It seems Greg Lehey wrote:
> >>
> >> I think the old wd code was broken on reporting. The reporting
> >> was changed from the original submission, when it was committed.
> >
> > Yep, but it does mean that people _belive_ they are running UDMA
> > where in fact they aren't, and now they blaim the ata driver
> > for doing worse :(
>
> Well, the wd driver definitely ran DMA on the 5591. And the fix
> Richard reports is not normally output; IIRC it's only on a verbose
> boot. On my machine, the driver reports:
>
> ide_pci0: <SiS 5591 Bus-master IDE Controller> irq 14 at device 0.1 on pci0
> ...
> wdc0 at port 0x1f0-0x1f7 irq 14 flags 0xb0ffb0ff on isa0
> wdc0: unit 0 (wd0): <ST51270A>, LBA, DMA, 32-bit, multi-block-32
> wd0: 1223MB (2504880 sectors), 621 cyls, 64 heads, 63 S/T, 512 B/S
> wdc1 at port 0x170-0x177 irq 15 flags 0xb0ffb0ff on isa0
> wdc1: unit 0 (wd2): <IBM-DHEA-36480>, LBA, DMA, 32-bit, multi-block-16
> wd2: 6197MB (12692736 sectors), 790 cyls, 255 heads, 63 S/T, 512 B/S
> wdc1: unit 1 (wd3): <IBM-DHEA-38451>, LBA, DMA, 32-bit, multi-block-16
> wd3: 8063MB (16514064 sectors), 1027 cyls, 255 heads, 63 S/T, 512 B/S
>
> DMA is definitely enabled. There's no mistaking the difference
> between DMA and no DMA.
That was not what I said, it may run DMA but it says UDMA when not
and vice versa, there is a difference between WDMA and UDMA....
> > I'm not sure the other values reported are valid either as they
> > are reported differently from what they are set to....
>
> I took a look at this a while back and never got round to committing
> the changes. I believe most of the discrepancies were minor. I'll
> take charge of fixing that if you like.
I dont care, but for 3.x that might be a good idea....
-Søren
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
