Thomas Gleixner [Tue, 5 Nov 2024 08:14:36 +0000 (09:14 +0100)]
posix-timers: Add a refcount to struct k_itimer
To cure the SIG_IGN handling for posix interval timers, the preallocated
sigqueue needs to be embedded into struct k_itimer to prevent life time
races of all sorts.
To make that work correctly it needs reference counting so that timer
deletion does not free the timer prematuraly when there is a signal queued
or delivered concurrently.
Thomas Gleixner [Tue, 5 Nov 2024 08:14:33 +0000 (09:14 +0100)]
posix-cpu-timers: Cleanup the firing logic
The firing flag of a posix CPU timer is tristate:
0: when the timer is not about to deliver a signal
1: when the timer has expired, but the signal has not been delivered yet
-1: when the timer was queued for signal delivery and a rearm operation
raced against it and supressed the signal delivery.
This is a pointless exercise as this can be simply expressed with a
boolean. Only if set, the signal is delivered. This makes delete and rearm
consistent with the rest of the posix timers.
Convert firing to bool and fixup the usage sites accordingly and add
comments why the timer cannot be dequeued right away.
Thomas Gleixner [Tue, 5 Nov 2024 08:14:32 +0000 (09:14 +0100)]
posix-timers: Make signal overrun accounting sensible
The handling of the timer overrun in the signal code is inconsistent as it
takes previous overruns into account. This is just wrong as after the
reprogramming of a timer the overrun count starts over from a clean state,
i.e. 0.
Don't touch info::si_overrun in send_sigqueue() and only store the overrun
value at signal delivery time, which is computed from the timer itself
relative to the expiry time.
Thomas Gleixner [Tue, 5 Nov 2024 08:14:31 +0000 (09:14 +0100)]
posix-timers: Make signal delivery consistent
Signals of timers which are reprogammed, disarmed or deleted can deliver
signals related to the past. The POSIX spec is blury about this:
- "The effect of disarming or resetting a timer with pending expiration
notifications is unspecified."
- "The disposition of pending signals for the deleted timer is
unspecified."
In both cases it is reasonable to expect that pending signals are
discarded. Especially in the reprogramming case it does not make sense to
account for previous overruns or to deliver a signal for a timer which has
been disarmed. This makes the behaviour consistent and understandable.
Remove the si_sys_private check from the signal delivery code and invoke
posix_timer_deliver_signal() unconditionally for posix timer related
signals.
Change posix_timer_deliver_signal() so it controls the actual signal
delivery via the return value. It now instructs the signal code to drop the
signal when:
1) The timer does not longer exist in the hash table
2) The timer signal_seq value is not the same as the si_sys_private value
which was set when the signal was queued.
This is also a preparatory change to embed the sigqueue into the k_itimer
structure, which in turn allows to remove the si_sys_private magic.
Thomas Gleixner [Tue, 5 Nov 2024 08:14:29 +0000 (09:14 +0100)]
posix-cpu-timers: Correctly update timer status in posix_cpu_timer_del()
If posix_cpu_timer_del() exits early due to task not found or sighand
invalid, it fails to clear the state of the timer. That's harmless but
inconsistent.
These early exits are accounted as successful delete. Move the update of
the timer state into the success return path, so all "successful" deletions
are handled.
Thomas Gleixner [Thu, 31 Oct 2024 12:04:08 +0000 (13:04 +0100)]
timekeeping: Always check for negative motion
clocksource_delta() has two variants. One with a check for negative motion,
which is only selected by x86. This is a historic leftover as this function
was previously used in the time getter hot paths.
Since 135225a363ae timekeeping_cycles_to_ns() has unconditional protection
against this as a by-product of the protection against 64bit math overflow.
clocksource_delta() is only used in the clocksource watchdog and in
timekeeping_advance(). The extra conditional there is not hurting anyone.
Remove the config option and unconditionally prevent negative motion of the
readout.
Thomas Gleixner [Thu, 31 Oct 2024 12:04:07 +0000 (13:04 +0100)]
timekeeping: Remove CONFIG_DEBUG_TIMEKEEPING
Since 135225a363ae timekeeping_cycles_to_ns() handles large offsets which
would lead to 64bit multiplication overflows correctly. It's also protected
against negative motion of the clocksource unconditionally, which was
exclusive to x86 before.
timekeeping_advance() handles large offsets already correctly.
That means the value of CONFIG_DEBUG_TIMEKEEPING which analyzed these cases
is very close to zero. Remove all of it.
Thomas Gleixner [Wed, 30 Oct 2024 07:53:51 +0000 (08:53 +0100)]
timers: Add missing READ_ONCE() in __run_timer_base()
__run_timer_base() checks base::next_expiry without holding
base::lock. That can race with a remote CPU updating next_expiry under the
lock. This is an intentional and harmless data race, but lacks a
READ_ONCE(), so KCSAN complains about this.
Add the missing READ_ONCE(). All other places are covered already.
Fixes: 79f8b28e85f8 ("timers: Annotate possible non critical data race of next_expiry") Reported-by: kernel test robot <oliver.sang@intel.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Reviewed-by: Frederic Weisbecker <frederic@kernel.org> Link: https://lore.kernel.org/all/87a5emyqk0.ffs@tglx Closes: https://lore.kernel.org/oe-lkp/202410301205.ef8e9743-lkp@intel.com
tick: Remove now unneeded low-res tick stop on CPUHP_AP_TICK_DYING
The generic clockevent layer now detaches and stops the underlying
clockevent from the dying CPU, unifying the tick behaviour for both
periodic and oneshot mode on offline CPUs. There is no more need for
the tick layer to care about that.
clockevents: Shutdown and unregister current clockevents at CPUHP_AP_TICK_DYING
The way the clockevent devices are finally stopped while a CPU is
offlining is currently chaotic. The layout being by order:
1) tick_sched_timer_dying() stops the tick and the underlying clockevent
but only for oneshot case. The periodic tick and its related
clockevent still runs.
2) tick_broadcast_offline() detaches and stops the per-cpu oneshot
broadcast and append it to the released list.
3) Some individual clockevent drivers stop the clockevents (a second time if
the tick is oneshot)
4) Once the CPU is dead, a control CPU remotely detaches and stops
(a 3rd time if oneshot mode) the CPU clockevent and adds it to the
released list.
5) The released list containing the broadcast device released on step 2)
and the remotely detached clockevent from step 4) are unregistered.
These random events can be factorized if the current clockevent is
detached and stopped by the dying CPU at the generic layer, that is
from the dying CPU:
a) Stop the tick
b) Stop/detach the underlying per-cpu oneshot broadcast clockevent
c) Stop/detach the underlying clockevent
d) Release / unregister the clockevents from b) and c)
e) Release / unregister the remaining clockevents from the dying CPU.
This part could be performed by the dying CPU
This way the drivers and the tick layer don't need to care about
clockevent operations during cpuhotplug down. This also unifies the tick
behaviour on offline CPUs between oneshot and periodic modes, avoiding
offline ticks altogether for sanity.
Adopt the simplification.
[ tglx: Remove the WARN_ON() in clockevents_register_device() as that
is called from an upcoming CPU before the CPU is marked online ]
When a new clockevent device is added and replaces a previous device,
the latter is put into the released list. Then the released list is
added back.
This may look counter-intuitive but the reason is that released device
might be suitable for other uses. For example a released CPU regular
clockevent can be a better replacement for the current broadcast event.
Similarly a released broadcast clockevent can be a better replacement
for the current regular clockevent of a given CPU.
Easwar Hariharan [Wed, 30 Oct 2024 17:47:35 +0000 (17:47 +0000)]
jiffies: Define secs_to_jiffies()
secs_to_jiffies() is defined in hci_event.c and cannot be reused by
other call sites. Hoist it into the core code to allow conversion of the
~1150 usages of msecs_to_jiffies() that either:
- use a multiplier value of 1000 or equivalently MSEC_PER_SEC, or
- have timeouts that are denominated in seconds (i.e. end in 000)
It's implemented as a macro to allow usage in static initializers.
This will also allow conversion of yet more sites that use (sec * HZ)
directly, and improve their readability.
Thomas Gleixner [Tue, 1 Oct 2024 08:42:09 +0000 (10:42 +0200)]
posix-timers: Add proper state tracking
Right now the state tracking is done by two struct members:
- it_active:
A boolean which tracks armed/disarmed state
- it_signal_seq:
A sequence counter which is used to invalidate settings
and prevent rearming
Replace it_active with it_status and keep properly track about the states
in one place.
This allows to reuse it_signal_seq to track reprogramming, disarm and
delete operations in order to drop signals which are related to the state
previous of those operations.
Thomas Gleixner [Tue, 1 Oct 2024 08:42:06 +0000 (10:42 +0200)]
posix-timers: Drop signal if timer has been deleted or reprogrammed
No point in delivering a signal from the past. POSIX does not specify the
behaviour here:
- "The effect of disarming or resetting a timer with pending expiration
notifications is unspecified."
- "The disposition of pending signals for the deleted timer is unspecified."
In both cases it is reasonable to expect that pending signals are
discarded. Especially in the reprogramming case it does not make sense to
account for previous overruns or to deliver a signal for a timer which has
been disarmed.
Drop the signal as that is conistent and understandable behaviour.
Thomas Gleixner [Tue, 1 Oct 2024 08:42:04 +0000 (10:42 +0200)]
signal: Allow POSIX timer signals to be dropped
In case that a timer was reprogrammed or deleted an already pending signal
is obsolete. Right now such signals are kept around and eventually
delivered. While POSIX is blury about this:
- "The effect of disarming or resetting a timer with pending expiration
notifications is unspecified."
- "The disposition of pending signals for the deleted timer is
unspecified."
it is reasonable in both cases to expect that pending signals are discarded
as they have no meaning anymore.
Prepare the signal code to allow dropping posix timer signals.
Thomas Gleixner [Tue, 1 Oct 2024 08:42:03 +0000 (10:42 +0200)]
posix-timers: Cure si_sys_private race
The si_sys_private member of the siginfo which is embedded in the
preallocated sigqueue is used by the posix timer code to decide whether a
timer must be reprogrammed on signal delivery.
The handling of this is racy as a long standing comment in that code
documents. It is modified with the timer lock held, but without sighand
lock being held. The actual signal delivery code checks for it under
sighand lock without holding the timer lock.
Hand the new value to send_sigqueue() as argument and store it with sighand
lock held. This is an intermediate change to address this issue.
The arguments to this function will be cleanup in subsequent changes.
Miguel Ojeda [Fri, 25 Oct 2024 11:01:41 +0000 (13:01 +0200)]
time: Fix references to _msecs_to_jiffies() handling of values
The details about the handling of the "normal" values were moved
to the _msecs_to_jiffies() helpers in commit ca42aaf0c861 ("time:
Refactor msecs_to_jiffies"). However, the same commit still mentioned
__msecs_to_jiffies() in the added documentation.
Miguel Ojeda [Fri, 25 Oct 2024 11:01:40 +0000 (13:01 +0200)]
time: Partially revert cleanup on msecs_to_jiffies() documentation
The documentation's intention is to compare msecs_to_jiffies() (first
sentence) with __msecs_to_jiffies() (second sentence), which is what the
original documentation did. One of the cleanups in commit f3cb80804b82
("time: Fix various kernel-doc problems") may have thought the paragraph
was talking about the latter since that is what it is being documented.
timekeeping: Rework do_adjtimex() to use shadow_timekeeper
Updates of the timekeeper can be done by operating on the shadow timekeeper
and afterwards copying the result into the real timekeeper. This has the
advantage, that the sequence count write protected region is kept as small
as possible.
Convert do_adjtimex() to use this scheme and take the opportunity to use a
scoped_guard() for locking.
That requires to have a separate function for updating the leap state so
that the update is protected by the sequence count. This also brings the
timekeeper and the shadow timekeeper in sync for this state, which was not
the case so far. That's not a correctness problem as the state is only used
at the read sides which use the real timekeeper, but it's inconsistent
nevertheless.
timekeeping: Rework timekeeping_suspend() to use shadow_timekeeper
Updates of the timekeeper can be done by operating on the shadow timekeeper
and afterwards copying the result into the real timekeeper. This has the
advantage, that the sequence count write protected region is kept as small
as possible.
While the sequence count held time is not relevant for the resume path as
there is no concurrency, there is no reason to have this function
different than all the other update sites.
Convert timekeeping_inject_offset() to use this scheme and cleanup the
variable declarations while at it.
As halt_fast_timekeeper() does not need protection sequence counter, it is
no problem to move it with this change outside of the sequence counter
protected area. But it still needs to be executed while holding the lock.
timekeeping: Rework timekeeping_resume() to use shadow_timekeeper
Updates of the timekeeper can be done by operating on the shadow timekeeper
and afterwards copying the result into the real timekeeper. This has the
advantage, that the sequence count write protected region is kept as small
as possible.
While the sequence count held time is not relevant for the resume path as
there is no concurrency, there is no reason to have this function
different than all the other update sites.
Convert timekeeping_inject_offset() to use this scheme and cleanup the
variable declaration while at it.
timekeeping: Rework timekeeping_inject_sleeptime64() to use shadow_timekeeper
Updates of the timekeeper can be done by operating on the shadow timekeeper
and afterwards copying the result into the real timekeeper. This has the
advantage, that the sequence count write protected region is kept as small
as possible.
Convert timekeeping_inject_sleeptime64() to use this scheme.
timekeeping: Rework timekeeping_init() to use shadow_timekeeper
For timekeeping_init() the sequence count write held time is not relevant
and it could keep working on the real timekeeper, but there is no reason to
make it different from other timekeeper updates.
timekeeping: Rework change_clocksource() to use shadow_timekeeper
Updates of the timekeeper can be done by operating on the shadow timekeeper
and afterwards copying the result into the real timekeeper. This has the
advantage, that the sequence count write protected region is kept as small
as possible.
timekeeping: Rework timekeeping_inject_offset() to use shadow_timekeeper
Updates of the timekeeper can be done by operating on the shadow timekeeper
and afterwards copying the result into the real timekeeper. This has the
advantage, that the sequence count write protected region is kept as small
as possible.
Convert timekeeping_inject_offset() to use this scheme.
That allows to use a scoped_guard() for locking the timekeeper lock as the
usage of the shadow timekeeper allows a rollback in the error case instead
of the full timekeeper update of the original code.
timekeeping: Rework do_settimeofday64() to use shadow_timekeeper
Updates of the timekeeper can be done by operating on the shadow timekeeper
and afterwards copying the result into the real timekeeper. This has the
advantage, that the sequence count write protected region is kept as small
as possible.
Convert do_settimeofday64() to use this scheme.
That allows to use a scoped_guard() for locking the timekeeper lock as the
usage of the shadow timekeeper allows a rollback in the error case instead
of the full timekeeper update of the original code.
Thomas Gleixner [Wed, 9 Oct 2024 08:29:08 +0000 (10:29 +0200)]
timekeeping: Provide timekeeping_restore_shadow()
Functions which operate on the real timekeeper, e.g. do_settimeofday(),
have error conditions. If they are hit a full timekeeping update is still
required because the already committed operations modified the timekeeper.
When switching these functions to operate on the shadow timekeeper then the
full update can be avoided in the error case, but the modified shadow
timekeeper has to be restored.
timekeeping: Split out timekeeper update of timekeeping_advanced()
timekeeping_advance() is the only optimized function which uses
shadow_timekeeper for updating the real timekeeper to keep the sequence
counter protected region as small as possible.
To be able to transform timekeeper updates in other functions to use the
same logic, split out functionality into a separate function
timekeeper_update_staged().
While at it, document the reason why the sequence counter must be write
held over the call to timekeeping_update() and the copying to the real
timekeeper and why using a pointer based update is suboptimal.
timekeeping: Add struct tk_data as argument to timekeeping_update()
Updates of the timekeeper are done in two ways:
1. Updating timekeeper and afterwards memcpy()'ing the result into
shadow_timekeeper using timekeeping_update(). Used everywhere for
updates except in timekeeping_advance(); the sequence counter protected
region starts before the first change to the timekeeper is done.
2. Updating shadow_timekeeper and then memcpy()'ing the result into
timekeeper. Used only by in timekeeping_advance(); The seqence counter
protected region is only around timekeeping_update() and the memcpy for
copy from shadow to timekeeper.
The second option is fast path optimized. The sequence counter protected
region is as short as possible.
As this behaviour is mainly documented by commit messages, but not in code,
it makes the not easy timekeeping code more complicated to read.
There is no reason why updates to the timekeeper can't use the optimized
version everywhere. With this, the code will be cleaner, as code is reused
instead of duplicated.
To be able to access tk_data which contains all required information, add a
pointer to tk_data as an argument to timekeeping_update(). With that
convert the comment about holding the lock into a lockdep assert.
timekeeping: Introduce tkd_basic_setup() to make lock and seqcount init reusable
Initialization of lock and seqcount needs to be done for every instance of
timekeeper struct. To be able to easily reuse it, create a separate
function for it.
timekeeping: Define a struct type for tk_core to make it reusable
The struct tk_core uses is not reusable. As long as there is only a single
timekeeper, this is not a problem. But when the timekeeper infrastructure
will be reused for per ptp clock timekeepers, an explicit struct type is
required.
Define struct tk_data as explicit struct type for tk_core.
timekeeper_lock protects updates to struct tk_core but is not part of
struct tk_core. As long as there is only a single timekeeper, this is not a
problem. But when the timekeeper infrastructure will be reused for per ptp
clock timekeepers, timekeeper_lock needs to be part of tk_core.
Move the lock into tk_core, move initialisation of the lock and sequence
counter into timekeeping_init() and update all users of timekeeper_lock.
As this is touching all lock sites, convert them to use:
guard(raw_spinlock_irqsave)(&tk_core.lock);
instead of lock/unlock functions whenever possible.
Thomas Gleixner [Wed, 9 Oct 2024 08:29:01 +0000 (10:29 +0200)]
timekeeping: Encapsulate locking/unlocking of timekeeper_lock
timekeeper_lock protects updates of timekeeper (tk_core). It is also used
by vdso_update_begin/end() and not only internally by the timekeeper code.
As long as there is only a single timekeeper, this works fine. But when
the timekeeper infrastructure will be reused for per ptp clock timekeepers,
timekeeper_lock needs to be part of tk_core..
Therefore encapuslate locking/unlocking of timekeeper_lock and make the
lock static.
Thomas Gleixner [Wed, 9 Oct 2024 08:29:00 +0000 (10:29 +0200)]
timekeeping: Move shadow_timekeeper into tk_core
tk_core requires shadow_timekeeper to allow timekeeping_advance() updating
without holding the timekeeper sequence count write locked. This allows the
readers to make progress up to the actual update where the shadow
timekeeper is copied over to the real timekeeper.
As long as there is only a single timekeeper, having them separate is
fine. But when the timekeeper infrastructure will be reused for per ptp
clock timekeepers, shadow_timekeeper needs to be part of tk_core.
So any access to via ktime_get*() except for access to CLOCK_MONOTONIC_RAW
will use either cachelines 0 + 1 or cachelines 0 + 2. Access to
CLOCK_MONOTONIC_RAW uses cachelines 0 + 1 + 3.
Reorder the members so that the result is more efficient:
Thomas Gleixner [Wed, 9 Oct 2024 08:28:58 +0000 (10:28 +0200)]
timekeeping: Simplify code in timekeeping_advance()
timekeeping_advance() takes the timekeeper_lock and releases it before
returning. When an early return is required, goto statements are used to
make sure the lock is realeased properly. When the code was written the
locking guard() was not yet available.
Use the guard() to simplify the code and while at it cleanup ordering of
function variables. No functional change.
Thomas Gleixner [Wed, 9 Oct 2024 08:28:55 +0000 (10:28 +0200)]
timekeeping: Don't stop time readers across hard_pps() update
hard_pps() update does not modify anything which might be required by time
readers so forcing readers out of the way during the update is a pointless
exercise.
The interaction with adjtimex() and timekeeper updates which call into the
NTP code is properly serialized by timekeeper_lock.
Julia Lawall [Sun, 13 Oct 2024 20:16:58 +0000 (22:16 +0200)]
posix-timers: Replace call_rcu() by kfree_rcu() for simple kmem_cache_free() callback
Since SLOB was removed and since commit 6c6c47b063b5 ("mm, slab: call
kvfree_rcu_barrier() from kmem_cache_destroy()"), it is not longer
necessary to use call_rcu() when the callback only performs
kmem_cache_free(). Use kfree_rcu() directly.
The documentation which tries to give advices how to properly inserting
delays or sleeps is outdated. The file name is 'timers-howto.rst' which
might be misleading as it is only about delay and sleep mechanisms and not
how to use timers.
Update the documentation by integrating the important parts from the
related function descriptions and move it all into a self explaining file
with the name "delay_sleep_functions.rst".
anysee driver was transformed to use usbv2 years ago. The comments in
anysee_ctrl_msg() still are referencing the old interfaces where msleep()
was used. The v2 interfaces also changed over the years and with commit 1162c7b383a6 ("[media] dvb_usb_v2: refactor dvb_usbv2_generic_rw()") the
usage of msleep() was gone anyway.
Remove FIXME comment and update also comment before call to
dvb_usbv2_generic_rw_locked().
powerpc/rtas: Use fsleep() to minimize additional sleep duration
When commit 38f7b7067dae ("powerpc/rtas: rtas_busy_delay() improvements")
was introduced, documentation about proper usage of sleep related functions
was outdated.
The commit message references the usage of a HZ=100 system. When using a
20ms sleep duration on such a system and therefore using msleep(), the
possible additional slack will be +10ms.
When the system is configured with HZ=100 the granularity of a jiffy and of
a bucket of the lowest timer wheel level is 10ms. To make sure a timer will
not expire early (when queueing of the timer races with an concurrent
update of jiffies), timers are always queued into the next bucket. This is
the reason for the maximal possible slack of 10ms.
fsleep() limits the maximal possible slack to 25% by making threshold
between usleep_range() and msleep() HZ dependent. As soon as the accuracy
of msleep() is sufficient, the less expensive timer list timer based
sleeping function is used instead of the more expensive hrtimer based
usleep_range() function. The udelay() will not be used in this specific
usecase as the lowest sleep length is larger than 1 millisecond.
Use fsleep() directly instead of using an own heuristic for the best
sleeping mechanism to use.
Function descriptions in iopoll.h, regmap.h, phy.h and sound/soc/sof/ops.h
copied all the same outdated documentation about sleep/delay function
limitations. In those comments, the generic (and still outdated) timer
documentation file is referenced.
As proper function descriptions for used delay and sleep functions are in
place, simply update the descriptions to reference to them. While at it fix
missing colon after "Returns" in function description and move return value
description to the end of the function description.
checkpatch: Remove links to outdated documentation
checkpatch.pl checks for several things related to sleep and delay
functions. In all warnings the outdated documentation is referenced. Also
in checkpatch kernel documentation the outdated documentation is
referenced.
Replace the links to the outdated documentation with links to the function
description.
Note: Update of the outdated checkpatch checks is done in a second step.
timers: Add a warning to usleep_range_state() for wrong order of arguments
There is a warning in checkpatch script that triggers, when min and max
arguments of usleep_range_state() are in reverse order. This check does
only cover callsites which uses constants. Add this check into the code as
a WARN_ON_ONCE() to also cover callsites not using constants and fix the
mis-usage by resetting the delta to 0.
mm/damon/core: Use generic upper bound recommondation for usleep_range()
The upper bound for usleep_range_idle() was taken from the outdated
documentation. As a recommondation for the upper bound of usleep_range()
depends on HZ configuration it is not possible to hard code it.
Use the define "USLEEP_RANGE_UPPER_BOUND" instead.
fsleep() simply implements the recommendations of the outdated
documentation in "Documentation/timers/timers-howto.rst". This should be a
user friendly interface to choose always the best timeout function
approach:
- udelay() for very short sleep durations shorter than 10 microseconds
- usleep_range() for sleep durations until 20 milliseconds
- msleep() for the others
The actual implementation has several problems:
- It does not take into account that HZ resolution also has an impact on
granularity of jiffies and has also an impact on the granularity of the
buckets of timer wheel levels. This means that accuracy for the timeout
does not have an upper limit. When executing fsleep(20000) on a HZ=100
system, the possible additional slack will be 50% as the granularity of
the buckets in the lowest level is 10 milliseconds.
- The upper limit of usleep_range() is twice the requested timeout. When no
other interrupts occur in this range, the maximum value is used. This
means that the requested sleep length has then an additional delay of
100%.
Change the thresholds for the decisions in fsleep() to make sure the
maximum slack which is added to the sleep duration is 25%.
Note: Outdated documentation will be updated in a followup patch.
udelay() as well as ndelay() are defines and no functions and are using
constants to be able to transform a sleep time into loops and to prevent
too long udelays/ndelays. There was a compiler error with non-const 8 bit
arguments which was fixed by commit a87e553fabe8 ("asm-generic: delay.h fix
udelay and ndelay for 8 bit args"). When using a function, the non-const 8
bit argument is type casted and the problem would be gone.
Transform udelay() and ndelay() into proper functions, remove the no longer
and confusing division, add defines for the magic values and add some
explanations as well.
timers: Update function descriptions of sleep/delay related functions
A lot of commonly used functions for inserting a sleep or delay lack a
proper function description. Add function descriptions to all of them to
have important information in a central place close to the code.
timers: Rename usleep_idle_range() to usleep_range_idle()
usleep_idle_range() is a variant of usleep_range(). Both are using
usleep_range_state() as a base. To be able to find all the related
functions in one go, rename it usleep_idle_range() to usleep_range_idle().
timers: Update schedule_[hr]timeout*() related function descriptions
schedule_timeout*() functions do not have proper kernel-doc formatted
function descriptions. schedule_hrtimeout() and schedule_hrtimeout_range()
have a almost identical description.
Add missing function descriptions. Remove copy of function description and
add a pointer to the existing description instead.
timers: Move *sleep*() and timeout functions into a separate file
All schedule_timeout() and *sleep*() related functions are interfaces on
top of timer list timers and hrtimers to add a sleep to the code. As they
are built on top of the timer list timers and hrtimers, the [hr]timer
interfaces are already used except when queuing the timer in
schedule_timeout(). But there exists the appropriate interface add_timer()
which does the same job with an extra check for an already pending timer.
Split all those functions as they are into a separate file and use
add_timer() instead of __mod_timer() in schedule_timeout().
While at it fix minor formatting issues and a multi line printk function
call in schedule_timeout().
Jeff Layton [Wed, 2 Oct 2024 21:27:17 +0000 (17:27 -0400)]
timekeeping: Add percpu counter for tracking floor swap events
The mgtime_floor value is a global variable for tracking the latest
fine-grained timestamp handed out. Because it's a global, track the
number of times that a new floor value is assigned.
Add a new percpu counter to the timekeeping code to track the number of
floor swap events that have occurred. A later patch will add a debugfs
file to display this counter alongside other stats involving multigrain
timestamps.
Jeff Layton [Wed, 2 Oct 2024 21:27:16 +0000 (17:27 -0400)]
timekeeping: Add interfaces for handling timestamps with a floor value
Multigrain timestamps allow the kernel to use fine-grained timestamps when
an inode's attributes is being actively observed via ->getattr(). With
this support, it's possible for a file to get a fine-grained timestamp, and
another modified after it to get a coarse-grained stamp that is earlier
than the fine-grained time. If this happens then the files can appear to
have been modified in reverse order, which breaks VFS ordering guarantees
[1].
To prevent this, maintain a floor value for multigrain timestamps.
Whenever a fine-grained timestamp is handed out, record it, and when later
coarse-grained stamps are handed out, ensure they are not earlier than that
value. If the coarse-grained timestamp is earlier than the fine-grained
floor, return the floor value instead.
Add a static singleton atomic64_t into timekeeper.c that is used to keep
track of the latest fine-grained time ever handed out. This is tracked as a
monotonic ktime_t value to ensure that it isn't affected by clock
jumps. Because it is updated at different times than the rest of the
timekeeper object, the floor value is managed independently of the
timekeeper via a cmpxchg() operation, and sits on its own cacheline.
Add two new public interfaces:
- ktime_get_coarse_real_ts64_mg() fills a timespec64 with the later of the
coarse-grained clock and the floor time
- ktime_get_real_ts64_mg() gets the fine-grained clock value, and tries
to swap it into the floor. A timespec64 is filled with the result.
The floor value is global and updated via a single try_cmpxchg(). If
that fails then the operation raced with a concurrent update. Any
concurrent update must be later than the existing floor value, so any
racing tasks can accept any resulting floor value without retrying.
[1]: POSIX requires that files be stamped with realtime clock values, and
makes no provision for dealing with backward clock jumps. If a backward
realtime clock jump occurs, then files can appear to have been modified
in reverse order.
Jeff Layton [Tue, 10 Sep 2024 17:43:34 +0000 (13:43 -0400)]
timekeeping: Don't use seqcount loop in ktime_mono_to_any() on 64-bit systems
ktime_mono_to_any() only fetches the offset inside the loop. This is a
single word on 64-bit CPUs, and seqcount_read_begin() implies a full SMP
barrier.
Use READ_ONCE() to fetch the offset instead of doing a seqcount loop on
64-bit and add the matching WRITE_ONCE()'s to update the offsets in
tk_set_wall_to_mono() and tk_update_sleep_time().
timekeeping: Add the boot clock to system time snapshot
For tracing purpose, the boot clock is interesting as it doesn't stop on
suspend. Export it as part of the time snapshot. This will later allow
the hypervisor to add boot clock timestamps to its events.
Thomas Gleixner [Wed, 11 Sep 2024 13:17:43 +0000 (15:17 +0200)]
ntp: Introduce struct ntp_data
All NTP data is held in static variables. That prevents the NTP code from
being reuasble for non-system time timekeepers, e.g. per PTP clock
timekeeping.
Introduce struct ntp_data and move tick_usec into it for a start.
Thomas Gleixner [Wed, 11 Sep 2024 13:17:42 +0000 (15:17 +0200)]
ntp: Read reference time only once
The reference time is required twice in ntp_update_offset(). It will not
change in the meantime as the calling code holds the timekeeper lock. Read
it only once and store it into a local variable.
Thomas Gleixner [Wed, 11 Sep 2024 13:17:41 +0000 (15:17 +0200)]
ntp: Convert functions with only two states to bool
is_error_status() and ntp_synced() return whether a state is set or
not. Both functions use unsigned int for it even if it would be a perfect
job for a bool.
Use bool instead of unsigned int. And while at it, move ntp_synced()
function to the place where it is used.
Code is partially formatted in a creative way which makes reading
harder. Examples are function calls over several lines where the
indentation does not start at the same height then the open bracket after
the function name.
Improve formatting but do not make a functional change.
Thomas Gleixner [Wed, 11 Sep 2024 13:17:39 +0000 (15:17 +0200)]
ntp: Clean up comments
Usage of different comment formatting makes fast reading and parsing the
code harder. There are several multi-line comments which do not follow the
coding style by starting with a line only containing '/*'. There are also
comments which do not start with capitals.
Clean up all those comments to be consistent and remove comments which
document the obvious.
leading to a build error when neither KVM_INTEL nor KVM_AMD support is
enabled:
arch/x86/kvm/x86.c: In function ‘kvm_arch_enable_virtualization’:
arch/x86/kvm/x86.c:12517:9: error: implicit declaration of function ‘cpu_emergency_register_virt_callback’ [-Wimplicit-function-declaration]
12517 | cpu_emergency_register_virt_callback(kvm_x86_ops.emergency_disable_virtualization_cpu);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
arch/x86/kvm/x86.c: In function ‘kvm_arch_disable_virtualization’:
arch/x86/kvm/x86.c:12522:9: error: implicit declaration of function ‘cpu_emergency_unregister_virt_callback’ [-Wimplicit-function-declaration]
12522 | cpu_emergency_unregister_virt_callback(kvm_x86_ops.emergency_disable_virtualization_cpu);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fix the build by defining empty helper functions the same way the old
cpu_emergency_disable_virtualization() function was dealt with for the
same situation.
Maybe we could instead have made the call sites conditional, since the
callers (kvm_arch_{en,dis}able_virtualization()) have an empty weak
fallback. I'll leave that to the kvm people to argue about, this at
least gets the build going for that particular config.
Fixes: 590b09b1d88e ("KVM: x86: Register "emergency disable" callbacks when virt is enabled") Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Sean Christopherson <seanjc@google.com> Cc: Kai Huang <kai.huang@intel.com> Cc: Chao Gao <chao.gao@intel.com> Cc: Farrah Chen <farrah.chen@intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
projects / thirdparty / kernel / linux.git / log
