Description: Adds ssleep_interruptible() to allow longer delays to occur in TASK_INTERRUPTIBLE, similarly to ssleep(). To be consistent with msleep_interruptible(), ssleep_interruptible() returns the
Description: Adds ssleep_interruptible() to allow longer delays to occur in TASK_INTERRUPTIBLE, similarly to ssleep(). To be consistent with msleep_interruptible(), ssleep_interruptible() returns the
After a discussion on IRC, I believe it is pretty clear that this function has serious issues. Mainly, that if I request a delay of 1 second, but msleep_interruptible() returns after 1 millisecond, t
Would making sure that the time slept was always rounded up to the nearest second resolve this problem. I believe that rounding up is a common approach to resolving this type of problem when changing
This is a good idea, but I have two issues: 1) A major reason for having msecs_to_jiffies() and co. is to avoid having to do this type of conversion ourselves. A weak argument, admittedly, but just s
If it is unacceptable to neither underestimate or overestimate the duration of a sleep to the nearest second (the unit of granularity of the sleep in this case) then I agree. That is unless you want
This is kind of my position. Overestimating leads to the potential, if a loop is used by the caller, of never leaving the loop, e.g. timeout = 1; while (timeout) { timeout = ssleep_interruptible(time