Mailing Lists: Apple Mailing Lists
Image of Mac OS face in stamp
 
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: zero filling pages

Am Donnerstag, 24.07.03 um 01:55 Uhr schrieb Clark Cox:


On Wednesday, July 23, 2003, at 17:50, Nat! wrote:
"Clark S. Cox III" <email@hidden>

Zero filling is my bottleneck currently.

Are you sure? how did you determine this.

Using sample. The first routine to touch a page is really expensive.

That doesn't mean that it's the zero-filling that's at fault. Even without zero-filling, the first access to the page would be the most expensive. When you first allocate a page, that page is only logical, it hasn't necessarily been mapped to any physical memory. Only after you first access an address inside that page can you know that the page has been mapped in.


I assumed that most of the work was already done by vm_allocate, but this might very well be not the case.

I can't quite imagine, what could be taking so long, if it isn't zero filling. As far as I know, an available RAM page must be grabbed (lots unused here on my machine) and mapped in the MMU to the virtual address. I would think this much less expensive than clearing 4K.

Ciao
Nat!
_______________________________________________
darwin-development mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/darwin-development
Do not post admin requests to the list. They will be ignored.
References: 
 >Re: zero filling pages (From: Clark Cox <email@hidden>)



Visit the Apple Store online or at retail locations.
1-800-MY-APPLE

Contact Apple | Terms of Use | Privacy Policy

Copyright © 2007 Apple Inc. All rights reserved.