Subject: Re: UVM aobj: Large VM objects.
To: None <pavel.cahyna@st.mff.cuni.cz>
From: YAMAMOTO Takashi <yamt@mwd.biglobe.ne.jp>
List: tech-kern
Date: 03/08/2006 14:24:18
> On Tue, Mar 07, 2006 at 10:09:22PM +0900, YAMAMOTO Takashi wrote:
> > > > however, i don't think long is appropriate here.
> > > > please introduce a 64-bit "page offset" type.
> > > > "typedef voff_t pgoff_t" should be fine.
> > > 
> > >  	Can't we use something similar to vm_offset_t, which is
> > >  	appropriately sized for the machine?
> > > 
> > >  	If we are ever going to need it to be 64bit on every arch with 32bit
> > >  	physical and virtual addresses then I agree it should be 64bit,
> > >  	but otherwise its yet another tiny performance loss/memory
> > >  	growth, continuing the pecking to death by ducks...
> > 
> > IMO, the appropriate size is arch-independent
> > as far as aobj is used for tmpfs or such.
> 
> So, if it was 32-bit, would it limit size of tmpfs files to 2^32 bytes, or
> 2^32 pages? The latter maybe isn't that serious.

the latter.

it might not be serious.  but i don't think the loss is serious either.

> P.S. the Cc: header of your message has really strange addresses. It is
> just me?

how?

YAMAMOTO Takashi