We have been facing some problems with self-suspending constrained
deadline tasks. The main reason is that the original CBS was not
designed for such sort of tasks.
One problem reported by Xunlei Pang takes place when a task
suspends, and then is awakened before the deadline, but so close
to the deadline that its remaining runtime can cause the task
to have an absolute density higher than allowed. In such situation,
the original CBS assumes that the task is facing an early activation,
and so it replenishes the task and set another deadline, one deadline
in the future. This rule works fine for implicit deadline tasks.
Moreover, it allows the system to adapt the period of a task in which
the external event source suffered from a clock drift.
However, this opens the window for bandwidth leakage for constrained
deadline tasks. For instance, a task with the following parameters:
runtime = 5 ms
deadline = 7 ms
[density] = 5 / 7 = 0.71
period = 1000 ms
If the task runs for 1 ms, and then suspends for another 1ms,
it will be awakened with the following parameters:
remaining runtime = 4
laxity = 5
presenting a absolute density of 4 / 5 = 0.80.
In this case, the original CBS would assume the task had an early
wakeup. Then, CBS will reset the runtime, and the absolute deadline will
be postponed by one relative deadline, allowing the task to run.
The problem is that, if the task runs this pattern forever, it will keep
receiving bandwidth, being able to run 1ms every 2ms. Following this
behavior, the task would be able to run 500 ms in 1 sec. Thus running
more than the 5 ms / 1 sec the admission control allowed it to run.
Trying to address the self-suspending case, Luca Abeni, Giuseppe
Lipari, and Juri Lelli [1] revisited the CBS in order to deal with
self-suspending tasks. In the new approach, rather than
replenishing/postponing the absolute deadline, the revised wakeup rule
adjusts the remaining runtime, reducing it to fit into the allowed
density.
A revised version of the idea is:
At a given time t, the maximum absolute density of a task cannot be
higher than its relative density, that is:
runtime / (deadline - t) <= dl_runtime / dl_deadline
Knowing the laxity of a task (deadline - t), it is possible to move
it to the other side of the equality, thus enabling to define max
remaining runtime a task can use within the absolute deadline, without
over-running the allowed density:
runtime = (dl_runtime / dl_deadline) * (deadline - t)
For instance, in our previous example, the task could still run:
runtime = ( 5 / 7 ) * 5
runtime = 3.57 ms
Without causing damage for other deadline tasks. It is note worthy
that the laxity cannot be negative because that would cause a negative
runtime. Thus, this patch depends on the patch:
df8eac8caf ("sched/deadline: Throttle a constrained deadline task activated after the deadline")
Which throttles a constrained deadline task activated after the
deadline.
Finally, it is also possible to use the revised wakeup rule for
all other tasks, but that would require some more discussions
about pros and cons.
Reported-by: Xunlei Pang <xpang@redhat.com>
Signed-off-by: Daniel Bristot de Oliveira <bristot@redhat.com>
[peterz: replaced dl_is_constrained with dl_is_implicit]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Juri Lelli <juri.lelli@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Luca Abeni <luca.abeni@santannapisa.it>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Romulo Silva de Oliveira <romulo.deoliveira@ufsc.br>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tommaso Cucinotta <tommaso.cucinotta@sssup.it>
Link: http://lkml.kernel.org/r/5c800ab3a74a168a84ee5f3f84d12a02e11383be.1495803804.git.bristot@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Instead of decreasing the runtime as "dq = -Uact dt" (eventually
divided by the maximum utilization available for deadline tasks),
decrease it as "dq = -max{u, (1 - Uinact)} dt", where u is the task
utilization and Uinact is the "inactive utilization".
In this way, the maximum fraction of CPU time that can be reclaimed
is given by the total utilization of deadline tasks.
This approach solves a fairness issue with "traditional" global GRUB
reclaiming: using the traditional GRUB algorithm, if tasks are
allocated to the various cores in a non-uniform way, the
reclaiming mechanism allows some tasks to reclaim more time than
others. This issue is visible starting 11 time-consuming tasks with
runtime 10ms and period 30ms (total utilization 3.666) on a 4-cores
system: some tasks will receive much more than the reserved runtime
(thanks to the reclaiming mechanism), while other tasks will receive
less than the reserved runtime.
Tested-by: Daniel Bristot de Oliveira <bristot@redhat.com>
Signed-off-by: Luca Abeni <luca.abeni@santannapisa.it>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Claudio Scordino <claudio@evidence.eu.com>
Cc: Joel Fernandes <joelaf@google.com>
Cc: Juri Lelli <juri.lelli@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mathieu Poirier <mathieu.poirier@linaro.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tommaso Cucinotta <tommaso.cucinotta@sssup.it>
Link: http://lkml.kernel.org/r/1495138417-6203-9-git-send-email-luca.abeni@santannapisa.it
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The newly introduced wrapper function only has one caller,
and this one is conditional, causing a harmless warning when
CONFIG_CPU_FREQ is disabled:
arch/x86/kernel/tsc.c:189:13: error: 'set_cyc2ns_scale' defined but not used [-Werror=unused-function]
My first idea was to move the wrapper inside of that #ifdef,
but on second thought it seemed nicer to remove it completely
again and rename __set_cyc2ns_scale back to set_cyc2ns_scale,
but leaving the extra argument.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 615cd03373 ("x86/tsc: Fix sched_clock() sync")
Link: http://lkml.kernel.org/r/20170517203949.2052220-1-arnd@arndb.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Some of the boot code in init_kernel_freeable() which runs before SMP
bringup assumes (rightfully) that it runs on the boot CPU and therefore can
use smp_processor_id() in preemptible context.
That works so far because the smp_processor_id() check starts to be
effective after smp bringup. That's just wrong. Starting with SMP bringup
and the ability to move threads around, smp_processor_id() in preemptible
context is broken.
Aside of that it does not make sense to allow init to run on all CPUs
before sched_smp_init() has been run.
Pin the init to the boot CPU so the existing code can continue to use
smp_processor_id() without triggering the checks when the enabling of those
checks starts earlier.
Tested-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20170516184734.943149935@linutronix.de
Signed-off-by: Ingo Molnar <mingo@kernel.org>
A customer has reported a soft-lockup when running an intensive
memory stress test, where the trace on multiple CPU's looks like this:
RIP: 0010:[<ffffffff810c53fe>]
[<ffffffff810c53fe>] native_queued_spin_lock_slowpath+0x10e/0x190
...
Call Trace:
[<ffffffff81182d07>] queued_spin_lock_slowpath+0x7/0xa
[<ffffffff811bc331>] change_protection_range+0x3b1/0x930
[<ffffffff811d4be8>] change_prot_numa+0x18/0x30
[<ffffffff810adefe>] task_numa_work+0x1fe/0x310
[<ffffffff81098322>] task_work_run+0x72/0x90
Further investigation showed that the lock contention here is pmd_lock().
The task_numa_work() function makes sure that only one thread is let to perform
the work in a single scan period (via cmpxchg), but if there's a thread with
mmap_sem locked for writing for several periods, multiple threads in
task_numa_work() can build up a convoy waiting for mmap_sem for read and then
all get unblocked at once.
This patch changes the down_read() to the trylock version, which prevents the
build up. For a workload experiencing mmap_sem contention, it's probably better
to postpone the NUMA balancing work anyway. This seems to have fixed the soft
lockups involving pmd_lock(), which is in line with the convoy theory.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20170515131316.21909-1-vbabka@suse.cz
Signed-off-by: Ingo Molnar <mingo@kernel.org>
When priority inheritance was added back in 2.6.18 to sched_setscheduler(), it
added a path to taking an rt-mutex wait_lock, which is not IRQ safe. As PI
is not a common occurrence, lockdep will likely never trigger if
sched_setscheduler was called from interrupt context. A BUG_ON() was added
to trigger if __sched_setscheduler() was ever called from interrupt context
because there was a possibility to take the wait_lock.
Today the wait_lock is irq safe, but the path to taking it in
sched_setscheduler() is the same as the path to taking it from normal
context. The wait_lock is taken with raw_spin_lock_irq() and released with
raw_spin_unlock_irq() which will indiscriminately enable interrupts,
which would be bad in interrupt context.
The problem is that normalize_rt_tasks, which is called by triggering the
sysrq nice-all-RT-tasks was changed to call __sched_setscheduler(), and this
is done from interrupt context!
Now __sched_setscheduler() takes a "pi" parameter that is used to know if
the priority inheritance should be called or not. As the BUG_ON() only cares
about calling the PI code, it should only bug if called from interrupt
context with the "pi" parameter set to true.
Reported-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
Tested-by: Laurent Dufour <ldufour@linux.vnet.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@osdl.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: dbc7f069b9 ("sched: Use replace normalize_task() with __sched_setscheduler()")
Link: http://lkml.kernel.org/r/20170308124654.10e598f2@gandalf.local.home
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Inter-Processor-Interrupt(IPI) is needed when a page is unmapped and the
process' mm_cpumask() shows the process has ever run on other CPUs. page
migration, page reclaim all need IPIs. The number of IPI needed to send
to different CPUs is especially large for multi-threaded workload since
mm_cpumask() is per process.
For smp_call_function_many(), whenever a CPU queues a CSD to a target
CPU, it will send an IPI to let the target CPU to handle the work.
This isn't necessary - we need only send IPI when queueing a CSD
to an empty call_single_queue.
The reason:
flush_smp_call_function_queue() that is called upon a CPU receiving an
IPI will empty the queue and then handle all of the CSDs there. So if
the target CPU's call_single_queue is not empty, we know that:
i. An IPI for the target CPU has already been sent by 'previous queuers';
ii. flush_smp_call_function_queue() hasn't emptied that CPU's queue yet.
Thus, it's safe for us to just queue our CSD there without sending an
addtional IPI. And for the 'previous queuers', we can limit it to the
first queuer.
To demonstrate the effect of this patch, a multi-thread workload that
spawns 80 threads to equally consume 100G memory is used. This is tested
on a 2 node broadwell-EP which has 44cores/88threads and 32G memory. So
after 32G memory is used up, page reclaiming starts to happen a lot.
With this patch, IPI number dropped 88% and throughput increased about
15% for the above workload.
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Link: http://lkml.kernel.org/r/20170519075331.GE2084@aaronlu.sh.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull crypto fix from Herbert Xu:
"This fixes a regression in the skcipher interface that allows bogus
key parameters to hit underlying implementations which can cause
crashes"
* 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6:
crypto: skcipher - Add missing API setkey checks
Pull pstore fix from Kees Cook:
"Marta noticed another misbehavior in EFI pstore, which this fixes.
Hopefully this is the last of the v4.12 fixes for pstore!"
* tag 'pstore-v4.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
efi-pstore: Fix write/erase id tracking
Pull ACPI fixes from Rafael Wysocki:
"These revert a 4.11 change that turned out to be problematic and add a
.gitignore file.
Specifics:
- Revert a 4.11 commit related to the ACPI-based handling of laptop
lids that made changes incompatible with existing user space stacks
and broke things there (Lv Zheng).
- Add .gitignore to the ACPI tools directory (Prarit Bhargava)"
* tag 'acpi-4.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
Revert "ACPI / button: Remove lid_init_state=method mode"
tools/power/acpi: Add .gitignore file
Pull power management fixes from Rafael Wysocki:
"These fix RTC wakeup from suspend-to-idle broken recently, fix CPU
idleness detection condition in the schedutil cpufreq governor, fix a
cpufreq driver build failure, fix an error code path in the power
capping framework, clean up the hibernate core and update the
intel_pstate documentation.
Specifics:
- Fix RTC wakeup from suspend-to-idle broken by the recent rework of
ACPI wakeup handling (Rafael Wysocki).
- Update intel_pstate driver documentation to reflect the current
code and explain how it works in more detail (Rafael Wysocki).
- Fix an issue related to CPU idleness detection on systems with
shared cpufreq policies in the schedutil governor (Juri Lelli).
- Fix a possible build issue in the dbx500 cpufreq driver (Arnd
Bergmann).
- Fix a function in the power capping framework core to return an
error code instead of 0 when there's an error (Dan Carpenter).
- Clean up variable definition in the hibernation core (Pushkar
Jambhlekar)"
* tag 'pm-4.12-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
cpufreq: dbx500: add a Kconfig symbol
PM / hibernate: Declare variables as static
PowerCap: Fix an error code in powercap_register_zone()
RTC: rtc-cmos: Fix wakeup from suspend-to-idle
PM / wakeup: Fix up wakeup_source_report_event()
cpufreq: intel_pstate: Document the current behavior and user interface
cpufreq: schedutil: use now as reference when aggregating shared policy requests
