Preemption is disabled in mark_wakeup_next_waiter(,) not in
rt_mutex_slowunlock().
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153944.341734608@linutronix.de
The indirection via a function pointer (which is at least optimized into a
tail call by the compiler) is making the code hard to read.
Clean it up and move the futex related trylock functions down to the futex
section.
Move the wake_q wakeup into rt_mutex_slowunlock(). No point in handing it
to the caller. The futex code uses a different function.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153944.247927548@linutronix.de
rtmutex is half __sched and the other half is not. If the compiler decides
to not inline larger static functions then part of the code ends up in the
regular text section.
There are also quite some performance related small helpers which are
either static or plain inline. Force inline those which make sense and mark
the rest __sched.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153944.152977820@linutronix.de
There is no value in having two header files providing just empty stubs and
a C file which implements trivial debug functions which can just be inlined.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153944.052454464@linutronix.de
The conditional debug handling is just another layer of obfuscation. Split
the function so rt_mutex_init_proxy_locked() can invoke the inner init and
__rt_mutex_init() gets the full treatment.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153943.955697588@linutronix.de
None of these functions are used when CONFIG_RT_MUTEXES=n.
Remove the gunk. Remove pointless comments and clean up the coding style
mess while at it.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153943.863379182@linutronix.de
There is no point for this wrapper at all.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153943.754254046@linutronix.de
No users or useless and therefore just ballast.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153943.549192485@linutronix.de
The rtmutex specific deadlock detector predates lockdep coverage of rtmutex
and since commit f5694788ad ("rt_mutex: Add lockdep annotations") it
contains a lot of redundant functionality:
- lockdep will detect an potential deadlock before rtmutex-debug
has a chance to do so
- the deadlock debugging is restricted to rtmutexes which are not
associated to futexes and have an active waiter, which is covered by
lockdep already
Remove the redundant functionality and move actual deadlock WARN() into the
deadlock code path. The latter needs a seperate cleanup.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153943.320398604@linutronix.de
The following debug members of 'struct rtmutex' are unused:
- save_state: No users
- file,line: Printed if ::name is NULL. This is only used for non-futex
locks so ::name is never NULL
- magic: Assigned to NULL by rt_mutex_destroy(), no further usage
Remove them along with unused inline and macro leftovers related to
the long gone deadlock tester.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153943.195064296@linutronix.de
rt_mutex_timed_lock() has no callers since:
c051b21f71 ("rtmutex: Confine deadlock logic to futex")
Remove it.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210326153943.061103415@linutronix.de
The ww_acquire_ctx structure for ww_mutex needs to persist for a complete
lock/unlock cycle. In the ww_mutex test in locktorture, however, both
ww_acquire_init() and ww_acquire_fini() are called within the lock
function only. This causes a lockdep splat of "WARNING: Nested lock
was not taken" when lockdep is enabled in the kernel.
To fix this problem, we need to move the ww_acquire_fini() after
the ww_mutex_unlock() in torture_ww_mutex_unlock(). This is done by
allocating a global array of ww_acquire_ctx structures. Each locking
thread is associated with its own ww_acquire_ctx via the unique thread
id it has so that both the lock and unlock functions can access the
same ww_acquire_ctx structure.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20210318172814.4400-6-longman@redhat.com
To allow the lock and unlock functions in locktorture to access
per-thread information, we need to pass some hint on how to locate
those information. One way to do this is to pass in a unique thread
id which can then be used to access a global array for thread specific
information.
Change the lock and unlock method to add a thread id parameter which
can be determined by the offset of the lwsp/lrsp pointer from the global
lwsa/lrsa array.
There is no other functional change in this patch.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20210318172814.4400-5-longman@redhat.com
In order to avoid false positive circular locking lockdep splat
when runnng the ww_mutex torture test, we need to make sure that
the ww_mutexes have the same lock class as the acquire_ctx. This
means the ww_mutexes must have the same lockdep key as the
acquire_ctx. Unfortunately the current DEFINE_WW_MUTEX() macro fails
to do that. As a result, we add an init method for the ww_mutex test
to do explicit ww_mutex_init()'s of the ww_mutexes to avoid the false
positive warning.
Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20210318172814.4400-3-longman@redhat.com
The use_ww_ctx flag is passed to mutex_optimistic_spin(), but the
function doesn't use it. The frequent use of the (use_ww_ctx && ww_ctx)
combination is repetitive.
In fact, ww_ctx should not be used at all if !use_ww_ctx. Simplify
ww_mutex code by dropping use_ww_ctx from mutex_optimistic_spin() an
clear ww_ctx if !use_ww_ctx. In this way, we can replace (use_ww_ctx &&
ww_ctx) by just (ww_ctx).
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Davidlohr Bueso <dbueso@suse.de>
Link: https://lore.kernel.org/r/20210316153119.13802-2-longman@redhat.com
Update wake_futex_pi() and kill the call altogether. This is possible because:
(i) The case of fixup_owner() in which the pi_mutex was stolen from the
signaled enqueued top-waiter which fails to trylock and doesn't see a
current owner of the rtmutex but needs to acknowledge an non-enqueued
higher priority waiter, which is the other alternative. This used to be
handled by rt_mutex_next_owner(), which guaranteed fixup_pi_state_owner('newowner')
never to be nil. Nowadays the logic is handled by an EAGAIN loop, without
the need of rt_mutex_next_owner(). Specifically:
c1e2f0eaf0 (futex: Avoid violating the 10th rule of futex)
9f5d1c336a (futex: Handle transient "ownerless" rtmutex state correctly)
(ii) rt_mutex_next_owner() and rt_mutex_top_waiter() are semantically
equivalent, as of:
c28d62cf52 (locking/rtmutex: Handle non enqueued waiters gracefully in remove_waiter())
So instead of keeping the call around, just use the good ole rt_mutex_top_waiter().
No change in semantics.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20210226175029.50335-1-dave@stgolabs.net
Adds defines for lock state returns from lock_is_held_type() based on
Johannes Berg's suggestions as it make it easier to read and maintain
the lock states. These are defines and a enum to avoid changes to
lock_is_held_type() and lockdep_is_held() return types.
Updates to lock_is_held_type() and __lock_is_held() to use the new
defines.
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/linux-wireless/871rdmu9z9.fsf@codeaurora.org/
Some kernel functions must be called without holding a specific lock.
Add lockdep_assert_not_held() to be used in these functions to detect
incorrect calls while holding a lock.
lockdep_assert_not_held() provides the opposite functionality of
lockdep_assert_held() which is used to assert calls that require
holding a specific lock.
Incorporates suggestions from Peter Zijlstra to avoid misfires when
lockdep_off() is employed.
The need for lockdep_assert_not_held() came up in a discussion on
ath10k patch. ath10k_drain_tx() and i915_vma_pin_ww() are examples
of functions that can use lockdep_assert_not_held().
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/linux-wireless/871rdmu9z9.fsf@codeaurora.org/
Drop repeated words in kernel/events/.
{if, the, that, with, time}
Drop repeated words in kernel/locking/.
{it, no, the}
Drop repeated words in kernel/sched/.
{in, not}
Link: https://lkml.kernel.org/r/20210127023412.26292-1-rdunlap@infradead.org
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Will Deacon <will@kernel.org> [kernel/locking/]
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Will Deacon <will@kernel.org>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: "Paul E. McKenney" <paulmck@kernel.org>
Cc: Juri Lelli <juri.lelli@redhat.com>
Cc: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
- Support for userspace to emulate Xen hypercalls
- Raise the maximum number of user memslots
- Scalability improvements for the new MMU. Instead of the complex
"fast page fault" logic that is used in mmu.c, tdp_mmu.c uses an
rwlock so that page faults are concurrent, but the code that can run
against page faults is limited. Right now only page faults take the
lock for reading; in the future this will be extended to some
cases of page table destruction. I hope to switch the default MMU
around 5.12-rc3 (some testing was delayed due to Chinese New Year).
- Cleanups for MAXPHYADDR checks
- Use static calls for vendor-specific callbacks
- On AMD, use VMLOAD/VMSAVE to save and restore host state
- Stop using deprecated jump label APIs
- Workaround for AMD erratum that made nested virtualization unreliable
- Support for LBR emulation in the guest
- Support for communicating bus lock vmexits to userspace
- Add support for SEV attestation command
- Miscellaneous cleanups
PPC:
- Support for second data watchpoint on POWER10
- Remove some complex workarounds for buggy early versions of POWER9
- Guest entry/exit fixes
ARM64
- Make the nVHE EL2 object relocatable
- Cleanups for concurrent translation faults hitting the same page
- Support for the standard TRNG hypervisor call
- A bunch of small PMU/Debug fixes
- Simplification of the early init hypercall handling
Non-KVM changes (with acks):
- Detection of contended rwlocks (implemented only for qrwlocks,
because KVM only needs it for x86)
- Allow __DISABLE_EXPORTS from assembly code
- Provide a saner follow_pfn replacements for modules
-----BEGIN PGP SIGNATURE-----
iQFIBAABCAAyFiEE8TM4V0tmI4mGbHaCv/vSX3jHroMFAmApSRgUHHBib256aW5p
QHJlZGhhdC5jb20ACgkQv/vSX3jHroOc7wf9FnlinKoTFaSk7oeuuhF/CoCVwSFs
Z9+A2sNI99tWHQxFR6dyDkEFeQoXnqSxfLHtUVIdH/JnTg0FkEvFz3NK+0PzY1PF
PnGNbSoyhP58mSBG4gbBAxdF3ZJZMB8GBgYPeR62PvMX2dYbcHqVBNhlf6W4MQK4
5mAUuAnbf19O5N267sND+sIg3wwJYwOZpRZB7PlwvfKAGKf18gdBz5dQ/6Ej+apf
P7GODZITjqM5Iho7SDm/sYJlZprFZT81KqffwJQHWFMEcxFgwzrnYPx7J3gFwRTR
eeh9E61eCBDyCTPpHROLuNTVBqrAioCqXLdKOtO5gKvZI3zmomvAsZ8uXQ==
=uFZU
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM updates from Paolo Bonzini:
"x86:
- Support for userspace to emulate Xen hypercalls
- Raise the maximum number of user memslots
- Scalability improvements for the new MMU.
Instead of the complex "fast page fault" logic that is used in
mmu.c, tdp_mmu.c uses an rwlock so that page faults are concurrent,
but the code that can run against page faults is limited. Right now
only page faults take the lock for reading; in the future this will
be extended to some cases of page table destruction. I hope to
switch the default MMU around 5.12-rc3 (some testing was delayed
due to Chinese New Year).
- Cleanups for MAXPHYADDR checks
- Use static calls for vendor-specific callbacks
- On AMD, use VMLOAD/VMSAVE to save and restore host state
- Stop using deprecated jump label APIs
- Workaround for AMD erratum that made nested virtualization
unreliable
- Support for LBR emulation in the guest
- Support for communicating bus lock vmexits to userspace
- Add support for SEV attestation command
- Miscellaneous cleanups
PPC:
- Support for second data watchpoint on POWER10
- Remove some complex workarounds for buggy early versions of POWER9
- Guest entry/exit fixes
ARM64:
- Make the nVHE EL2 object relocatable
- Cleanups for concurrent translation faults hitting the same page
- Support for the standard TRNG hypervisor call
- A bunch of small PMU/Debug fixes
- Simplification of the early init hypercall handling
Non-KVM changes (with acks):
- Detection of contended rwlocks (implemented only for qrwlocks,
because KVM only needs it for x86)
- Allow __DISABLE_EXPORTS from assembly code
- Provide a saner follow_pfn replacements for modules"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (192 commits)
KVM: x86/xen: Explicitly pad struct compat_vcpu_info to 64 bytes
KVM: selftests: Don't bother mapping GVA for Xen shinfo test
KVM: selftests: Fix hex vs. decimal snafu in Xen test
KVM: selftests: Fix size of memslots created by Xen tests
KVM: selftests: Ignore recently added Xen tests' build output
KVM: selftests: Add missing header file needed by xAPIC IPI tests
KVM: selftests: Add operand to vmsave/vmload/vmrun in svm.c
KVM: SVM: Make symbol 'svm_gp_erratum_intercept' static
locking/arch: Move qrwlock.h include after qspinlock.h
KVM: PPC: Book3S HV: Fix host radix SLB optimisation with hash guests
KVM: PPC: Book3S HV: Ensure radix guest has no SLB entries
KVM: PPC: Don't always report hash MMU capability for P9 < DD2.2
KVM: PPC: Book3S HV: Save and restore FSCR in the P9 path
KVM: PPC: remove unneeded semicolon
KVM: PPC: Book3S HV: Use POWER9 SLBIA IH=6 variant to clear SLB
KVM: PPC: Book3S HV: No need to clear radix host SLB before loading HPT guest
KVM: PPC: Book3S HV: Fix radix guest SLB side channel
KVM: PPC: Book3S HV: Remove support for running HPT guest on RPT host without mixed mode support
KVM: PPC: Book3S HV: Introduce new capability for 2nd DAWR
KVM: PPC: Book3S HV: Add infrastructure to support 2nd DAWR
...
[ NOTE: unfortunately this tree had to be freshly rebased today,
it's a same-content tree of 82891be90f3c (-next published)
merged with v5.11.
The main reason for the rebase was an authorship misattribution
problem with a new commit, which we noticed in the last minute,
and which we didn't want to be merged upstream. The offending
commit was deep in the tree, and dependent commits had to be
rebased as well. ]
- Core scheduler updates:
- Add CONFIG_PREEMPT_DYNAMIC: this in its current form adds the
preempt=none/voluntary/full boot options (default: full),
to allow distros to build a PREEMPT kernel but fall back to
close to PREEMPT_VOLUNTARY (or PREEMPT_NONE) runtime scheduling
behavior via a boot time selection.
There's also the /debug/sched_debug switch to do this runtime.
This feature is implemented via runtime patching (a new variant of static calls).
The scope of the runtime patching can be best reviewed by looking
at the sched_dynamic_update() function in kernel/sched/core.c.
( Note that the dynamic none/voluntary mode isn't 100% identical,
for example preempt-RCU is available in all cases, plus the
preempt count is maintained in all models, which has runtime
overhead even with the code patching. )
The PREEMPT_VOLUNTARY/PREEMPT_NONE models, used by the vast majority
of distributions, are supposed to be unaffected.
- Fix ignored rescheduling after rcu_eqs_enter(). This is a bug that
was found via rcutorture triggering a hang. The bug is that
rcu_idle_enter() may wake up a NOCB kthread, but this happens after
the last generic need_resched() check. Some cpuidle drivers fix it
by chance but many others don't.
In true 2020 fashion the original bug fix has grown into a 5-patch
scheduler/RCU fix series plus another 16 RCU patches to address
the underlying issue of missed preemption events. These are the
initial fixes that should fix current incarnations of the bug.
- Clean up rbtree usage in the scheduler, by providing & using the following
consistent set of rbtree APIs:
partial-order; less() based:
- rb_add(): add a new entry to the rbtree
- rb_add_cached(): like rb_add(), but for a rb_root_cached
total-order; cmp() based:
- rb_find(): find an entry in an rbtree
- rb_find_add(): find an entry, and add if not found
- rb_find_first(): find the first (leftmost) matching entry
- rb_next_match(): continue from rb_find_first()
- rb_for_each(): iterate a sub-tree using the previous two
- Improve the SMP/NUMA load-balancer: scan for an idle sibling in a single pass.
This is a 4-commit series where each commit improves one aspect of the idle
sibling scan logic.
- Improve the cpufreq cooling driver by getting the effective CPU utilization
metrics from the scheduler
- Improve the fair scheduler's active load-balancing logic by reducing the number
of active LB attempts & lengthen the load-balancing interval. This improves
stress-ng mmapfork performance.
- Fix CFS's estimated utilization (util_est) calculation bug that can result in
too high utilization values
- Misc updates & fixes:
- Fix the HRTICK reprogramming & optimization feature
- Fix SCHED_SOFTIRQ raising race & warning in the CPU offlining code
- Reduce dl_add_task_root_domain() overhead
- Fix uprobes refcount bug
- Process pending softirqs in flush_smp_call_function_from_idle()
- Clean up task priority related defines, remove *USER_*PRIO and
USER_PRIO()
- Simplify the sched_init_numa() deduplication sort
- Documentation updates
- Fix EAS bug in update_misfit_status(), which degraded the quality
of energy-balancing
- Smaller cleanups
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----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=pyaG
-----END PGP SIGNATURE-----
Merge tag 'sched-core-2021-02-17' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull scheduler updates from Ingo Molnar:
"Core scheduler updates:
- Add CONFIG_PREEMPT_DYNAMIC: this in its current form adds the
preempt=none/voluntary/full boot options (default: full), to allow
distros to build a PREEMPT kernel but fall back to close to
PREEMPT_VOLUNTARY (or PREEMPT_NONE) runtime scheduling behavior via
a boot time selection.
There's also the /debug/sched_debug switch to do this runtime.
This feature is implemented via runtime patching (a new variant of
static calls).
The scope of the runtime patching can be best reviewed by looking
at the sched_dynamic_update() function in kernel/sched/core.c.
( Note that the dynamic none/voluntary mode isn't 100% identical,
for example preempt-RCU is available in all cases, plus the
preempt count is maintained in all models, which has runtime
overhead even with the code patching. )
The PREEMPT_VOLUNTARY/PREEMPT_NONE models, used by the vast
majority of distributions, are supposed to be unaffected.
- Fix ignored rescheduling after rcu_eqs_enter(). This is a bug that
was found via rcutorture triggering a hang. The bug is that
rcu_idle_enter() may wake up a NOCB kthread, but this happens after
the last generic need_resched() check. Some cpuidle drivers fix it
by chance but many others don't.
In true 2020 fashion the original bug fix has grown into a 5-patch
scheduler/RCU fix series plus another 16 RCU patches to address the
underlying issue of missed preemption events. These are the initial
fixes that should fix current incarnations of the bug.
- Clean up rbtree usage in the scheduler, by providing & using the
following consistent set of rbtree APIs:
partial-order; less() based:
- rb_add(): add a new entry to the rbtree
- rb_add_cached(): like rb_add(), but for a rb_root_cached
total-order; cmp() based:
- rb_find(): find an entry in an rbtree
- rb_find_add(): find an entry, and add if not found
- rb_find_first(): find the first (leftmost) matching entry
- rb_next_match(): continue from rb_find_first()
- rb_for_each(): iterate a sub-tree using the previous two
- Improve the SMP/NUMA load-balancer: scan for an idle sibling in a
single pass. This is a 4-commit series where each commit improves
one aspect of the idle sibling scan logic.
- Improve the cpufreq cooling driver by getting the effective CPU
utilization metrics from the scheduler
- Improve the fair scheduler's active load-balancing logic by
reducing the number of active LB attempts & lengthen the
load-balancing interval. This improves stress-ng mmapfork
performance.
- Fix CFS's estimated utilization (util_est) calculation bug that can
result in too high utilization values
Misc updates & fixes:
- Fix the HRTICK reprogramming & optimization feature
- Fix SCHED_SOFTIRQ raising race & warning in the CPU offlining code
- Reduce dl_add_task_root_domain() overhead
- Fix uprobes refcount bug
- Process pending softirqs in flush_smp_call_function_from_idle()
- Clean up task priority related defines, remove *USER_*PRIO and
USER_PRIO()
- Simplify the sched_init_numa() deduplication sort
- Documentation updates
- Fix EAS bug in update_misfit_status(), which degraded the quality
of energy-balancing
- Smaller cleanups"
* tag 'sched-core-2021-02-17' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (51 commits)
sched,x86: Allow !PREEMPT_DYNAMIC
entry/kvm: Explicitly flush pending rcuog wakeup before last rescheduling point
entry: Explicitly flush pending rcuog wakeup before last rescheduling point
rcu/nocb: Trigger self-IPI on late deferred wake up before user resume
rcu/nocb: Perform deferred wake up before last idle's need_resched() check
rcu: Pull deferred rcuog wake up to rcu_eqs_enter() callers
sched/features: Distinguish between NORMAL and DEADLINE hrtick
sched/features: Fix hrtick reprogramming
sched/deadline: Reduce rq lock contention in dl_add_task_root_domain()
uprobes: (Re)add missing get_uprobe() in __find_uprobe()
smp: Process pending softirqs in flush_smp_call_function_from_idle()
sched: Harden PREEMPT_DYNAMIC
static_call: Allow module use without exposing static_call_key
sched: Add /debug/sched_preempt
preempt/dynamic: Support dynamic preempt with preempt= boot option
preempt/dynamic: Provide irqentry_exit_cond_resched() static call
preempt/dynamic: Provide preempt_schedule[_notrace]() static calls
preempt/dynamic: Provide cond_resched() and might_resched() static calls
preempt: Introduce CONFIG_PREEMPT_DYNAMIC
static_call: Provide DEFINE_STATIC_CALL_RET0()
...
Reduce rbtree boiler plate by using the new helpers.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Davidlohr Bueso <dbueso@suse.de>
Pull RCU updates from Paul E. McKenney:
- Documentation updates.
- Miscellaneous fixes.
- kfree_rcu() updates: Addition of mem_dump_obj() to provide allocator return
addresses to more easily locate bugs. This has a couple of RCU-related commits,
but is mostly MM. Was pulled in with akpm's agreement.
- Per-callback-batch tracking of numbers of callbacks,
which enables better debugging information and smarter
reactions to large numbers of callbacks.
- The first round of changes to allow CPUs to be runtime switched from and to
callback-offloaded state.
- CONFIG_PREEMPT_RT-related changes.
- RCU CPU stall warning updates.
- Addition of polling grace-period APIs for SRCU.
- Torture-test and torture-test scripting updates, including a "torture everything"
script that runs rcutorture, locktorture, scftorture, rcuscale, and refscale.
Plus does an allmodconfig build.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
include/asm-generic/qrwlock.h was trying to get arch_spin_is_locked via
asm-generic/qspinlock.h. However, this does not work because architectures
might be using queued rwlocks but not queued spinlocks (csky), or because they
might be defining their own queued_* macros before including asm/qspinlock.h.
To fix this, ensure that asm/spinlock.h always includes qrwlock.h after
defining arch_spin_is_locked (either directly for csky, or via
asm/qspinlock.h for other architectures). The only inclusion elsewhere
is in kernel/locking/qrwlock.c. That one is really unnecessary because
the file is only compiled in SMP configurations (config QUEUED_RWLOCKS
depends on SMP) and in that case linux/spinlock.h already includes
asm/qrwlock.h if needed, via asm/spinlock.h.
Reported-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Waiman Long <longman@redhat.com>
Fixes: 26128cb6c7 ("locking/rwlocks: Add contention detection for rwlocks")
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Ben Gardon <bgardon@google.com>
[Add arch/sparc and kernel/locking parts per discussion with Waiman. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
There are not users of mutex_trylock_recursive() in tree as of
v5.11-rc7.
Remove it.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210210085248.219210-2-bigeasy@linutronix.de
vmlinux.o: warning: objtool: lock_is_held_type()+0x107: call to warn_bogus_irq_restore() leaves .noinstr.text section
As per the general rule that WARNs are allowed to violate noinstr to
get out, annotate it away.
Fixes: 997acaf6b4 ("lockdep: report broken irq restoration")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Mark Rutland <mark.rutland@arm.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org> # build-tested
Link: https://lkml.kernel.org/r/YCKyYg53mMp4E7YI@hirez.programming.kicks-ass.net
Commit f6f48e1804 ("lockdep: Teach lockdep about "USED" <- "IN-NMI"
inversions") overlooked that print_usage_bug() releases the graph_lock
and called it without the graph lock held.
Fixes: f6f48e1804 ("lockdep: Teach lockdep about "USED" <- "IN-NMI" inversions")
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Waiman Long <longman@redhat.com>
Link: https://lkml.kernel.org/r/YBfkuyIfB1+VRxXP@hirez.programming.kicks-ass.net
This is a leftover from 7f26482a87 ("locking/percpu-rwsem: Remove the embedded rwsem")
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/r/20210126101721.976027-1-nborisov@suse.com
To fix the following issues:
kernel/locking/rtmutex.c:1612: warning: Function parameter or member
'lock' not described in '__rt_mutex_futex_unlock'
kernel/locking/rtmutex.c:1612: warning: Function parameter or member
'wake_q' not described in '__rt_mutex_futex_unlock'
kernel/locking/rtmutex.c:1675: warning: Function parameter or member
'name' not described in '__rt_mutex_init'
kernel/locking/rtmutex.c:1675: warning: Function parameter or member
'key' not described in '__rt_mutex_init'
[ tglx: Change rt lock to rt_mutex for consistency sake ]
Signed-off-by: Alex Shi <alex.shi@linux.alibaba.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/1605257895-5536-2-git-send-email-alex.shi@linux.alibaba.com
Nothing uses the argument. Remove it as preparation to use
pi_state_update_owner().
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
We generally expect local_irq_save() and local_irq_restore() to be
paired and sanely nested, and so local_irq_restore() expects to be
called with irqs disabled. Thus, within local_irq_restore() we only
trace irq flag changes when unmasking irqs.
This means that a sequence such as:
| local_irq_disable();
| local_irq_save(flags);
| local_irq_enable();
| local_irq_restore(flags);
... is liable to break things, as the local_irq_restore() would mask
irqs without tracing this change. Similar problems may exist for
architectures whose arch_irq_restore() function depends on being called
with irqs disabled.
We don't consider such sequences to be a good idea, so let's define
those as forbidden, and add tooling to detect such broken cases.
This patch adds debug code to WARN() when raw_local_irq_restore() is
called with irqs enabled. As raw_local_irq_restore() is expected to pair
with raw_local_irq_save(), it should never be called with irqs enabled.
To avoid the possibility of circular header dependencies between
irqflags.h and bug.h, the warning is handled in a separate C file.
The new code is all conditional on a new CONFIG_DEBUG_IRQFLAGS symbol
which is independent of CONFIG_TRACE_IRQFLAGS. As noted above such cases
will confuse lockdep, so CONFIG_DEBUG_LOCKDEP now selects
CONFIG_DEBUG_IRQFLAGS.
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210111153707.10071-1-mark.rutland@arm.com
The purpose of local_lock_t is to abstract: preempt_disable() /
local_bh_disable() / local_irq_disable(). These are the traditional
means of gaining access to per-cpu data, but are fundamentally
non-preemptible.
local_lock_t provides a per-cpu lock, that on !PREEMPT_RT reduces to
no-ops, just like regular spinlocks do on UP.
This gives rise to:
CPU0 CPU1
local_lock(B) spin_lock_irq(A)
<IRQ>
spin_lock(A) local_lock(B)
Where lockdep then figures things will lock up; which would be true if
B were any other kind of lock. However this is a false positive, no
such deadlock actually exists.
For !RT the above local_lock(B) is preempt_disable(), and there's
obviously no deadlock; alternatively, CPU0's B != CPU1's B.
For RT the argument is that since local_lock() nests inside
spin_lock(), it cannot be used in hardirq context, and therefore CPU0
cannot in fact happen. Even though B is a real lock, it is a
preemptible lock and any threaded-irq would simply schedule out and
let the preempted task (which holds B) continue such that the task on
CPU1 can make progress, after which the threaded-irq resumes and can
finish.
This means that we can never form an IRQ inversion on a local_lock
dependency, so terminate the graph walk when looking for IRQ
inversions when we encounter one.
One consequence is that (for LOCKDEP_SMALL) when we look for redundant
dependencies, A -> B is not redundant in the presence of A -> L -> B.
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
[peterz: Changelog]
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
In preparation for adding an TRACE_IRQFLAGS dependent skip function to
check_redundant(), move it below the TRACE_IRQFLAGS #ifdef.
While there, provide a stub function to reduce #ifdef usage.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Some __bfs() walks will have additional iteration constraints (beyond
the path being strong). Provide an additional function to allow
terminating graph walks.
Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
The local_lock_t's are special, because they cannot form IRQ
inversions, make sure we can tell them apart from the rest of the
locks.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
When the compiler doesn't feel like inlining, it causes a noinstr
fail:
vmlinux.o: warning: objtool: lock_is_held_type()+0xb: call to lockdep_enabled() leaves .noinstr.text section
Fixes: 4d004099a6 ("lockdep: Fix lockdep recursion")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210106144017.592595176@infradead.org
- A few extensions to the rwsem API and support for opportunistic
spinning and lock stealing
- lockdep selftest improvements
- Documentation updates
- Cleanups and small fixes all over the place
-----BEGIN PGP SIGNATURE-----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=uqcN
-----END PGP SIGNATURE-----
Merge tag 'locking-core-2020-12-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull locking updates from Thomas Gleixner:
"A moderate set of locking updates:
- A few extensions to the rwsem API and support for opportunistic
spinning and lock stealing
- lockdep selftest improvements
- Documentation updates
- Cleanups and small fixes all over the place"
* tag 'locking-core-2020-12-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
seqlock: kernel-doc: Specify when preemption is automatically altered
seqlock: Prefix internal seqcount_t-only macros with a "do_"
Documentation: seqlock: s/LOCKTYPE/LOCKNAME/g
locking/rwsem: Remove reader optimistic spinning
locking/rwsem: Enable reader optimistic lock stealing
locking/rwsem: Prevent potential lock starvation
locking/rwsem: Pass the current atomic count to rwsem_down_read_slowpath()
locking/rwsem: Fold __down_{read,write}*()
locking/rwsem: Introduce rwsem_write_trylock()
locking/rwsem: Better collate rwsem_read_trylock()
rwsem: Implement down_read_interruptible
rwsem: Implement down_read_killable_nested
refcount: Fix a kernel-doc markup
completion: Drop init_completion define
atomic: Update MAINTAINERS
atomic: Delete obsolete documentation
seqlock: Rename __seqprop() users
lockdep/selftest: Add spin_nest_lock test
lockdep/selftests: Fix PROVE_RAW_LOCK_NESTING
seqlock: avoid -Wshadow warnings
...
RCU:
- Avoid cpuinfo-induced IPI pileups and idle-CPU IPIs.
- Lockdep-RCU updates reducing the need for __maybe_unused.
- Tasks-RCU updates.
- Miscellaneous fixes.
- Documentation updates.
- Torture-test updates.
KCSAN:
- updates for selftests, avoiding setting watchpoints on NULL pointers
- fix to watchpoint encoding
LKMM:
- updates for documentation along with some updates to example-code
litmus tests
-----BEGIN PGP SIGNATURE-----
iQJHBAABCgAxFiEEQp8+kY+LLUocC4bMphj1TA10mKEFAl/Xon4THHRnbHhAbGlu
dXRyb25peC5kZQAKCRCmGPVMDXSYobXUD/92LJTI/TMgK6Z6EEQBiJZO/2mNKjK8
FEKc6AqTNMlZNsWCfQ5UgqtHpn+MkBZsX1x4u22gehE1qaCB8gnQ5wXgbXon8tQm
exxVk6vvQZjseeqCMqrsUYQlD7dNgHnf1qAmWXJvji4sA/1Opo6n2M74tqfE2ueV
S5hpQwSuK/6Zu2Hrr62HD8+Fx0in6ZuKRZxHGp1392l++DGbniJM3dzntRXB+JbZ
w3PDHFCQuGzTytyeKuQV48ot9IK+2YzmjIp/+4tHL6mvU38xeSu6gcYtqKPcfYWw
D6HXvDa965h5IrFdSA2JWSzjJ+VYgZVElk2HyXDNIae0fM/8GidgoIDQipT1WAur
sxW/Ke4U6Jm5MMqXqV8iMNduktkGD1/h6G/iB1Yis29xFdthorNpbHVAP+8cKXgf
1cR6RorOuBYv6XpyzygHtE7qfLY5ST352pJ4+UqNzboujOcuEnGaygttt0F/F8sA
ZH8NT5dyUfbGeqepdZWkbj116Hjeg3fyV3CZeyBhDeqpjf1Nn3nbJ1xRksPLfa3i
IKvN7HSzEg+vKnsJNnQeFlAmQ/W3n2bedzRqfaCg77pNhKI6jPuavY5f2YGFUj0y
yx0UzOYoI1Cln0keBMmynbyUKgJ7zstLkrt/JenjhtD3B+0df5BmYjkL+nqkP6ax
+XTCu7Xg+B061g==
=N/iO
-----END PGP SIGNATURE-----
Merge tag 'core-rcu-2020-12-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull RCU updates from Thomas Gleixner:
"RCU, LKMM and KCSAN updates collected by Paul McKenney.
RCU:
- Avoid cpuinfo-induced IPI pileups and idle-CPU IPIs
- Lockdep-RCU updates reducing the need for __maybe_unused
- Tasks-RCU updates
- Miscellaneous fixes
- Documentation updates
- Torture-test updates
KCSAN:
- updates for selftests, avoiding setting watchpoints on NULL pointers
- fix to watchpoint encoding
LKMM:
- updates for documentation along with some updates to example-code
litmus tests"
* tag 'core-rcu-2020-12-14' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (72 commits)
srcu: Take early exit on memory-allocation failure
rcu/tree: Defer kvfree_rcu() allocation to a clean context
rcu: Do not report strict GPs for outgoing CPUs
rcu: Fix a typo in rcu_blocking_is_gp() header comment
rcu: Prevent lockdep-RCU splats on lock acquisition/release
rcu/tree: nocb: Avoid raising softirq for offloaded ready-to-execute CBs
rcu,ftrace: Fix ftrace recursion
rcu/tree: Make struct kernel_param_ops definitions const
rcu/tree: Add a warning if CPU being onlined did not report QS already
rcu: Clarify nocb kthreads naming in RCU_NOCB_CPU config
rcu: Fix single-CPU check in rcu_blocking_is_gp()
rcu: Implement rcu_segcblist_is_offloaded() config dependent
list.h: Update comment to explicitly note circular lists
rcu: Panic after fixed number of stalls
x86/smpboot: Move rcu_cpu_starting() earlier
rcu: Allow rcu_irq_enter_check_tick() from NMI
tools/memory-model: Label MP tests' producers and consumers
tools/memory-model: Use "buf" and "flag" for message-passing tests
tools/memory-model: Add types to litmus tests
tools/memory-model: Add a glossary of LKMM terms
...
Reader optimistic spinning is helpful when the reader critical section
is short and there aren't that many readers around. It also improves
the chance that a reader can get the lock as writer optimistic spinning
disproportionally favors writers much more than readers.
Since commit d3681e269f ("locking/rwsem: Wake up almost all readers
in wait queue"), all the waiting readers are woken up so that they can
all get the read lock and run in parallel. When the number of contending
readers is large, allowing reader optimistic spinning will likely cause
reader fragmentation where multiple smaller groups of readers can get
the read lock in a sequential manner separated by writers. That reduces
reader parallelism.
One possible way to address that drawback is to limit the number of
readers (preferably one) that can do optimistic spinning. These readers
act as representatives of all the waiting readers in the wait queue as
they will wake up all those waiting readers once they get the lock.
Alternatively, as reader optimistic lock stealing has already enhanced
fairness to readers, it may be easier to just remove reader optimistic
spinning and simplifying the optimistic spinning code as a result.
Performance measurements (locking throughput kops/s) using a locking
microbenchmark with 50/50 reader/writer distribution and turbo-boost
disabled was done on a 2-socket Cascade Lake system (48-core 96-thread)
to see the impacts of these changes:
1) Vanilla - 5.10-rc3 kernel
2) Before - 5.10-rc3 kernel with previous patches in this series
2) limit-rspin - 5.10-rc3 kernel with limited reader spinning patch
3) no-rspin - 5.10-rc3 kernel with reader spinning disabled
# of threads CS Load Vanilla Before limit-rspin no-rspin
------------ ------- ------- ------ ----------- --------
2 1 5,185 5,662 5,214 5,077
4 1 5,107 4,983 5,188 4,760
8 1 4,782 4,564 4,720 4,628
16 1 4,680 4,053 4,567 3,402
32 1 4,299 1,115 1,118 1,098
64 1 3,218 983 1,001 957
96 1 1,938 944 957 930
2 20 2,008 2,128 2,264 1,665
4 20 1,390 1,033 1,046 1,101
8 20 1,472 1,155 1,098 1,213
16 20 1,332 1,077 1,089 1,122
32 20 967 914 917 980
64 20 787 874 891 858
96 20 730 836 847 844
2 100 372 356 360 355
4 100 492 425 434 392
8 100 533 537 529 538
16 100 548 572 568 598
32 100 499 520 527 537
64 100 466 517 526 512
96 100 406 497 506 509
The column "CS Load" represents the number of pause instructions issued
in the locking critical section. A CS load of 1 is extremely short and
is not likey in real situations. A load of 20 (moderate) and 100 (long)
are more realistic.
It can be seen that the previous patches in this series have reduced
performance in general except in highly contended cases with moderate
or long critical sections that performance improves a bit. This change
is mostly caused by the "Prevent potential lock starvation" patch that
reduce reader optimistic spinning and hence reduce reader fragmentation.
The patch that further limit reader optimistic spinning doesn't seem to
have too much impact on overall performance as shown in the benchmark
data.
The patch that disables reader optimistic spinning shows reduced
performance at lightly loaded cases, but comparable or slightly better
performance on with heavier contention.
This patch just removes reader optimistic spinning for now. As readers
are not going to do optimistic spinning anymore, we don't need to
consider if the OSQ is empty or not when doing lock stealing.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Davidlohr Bueso <dbueso@suse.de>
Link: https://lkml.kernel.org/r/20201121041416.12285-6-longman@redhat.com
If the optimistic spinning queue is empty and the rwsem does not have
the handoff or write-lock bits set, it is actually not necessary to
call rwsem_optimistic_spin() to spin on it. Instead, it can steal the
lock directly as its reader bias is in the count already. If it is
the first reader in this state, it will try to wake up other readers
in the wait queue.
With this patch applied, the following were the lock event counts
after rebooting a 2-socket system and a "make -j96" kernel rebuild.
rwsem_opt_rlock=4437
rwsem_rlock=29
rwsem_rlock_steal=19
So lock stealing represents about 0.4% of all the read locks acquired
in the slow path.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Davidlohr Bueso <dbueso@suse.de>
Link: https://lkml.kernel.org/r/20201121041416.12285-4-longman@redhat.com
The lock handoff bit is added in commit 4f23dbc1e6 ("locking/rwsem:
Implement lock handoff to prevent lock starvation") to avoid lock
starvation. However, allowing readers to do optimistic spinning does
introduce an unlikely scenario where lock starvation can happen.
The lock handoff bit may only be set when a waiter is being woken up.
In the case of reader unlock, wakeup happens only when the reader count
reaches 0. If there is a continuous stream of incoming readers acquiring
read lock via optimistic spinning, it is possible that the reader count
may never reach 0 and so the handoff bit will never be asserted.
One way to prevent this scenario from happening is to disallow optimistic
spinning if the rwsem is currently owned by readers. If the previous
or current owner is a writer, optimistic spinning will be allowed.
If the previous owner is a reader but the reader count has reached 0
before, a wakeup should have been issued. So the handoff mechanism
will be kicked in to prevent lock starvation. As a result, it should
be OK to do optimistic spinning in this case.
This patch may have some impact on reader performance as it reduces
reader optimistic spinning especially if the lock critical sections
are short the number of contending readers are small.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Davidlohr Bueso <dbueso@suse.de>
Link: https://lkml.kernel.org/r/20201121041416.12285-3-longman@redhat.com
