[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: WIndowsNT "snoop" cycles before doing DMA
- To: Mailing List Recipients <pci-sig-request@znyx.com>
- Subject: Re: WIndowsNT "snoop" cycles before doing DMA
- From: "Jeff" <jmd@audio2.engr.sgi.com>
- Date: Thu, 25 Mar 1999 12:36:06 -0800
- Delivered-To: pcisig@teleport.COM
- Resent-Date: Thu, 25 Mar 1999 13:09:49 -0800
- Resent-From: pci-sig-request@znyx.com
- Resent-Message-ID: <"ds_a11.0.72._tf-s"@electra.znyx.com>
- Resent-Sender: pci-sig-request@znyx.com
> OK, this is a question which more related to WindowsNT and not to PCI.
> Sorry for this. But perhaps somebody can help us because these
> "snooping" RW cycles
> produce a performance reduction and leads to a large RW activity (mix
> of DMA cycles and
> "snooping" RW cycles on the PCI bus) with a lot of side effects.
This touches on a Microsoft question I have had for some time
about PCI peer-to-peer DMA. Microsoft, has stated that they
wil _not_ support PCI peer-to-peer DMA. Unfortunately, They
don't say why, does anybody know? My guess is that it has
to do with coherency issues, but I would like to here it
from somebody that knows, any Microsoft people on this list?
Jeff