On 7/27/07, Brian Gupta <[EMAIL PROTECTED]> wrote:
> I knew that libthread was using it's own scheduler, but couldn't wrap my
> head around how it would go about making scheduling choices

The first edition of the "Solaris Internals" book has a few pages on
"scheduler activations".

For more info, just google for "scheduler activations" - you will find
quite a number of papers.

However, for some reasons, the page rank of this paper is low, so I
will give you the link here:
http://web.mit.edu/nathanw/www/usenix/freenix-sa/freenix-sa.html

>, and for that
> matter why the design of MxN was being used. (It seemed a bit black
> box to me)

I remember the "Dinosaurs OS book" talks about why M:N scales better
than 1:1. However, as hardware becomes more powerful, the difference
is not as big, yet 1:1 is a lot cleaner to implement, and has become
the default thread model in Solaris 9 and Linux 2.6.

> It would be interesting in an academic sense, if it's not to hard to find.

A site has archived the old pages of SunWorld, however some of the
links are dead if you just follow the index page:

http://sunsite.uakom.sk/sunworldonline/common/swol-backissues-columns.html#insidesolaris

I did a little hand URL editing, and the followings are supposed to be
all that are related to the old thread model in Solaris.
http://sunsite.uakom.sk/sunworldonline/swol-08-1998/swol-08-insidesolaris.html
http://sunsite.uakom.sk/sunworldonline/swol-09-1998/swol-09-insidesolaris.html
http://sunsite.uakom.sk/sunworldonline/swol-10-1998/swol-10-insidesolaris.html
http://sunsite.uakom.sk/sunworldonline/swol-12-1998/swol-12-insidesolaris.html
http://sunsite.uakom.sk/sunworldonline/swol-01-1999/swol-01-insidesolaris.html
http://sunsite.uakom.sk/sunworldonline/swol-02-1999/swol-02-insidesolaris.html
http://sunsite.uakom.sk/sunworldonline/swol-03-1999/swol-03-insidesolaris.html

Rayson


>
> Thank you for your answers.
>
> Also the Solaris 9 doc that Rayson linked to was very helpful.
>
> Thanks,
> -Brian
>
> _______________________________________________
> perf-discuss mailing list
> [email protected]
>
_______________________________________________
perf-discuss mailing list
[email protected]

Reply via email to