[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: (Fwd) Re: 66 MHz Capability bit in AGP

Monish Shah, quoting from the spec:

> "The 66 MHz PCI clock generation circuitry must connect to M66EN to
> generate the appropriate clock for the segment (33 to 66 MHz if M66EN is
> asserted, 0 to 33 MHz if M66EN is deasserted)."  (page 221)
> "All 66 MHz PCI agents must support a read-only 66MHZ_CAPABLE flag located
> in bit 5 of the PCI status register for that agent.  If set, the
> 66MHZ_CAPABLE bit signifies that the agent can operate in 66 MHz PCI mode."
> (page 220)

My interpretation: The first clause prevents a non-66-capable device from
being driven by a 66-MHz clock. The second clause allows the host to query
all the devices on the bus, and if they all report 66-MHz capability, the
host is entitled to assume they're all running at 66.

So I agree, we do need to use the 66MHZ_CAPABLE flag to distinguish between
66-MHz PCI and 66-MHz AGP silicon.

In retrospect, I think the spec should have required the host to read the
status of the M66EN line to indicate whether the bus is actually running at
66-- a low-level physical-layer kind of issue. Then the 66MHZ_CAPABLE flag
could have been used for the higher-level purpose of indicating whether the
device _can_ operate at 66 MHz, regardless of electrical signalling issues.

> But what about a true 66 MHz PCI chip that happens to be put on a board
> that grounds M66EN?  I don't know why you'd do this, but this actually
> seems legal, based on what I quoted above!

You'd get the same effect if someone plugs a 33-MHz PCI card into one slot
on a 66-MHz motherboard; the system can't tell who deasserts M66EN.

> So, I agree that it isn't clear from the spec that the host is free to look
> at either one to decide which frequency to run at.

The actual bus clock must be determined by M66EN, but the software doesn't
have to look at the actual bus clock to determine how much data it can
attempt to pump through the bus.

> You should order rev 2.1s of the spec.  The "s" at the end means that along
> with the paper copy, you get a couple of floppies that contain an
> electronic copy of the spec.  It is in Acrobat format and does give you
> rudimentary search capabilities.  That can be extremely useful.

Not as useful as a URL. But if I decide to make my own URL (for my personal
convenience) I'll buy the SIG's electronic version of the spec.

.                       png

P?