xfs
[Top] [All Lists]

Re: Errr... what's all this about then?

To: Bruce Tenison <btenison@xxxxxxxxxxxxx>
Subject: Re: Errr... what's all this about then?
From: Walt H <waltabbyh@xxxxxxxxxx>
Date: Sat, 26 May 2001 16:42:19 -0700
Cc: Juha Saarinen <juha@xxxxxxxxxxxx>, linux-xfs@xxxxxxxxxxx
References: <LGECJLHBGAPMGAJFMAFIOEEGCAAA.juha@saarinen.org> <3B0FCC24.50903@earthlink.net>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux 2.4.4-xfs i686; en-US; rv:0.9+) Gecko/20010525
Hello,

Just wanted to chime in with a FWIW post :)

I've also got a Promise PDC 20262 w/ an IBM Deskstar running in UDMA mode 4 w/ no problems whatsoever. Additionally a Maxtor Diamond Max connected to the stock BX MB w/ no problems. Early on (5-6 months ago) I did experience some minor corruption problems with the Maxtor connected directly to the mainboard, but never anything with the IBM on the Promise card. That thing's been solid. All fs's are XFS except boot which is reiserfs.

-Walt


Bruce Tenison wrote:

Strange. I'm experiencing the same problem on a Promise PDC20262 controller card too!
Another thing, is that I've never, well... mostly never ;) , experienced problems with other filesystems
on either chipset (I know there are issues with the VIA stuff, had problems with the USB stuff too,
makes me wonder why I bought the MB in the first place...)


Anyway, I'm gonna try and use a different drive and see if I experience the same problems. Using a seperate
partition, mounted on my ext2 root, I haven't been able to make it fail. Gonna install RH 71 on the XFS
partition and make it a root, and see if it'll fail there (as it does on my Fujitsu drive)


Juha Saarinen wrote:

:: using it on a
:: VIA vt82c596b (rev 23) chipset, both showing the same style corruption
:: problems. Something's amiss!


That's what we've got in common -- the VIA chip set. Nothing but trouble
with that, to be honest. Memory performance is slower than on 440BX chip
sets (trying to work out how to tweak the PCI registers with setpci so as to
enable memory interleaving) and the IDE interface works poorly.


Also discovered that there's an issue with the IO-APIC and Linux. If I
disable it (append="noapic" in lilo.conf or at the boot prompt), I don't see
the weird slowdown in disk performance.




-- Juha









<Prev in Thread] Current Thread [Next in Thread>