Guenter reported boot stalls on a emulated ARM 32-bit platform, which has a
24-bit wide clocksource.
It turns out that the calculated maximal idle time, which limits idle
sleeps to prevent clocksource wrap arounds, is close to the point where the
negative motion detection triggers.
max_idle_ns: 597268854 ns
negative motion tripping point: 671088640 ns
If the idle wakeup is delayed beyond that point, the clocksource
advances far enough to trigger the negative motion detection. This
prevents the clock to advance and in the worst case the system stalls
completely if the consecutive sleeps based on the stale clock are
delayed as well.
Cure this by calculating a more robust cut-off value for negative motion,
which covers 87.5% of the actual clocksource counter width. Compare the
delta against this value to catch negative motion. This is specifically for
clock sources with a small counter width as their wrap around time is close
to the half counter width. For clock sources with wide counters this is not
a problem because the maximum idle time is far from the half counter width
due to the math overflow protection constraints.
For the case at hand this results in a tripping point of 1174405120ns.
Note, that this cannot prevent issues when the delay exceeds the 87.5%
margin, but that's not different from the previous unchecked version which
allowed arbitrary time jumps.
Systems with small counter width are prone to invalid results, but this
problem is unlikely to be seen on real hardware. If such a system
completely stalls for more than half a second, then there are other more
urgent problems than the counter wrapping around.
Fixes: c163e40af9 ("timekeeping: Always check for negative motion")
Reported-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Link: https://lore.kernel.org/all/8734j5ul4x.ffs@tglx
Closes: https://lore.kernel.org/all/387b120b-d68a-45e8-b6ab-768cd95d11c2@roeck-us.net
Pull timer updates from Thomas Gleixner:
"A rather large update for timekeeping and timers:
- The final step to get rid of auto-rearming posix-timers
posix-timers are currently auto-rearmed by the kernel when the
signal of the timer is ignored so that the timer signal can be
delivered once the corresponding signal is unignored.
This requires to throttle the timer to prevent a DoS by small
intervals and keeps the system pointlessly out of low power states
for no value. This is a long standing non-trivial problem due to
the lock order of posix-timer lock and the sighand lock along with
life time issues as the timer and the sigqueue have different life
time rules.
Cure this by:
- Embedding the sigqueue into the timer struct to have the same
life time rules. Aside of that this also avoids the lookup of
the timer in the signal delivery and rearm path as it's just a
always valid container_of() now.
- Queuing ignored timer signals onto a seperate ignored list.
- Moving queued timer signals onto the ignored list when the
signal is switched to SIG_IGN before it could be delivered.
- Walking the ignored list when SIG_IGN is lifted and requeue the
signals to the actual signal lists. This allows the signal
delivery code to rearm the timer.
This also required to consolidate the signal delivery rules so they
are consistent across all situations. With that all self test
scenarios finally succeed.
- Core infrastructure for VFS multigrain timestamping
This is required to allow the kernel to use coarse grained time
stamps by default and switch to fine grained time stamps when inode
attributes are actively observed via getattr().
These changes have been provided to the VFS tree as well, so that
the VFS specific infrastructure could be built on top.
- Cleanup and consolidation of the sleep() infrastructure
- Move all sleep and timeout functions into one file
- Rework udelay() and ndelay() into proper documented inline
functions and replace the hardcoded magic numbers by proper
defines.
- Rework the fsleep() implementation to take the reality of the
timer wheel granularity on different HZ values into account.
Right now the boundaries are hard coded time ranges which fail
to provide the requested accuracy on different HZ settings.
- Update documentation for all sleep/timeout related functions
and fix up stale documentation links all over the place
- Fixup a few usage sites
- Rework of timekeeping and adjtimex(2) to prepare for multiple PTP
clocks
A system can have multiple PTP clocks which are participating in
seperate and independent PTP clock domains. So far the kernel only
considers the PTP clock which is based on CLOCK TAI relevant as
that's the clock which drives the timekeeping adjustments via the
various user space daemons through adjtimex(2).
The non TAI based clock domains are accessible via the file
descriptor based posix clocks, but their usability is very limited.
They can't be accessed fast as they always go all the way out to
the hardware and they cannot be utilized in the kernel itself.
As Time Sensitive Networking (TSN) gains traction it is required to
provide fast user and kernel space access to these clocks.
The approach taken is to utilize the timekeeping and adjtimex(2)
infrastructure to provide this access in a similar way how the
kernel provides access to clock MONOTONIC, REALTIME etc.
Instead of creating a duplicated infrastructure this rework
converts timekeeping and adjtimex(2) into generic functionality
which operates on pointers to data structures instead of using
static variables.
This allows to provide time accessors and adjtimex(2) functionality
for the independent PTP clocks in a subsequent step.
- Consolidate hrtimer initialization
hrtimers are set up by initializing the data structure and then
seperately setting the callback function for historical reasons.
That's an extra unnecessary step and makes Rust support less
straight forward than it should be.
Provide a new set of hrtimer_setup*() functions and convert the
core code and a few usage sites of the less frequently used
interfaces over.
The bulk of the htimer_init() to hrtimer_setup() conversion is
already prepared and scheduled for the next merge window.
- Drivers:
- Ensure that the global timekeeping clocksource is utilizing the
cluster 0 timer on MIPS multi-cluster systems.
Otherwise CPUs on different clusters use their cluster specific
clocksource which is not guaranteed to be synchronized with
other clusters.
- Mostly boring cleanups, fixes, improvements and code movement"
* tag 'timers-core-2024-11-18' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (140 commits)
posix-timers: Fix spurious warning on double enqueue versus do_exit()
clocksource/drivers/arm_arch_timer: Use of_property_present() for non-boolean properties
clocksource/drivers/gpx: Remove redundant casts
clocksource/drivers/timer-ti-dm: Fix child node refcount handling
dt-bindings: timer: actions,owl-timer: convert to YAML
clocksource/drivers/ralink: Add Ralink System Tick Counter driver
clocksource/drivers/mips-gic-timer: Always use cluster 0 counter as clocksource
clocksource/drivers/timer-ti-dm: Don't fail probe if int not found
clocksource/drivers:sp804: Make user selectable
clocksource/drivers/dw_apb: Remove unused dw_apb_clockevent functions
hrtimers: Delete hrtimer_init_on_stack()
alarmtimer: Switch to use hrtimer_setup() and hrtimer_setup_on_stack()
io_uring: Switch to use hrtimer_setup_on_stack()
sched/idle: Switch to use hrtimer_setup_on_stack()
hrtimers: Delete hrtimer_init_sleeper_on_stack()
wait: Switch to use hrtimer_setup_sleeper_on_stack()
timers: Switch to use hrtimer_setup_sleeper_on_stack()
net: pktgen: Switch to use hrtimer_setup_sleeper_on_stack()
futex: Switch to use hrtimer_setup_sleeper_on_stack()
fs/aio: Switch to use hrtimer_setup_sleeper_on_stack()
...
Since 135225a363 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.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241031120328.536010148@linutronix.de
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.
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-23-554456a44a15@linutronix.de
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.
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-22-554456a44a15@linutronix.de
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.
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-21-554456a44a15@linutronix.de
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.
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-17-554456a44a15@linutronix.de
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.
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-16-554456a44a15@linutronix.de
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.
Provide a helper function for that.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-15-554456a44a15@linutronix.de
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.
No functional change.
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-13-554456a44a15@linutronix.de
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.
No functional change.
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-12-554456a44a15@linutronix.de
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.
Suggested-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-9-554456a44a15@linutronix.de
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.
No functional change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: John Stultz <jstultz@google.com>
Link: https://lore.kernel.org/all/20241009-devel-anna-maria-b4-timers-ptp-timekeeping-v2-7-554456a44a15@linutronix.de
