On Wednesday, April 16, 2003, at 11:06  PM, Caffeinate The World wrote:

> What was the work-around? I am thinking of testing it out on a Seagate
> Barracuda 80 gig (ST380013AS):

 From the wd_quirk_table[] in sys/dev/ata/wd.c:

     /*
      * Some Seagate S-ATA drives have a PHY which can get confused	 * 
with the way data is packetized by some S-ATA controllers.
      *
      * The work-around is to split in two any write transfer who's
      * sector count % 15 == 1 (assuming 512 byte sectors).
      *
      * XXX This is an incomplete list.  There are at least a couple
      * XXX more model numbers.  If you have trouble with such transfers
      * XXX (8K is the most common) on Seagate S-ATA drives, please
      * XXX notify thorpej@netbsd.org.
      */
     { "ST3120023AS",
       WD_QUIRK_SPLIT_MOD15_WRITE },

So, basically, if you buy the drive, and you have trouble newfs'ing it 
or using the resulting file system (commands timing out), then add your 
drive's model string to the quirk table with this quirk.

> My apologies if I offended anyone with all this commercialism to the
> list.

Don't sweat it -- it's always good to know where you can get new 
hardware that NetBSD works with.

         -- Jason R. Thorpe <thorpej@wasabisystems.com>