CLOCK_MONOTONIC in Timers functions

[Yaakov (Cygwin/X)]

POSIX requires the Timers functions (timer_create, timer_[gs]ettime,
etc.) to also support CLOCK_MONOTONIC if the _POSIX_MONOTONIC_CLOCK
option is available. I'm willing to take a look at making this happen
when I have the chance. (Getting gcc 4.5 out is my first priority now.)

I see that timer.cc uses its own WaitForSingleObject calls instead of
using cancelable_wait(). Is there a good reason for this, or could this
be rewritten in terms of the latter?


Yaakov

[Corinna Vinschen]

Post by Yaakov (Cygwin/X)
POSIX requires the Timers functions (timer_create, timer_[gs]ettime,
etc.) to also support CLOCK_MONOTONIC if the _POSIX_MONOTONIC_CLOCK
option is available. I'm willing to take a look at making this happen
when I have the chance. (Getting gcc 4.5 out is my first priority now.)
I see that timer.cc uses its own WaitForSingleObject calls instead of
using cancelable_wait(). Is there a good reason for this, or could this
be rewritten in terms of the latter?

I don't see a reason to stick to WFMO if using cancelable_wait allows
to extend the timer functions as you propose.


Corinna

[Christopher Faylor]

Post by Corinna Vinschen

Post by Yaakov (Cygwin/X)
POSIX requires the Timers functions (timer_create, timer_[gs]ettime,
etc.) to also support CLOCK_MONOTONIC if the _POSIX_MONOTONIC_CLOCK
option is available. I'm willing to take a look at making this happen
when I have the chance. (Getting gcc 4.5 out is my first priority now.)
I see that timer.cc uses its own WaitForSingleObject calls instead of
using cancelable_wait(). Is there a good reason for this, or could this
be rewritten in terms of the latter?

I don't see a reason to stick to WFMO if using cancelable_wait allows
to extend the timer functions as you propose.

Ditto.

cgf