Re: Low-latency / scheduling latency in 2.5

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, 2002-12-12 at 04:42, Guennadi Liakhovetski wrote:

> I've searched to no avail the internet to clarify one question:
> my current understanding is, that was used to be the low-latency patch
> by Andrew Morton is now included (a modified version thereof) in 2.5, as
> well as the preemptible kernel patch by Robert Love (credits to Ingo
> Molnar and others are due too)? The preemptibility is enabled by a
> configuration option, but what about the low- (scheduler-) latency? Is it
> just always on, or are they enabled / disabled together or what?

Hi.  You are confusing a few things, I think.

The preemptive kernel was merged and is enabled via CONFIG_PREEMPT. 
This is the basis for low latency in 2.5.

A new scheduler, commonly called the O(1) scheduler, was merged.

Finally, the low latency patch was not merged.  Instead, much work went
in to tuning up parts of the VM and VFS that were slow and
unpreemptive.  A couple "low latency"-style lock breaking did go into
2.5 just recently, however, in areas where the latency problems could
not be be solved via the cleaner algorithmic rewrite.

All in all 2.5 is very nice.

	Robert Love

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/


[Index of Archives]     [Newbies FAQ]     [Linux Kernel Mentors]     [Linux Kernel Development]     [IETF Annouce]     [Git]     [Networking]     [Security]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux RAID]     [Linux SCSI]     [Linux ACPI]
  Powered by Linux