xfs
[Top] [All Lists]

Re: xfs block size

To: Andi Kleen <ak@xxxxxxx>
Subject: Re: xfs block size
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Wed, 17 Jan 2001 21:19:26 +0100
Cc: "Davida, Joe" <Joe_Davida@xxxxxxxxxx>, "'linux-xfs@xxxxxxxxxxx '" <linux-xfs@xxxxxxxxxxx>
In-reply-to: <20010117211319.A5369@gruyere.muc.suse.de>; from ak@suse.de on Wed, Jan 17, 2001 at 09:13:19PM +0100
References: <09D1E9BD9C30D311919200A0C9DD5C2C025370A4@mcaexc01.msj.maxtor.com> <20010117203852.B22384@caldera.de> <20010117211319.A5369@gruyere.muc.suse.de>
Sender: owner-linux-xfs@xxxxxxxxxxx
On Wed, Jan 17, 2001 at 09:13:19PM +0100, Andi Kleen wrote:
> > But it is not Linux.  They do organize things differently.
> > (Linux UFS support larger blocks too, but it seems pretty b0rken
> > currently)
> 
> iirc one of the plans for 2.5 is to move to bigger logical (and physical on
> systems that support it like ia64) page sizes similar to Irix.

IMHO this is a pretty good idea, and I think it's time to do it. 
Daniel Phillips' variable page sizes idea looks also pretty interesting in
this context.

> Hopefully that will help XFS too.

I bet.  On the other hand XFS does pretty well even with 4K blocks/pagesize 
due to the use of extends.

> 
> vmalloc got a bit cheaper in 2.4 now, no ipi needed on alloc anymore, just on
> free, but it's probably still not a real option for any regular executed 
> path. 

Agreed.


        Christoph

-- 
Whip me.  Beat me.  Make me maintain AIX.

<Prev in Thread] Current Thread [Next in Thread>