RE: Heads up: checking in change to ata-card.c

> -----Original Message-----
> From: Julian Elischer [mailto:julian_at_elischer.org]
> Sent: Thursday, June 26, 2003 7:26 PM
> To: Will Saxon
> Cc: David Xu; Soeren Schmidt; Bill Paul; imp_at_freebsd.org;
> current_at_freebsd.org
> Subject: RE: Heads up: checking in change to ata-card.c
> 
> 
> Nvidia (I previously didn't know they made full chipsets)
> IDE interfaces got broken around FreeBSD 4.3 and at 4.8 they are still
> not working.. There is however a workaround which is to 
> disable the use
> of DMA on the devices (in the loader.conf file). if you don't 
> the system
> hangs during boot.
> 
> I have only sporadic access to one, but I don't have one myself which
> makes it hard to
> test more..
> I have not tested it on 5.x yet..

Well I am pretty sure this isn't an Nvidia chipset, actually I know it isn't:

atapci0: <VIA 8235 ATA133 controller> port 0xb000-0xb00f at device 17.1 on pci0
ata0: at 0x1f0 irq 14 on atapci0
ata1: at 0x170 irq 15 on atapci0
ad0: 9797MB <QUANTUM FIREBALLP LM10.2> [19906/16/63] at ata0-master UDMA66
acd0: MODE_SENSE_BIG command timeout - resetting
ata1: resetting devices .. done
..repeat 19 more times..
acd0: CDROM <CD-956E/AKV> at ata1-master PIO4

Maybe there are some similarities though - Nvidia seemed to just pull a good chipset out of nowhere a couple of years ago, maybe they licensed some stuff from Via? I know they worked a lot on the Xbox but that is an Intel system - the nForce stuff is AMD-only.

I have 4 operable machines, 3 of them working with this 694D board, the other (ironically) a 230T also (which is working fine with a 4.8-PRERELEASE kernel from 2/19). In a few weeks they will all be replaced by new P4 machines and I should have the ability to basically do whatever I want with them. So if this turns into a protracted issue that requires testing, I will be more than happy to test patches for 4.x and 5.x, single and dual processor.

-Will

> 
> 
> 
> On Thu, 26 Jun 2003, Will Saxon wrote:
> 
> > 
> > 
> > > -----Original Message-----
> > > From: David Xu [mailto:davidxu_at_freebsd.org]
> > > Sent: Thursday, June 26, 2003 6:30 PM
> > > To: Soeren Schmidt; Bill Paul
> > > Cc: imp_at_freebsd.org; current_at_freebsd.org
> > > Subject: Re: Heads up: checking in change to ata-card.c
> > > 
> > > 
> > > I have another problem different with Bill Paul's.
> > > I have a Tyan Tiger 230T dual CPU MB. On second IDE,
> > > there are MITSUBISHI 52X CDROM as master and a SONY
> > > CD-RW CRX140E 8X/4X/32X as slave. FreeBSD never got
> > > them to work, while MS Windows and Linux and NetBSD and
> > > OpenBSD all work well with these devices. FreeBSD always
> > > stuck at "MODE SENSE BIG", this sucks, I must disable
> > > second IDE in BIOS before booting into FreeBSD and
> > > reenable it before booting into another OS. I changed
> > > configuration several times, both DMA and PIO mode don't
> > > work, it works only when CR-RW is pulled.
> > > 
> > > David Xu
> > 
> > I started seeing this behavior also, but it started when I 
> swapped drives from a dead Via dual chipset board machine 
> (MSI 694D Pro, I think) into an athlon machine with an Asus 
> A7Vsomething. I can get correct model numbers, etc. if 
> helpful. This is a 4.8-RELEASE installation. I had to reboot 
> the first time with an old 4.4 GENERIC leftover from 
> installation, and I saw no errors, but after compiling a 
> uniprocessor kernel I started seeing the above behavior. 
> Although, I left it for awhile and eventually the machine 
> does boot after about 30 of the MODE_SENSE_BIG, resetting 
> devices lines.
> > 
> > -Will
> > _______________________________________________
> > freebsd-current_at_freebsd.org mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-current
> > To unsubscribe, send any mail to 
> "freebsd-current-unsubscribe_at_freebsd.org"
> > 
> 
>