Re: Generic PCI IDE bus-mastering DMA support

john hood (cgull@smoke.marlboro.vt.us)
Sat, 20 Sep 1997 22:46:28 -0400 (EDT)


mlord writes:
> To boost your write performance to similar high levels,
> cautiously use this:
>
> ## turns on IDE write caching
> hdparm -W1 /dev/hd[a-d]

Mmm. How interesting. I hadn't even considered this problem-- it's
clearly coming up on one of my bad-sector junk drives, an HP-OEMed
Quantum FB640AT.

> I first need to hear back from VIA VP-1 chipset users though..

The generic code in 2.1.56 works OK on a PCChips VXPro motherboard,
with a relabeled VP-1. It appears we've now switched our stands: Your
code no longer initializes the timing on chipsets at all, whereas mine
is now going to considerable pains to do so :) Consider, for example,
a drive not configured in the BIOS. From what I can see, most BIOSes
will not bother to configure IDE timing then, leaving very slow
default values in place. The BIOS on my VXPro motherboard, and I
think my Biostar Intel VX motherboard, both do this.

(I'm one author of FreeBSD's IDE DMA code.)

I noticed also that Kim-Hoe Pang's ultra-DMA patch would turn on ultra
DMA mode on any ultra-DMA drive, even on older chipsets that don't
support it. That could be unpleasant.

--jh

-- 
John Hood				cgull@smoke.marlboro.vt.us

Predictably, they all eventually wandered away, rubbing their bruises and brushing mud out of their hair. Some went off to work for the ESA, launching much smaller rockets into low orbits, while others elected to sit on their front porches drinking Jim Beam from the bottle and launching bottle rockets from the empties. [Jordan Hubbard]