Re: Adaptive thread creation by the kernel

From: Chuck Lever (cel@monkey.org)
Date: Wed Feb 02 2000 - 15:37:26 EST


On Wed, 2 Feb 2000, Jamie Lokier wrote:
> Chuck Lever wrote:
> > if (mincore(map, readlength, vec))
> > perror("mincore");
> >
> > the vec[] contains a byte for each page you queried. each byte has value
> > "1" if the page is already in memory, otherwise "0".
>
> You do a syscall every time you're going to read part of a file?

you can choose to do this, yes.

> How do you ensure it's still there by the time you actually read it?

the only way to ensure a page is in memory is to use mlock(). mincore()
only provides a hint.

> How do you force the page in when it's not in memory -- spawn
> another thread?

madvise(WILLNEED) is one way to do this asynchronously.

see Vivek Pai's work on the FLASH web server.

        - Chuck Lever

--
corporate:	<chuckl@netscape.com>
personal:	<chucklever@netscape.net> or <cel@monkey.org>

The Linux Scalability project: http://www.citi.umich.edu/projects/linux-scalability/

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Mon Feb 07 2000 - 21:00:08 EST