The way I read that bit of code is this: The function "ata_get_capacity" first finds the capacity of the disk in 28-bit mode, then on line 454 it looks to find the maximum version of the ATA spec that the drive supports. If this is less than version 6, then it stops there and returns zero - i.e. success. So in this case the function would return the 28-bit capacity of the drive, which maxes out at 128GB.
If the drive *does* support 48-bit mode, then it goes on and does some checks, before actually returning the 48-bit capacity. The change to fix this for VMware is therefore trivial - ignore the check for version 6 of the ATA spec. Given that the code does some checks before it actually tries to compute the 48-bit size of the disk, it *should* be relatively safe to make this change. Perhaps this is one for Indiana to experiment with? What is the performance of VMWare like with 100GB+ drives? Cheers Andrew. This message posted from opensolaris.org _______________________________________________ opensolaris-help mailing list [email protected]
