Pull RCU fix from Paul McKenney:
"This contains a pair of commits that fix 282d8998e9 ("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 282d8998e9 ("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: 282d8998e9 ("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 282d8998e9 ("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: 282d8998e9 ("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 2bb2b7b57f.
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
Currently both expedited and regular grace period stall warnings use
a single timeout value that with units of seconds. However, recent
Android use cases problem require a sub-100-millisecond expedited RCU CPU
stall warning. Given that expedited RCU grace periods normally complete
in far less than a single millisecond, especially for small systems,
this is not unreasonable.
Therefore introduce the CONFIG_RCU_EXP_CPU_STALL_TIMEOUT kernel
configuration that defaults to 20 msec on Android and remains the same
as that of the non-expedited stall warnings otherwise. It also can be
changed in run-time via: /sys/.../parameters/rcu_exp_cpu_stall_timeout.
[ paulmck: Default of zero to use CONFIG_RCU_STALL_TIMEOUT. ]
Signed-off-by: Uladzislau Rezki <uladzislau.rezki@sony.com>
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Commit 9c7ef4c30f12 ("srcu: Make Tree SRCU able to operate without
snp_node array") initializes the local variable sdp differently depending
on the srcu's state in srcu_gp_start(). Either way, this initialization
overwrites the value used when sdp is defined.
This commit therefore drops this pointless definition-time initialization.
Although there is no functional change, compiler code generation may
be affected.
Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
If an SRCU reader blocks while a synchronize_srcu_expedited() waits for
that same reader, then that grace period will spawn an endless series of
workqueue handlers, consuming a full CPU. This quickly gets pointless
because consuming more CPU isn't going to make that reader get done
faster, especially if it is blocked waiting for an external event.
This commit therefore spawns at most one pair of back-to-back workqueue
handlers per expedited grace period phase, instead inserting increasing
delays as that grace period phase grows older, but capped at 10 jiffies.
In any case, if there have been at least 100 back-to-back workqueue
handlers within a single jiffy, regardless of grace period or grace-period
phase, then a one-jiffy delay is inserted.
[ paulmck: Apply feedback from kernel test robot. ]
Cc: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Reported-by: Song Liu <song@kernel.org>
Tested-by: kernel test robot <oliver.sang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit increases the sensitivity of contention detection by adding
checks to the acquisition of the srcu_data structure's lock on the
call_srcu() code path.
Co-developed-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Signed-off-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
This commit adds a srcutree.convert_to_big option of zero that causes
SRCU to decide at boot whether to wait for contention (small systems) or
immediately expand to large (large systems). A new srcutree.big_cpu_lim
(defaulting to 128) defines how many CPUs constitute a large system.
Co-developed-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Signed-off-by: Neeraj Upadhyay <quic_neeraju@quicinc.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Once kthread printing is available, console printing will no longer
occur in the context of the printk caller. However, there are some
special contexts where it is desirable for the printk caller to
directly print out kernel messages. Using pr_flush() to wait for
threaded printers is only possible if the caller is in a sleepable
context and the kthreads are active. That is not always the case.
Introduce printk_prefer_direct_enter() and printk_prefer_direct_exit()
functions to explicitly (and globally) activate/deactivate preferred
direct console printing. The term "direct console printing" refers to
printing to all enabled consoles from the context of the printk
caller. The term "prefer" is used because this type of printing is
only best effort. If the console is currently locked or other
printers are already actively printing, the printk caller will need
to rely on the other contexts to handle the printing.
This preferred direct printing is how all printing has been handled
until now (unless it was explicitly deferred).
When kthread printing is introduced, there may be some unanticipated
problems due to kthreads being unable to flush important messages.
In order to minimize such risks, preferred direct printing is
activated for the primary important messages when the system
experiences general types of major errors. These are:
- emergency reboot/shutdown
- cpu and rcu stalls
- hard and soft lockups
- hung tasks
- warn
- sysrq
Note that since kthread printing does not yet exist, no behavior
changes result from this commit. This is only implementing the
counter and marking the various places where preferred direct
printing is active.
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Reviewed-by: Petr Mladek <pmladek@suse.com>
Acked-by: Paul E. McKenney <paulmck@kernel.org> # for RCU
Signed-off-by: Petr Mladek <pmladek@suse.com>
Link: https://lore.kernel.org/r/20220421212250.565456-13-john.ogness@linutronix.de
Currently, a CONFIG_PREEMPT_NONE=y kernel substitutes normal RCU for
RCU Tasks Rude and RCU Tasks Trace. Unless that kernel builds rcuscale,
whether built-in or as a module, in which case these RCU Tasks flavors are
(unnecessarily) built in. This both increases kernel size and increases
the complexity of certain tracing operations. This commit therefore
decouples the presence of rcuscale from the presence of RCU Tasks Rude
and RCU Tasks Trace.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, a CONFIG_PREEMPT_NONE=y kernel substitutes normal RCU for
RCU Tasks. Unless that kernel builds rcuscale, whether built-in or as
a module, in which case RCU Tasks is (unnecessarily) built. This both
increases kernel size and increases the complexity of certain tracing
operations. This commit therefore decouples the presence of rcuscale
from the presence of RCU Tasks.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, a CONFIG_PREEMPT_NONE=y kernel substitutes normal RCU for
RCU Tasks Rude and RCU Tasks Trace. Unless that kernel builds refscale,
whether built-in or as a module, in which case these RCU Tasks flavors are
(unnecessarily) built in. This both increases kernel size and increases
the complexity of certain tracing operations. This commit therefore
decouples the presence of refscale from the presence of RCU Tasks Rude
and RCU Tasks Trace.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, a CONFIG_PREEMPT_NONE=y kernel substitutes normal RCU for
RCU Tasks. Unless that kernel builds refscale, whether built-in or as a
module, in which case RCU Tasks is (unnecessarily) built in. This both
increases kernel size and increases the complexity of certain tracing
operations. This commit therefore decouples the presence of refscale
from the presence of RCU Tasks.
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Unless a kernel builds rcutorture, whether built-in or as a module, that
kernel is also built with CONFIG_TASKS_RUDE_RCU, whether anything else
needs Tasks Rude RCU or not. This unnecessarily increases kernel size.
This commit therefore decouples the presence of rcutorture from the
presence of RCU Tasks Rude.
However, there is a need to select CONFIG_TASKS_RUDE_RCU for testing
purposes. Except that casual users must not be bothered with
questions -- for them, this needs to be fully automated. There is
thus a CONFIG_FORCE_TASKS_RUDE_RCU that selects CONFIG_TASKS_RUDE_RCU,
is user-selectable, but which depends on CONFIG_RCU_EXPERT.
[ paulmck: Apply kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, a CONFIG_PREEMPT_NONE=y kernel substitutes normal RCU for
RCU Tasks. Unless that kernel builds rcutorture, whether built-in or as
a module, in which case RCU Tasks is (unnecessarily) used. This both
increases kernel size and increases the complexity of certain tracing
operations. This commit therefore decouples the presence of rcutorture
from the presence of RCU Tasks.
However, there is a need to select CONFIG_TASKS_RCU for testing purposes.
Except that casual users must not be bothered with questions -- for them,
this needs to be fully automated. There is thus a CONFIG_FORCE_TASKS_RCU
that selects CONFIG_TASKS_RCU, is user-selectable, but which depends
on CONFIG_RCU_EXPERT.
[ paulmck: Apply kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Unless a kernel builds rcutorture, whether built-in or as a module, that
kernel is also built with CONFIG_TASKS_TRACE_RCU, whether anything else
needs Tasks Trace RCU or not. This unnecessarily increases kernel size.
This commit therefore decouples the presence of rcutorture from the
presence of RCU Tasks Trace.
However, there is a need to select CONFIG_TASKS_TRACE_RCU for
testing purposes. Except that casual users must not be bothered with
questions -- for them, this needs to be fully automated. There is thus
a CONFIG_FORCE_TASKS_TRACE_RCU that selects CONFIG_TASKS_TRACE_RCU,
is user-selectable, but which depends on CONFIG_RCU_EXPERT.
[ paulmck: Apply kernel test robot feedback. ]
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Currently, any kernel built with CONFIG_PREEMPTION=y also gets
CONFIG_TASKS_RCU=y, which is not helpful to people trying to build
preemptible kernels of minimal size.
Because CONFIG_TASKS_RCU=y is needed only in kernels doing tracing of
one form or another, this commit moves from TASKS_RCU deciding when it
should be enabled to the tracing Kconfig options explicitly selecting it.
This allows building preemptible kernels without TASKS_RCU, if desired.
This commit also updates the SRCU-N and TREE09 rcutorture scenarios
in order to avoid Kconfig errors that would otherwise result from
CONFIG_TASKS_RCU being selected without its CONFIG_RCU_EXPERT dependency
being met.
[ paulmck: Apply BPF_SYSCALL feedback from Andrii Nakryiko. ]
Reported-by: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Tested-by: Hyeonggon Yoo <42.hyeyoo@gmail.com>
Tested-by: Zhouyi Zhou <zhouzhouyi@gmail.com>
Cc: Andrii Nakryiko <andrii@kernel.org>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Acked-by: Masami Hiramatsu <mhiramat@kernel.org>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
When booting kernels built with both CONFIG_RCU_STRICT_GRACE_PERIOD=y
and CONFIG_PREEMPT_RT=y, the rcu_read_unlock_special() function's
invocation of irq_work_queue_on() the init_irq_work() causes the
rcu_preempt_deferred_qs_handler() function to work execute in SCHED_FIFO
irq_work kthreads. Because rcu_read_unlock_special() is invoked on each
rcu_read_unlock() in such kernels, the amount of work just keeps piling
up, resulting in a boot-time hang.
This commit therefore avoids this hang by using IRQ_WORK_INIT_HARD()
instead of init_irq_work(), but only in kernels built with both
CONFIG_PREEMPT_RT=y and CONFIG_RCU_STRICT_GRACE_PERIOD=y.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Linus Torvalds
Neeraj Upadhyay
Paul E. McKenney
Petr Mladek
Kalesh Singh
Uladzislau Rezki
Lukas Bulwahn
John Ogness
Zqiang