Clarification on nanosecond() sleep in linux

Question

0.00/5 (No votes)

See more:

Using nanosecond sleep , will i able to sleep for 1 nanosecond?.
Is that much finer timer granularity is supported on linux?
Mostly timer granularity will be 10mSec(scheduler tick), so my understanding is that it is not possible to achieve timer granularity less than 10mSec.
can someone clarify please?

If suppose finer resolution is supported , my task will be put into sleep or kernel will do busy loop on behalf of my task to pass time(say 100 microsecond or 500 nanosecond)?

Can someone please clarify?

What I have tried:

If suppose finer resolution is supported , my task will be put into sleep or kernel will do busy loop on behalf of my task to pass time(say 100 microsecond or 500 nanosecond)?

Can someone please clarify?

Posted 22-Jul-16 2:37am

chandrAN2&

Updated 22-Jul-16 3:04am

Add a Solution

Comments

[no name] 22-Jul-16 8:43am

Do you mean nanosleep, like described here nanosleep(2): high-resolution sleep - Linux man page[^]?

2 solutions

Add a Solution

Add your solution here

Treat my content as plain text, not as HTML

Preview 0

…

Existing Members

Sign in to your account

...or Join us

Download, Vote, Comment, Publish.

Your Email
Password
Forgot your password?

Your Email
This email is in use. Do you need your password?
Optional Password

I have read and agree to the Terms of Service and Privacy Policy
Please subscribe me to the CodeProject newsletters

When answering a question please:

Read the question carefully.
Understand that English isn't everyone's first language so be lenient of bad spelling and grammar.
If a question is poorly phrased then either ask for clarification, ignore it, or edit the question and fix the problem. Insults are not welcome.
Don't tell someone to read the manual. Chances are they have and don't get it. Provide an answer or move on to the next question.

Let's work to help developers, not make them feel stupid.

This content, along with any associated source code and files, is licensed under The Code Project Open License (CPOL)

Richard MacCutchan · Answer 1 · 2016-07-22T02:44:00

Solution 1

You cannot use timers for absolute timing in a multi-activity operating system. When a timer expires that merely means that the operating system places your activity (process) back on the execution queue, so it will resume the next time it reaches the top of the queue. This will be after 1 or more nanoseconds.

See nanosleep(2): high-resolution sleep - Linux man page[^].

Posted 22-Jul-16 2:44am

Richard MacCutchan

Updated 22-Jul-16 2:46am

v2

Comments

chandrAN2& 22-Jul-16 9:00am

Thanks for your reply. I am not talking about absolute timing. My question here is kernel supports 1ns timer granularity? If so how via sleep or busyloop(bogus mips?)

Richard MacCutchan 22-Jul-16 9:52am

I don't know the details without looking at the Linux source.

Jochen Arndt · Answer 2 · 2016-07-22T03:04:00

You will probably not be able to sleep for a single nano second because the available resolution depends on the granularity of the underlying clock which is system dependant. If the underlying clock has not a resolution of a nano second or less, the passed value will be by rounded up.

See also the OpenGroup specification nanosleep[^] and the nanosleep(2): high-resolution sleep - Linux man page[^].

To get the resolution of the hardware clock see clock_getres(2): clock/time functions - Linux man page[^]. Note that there is also the clock_nanosleep(2) - Linux man page[^] function which allows specification of the clock to be used.

When using sleep functions only the current thread (or the current process when it is single threaded) will be waiting. Other threads including the kernel will continue operation.