user/sven/linux.git/include/linux/timer.h, branch v5.5.8

timers: Prepare support for PREEMPT_RT

2019-08-01T18:51:22Z

When PREEMPT_RT is enabled, the soft interrupt thread can be preempted. If the soft interrupt thread is preempted in the middle of a timer callback, then calling del_timer_sync() can lead to two issues: - If the caller is on a remote CPU then it has to spin wait for the timer handler to complete. This can result in unbound priority inversion. - If the caller originates from the task which preempted the timer handler on the same CPU, then spin waiting for the timer handler to complete is never going to end. To avoid these issues, add a new lock to the timer base which is held around the execution of the timer callbacks. If del_timer_sync() detects that the timer callback is currently running, it blocks on the expiry lock. When the callback is finished, the expiry lock is dropped by the softirq thread which wakes up the waiter and the system makes progress. This addresses both the priority inversion and the life lock issues. This mechanism is not used for timers which are marked IRQSAFE as for those preemption is disabled accross the callback and therefore this situation cannot happen. The callbacks for such timers need to be individually audited for RT compliance. The same issue can happen in virtual machines when the vCPU which runs a timer callback is scheduled out. If a second vCPU of the same guest calls del_timer_sync() it will spin wait for the other vCPU to be scheduled back in. The expiry lock mechanism would avoid that. It'd be trivial to enable this when paravirt spinlocks are enabled in a guest, but it's not clear whether this is an actual problem in the wild, so for now it's an RT only mechanism. As the softirq thread can be preempted with PREEMPT_RT=y, the SMP variant of del_timer_sync() needs to be used on UP as well. [ tglx: Refactored it for mainline ] Signed-off-by: Anna-Maria Gleixner Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Thomas Gleixner Acked-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20190726185753.832418500@linutronix.de

timer: Document TIMER_PINNED

2019-07-07T09:50:03Z

The flag hints the user that the pinned timers will always be run on a static CPU (because that should be what "pinned" means...) but that's not the truth, at least with the current implementation. For example, currently if a pinned timer is set up but later mod_timer() upon the pinned timer is invoked, mod_timer() will still try to queue the timer on the current processor and migrate the timer if necessary. Document it a bit with the definition of TIMER_PINNED so that all future users will use it correctly. Signed-off-by: Peter Xu Signed-off-by: Thomas Gleixner Cc: Marcelo Tosatti Cc: Luiz Capitulino Link: https://lkml.kernel.org/r/20190628105942.14131-1-peterx@redhat.com

timers: Remove stale struct tvec_base forward declaration

2018-04-17T15:18:04Z

struct tvec_base is a leftover of the original timer wheel implementation and not longer used. Remove the forward declaration. Signed-off-by: Liu Changcheng Signed-off-by: Thomas Gleixner Cc: akpm@linux-foundation.org Link: https://lkml.kernel.org/r/20180412075701.GA38952@sofia

timers: Reinitialize per cpu bases on hotplug

2017-12-29T22:13:09Z

The timer wheel bases are not (re)initialized on CPU hotplug. That leaves them with a potentially stale clk and next_expiry valuem, which can cause trouble then the CPU is plugged. Add a prepare callback which forwards the clock, sets next_expiry to far in the future and reset the control flags to a known state. Set base->must_forward_clk so the first timer which is queued will try to forward the clock to current jiffies. Fixes: 500462a9de65 ("timers: Switch to a non-cascading wheel") Reported-by: Paul E. McKenney Signed-off-by: Thomas Gleixner Cc: Peter Zijlstra Cc: Frederic Weisbecker Cc: Sebastian Siewior Cc: Anna-Maria Gleixner Cc: stable@vger.kernel.org Link: https://lkml.kernel.org/r/alpine.DEB.2.20.1712272152200.2431@nanos

treewide: Remove TIMER_FUNC_TYPE and TIMER_DATA_TYPE casts

2017-11-22T00:35:54Z

With all callbacks converted, and the timer callback prototype switched over, the TIMER_FUNC_TYPE cast is no longer needed, so remove it. Conversion was done with the following scripts: perl -pi -e 's|$TIMER_FUNC_TYPE$||g' \ $(git grep TIMER_FUNC_TYPE | cut -d: -f1 | sort -u) perl -pi -e 's|$TIMER_DATA_TYPE$||g' \ $(git grep TIMER_DATA_TYPE | cut -d: -f1 | sort -u) The now unused macros are also dropped from include/linux/timer.h. Signed-off-by: Kees Cook

timer: Remove redundant __setup_timer*() macros

2017-11-21T23:57:15Z

With __init_timer*() now matching __setup_timer*(), remove the redundant internal interface, clean up the resulting definitions and add more documentation. Cc: Thomas Gleixner Cc: Tejun Heo Cc: Lai Jiangshan Cc: Shaohua Li Cc: Jens Axboe Cc: Andrew Morton Signed-off-by: Kees Cook

timer: Pass function down to initialization routines

2017-11-21T23:57:14Z

In preparation for removing more macros, pass the function down to the initialization routines instead of doing it in macros. Cc: Thomas Gleixner Cc: John Stultz Cc: Stephen Boyd Signed-off-by: Kees Cook

timer: Remove unused data arguments from macros

2017-11-21T23:57:14Z

With the .data field removed, the ignored data arguments in timer macros can be removed. Cc: Thomas Gleixner Cc: Tejun Heo Cc: Lai Jiangshan Cc: Jens Axboe Cc: Andrew Morton Cc: Shaohua Li Signed-off-by: Kees Cook

timer: Switch callback prototype to take struct timer_list * argument

2017-11-21T23:57:13Z

Since all callbacks have been converted, we can switch the core prototype to "struct timer_list *" now too. Cc: Thomas Gleixner Cc: John Stultz Cc: Stephen Boyd Signed-off-by: Kees Cook

timer: Pass timer_list pointer to callbacks unconditionally

2017-11-21T23:57:12Z

Now that all timer callbacks are already taking their struct timer_list pointer as the callback argument, just do this unconditionally and remove the .data field. Cc: Thomas Gleixner Cc: John Stultz Cc: Stephen Boyd Signed-off-by: Kees Cook