot to the Uniontech and Huawei folks for doing some of this nasty work" * tag 'sysctl-5.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/linux: (28 commits) sched: Fix build warning without CONFIG_SYSCTL reboot: Fix build warning without CONFIG_SYSCTL kernel/kexec_core: move kexec_core sysctls into its own file sysctl: minor cleanup in new_dir() ftrace: fix building with SYSCTL=y but DYNAMIC_FTRACE=n fs/proc: Introduce list_for_each_table_entry for proc sysctl mm: fix unused variable kernel warning when SYSCTL=n latencytop: move sysctl to its own file ftrace: fix building with SYSCTL=n but DYNAMIC_FTRACE=y ftrace: Fix build warning ftrace: move sysctl_ftrace_enabled to ftrace.c kernel/do_mount_initrd: move real_root_dev sysctls to its own file kernel/delayacct: move delayacct sysctls to its own file kernel/acct: move acct sysctls to its own file kernel/panic: move panic sysctls to its own file kernel/lockdep: move lockdep sysctls to its own file mm: move page-writeback sysctls to their own file mm: move oom_kill sysctls to their own file kernel/reboot: move reboot sysctls to its own file sched: Move energy_aware sysctls to topology.c ...
Pull printk updates from Petr Mladek: - Offload writing printk() messages on consoles to per-console kthreads. It prevents soft-lockups when an extensive amount of messages is printed. It was observed, for example, during boot of large systems with a lot of peripherals like disks or network interfaces. It prevents live-lockups that were observed, for example, when messages about allocation failures were reported and a CPU handled consoles instead of reclaiming the memory. It was hard to solve even with rate limiting because it would need to take into account the amount of messages and the speed of all consoles. It is a must to have for real time. Otherwise, any printk() might break latency guarantees. The per-console kthreads allow to handle each console on its own speed. Slow consoles do not longer slow down faster ones. And printk() does not longer unpredictably slows down various code paths. There are situations when the kthreads are either not available or not reliable, for example, early boot, suspend, or panic. In these situations, printk() uses the legacy mode and tries to handle consoles immediately. - Add documentation for the printk index. * tag 'printk-for-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/printk/linux: printk, tracing: fix console tracepoint printk: remove @console_locked printk: extend console_lock for per-console locking printk: add kthread console printers printk: add functions to prefer direct printing printk: add pr_flush() printk: move buffer definitions into console_emit_next_record() caller printk: refactor and rework printing logic printk: add con_printk() macro for console details printk: call boot_delay_msec() in printk_delay() printk: get caller_id/timestamp after migration disable printk: wake waiters for safe and NMI contexts printk: wake up all waiters printk: add missing memory barrier to wake_up_klogd() printk: cpu sync always disable interrupts printk: rename cpulock functions printk/index: Printk index feature documentation MAINTAINERS: Add printk indexing maintainers on mention of printk_index
exp.2022.05.11a: Expedited-grace-period latency-reduction updates.
Enabling CONFIG_RCU_BOOST did not reduce RCU expedited grace-period latency because its workqueues run at SCHED_OTHER, and thus can be delayed by normal processes. This commit avoids these delays by moving the expedited GP work items to a real-time-priority kthread_worker. This option is controlled by CONFIG_RCU_EXP_KTHREAD and disabled by default on PREEMPT_RT=y kernels which disable expedited grace periods after boot by unconditionally setting rcupdate.rcu_normal_after_boot=1. The results were evaluated on arm64 Android devices (6GB ram) running 5.10 kernel, and capturing trace data in critical user-level code. The table below shows the resulting order-of-magnitude improvements in synchronize_rcu_expedited() latency: ------------------------------------------------------------------------ | | workqueues | kthread_worker | Diff | ------------------------------------------------------------------------ | Count | 725 | 688 | | ------------------------------------------------------------------------ | Min Duration (ns) | 326 | 447 | 37.12% | ------------------------------------------------------------------------ | Q1 (ns) | 39,428 | 38,971 | -1.16% | ------------------------------------------------------------------------ | Q2 - Median (ns) | 98,225 | 69,743 | -29.00% | ------------------------------------------------------------------------ | Q3 (ns) | 342,122 | 126,638 | -62.98% | ------------------------------------------------------------------------ | Max Duration (ns) | 372,766,967 | 2,329,671 | -99.38% | ------------------------------------------------------------------------ | Avg Duration (ns) | 2,746,353 | 151,242 | -94.49% | ------------------------------------------------------------------------ | Standard Deviation (ns) | 19,327,765 | 294,408 | | ------------------------------------------------------------------------ The below table show the range of maximums/minimums for synchronize_rcu_expedited() latency from all experiments: ------------------------------------------------------------------------ | | workqueues | kthread_worker | Diff | ------------------------------------------------------------------------ | Total No. of Experiments | 25 | 23 | | ------------------------------------------------------------------------ | Largest Maximum (ns) | 372,766,967 | 2,329,671 | -99.38% | ------------------------------------------------------------------------ | Smallest Maximum (ns) | 38,819 | 86,954 | 124.00% | ------------------------------------------------------------------------ | Range of Maximums (ns) | 372,728,148 | 2,242,717 | | ------------------------------------------------------------------------ | Largest Minimum (ns) | 88,623 | 27,588 | -68.87% | ------------------------------------------------------------------------ | Smallest Minimum (ns) | 326 | 447 | 37.12% | ------------------------------------------------------------------------ | Range of Minimums (ns) | 88,297 | 27,141 | | ------------------------------------------------------------------------ Cc: "Paul E. McKenney" <paulmck@kernel.org> Cc: Tejun Heo <tj@kernel.org> Reported-by: Tim Murray <timmurray@google.com> Reported-by: Wei Wang <wvw@google.com> Tested-by: Kyle Lin <kylelin@google.com> Tested-by: Chunwei Lu <chunweilu@google.com> Tested-by: Lulu Wang <luluw@google.com> Signed-off-by: Kalesh Singh <kaleshsingh@google.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org>