We've been seeing hard-to-trigger psi crashes when running inside VM
instances:
divide error: 0000 [#1] SMP PTI
Modules linked in: [...]
CPU: 0 PID: 212 Comm: kworker/0:2 Not tainted 4.16.18-119_fbk9_3817_gfe944c98d695 #119
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 0.0.0 02/06/2015
Workqueue: events psi_clock
RIP: 0010:psi_update_stats+0x270/0x490
RSP: 0018:ffffc90001117e10 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: ffff8800a35a13f8
RDX: 0000000000000000 RSI: ffff8800a35a1340 RDI: 0000000000000000
RBP: 0000000000000658 R08: ffff8800a35a1470 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 00000000000f8502
FS: 0000000000000000(0000) GS:ffff88023fc00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fbe370fa000 CR3: 00000000b1e3a000 CR4: 00000000000006f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
psi_clock+0x12/0x50
process_one_work+0x1e0/0x390
worker_thread+0x2b/0x3c0
? rescuer_thread+0x330/0x330
kthread+0x113/0x130
? kthread_create_worker_on_cpu+0x40/0x40
? SyS_exit_group+0x10/0x10
ret_from_fork+0x35/0x40
Code: 48 0f 47 c7 48 01 c2 45 85 e4 48 89 16 0f 85 e6 00 00 00 4c 8b 49 10 4c 8b 51 08 49 69 d9 f2 07 00 00 48 6b c0 64 4c 8b 29 31 d2 <48> f7 f7 49 69 d5 8d 06 00 00 48 89 c5 4c 69 f0 00 98 0b 00 48
The Code-line points to `period` being 0 inside update_stats(), and we
divide by that when calculating that period's pressure percentage.
The elapsed period should never be 0. The reason this can happen is due
to an off-by-one in the idle time / missing period calculation combined
with a coarse sched_clock() in the virtual machine.
The target time for aggregation is advanced into the future on a fixed
grid to prevent clock drift. So when an aggregation runs after some idle
period, we can not just set it to "now + psi_period", but have to
calculate the downtime and advance the target time relative to itself.
However, if the aggregator was disabled exactly one psi_period (ns), we
drop one idle period in the calculation due to a > when we should do >=.
In that case, next_update will be advanced from 'now - psi_period' to
'now' when it should be moved to 'now + psi_period'. The run finishes
with last_update == next_update == sched_clock().
With hardware clocks, this exact nanosecond match isn't likely in the
first place; but if it does happen, the clock will still have moved on and
the period non-zero by the time the worker runs. A pointlessly short
period, but besides the extra work, no harm no foul. However, a slow
sched_clock() like we have on VMs might not have advanced either by the
time the worker runs again. And when we calculate the elapsed period, the
result, our pressure divisor, will be 0. Ouch.
Fix this by correctly handling the situation when the elapsed time between
aggregation runs is precisely two periods, and advance the expiration
timestamp correctly to period into the future.
Link: http://lkml.kernel.org/r/20190214193157.15788-1-hannes@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: Łukasz Siudut <lsiudut@fb.com
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull cpu hotplug fixes from Thomas Gleixner:
"Two fixes for the cpu hotplug machinery:
- Replace the overly clever 'SMT disabled by BIOS' detection logic as
it breaks KVM scenarios and prevents speculation control updates
when the Hyperthreads are brought online late after boot.
- Remove a redundant invocation of the speculation control update
function"
* 'smp-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
cpu/hotplug: Fix "SMT disabled by BIOS" detection for KVM
x86/speculation: Remove redundant arch_smt_update() invocation
psi has provisions to shut off the periodic aggregation worker when
there is a period of no task activity - and thus no data that needs
aggregating. However, while developing psi monitoring, Suren noticed
that the aggregation clock currently won't stay shut off for good.
Debugging this revealed a flaw in the idle design: an aggregation run
will see no task activity and decide to go to sleep; shortly thereafter,
the kworker thread that executed the aggregation will go idle and cause
a scheduling change, during which the psi callback will kick the
!pending worker again. This will ping-pong forever, and is equivalent
to having no shut-off logic at all (but with more code!)
Fix this by exempting aggregation workers from psi's clock waking logic
when the state change is them going to sleep. To do this, tag workers
with the last work function they executed, and if in psi we see a worker
going to sleep after aggregating psi data, we will not reschedule the
aggregation work item.
What if the worker is also executing other items before or after?
Any psi state times that were incurred by work items preceding the
aggregation work will have been collected from the per-cpu buckets
during the aggregation itself. If there are work items following the
aggregation work, the worker's last_func tag will be overwritten and the
aggregator will be kept alive to process this genuine new activity.
If the aggregation work is the last thing the worker does, and we decide
to go idle, the brief period of non-idle time incurred between the
aggregation run and the kworker's dequeue will be stranded in the
per-cpu buckets until the clock is woken by later activity. But that
should not be a problem. The buckets can hold 4s worth of time, and
future activity will wake the clock with a 2s delay, giving us 2s worth
of data we can leave behind when disabling aggregation. If it takes a
worker more than two seconds to go idle after it finishes its last work
item, we likely have bigger problems in the system, and won't notice one
sample that was averaged with a bogus per-CPU weight.
Link: http://lkml.kernel.org/r/20190116193501.1910-1-hannes@cmpxchg.org
Fixes: eb414681d5 ("psi: pressure stall information for CPU, memory, and IO")
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: Suren Baghdasaryan <surenb@google.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Lai Jiangshan <jiangshanlai@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
With the following commit:
73d5e2b472 ("cpu/hotplug: detect SMT disabled by BIOS")
... the hotplug code attempted to detect when SMT was disabled by BIOS,
in which case it reported SMT as permanently disabled. However, that
code broke a virt hotplug scenario, where the guest is booted with only
primary CPU threads, and a sibling is brought online later.
The problem is that there doesn't seem to be a way to reliably
distinguish between the HW "SMT disabled by BIOS" case and the virt
"sibling not yet brought online" case. So the above-mentioned commit
was a bit misguided, as it permanently disabled SMT for both cases,
preventing future virt sibling hotplugs.
Going back and reviewing the original problems which were attempted to
be solved by that commit, when SMT was disabled in BIOS:
1) /sys/devices/system/cpu/smt/control showed "on" instead of
"notsupported"; and
2) vmx_vm_init() was incorrectly showing the L1TF_MSG_SMT warning.
I'd propose that we instead consider #1 above to not actually be a
problem. Because, at least in the virt case, it's possible that SMT
wasn't disabled by BIOS and a sibling thread could be brought online
later. So it makes sense to just always default the smt control to "on"
to allow for that possibility (assuming cpuid indicates that the CPU
supports SMT).
The real problem is #2, which has a simple fix: change vmx_vm_init() to
query the actual current SMT state -- i.e., whether any siblings are
currently online -- instead of looking at the SMT "control" sysfs value.
So fix it by:
a) reverting the original "fix" and its followup fix:
73d5e2b472 ("cpu/hotplug: detect SMT disabled by BIOS")
bc2d8d262c ("cpu/hotplug: Fix SMT supported evaluation")
and
b) changing vmx_vm_init() to query the actual current SMT state --
instead of the sysfs control value -- to determine whether the L1TF
warning is needed. This also requires the 'sched_smt_present'
variable to exported, instead of 'cpu_smt_control'.
Fixes: 73d5e2b472 ("cpu/hotplug: detect SMT disabled by BIOS")
Reported-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Joe Mario <jmario@redhat.com>
Cc: Jiri Kosina <jikos@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: kvm@vger.kernel.org
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/e3a85d585da28cc333ecbc1e78ee9216e6da9396.1548794349.git.jpoimboe@redhat.com
Notable cmpxchg() does not provide ordering when it fails, however
wake_q_add() requires ordering in this specific case too. Without this
it would be possible for the concurrent wakeup to not observe our
prior state.
Andrea Parri provided:
C wake_up_q-wake_q_add
{
int next = 0;
int y = 0;
}
P0(int *next, int *y)
{
int r0;
/* in wake_up_q() */
WRITE_ONCE(*next, 1); /* node->next = NULL */
smp_mb(); /* implied by wake_up_process() */
r0 = READ_ONCE(*y);
}
P1(int *next, int *y)
{
int r1;
/* in wake_q_add() */
WRITE_ONCE(*y, 1); /* wake_cond = true */
smp_mb__before_atomic();
r1 = cmpxchg_relaxed(next, 1, 2);
}
exists (0:r0=0 /\ 1:r1=0)
This "exists" clause cannot be satisfied according to the LKMM:
Test wake_up_q-wake_q_add Allowed
States 3
0:r0=0; 1:r1=1;
0:r0=1; 1:r1=0;
0:r0=1; 1:r1=1;
No
Witnesses
Positive: 0 Negative: 3
Condition exists (0:r0=0 /\ 1:r1=0)
Observation wake_up_q-wake_q_add Never 0 3
Reported-by: Yongji Xie <elohimes@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Waiman Long <longman@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The only guarantee provided by wake_q_add() is that a wakeup will
happen after it, it does _NOT_ guarantee the wakeup will be delayed
until the matching wake_up_q().
If wake_q_add() fails the cmpxchg() a concurrent wakeup is pending and
that can happen at any time after the cmpxchg(). This means we should
not rely on the wakeup happening at wake_q_up(), but should be ready
for wake_q_add() to issue the wakeup.
The delay; if provided (most likely); should only result in more efficient
behaviour.
Reported-by: Yongji Xie <elohimes@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Waiman Long <longman@redhat.com>
Cc: Will Deacon <will.deacon@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Currently, CONFIG_JUMP_LABEL just means "I _want_ to use jump label".
The jump label is controlled by HAVE_JUMP_LABEL, which is defined
like this:
#if defined(CC_HAVE_ASM_GOTO) && defined(CONFIG_JUMP_LABEL)
# define HAVE_JUMP_LABEL
#endif
We can improve this by testing 'asm goto' support in Kconfig, then
make JUMP_LABEL depend on CC_HAS_ASM_GOTO.
Ugly #ifdef HAVE_JUMP_LABEL will go away, and CONFIG_JUMP_LABEL will
match to the real kernel capability.
Signed-off-by: Masahiro Yamada <yamada.masahiro@socionext.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
Tested-by: Sedat Dilek <sedat.dilek@gmail.com>
Merge more updates from Andrew Morton:
- procfs updates
- various misc bits
- lib/ updates
- epoll updates
- autofs
- fatfs
- a few more MM bits
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (58 commits)
mm/page_io.c: fix polled swap page in
checkpatch: add Co-developed-by to signature tags
docs: fix Co-Developed-by docs
drivers/base/platform.c: kmemleak ignore a known leak
fs: don't open code lru_to_page()
fs/: remove caller signal_pending branch predictions
mm/: remove caller signal_pending branch predictions
arch/arc/mm/fault.c: remove caller signal_pending_branch predictions
kernel/sched/: remove caller signal_pending branch predictions
kernel/locking/mutex.c: remove caller signal_pending branch predictions
mm: select HAVE_MOVE_PMD on x86 for faster mremap
mm: speed up mremap by 20x on large regions
mm: treewide: remove unused address argument from pte_alloc functions
initramfs: cleanup incomplete rootfs
scripts/gdb: fix lx-version string output
kernel/kcov.c: mark write_comp_data() as notrace
kernel/sysctl: add panic_print into sysctl
panic: add options to print system info when panic happens
bfs: extra sanity checking and static inode bitmap
exec: separate MM_ANONPAGES and RLIMIT_STACK accounting
...
Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument
of the user address range verification function since we got rid of the
old racy i386-only code to walk page tables by hand.
It existed because the original 80386 would not honor the write protect
bit when in kernel mode, so you had to do COW by hand before doing any
user access. But we haven't supported that in a long time, and these
days the 'type' argument is a purely historical artifact.
A discussion about extending 'user_access_begin()' to do the range
checking resulted this patch, because there is no way we're going to
move the old VERIFY_xyz interface to that model. And it's best done at
the end of the merge window when I've done most of my merges, so let's
just get this done once and for all.
This patch was mostly done with a sed-script, with manual fix-ups for
the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form.
There were a couple of notable cases:
- csky still had the old "verify_area()" name as an alias.
- the iter_iov code had magical hardcoded knowledge of the actual
values of VERIFY_{READ,WRITE} (not that they mattered, since nothing
really used it)
- microblaze used the type argument for a debug printout
but other than those oddities this should be a total no-op patch.
I tried to fix up all architectures, did fairly extensive grepping for
access_ok() uses, and the changes are trivial, but I may have missed
something. Any missed conversion should be trivially fixable, though.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Zhipeng Xie, Xie XiuQi and Sargun Dhillon reported lockups in the
scheduler under high loads, starting at around the v4.18 time frame,
and Zhipeng Xie tracked it down to bugs in the rq->leaf_cfs_rq_list
manipulation.
Do a (manual) revert of:
a9e7f6544b ("sched/fair: Fix O(nr_cgroups) in load balance path")
It turns out that the list_del_leaf_cfs_rq() introduced by this commit
is a surprising property that was not considered in followup commits
such as:
9c2791f936 ("sched/fair: Fix hierarchical order in rq->leaf_cfs_rq_list")
As Vincent Guittot explains:
"I think that there is a bigger problem with commit a9e7f6544b and
cfs_rq throttling:
Let take the example of the following topology TG2 --> TG1 --> root:
1) The 1st time a task is enqueued, we will add TG2 cfs_rq then TG1
cfs_rq to leaf_cfs_rq_list and we are sure to do the whole branch in
one path because it has never been used and can't be throttled so
tmp_alone_branch will point to leaf_cfs_rq_list at the end.
2) Then TG1 is throttled
3) and we add TG3 as a new child of TG1.
4) The 1st enqueue of a task on TG3 will add TG3 cfs_rq just before TG1
cfs_rq and tmp_alone_branch will stay on rq->leaf_cfs_rq_list.
With commit a9e7f6544b, we can del a cfs_rq from rq->leaf_cfs_rq_list.
So if the load of TG1 cfs_rq becomes NULL before step 2) above, TG1
cfs_rq is removed from the list.
Then at step 4), TG3 cfs_rq is added at the beginning of rq->leaf_cfs_rq_list
but tmp_alone_branch still points to TG3 cfs_rq because its throttled
parent can't be enqueued when the lock is released.
tmp_alone_branch doesn't point to rq->leaf_cfs_rq_list whereas it should.
So if TG3 cfs_rq is removed or destroyed before tmp_alone_branch
points on another TG cfs_rq, the next TG cfs_rq that will be added,
will be linked outside rq->leaf_cfs_rq_list - which is bad.
In addition, we can break the ordering of the cfs_rq in
rq->leaf_cfs_rq_list but this ordering is used to update and
propagate the update from leaf down to root."
Instead of trying to work through all these cases and trying to reproduce
the very high loads that produced the lockup to begin with, simplify
the code temporarily by reverting a9e7f6544b - which change was clearly
not thought through completely.
This (hopefully) gives us a kernel that doesn't lock up so people
can continue to enjoy their holidays without worrying about regressions. ;-)
[ mingo: Wrote changelog, fixed weird spelling in code comment while at it. ]
Analyzed-by: Xie XiuQi <xiexiuqi@huawei.com>
Analyzed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reported-by: Zhipeng Xie <xiezhipeng1@huawei.com>
Reported-by: Sargun Dhillon <sargun@sargun.me>
Reported-by: Xie XiuQi <xiexiuqi@huawei.com>
Tested-by: Zhipeng Xie <xiezhipeng1@huawei.com>
Tested-by: Sargun Dhillon <sargun@sargun.me>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Acked-by: Vincent Guittot <vincent.guittot@linaro.org>
Cc: <stable@vger.kernel.org> # v4.13+
Cc: Bin Li <huawei.libin@huawei.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: a9e7f6544b ("sched/fair: Fix O(nr_cgroups) in load balance path")
Link: http://lkml.kernel.org/r/1545879866-27809-1-git-send-email-xiexiuqi@huawei.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Caused by making the variable static:
kernel/sched/fair.c:119:21: warning: 'capacity_margin' defined but not used [-Wunused-variable]
Seems easiest to just move it up under the existing ifdef CONFIG_SMP
that's a few lines above.
Fixes: ed8885a144 ('sched/fair: Make some variables static')
Signed-off-by: Olof Johansson <olof@lixom.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull scheduler updates from Ingo Molnar:
"The main changes in this cycle were:
- Introduce "Energy Aware Scheduling" - by Quentin Perret.
This is a coherent topology description of CPUs in cooperation with
the PM subsystem, with the goal to schedule more energy-efficiently
on asymetric SMP platform - such as waking up tasks to the more
energy-efficient CPUs first, as long as the system isn't
oversubscribed.
For details of the design, see:
https://lore.kernel.org/lkml/20180724122521.22109-1-quentin.perret@arm.com/
- Misc cleanups and smaller enhancements"
* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits)
sched/fair: Select an energy-efficient CPU on task wake-up
sched/fair: Introduce an energy estimation helper function
sched/fair: Add over-utilization/tipping point indicator
sched/fair: Clean-up update_sg_lb_stats parameters
sched/toplogy: Introduce the 'sched_energy_present' static key
sched/topology: Make Energy Aware Scheduling depend on schedutil
sched/topology: Disable EAS on inappropriate platforms
sched/topology: Add lowest CPU asymmetry sched_domain level pointer
sched/topology: Reference the Energy Model of CPUs when available
PM: Introduce an Energy Model management framework
sched/cpufreq: Prepare schedutil for Energy Aware Scheduling
sched/topology: Relocate arch_scale_cpu_capacity() to the internal header
sched/core: Remove unnecessary unlikely() in push_*_task()
sched/topology: Remove the ::smt_gain field from 'struct sched_domain'
sched: Fix various typos in comments
sched/core: Clean up the #ifdef block in add_nr_running()
sched/fair: Make some variables static
sched/core: Create task_has_idle_policy() helper
sched/fair: Add lsub_positive() and use it consistently
sched/fair: Mask UTIL_AVG_UNCHANGED usages
...
Pull locking updates from Ingo Molnar:
"The main change in this cycle are initial preparatory bits of dynamic
lockdep keys support from Bart Van Assche.
There are also misc changes, a comment cleanup and a data structure
cleanup"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Clean up comment in nohz_idle_balance()
locking/lockdep: Stop using RCU primitives to access 'all_lock_classes'
locking/lockdep: Make concurrent lockdep_reset_lock() calls safe
locking/lockdep: Remove a superfluous INIT_LIST_HEAD() statement
locking/lockdep: Introduce lock_class_cache_is_registered()
locking/lockdep: Inline __lockdep_init_map()
locking/lockdep: Declare local symbols static
tools/lib/lockdep/tests: Test the lockdep_reset_lock() implementation
tools/lib/lockdep: Add dummy print_irqtrace_events() implementation
tools/lib/lockdep: Rename "trywlock" into "trywrlock"
tools/lib/lockdep/tests: Run lockdep tests a second time under Valgrind
tools/lib/lockdep/tests: Improve testing accuracy
tools/lib/lockdep/tests: Fix shellcheck warnings
tools/lib/lockdep/tests: Display compiler warning and error messages
locking/lockdep: Remove ::version from lock_class structure
Pull RCU updates from Ingo Molnar:
"The biggest RCU changes in this cycle were:
- Convert RCU's BUG_ON() and similar calls to WARN_ON() and similar.
- Replace calls of RCU-bh and RCU-sched update-side functions to
their vanilla RCU counterparts. This series is a step towards
complete removal of the RCU-bh and RCU-sched update-side functions.
( Note that some of these conversions are going upstream via their
respective maintainers. )
- Documentation updates, including a number of flavor-consolidation
updates from Joel Fernandes.
- Miscellaneous fixes.
- Automate generation of the initrd filesystem used for rcutorture
testing.
- Convert spin_is_locked() assertions to instead use lockdep.
( Note that some of these conversions are going upstream via their
respective maintainers. )
- SRCU updates, especially including a fix from Dennis Krein for a
bag-on-head-class bug.
- RCU torture-test updates"
* 'core-rcu-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (112 commits)
rcutorture: Don't do busted forward-progress testing
rcutorture: Use 100ms buckets for forward-progress callback histograms
rcutorture: Recover from OOM during forward-progress tests
rcutorture: Print forward-progress test age upon failure
rcutorture: Print time since GP end upon forward-progress failure
rcutorture: Print histogram of CB invocation at OOM time
rcutorture: Print GP age upon forward-progress failure
rcu: Print per-CPU callback counts for forward-progress failures
rcu: Account for nocb-CPU callback counts in RCU CPU stall warnings
rcutorture: Dump grace-period diagnostics upon forward-progress OOM
rcutorture: Prepare for asynchronous access to rcu_fwd_startat
torture: Remove unnecessary "ret" variables
rcutorture: Affinity forward-progress test to avoid housekeeping CPUs
rcutorture: Break up too-long rcu_torture_fwd_prog() function
rcutorture: Remove cbflood facility
torture: Bring any extra CPUs online during kernel startup
rcutorture: Add call_rcu() flooding forward-progress tests
rcutorture/formal: Replace synchronize_sched() with synchronize_rcu()
tools/kernel.h: Replace synchronize_sched() with synchronize_rcu()
net/decnet: Replace rcu_barrier_bh() with rcu_barrier()
...
Energy-aware scheduling is only meant to be active while the system is
_not_ over-utilized. That is, there are spare cycles available to shift
tasks around based on their actual utilization to get a more
energy-efficient task distribution without depriving any tasks. When
above the tipping point task placement is done the traditional way based
on load_avg, spreading the tasks across as many cpus as possible based
on priority scaled load to preserve smp_nice. Below the tipping point we
want to use util_avg instead. We need to define a criteria for when we
make the switch.
The util_avg for each cpu converges towards 100% regardless of how many
additional tasks we may put on it. If we define over-utilized as:
sum_{cpus}(rq.cfs.avg.util_avg) + margin > sum_{cpus}(rq.capacity)
some individual cpus may be over-utilized running multiple tasks even
when the above condition is false. That should be okay as long as we try
to spread the tasks out to avoid per-cpu over-utilization as much as
possible and if all tasks have the _same_ priority. If the latter isn't
true, we have to consider priority to preserve smp_nice.
For example, we could have n_cpus nice=-10 util_avg=55% tasks and
n_cpus/2 nice=0 util_avg=60% tasks. Balancing based on util_avg we are
likely to end up with nice=-10 tasks sharing cpus and nice=0 tasks
getting their own as we 1.5*n_cpus tasks in total and 55%+55% is less
over-utilized than 55%+60% for those cpus that have to be shared. The
system utilization is only 85% of the system capacity, but we are
breaking smp_nice.
To be sure not to break smp_nice, we have defined over-utilization
conservatively as when any cpu in the system is fully utilized at its
highest frequency instead:
cpu_rq(any).cfs.avg.util_avg + margin > cpu_rq(any).capacity
IOW, as soon as one cpu is (nearly) 100% utilized, we switch to load_avg
to factor in priority to preserve smp_nice.
With this definition, we can skip periodic load-balance as no cpu has an
always-running task when the system is not over-utilized. All tasks will
be periodic and we can balance them at wake-up. This conservative
condition does however mean that some scenarios that could benefit from
energy-aware decisions even if one cpu is fully utilized would not get
those benefits.
For systems where some cpus might have reduced capacity on some cpus
(RT-pressure and/or big.LITTLE), we want periodic load-balance checks as
soon a just a single cpu is fully utilized as it might one of those with
reduced capacity and in that case we want to migrate it.
[ peterz: Added a comment explaining why new tasks are not accounted during
overutilization detection. ]
Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com>
Signed-off-by: Quentin Perret <quentin.perret@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: adharmap@codeaurora.org
Cc: chris.redpath@arm.com
Cc: currojerez@riseup.net
Cc: dietmar.eggemann@arm.com
Cc: edubezval@gmail.com
Cc: gregkh@linuxfoundation.org
Cc: javi.merino@kernel.org
Cc: joel@joelfernandes.org
Cc: juri.lelli@redhat.com
Cc: patrick.bellasi@arm.com
Cc: pkondeti@codeaurora.org
Cc: rjw@rjwysocki.net
Cc: skannan@codeaurora.org
Cc: smuckle@google.com
Cc: srinivas.pandruvada@linux.intel.com
Cc: thara.gopinath@linaro.org
Cc: tkjos@google.com
Cc: valentin.schneider@arm.com
Cc: vincent.guittot@linaro.org
Cc: viresh.kumar@linaro.org
Link: https://lkml.kernel.org/r/20181203095628.11858-13-quentin.perret@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
::smt_gain is used to compute the capacity of CPUs of a SMT core with the
constraint 1 < ::smt_gain < 2 in order to be able to compute number of CPUs
per core. The field has_free_capacity of struct numa_stat, which was the
last user of this computation of number of CPUs per core, has been removed
by:
2d4056fafa ("sched/numa: Remove numa_has_capacity()")
We can now remove this constraint on core capacity and use the defautl value
SCHED_CAPACITY_SCALE for SMT CPUs. With this remove, SCHED_CAPACITY_SCALE
becomes the maximum compute capacity of CPUs on every systems. This should
help to simplify some code and remove fields like rd->max_cpu_capacity
Furthermore, arch_scale_cpu_capacity() is used with a NULL sd in several other
places in the code when it wants the capacity of a CPUs to scale
some metrics like in pelt, deadline or schedutil. In case on SMT, the value
returned is not the capacity of SMT CPUs but default SCHED_CAPACITY_SCALE.
So remove it.
Signed-off-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1535548752-4434-4-git-send-email-vincent.guittot@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull RCU changes from Paul E. McKenney:
- Convert RCU's BUG_ON() and similar calls to WARN_ON() and similar.
- Replace calls of RCU-bh and RCU-sched update-side functions
to their vanilla RCU counterparts. This series is a step
towards complete removal of the RCU-bh and RCU-sched update-side
functions.
( Note that some of these conversions are going upstream via their
respective maintainers. )
- Documentation updates, including a number of flavor-consolidation
updates from Joel Fernandes.
- Miscellaneous fixes.
- Automate generation of the initrd filesystem used for
rcutorture testing.
- Convert spin_is_locked() assertions to instead use lockdep.
( Note that some of these conversions are going upstream via their
respective maintainers. )
- SRCU updates, especially including a fix from Dennis Krein
for a bag-on-head-class bug.
- RCU torture-test updates.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull STIBP fallout fixes from Thomas Gleixner:
"The performance destruction department finally got it's act together
and came up with a cure for the STIPB regression:
- Provide a command line option to control the spectre v2 user space
mitigations. Default is either seccomp or prctl (if seccomp is
disabled in Kconfig). prctl allows mitigation opt-in, seccomp
enables the migitation for sandboxed processes.
- Rework the code to handle the conditional STIBP/IBPB control and
remove the now unused ptrace_may_access_sched() optimization
attempt
- Disable STIBP automatically when SMT is disabled
- Optimize the switch_to() logic to avoid MSR writes and invocations
of __switch_to_xtra().
- Make the asynchronous speculation TIF updates synchronous to
prevent stale mitigation state.
As a general cleanup this also makes retpoline directly depend on
compiler support and removes the 'minimal retpoline' option which just
pretended to provide some form of security while providing none"
* 'x86-pti-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (31 commits)
x86/speculation: Provide IBPB always command line options
x86/speculation: Add seccomp Spectre v2 user space protection mode
x86/speculation: Enable prctl mode for spectre_v2_user
x86/speculation: Add prctl() control for indirect branch speculation
x86/speculation: Prepare arch_smt_update() for PRCTL mode
x86/speculation: Prevent stale SPEC_CTRL msr content
x86/speculation: Split out TIF update
ptrace: Remove unused ptrace_may_access_sched() and MODE_IBRS
x86/speculation: Prepare for conditional IBPB in switch_mm()
x86/speculation: Avoid __switch_to_xtra() calls
x86/process: Consolidate and simplify switch_to_xtra() code
x86/speculation: Prepare for per task indirect branch speculation control
x86/speculation: Add command line control for indirect branch speculation
x86/speculation: Unify conditional spectre v2 print functions
x86/speculataion: Mark command line parser data __initdata
x86/speculation: Mark string arrays const correctly
x86/speculation: Reorder the spec_v2 code
x86/l1tf: Show actual SMT state
x86/speculation: Rework SMT state change
sched/smt: Expose sched_smt_present static key
...
Mel Gorman reports a hackbench regression with psi that would prohibit
shipping the suse kernel with it default-enabled, but he'd still like
users to be able to opt in at little to no cost to others.
With the current combination of CONFIG_PSI and the psi_disabled bool set
from the commandline, this is a challenge. Do the following things to
make it easier:
1. Add a config option CONFIG_PSI_DEFAULT_DISABLED that allows distros
to enable CONFIG_PSI in their kernel but leave the feature disabled
unless a user requests it at boot-time.
To avoid double negatives, rename psi_disabled= to psi=.
2. Make psi_disabled a static branch to eliminate any branch costs
when the feature is disabled.
In terms of numbers before and after this patch, Mel says:
: The following is a comparision using CONFIG_PSI=n as a baseline against
: your patch and a vanilla kernel
:
: 4.20.0-rc4 4.20.0-rc4 4.20.0-rc4
: kconfigdisable-v1r1 vanilla psidisable-v1r1
: Amean 1 1.3100 ( 0.00%) 1.3923 ( -6.28%) 1.3427 ( -2.49%)
: Amean 3 3.8860 ( 0.00%) 4.1230 * -6.10%* 3.8860 ( -0.00%)
: Amean 5 6.8847 ( 0.00%) 8.0390 * -16.77%* 6.7727 ( 1.63%)
: Amean 7 9.9310 ( 0.00%) 10.8367 * -9.12%* 9.9910 ( -0.60%)
: Amean 12 16.6577 ( 0.00%) 18.2363 * -9.48%* 17.1083 ( -2.71%)
: Amean 18 26.5133 ( 0.00%) 27.8833 * -5.17%* 25.7663 ( 2.82%)
: Amean 24 34.3003 ( 0.00%) 34.6830 ( -1.12%) 32.0450 ( 6.58%)
: Amean 30 40.0063 ( 0.00%) 40.5800 ( -1.43%) 41.5087 ( -3.76%)
: Amean 32 40.1407 ( 0.00%) 41.2273 ( -2.71%) 39.9417 ( 0.50%)
:
: It's showing that the vanilla kernel takes a hit (as the bisection
: indicated it would) and that disabling PSI by default is reasonably
: close in terms of performance for this particular workload on this
: particular machine so;
Link: http://lkml.kernel.org/r/20181127165329.GA29728@cmpxchg.org
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Tested-by: Mel Gorman <mgorman@techsingularity.net>
Reported-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now that synchronize_rcu() waits for preempt-disable regions of code
as well as RCU read-side critical sections, synchronize_sched() can be
replaced by synchronize_rcu(). This commit therefore makes this change.
Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Merge misc fixes from Andrew Morton:
"16 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
mm/memblock.c: fix a typo in __next_mem_pfn_range() comments
mm, page_alloc: check for max order in hot path
scripts/spdxcheck.py: make python3 compliant
tmpfs: make lseek(SEEK_DATA/SEK_HOLE) return ENXIO with a negative offset
lib/ubsan.c: don't mark __ubsan_handle_builtin_unreachable as noreturn
mm/vmstat.c: fix NUMA statistics updates
mm/gup.c: fix follow_page_mask() kerneldoc comment
ocfs2: free up write context when direct IO failed
scripts/faddr2line: fix location of start_kernel in comment
mm: don't reclaim inodes with many attached pages
mm, memory_hotplug: check zone_movable in has_unmovable_pages
mm/swapfile.c: use kvzalloc for swap_info_struct allocation
MAINTAINERS: update OMAP MMC entry
hugetlbfs: fix kernel BUG at fs/hugetlbfs/inode.c:444!
kernel/sched/psi.c: simplify cgroup_move_task()
z3fold: fix possible reclaim races
Pull scheduler fix from Ingo Molnar:
"Fix an exec() related scalability/performance regression, which was
caused by incorrectly calculating load and migrating tasks on exec()
when they shouldn't be"
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Fix cpu_util_wake() for 'execl' type workloads
A ~10% regression has been reported for UnixBench's execl throughput
test by Aaron Lu and Ye Xiaolong:
https://lkml.org/lkml/2018/10/30/765
That test is pretty simple, it does a "recursive" execve() syscall on the
same binary. Starting from the syscall, this sequence is possible:
do_execve()
do_execveat_common()
__do_execve_file()
sched_exec()
select_task_rq_fair() <==| Task already enqueued
find_idlest_cpu()
find_idlest_group()
capacity_spare_wake() <==| Functions not called from
cpu_util_wake() | the wakeup path
which means we can end up calling cpu_util_wake() not only from the
"wakeup path", as its name would suggest. Indeed, the task doing an
execve() syscall is already enqueued on the CPU we want to get the
cpu_util_wake() for.
The estimated utilization for a CPU computed in cpu_util_wake() was
written under the assumption that function can be called only from the
wakeup path. If instead the task is already enqueued, we end up with a
utilization which does not remove the current task's contribution from
the estimated utilization of the CPU.
This will wrongly assume a reduced spare capacity on the current CPU and
increase the chances to migrate the task on execve.
The regression is tracked down to:
commit d519329f72 ("sched/fair: Update util_est only on util_avg updates")
because in that patch we turn on by default the UTIL_EST sched feature.
However, the real issue is introduced by:
commit f9be3e5961 ("sched/fair: Use util_est in LB and WU paths")
Let's fix this by ensuring to always discount the task estimated
utilization from the CPU's estimated utilization when the task is also
the current one. The same benchmark of the bug report, executed on a
dual socket 40 CPUs Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz machine,
reports these "Execl Throughput" figures (higher the better):
mainline : 48136.5 lps
mainline+fix : 55376.5 lps
which correspond to a 15% speedup.
Moreover, since {cpu_util,capacity_spare}_wake() are not really only
used from the wakeup path, let's remove this ambiguity by using a better
matching name: {cpu_util,capacity_spare}_without().
Since we are at that, let's also improve the existing documentation.
Reported-by: Aaron Lu <aaron.lu@intel.com>
Reported-by: Ye Xiaolong <xiaolong.ye@intel.com>
Tested-by: Aaron Lu <aaron.lu@intel.com>
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Dietmar Eggemann <dietmar.eggemann@arm.com>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Morten Rasmussen <morten.rasmussen@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Quentin Perret <quentin.perret@arm.com>
Cc: Steve Muckle <smuckle@google.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Todd Kjos <tkjos@google.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Fixes: f9be3e5961 (sched/fair: Use util_est in LB and WU paths)
Link: https://lore.kernel.org/lkml/20181025093100.GB13236@e110439-lin/
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull scheduler fixes from Thomas Gleixner:
"Two small scheduler fixes:
- Take hotplug lock in sched_init_smp(). Technically not really
required, but lockdep will complain other.
- Trivial comment fix in sched/fair"
* 'sched/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
sched/fair: Fix a comment in task_numa_fault()
sched/core: Take the hotplug lock in sched_init_smp()
Now that synchronize_rcu() waits for preempt-disable regions of code
as well as RCU read-side critical sections, the synchronize_sched()
in sys_membarrier() can be replaced by synchronize_rcu(). This commit
therefore makes this change.
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Johannes Weiner