Weird %busy in systat

[Joao Barros]

I'm doing a zpool scrub to exercise my system and noticed one strange
thing when looking at the output of 'systat -vm':

Disks   ad0   ad1   ad4   ad6   ar0   ar1
KB/t  44.87 44.71 45.47 45.44  0.00  0.00
tps     162   163   160   162     0     0
MB/s   7.11  7.13  7.11  7.18  0.00  0.00
%busy   100   100    63    62     0     0

where:

ad0: 305245MB <Seagate ST3320620AS 3.AAK> at ata0-master SATA150
ad1: 305245MB <Seagate ST3320620AS 3.AAK> at ata0-slave SATA150
ad4: 305245MB <WDC WD3200KS-00PFB0 21.00M21> at ata2-master SATA150
ad6: 305245MB <Seagate ST3320620AS 3.AAJ> at ata3-master SATA150

are connected to:

atapci0: <Promise PDC20319 SATA150 controller> port
0xb000-0xb03f,0xb400-0xb40f,0xb800-0xb87f mem
0xfc024000-0xfc024fff,0xfc000000-0xfc01ffff irq 23 at device 4.0 on
pci4
atapci1: <Intel 6300ESB SATA150 controller> port
0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xf000-0xf00f irq 18 at device
31.2 on pci0

and my zpool is:

config:

        NAME        STATE     READ WRITE CKSUM
        r4x320      ONLINE       0     0     0
          raidz1    ONLINE       0     0     0
            ad0s1d  ONLINE       0     0     0
            ad1s1d  ONLINE       0     0     0
            ad4     ONLINE       0     0     0
            ad6     ONLINE       0     0     0

Why is %busy so different when the disks are doing the same thing?
The only difference is the SATA controllers and zfs using a slice on
half of them and a whole disk with the other half.
What is taken into account when calculating %busy? I can go read code
but was hoping for someone with specific knowledge to answer :)

-- 
Joao Barros