I was just playing around running lmbench on various machines around
here and saw what seem to be odd results. I'm wondering if anybody could
shed some light on them. Both machines below are running 1.2 release,
and are P5-166's.

On one system that has:

   wdc0 at isa0 port 0x1f0-0x1f7 irq 14
   wd0 at wdc0 drive 0: 2014MB, 4092 cyl, 16 head, 63 sec, 512 bytes/sec
   <QUANTUM FIREBALL_TM2100A>
   wd0: using 16-sector 16-bit pio transfers, lba addressing

I get the following figures (extracted from the raw data file created by
lmbench):

   Disk /dev/rwd0a bandwidth          4474 KB/sec
   File /usr/tmp/XXX write bandwidth: 3595 KB/sec

On the other hand, on a machine with:

   ahc0 at pci0 dev 18 function 0
   ahc0: interrupting at irq 11
   ahc0: Reading SEEPROM...done.
   ahc0: aic7880 Wide Channel, SCSI Id=7, 16 SCBs
   ahc0: Reseting Channel A
   ahc0: Downloading Sequencer Program...Done
   scsibus0 at ahc0
   ahc0: target 0 using 16Bit transfers
   ahc0: target 0 synchronous at 10.0MHz, offset = 0x8
   sd0 at scsibus0 targ 0 lun 0: <SEAGATE, ST15150W, 0023> SCSI2 0/direct fixed
   sd0: 4095MB, 3712 cyl, 21 head, 107 sec, 512 bytes/sec

I get:

   Disk /dev/rsd0a bandwidth          6677 KB/sec
   File /usr/tmp/XXX write bandwidth: 1466 KB/sec

So how is it that raw read performance is so much better on the SCSI
drive, but file system write performance is so much better on the IDE?
Does this have something to do with what lmbench is doing, or is it
something in interaction between FFS and the drivers? Clues?