WRITE_DMA problem again

[Kendall Gifford]

On 4/22/05, arax <cezar at arax.md> wrote:
> 
> Oh, no. Hope that's not true. I jsut finished installing and configure
> all the software, it would be a big pain starting over again. So, as
> far as I get it, there isn't a definite answer to this issue. You're
> suggesting that it's the drive, I saw some threads in the archives
> suggesting to apply some patches but they are still under development.
> I saw a lot of complaints regarding this write_dma thing, is it
> possible that all those people have problems with drives? Who knows?
> Maybe.

Howdy, I may, or may not, be experiencing the same problem as you. I
recently posted on freebsd-hardware yesterday but since I haven't
received any response yet, and since I noticed your post, I thought
I'd post here as well.

In my experience, my drives are fine (in addition to my drives,
cables, and motherboard being new, working fine with the FreeBSD 4.x
branch, being fine according to smartmontools, they're also fine
according to Maxtor's disk diagnostic utility). My previous post gives
the details of my situation if you're interested in comparing (or if
anyone has info for me):

http://groups-beta.google.com/group/muc.lists.freebsd.hardware/browse_thread/thread/4fbe97ece78c6d78/55635dcc817dc4c6?rnum=1&hl=en#55635dcc817dc4c6

Anyhow, my understanding is that the current work-around to this
problem is to turn DMA off for your ata drives. I guess this can be
done manually with the atacontrol command as I've been doing (though
as you'll see in my message linked above, I apparently need to do it
differently/sooner). I also understand that you can do so by setting 
hw.ata.ata_dma="0"  in  /boot/loader.conf  .

So, from doing a search on the problem, apparently there are those who
have hardware at fault and those of us who don't seem to have hardware
problems, but who know (I don't). If someone with more knowledge would
like to elighten us, it would be much appreciated.

--
Kendall Gifford