As Jos Backus wrote ...
> On Tue, Jun 08, 1999 at 12:53:54AM -0600, Warner Losh wrote:
> > Isn't that 24 bits for addresses? You can dma from an ISA card to
> > anywhere in the first 16M...
>
> Arrgh, yes, I'm terribly confused. Sorry 'bout that.
Be happy, you're not alone ;-)
| / o / / _
: Tuesday, June 08, 1999 1:54 AM
Subject: Re: The choice of MAXPHYS
>In message <19990603231216.a36...@hal.mpn.cp.philips.com> Jos Backus
writes:
>: On Thu, Jun 03, 1999 at 07:30:20PM +0200, Wilko Bulte wrote:
>: > 20 bits. But older cards can do no more than 64 kB.
>:
>: I
On Tue, Jun 08, 1999 at 12:53:54AM -0600, Warner Losh wrote:
> Isn't that 24 bits for addresses? You can dma from an ISA card to
> anywhere in the first 16M...
Arrgh, yes, I'm terribly confused. Sorry 'bout that.
--
Jos Backus _/ _/_/_/ "Reliability means never
In message <19990603231216.a36...@hal.mpn.cp.philips.com> Jos Backus writes:
: On Thu, Jun 03, 1999 at 07:30:20PM +0200, Wilko Bulte wrote:
: > 20 bits. But older cards can do no more than 64 kB.
:
: Indeed, 20 bits (=1 Mbyte) for the address, 16 bits for the transfer counter
: (offset).
Isn't th
In message
Zhihui Zhang writes:
: The value of MAXPHYS is chosen to be 64K for the maximum raw I/O transfer
: size. I am wondering why it is not set larger.
I don't think that it is possible to guarantee that you can do a
larger write than 64k on a aha-1542 card given the worst case scatter
gath
>
> MAXPHYS is ultimately a limitation on the size of the b_pages array in
> the struct buf structure. The default is 128 KBytes. Device Drivers are
> supposed to break up I/O's into manageable sections themselves.
>
> Maybe when the buffer I/O / VFS subsystem is rewritten late
:On Thu, Jun 03, 1999 at 07:30:20PM +0200, Wilko Bulte wrote:
:> 20 bits. But older cards can do no more than 64 kB.
:
:Indeed, 20 bits (=1 Mbyte) for the address, 16 bits for the transfer counter
:(offset).
:
:--
:Jos Backus _/ _/_/_/ "Reliability means never
MAXPHY
On Thu, Jun 03, 1999 at 07:30:20PM +0200, Wilko Bulte wrote:
> 20 bits. But older cards can do no more than 64 kB.
Indeed, 20 bits (=1 Mbyte) for the address, 16 bits for the transfer counter
(offset).
--
Jos Backus _/ _/_/_/ "Reliability means never
As Zhihui Zhang wrote ...
>
> The value of MAXPHYS is chosen to be 64K for the maximum raw I/O transfer
> size. I am wondering why it is not set larger. The maxcontig value of FFS
> is default to be 16, which means 16*8192 or 128K bytes (twice as big as
> 64K) . If we raise the value of MAXPHYS,
As Jos Backus wrote ...
> On Thu, Jun 03, 1999 at 10:40:10AM -0400, Zhihui Zhang wrote:
> > The value of MAXPHYS is chosen to be 64K for the maximum raw I/O transfer
> > size. I am wondering why it is not set larger.
>
> Just a guess: maybe this has something to do with DMA address counters on ISA
On Thu, Jun 03, 1999 at 10:05:46AM -0600, Kenneth D. Merry wrote:
> I think the 64K value may have been chosen because cards like the Adaptec
> 1542 can't handle more than that.
That's exactly the card I was thinking about (I used to own one). So I was
close :)
--
Jos Backus
Jos Backus wrote...
> On Thu, Jun 03, 1999 at 10:40:10AM -0400, Zhihui Zhang wrote:
> > The value of MAXPHYS is chosen to be 64K for the maximum raw I/O transfer
> > size. I am wondering why it is not set larger.
>
> Just a guess: maybe this has something to do with DMA address counters on ISA
> c
On Thu, Jun 03, 1999 at 10:40:10AM -0400, Zhihui Zhang wrote:
> The value of MAXPHYS is chosen to be 64K for the maximum raw I/O transfer
> size. I am wondering why it is not set larger.
Just a guess: maybe this has something to do with DMA address counters on ISA
cards being 16 bit? (ducks) Or is
13 matches
Mail list logo
