Subject: Re: is SUNW,fas an esp(4)?
To: Matthew Jacob <mjacob@feral.com>
From: Eduardo Horvath <eeh@turbolinux.com>
List: port-sparc
Date: 08/17/2000 09:59:12
On Wed, 16 Aug 2000, Matthew Jacob wrote:

> No. It's not. It's *close*, but I believe it's either a ESP-266 or
> ESP366 chip, not the ESP-100A that esp(4) drives. I have documentation if
> somebody wants to take this on (ENOTIME). 

It's a FAS366/ESP366 chip.

> It *should* be driven by the esp(4) driver. Insofar as my documents tell me,
> it's the same logic, but wide. The same logic, btw, which is then wrapped
> inside of a little RISC sequencer which is that which makes up the isp(4)
> cards.

The main issue with getting the esp driver to work with fas is that esp
expects the DMA engine to be on a separate OBP node but fas has only one
OBP node and two sets of registers, one for the SCSI chip and the other
for the DMA engine.  Otherwise it should be trivial to get to work.  The
FAS366 docs should be available at the Sun website.  ISTR it describes the
entire ASIC with the HME, FAS366, and DMA engine.

I've been considering writing a driver for it but don't have the H/W.

> On Wed, 16 Aug 2000, Erik E. Fair wrote:
> 
> > I have a SunSwift Fast Ethernet (hme)/Wide SCSI (fas) Sbus card, sun 
> > part # 501-2739. Alas, NetBSD pops up this SCSI controller as "not 
> > configured" at boot. Is this not another variant of the "esp" and 
> > therefore a trivial thing to enable?

Eduardo Horvath