[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: Some questions
- To: Mailing List Recipients <pci-sig-request@znyx.com>
- Subject: RE: Some questions
- From: "Kimmery, Clifford (FL51)" <kimmery@space.honeywell.com>
- Date: Sun, 8 Dec 1996 16:42:27 -0500
- Encoding: 26 TEXT
- Resent-Date: Sun, 8 Dec 1996 16:42:27 -0500
- Resent-From: pci-sig-request@znyx.com
- Resent-Message-Id: <"eIe7d3.0.RZ4.tIpgo"@dart>
- Resent-Sender: pci-sig-request@znyx.com
I would define a FIFO as not prefetchable for exactly that reason.
Cliff Kimmery
Honeywell Inc.
kimmery@space.honeywell.com
>----------
>From: John R Pierce[SMTP:pierce@hogranch.com]
>Sent: Friday, December 06, 1996 9:58 PM
>To: Kimmery, Clifford (FL51); pci-sig@znyx.com
>Subject: Re: Some questions
>
>Kimmery, Clifford (FL51) <kimmery@space.honeywell.com> writes...
>>
>> Does this mean that most CPU-PCI bridges do not attempt to prefetch
>> unless the target address is known to be cacheable?
>>
>> An agent can be prefetchable (no read side-effects) without being
>> cacheable (no potential for hidden changes).
>
>prefetch CAN have side effects if it reads excessive words from a data
>transfer
>fifo (such as many graphics chips use).
>
>-jrp
>
y ø æ