My best guess as to the reason for this bug is that the kernel identifies some disks as "SEMB devices", whatever those are. See drivers/ata/libata-core.c, function ata_dev_classify (line 1082 in 2.6.27):
I modified the printk to report the values (as shown above), and got the values lbam=0x3c and lbah=0xc3. I have no idea what those mean, but maybe they're helpful.
My best guess as to the reason for this bug is that the kernel identifies some disks as "SEMB devices", whatever those are. See drivers/ ata/libata- core.c, function ata_dev_classify (line 1082 in 2.6.27):
if ((tf->lbam == 0x3c) && (tf->lbah == 0xc3)) {
printk( KERN_INFO "ata: SEMB device ignored (lbam=0x%lx lbah=0x%lx\n", (unsigned long) tf->lbam, (unsigned long) tf->lbah);
return ATA_DEV_SEMB_UNSUP; /* not yet */
}
I modified the printk to report the values (as shown above), and got the values lbam=0x3c and lbah=0xc3. I have no idea what those mean, but maybe they're helpful.