TimerSettime(), TimerSettime_r()
Arm a timer with the given expiration time, disarm it, or set its tolerance
Synopsis:
#include <sys/neutrino.h>
int TimerSettime( timer_t id,
int flags,
const struct _itimer * itime,
struct _itimer * oitime );
int TimerSettime_r( timer_t id,
int flags,
const struct _itimer * itime,
struct _itimer * oitime );
Arguments:
- id
- The ID of the timer that you want to modify, as returned by TimerCreate().
- flags
- Flags that specify what you want to set for the timer; at most one of the following:
- TIMER_ABSTIME — set an absolute expiration time.
- TIMER_TOLERANCE — specify the amount of the tolerance to allow the kernel in low-power situations.
If you don't set either of these flags, the function sets a relative expiration time. For more information, see below.
If you're setting an expiration time (i.e., you specify TIMER_ABSTIME or neither bit), you can OR in TIMER_PRECISE to override any default tolerance that was set for the process (see procmgr_timer_tolerance()).
- itime
- A pointer to an _itimer structure that specifies the expiration time or tolerance, depending on the flag settings. If you specify TIMER_TOLERANCE for that other argument, itime can be NULL.
- oitime
- NULL, or a pointer to an _itimer structure where the function can store the previous data.
Library:
libc
Use the -l c option to qcc to link against this library. This library is usually included automatically.
Description:
The TimerSettime() and TimerSettime_r() functions either set the expiration time of the timer specified by id, or set the tolerance for that timer, depending on the flags provided.
These functions are identical except in the way they indicate errors. See the Returns section for details.
The _itimer structure contains at least the following members:
- uint64_t nsec
- uint64_t interval_nsec
The meaning of these fields depends on the flags specified.
Expiration date
If you specify the TIMER_ABSTIME flag, then the nsec member represents an absolute expiration date, in nanoseconds from the start of the epoch of the source clock. If the specified date has already passed, then the expiration event is delivered immediately. For CLOCK_REALTIME, the start of the epoch is 00:00:00 January 1, 1970 UTC. For CLOCK_MONOTONIC, it's the kernel initialization time. For a CPU-time clock for a process or thread, it's the startup time of that execution unit.
If you don't specify TIMER_ABSTIME or TIMER_TOLERANCE, nsec represents a relative expiration period that's offset from the given clock's current system time, in nanoseconds.
If nsec is zero, then the timer is disarmed.
If interval_nsec is nonzero, then it specifies a repeat rate which is added to the timer once the nsec period has expired. Subsequently, the timer is automatically rearmed, causing it to become repetitive with a period of interval_nsec.
If the timer is already armed when you call TimerSettime(), this call discards the previous setting and sets a new setting.
If the event notification specified by TimerCreate() has a sigev_code of SI_TIMER, then at most one event is queued. In this case, if an event is pending from a previous timer when the timer fires again, a timer overrun occurs. You can use the TimerInfo() kernel call to obtain the number of overruns that have occurred on this timer.
Understanding the Microkernel's Concept of Timechapter of the QNX OS Programmer's Guide.
If the oitime argument isn't NULL, the function fills in the structure that it points to with the interval timer period (i.e., the previous amount of time left before the timer was to have expired), or zero if the timer was disarmed at the time of the call. The previous repeat rate is stored in the interval_nsec member.
Timer tolerance
If you specify the TIMER_TOLERANCE flag, then:
- If itime isn't NULL, the nsec member specifies the
amount of tolerance, in nanoseconds, that the kernel is allowed with regard to the expiry time.
You can specify a large value (e.g.,
~0ULL) for infinite tolerance. - If oitime isn't NULL, the function sets the nsec member to the previous tolerance.
In order to set the tolerance to a value between 0 and the clock period, you need to have the PROCMGR_AID_HIGH_RESOLUTION_TIMER ability enabled. For more information, see procmgr_ability().
In this release, you can set the tolerance for only an inactive timer,
that is, if you haven't called TimerSettime() with a non-zero expiration time
and without the TIMER_TOLERANCE flag.
If you attempt to set the tolerance after such a call, this function fails with
EBUSY. For more information, see High-resolution timers
in the Understanding the Microkernel's Concept of Time
chapter of the QNX OS Programmer's Guide.
Blocking states
This call doesn't block.
Returns:
The only difference between these functions is the way they indicate errors:
- TimerSettime()
- If an error occurs, -1 is returned and errno is set. Any other value returned indicates success.
- TimerSettime_r()
- EOK is returned on success. This function does NOT set errno. If an error occurs, any value in the Errors section may be returned.
Errors:
- EBUSY
- An attempt was made to set the tolerance of the timer timerid after it was armed.
- EFAULT
- A fault occurred when the kernel tried to access itime or oitime.
- EINVAL
- One of the following is true:
- The timer specified by id doesn't exist.
- The timer specified by id was for a CPU-time clock and TIMER_TOLERANCE was passed in flags.
- EOVERFLOW
- The sum of the current time plus the provided relative time results in a value that exceeds UINT64_MAX.
- EPERM
- The calling process doesn't have the required permission; see procmgr_ability().
Classification:
| Safety: | |
|---|---|
| Cancellation point | No |
| Signal handler | Yes |
| Thread | Yes |