Pull scheduler updates from Ingo Molnar:

 - Cleanups for SCHED_DEADLINE

 - Tracing updates/fixes

 - CPU Accounting fixes

 - First wave of changes to optimize the overhead of the scheduler
   build, from the fast-headers tree - including placeholder *_api.h
   headers for later header split-ups.

 - Preempt-dynamic using static_branch() for ARM64

 - Isolation housekeeping mask rework; preperatory for further changes

 - NUMA-balancing: deal with CPU-less nodes

 - NUMA-balancing: tune systems that have multiple LLC cache domains per
   node (eg. AMD)

 - Updates to RSEQ UAPI in preparation for glibc usage

 - Lots of RSEQ/selftests, for same

 - Add Suren as PSI co-maintainer

* tag 'sched-core-2022-03-22' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (81 commits)
  sched/headers: ARM needs asm/paravirt_api_clock.h too
  sched/numa: Fix boot crash on arm64 systems
  headers/prep: Fix header to build standalone: <linux/psi.h>
  sched/headers: Only include <linux/entry-common.h> when CONFIG_GENERIC_ENTRY=y
  cgroup: Fix suspicious rcu_dereference_check() usage warning
  sched/preempt: Tell about PREEMPT_DYNAMIC on kernel headers
  sched/topology: Remove redundant variable and fix incorrect type in build_sched_domains
  sched/deadline,rt: Remove unused parameter from pick_next_[rt|dl]_entity()
  sched/deadline,rt: Remove unused functions for !CONFIG_SMP
  sched/deadline: Use __node_2_[pdl|dle]() and rb_first_cached() consistently
  sched/deadline: Merge dl_task_can_attach() and dl_cpu_busy()
  sched/deadline: Move bandwidth mgmt and reclaim functions into sched class source file
  sched/deadline: Remove unused def_dl_bandwidth
  sched/tracing: Report TASK_RTLOCK_WAIT tasks as TASK_UNINTERRUPTIBLE
  sched/tracing: Don't re-read p->state when emitting sched_switch event
  sched/rt: Plug rt_mutex_setprio() vs push_rt_task() race
  sched/cpuacct: Remove redundant RCU read lock
  sched/cpuacct: Optimize away RCU read lock
  sched/cpuacct: Fix charge percpu cpuusage
  sched/headers: Reorganize, clean up and optimize kernel/sched/sched.h dependencies
  ...

On some microarchitectures, clearing PSTATE.TCO is expensive. Clearing
TCO is only necessary if in-kernel MTE is enabled, or if MTE is
enabled in the userspace process in synchronous (or, soon, asymmetric)
mode, because we do not report uaccess faults to userspace in none
or asynchronous modes. Therefore, adjust the kernel entry code to
clear TCO only if necessary.

Because it is now possible to switch to a task in which TCO needs to
be clear from a task in which TCO is set, we also need to do the same
thing on task switch.

Signed-off-by: Peter Collingbourne <pcc@google.com>
Link: https://linux-review.googlesource.com/id/I52d82a580bd0500d420be501af2c35fa8c90729e
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Link: https://lore.kernel.org/r/20220219012945.894950-2-pcc@google.com
Signed-off-by: Will Deacon <will@kernel.org>

This patch enables support for PREEMPT_DYNAMIC on arm64, allowing the
preemption model to be chosen at boot time.

Specifically, this patch selects HAVE_PREEMPT_DYNAMIC_KEY, so that each
preemption function is an out-of-line call with an early return
depending upon a static key. This leaves almost all the codegen up to
the compiler, and side-steps a number of pain points with static calls
(e.g. interaction with CFI schemes). This should have no worse overhead
than using non-inline static calls, as those use out-of-line trampolines
with early returns.

For example, the dynamic_cond_resched() wrapper looks as follows when
enabled. When disabled, the first `B` is replaced with a `NOP`,
resulting in an early return.

| <dynamic_cond_resched>:
|        bti     c
|        b       <dynamic_cond_resched+0x10>     // or `nop`
|        mov     w0, #0x0
|        ret
|        mrs     x0, sp_el0
|        ldr     x0, [x0, #8]
|        cbnz    x0, <dynamic_cond_resched+0x8>
|        paciasp
|        stp     x29, x30, [sp, #-16]!
|        mov     x29, sp
|        bl      <preempt_schedule_common>
|        mov     w0, #0x1
|        ldp     x29, x30, [sp], #16
|        autiasp
|        ret

... compared to the regular form of the function:

| <__cond_resched>:
|        bti     c
|        mrs     x0, sp_el0
|        ldr     x1, [x0, #8]
|        cbz     x1, <__cond_resched+0x18>
|        mov     w0, #0x0
|        ret
|        paciasp
|        stp     x29, x30, [sp, #-16]!
|        mov     x29, sp
|        bl      <preempt_schedule_common>
|        mov     w0, #0x1
|        ldp     x29, x30, [sp], #16
|        autiasp
|        ret

Since arm64 does not yet use the generic entry code, we must define our
own `sk_dynamic_irqentry_exit_cond_resched`, which will be
enabled/disabled by the common code in kernel/sched/core.c. All other
preemption functions and associated static keys are defined there.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lore.kernel.org/r/20220214165216.2231574-8-mark.rutland@arm.com

For historical reasons, the decision of whether or not to preempt is
spread across arm64_preempt_schedule_irq() and __el1_irq(), and it would
be clearer if this were all in one place.

Also, arm64_preempt_schedule_irq() calls lockdep_assert_irqs_disabled(),
but this is redundant, as we have a subsequent identical assertion in
__exit_to_kernel_mode(), and preempt_schedule_irq() will
BUG_ON(!irqs_disabled()) anyway.

This patch removes the redundant assertion and centralizes the
preemption decision making within arm64_preempt_schedule_irq().

Other than the slight change to assertion behaviour, there should be no
functional change as a result of this patch.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lore.kernel.org/r/20220214165216.2231574-7-mark.rutland@arm.com

Some thread flags can be set remotely, and so even when IRQs are disabled,
the flags can change under our feet. Generally this is unlikely to cause a
problem in practice, but it is somewhat unsound, and KCSAN will
legitimately warn that there is a data race.

To avoid such issues, a snapshot of the flags has to be taken prior to
using them. Some places already use READ_ONCE() for that, others do not.

Convert them all to the new flag accessor helpers.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Will Deacon <will@kernel.org>
Acked-by: Paul E. McKenney <paulmck@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Link: https://lore.kernel.org/r/20211129130653.2037928-7-mark.rutland@arm.com

In preparation for removing HANDLE_DOMAIN_IRQ_IRQENTRY, have arch/arm64
perform all the irqentry accounting in its entry code.

As arch/arm64 already performs portions of the irqentry logic in
enter_from_kernel_mode() and exit_to_kernel_mode(), including
rcu_irq_{enter,exit}(), the only additional calls that need to be made
are to irq_{enter,exit}_rcu(). Removing the calls to
rcu_irq_{enter,exit}() from handle_domain_irq() ensures that we inform
RCU once per IRQ entry and will correctly identify quiescent periods.

Since we should not call irq_{enter,exit}_rcu() when entering a
pseudo-NMI, el1_interrupt() is reworked to have separate __el1_irq() and
__el1_pnmi() paths for regular IRQ and psuedo-NMI entry, with
irq_{enter,exit}_irq() only called for the former.

In preparation for removing HANDLE_DOMAIN_IRQ, the irq regs are managed
in do_interrupt_handler() for both regular IRQ and pseudo-NMI. This is
currently redundant, but not harmful.

For clarity the preemption logic is moved into __el1_irq(). We should
never preempt within a pseudo-NMI, and arm64_enter_nmi() already
enforces this by incrementing the preempt_count, but it's clearer if we
never invoke the preemption logic when entering a pseudo-NMI.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Reviewed-by: Pingfan Liu <kernelfans@gmail.com>
Reviewed-by: Marc Zyngier <maz@kernel.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will@kernel.org>

When handling an exception from EL0, we perform the entry work in that
exception's C handler, and once the C handler has finished, we return
back to the entry assembly. Subsequently in the common `ret_to_user`
assembly we perform the exit work that balances with the entry work.
This can be somewhat difficult to follow, and makes it hard to rework
the return paths (e.g. to pass additional context to the exit code, or
to have exception return logic for specific exceptions).

This patch reworks the entry code such that each EL0 C exception handler
is responsible for both the entry and exit work. This clearly balances
the two (and will permit additional variation in future), and avoids an
unnecessary bounce between assembly and C in the common case, leaving
`ret_from_fork` as the only place assembly has to call the exit code.
This means that the exit work is now inlined into the C handler, which
is already the case for the entry work, and allows the compiler to
generate better code (e.g. by immediately returning when there is no
exit work to perform).

To align with other exception entry/exit helpers, enter_from_user_mode()
is updated to take the EL0 pt_regs as a parameter, though this is
currently unused.

There should be no functional change as a result of this patch. However,
this should lead to slightly better backtraces when an error is
encountered within do_notify_resume(), as the C handler should appear in
the backtrace, indicating the specific exception that the kernel was
entered with.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Link: https://lore.kernel.org/r/20210802140733.52716-5-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>

In `ret_to_user` we perform some conditional work depending on the
thread flags, then perform some IRQ/context tracking which is intended
to balance with the IRQ/context tracking performed in the entry C code.

For simplicity and consistency, it would be preferable to move this all
to C. As a step towards that, this patch moves the conditional work and
IRQ/context tracking into a C helper function. To aid bisectability,
this is called from the `ret_to_user` assembly, and a subsequent patch
will move the call to C code.

As local_daif_mask() handles all necessary tracing and PMR manipulation,
we no longer need to handle this explicitly. As we call
exit_to_user_mode() directly, the `user_enter_irqoff` macro is no longer
used, and can be removed. As enter_from_user_mode() and
exit_to_user_mode() are no longer called from assembly, these can be
made static, and as these are typically very small, they are marked
__always_inline to avoid the overhead of a function call.

For now, enablement of single-step is left in entry.S, and for this we
still need to read the flags in ret_to_user(). It is safe to read this
separately as TIF_SINGLESTEP is not part of _TIF_WORK_MASK.

There should be no functional change as a result of this patch.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Link: https://lore.kernel.org/r/20210802140733.52716-4-mark.rutland@arm.com
[catalin.marinas@arm.com: removed unused gic_prio_kentry_setup macro]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>

When entering an exception, we must perform irq/context state management
before we can use instrumentable C code. Similarly, when exiting an
exception we cannot use instrumentable C code after we perform
irq/context state management.

Originally, we'd intended that the enter_from_*() and exit_to_*()
helpers would enforce this by virtue of being the first and last
functions called, respectively, in an exception handler. However, as
they now call instrumentable code themselves, this is not as clearly
true.

To make this more robust, this patch splits the irq/context state
management into separate helpers, with all the helpers commented to make
their intended purpose more obvious.

In exit_to_kernel_mode() we'll now check TFSR_EL1 before we assert that
IRQs are disabled, but this ordering is not important, and other than
this there should be no functional change as a result of this patch.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Link: https://lore.kernel.org/r/20210802140733.52716-3-mark.rutland@arm.com
[catalin.marinas@arm.com: comment typos fix-up]
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>

To make the various entry/exit helpers easier to understand and easier
to compare, this patch moves all the entry/exit helpers to be adjacent
at the top of entry-common.c, rather than being spread out throughout
the file.

There should be no functional change as a result of this patch.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Joey Gouly <joey.gouly@arm.com>
Cc: Marc Zyngier <maz@kernel.org>
Cc: Will Deacon <will@kernel.org>
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Link: https://lore.kernel.org/r/20210802140733.52716-2-mark.rutland@arm.com
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>