Re: 2.2.16/17pre6 nanosleep stops my X pointer from moving

[George Anzinger]

Not to throw cold water on this, but as a real time professional, I
think timing loops like this should be removed from the kernel.  A FAR
better approach would be to set up an interrupt, even if it is sloppy. 
Those who need real time understand this and would rather have the
cycles than the accuracy.  If it is a real issue they can use an
interface that explititly busy waits.  At least this is clear when used.

nanosleep() should bridge the gap from sleep() (1 sec.) to jiffy with
further refinement only if supported by the hardware as an interrupt.

George

Alan Cox wrote:
> 
> > Long calls to nanosleep cause my X pointer to stop moving.
> > nanosleep({0, 1000000}, NULL)           = 0
> >
> > Is this the expected behavior?
> 
> If the task is real time then yes the kernel will choose to do a busy
> loop for accuracy.
> 
> Alan
> 
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [e-mail]
> Please read the FAQ at http://www.tux.org/lkml/