[lkml]   [1996]   [May]   [1]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: IDE Disk speed enhancements

On Mon, 29 Apr 1996, David A Willmore wrote:

> On Apr 27, 3:59pm, mlord wrote:
> > David A Willmore wrote:
> > > I have tried this recently, but with my qd6580 VLB IDE interface, this
> seems to
> > > make no difference. I have the append "ide0=qd6580" in my lilo config and
> the
> > > kernel parses it correctly. Does this interface serialize and I don't know
> > > about it?
> >
> > See qd6560.c for this info. The kernel does NOT serialize this
> > interface, but then the kernel also only knows how to tune ide0
> > for the qd6560 -- does it actually have a second set of IDE ports??
> Right, I looked at that, but, from my 'observation' of how the qd6580 DOS
> driver works, the qd6580.c driver may be doing more than it thinks. I'll know
> more when I get a copy of DOSEMU which logs I/O accesses.
> Yes, my 'Vision' brand VLB super I/O card has two IDE ports on it. One seems
> to be an EIDE port and the second seems to be a normal ISA IDE port. I haven't
> fully verified this.

This is basically how I got the information for this interface. My Vision
VLB only had the primary ide port on it. I used a combination of tools to
get the info for programming this interface, but dosemu IO logging will not
allow the QD driver to finish loading. So I had to goto some DOS tools in
order to get the rest of the information. The information I send to the
card to initialize it is what the DOS driver sends for the primary ide
port. On my system the speed increase was noticable so I submitted the
info to Mark Lord. It seems to work for most people.

Colten Edwards

 \ /
  Last update: 2005-03-22 13:37    [W:0.041 / U:0.032 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site