Date: Wed, 8 May 1996 21:09:34 -0400 (EDT) From: Richard Toren <rpt@miles.sso.loral.com> To: questions@freebsd.org Cc: techsup@buslogic.com Subject: Buslogic 445C & AMD DX/4-133 fails Message-ID: <Pine.SUN.3.91.960508204153.21300A-100000@miles>
next in thread | raw e-mail | index | archive | help
Oh SCSI masters...
I have been using the BusLogic 445C controller with my DX-66 system
for the past 2 years. I am now trying to upgrade my system with a new
motherboard and processor.
The bottom line is that all tries have ended in a situation where the
'post' completes, but the boot hangs where it should be accessing the disc.
I have been through 3 motherboards and 2 processors and they all act
the same way.
? has anyone used a BT445C at a speed above 66 MHZ?
The different components have been a:
'Pine' motherboard (that turned out not to do bus mastering).
'HOT-419' 486 VL-Bus Green Main Board.
2 different AMD Am5x86 -P75 \n AMD-X5-133ADW processors.
With each of the possible combinations, I did find that if I disabled
the internal cache, occasionally it would boot (even the non-mastering) one.
The HOT-419 let me adjust the VLB speed (>33, <33) insert a VLB write wait
state, adjust the 'RDY# delay' and the 'ADS#" signal. All to no avail.
The response from BusLogic was :
>Dear Sir: There have been numerous reports of problems with L1 caching
>with some of the 5x86 and AMD chips when using a busmastering controller,
>which results in a non-boot situation unless the cache is disabled. That
>sounds like what you're running into. The typical external VL bus speed
>is 33 MHz, and the ISA bus speed should still be around 8 MHz.
>Busmastering is required in the slot the card's using, so you might check
>to be sure that it's enabled. (It's also possible that the L1 cache
>enable turns off busmastering.)
>
>The adapter itself should be able to handle 40 MHz in a stable VL
>motherboard, but we have had reports of some of the clone motherboards
>becoming unstable at bus speeds over 33 MHz.
>
>Your bios and firmware are not current, but that should not prevent the
>card from booting.
>
>Regards,
>Jo Kellner
>BusLogic Technical Support
? Does this sound reasonable ? What are the odds of two defective AMD
processors?
? Should I just get a new SCSI controller? If so what brand and model?
Needs to be bootable, handle 2+ drives, 2 GB drive size, Work with
FreeBSD, DOS / Windoze, prossibly bus mastering for performance?
? Does the bus mastering add significant performance?
? Shud I set this mb up for the kids with an EIDE disc as a screeming
'Doom' machine?
====================================================
Rip Toren | The bad news is that C++ is not an object-oriented |
rpt@miles.sso.loral.com | programming language. .... The good news is that |
| C++ supports object-oriented programming. |
| C++ Programming & Fundamental Concepts |
| by Anderson & Heinze |
====================================================
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.SUN.3.91.960508204153.21300A-100000>