This message is from the T13 list server.

 Not exactly.  1510D includes ADMA, which your information is correct.
However non-ADMA HBA's are harder to detect.  I can tell you everything
with Class+SubClass 0101 and VID=8086 are generally compliant with this,
but UDMA capabilities are dependent on the DevID found.  In general you
have to build a VID/DID/CC/SCC table in a driver/BIOS to determine what
maps to the non-ADMA features described in the document.  We put this
stuff in to just document existing designs.  MKE.



-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of
Curtis E. Stevens
Sent: Wednesday, February 18, 2004 9:33 AM
To: [EMAIL PROTECTED]
Subject: Re: [t13] How to detect whether a host adapter complies with
1510D 

This message is from the T13 list server.


Thomas

    You detect compliance by looking at the PCI Class and Subclass.
Class =
01, Subclass = 05, indicates that the HBA is compliant with INCITS 370
(1510D).
------------------------------------------------
Curtis E. Stevens
29 Dewey
Irvine, Ca. 92620

Home Phone: 949-552-4777
Cell: 949-307-5050
E-Mail: [EMAIL PROTECTED]
----- Original Message ----- 
From: "Thomas Jansen, WTY Soft" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Wednesday, February 18, 2004 5:37 AM
Subject: [t13] How to detect whether a host adapter complies with 1510D


> This message is from the T13 list server.
>
>
>  >Hi,
>  >I have been playing with host adapter programming a bit lately and
found
> document 1510D. By now, I know that the bus-master part works on most
host
> adapters I've seen but the setup >registers in annex B are not. The
way I
> read the spec they don't have to do, so I guess that's ok. The only
thing
> I'm missing in the spec is how to detect that the host adapter
> supports >the bus-master programming and / or setup registers.
>  >
>  >Can anyone clarify this to me ?
>  >
>  >Thanks in advance,
>  >Thomas
>
> What I mean is how can I detect whether a host adapter complies with
this
> standard.
>
> Thomas
>
>
>
>
>

Reply via email to