Subject: Re: Problem with aic7899
To: Justin T. Gibbs <gibbs@scsiguy.com>
From: Brian Buhrow <buhrow@lothlorien.nfbcal.org>
List: port-i386
Date: 02/13/2003 18:34:07
Hello. I've been following this discussion with some interest and I'd
like to ask what will probably turn out to be a silly question.
Justin, do you know if your driver uses more of the scratch RAM than the
Windows driver? Since Frank says the troubled machine is using an on-board
87899, I'm presuming that the one which works is on an external card? I'm
wondering if the mother board manufacturer might have scrimped with respect
to the amount of RAM provided for the chip's use, versus what's available
on most external cards? Is the amount of RAM supposed to be a fixed
number, or is the driver supposed to query the chip before it downloads the
sequencer code?
The reason I'm running along this line of thought is that when the
87899 first came out, I worked at SCO, in their corporate tech support
department. At the time, the ADaptec engineer charged with producing a SCO
driver for the chip ran into some interesting issues when faced with chips
installed on motherboards, as opposed to chips provided on PCI peripheral
cards. I don't remember the details of the issue, but I could believe it
had something to do with possibly different RAM configurations for the two
types of installation.
-Brian
On Feb 13, 7:30pm, "Justin T. Gibbs" wrote:
} Subject: Re: Problem with aic7899
} > I actually did patch the driver to use less SCBs at one point, but
} > I recall still seeing the problem. What would be the easiest way to
} > make it use less scratch ram?
}
} SCB ram is easy. Scratch ram is harder. A concerted effort would have
} to be put into adjusting the firmware to use less scratch ram. Scratch
} ram is used to store firmware state. The scratch ram layout is documented
} in aic7xxx.reg.
}
} --
} Justin
}
>-- End of excerpt from "Justin T. Gibbs"